Mansour Gholami. (M.Sc. Horticulture, Tehran) l'[/ aite Agriculture Re s earch Institute. The University of Adelaide South Australia

Size: px

Start display at page:

Download "Mansour Gholami. (M.Sc. Horticulture, Tehran) l'[/ aite Agriculture Re s earch Institute. The University of Adelaide South Australia"

Jennifer Briggs
5 years ago
Views:

aite Agriculture Re s earch Institute The University f Adelaide Suth

1 {ù. ;"$."t c,þrûfu$ Qq*i'. 21 ci 1996 a Bisynthesis and translcatin f secndary ^"fu rr" glycsides in the grapevin e Vitis viniíera - '%Sry"V # "gð"/ by Mansur Ghlami (M.Sc. Hrticulture, Tehran) Department f Hrticulture, Viticulture and Oenlgt l'[/ aite Agriculture Re s earch Institute The University f Adelaide Suth Australia Thesis Submitted fr the Degree f Dctr f Philsphy ln The University f Adelaide Facttlty f Agriculture and Natural Resurces Science March,1996

2 Table f cntents Summary...- Declaratin.. Acknwledgements.. List f tables List f figures. List f publicatins. Abbreviatins Chapter l. I -l.y v...1{l.yrii..xu..eu I Chapter 2. Chapter 3 Chapter 4 J Chapter 5 Chapter 6 Chapter 7 Chapter I References Appendix. Publicatins lt7 t2t

4 Summary In this study different experiments were cnducted t investigate the site f bisynthesis f flavur cmpunds in the grapevine. Mst f the secndary metablites, including flavur cmpunds, are glycsylated and stred in plant tissues as glycsides. The chemical prperties f these cmpunds, especially their water slubility, suggests that glycsides might be pssible frms f translcated secndary metablites in plants. The levels f ttal glycsides in grapevine fruits and leaves was determined. Whle berries, leaf blades and petiles f three grape cultivars were sampled and extracted thrughut the grwing perid. The glycsidically bund fractin was islated by retentin n Cl8 reverse-phase silica gel, and after hydrlysis with sulfuric acid at 1 'C, the released glycsyl-glucse (G-G) was enzymatically determined. There were large quantities f glycsidically bund secndary metablites in the extracts f leaf blades but much lwer levels f glycsides were detected in petiles and berries. Seasnal changes in the amunts f glycsides, and als varietal differences between cultivars were bserved. The glycside levels increased n a per berry basis but decreased n a fresh weight basis during berry develpment. Muscat Grd, as a flral grape, and Sultana, as a neutral grape, had different patterns f ttal glycside accumulatin during ripening. The effect f leaf gentype n the berry flavu cnstituent prfile and levels was investigated b1, grafting bunches between flral and nn-flral grape cultivars. Inflrescences f Muscat Grd and Shiraz grapevines grwn in a glasshuse were grafted nt shts f each ther befre flwering, and grwn until ripe. Mnterpene glycsides were islated frnr the fruits, enzvmatically hydrlysed and the released mnterpene aglycnes liquid-liquid extracted and analysed by GC-MS. Muscat Grd fruit frm bth ungrafted cntrl vines and frm fruit grafted nt Shiraz vines yielded fruit with mnterpenes at levels and types typical f this flral grape variety. Cntrl Shiraz fruit (ungrafted) and Shiraz fìuit grafted nt Muscat Grd vines cntained nly lw levels f mnterpene glycsides. Similar results were btained in a field experiment with Muscat Grd and Sultana fruits similarly grafted, The lack f difference between flavur cmpunds in grafted and nn-grafted f uit suggests that berries have the ability t synthesise arma cmpunds independently frm the rest f the vine. Rse clratin f skin was seen t develp late during ripening n the nrmally white grape berries f cv. Muscat Grd Blanc. The nature f the pigment was investigated by HPLC analysis f skin extracts f single ber ies. The predminant anthcyanin was identified as cyanidin-3-glucside with minr amunts f delphinidin- and penidin-3-

ii glucsides. This cmpsitin resembles the skin cmpsitin f clured, Muscat a petits grains cultivars which it als resembles by the mnterpene cmpsitin f the juice.

5 ii glucsides. This cmpsitin resembles the skin cmpsitin f clured, Muscat a petits grains cultivars which it als resembles by the mnterpene cmpsitin f the juice. The pigments ccurred nly in berries with levels f ttal sluble slids in excess f 24"Brix in the juice and such berries tended t have smaller fresh weight. Berry pigmentatin ccurred n vines with varius rt systems. The specific cnditins under which pigment develped in Muscat Grd berries may ffer a useful tl in the study f anthcyanin bisynthesis. As grafting bunches f Muscat Grd nt Shiraz and als grafting f Shiraz nt Muscat Grd did nt effect the anthcyanin prfile f the berries, it appears that anthcyanins are anther grup f secndary metablites synthesised in the berries independent f the leaves. Phtsynthetically fixed assimilates translcate frm surce leaves t sinks via phlem tissue in vascular bundles. The phlem sap cmpsitin was analysed fr three grapevine cultivars. Phlem sap was cllected frm the cut end f grape bunch laterals using an EDTA-facilitated exudatin technique. Samples f phlem sap exudates were cllected at different berry develpmental stages and als cntinuusly ver a few days and nights at different perids f berry ripening. Sucrse, amin acids and mineral ins were analysed in these phlem exudates. The sucrse cncentratin in the phlem exudates cllected frm different cultivars was fund t be high while glucse and fructse were lw. In additin t sucrse, substantial levels f amin acids and ptassium were als detected in the exudates. Glutamine was the principal amin acid in the phlem exudates f grapevines fllwed by glutamic and aspartic acids, and alanine. On'rissin f EDTA frm the buffer slutin r girdling f canes either side f the bunch greatly decreased the levels f all these sugars and amin acids in the exudate. These cntrls, tgether with the bserved cmpsitin f the cllected material, suggested that the exudates were predminantl;- derived frm phlem sap entering the berries in the laterals. The metablites exuded frm the phlem f fruit bunch stem shwed seasnal and diurnal variatins. The diurnal pattern f sucrse exudatin shwed high levels f sucrse exuded at night. Ttal amin acids and sme individual amin acids als shwed diurnal variatin. This is the first time that the diurnal pattern f phlem sap flux in grapevines has been reprted. Using a similar EDTA-lacilitated cllectin technique, the phlem exudates f grapevine leaves were cllected frm the cut end f petiles f excised leaves. Sucrse, ptassium and amin acids were present in the samples cllected frm leaves. The results indicate that sucrse is the main sugar translcating in grapevine phlem, and that amin acids and ptassium are als majr cnstituents f the phlem sap.

$iii T investigate the site f bisynthesis f flavur cmpunds in grape berries, the glycsidically bund fractin in phlem sap cllected frm fruits and leaves f three cultivars was qualitatively and$

6 iii T investigate the site f bisynthesis f flavur cmpunds in grape berries, the glycsidically bund fractin in phlem sap cllected frm fruits and leaves f three cultivars was qualitatively and quantitatively analysed by GC-MS. Ttal glycside levels in the cllected slutin were quantified by determinatin f G-G. The aglycnes identified were mstly mnterpenes, C13-nrisprenids, and benzene derivatives. There were varietal differences in the GC-MS prfile f cmpunds exuded int the buffer slutin. Glycsylated secndary metablite cncentratins were nt decreased in exudates frm bunches after girdling f the canes, indicating that the cmpunds being analysed may nt have been derived frm phlem sap. Analysis f grape bunch peduncle tissue shwed that sme f mst abundant glycsylated cmpunds in the exudates, such as thse f geranil, were als present in the rachis tissue in high cncentratins. It seems that these cmpunds reside in the tissues surrunding the phlem tissue and, due t their slubility, they may easily leak ut and enter the buffer slutin used fr sap cllectin. The experimental fîndings in this thesis lead t the interpretatin that, wilh regard t secndary metablite prductin, grape berries are metablically independent f leaves and ther parts f the plant and all secndary metablic pathways are a part f the berry's genetic identity. Althugh dependent n the rest f the vine fr supply f precursr cmpunds and primary metablites, it appears that the berr), itself is the site f synthesis f secndary metablites including anthcyanins and flavur cmpunds.

IV I)eclaratin I hereby declare that this wrk cnducted in the Department f Hrticulture, Viticulture and Oenlgy, within the University f Adelaide and cntains n material which has been previusly

7 IV I)eclaratin I hereby declare that this wrk cnducted in the Department f Hrticulture, Viticulture and Oenlgy, within the University f Adelaide and cntains n material which has been previusly accepted fr the award f any ther degree r diplma at any University. T the best f my knwledge this thesis cntains n material previusly published r written by anther persn except where due references is made in the text. I cnfent t this thesis being made available fr lan and phtcpying, when depsited in the university library. Mansur Ghlami

V Acknwledgement My sincere appreciatin ges t my supervisrs, Dr Patrick Williams, Dr Simn Rbinsn, and Dr Jhn Jacksn fr their excellent advice, patience and encuragement thrughut this prject and

8 V Acknwledgement My sincere appreciatin ges t my supervisrs, Dr Patrick Williams, Dr Simn Rbinsn, and Dr Jhn Jacksn fr their excellent advice, patience and encuragement thrughut this prject and during the preparatin f the thesis. I wuld like t express my prfund gratitude t Dr Bryan Cmbe fr his cnstant guidance, cnstructive criticism and assistance during the investigatin and in preparatin f this manuscript. I wish t acknwledge gratefully the supprt f the University f Bu-Ali Sina, Hamadan, Islamic Republic f Iran f awarding a schlarship which enabled me t persue my studies at the University f Adelaide, Australia. Special thanks are due t Mr Yji Hayasaka fr GC-MS analysis. Dr Wies Cynkar, Ms Marila Kwiatkwski, and Mr Ken Pcck f Australian Wine Resea ch Institute fr their technical supprt fr the chemical analyses. I wuld like t thanii the Cperative Research Centre fr Viticulture fr their financial supprt all thse peple assciated within CRCV Prgram 3 fr their friendship. Finally, I wish t express my appreciatin t my wife and children fr their patience and encuragement thrughut my studies.

VI List f tables Table 2.1 Translcatin f phtsynthetic prducts frm leavesat immature benies (frm Peynaud and Ribéreau-Gayn, 197 I). Table 4.1. Cncentratins f mnterpene cmpunds identified in ripe grafted and nngrafted berries frm the glasshuse experiments.

9 VI List f tables Table 2.1 Translcatin f phtsynthetic prducts frm leavesat immature benies (frm Peynaud and Ribéreau-Gayn, 197 I). Table 4.1. Cncentratins f mnterpene cmpunds identified in ripe grafted and nngrafted berries frm the glasshuse experiments. Peak numbers are shwn in Fig Table 5.1 Ttal plyphenlics and anthcyanins in the skin f Muscat Grd Blanc in tw grups: white (nn-pigmented) and pigmented (shwing a light rse r pink clur). The fruits were frm vines grwn in the field. Table 5.2 Cncentratins f mnterpene cmpunds in whle nn-pigmented (White) and rse- r pink-clured (Clur) Muscat Grd berries and white juice. Cmpunds are numbered as in Figure 4.3 f chapter fur. Table 6.1 Sugars, amin acids and ptassium measured in phlem sap exudate f three grapevine cultivars during the grwing seasn (p ml / 1.5 ml f buffer slutin). Table 6.2 The seasnal changes in phlem exudate amin acids f Muscat Grd, Sultana and Shiraz. Values iven are Y f ttal amin acids befre veraisn (B), at veraisn (V). and at ripening time (R). Table 6.3 Cmparisn f sugars exuded frm fruit bunches n the vine with excised bunchstems as in Fig. 6.3, (cv. Muscat Grd at pre veraisn, "Brix : 4'5). Tabte 6.4 The effect f exclusin f EDTA frm the buffer slutin used fr sap cllectin n the levels f different metablites in cllected sap samples (cv. Muscat Grd at veraisn, "Brix :6.4). TTable 6.5 The effect f girdling the phlem tissue abve and belw the bunch n the levels f different metablites in cllected sap samples (cv. Muscat Grd at ripening. "Brix : 1.2). Table 6.6 The effect f girdling and exclusin f EDTA n the phlem sap exudatin (cv Muscat Grd at ripening, 'Brix = 15.8).

v'ii Table 6.7 The effect f exclusin f EDTA n the phlem sap exudatin (cv. Sultana at ripening 'Brix: 18.). Table 6.8 The effect f exclusin f EDTA n the phlem sap exudatin (cv.

1 Glycsyl glucse levels measured in phlem sap exuded int buffer slutin frm peduncle laterals and leaves, and als in berry hmgenates f Muscat Grd, Sultana and Shiraz at three berry develpment stages.

10 v'ii Table 6.7 The effect f exclusin f EDTA n the phlem sap exudatin (cv. Sultana at ripening 'Brix: 18.). Table 6.8 The effect f exclusin f EDTA n the phlem sap exudatin (cv. Sultana, ripe 'Brix = 23.8). Table 6.9 The levels f sucrse, glucse, fructse, ptassium and ttal amin acids f leaf phlem sap exudated under light and dark, cllected at ripening perid. Table 7.1 Glycsyl glucse levels measured in phlem sap exuded int buffer slutin frm peduncle laterals and leaves, and als in berry hmgenates f Muscat Grd, Sultana and Shiraz at three berry develpment stages. Table 7.2 The aglycnes determined in hydrlysates f phlem exudate cllectins and tissue extracts Appendix Table 1. The seasnal changes in phlem sap amin acids f Muscat Grd, Sultana and Shiraz. Values given are/ f ttal amin acids befre veraisn (B), at veraisn (V), and at ripening time (R). Appendix Table 2. The differences in amin acid prfiles f leaf exudates in light and dark cllected at berrv ripening time (% f ttal). Appendix Table 3. The seasnal changes in amin acid prfiles (Y f ttal) f Muscat Grd peduncle exudates. Appendix Table 4. The seasnal changes in amin acid prfiles (% f ttal) f Sultana peduncle exudates. Appendix Table 5. The seasnal changes in amin acid prhles (% f ttal) f Shiraz peduncle exudates.

viii List f figures Figure 2.1 Water and dry matter (1) and type f changes in the amunts f the sugils (2) in grape berries during berry develpment. (Frm cmbe, 1992). Figure 2.2 Cncept f synthesis and hydrlysis f secndary metablite glycsides in plants (frm Hsel, 1981)' Figure 2.

1 t 6 Evlutin f G-G in leaf blades ( I ), leaf petile ( 2 ) and fruit ( 3 ) and changes in berry vlume ( 4 ), berry density ( 5 ), and fruit sluble slids ( 6 ) in Muscar Grd with time in weeks after

11 viii List f figures Figure 2.1 Water and dry matter (1) and type f changes in the amunts f the sugils (2) in grape berries during berry develpment. (Frm cmbe, 1992). Figure 2.2 Cncept f synthesis and hydrlysis f secndary metablite glycsides in plants (frm Hsel, 1981)' Figure 2.3 Types f secndary metablite glyccnjugates that have been identified as flavur precursrs in fruits (frm Williams 1993). Figure 3.1 G-G assay steps. Figure t 6 Evlutin f G-G in leaf blades ( I ), leaf petile ( 2 ) and fruit ( 3 ) and changes in berry vlume ( 4 ), berry density ( 5 ), and fruit sluble slids ( 6 ) in Muscar Grd with time in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in macerated whle bunch n a fresh weight basis (left axis) and in mg per separated berry (right axis)' Vertical bars thrugh mean values represent SE. : inceptin f "Brix increase. Figure t 6 Evlutinf G-Ginleaf blades( I ), leaf petile(2)andfruit(3 ) and changes in berry vlume ( 4 ), berry density ( 5 ), and fruit sluble slids ( 6 ) in Sultana with time in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in whle bunches n a fresh weight basis (left axis) and in mg per separated berry (right axis). Vertical bars thrugh mean values represent SE. : inceptin f 'Brix increase Figure t 6 Evlutin f G-G in leaf blades ( I ), leaf petile (.2 ) and fruit ( 3 ) and changes in ben-v vlume ( 4 ), berry density ( 5 ), and fruit sluble slids ( 6 ) in Flame Seedless with time in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in whle bunches n a fresh weight basis (left axis) and in nig per separated berry (right axis). Vertical bars thrugh mean values represent SE. : indicatin f 'Brix increase. Figure t Glycsyl glucse (G-G) cncentratin in different tissues f all three cultivars cmparing G-G in leaf blades (Fig.1), in leaf petiles (Fig.2 ) and whle bunches (Fig.3). M, Muscat Grd; S, Sultana; F, Flame Seedless.

ix Figure 3.6 Glycsyl glucse cncentratin in berries f Muscat Grd, Sultana and Flame Seedless pltted against juice 'Brix (data frm 1992-93)' Figure 3.

12 ix Figure 3.6 Glycsyl glucse cncentratin in berries f Muscat Grd, Sultana and Flame Seedless pltted against juice 'Brix (data frm )' Figure 3.7 Glycsyl glucse cncentratin in berries f Muscat Grd and Sultana pltted against juice "Brix (data frm 1994'95). Figure 4.1 Transfer f grape bunches by apprach-grafting. Befre flwering, the dnr vine with bunch (A) was grafted nt the hst vine (C). The dnr vine was cut ff five weeks later (D) and the fruit was allwed t ripen n the hst vine (E). Ungrafted vines were grwn under the same cnditins as a cntrl (B). Figure 4.2 Glasshuse grwn Muscat Grd grapes bunch grafted nt Shiraz vine. Figure 4.3 Glasshuse grwn Shiraz grapes bunch grafted nt a Muscat Grd vine. Figure 4.4Typical GC-MS chrmatgrams f the aglycns frm Shiraz cntrl berries and frm fruit f Muscat Grd cntrl and Muscat Grd grafted nt Shiraz. The y- axis scale is 'relative in current' (RIC). Figure 4.5. Identity f 16 mnterpene cmpunds frm GC-MS chrmatgrams Figure 4.6 Histgrams f mnterpene aglycnes in berries (shwn in Fig. 4.2) frm the field experiment f Sultana ungrafted (A), Sultana grafted n Muscat Grd (B), Muscat Grd ungrafted (C) and Muscat Grd grafted n Sultana (D), at three ripening stages. Figure 4.7 Typical GC-MS chrmatgrams f the aglycns frm Sultana cntrl berries and frm fruit f Muscat Grd cntrl and Muscat Grd grafted nt Sultana. The y-axis scale is 'relative in current' (RIC). Figure 4.8 Typical GC-MS chrmatgrams f the aglycns frm Muscat grd cntrl berries and frm fruit f Sultana cntrl and Sultana grafted nt Muscat Grd. The y-axis scale is 'relative in current' (2C). Figure 5.1 Muscat Grd berries shwing pink clur Figure 5.2 HPLC chrmatgrams f anthcyanin pigments prfile f the extracts fr Muscat Grd berries skin (a) frm wn-rted vines, (b), frm bunches grafted nt Shiraz vines. chrmatgram (c) shws authentic cyanidin-3-glucside.

13 X Chrmatgram (b) shwed a faster elutin prfile due t different instrument cnditins Figure 5.3 Relatinship fr 221Muscat Grd berries f: (a) ttal sluble slids f juice ('Brix) and absrbance f skin extracts at 52 nm, (b) berry weight and absrbance f skin extracts at52 nm and (c) berry weight and "Brix. Figure 6.1 Phlem sap cllectin frm a peduncle lateral. Figure 6.2 Phlem sap cllectin frm detached leaves. Figure 6.3 Exudate cllectin frm peduncle tissue ff the vine. Ped: peduncle; BL: bunch lateral; ET: Eppendrf tube; HB: hlding blck. Figure 6.4 Girdling the cane abve and belw the fruit bunch in rder t stp phlem sap flux t the bunch. G: girdled area; Ped. L: peduncle lateral frm which sap was cllected. Figure 6.5 The diurnal pattern f sucrse (A), ttal amin acids (B) and ptassium (C) levels in phlem sap exudates cllected frm pre-veraisn grape bunches during the day (vertical bars indicate SE). Figure 6.6 (A-F) The diurnal patterns and the levels f majr amin acids detected in the grapevine phlem sap þre-veraisn). Figure 6.7 The diurnal pattern f cntinuus sucrse exudatin frm pst-veraisn Muscat Grd bunch laterals during light and dark perids with different starting times (each value represents mean f l bunches), "Brix:1.2. A, B, and C represent three samples where the bunch laterals were remved at different times. Fr girdled sample (D) the canes were girdled the day befre sampling. The values are fr tw hurs f each sampling perid, Figure 6.8 The diurnal pattern f cntinuus sucrse exudatin frm ripe Muscat Grd bunch laterals during light and dark perids with different starting times (each value represents mean f l bunches), "Brix : Figure 6.9 The diurnal pattern f ttal amin acids measured in ripe Muscat Grd grapevine exudates during days and nights (samplings started at different times

14 xt and each value represents mean f l bunches), (first and secnd series f samples presented in Fig. 6.8). Figure 6.1 The diurnal pattern f ptassium measured in ripe Muscat Grd grapevine exudates during days and nights (samplings started at different times and each value represents mean f l bunches)' Figure 6.11 (A-F) The diurnal pattern f individual amin acids measured in ripe Muscat Grd grapevine phlem sap exudates during days and nights (samplings started at different times and each value represents mean f 1 bunches). Figure 7.f GC-MS chrmatgrams f the aglycnes islated frm phlem exudates cllected frm ripe fruit bunch laterals f Muscat Grd (A), Sultana (B), and Shiraz (C) grapes. Figure 7.2The cncentratin f secndary metablite aglycnes shwn as bars (right axis) in the phlem exudate samples cllected ver a 36 h perid. The sucrse cncentratin (left axis) in the same samples als shwn.

x'ii List f Publicatins Published; Ghlami, M., Hayasaka Y., Cmbe, 8.G., Jacksn, J.F., Rbinsn, S.P. and Williams, P.J' (1995) Bisynthesis f flavur cmpunds in Muscat Grd Blanc grape berries.

15 x'ii List f Publicatins Published; Ghlami, M., Hayasaka Y., Cmbe, 8.G., Jacksn, J.F., Rbinsn, S.P. and Williams, P.J' (1995) Bisynthesis f flavur cmpunds in Muscat Grd Blanc grape berries. Australian Jurnal f Grape and Wine Research, I(L): Ghlami, M., and B.G. Cmbe, 1995, Occurrence f anthcyanin pigments in berries f the white cultivar Muscat grd Blanc (Vitis vinifera L.). Australian Jurnal f Grape and Wine Research l(2):67-7. Manuscripts in preparatin; N. I Title: Distributin f glycsides in the grapevine tissues during berry grwth and develpment (Chapter 3). Authrs: M.Ghlami, B. Cmbe, and P.J. V/illiams N.2 Title: Grapevine phlem sap analysis; I. Sucrse, amin acids, ptassium cncentratins, seasnal and diurnal patterns (Chapter 6). Authrs: M. Ghlami. S. Rbinsn, and B. Cmbe N.3 Title: Grapevine phlem sap analysis; II. Secndary metablite flavur cmpunds (Chapter 7). Authrs: M.Ghlami. P.J. Williams, Y. Hayasaka, B. Cmbe, and S. Rbinsn Frm this wrk was presented; In Australian Plant Physilgy cnference (APP) 1994, ral presentatin and abstract was published in the prceeding (abstract n. 162). Psters in; CRCV annual meeting, CRCV annual meeting, 1994 (tw psters). Wine industry technical cnference, 1995.

16 xii i Abbreviatins G-G HEPES EDTA TSS P-prtein ASP GLU ASN SER GLN HIS GLY THR ALA ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU LYS PRO Glycsyl Glucse N-[2-Hydrxy Ethyl] Piperazine-N' [2-Ethan Sulfnic acid] EthyleneDiamineteTraacetic Acid) Ttal Slubl Slids. Phlem prtein Aspartic acid Glutamic acid Asparagine Serine Glutamine Histidine Glycine Threnine Alanine Arginine Tyrsine Cysteine Valine Methinine Tryptphan Phenylalanine Isleucine Leucine Lysine Prline

17 This thesis dedicated t : My parents wh supprted me thrughut my studies and passed away while I was cmpleting my Ph D and befre I culd see them again. My wife and my children fr their patience and encuragement during this wrk.

18 Chapter 1.:rl'fË t-,,,itiixiili i ir.i:,ê r :" 1 i ii;it,-iì;lili ',1ì:;lüí:i,,^,,, ' Intrductin and aims 1.1 Intrductin The imprtance f flavur cmpunds in grapes, grapejuice and derived beverages has attracted much research interest. This has led t the identificatin and classificatin f many different cmpunds imprtant t the flavur f grapes (Strauss et al., 1986; V/illiams et al., 1993; Nble, 1994). Terpenids, particularly mnterpenes, have been extensively studied bth fr their evlutin during berry grwth and develpment, and as cnstituents f different cultivars (Wilsn et al., 1984; Gunata et al., 1985b; Park et al., 1991). It is knw appreciated that the majrity f flavur cmpunds ccur in grape berries as flavurless glycsides which culd readily release the flavur cmpunds by hydrlysis. Glycsidically cnjugated mnterpenes were the earliest recgnised as flavur precursr in grape berries. Sme research has shwn that leaves have an imprtant rle in the develpment and the cmpsitin f fruits, and decreasing the leaf area t lw levels culd result in a decrease in yield and quality (Kliewer, 197a; Sidahmed and Kliewer, 198; Bledse et al., 1988; Hunter and Visser, 1989; Candlfi-Vascncels and Kblet, 199; Hunter and Visser, 199 ). Viticultural practices and envirnmental factrs have als been shwn t have an influence n grape cmpsitin and flavur Qllble, 1994). Few data are available cncerning the presence f glycsides f secndary metablites in leaves r ther parts f the grapevine. Gunata et al., (1986) reprted the presence f mnterpene glycsides in grapevine leaves and suggested the pssible translcatin f these cmpunds frm leaves t berries. There is n infrmatin in the literature abut the translcatin f glycsides synthesised in leaves t the berries. In the absence f reliable evidence abut the mbility f glycsylated secndary metablites in general, and mnterpene glycsides in particular, the mechanism f Their translcatin remains an assumptin. 1.2 Aims The general aim f this wrk was t investigate the relatinship between leaves and berries f grapevines in terms f the translcatin f secndary metablite, and in particular mnterpene flavur cmpunds. The bjectives f this prject are as fllws:

2 l. T investigate the levels f ttal glycsides in differentparts f the grapevine and their cncentratin changes at va ius grwth and physilgical stages. 2.

19 2 l. T investigate the levels f ttal glycsides in differentparts f the grapevine and their cncentratin changes at va ius grwth and physilgical stages. 2. T determine the cntributin f leaves, as a site f metablite synthesis and exprt, t the berry cmpsitin. 3. T evaluate the rle and imprtance f the berry itself in terms f utilisatin and metablism f the imprted metablites. 4. T cllect and analyse phlem sap translcating in the grapevine phlem tissue frm leaves t the berries fr primary metablites (carbhydrates and amin acids)' 5. T analyse grapevine phlem sap, exuded frm the leaf petile and peduncle lateral, fr secndary metablites thrughut the grwing seasn.

3 Chapter 2 General literature review Cntents Page 2.1 Berry cmpsitin... 2.1.1 Primary metablites... 2.l.l.l CarbhYdrates... 2.1.1.2 Organic acids.-... 2.1.1.3 Amin acids... 2.1.1.4 Minerals 2. 1.

20 3 Chapter 2 General literature review Cntents Page 2.1 Berry cmpsitin Primary metablites... 2.l.l.l CarbhYdrates Organic acids Amin acids Minerals Secndary metablites and flavur cmpunds Mnterpenes Nrisprenids Sesquiterpenids Shikimic acid derivatives Anthcyanins Secndary metablite glycsides Glyc sy I atin f secndary metablites Glycsides in fruits Glycsides as precursrs f vlatile flavur cmpunds 2.2 Assimilate transprt in grapevine (Vitis vinifera L.) Analmical aspects Teclniques f phlem sap cllectin Leaves as surces f assimilates Relranslcatin frm reserves Transprt f carbhydrates Sucrse Hexses T ransprt f nitrgen cmpunds Transprt f rganic acids Transprt f minerals Transprt f secndary metablites t 11 t ls l5 l6 t7 l l 2l 22 23

4 2.1B,erry cmpsitin The grape berry cntains a mixture f primary and secndary metablites' Primary metablites are essential fr plant grwth and develpment and include lipids, carbhydrates, amin acids,

21 4 2.1B,erry cmpsitin The grape berry cntains a mixture f primary and secndary metablites' Primary metablites are essential fr plant grwth and develpment and include lipids, carbhydrates, amin acids, prteins and nucletides (Mann, 1987; Taiz and Zeiger, 1991). Plant secndary metablites are a large and diverse array f rganic cmpunds that d nt appear t have a direct functin in grwth and develpment. Unlike primary metablites, these cmpunds may have a restricted distributin in the plant kingdm; smetimes a particular secndary plant prduct is typical fr nly ne plant species r a taxnmically-related grup f species (Taiz and Zeiget,l99l). 2.1.L Primary metablites 2.l.l.l Carbhydrates Grape juice, which is largely the vacular cntent f the pericarp cells f ripened grape berries, cntains large amunts f tw hexses, glucse and fructse, and smewhat lesser amunts f tartaric acid. malic acid. plus a wide variety f ther cmpunds. Ptassium is the dminant mineral in the juice. The majr cnstituent f berries is water fllwed by, in the case f ripe berries. D(+)-glucse and D(-)-fructse which tgether amunt t abut grams per litre fjuice. These tw carbhydrates cnstitute a large part f berry dry matter (Peynaud and Ribéreau-Gayn, 1971, Cmbe, 1992). Berry sugar level increases at the inceptin f ripening (veraisn) alng with sftening f the berry, decline in acidity, change in skin clur, and increase in berry vlume (Cmbe, 1989, 1992). Glucse cnstitutes 85/ f the grape berry sugar in the hrst grwth perid, but after veraisn the rati f glucse t fructse gets appraches t unity, while in very ripe grapes there is generally a slight excess f fructse (Kliewer, 1967a; Hanis et al., 1968; Cmbe, 1975) (Fig. 2.1). It is knwn that sucrse, which is an end prduct f phtsynthesis, is translcated t the berries (Swansn and El-Shishini, 1958; Glad et al., 1992a), but the cncentratins f glucse and fructse are apprximately 3-fld that f sucrse in ripe berries (Hrazdina et al., 1984; Cmbe, 1987b). It seems that sucrse is metablised fllwing unlading int the berry, first by hydrlysis t its hexse mieties and then utilised in different metablic pathways (Hardy, 1967,1968). In immature

22 5 berries, carbhydrates are transfrmed int malic acid but this reactin decreases as the berry ripens (Peynaud and Ribéreau-Gayn, L97l). After veraisn, malate decline tends t be initially rapid while tartaric acid per berry tends t remain cnstant. In green grapes glucse can be cnverted t malate, just as the reverse may ccur in ripe grapes by glucnegenesis thrugh PEP-carbxykinase (Ruffner and Kliewer,1975; Ruffner et al., tets). r!t l lf :.3 wltcil,1. ll al ::! l?i 1 2 xv xy Ok Fig. 2.1 Water and dry matter (l) and type f changes in the amunts f the sugars (2) in grape berries.during berry develpment. (Frm Cmbe, 1992) There are sme ther sugars present in grape juice but in very lw cncent atin cmpared t glucse and fructse. At the inceptin f ripening intense physilgical activities ccur and a cnsiderable accumulatin f sugar takes place, fr example, the sugar cntent f a berry increases abut seven fld within a week r s. This increase is difficult t explain in terms f increase in phtsynthesis rates. Brwn (1981) suggested that the nset f rapid sugar accumulatin in the grape be.ry results frm changes in phlem unlading rather than frm cncentrative accumulatin by the cell. This authr als fund that the skin f a pre-veraisn berry is a glucse accumulatr, but pst-veraisn it becmes a fructse and glucse accumulatr. Cmbe and Matile (198) shwed that cncentratins f fructse, bth diffr, sible and cmpartmented, were duble that f glucse in green skin but there was n difference in the cncent atins f the hexses in ripening berries.

23 Organic acids The main rganic acids in the grape berries are L(+)-tartaric acid and L(-)-malic acid; tgether these acids accunt fr 9Y f ttal berr)' acidity with citric' ascrbic and phsphric acids cntribuuting t the ther 1% (Winkler et al., 1974). In plant tissues generally, carbhydrates are brken dwn thrugh the pentse phsphate cycle r thrugh glyclysis which transfrms the glucse mlecule int pyruvic acid. Pyruvic acid can give rise t malic acid thrugh carbxylatin r the peratin f the tricarbxylic acid (TCA) cycle. Intrductin f rac-acetic acid int the grape berries resulted in the frmatin f highly labelled glutamic acid frmed by aminatin f cr-xglutarate (Peynaud and Ribéreau-Gayn, lgtl). There is n direct relatin between the peratin f the cycle as a whle and the accumulatin f malic acid (Peynaud and Ribéreau-Gayn, l97l). If it is assumed that sucrse is the sle sugar transprted int the berry, then hydrlysis f sucrse int glucse and fructse under the effect f invertase culd be the key reactin t supply carbn fr all metablic reactins. The cnversin f glucse t malate via phsphglycerate and phsphenlpyruvate can be demnstrated by supplying immature berries with unifrmly labelled glucse thus emphasising the clse intercnnectins between the metablism f malic acid and f glucse (Peynaud and Ribéreau-Gayn, 1971). In immature berries, carbhydrates are transfrmed int malic acid but the intensity f this reactin decreases as the berry ripens and in ripe grapes, the reverse prcess (glucnegenesis) may take place (Peynaud and Ribéreau-Gayn, l97l)' Hwever, the activity f glucnegenesis enzymes in cnverting malic acid t sugar in ripening berry culd accunt fr nly 5/ f sugff accumulatin if there is n fresh malic acid prductin after veraisn (Ruffner and Hawker, 1977). Accumulatin f majr rganic acids in the berry ccurs during the first stage f berr), grwth and ttal acid cntent f the berries increases significantly until veraisn time. Tartrate accumulates in grape berries befre veraisn and during ripening tartrate cntent per berr1, remains cnstant (Hale, 1977; Carrll and Marcy, 1982). The metablism f tanaric acid differs frm that f malic acid and, nce frmed, it remains relatively stable (Hrazdina et al ). It is als knwnthat immature berries are able t synthesise their wn tartaric acid during the early stages f develpment. At this perid, tafaric acid synthesis is as active as malic acid synthesis but it decreases rapidly until at veraisn it practicall),ceases (Hrazdina et al., 1984). Accrding t Hardy (1968), the high percentages f rc fund in rganic acids nly a few hurs after supplying immature grapes with labelled sugars indicated that mst f the malate and the tartrate in the berries riginates frm glucse and fructse (mst prbably supplied by hydrlysis f the sucrse translcated via the phlem). The malate cntent f the berry declines sharply during ripening. The free acids tend t cnvert t salts (mn and di-basic) during ripening (Iland and Cmbe. 1988). The decline in the acid cntent f berries culd be als due t

24 7 decreased synthesis and increased metablism. Malate is a majr tricarbxylic cycle acid which is metablised under the effect f malate dehydrgenase enzyme and cnverted t carbhydrates r recycled in the tricabxylic acid cycle. It can als be cnverted t acetyl C-A and participate in the metablism f fatty acids r enter int transfrmatins leading t the frmatin f secndary metablites (Nii and Cmbe, 1983; Pssner et al., 1983; Hrazdtnaet al., 1984) Amin acids Ttal nitrgen cncentratin in mature grape berries ranges frm 1-2 mg/ 1 ml, but nitrate and prtein nly accunt fr a small prprtin f the ttal. The prtein cntent f grape juice is reprted t be 5-1 mg/l fjuice (Smers andziemelis, 1973; Murphey et al., 1989; Puey et al., 1993). Twards the end f fruit ripening large amunts f free amin acids accumulate in the berries and at maturity make up 5% t 9/ f the ttal N, with arginine and prline being the mst prevalent amin acids n a mlar basis (Kliewer, 1969, I97b; Wermelinger, 1991). The amin acid cmpsitin f ripe grape berries f numerus cultivars f Vitis spp have been studied in past years and is reprted t be: glutamic acid I mm; arginine frm.3 - I 1.2 mm; prline mm; serine mm; threnine mm; alanine I ;1-aminbutyric acid m}y'r; aspartic acid mm (Rubelakis-Angelakis,l99l; Kliewer, 1969, 197b) and the ttal free amin acid cncentratin ranges frm 1.4 t 64.5 mm. Arginine accunted fr 5YrT 43% f ttal nitrgen in grape juice. Prline levels in grapes ften increase mre than arginine late in the seasn (Kliewer, 1969, 197b; Bath et al., 1991 ). In summary, althugh the cncentratins f free amin acids in ripe grape berries may depend n several factrs. including cultivar, envirnmental cnditins, sil characteristics, and cultural practices. it seems that in l/itis vinifera cultivars, prline and arginine, and t a lesser extent glutamic acid and alanine, are the predminant amin acids (Kliewer, 1967b, l e6e) Minerals Cnstant uptake f anins and catins and distributin f minerals t the varius part f the plants including the berries ccurs all thrugh the grwth perid. The berries are relatively lw in mineral elements cmpared with ther parts f the plant. During the secnd grwth perid there is an increase per berry in catins such as K*, Na+, Ca**, Mg**, Cr**, Mn** and in anins such as phsphate (POa-l (Hrazdina et al., 1984; Bselli et al., 1995). Phsphric acid (which increases all thrugh the berry grwth

8 Bselli et al., 1995). Phsphric acid (which increases all thrugh the berry grwth perid) is the third majr acid present in grapes. The catin cntent per berry increases 2-3 times in the skin and 1.2-l.

25 8 Bselli et al., 1995). Phsphric acid (which increases all thrugh the berry grwth perid) is the third majr acid present in grapes. The catin cntent per berry increases 2-3 times in the skin and 1.2-l.g times in the pulp during berr),ripening (Hale, I977;lland and Cmbe, 1988). On the basis f berry fresh weight' ptassium remains almst cnstant during the beny develpment perid whereas levels f ther catins such as calcium, manganese, cpper and magnesium decrease during berry develpment which is prbably due t berry expansin (Hrazdina et al., 1984; Pssner and Kliewer' 1985; Bselli et al., 1995). Ptassium is the mst abundant catin in the grape berry accunting fr 45V-7% f themineral catin cntent f the berry (Smers, 1975; Hale, 1977). During the same perid heavy metals increase by 5%. There is an increase in anins, particularly phsphate, which is mstly cncentrated in seeds. The phsphate cntent f skin and pulp als increases steadily as the fruit ripens (Peyanud and Ribéreau-Gayn, r97r) Secndary metablites and flavur cmpunds The chemical structure f thusands f secndary metablites are knwn (Lukner 199), and every mnth there are reprts in phytchemical jurnals f mre nvel structures. Plant secndary prducts can be divided int three grups accrding t their mde f bisynthesis: terpenes, phenlic and nitrgen cntaining cmpunds' Terpenes are synthesised frm acetyl CA via the mevalnic acid pathway. Phenlic cmpunds are armatic substances frmed via the shikimic acid and/r the plyketide pathways. Nitrgen cntaining secndary prducts, such as alkalids, are bisynthesised primarily frm amin acids. One f the defining characteristics f secndary metablites is their prpensity t accumulate in particular rgans r tissues f the plant (Haslam, 1975). Many vlatile cmpunds f different metablic rigins have been identified in grapes and related prducts (Rapp, 1988b). Vlatile flavur cmpunds f grapes can be classified int three grups accrding t their metablic rigin; l) mevalnic acid derivatives, such as terpenids, 2) shikimic acid pathway derivatives which includes vlatile phenls and 3) fatty acid derived cmpunds (Williams et al., 1989). Terpenids are a large. diverse and widespread grup f secndary metablites with structures made up f C5 isprenid units. In grapes, mnterpenes (C I ), nrisprenids (C7.8. tr. r ), shikimate-derived cmpunds, sesquiterpenes (C 5) and cartenids (Cae) have been identified (Schreier et al., 1976; Strauss et al., 1986; Versini et al.,1988; Razungles et al ; Williams et al., 1989; Seftn et al., 1989, 7993,1994; Banthrpe, 1991 Williams et al., 1992; Stahl-Biskup et al., 1993). Cmpunds respnsible fr grape varietal flavur can be classified int tw grups ie, l) free vlatiles, which are respnsble fr the perceived flavur f the fruit, and 2) mre

I abundant glycsidically-bund, flavurless cmpunds. These nn-vlatile cmpunds are frmed by cupling f free vlatiles t a glucse, r t a disaccharide, the latter cmprising a glucse and ne ther sugar unit.

26 I abundant glycsidically-bund, flavurless cmpunds. These nn-vlatile cmpunds are frmed by cupling f free vlatiles t a glucse, r t a disaccharide, the latter cmprising a glucse and ne ther sugar unit. Imprtantly, these glycsides can cntribute significantly t arma upn hydrlysis (Nble et al., 1988; Williams et al., 1989; Shseyv et al., 199; Fng et al.,l99l;nble, 1994) and thus represent a ptential surce f flavur cmpunds Mnterpenes The frmatin f significant quantities f mnterpenes (>.I / fresh weight ) appears t be cnfined t sme 5 families f higher plants and mre than 1 mnterpenes are presently knwn as natural prducts f higher plants (Charlwd and Banthrpe,l9ST; Crteau, 1987; 'West, 199; Charlwd and Charlwd, 1991). Mnterpenes have been identified as imprtant cntributrs t the arma f the essential ils f many plant species and they ften ccur as glycsidically bund cmpnents f plant tissues. The cmpsitin and flavur rle f the mnterpene glycsides f plant-derived fds have been the subject f many studies in the last few years (Williams, 1993). The mnterpene flavur cmpunds are an extensively studied grup f secndary metablites in grapes (Ribéreau-Gayn et al.,1975; Williams et al., 198; Williams et al., 1983; Maris, 1983; Rapp et al., 1983; Wilsn et al., 1984; Strauss et al., 1986, 1988, Rapp, 1988b). Mnterpene cmpunds, alne amng the many and varied cnstituents f grapes, shw a relatin with varietal flavur f Vitis vinifera (Rapp 1988a). In muscat grapes they were fund in bth vlatile and nn-vlatile frms (Strauss et al., 1986; Park and Nble, 1993). Crdnnier and Baynve (1974) first suggested the existence f mnterpene glycsides as precursrs f free mnterpenes in muscat grapes. Subsequent wrk shwed that mnterpene vlatiles in grapes are generated by hydrlysis f glycsidic cmpunds and f plyhydrxylated mnterpenes (ie plyls). The latter plyls may be glycsylated r free (Strauss et al., 1986). Mnterpene cmpunds accumulate in the grape berries during berry ripening in bth free and glycsylated frms (Wilsn et al ; Williams et al., 1985; Park et al., l99l) Nrisprenids Nrisprenids with l3 carbn atms (C13) are anther imprtant class f flavur and arma cmpnents(winterhalter and Schreier, 1994; Ohlfl 1978). Cartenids are assumed t be the precursr f nrisprenids in grapes (Schreier et al., 1976: Razungles et al., 1988; Seftn et al ; Kanasawud and Cruzet, 199; Williams et al., 1991)

27 1 ccurrence f nrisprenids and cartenids in grapes (Williams et al., 1982c). Hwever, little is knwn abut the precursrs f the nrisprenid cmpunds and the reactins by which they are frmed (Winterhalter and Schreir, 1938). Nrisprenids cntribute t the flavur and arma f flral grape varieties alng with mnterpenes. In additin t free frms they are als present in nn-vlatile glycsylated frms (V/illiams et al., I982a;Williams et al., 1989; Winterhalter et al., 199b) and in sme flral varieties such as Riesling levels f nrisprenid glycsides are higher than mnterpene glycsides (Williams et al., 1987) Sesquiterpenids Sesquiterpenes are the largest class f terpenids, with mre than 1 sesquiterpenid structures knwn and several thusand cmpunds f the class islated and identified (Fraga, l99l). Sesquiterpenes are C 5 terpenids and have been identified as imprtant cmpnents f arma in plants (Ramaswami et al., 1986). Frm the sesquiterpenids fund in nature (Ramaswami et al., 1986), several have been identified as cnstituent f grapesandgrapejuice(schreieret a1.,1976;maris, 19S3). Ohlff (1978b)inareviewn naturally ccurring arma cmpunds has given many ther examples f sesquiterpenids with fl avur prperties Shikimic acid derivatives The shikimic acid pathway gives rise t large number f armatic cmpunds including many plyphenls. Since vlatile phenls were nt generally detected in grape juices, the rigin f shikimate-derived phenls in wines was suggested t be berry skin, seeds and cluster stem tissues during alchl fermentatin (Dzhakhua et al., 1978). The shikimatederived vlatile phenls were later identified in the hydrlysate f Cl8 reversed-phase islates f grape juices (Strauss et al., 1987; Williams et al., 1989; Winterhalter et al., l e9). 2.L.2.5 Anthcyanins Anthcyanins as a flavnid subclass are secndary metablites widespread amng plants (Hahlbrck and Scheele, 1989). Grape berries f red varieties cntain relatively large amunts f anthcyanins and ther plyphenls which cntribute greatly t their appearance, and quality. Wulf and Nagel (1978) have separated and quantified abut 2 different anthcyanin cmpunds frm Vitis vinifera Z. berries. The distributin f anthcyanins in the fruit is cmplex and varies accrding t variety.

28 11 The anthcyanins are fundamentally respnsible fr all clur differences between grapes (Ribereau-Gayn, lg82). The level f anthcyanins in the berry skin is ne f the parameters being used fr quality assessment in black grapes. The accumulatin f anthcyanin begins at veraisn, an event signalling the end f the develpment perid f the berries and the nset f the ripening prcess which is cincident with r shrtly after the increased accumulatin f sugars (Cmbe, 1973; Pirie and Mullins, Ig77, lg8; Hrazdina et al., 1984; Crippen and Mrrisn, 1986; Rgger et al', 1986; Darné, 1993) Secndary metablite glycsides Many secndary prducts ccur in plants in the free frm (i.e. as aglycnes) t nly a limited extent r nt at all. Usually they are cnjugated with a variety f mnmeric r ligmeric saccharide mieties. Aliphatic and phenlic hydrxyl grups, carbxyl functins and amin and mercapt grups f aglycnes are invlved in cnjugatin reactins. Glycsides f terpenids, phenls r ther alchls are widely distributed in nature. Flavnids and anthcyanins have their hydrxyl grups linked t sugars. The N- glycsides are f utmst imprtance as they are structural units in cenzymes, nucleic acids, nucletides. etc.(kurt and Trssell, 1983). Frm a physilgical pint f view, cnjugatin reactins are f imprtance fr a number f reasns (Barz and Kster, 1981): a ) they drasrically alter the chemical (i.e. slubility) r physilgical (i.e. transprt thrugh cells r membranes and bilgical activity) prperties f secndary metablites; b ) they may result in a site f accumulatin different frm that ccupied by the aglycne thus resulting in the frmatin f several interrelated metablic pls (Ackerman et al., 1989); c ) the cnjugated cmpund may enter a different metablic pathway t the crrespnding aglycne; and d) they can determine whether a cmpund is a metablically active species r a metablically inactive (end) prduct. Therefre. cnjugatin reactins are cnsidered an imprtant way t transprt and stre materials r t detxiff unwanted materials (Barz and Kster, 1981;Hsel. l98l;crteau. 1984; Crteau, 1987). Glycsylatin has been bserved t effectively redistribute mnterpenes thrughut skin and juice fractins f grape berries (Wilsn et al : Park et al., l99l). C 3 nrisprenids and shikimiatederived cmpunds als have been islated in glyccnjugated frms frm grape juice (Seftn et al., 1989; Williams et al., 1989).

29 Glycsylatin f secndary metablites Transglycsylatin frm a glycsylating agent t the secndary metablite aglycne is always catalyzed by transferases utilizing a nucletide sugar (NDP-sugar) as a cfactr, usually uridine diphsphate-d-glucse (UDPG) (Luckner, 1972; Hsel, l98l; Gdwin and Mercer, 1983). UDPG is a transfer agent which leads t the frmatin f a ß- glucside. Althugh the mechanism f glycsylatin is understd, the trigger fr this cnjugatin reactin in plants is nt knwn. Frmatin and breakdwn f glycsides are a cnsequence f the apprpriate enzymes present in plants and can be represented in terms f enzymatic substrate specificities (Hsel, 1981). The glycsylatin steps are usually at the end f the particular bisynthetic pathways (Hsel, 1981)' Nucletide suga Transferase R-OH R-O-sugar Glycsidase Sugar Fig Cncept f synthesis and hydrlysis f secndary metablite glycsides in plants (frm Hsel. l98l ).

13 There is evidence that glycsylatin may be enhanced at sme particular physilgical stages in plants, e.g. after flwering in peppermint leaves (Crteau and Martinkus, 1979; Crteau, 1984) and after veraisn (berrl'sftening) in grape berries (Wilsn et al.

30 13 There is evidence that glycsylatin may be enhanced at sme particular physilgical stages in plants, e.g. after flwering in peppermint leaves (Crteau and Martinkus, 1979; Crteau, 1984) and after veraisn (berrl'sftening) in grape berries (Wilsn et al., 1984). Recent evidence suggests that glycsylatin takes place after extra hydrxyl grups ìre intrduced int a mnterpene skeletn in the frmatin f plyls (Strauss et al., 1988). Cntrary t these findings, Ackerman et al. (1989) demnstrated that fr tw rse species, Rsa damascena and Rsa gallica, the cncentratin f glycsidically bund cmpunds was greatest at the early stage f flwering, and decreased as flwers develped; at the same time they bserved an increase in the cncentratin f free vlatiles. The mechanism causing these differences in the time and pattern f glycside accumulatin fr these plant species is nt knwn. Study f the enzymes invlved in the cnjugatin reactins wuld help understanding f hw glycsides accumulate. It wuld als prduce a pssible means by which glycsylatin and thus ther related pathways may be cntrlled Glycsides in fruits The changes in the chemical prperties f glycsylated secndary prducts make them different frm free aglycnes in water slubility and chemical reactivity. This may explain why glycsylated cmpunds, rather than the free aglycnes, are accumulated in the plant vacule and are less reactive twards ther cellular cmpnents than are the aglycnes, Glycsylated cmpunds are therefre ften thught f as excretin prducts r as physilgically inactive plant strage frms. Hydrlysis f secndary plant glycsides, wuld release the physilgically active aglycnes (Hsel, l98l). The cmpsitin f the mnterpene glycsides f plants and their pssible bichemical functin have been recently reviewed (Salles et al., 1988; Williams, 1993; Stahl-Biskup et al., 1993) Glycsides as precursrs f vlatile flavur cmpunds Many glycsylated secndary metablites are knwn t be flavu less (Nble et al., 1988; Fng et al., 1991). Nevertheless, their ability t indirectly influence grape juice flavur, ie their precursr prperties (see abve), justifies the interest in their analysis, structural elucidatin and understanding f their sensry aspects. Apart frm this, structural elucidatin f glycsidic secndary metablites may prve useful t plant physilgists and thers. The bichemical and physilgical prperties f many secndary prducts cannt be understd unless their cnjugatin mieties are knwn. Analysis f plant extracts fr aglycne alne will nt reveal the full cmplexity and

31 14 pssible cnjugate-caused cmpartmentatin f secndary metablism (Barz and Kster, re81). Hwever, n infrmatin abut precursrs f flavur cmpunds in fruits was btained until acid and enzyme hydrlysis studies n Muscat grapes indicated that glycsidic derivatives f mnterpenes may have been invlved (Crdnnier and Baynve, 1974). Since then a grwing number f reprts have been published cncerning flavu precursrs in a range f fresh and prcessed fruits, as well as in related beverages, and in sme ther cnsumed plant and leaf prducts. Mst f the precursr cmpunds that have been islated have been fund t be glycsidic derivatives (V/illiams, 1993 ). The cmmn structural element f many glyccnjugates that have been identified in grapes and many ther plant prducts is a glucpyransyl unit attached thrugh ß- glucsidic linkage, t an aglycne which, in turn, can be ne f several well-knwn secndary metablites (Fig. 1.3). Bth glycsides and plyls must be cnsidered as precursrs f flavur which culd be released by hydrlysis f these cmpunds. Thus, enrymatic and chemical hydrlysis f glycsides and acid hydrlysis f plyls is very imprtant fr recvering the nnvlatile flavur cnstituents f fiuits (Williams, 1993). R HO- i OH -L-arabinñ:ransyl R AlÞbatic rcsiduæ c1, c5, c6, cs Mnærpencs Scsquitcrpcnes -L - riramnp yra. s y I Nrisprcnids C, C 1 OH Shikinic acid ctabliæs þd-xylpyransyl OH OH HOOrrfl -l - þapituransyl i Fig. 1.3 Types f secndary metablite glyccnjugates that have been identified as flavur precursrs in fruits (frm Williams, 1993)

15 2.2 Assimilate transprt in grapevin e (Vitis viniferø L.) Transprt in higher plants is a brad tpic and has been the subject f many studies.

32 Assimilate transprt in grapevin e (Vitis viniferø L.) Transprt in higher plants is a brad tpic and has been the subject f many studies. In vascular plants lng distance transprt f water ccurs thrugh the transpiratin stream via primary and secndary xylem. Inrganic in transprt ccurs principally by the same rute. Lng distance transprt f assimilates ccurs predminantly thrugh the phlem' The phlem tissue transprts carbhydrates, and ther substances, t meristematic grwing pints, develping fruits, strage rgans, and ther sites f carbhydrate utilisatin. It is cnsidered by sme researchers that translcatin takes place n a surce t sink pathway, that the lading f the nutrients int the phlem cntrls the rate f mvement, and that the directin f the mvement is cntrlled by unlading at the sink and cnsequence effects n cncentratin gradients. Carbhydrates are knwn as the majr substances translcating in phlem f plants (Kursanv, 1984). During the final grwth phase f the grape berry water and slutes, principally the hexses glucse and fructse, accumulate in large amunts in pericarp tissue- Frm the beginning f the ripening phase (veraisn) until 4 t 8 weeks later when it slws r ceases, the ttal sluble slids may increase frm 5Yt mre than 2%, smetimes nearly 3/; fructse may increase frm 5-1 t mre than 16 mg g-t (Cmbe and Philips, 1982l. Cmbe, l9s7). Phlem transprt in the sieve tubes ensures the transfer f C and N assimilates frm leaves t the different sink tissues, i.e. fruits, grwing shts, and perennial wdy parts (Pate, 198). Infrmatin abut grapevine phlem sap cmpsitin is limited t that published by Swansn and El-Shishiny (1958) and Glad et al, (1992a) which shw that sucrse, which is the principal sugar. is present alng with amin acids in the sap. A precise knwledge f phlem sap cmpsitin is imprtant fr the understanding f the mechanisms f assimilate transprt. Analysis f phlem exudate culd prvide a better picture f the substances present in the mature sieve cells and the flux f sucrse and ther metablites int the berry during ripening Ãnatmical aspects The specialised phlem tissue f vascular plants transprts carbhydrates (prduced as a result f phtsynthesis) and ther substances, t meristems, develping fruits, strage rgans, and ther sites f carbhydrate utilisatin. Accrding t Flwers and Ye, (1992) sieve tubes were discvered in 1837 by Harting, he als descriptin f exudatin frm the phlem was made in 186 by the same researcher. Phlem is ften difficult t identify in transverse sectins f plants but is generally spatially assciated with the xylem; tgether they cmprise the vascular tissue f the plant. The phlem tissue is

33 cmpsed f cnducting sieve elements, cmpanin cells parenchyma cells, fibers and sclereids. Bth f the latter a e supprting cells' Sieve tube elements are highly specialised cells thrugh which translcatin actually takes place. Mature sieve elements a e unique amng living plant cells and lack many structures nrmally fund in ther living cells such as nucleus, tnplast, glgi bdies, ribsmes, micrfilaments and micr tubules. In additin t the plasma membrane, rganelles that are retained include smewhat mdified mitchndria, plastids, and smth endplasmic reticulum (Taíz andzeigel l99l). Sieve tube members f the dicts are usually rich in phlem prteins called P-prteins which, with callse, appears t functin in sealing ff damaged sieve elements by plugging up the sieve plate pres (Taiz and Zeiger, 1991). Phlem prteins, which culd be visualised by light and electrn micrscpy, were als fund in the phlem exudates frm cucurbitaceae (Eschrich and Heyser, 1975; Crnshaw and Sabnis, 199). Sme f the P-prteins culd have imprtant rles in the lading f sugars and amin acids int sieve tubes and even in cell divisin at the meristematic regins cnnected t the end f sieve elements (Ishiwatari et al., 1995). The end walls f the sieve elements are mdified t frm sieve plates which are perfrated s as t allw the massive rapid flw f nutrient assimilates (Wding, 1978). Each sieve tube is assciated with ne r mre cmpanin cells. The tw types f cells arise in pairs frm the same parent cell. Numerus intercellular cnnectins, the plasmdesmata, penetrate the walls between sieve tube members and their cmpanin cells, suggesting a clse functinal relatinship and ease f transprt between the tw cells (Esau, 1965). Structurally, cmpanin cells appear less specialised than sieve elements; they retain a nucleus and have numerus ribsmes and rganelles which indicates a high level f metablic activity (Esau, 1965; Ridge, 1991). The phlem parenchyma serves as 'packing' but it may als be mdified t frm transfer cells and is then invlved in the lcal transfer f materials (Ridge, 1991). Smetimes cmpanin cells in the small leaf veins, which are the mst active sites f slute lading int sieve tubes, becme specialised t frm transfer cells (fudge, 1991) Techniques f phlem sap cllectin Varius techniques have been used in the past t cllect phlem sap frm many plant species. One f the ldest methds uses the natural exudatin prperty f certain plant species, which prduce phlem sap after an incisin is made t the tubes. This methd is easy t use but is limited t nly a small number f species, mstly wdy dictyledns (Zimmermann, 1957,196; Milburn 197; Richardsn et al., 1982; Chin et al., l99l). Sme researchers have used islated sieve elements fr their investigatins n phlem transprtatin f metablites. Islating sieve elements frm phlem tissue s that they

wuld be separated frm ther cells withut lsing their prperties is an extremely difficult task, especially when accunt is taken f the high lability f the differentiating sieve tubes and their strng

34 wuld be separated frm ther cells withut lsing their prperties is an extremely difficult task, especially when accunt is taken f the high lability f the differentiating sieve tubes and their strng plasmdesmatal links with cmpanin cells' Successful tapping f sap frm phlem has been achived by the use f insects such as aphids and cccids which parasitize the phlem by inserting their stylet tips int the sieve tubes. Upn cutting the aphid away frm the stylet which is left penetrating the sieve tubes, sap exudates can be cllected. The stylectmy methd is unsurpassed in many respects. Using this methd it is pssible t btain pure sap, uncntaminated by traces f substances frm the cut surface f ther tissues. On the ther hand there are limitatins fr this methd such as lw flw rate and lng sampling time as well as that f finding a suitable insect. Anther methd f identiffing translcated material is the replacement f translcates by labelled substances f the same type. ßCOz is usually supplied t ne leaf and va ius plant parts are subsequently analysed fr radiactive substances (Arnff, 1955; Swansn and El-Shishiny, 1958;Hale and Weaver, 1962). An alternative technique invlves tapping the sap (phlem exudate) frm differentiated sieve tubes dissected r punctured fr the purpse. Phlem exudate prvides a mre cmplete picture f the cmpsitin f the substances present in the lumen f the mature sieve elements. The facilitated exudatin technique was initiated by King and Zeevaart (1g74) and is based n the use f chelating agents e.g. (EDTA) in the cllecting slutin in which the cut end f the severed tissue is immersed. Chelatrs prevent blckage f the sieve-plate pres after wunding by their actin in absrbing calcium, which is an essential element in the presence f P-prteins fr the synthesis f the plugging material, callse' This technique is valuable in numerus herbaceus and wdy species (Pate et a1.,1974l' GIad et al..l992a). It is easy t use and allws the cllectin f relatively large vlumes f phlem sap cnsidered t represent true phlem sap cmpsitin (Weibull et al., 199; Girusse et al.. l99l) and sufficient fr bichemical analysis particularly in plants n which insect stylets cannt be used. It is generally accepted that the exudates cllected by these techniques are samples f sieve tubes cntents, and using these methds, many analytical studies have been carried ut t determine the cmpsitin f phlem exudates Leaves as surces f assimilates Phtsynthesis in the prductive leaves f higher plants prvides the assimilatary driving frce fr metablism f the sink rgans and sustains plant grwth and develpment. Phlem transprt in sieve tubes enables the transfer f assimilates frm surce tissues (mature leaves) t sink tissues elsewhere in the plant. Leaves are sites fr phtsynthetic

18 activity and the main surce f carbhydrates. Assimilate prductin in the leaves depends n bth cmpartmentatin and distributin f assimilates between strage and transprt cmpartments (Wardlaw, 199).

35 18 activity and the main surce f carbhydrates. Assimilate prductin in the leaves depends n bth cmpartmentatin and distributin f assimilates between strage and transprt cmpartments (Wardlaw, 199). Exprt f phtsynthate assimilates frm surce leaves is therefre dependent n the functining f many cmplex metablic events that cntrl the prductin f slutes such as sucrse, which translcate in phlem, and delivery f these slutes t the phlem (Stitt and Quick, 1989; Hunter and Visser, 1989, 1988 a;geiger and Fndy, 1991; Candlfi-Vascncels and Kblet, 199). In sme plants there is a relatinship between sucrse synthesis and the rate f exprt frm the leaf which suggests that the rate f phlem transprt is cntrlled directly by the rate f sucrse synthesis (Stitt and Quick, 1989; Geiger and Fndy, 1991). In ther plants leaf sucrse levels have n crrelatin with rates f exprt, suggesting that it may be phlem lading rather than sucrse synthesis which is imprtant in determining exprt rates (Geiger and Fndy,lggl; Gldschmidt and Huber, 1992). Develping leaves, including thse f grapevine, change frm assimilate imprters t exprters f phtsynthetic assimilates at the time they achieve 3-5% f their maximal size (Hale and Weaver,1962: Turgen and Webb, 1973; Fellws and Dalling, 1974;Twgen et a1.,1975) Retranslcatin frm reserves At the early grwing stage befre leaf phtsynthesis is sufficient, sluble carbhydrate reserves are mbilised frm the wdy tissues and translcated twards the grwing buds via the xylem stream. Sme carbhydrates may riginate frm the breakdwn f the callse lcated in drmant wdy tissues (Glad et al., 1992b; Alni et al., 1991). The ripening grape berry is a strng sink fr dry matter supplied by current phtsynthesis r reserves in wd (Cmbe, 1989), and mbilisatin f sucrse frm reserves t fruit can cntribute signifìcant amunts f carbhydrates during fruit maturatin and ripening (Candlfi-Vascncels et al., 1994; Williams and Biscay, l99l). Sucrse translcating in the grapevine culd riginate frm either f carbn fixatin r reserve tissue in the leaves r ther perennial parts f the vine. Current leaf phtsynthesis is the main surce f sugar fr maturatin f the berry but rembilisatin frm stem reserves ccurs under abnrmal cnditins such as defliatin (Matsui et al., 1979). Cnradie (198) analysing all parts f tw-year-ld vines shwed that the majrity f dry mass increase (abut fur-fld) in the ripening grapes was frm phtsynthesis with little change in the dry mass f ther parts f the vine. Hunter and Visser (1988) shwed that redistributin f phtsynthates t vegetative rgans is likely t take place at berry ripeness and that bunches were mainly supprted by basal leaves. Hwever, mbilisatin f strage reserves in the wd f stems and rts may als cntribute in the dry mass accumulatin (Kliewer and Antcliff, 197).

19 2.2.5 Transprt f carbhydrates V/ater is quantitatively the mst abundant substance transprted in the phlem' Disslved in the water are the translcated slutes, which cnsist mainly f carbhydrates in

36 Transprt f carbhydrates V/ater is quantitatively the mst abundant substance transprted in the phlem' Disslved in the water are the translcated slutes, which cnsist mainly f carbhydrates in almst all f the plants which have been studied s far. The sugar cmpsitin f phlem exudates is highly variable frm species t species. It has been firmly established that sucrse is the main mbile cmpnent transprted in the phlem f mst plants, with ligsaccharides (e.g. raffinse, stachise, verbascse), and sug Ìr alchls (e'g. srbitl, manitl), are translcated in sme species (Gamalei, 1985; Madre, l99i; Flwers and ye, 1992; Van Bel, 1993). In the sieve element these accunt fr at least 85% f all mbile cmpunds (Kursanv, 1984). Sucrse is the main mbile cmpnent, ccurring at cncentratins f.3-.9 M. Once carbhydrate is transprted frm, SâY, leaves t fruits, cnsiderable transfrmatin may ccur. Fr example, Table 2.1 shws the different metablites in the leaves and immature berries f grapes after feeding the leaves with labelled tc}z- Table 2.1 Translcatin f phtsynthetic prducts frm leavesa t immature berries (frm Peynaud and Ribéreau-Gayn, l97i). leaves Radiactivity (1 s) Berries Cmpund 48 hurs 6 hurs 24 hurs 48 hurs Ttal carbhydrates Ttal rganic acids Ttal amin acids Fructse Glucse Sucrse Malic acid 265 I l3 l l I tr. tr. tr. 5 8 I t7 l4 t t7 19 6l 38 4t Tartaric acid 1 I 14 Citric acid 2 tr. I u Leaves were fed 'COz during a 48 hur perid. Analyses were made f the leaves and adjacent grape berries f the same vine-stck after varius times f expsure as shwn. tr. : trace.

37 Sucrse Ziegler (1956), reprted paper chrmatgraphic analysis f sieve tube exudates frm sme Eurpean trees which shwed that the nly sugar fund was sucrse, and that hexses and their prducts were absent. Zimmermann (1957), analysed exudates f l6 Nrth American tree species and fund sucrse t be predminant in a number f them, with sme carrying sugars such as raffinse besides sucrse. In many plants, sucrse is the nly sugar present, but ther sugars, namely raffinse, stachyse, and verbascse, and ccasinally, the sugar alchls manitl and srbitl may ccur in sme species (Zimmermann, 196). Sucrse is the nly sugar detected in the phlem sap f maize (Mss and Rasmussen, 1969, Maillard and Prcher l99l), Ricinus cummunis L.(Hall and Baker, lg72), sugil beet (Bet vulgaris), ( Fndy and Geiger,1977); bean, (Giaquinta and Geiger, lg77), Pisume sativum (Barlw and Randlph, 1978), Oryza sativa (Kawabe et al., 198), Vicia faba, (Delrt, 1981), Triticum aestivum (Hayashi and Chin, 1986), Medicag sativa (Girusse et al., l99l; Chin et al., 1991), and Banksia printes (Jeschke and Pate, 1gg5). The cncentratin f sucrse in the phlem sap f alfalfa is reprted t be abutg}y f assayed sugars (Girusse et al., 1991). Sucrse is the main sugar prduced by phtsynthesis and translcated in grapevines. This cmpund may be derived frm recent phtsynthesis carbn fixatin r be mbilised frm reserve tissues in leaf r ther perennial parts f the grapevine (Swansn and Elshishiny, i958; Kblet, 1977; Glad et al-,1992a) Hexses Besides sucrse. hwever, glucse and fructse have been reprted in amunts that seem t depend largely n the sampling site, experimental prcedures, and times. All reprts cncerning sieve tube exudates, cvering a ttal f abut 45 species, stress the cmplete absence f hexses even in chrmatgraphically detectable traces and reprts n displacement f naturally ccurring translcates with 'COr-labelled phtsynthates supprts this (Zimmermann. 196; Flwers and Ye, 1992). There are indicatins that sucrse is the nly translcated sugar in the grapevine, and that the hexses are secndary prducts, derived frm hydrlysis after sucrse has mved ut f the actual translcatinal channels (Swansn and El-Shishini, 1958). It is pssible that sme sucrse mlecules diffuse ut f sieve tubes during nrmal translcatin and are hydrlysed in adjacent cells t glucse and f uctse which are nt nrmally mbile in phlem (Swansn and El-Shishini, I 958).

21 2.2.6 Transprt f nitrgen cmpunds Althugh nitrgen slutes frequently are the majr cmpnents f dry matter in grapevine xylem exudates (Andersn and Brdbeck, 1989a; 1989b, Glad et al-, 1992b; lg94) they

38 Transprt f nitrgen cmpunds Althugh nitrgen slutes frequently are the majr cmpnents f dry matter in grapevine xylem exudates (Andersn and Brdbeck, 1989a; 1989b, Glad et al-, 1992b; lg94) they are als imprtant in the phlem, being secnd nly t carbhydrates (Pate, 198; Glad et al., 1992a; Rubelakis-Angelakis and Kliewer, 1992). Phlem transprt invlves mstly the same nitrgen frms as xylem transprt (Pate, 198; Bray, 1983), but phlem exudates usually cntain 1-2 times the xylem cncentratin f nitrgenenus slutes Althugh nitrate cncentratin may be high in xylem, it is ften absent r much lwer in phlem (Pate, 198). Phlem translcatin f nitrgen is imprtant t supply these cmpunds t fruits (Pate, 1983). Nitrgen ccurs in phlem largely in the frm f amin acids and amides especially glutamate/glutamine and aspartate/asparagine. The amin acids glutamine, asparagine, glutamate, aspartate, alanine, and serine have been reprted t be the main nitrgen-cntaining cmpunds in phlem sap f plants (Hall and Baker, 1972; Leckstein and Llewellyn, 1975; Pate, 198; Hking, 198; Simpsn and Dalling, 1981; Urquhart and Ly, 1981;Fukumrita and Chin, 1982; Fisher and Mac- Nicl, 1986; Chin et al., l99l;feller and Helzer,Iggt; Girusse et al., 1991; Wiener et al.,l99l; Glad ef al.,1992b; Jeschke and Pate, 1995). Glad et al. (1992a), reprted the nitrgenus slute cmpsitin f grapevine phlem sap (Vitis vinifera L. cv. Pint nir). During flwering, glutamine was the majr frm f transprted nitrgen in the phlem sap f the grapevine (6%) and prline the secnd (1%) Transprt f rganic acids Organic acids as phtsynthesis assimilates translcate int cnducting cells f leaves alng with amin acids. but in a lwer cncentratin. Malic acid, tartaric acid, citric acid and ket acids have been detected in phlem sap f sme wdy plants (Ziegler, 1962, cited in Kursanv. 1984). Hwever, it is nt clear whether these rganic acids are transprted directly frm leaves as phtsynthetic prducts r whether they are synthesised in sieve tubes themselves as a result f glyclysis reactins and tricarbxylic acid cycle (Kursanv, 1984). In phlem transprt studies, experiments withlacoz have shwn that labelled rganic acids are the earliest phtsynthetic prducts emerging in cnducting bundles after the leaves have been supplied with laco2. The relative amunt f rganic acids amng the lac assimilates entering the cnducting cells is small, being equal t nly 2.5/ f labelled cmpunds cmpared with carbhydrates, (9l), and amin acids, (7.3%) (Kurasanv, 1984).

22 Tartaric acid, malic acid and citric acid were detected at levels f 17, 11, and 25 nml respectively frm phlem sap exudate f grape bunches at flwering time (Glad et al., 1992a). 2.2.8 T ransprt f minerals Ptassium is the predminant catin in plant phlem exudates nrmally accunting fr mre than 6% f the ttal catins (Kursanv, 1984).

39 22 Tartaric acid, malic acid and citric acid were detected at levels f 17, 11, and 25 nml respectively frm phlem sap exudate f grape bunches at flwering time (Glad et al., 1992a) T ransprt f minerals Ptassium is the predminant catin in plant phlem exudates nrmally accunting fr mre than 6% f the ttal catins (Kursanv, 1984). Bukvac and V/ittwet (1957) studied absrptin and translcatin f a number f fliar-applied radiactive istpes. Using recvery in untreated plant parts as the cntrl, they fund Rb, Na, and K ins were the mst readily absrbed and mst highly mbile; P, Cl, S,Zn, Cu, Mn, Fe, and M ins were intermediate with decreasing mbility in the rder given; while Ca, Sr, and Ba ins were absrbed int the leaf but nt exprted. In ripe grape berries, K+ cnstitutes 7/ f the mineral catin cntent f the berry and 45/ f this ttal accumulates in the berry skin (Smers, 1975; Hale, 1977; Iland and Cmbe, 1988). Changes in K+ cntent in fruits is generally assciated with phlem flw and in grapes is usually cntrasted with Ca2+ accumulatin. Therefre, change in xylem flw after veraisn results in decreasing calcium accumulatin in the berry,. Calcium ins mve passively within the xylem via the transpiratin stream, and the Ca2+ cntent in fruits has been used as an indicatr f cumulative xylem flw ( Lang and Thrp, 1989; Mrrisn and Idi. 199). Creasy et al. (1993), using Pint nir grapevines, shwed that K+ in the berry increased rapidly after veraisn whereas this was nt the same with Ca2+ because f berr.v xylem discntinuities at the veraisn stage. Calcium is well knwn t be transprted nly in the xylem (Marschner, 1983) while phlem sap cntains little r n Ca and is nt cnsidered t be a medium fr Ca mvement (Pate, 1975). Additinally, amng the catins, ptassium has a rle in stimulating the unlading f assimilates (H, I 988). During the secnd grwth perid f grapevine there is an increase per berry in catins such as K*, Na*, Ca2* and Mg2- and in anins such as PO-1. During fruit maturatin and ripening, mbilisatin f K* frm reserves t fruit increases in significant amunts (Williams and Biscay, 1991). Ptassium was the main mineral in fund in the Chardnnay grapevine xylem exudate (73pM), fllwed by calcium, phsphate, nitrate, sulphate, magnesium, and chlride (Glad et al., 1992b). Ptassium was exuded at levels clse t 8 nml/ sample whereas nitrate was nt detected (Glad et al., 1992a). They fund that phlem sap was slightly alkaline and quted the ph value f grapevine phlem sap exudate as being 7.7 (Glad et al., 1992a). Bselli et al. (1995) reprted that the ptassium cntent f berries f tw cultivars inc eased t apprximately.2-.3 mg berry-r d-r during the entire perid f fruit

40 23 develpment, but the cncentratin f calcium and magnesium in the berry reached maximum at veraisn and then decreased. Their results shw that, n a fresh weight basis' the ptassium level was cnstant during the berry develpment perid whereas calcium and magnesium levels decreased during the whle berry develpment perid' Transprt f secndary metablites There is little infrmatin regarding the translcatin f secndary metablites in plants. Secndary metablites such as plyphenls d nt seem t be typically invlved in phlem transprt. Nevertheless, accrding t sme reprts small amunts f phenlic substances have been reprted in phlem exudates (Kursanv, 1984 and references cited therein). The pssibility f transprt f large amunts f plyphenls in the phlem seems unlikely because, in differentiated sieve elements, the cytplasm is nt prtected against the cntents f the central lumen by a tnplastic membrane but is partially dispersed in it; this creates cnditins cnducive t prtein precipitatin (Kursanv' 1 984). Cnsiderable quantities f different secndary metablites accumulate in grape berries during grwth and ripening. The ability f the varius tissues t accumulate secndary metablites is specific t the develpmental stage, the tissue, and cell type in the mature grape berry, and it seems that epidermal cells are the mst apprpriate site fr the accumulatin f secndary metablites such as pigments, ther nn-pigment plyphenls, and flavur cmpunds. Cnsidine ( I 979) remarks upn the ccurrence f plyphenls in mst pericarp cells at anthesis; thse f the uter epidermis and the mescarp stained differently. By day 16 after anthesis there were tw cell types between the hypdermal layers and the psitin f the vascular bundles, ne cntaining plyphenls and the ther nt. By day 26 there was a reductin in plyphenls in all except the dermal system. It seems that, despite the similar functin f inner and uter mescarp, cells f these tissues, they are metablically dilferentiated in the hrst grwth phase. The imprtant questin which arises here is whether r nt secndary metablites are actually translcating int grape berries via phlem sap and, if s, is this the mechanism respnsible fr the accumulatin f flavur cmpunds in the grape berries.

24 Chapter 3 Distributin f glycsides in grapevine tissues Cntents Page 3.1 Intrductin... 3.2 Materials and methds... 3.2.1 P1ants... 3.2.2 F ruit grwth... 3.2.3 Plant materials... 3.2.4 Tissue extractin 3.

41 24 Chapter 3 Distributin f glycsides in grapevine tissues Cntents Page 3.1 Intrductin Materials and methds P1ants F ruit grwth Plant materials Tissue extractin Glycsyl-glucse determinatin 3.3 Results Discussin Leaves Bunches and berries Cnclusin

25 3.1 Intrductin The aim f this study was t investigate the amunts f ttal glycsides as an indicatr f ttal secndary metablites in different tissues f grapevines.

42 Intrductin The aim f this study was t investigate the amunts f ttal glycsides as an indicatr f ttal secndary metablites in different tissues f grapevines. In particular, studies were made f the changes in cncentratin f glycsides in the leaf blades and petiles as well as in berries at varius grwth and physilgical stages thrughut the grwing seasn fr cvs Muscat Grd, Suløna and Flame Seedless vines. 3.2 Materials and methds Plants Mature (apprximately 1l year-ld) vines f Vitis vinifera L., Muscat Grd Blanc (syn. Muscat f Alexandria), Sultana (syn. Thmpsn Seedless) and Flame Seedless grapes were used in this experiment. The vines were grwn in the V/aite Agricultural Research Institute experimental vineyard at Glen Osmnd, Suth Australia Fruit grwth Inflrescences shwin g 5% flwer cap fall were selected n the same day (4-6 bunches fr each cultivar), labelled, and the date f flwering recrded. Fr berry grwth measurements, 25-3 berries f different size and frm different sides f the bunch (4 t 6 berries n each) were labled and measured weekly n the vine fr diameter. length. and berry defrmability. At the same time every week, 1-15 berries were picked randmly frm clusters and were used t determine berry dimensins, berry weight and 'Brix (ttal sluble slids) in all the three cultivars. Berry density was calculated by dividing berry fresh weight by berry vlume calculated frm measurement f th ee berry axes, length, smaller diameter and larger diameter (Cmbe et al. 1987). BerÐ, defrmability was measured with callipers and juice 'Brix was measured by hand refractmeter.

26 3.2.3 Plant materials Beginning at three weeks after flwering in the 1992-93 seasn, triplicate samples f fruits and leaves were taken at three week intervals up t veraisn and then a sample at

43 Plant materials Beginning at three weeks after flwering in the seasn, triplicate samples f fruits and leaves were taken at three week intervals up t veraisn and then a sample at veraisn, at early ripening and at ripeness. Mature leaves at the furth t sixth ndes frm the base f the sht were picked and separated t leaf blades and leaf petiles. V/hle unifrm bunches, including berries and bunchstem, were used as a fruit sample in this experiment and are labelled whle bunch t distinguish frm 'berries'. A separate sample f berries was taken in fr the determinatin f glycsyl glucse. A series f berry samples was als cllected fr cvs Muscat Grd and Sultana in the seasn frm befre veraisn until ripeness. All plant samples were frzen in liquid nitrgen and stred at'2 'C until extractin' Tissue extractin Extractin was carried ut as fllws: Samples were immersed in liquid nitrgen and the slid frzen tissue was grund, first in a labratry blender and then in a cffee grinder t give a ñne pwder. Fr extractin f glycsides, the methd used by Gunata et al. (1985a) was adpted after mdificatin when it was fund that.3 M sdium acetate buffer, ph 5. imprved the recvery. Grund tissue sample (5g) was added t acetate buffer (4 ml) at 9 'C. After ph adjustment t 5. by additin f NaOH (.5 M), the slutin was held fr 5 minutes at 9'C, then cled and centrifuged at 9 x g fr l5 min. at2 C in a Srvall superspeed RC2-B Autmatic refrigerated centrifuge. After remval f the supernatant, the slids were re-extracted with 3 ml f ht buffer in the same way. The supernatants were cmbined fr islatin f glycsides. A different prcedure was used fr ripe fruit. Whle bunch samples, which included the bunchstem, skin and flesh f the ripe berries were grund under liquid nitrgen then hmgenized with an Ultra-Turrax Tl25 high speed hmgenizer with an F25N dispersing head (Janke & Kunkel GmbH & C., Germany). This mix was extracted with ht buffer as fr ther tissue samples. All extracts and juice samples were stred at. -2'C until used fr assay f glycsyl glucse (G-G). Muscat Grd and Sultana berry samples frm the 1995 seasn were prcessed differently. In this case a sample f berries was cunted, weighed, hmgenised and extracted with 5% ethanl as described in Williams et al. (1995) fr G-G determinatin.

$An utline f the prcedure is shwn in Fig. 3.1. Sample Hmgenized fruit extract r grund tissue extact r Juice + Islatin f glycsides Cl8 Reverse Phase cartridge G ide fractin + Hydrlysis 1.$

44 Glycsyl-glucse determinatin The glycside assay fr determinatin f glycsyl-glucse accrding t the principles utlined by Williams et al. (1995) was adpted in this wrk. An utline f the prcedure is shwn in Fig Sample Hmgenized fruit extract r grund tissue extact r Juice + Islatin f glycsides Cl8 Reverse Phase cartridge G ide fractin + Hydrlysis 1.5 M H2SOa, t h, 1'C Glucse + Glucse analysis enzymlc assay Glycsyl Glucse value G-G Fig. 3.1 G-G assay steps

28 3.3 Results In this study glycsidically bund secndary metablites were islated and measured fr Muscat Grd Blanc, Sultana and Flame Seedless cultivars.

45 Results In this study glycsidically bund secndary metablites were islated and measured fr Muscat Grd Blanc, Sultana and Flame Seedless cultivars. The measurements fr each sample were calculated in pml f glycsyl glucse per gram f tissue fresh weight. The means f measurements fr each cultivar during the grwth perid are pltted against time after flwering in Figures t 3, 3.3 '1.t 3, and t 3. Figures3.2.4t 6, t 6, and t 6 shw changes in berry density, sluble slids and berry vlume. G-G accumulatin in berries was calculated n a per berry basis based n berry samples separate frm the whle bunch samples (Figs.3.2.3, and 3.4.3). In sme graphs standard enr (SE) f the mean is given by a vertical bar. In Muscat Grd Blanc the cncentratin f glycsides in leaf blades and petiles increased gradually in weeks th ee t 14 after flwering and then increased sharply frm week 14 t 18 (Fig and 3.2.2). There was a sharp decrease in the cncentratin f glycsides in whle bunch samples with the initial acceleratin in berry vlume, but the secnd phase f rapid berry grwth (see Fig ) had n such effect. In Sultana the cncentratin f glycsides in leaf blades and petiles increased in a parallel fashin thrughut the perid f berry develpment (Fig and 3.3.2)' Leaf blades had seven-fld higher G-G cncentratin cmpared t petiles. The changes in whle bunch glycsides shwed a similar trend t that f Muscat Grd berries, althugh the cncentratins f G-G in Sultana were higher than were fund in the Muscat berries in the first six weeks and declined t lwer cncentratin at berry ripeness (cmpare Fig with Fig ). In cntrast t the patterns seen in Muscat Grd and Sultana, the rate f increase in G-G cncentratin f Flame Seedless leaves and petiles shwed n increase until the sample f ten weeks after flwering when berries were ripe (Fig. 3.a.1,3.4.2); cncentratins then declined in the furth sample. The G-G cncentratin f Flame Seedless whle bunches shwed a decrease frm befre t after veraisn fllwed by a small increase thereafter (Fig ). The graphs f G-G cncentratin in leaf blades, petiles andbunches frm Figs.3.2, 3.3 and 3.4 are regruped in Figs. 3.5 t illustrate the cmparative changes in the three cultivars. Cncentratins f G-G were much higher in leaf blades cmpared with petiles and fruit (Figs t 3). Of the cncentratins in leaf blades f the three cultivars (Fig ) levels were greatest in Sultana, intermediate in Muscat Grd and lwest in Flame Seedless (except fr ne high value at week l when berries were 2OBrix). G-G cncentratins tended t increase in leaf blades during the berry develpment perid in all cultivars, e.g. by 95/ and 53Y in Sultana and Muscat Grd respectively.

The changes in the cncentratins f G-G in petiles (Fig. 3.5.2) were, in general, simila t thse in leaf blades but at a lwer level - at abut ne fifth.

46 The changes in the cncentratins f G-G in petiles (Fig ) were, in general, simila t thse in leaf blades but at a lwer level - at abut ne fifth. The levels in Sultana were nt predminant as in leaf blades being lwer at the early and late sampling times. Again, there was a single higher value at week 1 in the petiles f Flame Seedless, as seen in leaf blades, and als petiles shwed the same tendency t increase ver the measurement perid, Sultana by 98% and Muscat Grd by 157%' The pattern f change in the G-G cncentratin in whle bunches (Fig ) was quite different t that shwn by leaf blades and petiles. In all three cultivars the levels declined frm their highest level at the first sampling (three weeks after flwering, just after berry setting). The decline was mst rapid until veraisn (eight weeks after flwering in Muscat Grd and Sultana and five weeks in Flame Seedless) after which the levels steadied during berry ripening. Cncentratins were similar in Muscat Grd and Flame Seedless but shwed greater extremes in Sultana fruit, being highest early and declining t lwest levels at berry ripeness. This pattern f decrease in G-G cncentratin in whle bunches was shwn t be reversed when the G-G was determined n as a per berry basis (Figs. 3.6 and 3.7)' This determinatin was made n berry samples taken separately frm the whle bunch samples but frm the same vines. They are graphed in Fig. 3.6 against "Brix f juice (as an alternative indicatr f berry develpment). The graphs shw that levels f G-G per berry at the first sampling (5"Brix) started at levels f <.2 pmles, then increased cnsiderably. The increase was greatest in Muscat Grd eventually reaching a level f.74 pmles G-G per berry. On the ther hand the amunt in Sultana berries at the secnd sampling (6'Brix) had risen t.2 pmles but, unexpectedly, shwed n further increase thereafter. As a check n these results, a secnd series f samples f Muscat Grd and Sultana were taken in the seasn and are shwn in Fig 3.7. Thcsc results shwed similar trends t thse in Fig. 3.6 except that the values f G-G per berry were smewhat higher than thse in

$Ì Er E ct (5 t6 t5 l4 l3 12 11 1 1 5(Xt E E 5.* \ t 2 q 1 4 3 I 2 f 6 I l 12 14 16 18 2 22 24 Weeks alter f lwerlng 2 4 6 I r 12 14 16 18 2 Weeks after llwering 4., 3.5 È i 3. Er õ 2.s E 9(5 2. l5 1.$

47 Ì Er E ct (5 t6 t5 l4 l (Xt E E 5.* \ t 2 q I 2 f 6 I l Weeks alter f lwerlng I r Weeks after llwering 4., 3.5 È i 3. Er õ 2.s E 9(5 2. l cl cd ad c t, t 1.t Weeks after flwering I 1 12 Weeks after llwering 14 't6 + ì 12 cd E (,.6 F É F ct! tt x?5 2 :\ O L r< E c'" =- Or^ (J I 1t Weeks alter llwerlng Weeks atter llwerlng Figs3.2.1 t6evlutinfg-ginleafblades(l), leafpetile(2)andfruit(3)and changesinberryvlume(4),berrydensity(5),andfruitslubleslids(6)inmuscat Grd with time in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in macerated whle bunch n a fresh weight basis (left axis) and in mg per separated berry (right axis). Vertical bars thrugh mean values represent SE.A: inceptin f "Brix increase.

$3.3 3 --+-- 5 25 6 lj an E an '" 1. g, P ó ô z j 2 x \ n,. E ; 1s 1.1 ó 1.5.5 5 ó.. 2 4 6 8 l 12 14 16 18 2 2 4 6 I t 12.t4 16.tI 2 22 Weeks after flwerlng weeks afler llwerlng Figs 3.3.1 t 6 Evlutin$

48 ' Þtg l = Et 2 1 t E ts Ër* E t a 4 31 (t I ct l t I l I t 12 Weeks after f lwering Weeks alter f lwering 3 ì z.s j Et È z. E (5 (, 1.s 2 5 E cd =an c t 1.2 I.'l I l I f Weeks after f lwering Weeks after f lwering + È lj an E an '" 1. g, P ó ô z j 2 x \ n,. E ; 1s 1.1 ó ó l I t 12.t4 16.tI 2 22 Weeks after flwerlng weeks afler llwerlng Figs t 6 Evlutin f G-G in leaf blades ( I ), leafpetile ( 2 ) and fruit ( 3 ) and changes in berry vlume ( 4 ), berry density ( 5 ), and fruit sluble slids ( 6 ) in Sultana with time in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in whle bunches n a fresh weight basis (left axis) and in mg per separated berry (right axis). Vertical bars thrugh mean values represent SE. A: inceptin f "Brix increase.

t6 I j ctt E 18 (, 2 12 4 1 r2 t E s E : 6 à c 4 2 4 32 2 4 6 I 1 12 14 Weeks after flwering 2 4 6 I t Weeks after llwering 12 È Eì E (5 3. 2.6 22 1.8 1.4 2 5 E Et =al, tr! lt 1,2 1.1 1. 1. 2 4 6 I 't 12 14.

49 t6 I j ctt E 18 (, r2 t E s E : 6 à c I Weeks after flwering I t Weeks after llwering 12 È Eì E ( E Et =al, tr! lt 1, I 't I Weeks af ter f lwering Weeks after flwering D b = j.4 2 ct) : ō E (5 ( ó q) - uj _ x sù r I(t I l 12 1{ Weeks af ter f lwer ng 4 6 I 't Weeks after llwering Figs 3.-f.1 t 6 Evlutin f G-G in leaf blades ( I ), Ieaf petile (.2 ) and fruit ( 3 ) andchangesinberryvlume(4),berr,vdensity(5),andfruitslubleslids(6)in Flame Seedless with tinre in weeks after flwering. G-G data in Fig. ( 3 ), are expressed as cncentratin in r.vhle bunches n a fresh weight basis (left axis) and in mg per separated berry (right axis). Vertical bars thrugh mean values represent SE.A: indicatin f 'Brix increase.

21 1 33 2 = j Ên 16 E 1p t5 4 2 4 6 I 1 12 14 16 18 2 22 24 Weeks atter flwerlng 4 2 I 3 Et ts a 2 4 6 I 1 12 14 t6 18 2 22 24 Weeks after f lwering 3 3-3 õ) E (J (, - ts -.

50 = j Ên 16 E 1p t I Weeks atter flwerlng 4 2 I 3 Et ts a I t Weeks after f lwering õ) E (J (, - ts -.È + Muscal Grd Sultana Flame Seedless I Weeks af ter llwerlng Figs t Glycsyl glucse (G-G) cncentratin in different tissues f all th ee cultivars cmparing G-G in leaf blades (Fig.l), in leaf petiles (Fig.2 ) and whle bunches (Fig.3). M, Muscat Grd; S, Sultana; F, Flame Seedless.

51 '.c =\ E (, I ( Muscat Grd Flame Seedless Sultana 'Brix Fig. 3.6 Glycsyl glucse cncentratin in berries f Muscat Grd, Sultana and Flame Seedless pltted against juice 'Brix (data frm ) ) 8 Muscat Grd E J (, I (5.6.4 Sultana.2. E 'I 'I 'Brix Fig. 3.7 Glycsyl glucse cncentratin in berries f Muscat Grd and Sultana pltted against juice 'Brix (data frm ).

$(1993) reprted that glycsides are a majr fractin f leaf extracts in a wide variety f plants. In this study the results shwed that grapevine leaves accumulate large quantities f glycsylated cmpunds.$

52 Discussin Leaves Leaf btade. Leaf blades are knwn t be highly active sites f phtsynthesis and carbn assimilatin. Tazaki et al. (1993) reprted that glycsides are a majr fractin f leaf extracts in a wide variety f plants. In this study the results shwed that grapevine leaves accumulate large quantities f glycsylated cmpunds. Althugh it is als knwn that grapevine leaves begin t transprt assimilates after gaining abut 5Y, ftheir final size (Hale and Weaver,1962; Turgen and Webb, 1973; Flwers and Geiger, I974; Turgen et al., I975), the results reprted in this chapter shw that the strage f glycsylated cmpunds cntinues after the leaves mature and reach their maximum size (see Fig. 3.2.I,3.3.1 and 3.4.1). This culd be due t strage f excess f translcating metablites r it may be due t physilgical changes in the leaves during their senescence develpment. At the end f the grwth perid and during fruit ripening, leaves shwed different patterns f glycside evlutin depending n the cultivar. Muscat Grd and Sultana leaves cntained highest levels at the late berry ripening stage with the frmer shwing the sharpest increase in cncentratin (see Fig and 3.31). In cntrast, glycside cncentratin in Flame Seedless leaves decreased at the late ripening stage (Fig. 3.a.1). Because it is an early ripening cultivar the leaves f Flame Seedless may have reached advanced senescence at the last sampling time. The G-G levels f leaf tissue are cmpa ed in Fig Leaf petile. Leaf petile is cnsidered as a passage fr the assimilates transprted frm the leaves t ther parts f plant. In this experiment leaf petiles fr all three cultivars analysed. shwed a pattern f cncentratin f glycsides similar t that f the crrespnding leaf blades. hwever, the cncentratins were at levels f abut ne fifth f thse fund in the leaf blades. (see Fig , and ). Reasns fr the lwer levels f glycsides in the petiles culd be due t firstly, petile tissue is nt as active phtsynthetically as leaves are, and secndly, the petile tissue, which cnnects the vascular bundles f leaves t the shts, is nt knwn as a strage tissue. The G-G levels f petile tissue are cmpared in Fig Bunches and berries Measurement f glycsides in whle bunches cannt be cnverted directly t per berry because f the presence f peduncle tissues in bunch samples. The data reprtted in Chapter 7 suggest that secndary metablites are abundant in peduncle tissue. This bservatin is supprted by calculatin f cncentratin f G-G per g berry weight

53 (dividing the amunt per berry by berry fresh weight- data nt shwn) with G-G per g fesh weight f whle bunch. Such cmparisn shws that the berry samples had cncentatins 18, 19,23,26,34,and32%f bunch samples at successive sampling dates. These calculatins d nt detract frm the fllwing cnclusins. Glycside cncentratin in the whle bunches f all three cultivars decreased frm berry set until veraisn (Fig 3.4.3) but glycside amunt per berry increased (Fig. 3.7). This reversal is explained by the large amunt f berry grwth that ccurs during this perid f berry develpment, especially in the large-berried Muscat Grd fr example Muscat Grd berries at the first sampling averaged.47 g fresh weight but 3.41 g at the last sampling. In ther wrds, the decline in G-G cncentratin in fruit, which differs s markedly frm the increase shwn in leaf blades and petiles, \ryas nt reflected in a decrease in the amunt f G-G per berry but t the dilutin f an increasing weight f G- G by a prprtinally greater increase in water and dry mass in the develping berry. Leaf blades and petiles shwed small changes in fresh weight per rgan; fr example, the range f fresh weights f Muscat Grd leaf blades and petiles nly fluctuated + 9y and t 4/ in all samples. These small differences wuld nt have altered the general pattern f increase in G-G shwn in the cncentratin values shwn in Figs. 3.5'l and Thus all three rgans - leaf blades, petiles and fruit - shwed an accumulatin f G-G per rgan during the berry develpment phase f these vines. This differs frm reprt f the high accumulatin f secndary metablites in immature plant rgans (Stahl- Biskup et al. 1993). Given the high percentage f secndary metablites that are glycsylated (Barz and Kster, l98l;kurt and Trssell, 1983), the measurement f glucse released frm islated glycsides, as dne in the G-G assay, appears t be a cnvenient index f ttal secndary metablites in plant tissue' Cnclusins During the berry ripening perid f a grapevine's yearly cycle f develpment, the cncentratins f glycsyl-glucse (G-G) increase in leaf blades and petiles but decline in fruit. This difference in pattern is shwn t be due t the large weight increase that takes place in berries during this perid cmpared with leaves; in fact n a per berry basis, the weight f G-G als increases during its develpment. Cmparing cultivars, the amunt f G-G per Muscat Grd berry abut dubles as Brix increases frm 6 t23, while fr Sultana there is nly a small increase ver this perid; this reflects their relative flavur status. The glycsyl-glucse assay shws prmise as an indicatr f the cntent f ttal secndary metablites in plant tissues.

37 Chapter 4 The invlt'ment f leaves in the accumulatin f flavur cmpunds in Muscat Grd grape berries Cntents Pase 4.1 Intrductin... 4.2 Materials and methds... - 4.2.1 Plarft materials.

54 37 Chapter 4 The invlt'ment f leaves in the accumulatin f flavur cmpunds in Muscat Grd grape berries Cntents Pase 4.1 Intrductin Materials and methds Plarft materials.. - Glas s huse experiment. Field experiment. A.2.2Methds f grwing plants and grafting bunches.. P re-r ted cuttings... Grafting bunches frm field grwn vines nt ptted vines Chip-budding nt pre-rted cutting... Bench-grafiin Sample preparatin G as chrmatgraphy-mass spectrmetry (GC-MS) Results 4.4 Discussin Cnclusin l l 52

38 4.1 Intrductin Mnterpene cmpunds cntribute significantly t the characteristic flavur f flral grapes and are generally present nly at lw levels in nn-flral varieties.

55 Intrductin Mnterpene cmpunds cntribute significantly t the characteristic flavur f flral grapes and are generally present nly at lw levels in nn-flral varieties. These arma cmpnents, which are crnmn cnstituents f many fiuits, are present in free frm and, mre abundantly, as nn-vlatile glycsides (Williams et a1.,1982a, V/ilsn et al., 1984, Gunata et al., 1985a). The sites f synthesis f these flavur cmpunds in grapevines are unknwn. There is limited infrmatin cncerning the cmpsitin f secndary metablites in grapevine leaves. Wildenradt et al. (1975) reprted six-carbn cmpunds and als terpenes and their derivatives at high levels in grapevine leaves. Gunata et al. (1986) reprted the presence f terpenls and armatic alchls in free and bund frms in Muscat f Alexandria (syn. Muscat Grd Blanc) leaves in much higher cncentratin than in berries. Gunata et al. (1986) als analysed Muscat Grd leaf blades and petiles fr terpenls and reprted that, despite having the same prfile f these cmpunds, the changes in levels f terpenls in the leaf blade and petile were different during the perid f fruit develpment. In the leaf blade the levels f terpenyl glycsides increased prgressively during grape maturatin, whilst in the stalk bth free and bund frms f terpenls decreased. Their results als shwed that the leaf blade was higher in free and bund armatic alchls than either the petile r the berry. They raised the questin that if the petile is a passage fr metablites transprted frm leaves t ther parts f the plant, the changes in cncentratin f terpenls in the petile wuld be similart thse in the blade and if the glvcsides are cnsidered a preferential frm f transprt then the petile wuld cntain mre f the bund frms. They cncluded that if these cmpunds are translcated frm the leaves t the berries, this was nt reflected in the levels f the cmpunds in the varius plant parts during the seasn. Bravd et al. (1989) after feeding Muscat Grd grapevine leaves with lac-labelled mevalnic acid fund that bth leaves and berries cntained labelled mnterpene glycsides l2 t 72 h after feeding. The authrs suggested that mnterpene glycsides are translcated frm leaves t berries. Skurumunis and Winterhalter (1994) fund C 3 nrisprenids present in Riesling grapevine leaves in cncentratins l-1 times higher in Riesling wine. Nne f the references suggested anv clear mechanism fr translcatin f secndary metablite cmpunds between leaves and berries.

In a textbk n viticulture, Winkler et al. (1974, pp.l7-171) described results f an unpublished experiment which implied that grape berries have the capacity t prduce flavur cmpunds directly.

56 In a textbk n viticulture, Winkler et al. (1974, pp.l7-171) described results f an unpublished experiment which implied that grape berries have the capacity t prduce flavur cmpunds directly. The experiments repfed, thrugh infrmal tastings nly, the retentin f flral flavur characteristics in clusters f Muscat berries grafted and grwn n vines f a nn-flral cultivar, and the nn-develpment f Muscat flavur in neutral fruit grafted and grwn n Muscat vines' Since knwledge abut translcatin f secndary metablites in Vitis vinifera grapevine cultivars is lacking, the aim f this study was t investigate the invlvement f leaves as a surce f metablites and their influence n the cmpsitin f the fruits. The develpment f analytical techniques fr the analysis in grapes f glycsidically-bund mnterpenes (Wilsn et al., 1984) prvides methds fr bjective investigatins int the sites f bisynthesis f flavur cmpunds in Vitis vinifera L. The experiments described in this chapter used these techniques cmbined with bunch grafting f fruit t determine if mnterpene flavur cmpunds are synthesised in grape berries r translcated int them Materials and methds 4.2.1Plant materials Glasshuse experimenl Cuttings f Muscat Grd Blanc and Shiraz vines were gr\ rrt in the glasshuse by the methd f Mullins and Rajasekaran (1981) t prduce fruit. Tw pre-rted cuttings, ne f each cultivar, u,ere planted in a pt clse tgether. One f the rtlings in each pt was allwed t prduce a sht and the ther trained t prduce an inflrescence. Abut tw weeks befre flwering the inflrescence f ne vine was grafted nt the sht f the ther by apprach-grafting (Fig. a.l ). After five weeks, by which time berries had frmed, the bunch was cut ff frm its parent plant and allwed t develp n the ther cultivar's sht until the berries were ripe (Fig. a.2). Cntrl plants were left ungrafted and grwn under the same cnditins. Berries were harvested and kept frzen at -2"C until they were used fr extractin. Field experiment In the fllwing seasn. bunches were grafted using pairs f shts n mature Muscat Grd and Sultana vines grwn side by side in the same rw s that their shts intermingled. The timing and methd f grafting were similar t that used in the Muscat Grd/Shiraz glasshuse experiment. Six bunches were set up fr each f fur treatments (Muscat n Sultana. Muscat ungrafted, Sultana n Muscat and Sultana ungrafted).

4 Duplicate bunches were harvested at three ripening stages þre-ripe, early ripe and very ripe). 4.2.

57 4 Duplicate bunches were harvested at three ripening stages þre-ripe, early ripe and very ripe) Methds f grwing plants and grafting bunches Pre-rted cuttings Fr the glasshuse experiment drmant cuttings were btained frm the field befre bud burst. The cuttings \ryere the basal part f canes, abut 25 cm in length. All the buds were remved except the tw uppermst buds, and cuttings were planted in a heated rtingbed cntaining heating cable and carse sand in a bx. The rting bx was in a rm with the air temperature held at 4 "C. The cuttings were planted s that the tw upper buds remained ut f the sand and the lwer end was in the wann sand. Tap water was used fr watering every tw days. After abut fur weeks, tw f the rted cuttings, ne frm each cultivar, were planted in a pt f 2 cm diameter filled with a6:3:l mixture f perlite, vermiculite and peat mss and placed in the glasshuse and watered with Hagland slutin. When buds burst the methd f Mullins and Rajasekaran (1981) was used retaining nly ne bud t btain inflrescences n the dnr vine. Fr the hst vine, ne f the buds was allwed t grw and the inflrescences were severed when they appeared n the newly grwing sht. Grafting bunches frmfield grwn vines nt ptted vines Because f the lw rate f fruiting Muscat Grd cuttings btained by the methd described abve anther methd was develped t graft bunches frm field grwn vines nt ptted vines. In this methd, tw-year-ld ptted vines were kept utdrs during winter, and at the time f bud burst, were trained t give ne r tw shts, depending n the vine's vigur. Befre the clusters n the field grwn vines had flwered, these pts were tied under the vines s that their shts culd intermingle with the field vines. The fruit cluster frm a freld vine (dnr) was grafted nt the ptted plant (hst) by apprach-grafting and after three weeks the dnr sht tip was remved. Five weeks frm grafting, the bunch was cut ff frm the dnr plant and the ptted plant with grafted bunch was transferred t the glasshuse and grwn until ripening (Fig. 4.2 and 4.3). This methd was mst cnvenient and was virtually 1% successful. Chip-b udding nt pre-rted cutting A chip-budding methd was used n pre-rted cuttings and, after transferring a drmant bud frm ne cultivar nt a pre-rted cutting f the ther cultivar, the grafted plants were kept in a glasshuse and treated accrding t the methd f Mullins and Rajasekaran (1981) and maintained as described befre. Using this methd, inflrescences

$41 \/ere btained frm the scin after the grafted bud grew (Appendix Fig.l) but they failed t develp int berries after setting.$

58 41 \/ere btained frm the scin after the grafted bud grew (Appendix Fig.l) but they failed t develp int berries after setting. Bench-grafting Cuttings f different cultiva s were grafted using the bench-grafting technique and then the graft regin and the basal end f the grafted cuttings was held at 25" C under wet sand fr tw weeks t prduce a callse at the graft area t imprve the success f grafting. After a callse was develped grafted cuttings were planted in pts and placed in the glasshuse. This methd failed t prduce any successfully grafted plants. Ã Hst vine Dnr vine I I Dnr vine cut ff + IUngrafted fru t ripens Grafted lru t r pens Analyse frurt I Analyse fruit Fig. 4.1 Transfer f grape bunches by apprach-grafting. Befre flwering, the dnr vine with bunch (A) was grafted nt the hst vine (C). The dnr vine was cut ff five weeks later (D) and the fruit was allwed t ripen n the hst vine (E). Ungrafted vines were grwn under the same cnditins as a cntrl (B).

59 lig.4.2 Glasshuse grwn Muscat Grd grapes bunch grafted nt Shiraz vine 42

60 f ig. 4.3 Glasshuse grwn Shiraz grapes bunch grafted nt a Muscat Grd vine. 43

44 4.2.3 Sample preparatin The fruits f the Muscat Grd grafted nt a Shiraz sht and Shiraz grafted nt a Muscat Grd sht, plus ungrafted Grd and Shiraz (cntrl), were used fr mnterpene analysis.

61 Sample preparatin The fruits f the Muscat Grd grafted nt a Shiraz sht and Shiraz grafted nt a Muscat Grd sht, plus ungrafted Grd and Shiraz (cntrl), were used fr mnterpene analysis. Berries were weighed after the seeds had been remved and were hmgenised in redistilled water purified in a Milli-Q apparatus (Waters, Millipre Crp.) using an Ultra-Turrax hmgeniser. The hmgenate was centrifuged at 9, g at 4'C fr l5 min and the supernatant decanted and retained. The centrifuged pellet was resuspended in redistilled water, hmgenised, centrifuged, decanted and the tw lts f supernatants were cmbined. After the additin f 2 nml each f n-ctyl ß-D-glucpyranside and n-ddecyl ß-Dglucpyranside as internal standards (IS) (1 nml fr the ungrafted Shiraz), the supernatant was passed thrugh a C-18 reversed-phase adsrbent cartridge (Sep-Pak, Waters) t adsrb free mnterpenes and their glycsidically bund frms. The C-l8 cartridge was washed three times with l5 ml f the Milli-Q water. Then the sample was eluted twice with.5 ml f redistilled ethanl and evaprated t dryness under a stream f nitrgen. The residue was redisslved in l ml f an aqueus ph 5 buffer [mixture f.1 M f aqueus citric acid slutin,.2 M f aqueus disdium hydrgen phsphate slutin and sdium azide (final cnc. :.2%)l and then incubated at37"cfr 17 h with.5 ml f an e yme slutin made up with I ml f Nvferm 12 ( Nvnrdisk, Switzerland) in 25 ml f redistilled water. The hydrlysate was diluted with redistilled water t 2 ml and then extracted cntinuusly with redistilled dichlrmethane at 55"C fr 24 h using a liquid/liquid extractr, The slvent extract was cncentrated by distillatin thrugh a clumn packed with Fenske's helices and then subjected t GC-MS analysis. 4.2,4 G as ch r m at gra p h-v-- m as s s pe ct r m etry (G C-MS) A Finnigan Mat TSQ-7 mass spectrmeter was interfaced with a Varian mdel 34 gas chrmatgraph. A DB-171 capillary clumn (J&W Scientific) f 3 m x.25 mm and frlm thickness f.25 pm, was used. The injectr and transfer line temperatures were 2'and 25'C, respectively. The cncentrated extract was injected using splitless mde venting after.5 min. The clumn was held at 5'C fr I min, prgrammed at 5"C/min t 25"C, and held at this temperature fr 2 min. The inisatin energy and emissin current fr electrn impact mde vn'ere 7 ev and.2 ma. Mass spectra were taken ver mlz in.5 sec. Mnterpenes were identified frm their mass spectra and retentin times by cmparisn with data btained previusly in this labratry (Strauss et al., 1986, 1988; Winterhalte et al.. 199).

45 The cncentratin f the identified mnterpenes wíls estimated by cmparing the peak area f the individual cmpunds relative t thse f internal standards (IS).

62 45 The cncentratin f the identified mnterpenes wíls estimated by cmparing the peak area f the individual cmpunds relative t thse f internal standards (IS). The calculatins were based n the tw assumptins that the individual mnterpenes had the same behaviur during sample preparatin and the same respnse factr in GC-MS analysis as thse f IS. Analyses f all samples frm the first experiment and sme f the grafted samples frm the secnd set were dne in duplicate' 4.3 Results Figure 4.3 shws typical GC-MS chrmatgrams f the aglycnes frm berries f Muscat Grd bunches grafted nt Shiraz shts, ungrafted Muscat Grd fruit and ungrafted Shiraz fruit frm the glasshuse experiment. The differences in amunts f the mnterpenes in the samples are evident frm these chrmatgrams. The cmpunds were identified frm their mass spectra and are shwn in Fig. 4.4 with their cncentratins in Table 4.1. The large differences in mnterpene cmpsitin between Muscat Grd and Shiraz is emphasised by the data fr the respective cntrls. Summatin f the data in Table 4.1 shwed an apprximately 5 times higher level f ttal mnterpenes in Muscat Grd cmpared with Shiraz. Mst mnterpenes were either nt detected r were present in nly trace amunts in Shiraz grapes. Imprtantly, the results shw that the Muscat Grd berries prduced the same range f mnterpene cmpunds regardless f the type f vine n which they were grafted. In fact, fr this experiment the levels f mst cmpunds determined were 2-8 times higher in the grafted fruit, pssibly relating t differences in ripeness between the grafted (26.2'Brix) and the cntrl (21. 'Brix) grapes. Shiraz fruit frm the grafted vines, which was available in nly a small amunt, was essentially unchanged in cmpsitin frm its cntrl. GC-MS analysis f Shiraz berries grafted nt Muscat Grd shts was repeated in 1995 using mre berries fr extractin. Chrmatgrams f aglycnes frm GC-MS analysis f Shiraz berries grafted nt Muscat Grd and ungrafted Shiraz berries frm this experiment are shwn in Fig.4.5. The results shwed the same pattern as the previus results with detected cmpunds in higher cncentratin in the Shiraz fruit (Table 4.1). Results f the field experiment using grafted and ungrafted (cntrl) Muscat Grd and Sultana fruits are shwn in Fig Like Shiraz, the Sultana cntained mnterpenes at levels f less than ne tenth f Muscat Grd berries. As in the glasshuse experiment, the Muscat Grd fruit had similar mnterpene prfiles in the grafted and ungrafted berries. Grafting f Sultana nt Muscat Grd vines had n influence n the mnterpene cmpsitin. The cncentratin f individual cmpunds varied during ripening f the

46 berries (Fig. 4.6), generally shwing a large increase, especially in the mre abundant cmpunds, as ttal sluble slids rse frm 1-15 t 25-28 'Brix.

63 46 berries (Fig. 4.6), generally shwing a large increase, especially in the mre abundant cmpunds, as ttal sluble slids rse frm 1-15 t 'Brix. GC-MS chrmatgrams f these samples are shwn in figures 4.7 and 4.8. Table 4.1. Cncentratins f mnterpene cmpunds identified in ripe grafted and nn-grafted berries frm the glasshuse experiments. Peak numbers are shwn in Fig Peak n. I 2 J ll t2 13 l4 l5 l6 Muscat Grd n Shiraz Muscat Grd Cntrl n.d..8 t n.d. 1.7 Levels (nml / e fresh weieht) Shiraz n Muscat Grd * n.d. n.d. n.d. n.d. n.d. n.d. n.d. n.d. trace n.d. n.d. n.d. n.d. n.d. n.d. n.d. Shiraz Cntrl trace n.d. trace n.d. n.d. n.d. n.d. trace n.d. n.d. n.d..2.5 n.d. n.d..1 Shiraz n Muscat Grd (1995) trace Ttal I trace "Brix l n.d.: nt detected trace: less than.5 + Due t the small amunt f sample the cmpunds levels were lw in the extract n.d. n.d. n.d. n.d. n.d. n.d..4.2 n.d. n.d n.d. n.d. 3.4

64 4 Shiraz cntrt 47 3 I I f 'r I tjîtf: I 5 9 IS AO Muscat Grd cntrl 1 I 2 1 ) I1 I 4 \ 7 1 Iy )UU 5 la_, fs I q 12 IS Muscat Grd grafted )O 2n > 4 \ 7 11 / t \ 6 \ 1 { ü c'ct l( î:r. t: f ig. 4.4 Typical GC-MS chrmatgrams f the aglycns frm Shiraz cntrl berries and frm fruit f Muscat Grd cntrl and Muscat Grd grafted nt Shiraz. The y-axis scale is 'relative in current' (RIC).

48 OH I ) 3 416 OH OH OH OH OH OH OH OH 7 OH 8 OH 9 r ll OH OH HO OH OH OH OH t2 l3 t4 15 t6 Fig. 4.4. Identity f l6 mnterpene cmpunds frm GC-MS ch matgrams (see Fig.

65 48 OH I ) OH OH OH OH OH OH OH OH 7 OH 8 OH 9 r ll OH OH HO OH OH OH OH t2 l3 t4 15 t6 Fig Identity f l6 mnterpene cmpunds frm GC-MS ch matgrams (see Fig. l): 1: Furan linall xide, 2: Linall, J: Htrienl, 4: Pyran linall xide ismer, 5: cr- Terpinel, ó: Pyran linall xide ismer, Z: Nerl, 8: Geranil,9:2,6-Dimethylcta-3,7- diene-2,6-dil,i:3,7-dimethylcta-1,6-diene-3,5-dil,1122,6-dimetþlcta-1,7-diene- 3,6-dil, I2:Z-2,6-Dimethylcta-2,7-diene-1,6-dil,I3:E-2,6-Dimethylcta-2,7-diene- 1,6-dil, I4z 3,7-Dimethylcta-2-ene-1,7-dil, 15: p-menth-l-ene-6,8-dil, 1ó: p-menthl-ene- 7,8-dil.

66 s { I rs Muscat cntrl 49! I 2 l? I I 4 6? 11! \ l 12 t{ f-, t., 5 a1 Shiraz cntrl ) 12 IS t6 l r.ð l7(tt: 15 S S Shiraz grafted I 1i * I 9 I tl Figure 4.5 Typical GC-MS chrmatgrams f the aglycns frm Muscat Grd cntrl berries and frm fruit f Shiraz cntrl and Shiraz grafted nt Muscat Grd. The y-axis scale is 'relative in current' (RIC).

67 tz -c tj).) 8 4 E (õ 9^ c 82 c.9 Er c () c (J 12 tt q) c e8 q ( ) 54 c A I B D 'Brix ABCD Pre-ripe 15., 14.4, 11.2, 1.4 El Early ripe 18., 17.8, 17.8, 17.8 fl Ripe 28., 25., 26.2,28. -",Jl JI rjlr* I I I I 't5 16 Peak number Figure 4.6. Histgrams f mnterpene aglycnes in berries (shwn in Fig. 4.2) frm the field experiment f Sultana ungrafted (A), Sultana grafted n Muscat Grd (B), Muscat Grd ungrafted (C) and Muscat Grd grafted n Sultana (D), at three ripening stages.

51 4.4 Discussin Grape berries cntain a cmplex mixture f secndary metablites which cntribute t the a ma and flavur f the fruit and prvide the chemical basis fr varietal characteristics f the grapes.

68 Discussin Grape berries cntain a cmplex mixture f secndary metablites which cntribute t the a ma and flavur f the fruit and prvide the chemical basis fr varietal characteristics f the grapes. Mnterpene cmpunds are particularly abundant in flral grape varieties such as Muscat Grd and mst f these cmpunds are present as glycsides (Gunata et al., 1985b, V/illiams et al., 1982a, Wilsn et al., 1934). In this present study, the ttal levels f mnterpene glycsides in Muscat Grd fruit were apprximately 5 times that f the nn-flral varieties. Indeed, many f the mnterpene glycsides present in the Muscat Grd fruit were nt detected in berries f the nn-flral varieties. In the grafting experiments, the same spectrum f mnterpene glycsides was present in Muscat Grd fruit, irrespective f whether the fruit develped n its wn vine r fllwing bunch grafting nt Shiraz vines (Table 4.I, Fig. a.3) r Sultana vines (Fig. 4.5, 4.6). The amunt f each f these mnterpene glycsides present in grafted Muscat Grd fruit was equal t r even greater than that in ungrafted fruit. The quantitative differences bserved, prbably relate t differences in ripeness f the fruit, as these cmpunds are knwn t cntinue t accumulate as the fruit ripens (Gunata et al., 1985b, Wilsn et al., 1984). Of equal imprtance, grafting f fruit f the nn-flral varieties Sultana and Shiraz nt Muscat Grd vines did nt increase the lw level f mnterpene glycsides in these grapes, nr did it induce accumulatin f any mnterpene cmpunds different frm thse f ungrafted berries (Table 4. l, Fig. 4.3 and Fig. 4.6). These results cnfirm the earlier deductins made by Winkler (Winkler et al., 1974) and suggest that the ability t accumulate mnterpene glycsides resides within grape berries and is independent f the rest f the vine. As with many secndary cmpuncls, mnterpene glycsides are als present in grapevine leaves (Gunata et al., 1986; Bravd et al., 1989). These cmpunds ma),be synthesised in the leaves and transprted t the berries r they may be fbrmed in the berries themselves frm precursr mlecules imprted via the phlem sap. The results indicate that the mnterpene glycsides are actually synthesised in the berries. This wuld suggest that the enzymes required t synthesise these cmpunds are present in the berries and that sme r all f these enzymes are absent r inactive in fruit f nn-flral varieties. It is, hwever. still pssible that the mnterpene glycsides are synthesised elsewhere in the grapevine and imprted int the fruit. If this is s, the Muscat Grd berries must have mechanisms t selectively imprt and accumulate these cmpunds while these mechanisms are presumabl - nt active in grapes f the nn-flral varieties. The grafting experiments d indicate that the inability f the nn-flral grapes t accumulate mnterpene glycsides des nt result frm a lack f these cmpunds r their

52 precgrsrs in the phlem sap circulating arund the vine, since fruit f Muscat Grd grafted nt these varieties still prduced the nrmal levels f mnterpene glycsides.

69 52 precgrsrs in the phlem sap circulating arund the vine, since fruit f Muscat Grd grafted nt these varieties still prduced the nrmal levels f mnterpene glycsides. The dependence n the rest f plant fr synthesis f metablites in fruit may be less than is currently accepted. 4.5 Cnclusin These experiments reprted in this chapter have demnstrated thrugh fruit cmpsitin analysis that bisynthesis f mnterpene glycside flavur cmpunds in the grape berry is independent f ther parts f the vine.

70 Muscat Grd ungr ltcd (cntrl) / II rtc Sullane grafted n Musc t Grd aaa rst c Sulaana ungr flcd (cntrl) lsl l fig.4.7 Typical GC-MS chrmatgrams f the aglycns frm Sultana cntrl berries and frm fruit f Muscat Grd cntrl and Muscat Grd grafted nt Sultana. The y-axis scale is 'relative in current' (RIC).

71 Sullenr ungreftcd (cntrl) I tlc lsf Muscat Grd greftcd n Sult n Muscat Grd ungrafted (cntrl) I I l 't 1 1 Fig. 4.8 Typical GC-MS chrmatgrams f the aglycns frm Muscat grd cntrl berries and frm fruit f Sultana cntrl and Sultana grafted nt Muscat Grd. The y- æris scale is 'relative in current' (RIC).

55 Chapter 5 Occurrence f anthcyanin pigments in berries f the white cultivar Muscat Grd Blanc Cntents 5.1 Intrductin... 5.2 Materials and methds... 5.2.1 Plant material 5.

72 55 Chapter 5 Occurrence f anthcyanin pigments in berries f the white cultivar Muscat Grd Blanc Cntents 5.1 Intrductin Materials and methds Plant material ExÛ ctin HPLC analysis GC-MS analysis Results Discussin 5.5 Cnclusin Page

56 5.1 Intrductin The clur f berries is an bvius distinguishing factr between grape cultivars and clnes and, in clured types, is mainly caused by differences in type and amunt f anthcyanins in berry

73 Intrductin The clur f berries is an bvius distinguishing factr between grape cultivars and clnes and, in clured types, is mainly caused by differences in type and amunt f anthcyanins in berry skin. rwulf and Nagel (197S) separated and quantified abut twenty anthcyanin cmpunds by high-perfrmance liquid chrmatgraphy (HPLC) analysis. This technique is still the standard and has been used by many t gain infrmatin abut the nature f the clur differences between cultivars (Baker and Timberlake, 1985,'Wenzel et al., 1987, Mattivi et al., 1989, Craver et al., 1994) and berry ripening stage (Smers, 1976, Crippen and Mrrisn, 1986, Ferna'ndez-L'pez et al., 1992). Recently, Watanabe et al. (1991) used it t classiff grape cultivars int five types based n their 3- glucsides: (a) mstly cyanidin (b) penidin plus cyanidin (c) delphinidin plus cyanidin (d) fur tgether - cyanidin, penidin, delphinidin and petunidin and (e) mainly malvidin. Types (a) and (b) have skin clurs with high hue angle (rse) and type (e) have lw hue angle (blue-black). The muscat grup f grape cultivars is knwn fr its large array f berry clurs. Cllectively, muscats are characterised by similar leaf mrphlgy and a distinctive and widely-appreciated berry flavur and arma (Galet, 1958, 1962), attributable largely t their cntent f a number f xygenated mnterpene cmpunds (Ribereau-Gayn et al., 1975). The mst widely planted member f the grup, cv. Muscat Grd Blanc, has been the subject f much physilgical and bichemical research; hwever this research has nt included a studv f anthcyanins within its berry skin-nt surprising cnsidering that it is a white cultivar. During the studies n the bisynthesis f flavur cmpunds in berries f Muscat Grd (Chapter 4) it was frrst nticed that very ripe Muscat Grd berries n wnrted vines r n bunches which were grafted nt Shiraz vine shts develped a pink r rse clratin. The latter benres rvere fed by leaves frm Shiraz grapevine which has clured berries. The questin arse as t whether this cluratin f nrmally white berries was due t the effèct f leaves and whether the precursrs f anthcyanins were translcated frm leaves t the berries. T gain an understanding f this phenmenn, the cmpnents f skin clur, the levels f plyphenlic cmpunds and juice arma were studied. Muscat Grd berries at different stages f ripening and develpment were used and the clur pigment f the

74 57 berries that accumulates ccasinally in ver-ripe berries was extracted, analysed and identified. 5.2 Materials and methds Plant materials Muscat Grd grape bunches were picked frm tw lcatins: the Waite experimental vineyard, which were wn-rted; and Penfld's vineyard at Waikerie, Riverland S.4., which were grafted n t cv. Matar (syn. Muvedre), and als frm bunches grafted nt Shiraz vines and grwn in the glasshuse (methd in chapter 4.2). Tw types f bunches were taken (l) thse with berries clured pink r rse (Fig. 5.1) and (if) thse withut any pigmented berries. Frm abut ten bunches f each type, a pl f 221 berries were picked at late ripening (E-L stage 39, Cmbe, 1995) and mixed Extractin All benies were individually weighed, berry length and tw diameters measured t enable skin surface area t be calculated, and the ttal sluble slids (TSS) f the juice was determined as "Brix after peeling the skin. The skin f each berry was placed in a test tube cntaining 5% ethanl, acidified t phcl with HCl, the tubes mixed every hur fr 4 hurs and then left t stand vernight. After centrifuging at 4 rpm (15 g) fr 1 minutes, the clear supernatants were used fr spectrphtmetric measurements (using a Varian/DMS 2 UV Visible spectrphtmeter) at 52 and 28 nm. and anthcyanin and ttal plyphenl cntent f the skin calculated per berry, and per gram fresh weight f skin. Shiraz berries grafted nt Muscat Grd and nn-grafted Shiraz berries were extracted fr anthcyanin pigments and spectrphtmetric measurements were prefrmed in the same way HPLC analysis HPLC analysis was dne in triplicate n the extracts f each type f samples after being cncentrated b_'" evaprating the ethanl under nitrgen flw at 35'C. HPLC analysis fr berries frm wn-rted vines, frm vines grafted n t cv. Matar (syn. Muvedre), and als frm bunches grafted nt Shiraz vines (glasshuse experiment) were dne separately.

The fllwing elutin prtcl was used: A;2Y,-1 minutes, A; 2/û 35/,1-15miuutes, A;35%-7y,15-3minutesandA;7%-Iy,3-35minutes.

75 58 The separatin f anthcyanins was made n a C18 clumn (LC.I. ODS 2; 5 pm, 25 mm x 4.6 mm) with preclumn (Cl8 cartridge 1 mm). Elutin was with tw slvents, slvent gradients were varying prprtins f (A): ly methanl, and (B) : O-6 V aqueus perchlric acid, ph 1.3 t 1.4. The flw rate was 1.5 mu min., and run time; 35 minutes. The fllwing elutin prtcl was used: A;2Y,-1 minutes, A; 2/û 35/,1-15miuutes, A;35%-7y,15-3minutesandA;7%-Iy,3-35minutes. The clumn was then washed, first fr 4-5 min with A-1% and then fr 5-6 min with A-2% thus restring it t its riginal cnditin befre each run. The eluant was mnitred at 52 nm using a dide array detectr. The peaks f the chrmatgrams were identified by cmparing retentin times with thse frm knwn cmpnents in Shiraz grape juice, Shiraz wine and authentic cyanidin- 3-glucside. Quantificatin amunts f cyanidin-3 -glucside. was based n peak areas f unknwns against that f knwn GC-MS analysis Mnterpene cmpunds were determined by GC-MS analysis f 2-25 g f berries accrding t the methd described in chapter fur (4.2.4). Fig. 5.1 Muscat Grd berries shwing pink clur

59 5.3 Results Un-pigmented berries f Muscat Grd, at 18 Brix, shwed n detectable level f anthcyanins in berry skin, but the sample f pigmented berries (which had a high Brix level f 28.

The cmpunds cntributing t the clur were predminated by cyanidin-3-glucside, plus minr amunts f penidin- and delphinidin-3-glucsides. The ch matgrams shwn in Fig. 5.

76 Results Un-pigmented berries f Muscat Grd, at 18 Brix, shwed n detectable level f anthcyanins in berry skin, but the sample f pigmented berries (which had a high Brix level f 28.5") had a significant thugh lw level f ttal anthcyanins, tgether with a high cntent f ttal phenlics (Table 5.1). The cmpunds cntributing t the clur were predminated by cyanidin-3-glucside, plus minr amunts f penidin- and delphinidin-3-glucsides. The ch matgrams shwn in Fig. 5.2 present the separatin prfile f anthcyanins by the methd described abve with the Muscat Grd berry skin extracts. The majr peak was identified by cmparisn with cyanidin-3-glucside as a standard. The remaining peaks crrespnding t ther anthcyanins, cntributed less than l% f the ttal pigments in the berries (Table 5.1). Table 5.1 Ttal plyphenlics and anthcyanins in the skin f Muscat Grd Blanc in tw grups: white (nn-pigmented) and pigmented (shwing a light rse r pink clur)' The fruits were frm vines grwn in the field. White Pismented Juice Brix Cncentratin mgikg skin fr. wt. Ttal phenlics (GAE*) Ttal anthcyanins Anthcyanin %n f ttal Delphinidin-3 - glucside Cyanidin-3-glucside Petunidin-3 - gl ucside Penidin-3-glucside Malvidin-3-glucside p-cumarate esters Acetate esters 93 n.d. r57 ** 7.7 Ttal 1 n.d.: nt detected +GAE: gallic acid equivalents ** Values are means f three: standard elrrs were +12-l4Y n.d. n.d. n.d. n.d. n.d. n.d. n.d n.d. 3.6 n.d. n.d. n.d. l t

6 The absrbance, fresh weight and Brix f an unselected ppulatin f 221 individual berries harvested at randm frm ten bunches frm bth sites a e shwn in Fig' 5.3.

77 6 The absrbance, fresh weight and Brix f an unselected ppulatin f 221 individual berries harvested at randm frm ten bunches frm bth sites a e shwn in Fig' 5.3. The absrbances at 52 nm f skin extracts shw that pigment nly ccurred in berries with Brix abve 2a Gig.5.3 a); thse with Brix values f 24 r less gave lw (<.18 units) absrbance readings which were shwn by spectrphtmeter scanning nt t be due t specific 52 nm absrbance. Skin extract absrbance at 52 nm is cmpared with berry weight in Fig. 5.3 b. It is clear that pigmentatin ccurred in a wide array f berry sizes, but especially in smaller berries. Fig. 5.3c shws that small berries have higher Brix' in fact, berries f < 2.3 g all had juice with Brix > 25. N study was made f the relatinship f pigmentatin t seed develpment. Regarding the absrbance values at 52 nm, the highest ttal anthcyanin cntent f an individual Muscat Grd ber{ recrded in this experiment was 33.9 t 1.6 Pg per berry. By cmparisn, the anthcyanin cntent f Shiraz berries grafted nt Muscat Grd shts and nn-grafted Shiraz berries were I.33 mglg berry ("Brix : 23) and 1.26 mglg berry' ("Brix : 2) respectively. Examinatin f berry surfaces under the mic scpe shwed that pigmented skin had a mix f clured and unclured cells (data nt shwn) and suggested that greater pigmentatin was due mre t a higher prprtin f clured cells than t a deeper clur per cell. Values fr the cncentratin f mnterpenes in hmgenates f whle berries r in the berry juice are shwn in Table 5.2. The increasing values f TSS inthe three samples used fr whle berry extractin shwed a cr elated increase in all 16 mnterpenes, in many cases shwing a ten-fld range. Hwever their relative prprtins remained generally similar.

61.? c t.a b.l ra -6.2 a. 12. a.1.1. I.3.2 È 1 2...6. -ó. 2 : '- i I 3,.!.. 1. 2. 1. Xlôuced Fig. 5.

78 61.? c t.a b.l ra -6.2 a. 12. a.1.1. I.3.2 È ó. 2 : '- i I 3,.! Xlôuced Fig. 5.2 HPLC chrmatgrams f anthcyanin pigments prfile f the extracts fr Muscat Grd berries skin (a) frm wn-rted vines, (b), frm bunches grafted nt Shiraz vines. chrmatgram (c) shws authentic cyanidin-3-glucside. Chrmatgram (b) shwed a faster elutin prfile due t different instrument cnditins.

79 E c ôl tf a 62 G ọ J tr (ú at, '.'j ld". 1 t Ëflx ' ì... "'a' a at '....t...tr rl.d Ë1.Â: " -' l a t^ -]l l.ì;...rrl" I^, b E c ôt at 1.5 a) tr 6 lt at l '.{... t...a?t' ï:.,...âfj.é i t 345 f :ri t d. î,. 678 Berry Weight g 4 c x fd 3 25 dl 2.t t 't Berry Welght Fig. 5.3 Relatinship fr 221 Muscat Grd berries f: (a) ttal sluble slids fjuice ('Brix) and absrbance f skin exûacts at 52 nm, (b) berry weight and absrbance f skin extracts at 52 nm and (c) berry weight and Brix.

63 Table 5.2 Cncentratins f mnterpene cmpunds in whle nn-pigmented (White) and rse- r pink-clured (Clur) Muscat Grd berries and white juice' Cmpunds are numbered as in Figure 4.

80 63 Table 5.2 Cncentratins f mnterpene cmpunds in whle nn-pigmented (White) and rse- r pink-clured (Clur) Muscat Grd berries and white juice' Cmpunds are numbered as in Figure 4.3 f chapter fur' Whle berries White Clur Clur Juice White "Brix nmvet nmvmll N Cmpunds I Furan linall xide, Linall Htrienl r 4 Pyran linall xide ismer.58 2.t r6 5 a-terpinel Pyran linall xide ismer I 7 Nerl Geranil 1.8 r t ,6-Dimethylcta-3,7-diene-2,6-dil l 3,7-Dimethylcta-I.6-diene-3,5-dil I 2,6-Dimethylcta-1,7-diene-3,6-dil Z-2,6-Dimethylcta-2.7-diene- 1.6-dil.44 t I 3 E-2,6-Dimethylcta-2.7-diene- 1.6-dil l4 3,7-Dimethylcta-2-ene-1.7-dil r6 15 p-menth- 1 -ene-6,8-dil nd 16 p-menth-1-ene- 7.8-dil Ttal tt I t nmvg f berry hmgenate +nml/ ml fjuice

64 5.4 Discussin The first bservatins f Muscat Grd pigmentatin were made during the experiments described in Chapter fur in which bunches f this cultivar were grafted nt Shiraz shts t study

81 Discussin The first bservatins f Muscat Grd pigmentatin were made during the experiments described in Chapter fur in which bunches f this cultivar were grafted nt Shiraz shts t study bisynthesis f flavur cmpunds. Samples frm that experiment shwed that the anthcyanin prfile fr Muscat Grd berries grafted nt Shiraz sht was the same as nngrafted berries (Fig. 5.2). Als Shiraz berries develped full berry clur (1.33 gm/ g berrl') when grafted n t Muscat Grd shts cmpared with the nn-grafted berries (1.26 mgl g berry). The slightly higher value fr nn-grafted berries was prbably due t smaller berry in size with higher "Brix. Similarly, there were n differences between the pigment prfile f berries frm wn-rted Muscat Grd and berries frm vines grwn n the rts f the black cultivar Matar (chrmatgram is nt shwn). All f these results supprt the view that berries are independent f leaves and rtstck fr synthesis f anthcyanins as well as ther secndary metablites, as was suggested earlier in Chapter 4. The results shw that light pink r rse pigmentatin is able t develp in the nrmally white (un-pigmented) berries f cv Muscat Grd. The amunts are small - in ne test we fund a cncentratin f ttal anthcyanins f 8 mg per kg skin fresh weight which is lwer than the smallest cncentratin f the range (varying frm 28 t 685 mg/kg) fund by Craver et al. (1994)in skins f 22 clured muscat cultivars. Distinctive features f berry clratin in Grd are its late develpment---clured berries had juice f 24 "Brix r higher and the ht cnditins under which it develped. This cntrasts with sme malvidin-3-glucside-accumulating, black cultivars such as Cabemet Sauvignn and Shiraz in which cluring develps at abut l Brix (e.g. Rgger et al., 1986) and is mre intense if the weather is cl (Kliewer, 197c). It is pssible that the Muscat Grd pigmentatin was augmented by the lateness f its develpment in autumn when nights were cler. Cyanidin-3-glucside was the predminant anthcyanin (Fig. 5.2), making up 94 per cent f the ttal (Table 5.1). Accrding t the classificatin made by Craver et al. (1994), this result puts it in the same range as th ee ther muscat cultivars - Perle de Csaba ruge, Siegerrebe and Marangs. These authrs als shwed that fur ther cultivars, which had abut 7 per cent f the ttal as cyanidin-3-glucside, belnged t the general grup f rse-clured, small-seeded muscats; these are clured variants f 'Muscat a petit grains' (syn., Muscat de Frntignan, Muscat Canelli, r, in Australia, Frntignac). Of the 22 clured Muscats examined by Craver et al. (1994), fur had a predminance f penidin-3-glucside while, in half f them, malvidin-3-glucside predminated; many f these tw grups f cultivars have blue r black berries. They nted that thse cultivars with a high level f cyanidin-3-glucside in their berry skin als

65 have a high level f mnterpenes in their juice and are cmmensurately armatic (in the style f the small-seeded muscat grup). Values f mnterpenes in juice f Muscat Grd (clumn 4 f Table 5.

82 65 have a high level f mnterpenes in their juice and are cmmensurately armatic (in the style f the small-seeded muscat grup). Values f mnterpenes in juice f Muscat Grd (clumn 4 f Table 5.2), while difficult t cmpare with the data f Craver et al. (1994)(which lack units), nevertheless shw similarities with, fr instance, cv. Muscat de Frntignan ruge in the predminance f dil N. 9 (: their DDI) plus high levels f linall (Muscat Grd als had high levels f terpene N.3, htrienl). The higher cncentratin f ttal mnterpenes in whle berries at high "Brix (Table 5.2) is reminiscent f the same phenmenn shwn by ttal phenlics and anthcyanins in skin (Table 5.1 and Fig.5.3 a). Althugh different bisynthetic pathways wuld be invlved, Craver et al. (1994) suggested that the relatinship between juice mnterpenes and skin anthcyanins warranted further investigatin. Any cnnectin t seed develpment appears unlikely in view f the difference in seed size f Muscat Grd and the small-seeded muscats, but this aspect als needs explratin. The cmpund cyanidin-3-glucside is regarded by Rgger et al. (1986) as a primitive pigment in the bisynthetic pathway f anthcyanins (alng with pelargndin-3- glucside, which des nt ccur in Vitis vinifera). This less stable and primitive grape pigment culminates in the stable pigments pendin-3-glucside thrugh hydrxylating and methylating enzymes. Only penidin- and delphinidin-3-glucsides were fund in Muscat Grd and these at nly a few per cent. Thus the pathway in Muscat Grd, under the specific cnditins f very high levels f TSS and warm weather during berry develpment, enabled bisynthesis f cyanidin (r its glucside) frm its flavanne, but it seems that the further develpment and transfrmatin fails t functin prperly and the enzymes fr further metablism (flavnid 3'-5'-hydrxylase and methyl transferase) were relatively inactive. It is interesting that Watanabe et al. (1992) fund, frm csses utilising ne parent with white fruit clur and ne with clured fruit, that Muscat Grd has a latent but dminant gene fr methylatin f the B ring f anthcyanins. The unusual and specific cnditins under which anthcyanin synthesis has been fund t be triggered in Muscat Grd berries might be useful fr future studies f anthcyanin bisynthesis in grape berries. The results presented in this chapter supprt the idea that the pigmentatin which ccurs in Muscat Grd berries is genetically cntrlled by the berry itself. This phenmenn ccurs nl1, in the ver-ripe berries and there is n relatinship between grafting practice and this type f berry pigmentatin. Feeding frm leaves f vines with clured fruit was nt the cause f the pigmentatin, and the clur f Shiraz berries was nt significantly decreased when grwn n shts f a white-fruited cultivar. As was cncluded frm the experiments f flavur cmpunds in chapter fur, the benies appear t be independent f leaves in the synthesis f anthcyanin pigments.

bþ 5.5 Cnclusin Pink pigmentatin can develp in the skin f the nrmally white berries f cv Muscat Grd. The principal causal anthcyanin is cyanidin-3-glucside.

83 bþ 5.5 Cnclusin Pink pigmentatin can develp in the skin f the nrmally white berries f cv Muscat Grd. The principal causal anthcyanin is cyanidin-3-glucside. The clur develps nly in berries with juice "Brix > 24 mst f such berries being small. Their cmpsitin f anthcyanin and mnterpene resembles that f berries f 'Muscat a petits grains' cultivars. Bunch grafting shwed that leaves f black berried cultivar (Shiraz) did nt alter the phenmenn f pigmentatin in Muscat Grd berries. Similarly, the anthcyanin pigments f Shiraz berries were nt altered by being grafted nt shts f Muscat Grd. As was cncluded frm the lack f differences in mnterpene cmpsitin in Chapter fur, berries appeff t be independent f leaves in the synthesis f anthcyanin pigments as well.

67 Chapter 6 Assimilate transprt in the grapevine ; carbhydrates, amin acids and ptassium Cntents Page 6.1 6.2 Intrductin...-... Materials and methds... -... 6.2.1 Plant materials... 6.2.2 Other materials.

84 67 Chapter 6 Assimilate transprt in the grapevine ; carbhydrates, amin acids and ptassium Cntents Page Intrductin Materials and methds Plant materials Other materials Exudate cllectin techniques Exudate cllectinfrm the peduncle.. Exudate clle ctin frm the leaves Exudate cllectin fr m excis ed peduncle.. Girdling... Cntrls f bufer slutinwithut EDTA. S ampling t e s tabli sh diurnal patterns Sugar analysis... Amin acid anaiysis Results Mineral elements measurement Grapevine phlem sap cmpsitin... Seasnal changes in the phlem sap cmpsitin Validatin f the sap cllecting technique, Phlem sap diurnal patterns... Pattern f sucrse in exudøles during a day (befre veraisn)... Dalt snd night diurnal patlerns (after veraisn)... Experimentl... Experiment II Grapevine leaf exudates Discussin Validatin f the sap cllecting methd Grapevine exudate cmpsitin; cmparisn with tissue cmpsitin ;.;.; ;;-," n". "^,"t"," u.."*"i",t";t;;;";; ;;;t Seasnal changes in exudate flux Diurnal chances in exudatin Grapevine leaf exudate Cnclusin...

68 6.1 Intrductin Phtsynthesis in the mature leaves f higher plants prvides the metablites fr the sink rgans and maintains the assimilates required fr plant grwth and develpment.

85 Intrductin Phtsynthesis in the mature leaves f higher plants prvides the metablites fr the sink rgans and maintains the assimilates required fr plant grwth and develpment. Abut 8% f phtsynthetically fixed carbn is used fr carbhydrate synthesis (cellulse, starch, sucrse in higher plants, Grdnet al., 198) and the remainder is used in synthesis f amin acids, rganic acids, and ther primary and secndary metablites (Huber et al. 1992). The mature leaves prduce assimilates such as sucrse and certain amin acids frm inrganic carbn and nitrgen, which then can be transprted t and utilised in ther parts f the plant. Phlem tissue cnducts the slute transprt frm surce tissues t sinks. Many inrganic ins and rganic cmpunds such as sugars, amin acids, rganic acids, and plant hrmnes as well as macrmlecules such as nucleic acids and prteins have been detected in phlem exudates cllected frm varius plants (Mss and Rasmussen, 1969; Hall and Baker,1972;Ziegler, 1975; Fndy and Geiger, 1977; Giaquinta and Geiger, 1977, Barlw and Randlph, 1978; Kawabe et al.' 198; Delrt, l98l;hayashi and Chin, 1986; Chin etal., 1991; Girusse, l99l: Jeschke and Pate. 1995;Nakamura et al. 1995; Wijayanti et al., 1995), and frm Vitis vinifera by Glad et al. (1992a). The cncentratin f dry matter in phlem sap varies but values f 25/" w/v have been measured (Pate, 1975). The cncentratin f ttal slutes inphlem sap f trees is usually.4 t.6 Mlar (Zimmermann, 196; Nbel, 1991). The types and cncentratins f slutes exhibit daily and seasnal variatins, and als depend n the tissues that the phlem slutin is flwing tward r away frm (Nbel, l99l). The amunts f phlem sap assimilates allcated t fruits ver the whle perid f their develpment are imprtant factrs affecting the quality and quantity f the fruit. A large amunt f glucse and fructse. and smewhat lesser amunts f tartrate, malate, ptassium and a wide variety' f ther cmpunds accumulate in grape berries during berry develpment and ripening (Kiiewer. 1967a; Harris et al., 1968; Peynaud and Ribéreau- Gayn, 1971; Cmbe and Matile, 198; Cmbe, 1975,1989, 1992). The mature and well develped leaves n the sht are the main surces f sugar supplied t the berries and t strage tissues (Candlfi-Vascncels and Kblet, 199). The surplus energy frm phtsynthesis in surce leaves can be stred temprarily in the chlrplast either as carbhydrate r as lipid depending n temperature (Buttrse and Hale, l97l). Starch is an imprtant end prduct f phtsynthesis that serves as a temprary reserve frm f

86 reduced ca bn stred in leaves (Huber et al., 1992), and als in rts (Hunter et al., 1995). The grape berries have a directinal influence n translcatin in grapevines and many studies shwed a predminant mvement f phtsynthates twards the grape berries nce a certain fruit size is attained (Hale and Weaver,1962; Quinlan and Weaver,197; Kriedemann, 1977; Hunter and Visser, 1988 Cmbe, 1989; Mtmra, 1993). Remval f grapes frm the vine, and als girdling f the phlem, results in rapid decrease in the phtsynthesis rate and an increase in the stmatal diffusive resistance, but tpping des nt, thus demnstrating the rle f berries as a strng sink in activating transprt frm leaves and suggesting that the grwing pint f the sht at this stage f its develpment is nt as strng a cmpetitr fr phtsynthate metablites (HfÌicker, 1978). Sucrse is the majr carbhydrate prduced in the leaves and transprted int grape berries (Swansn and El-Shishini, 1958; Kblet, 1977) and accunts fr 7% f carbhydrates f grapevine phlem sap (Glad et al., 1992a). Sucrse translcating in the grapevine culd riginate frm any current phtsynthesis r any reserve tissues in the leaf r frm stem, wd, rts r ther perennial parts f the vine (Matsui et al., 1979; Cmbe, 1989; Williams and Biscay, 1991; Candlfi-Vascncels et a1.,1994). The levels fr hexses were reprted t be lw in phlem sap exudates f grapevine analysed by Glad et al., (1992a). Besides sucrse, amin acids and amides are majr cnstituents f the grapevine sieve tube sap (Pate, 198; Glad et a1.,7992a; Rubelakis-Angelakis and Kliewer, 1992). The transprt f nitrgen in the phlem seems t fllw the bulk flw thery fr carbhydrate mvement (Bidwell 1979). Glad et al. (1992a} reprted the nitrgen slute cmpsitin f grapevine (Vitis vinifera L.), cv. Pint Nir phlem sap, glutamine being the majr (6%) transprted frm f nitrgen and prline the secnd (1%) at flwering. Ptassium is the mst prevalent catin in ripe grape berries and accumulates in cnsiderable amunts after veraisn and during ripening f the berry (Hale, 1977; Cmbe, 1987a Iland and Cmbe. 1988; Mnisn and ldi, 199; Williams and Biscay, 1991; Guatiérez-Guanda and Mrrisn. 1992: Creasy et al., 1993; Bselli et al., 1995). The phlem sap is the surce f metablites accumulated int the grape berries during berry grwth and develpment. mst extensively after the inceptin f ripening. Yet little is knwn abut the cmpsitin f grapevine phlem sap. In this study, phlem sap was cllected frm the peduncle f different grapevine cultivars and analysed fr sugars, amin acids and ptassium in rder t reveal the grapevine phlem sap cmpsitin and the seasnal and diurnal changes in the levels f metablites present in the phlem sap and als t check the validity f the technique fr grapevine phlem sap cllectin. 69

7 6.2 Materials and methds 6.2.1 Plant materials Mature grapevines f Vitis vinifera L.

87 7 6.2 Materials and methds Plant materials Mature grapevines f Vitis vinifera L. cultivars; Muscat Grd Blanc, Sultana, and Shiraz were used frm the Waite Institute experimental vineyards at Glen Osmnd, Suth Australia. The samples were cllected at three times during berry grwth and develpment being befre veraisn (three t fur weeks after fruit set), at veraisn, and at ripening time. Samples als were cllected frm Muscat Grd frm early veraisn up t ripening time Other materials HEPES {N-[2-hydrxyethyl] piperazine-n'[2-ethansulfnic acid]] buffer slutin was made f 1 mm HEPES disslved in distilled water and EDTA (ethylenediaminetetraacetic acid) was added t the slutin at the cncentratin f 1 mm (final slutin 1 mm f each cmpund) and then the ph was adjusted t 7.5 using sdium hydrxide (1. N). Eppendrf tubes filled with 1.5 ml f this slutin were used fr phlem sap cllectin Exudate cllectin techniques The facilitated exudatin technique (King andzeevaart,1974; Hansn and Chen 1985; Glad et al.,1992a) was used t cllect phlem sap frm severed peduncle laterals and frm the cut end f leaf petiles. Exudate cllectin frm the peduncle The phlem exudate frm peduncle was cllected in situ. The first lateral n the peduncle was excised under a drp f buffer slutin and immediately the cut end was dipped int an Eppendrf tube filled with buffer slutin (Fig. 6.1). The tube was fixed t (Australia) Pty. Ltd.]. This stay n the bunch using blue-tack pliable adhesive [Bstik, material was fund t have n harmful effect n the plant tissue. The cut end f the peduncle laterals were rinsed with the buffer slutin fr abut l minutes and then the tubes were replaced with tubes cntaining fresh buffer slutin which were left t stay fr all f the sampling perid. The sampling time curse varied between experiments. Sampling n each day started frm 9 am until 5 pm particularly when a large number f tubes were cllected. The tubes were transfered int ice after being remved frm the vines and then transfered t -2 "C until chemical analysis.

$71 Fruits frm cut laterals were cllected and weighed, and mean berry weight was determined. "Brix fjuice frm a pl f fruits was measured with a hand refractmeter.$

88 71 Fruits frm cut laterals were cllected and weighed, and mean berry weight was determined. "Brix fjuice frm a pl f fruits was measured with a hand refractmeter. Exudate cllectinfrm the leaves Mature, healthy leaves frm ndes 4 t 6 n the sht were used fr phlem exudate cllectin. The leaves were severed frm the vines at the basal end f the petile clse t the cane under a drp f buffer slutin. They were immediately transferred int buffer slutin befre taking them t a cntrlled envirnment fr exudate cllectin. Petiles were placed in the tubes cntaining buffer slutin (1 mm, HEPES buffer cntaining EDTA as used fr sap cllectin frm bunch laterals), and phlem exudate f the leaves was cllected (King andzeevaart,lgt4). The leaves were fixed t stand in a bx with a tp lid, made f clear perspex (Fig. 6.2). The leaves were sprayed with distilled water t maintain high humidity and s minimise transpiratin. The exudate samples cllected were kept at -2'C until analysis. The first year sampling (1994) was dne under lights, with a light intensity f 4 pe m-2s-' and the temperature near the leaves was between "C. The secnd year sampling (1995) was dne in the dark with ambient temperatures f "C. Exudate cllectin frm excised peduncle Exudate was cllected frm the grape peduncle after being cut ff the vine t check the cntributin made by this tissue. The first bunch lateral n a grape bunch was cut under a drp f buffer slutin (as described abve), then a sap-cllecting tube was fixed n the bunch. The whle bunch was severed frm the vine and all berries, laterals and mre than half f the peduncle were remved. The remaining bunch stem with the tube attached t it was fixed n a hlder (Fig. 6.3) and left n the bench fr 6 hurs and the exudate cllected and analysed fr metablites. Exudate was cllected frm three replicate bunchstems with a mean fresh weight f I.5 gr. Girdling T prvide further assessment f phlem sap flux, girdling was dne n the sht abve and belw the bunch (Fig. 6.a) thus excluding the fruit bunch frm phlem flux and stpping phlem sap transprt int the berries. The leaf ppsite the bunch was als remved and cllectins were made frm the girdled bunches as described abve and analysed fr metablites. Cntrls f buffer slutin witltut EDTA A series f cntrl samples was cllected fr exudatin by excluding EDTA frm the buffer slutin. The methd fr cllecting sap samples was the same as described abve.

89 72 Fig. 6.1 Phlem sap cllectin frm a peduncle lateral Fig. 6.2 Phler-n sap cllectin frm cletached leaves.

90 73 BL ET IIB Fig.6.3 Exudate cllectin frmpeduncletissue ff the vine. Ped: peduncle; BL: bunch lateral; ET: Eppendrf tube; HB: hlding blck. G Ped. L Fig. 6.4 Girdling the cane abve and belw the fruit bunch in rder t stp phlem sap flux t the bunch, G: girdled area; Ped. L: peduncle lateral frm which sap was cllected.

91 74 Sømpling t estøblish díurnal patterns Samples were cllected shrtly befre veraisn, after veraisn, and at early ripening time. The samples frm each f ten separate bunches were cllected successively during the day and night at different time curse intervals and thse frm each sampling time were pled t give an adequate sample size fr analyses Sugar analysis Sluble sugars were assayed enzymatically using a Behringer Mannheim enzyme kit and quantified by spectrphtmetric detectin. The slutin cntaining phlem exudate was used directly withut cncentrating but the samples with high sucrse levels were diluted t imprve accuracy. The levels f sucrse, glucse and fructse were determined fr each sample Amin acid analysis The phlem exudates in 1 mm HEPES buffer slutin were freeze dried. Lyphilised samples were redisslved in 1p1 f HCI (.1 N) cntaining.5 mm nrvaline and sarcsine as internal standards. Redisslved samples were centrifuged fr 5 minutes at 14 rpm in an Eppendrf 5415c centrifuge. Analysis was perfrmed with ne tll f the supernatant. Amin acid analysis was carried ut n a Hewlett-Packard Amin Quant amin acid analyser, cnsisting f an HP 19 series II liquid chrmatgraph cntrlled by HP Chem Statin sftware. An autsampler was used t derivatise the amin acids which were then separated by reversed-phase HPLC n an Amin quant Cl8 clumn. Derivatisatin was with OPA (rth-phthalaldehyde) (l'amin acids) and FMOC (9- flurenylmethyl chlrfrmate) (2' amin acids) and quantified by U.V. absrbance. Amin acid calibratin standards. derivatising regents, and the amin acid analysis clumn were btained frm Hewlett-Packard Mineral elements measurement Inductively cupled plasma spectrmetery (ICPS) enables the determinatin f ttal P, K, S, Ca, Mg, Na, Al, Zn, Mn, Fe, C, M, and B. ICPS has the ptential t simultaneusly determine all the nutritinal elements (except nitrgen) with a plychrmatr in a single digestin. A knwn vlume f exudate in buffer slutin was lyphilised and disslved in apprximately 1 ml f nitric acid (Univar A.R.71/ w/w). The mixture was heated in a water bath fr 2 minutes, until brwn fumes were n lnger

75 given ff. Digests were remved frm the water bath and allwed t cl. After cling, the digest was diluted t the initial vlume with l% v/v nitric acid and decanted int a plystyrene tube.

92 75 given ff. Digests were remved frm the water bath and allwed t cl. After cling, the digest was diluted t the initial vlume with l% v/v nitric acid and decanted int a plystyrene tube. The digest std vernight befre ICP analysis t allw precipitatin f any suspended slids. The ICP instrument used was a 3588 ICP Spectrmeter Sim/Seq. with 21 simultaneus channels manufactured by Apptied Research Labratries (Switzerland). As sdium hydrxide (NaOH ) was used fr ph adjustment in the buffer slutin and sulphur was ne f cnstituents f HEPES buffer, and als because f the presence f EDTA in the slutin, the values fr Na, S, Ca and Mg are nt presented. 6.3 Results The grapevine phlem sap was cllected frm bunch lateral branches in situ fr three cultivars at different physilgical stages during the grwing seasn and analysed fr sugars (sucrse, glucse and fructse), amin acids and minerals Grapevine phlem sap cmpsitin Tables 6.1 and 6.2 shw the levels f sugars, ttal amin acids and K in the phlem sap cllected frm cultivars Muscat Grd, Sultana and Shiraz befre veraisn, at veraisn and at berry ripeness. The results presented are the cncentratins f each cmpund in the buffer slutin used t cllect the exudates. This was 1.5 ml per bunch lateral and the values crrespnd t the cncentratin f the metablite in slutin taken frm the pl f samples cllected frm 12 bunches (1.5 ml/ bunch) during 4 hurs between 9. am t 5. pm. Fr all three cultivars. sugars and K were the predminant slutes in the exudates cllected during the grwing seasn. Sucrse levels in the sap cllecting buffer slutin were highest befre veraisn and the lwest at veraisn. The majr sugar detected in all samples was sucrse. but glucse and fructse were als present althugh cnstantly in lw amunts and shwing n crrelatin with the crrespnding sucrse levels (Table 6.1).

93 76 Table 6.1 Sugars, amin acids and ptassium measured in phlem sap exudate f three grapevine cultiva s during the grwing seasn (p ml I 1.5 rnl f buffer slutin)' Cultiva Sampling time Sucrse Glucse Fructse Ptassium Ttal amin "Brix acids bef-re veraisn Muscat Grd veraisn l.l npe Sultana befre veraisn 4.7 veraisn 8.2 ripe t t Shiraz befre veraisn veraisn n D e 2t ,l6l Exuded fr 4 hurs. (12 bunches; 1.5 ml / bunch/ 4h.) l' Data frm 1994 samples Sucrse levels in the exudates frm Sultana were higher than fr the ther tw varieties. The phlem sap exudates were analysed fr I I minerals but nly results fr ptassium are presented here. The K levels in the exudates were similar in all th ee cultivars and did nt change greatly during the grwing seasn (Table 6.1). Ttal amin nitrgen was determined and as shwn in Table 6.1. cncentratin f amin acids was lwer than fr sucrse and K n a mlar basis. There were als seasnal changes and varietal differences in ttal amin acids but they did nt parallel thse fr sucrse (Table 6.1). Fr example, while there are parallel changes in the cncentratins f sucrse and amin acids in Muscat Grd phlem exudates. Sultana and Shiraz phlem exudates did nt shw similar fluctuatins in the cncentratins f sucrse and amin acids Seasnal changes in the phlem sap cmpsitin The predmihant amin acid in phlem sap was glutamine, with smaller amunts f glutamate, aspartate, serine and prline. Glutamine and glutamic acid tgether represented abut 5% - 8% f the amin acid prfìle all thrugh the grwing seasn and amin acids levels shwed different seasnal patterns in varius cultivars (Table 6.2). In Muscat Grd amin acid levels shwed a seasnal pattern similar t that f sucrse whereas Sultana and Shiraz each had different seasnal pattems f amin acids levels in the buffer slutin (Table 6.2 and Appendix Tables 1,3,.4, and 5).

Table 6.2The seasnal changes in phlem exudate amin acids f Muscat Grd, Sultana and Shiraz. Values given are/ f ttal amin acids befre veraisn (B), at veraisn (V), and at ripening time (R).

94 Table 6.2The seasnal changes in phlem exudate amin acids f Muscat Grd, Sultana and Shiraz. Values given are/ f ttal amin acids befre veraisn (B), at veraisn (V), and at ripening time (R). 77 Muscat Grd Sultana Shiraz Amin acids (B) CÐ ß) (B) ry) ß) (B) (v) (R) ASP GLU ASN SER GLN HIS GLY THR ALA ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU LYS PRO I r t4 9.3 tt.7 I 7) 42.2 r J.J r I.9 1 t Ttal * "Brix J.J 69.3 *Ttul values are in rml amin acids/ l.5ml. f buffer slutin/ 4 h. r t I r " t t t t t Abbreviatins fr amin acids presented ASP Aspartic acid, GLU Glutamic acid, ASN Asparagine, SER Serine, GLN Glutamine, HIS Histidine, GLY Glycine, THR Threnine, ALA Alanine, ARG Arginine, TYR Tyrsine, CYS-CYS Cysteine, VAL Valine, MET Methinine, TRP Tryptphan, PHE Phenylalanine,ILE Isleucine, LEU Leucine, LYS Lysine, PRO Prline.

95 Validatin f the sap cllecting technique T assure the validity f the technique used in these experiments, cntrls were set up by applying different treatments designed t alter phlem sap exudatin. Table 6.3 shws the cmparisn between the sugar levels exuded frm cut bunch stems n detached btmches (Fig. 6.3) and the exudate cllected frm bunches still cnnected t the vine. The results shw mre than 14 times mre sucrse exuded frm the bunches which were cnnected t the vine cmpared t the cut bunch stems. The levels f glucse and fructse were very lw under bth treatments. Table 6.3 Cmparisn f sugars exuded frm fruit bunches n the vine with excised bunchstems as in Fig. 6.3, (cv. Muscat Grd at pre veraisn, 'Brix:4.5). Treatments Standard sampling Sucrse Glucse Fructse ttml/ L5 ml f buffer slutin* ** Bunchstem nly *The exudatin perid was 6 h..26! Values are meant SD f three replicates f l bunches in each. As EDTA was used as a chelating agent t keep sieve tubes pen, ne f the treatments t btain cntrl samples t test the cntinuatin f phlem sap flux int the buffer slutin was exclusin f the EDTA frm the buffer. Table 6.4 presents the results f the samples cllected b1, using HEPES buffer cntaining EDTA (buffer + EDTA) and the buffer withut EDTA (buffer - EDTA) and als using water nly. In the samples cllected in the buffer slutin withut EDTA, and in water, n sucrse was detected and the levels f amin acids and ptassium were als lw in cmparisn with the samples cllected in standard buffer slutin.

79 Table 6.4 The effect f exclusin f EDTA frm the buffer slutin used fr sap cllectin n the levels f different metablites in cllected sap samples (cv. Muscat Grd at veraisn, 'Brix :6.4).

96 79 Table 6.4 The effect f exclusin f EDTA frm the buffer slutin used fr sap cllectin n the levels f different metablites in cllected sap samples (cv. Muscat Grd at veraisn, 'Brix :6.4). Treatment Sucrse Ttal amin acids Ptassium Buffer + EDTA Buffer - EDTA Water.342 <.4 <.4 unl/ 1.5 ml fbuffer slutin* *The exudatin perid was 4 hurs. Subsampled frm pl f samples cllected frm 3 bunches Girdling abve and belw the fruit bunch n the cane (Fig. 6.4) was dne t cut the phlem tissue and thus stp phlem sap transprt t the bunch. The phlem sap cllected frm girdled bunches was analysed fr sucrse, amin acids and ptassium and the results are presented in Table 6.5. Table 6.5 The effect f girdling the phlem tissue abve and belw the bunch n the levels f different metablites in cllected sap samples (cv. Muscat Grd at ripening, 'Brix: 1.2). Treatment Sucrse Ttal amin acids ptassium ttml/ L5 ml fbuffer slutin* Nn-girdled Girdled t9 *The exudatin perid was 4 hurs. Subsampled frm pl f samples cllected frm l bunches. Girdling treatment reduced the amunt f each f the cmpunds exuded cmpared with the nn-girdled bunches. Amin acid levels in the buffer slutin were particularly affected by girdling but sucrse was als reduced t lw levels and ptassium was decreased. In anther sampling with bth girdling and exclusin f EDTA frm the buffer slutin the levels f sugars exudated frm phlem was measured. The treatments shwed a prnunced effect n the sugar levels exudated int the buffer slutin with n sucrse and glucse detected in the buffer slutin withut EDTA and the levels f sugars exuded frm girdled bunches were at the lwer limits f detectin (Table 6.6).

8 Table 6.6 The effect f girdling and exclusin f EDTA n the phlem sap exudatin (cv Muscat Grd at ripening, 'Brix: 15.8). Treatments Sucrse Glucse Fructse Nn-girdled Girdled 1.969.45 ttnl/ 1.

97 8 Table 6.6 The effect f girdling and exclusin f EDTA n the phlem sap exudatin (cv Muscat Grd at ripening, 'Brix: 15.8). Treatments Sucrse Glucse Fructse Nn-girdled Girdled ttnl/ 1.5 ml f buffer slutin* <.39 Buffer - EDTA <.39 < *The exudatin perid was 6-7 h, Subsampled frm pl f samples cllected frm 6 bunches. Later in the seasn the phlem flux cntrl treatments were als applied t ther cultivars. Tables 6.7 and 6.8 shw that simila results were btained using cv. Sultana with relatively high sucrse levels exuded vernight frm the bunchstem f ripe bunches. Exclusin f EDTA frm the buffer slutin greatly decreased the levels f sugars, amin acids and ptassium exuded. Table 6.7 The effect f exclusin f EDTA n the phlem sap exudatin (cv. Sultana at ripening "Brix : 18.). Treatment Sucrse Glucse Fructse Wl/ 1.5 ml f buffer slutin* Buffer + EDTA r Buffer - EDTA <.39 < *The exudatin perid was 5 h. Subsampled frm pl f samples cllected frm 6 bunches. Table 6.8 The effect f exclusin f EDTA n the phlem sap exudatin (cv. Sultana, ripe "Brix:23.8). Treatment Sucrse Glucse Fructse Amin acids Ptassium Uml/ 1.5 ml f buffer slutin+ Buffer + EDTA Buffer - EDTA < *The exudatin perid was l6 h (vernight), Subsampled frm pl f samples cllected frm 6 bunches

$4 Phlem sap diurnal Patterns Pøttern f sucrse in exudates during a day (befre veraisn) Fluctuatins in cncentatin f sucrse \ /ere seen in phlem exudates taken during ne day frm Muscat Grd grape$

98 Phlem sap diurnal Patterns Pøttern f sucrse in exudates during a day (befre veraisn) Fluctuatins in cncentatin f sucrse \ /ere seen in phlem exudates taken during ne day frm Muscat Grd grape bunches befre veraisn ("Brix : 4.5). The samples were taken every 1.5 h. during the day and analysed fr sucrse, amin acids and ptassium. The results a e the mean f three replicates with l bunches in each. The exudate samples cllected frm each grup f 1 bunches were pled and analysed. Sucrse level increased during the day reaching its highest at midday ( ) achieving a maximal level f 2.68 pmles/ 1.5 ml f the buffer slutin and then shwed a steady decrease during the afternn (Fig. 6.5 A). Levels f hexses (glucse and fructse) were fund nly in trace quantities in the cllecting buffer slutin and shwed n distinct diurnal patterns (data is nt shwn). Ptassium levels als shwed diurnal variatin during the sampling perid and gained the maximum value f 2.16 pmles/ 1.5 ml f the buffer slutin at h. (Fig' 6.5 C). The pattern f ptassium was similar t sucrse but the variatin was less prnunced. Amin acid levels were lwer than sucrse r ptassium. Ttal amin acids als shwed diurnal variatin with a similar pattern t ptassium. Hwever, the cntributin f specific amin acids t the ttal pl varied during the diurnal cycle (Fig. 6.6). The principal amin acids were glutamine, glutamate, aspartate, arginine, alanine, and prline (Fig. 6.6 A-F). All the principal amin acids shwed a diurnal pattern, but alanine stayed high in the afternn samples (Fig. 6.6 E). The sampling time in this experiment was nly during daylight and stpped in the afternn.

99 82 Sucr.a A 3. 2.O 1. tr I. I Amln rcld B ã.6 J t.5, l ts u?.3 g E 2.1 Pl...lum c q q N c; qq q qrì NO Sam pllng tlme (hrs.) q 6 q rì ( Fig. 6.5 The diurnal pattern f sucrse (A), ttal amin acids (B) and ptassium (C) levels in phlem sap exudates cllected frm pre-veraisn grape bunches during the day (vertical ba s indicate SE).

100 83 u? ) E rq õ E c,r(x 35 3?.9 2ú + G J{ A ) E ll l õ E c a ^RG D r G.IJ B 11 +A.À E G c? ) E UÌ 6 õ E c t 4 I t8 1 U? E u?! Ec l I 6 I 2 2 E uì ) E úì 2A ASP c E t? È ul PRO F õ E c 1 6 ú s E c 2 ô Sampllng tlme (hrs.) Sampllng tlm (hrs.) Fig. 6.6 (A-F) The diurnal patterns and the levels f majr amin acids detected in the grapevine phlem sap (pre-veraisn).

101 Døy and night diurnal patterns (øfter veraisn) Experiment I. T investigate the effect f time f cutting ff the bunch laterals n the appearance f the sucrse exudatin peak, anther set f samples with different starting times were cllected frm Muscat Grd grape bunches when the berries had just started ripening ("Brix : 1.2). The whle sampling perid cvered a day, a night, and the next day. Three series f sampling (4, B, and C) were dne at tw hurly intervals, the first set started at 6. hr. Each f the ther tw sets started fur hurs later than the previus ne. The analysis f cllected samples shwed that regardless f the time that sampling started, sucrse cntent in the buffer slutin increased frm late afternn and reached its highest during the dark perid (Fig. 6.7 A, B, and C). Frm 22. hr. t next midday (12. hr) samples were cllected fr 8, 6, and 6 h perids frm the same grup f bunches. The results shwed that after midnight, sucrse levels decreased t very lw levels during the light perid (Fig. 6.7 A, B, and C). A series f samples was als cllected frm the bunches that were pre-treated with girdling as described befre in The results f sugar analysis shwed very lw levels f sucrse and n diurnal pattern f sucrse in the exudates cllected frm girdled bunches (Fig' 6.7 D). Experiment II. The samples that were cllected after veraisn (Fig. 6.7) shwed a different sucrse diurnal pattern cmpared with thse frm befre veraisn (Fig. 6.5). T investigate diurnal patterns further, a third series f samples were cllected at 6 hur intervals frm Muscat Grd bunches later during ripening ("Brix : 15.6). Each set f sampling series started 12 hurs later than the previus ne. In line with previus results, the sucrse exudatin was highest in the samples cllected at night and then decreased in samples which were cllected during the day (Fig. 6.3). Each sampling grup shwed the highest sucrse flux during 24. hr t 6. hr regardless f the time f beginning f the sampling. The sucrse exudatin cntinued fr 48 hurs fr each sampling grup and then decreased. There was n diurnal pattern bserved fr sucrse levels in the exudates cllected frm girdled bunches in this experiment (Fig. 6.8). Tw series f the samples which had been started first and secnd, were als analysed fr ptassium and fr amin acids. It was fund that ptassium levels in bth samples analysed shwed large variatins but in a different pattern t that fr sucrse. Ptassium exudatin was high during the light perid when sucrse was lw in nearly all samples (Fig. 6.1). Ttal amin acids als shwed a diurnal pattern fr the first tw days (Fig. 6.9). Individual amin acids displayed different patterns f exudatin during the sampling perid. In bth f the light and dark perids glutamine, aspartate, and glutamate were the principal cmpnents f the phlem sap amin acid pl (Fig.6.1l A-C). Levels f arginine shwed n significant changes in respnse t light r da k perids (Fig. 6.1I D). Levels f arginine were higher in the later samples than in the samples cllected earlier in the hrst tw days f sap cllectin. Arginine changed asynchrnusly reaching a very 84

$85 high level at the third day whereas exudatin f ther amin acids ceased by that time (Fig. 6. ll D). 3.s 3. ô 2.5 J Þ vì 2. f r.s q - -A ----{-B ------<,- c -ô- D (Girdlcd) L : 1. (t).5.? ( c q @ ñ O q ç c N @ I ç q Õ q @ c; N q CO Sampling time curse q N (\ q N c( q ñ N q N c( I q ÔJ Fig.$

102 85 high level at the third day whereas exudatin f ther amin acids ceased by that time (Fig. 6. ll D). 3.s 3. ô 2.5 J Þ vì 2. f r.s q - -A ----{-B <,- c -ô- D (Girdlcd) L : 1. (t).5.? ( c ñ O q ç c I ç q Õ c; N q CO Sampling time curse q N (\ q N c( q ñ N q N c( I q ÔJ Fig. 6.7 The diurnal pattern f cntinuus sucrse exudatin frm pst-veraisn Muscat Grd bunch laterals during light and dark perids with different starting times (each value represents mean f l bunches), 'Brix = 1.2. A, B, and C represent three samples where the bunch laterals were remved at different times. Fr girdled sample (D) the canes were girdled the day befre sampling. The values are fr tw hurs f each sampling perid.

$( @ \ Þl E ra E q) U' r È (â 6. 5. 4. 3. 2.OO 1. Started at 6 pm day I (series l) Started at 6 am day 2 (series 2) Started at ó pm day 2 Started at 6 am day 3 Girdled and started at day I.$

103 \ Þl E ra E q) U' r È (â OO 1. Started at 6 pm day I (series l) Started at 6 am day 2 (series 2) Started at ó pm day 2 Started at 6 am day 3 Girdled and started at day I. + (\ I q ( a e s C\ q (\l I q ( q c) I q q $ a F q ( t q $ (\ q $j I q ( q.t r + C\.1 I O q ( I q sf c! q ôj - q ( q c r I q.t Sampling time curse Fig. 6.8 The diurnal pattern f cntinuus sucrse exudatin frm ripe Muscat Grd bunch laterals during light and dark perids with diflerent starting times (each value represents mean f l bunches), "Brix: 15.6.

$87 t I E ul.8.6 Serles I Serles 2 g È 4 a g ( g E E.2. q ç q TD q ( q \t N q N c( c.d q N rl î c(d Sam P (N cq rt@ NO ng tlme q @ c C\ c!l N c @ c( q t N N q ( q ( c 6$

104 87 t I E ul.8.6 Serles I Serles 2 g È 4 a g ( g E E.2. q ç q TD q ( q \t N q N c( c.d q N rl î c(d Sam P (N cq rt@ NO ng tlme c C\ c!l N c( q t N N q ( q ( c 6 Fig. 6.9 The diurnal pattern f ttal amin acids measured in ripe Muscat Grd grapevine exudates during days and nights (samplings started at different times and each value represents mean f l bunches), (fìrst and secnd series f samples presented in Fig. 6.8). ( I -_F S riß 1 J E ul - S ries 2 t g E 4 E r gl õ. 2 N N N ñ ç.þ qc Fô, Sampllng q N t me q (D q N q!t N 9 q) (D ñj N q ( q N Fig. 6.1 The diurnal pattern f ptassium measured in ripe Muscat Grd grapevine exudates during days and nights (samplings started at different times and each value represents mean f l bunches).

105 88 E t a ã E l l ì! É, Ê l(x) 3 2 r A (ln + 6.ttðl Ssiö2-4 a = J E ul a E Ê 2g 2 15 t 5 D Arg + SaDt I Scrb 2 - G ( ã. J E ul ì E t 12 1 EO 1 2 B G J + Seds I sefis Ã J E ul a! E c E Ala + S.rbs I S rbs 2 t J E ul c A5p - Sems 1 SE 2 i f J E ul F Pr - - Sneg 1 Senes 2 a! E Ê 1 2 6! E waô -NO?-N -?NO Sampling timc çø N@NPN, 6?6 -N Sampling tlnc N N Fig (A-F) The diurnal pattern f individual amin acids measured in ripe Muscat Grd grapevine phlem sap exudates during days and nights (samplings started at different times and each value represents mean f l bunches).

106 Grapevine leaf exudates The exudate samples cllected frm grapevine leaves f cvs Muscat Grd, Sultana and Shiraz were analysed fr sucrse, glucse, fructse, amin acids and mineral ins and the results are presented in Table 6.9. The values represent the exudate cllected frm the cut end f petiles f 15 leaves ver 16 hurs. Lw levels f metablites were exuded frm grapevine leaves cmpared t grape bunchstems. In leaf samples sucrse was high cmpared t glucse and fructse which in mst samples were belw detectin levels (Table 6.9). Higher sucrse was measured in samples cllected in 1995 (in dark) than thse f 1994 (in light). The samples with higher sucrse levels had higher ptassium and amin acids as well (Table 6.9). In the leaf exudates the same amin acids predminated as were fund predminant in the bunchstem exudate i.e. glutamine, glutamate, aspartate, and alanine (Appendix I.3). When the leaves were kept under light cnditins the prline levels were higher and alanine levels decreased under the same cnditins (Appendix I.3). Table 6.9 The levels f sucrse, glucse, fructse, ptassium and ttal amin acids f leaf phlem sap exudated under light and dark, cllected at ripening perid. Cultiva s Cmpunds Sucrse Muscat Grd.37 Sultana Shiraz Muscat Suløna Shiraz Grd r995 t994 I rml/ 1.5 ml f buffer slutin/ 16 h.1l < <.4 Glucse <.4 <.4.88 < Fructse <.4 <.4.t32 <.4.43 <.4 Ptassium r Ttal amin acids I The 1994 samples were cllected in the light and the samples fr 1995 were cllected in dark.

9 6.4 Discussin 6.4.1 Validatin f the sap cllecting methd Althugh the EDTA-facilitated methd f phlem sap cllectin has been used previusly (King and Zeevaart, 1974; Hansn and Chen, 1985; Glad et al.

107 9 6.4 Discussin Validatin f the sap cllecting methd Althugh the EDTA-facilitated methd f phlem sap cllectin has been used previusly (King and Zeevaart, 1974; Hansn and Chen, 1985; Glad et al., 1992a), a number f cntrlled experiments were cnducted t further validate the methd (Tables ) The majr cnstituents f phlem sap exudates were sucrse, ptassium, and amin acids (Table 6.1,6.2) which is cnsistent with ther analyses f phlem sap (Girusse et al., I99I; V/iener et al., l99l; Glad et al.,!992a). When EDTA was mitted frm the buffer slutin, the amunts f each f thse cmpunds detected in the exudates was greatly decreased (Tables 6.5, 6.7,6.8, and 6.9), indicating that EDTA has a key rle in facilitating exudatin f phlem sap frm the bunch laterals. This has als been bserved in ther plants (King andzeevaart, 1974 Hansn and Chen, 1985) and is thught t invlve chelatin f calcium by the EDTA t stp callse frmatin in the sieve tubes. Girdling f the cane als reduced the levels f sucrse, amin acids, and ptassium in the exudates (Tabls 6.6, 6.7, Figs. 6.7, 6.8), as wuld be expected if these cmpunds were derived frm phlem sap entering the bunches. All f these results supprt the ntin that the material cllected frm bunch laterals by this technique is derived frm the phlem sap entering the bunch fr distributin t the develping berries Grapevine exudate cmpsitin; cmparisn with tissue cmpsitin The data frm Table 6.1 shws that in the grapevine phlem sap cllected by the facilitated technique, sucrse is the main sugar translcated this was als bserved by Swansn and El-Shishini (1958), Kblet (1977) and by Glad et al. (1992a). The levels f glucse and fructse were cnsistently lw in all samples cllected at different times and this may suggest a nn-phlem rigin fr the hexses. It is pssible that sme sucrse mlecules are hydrlysed r inverted after mving int the cllectin slutin r diffuse ut f sieve tubes during nrmal translcatin and are hydrlysed in adjacent cells t glucse and fructse which are nt mbile in the phlem. The values fr glucse and fructse were lwer than that reprted by Glad et al. (1992a), prbably due t m e lignified and mature rachis tissue used in this experiment cmpared t the yung tissue (at flwering time) used by Glad and cwrkers. In all three cultivars the mst abundant amin acid present in the phlem sap was glutamine fllwed by glutamic acid, aspartic acid, alanine, serine and prline (Table 6.3 and Appenix l). These amin acids have als been reprted as a majr cmpnent f phlem sap in sme ther plant species (Hall and Baker, 1972; Leckstein and Llewellyn, 1975; Pate, 198; Hking. 198; Simpsn and Dalling, 1981; Urquhart and Ly, l98l; \

108 91 Fukumrita and Chin, 1982;Fisher and MacNicl, 1986; Girusse etal.,l99i; Wiener et al., 1991). Since glutamine was predminant all thrugh the grwing perid and in all cultivars studied it appears t be a majr translcating amin N frm in grapevine phlem sap which was als reprted by Glad et al. (1992a). It is als reprted t be a majr amide in grapevine xylem sap (Andersn and Brdbeck, 1989a; Glad et al', 1992b; Glad et al'' 1994; Andersn et al., 1995). ptassium accumulates in cnsiderable amunts in the grape berries after veraisn and during berry maturatin and ripening (Smers, 1975; Hale, 1977; Hrazdina and Mskwitz, 198; Iland and cmbe, 1988; Lang and Thrp, 1989; Mrrisn and Idi, 199; Witliams and Biscay,IggI; Creasy et al., 1993; Bselli et al., 1995). In ripe grape berries K cnstitutes abut 7% f mineral catin cntent f the berry (Hale, 1977) and, f the ttal berry K. nearly half may be lcated in the skin (Smers, 1975; Iland and Cmbe, 1988; Gutiérrez-Granda and Mrrisn, lgg2). Amng the catins, ptassium has a rle in stimulating the unlading f assimilates (H, lgss) and was the majr mineral fund in the grapevine phlem exudate, which is in agreement with the results reprted by Glad et al. (1992a). The ptassium levels in Muscat Grd sap exudates were always higher than in Sultana and Shiraz (Table 6.2). Althugh xylem exudate is reprted t cntain cnsiderable levels f ptassium (73pM) (Glad et al. 1992b), its presence in the phlem sap in relatively high cncentratin indicates that phlem sap is als an imprtant surce f ptassium accumulating in the ripening berries. In additin the high levels f ptassium and ther catins in the phlem sapprbably serves t balance the electrical charge due t the presence f sme anins. Glad and cwrkers (1992a) reprted rganic acids in grapevine phlem exudates and the ph f phlem sap t be alkaline (ph= 7.7). Changes in K cntent in fruits is generally assciated with phlem flw and in grapes is usually cntrasted with Ca accumulatin (Lang and Thrp, 1989; Mrrisn and ldi, 199) which is thught t be transprted nly in the xylem (Marschner, 1983). Further studies will be required t quantif, phlem sap cmpsitin at different stages f berry develpment such data culd then be used t determine the flux f carbn and nitrgen int the fruits Phlem sap flux and slute accumulatin in grape berries The influx int grape berries f nearly 25 mm3 -l per berry f phlem sap was measured by Lang and Düring, (1991). The authrs fund thatthe stream shwed little indicatin f sensitivity t the day/night cycle and cmpared with a rate f transpiratin f abut 2 mm3 -l per berd,. They als measured backflwf xylem f at least 5 mm3 d-l frm the berry t the vine. As a result, the influx f phlem sap is mre r less in balance with the efflux f xylem sap and transpiratinal lss t the air ( Lang and Thrpe,

92 19g9; Lang and Düring, 1991). Phlem sap inflw f.52 +.28 mm3 h-l in the light and.32 +.15 mm3 h-l in the dark at pre-veraisn, and 5.79 t 1.99 mm3 h-l in light and 1.62 +.

109 92 19g9; Lang and Düring, 1991). Phlem sap inflw f mm3 h-l in the light and mm3 h-l in the dark at pre-veraisn, and 5.79 t 1.99 mm3 h-l in light and mm3 h-l in dark at pst-veraisn, was reprted by Greenspan et al. (1994). The net flw fr the xylem during sugar accumulatin is frm fruit t vine ( Lang and Thrpe, 1989), and there are a number f reprts f xylem backflw at sme stage in fruit develpment (Hamiltn and Davis, 1988; Eh et and H, 1986; Lang and Thrpe, 1989)' 'Water backflw frm the xylem als was bserved by Greenspan et al. (1994) after veraisn in water stressed berries. With the assumptin that after veraisn all slutes accumulated in the berry are imprted via phlem, estimates were made f the required rate f phlem sap flux int Muscat Grd berries frm the knwn rate f increase in slute increase rate in the berries ver time. I. Frm what is reprted by Cmbe (198), the slute levels per berry f Muscat Grd were 12 mg at day 7 after flwering (at the beginning f ripening) and reached 52 mg per berry at day 1. This increase in slutes ver 3 days ccurred when berry vlume increased by ne cm3 and it is equal t 4mg/ berryl3 days r 13.3 mg/ berryi day = 73.9 pmles/ berry / day as glucse equivalents. il. The same calculatins were dne using the data frm Cmbe (1989). Glucse * fructse cncentratin in the Muscat Grd berry increased by 13'4 mgl ml f juice per day between the secnd day after veraisn and day 12. If we assume ne ml increment f juice per Muscat Grd berry this cnespnds t an increase f: 13.4 mglberry/ day : 74.4 tmles /berry/ day as glucse equivalents. III. Frm data presented in chapter three (Fig. 3.2 graphs 4,5,6) slute increase in Muscat Grd berries was calculated, using the frmula suggested by Cmbe (198). Between day 5 after flwering ('Brix: 8.), and day 1 ('Brix =22.) berry vlume increased by apprximately ne cm3. The amunt f slutes per berry was249.6 mg at day 5 and increased t mg/ berrl'at da1' 1. The amunt gained per berry was mg/ berry during 5 days r: 13.3 mg/ day/ berry, =73.9 rmles/ berry/ day as glucse equivalents. y. Anther calculatin was dne using the data recrded during phlem exudate cllectin between tw sampling times. The slute increase n a per berry basis was I 1.56 mg/ berry /day ver 14 days when 'Brix increased frm 1.2 t 15.6 and berry vlume increased.6 cm3 during the same time which crrespnds t an increase f: mg daylberry :64-2 pmles/ berla/ day as glucse equivalent' The similar values resulting frm the use f different surces f data indicates that the methd f calculatin is reasnable and suggest that the rate f increase in slutes f the berries is relatively cnstant during ripening and in the range pmles glucse equivalents I berrylday. T cmpare the values f sucrse and ther metablites measured in phlem exudate cllected in this experiment, similar calculatins were dne fr the

93 amunt f metablites exuded frm phlem int the buffer slutin using the number f berries n the laterals befre they were remved t cllect phlem sap exudates: Using the data in Fig. 6.

110 93 amunt f metablites exuded frm phlem int the buffer slutin using the number f berries n the laterals befre they were remved t cllect phlem sap exudates: Using the data in Fig. 6.8, fr first24 h, the ttal sucrse exuded fr the samples f series I was calculated t be 9.78 pmles per lateral hlding 33 berries. This is equal t.59 pmles/ berry/ day glucse equivalent. Als ttal sucrse exuded fr the samples f series 2 (Fig.6.8) was calculated t be pmles/ lateral hlding 25 berries /day. This was equal t.92 pmles/ berry/ day glucse equivalent. Using data presented in Fig. 6.7.(A) fr first the 24 h and with similar calculatins and 32 benies per lateral a value calculated was.73 pmles/ berry/ day glucse equivalent. Thus the daily rate f sugar efflux frm cut laterals was in the range pmles glucse equivalents I berrylday. The maximum rate f efflux can als be calculated using the peak rates recrded. Using the maximum values f sucrse exuded during exudate cllectin the fllwing values were btained: Fig. 6.8 (Series 1);4.56 pmlesi 4 h: pmles/ dayl 24hl 33 berries: 1'l pmles/ berry / day glucse equivalent. Fig. 6.8 (Series 2); 5.4 pmles/ 4 h : 2.16 pmles/ dayl 24 hl 25 berries : l'6 pmles/ berry / day glucse equivalent. Fig. 6.7.(C) 3.29 pmles/ 2h:39.48 pmles/ dayl 24W 32 berries: 2.5 pmles/ berry/ day glucse equivalent. The data presented in Table 6.1 fr sucrse exuded frm Muscat Grd laterals befre veraisn was 3.24 pmles/ 4h equal t pmles/ day I 25 berries : 1.5 pmles/ berryi day glucse equivalent. The data presented in Table 6.3 fr sucrse exuded frm Muscat Grd laterals at befre veraisn was 3.77 rmles/ 4h equal t 15.8 pmles/ dayl22 berries: 1.4 pmles/ berry/ day glucse equivalent. Mean value fr the calculatins dne abve was 1.6 pmles/ berry/ day. Cmpared with the abve calculatin f sugar flw in each berry f abut 8 pmles/ berryl day, this suggests that the phlem exudatin technique has cllected 2/ f that which is nrmally accumulating in the berry in situ. Similar calculatins can be dne fr amin acids and ptassium. Amin acids were shwn t increase in Muscat Grd berries during apprximately 14 days frm 173 pmles/ 1 ml f juice t 323 rmles/ 1 ml (Kliewer 1969) during the time when 'Brix increased frm 18.4 t 21.5 which was assumed t take abut tw weeks. Then the calculated value f amin acid accumulatin wuld be 1.7 pmles/ ml /day. If we

94 assume that berry vlume increases by.6 cm3 during 14 days then it will be equal t.643 pmles / berry/ day. Frm data presented in Table 6.1 amin acids exuded was 2.7 pmles/ 25 berriesl day : 18.

111 94 assume that berry vlume increases by.6 cm3 during 14 days then it will be equal t.643 pmles / berry/ day. Frm data presented in Table 6.1 amin acids exuded was 2.7 pmles/ 25 berriesl day : 18.7 nmles/ berry/ day. Frm the data given in Fig. 6.9 (series l) the value fr amin acids exuded was 38.6 nmles/ berry / day during first 24 h. Frm the data given in Fig. 6.9 (series 2) the value fr amin acids exuded was 34.4 nmles/ berry / day during first 24 h and als the value fr series I, cllected at the same time parallel t series 2 was 65.6 nmles/ berry/ day Mean value fr the calculatins dne abve was 61.8 nmles/ berra/ day which is equal t l/ f amin acids calculating t be accumulating int a berry at the same time. Using maximum values fr amin acids exuded during 24 h Fig. 6.8 (series l) 82.4 nmles/ berrylday: l3y f amin acids accumulating int a berry at the same time. Fig.6.8(series2)68.3nmles/berry/day= ll/faminacidsaccumulatinginta berry at the same time. Bselli et al. (1995), reprted that ptassium cntent increased apprximately.2-.3 mg b..ry-' d-r during the entire perid f fruit develpment. =.51 '.17 pmles/ berry/ day. Ptassium cncentratins were shwn t increase abut 1 mg/ L in berry extract frm veraisn t ripe stage during 56 days in Chaunac grapes (Hrazdina et al. 1984)' =.92 pmles /berry /day Ptassium increased frm 1.5 mg/berry't 4.7 mglberry during 65 days inchardnnay grapes (Pssner and Kliewer 1985). :1.26 pmles/ berry idar'. In Shiraz berries ptassium per berry increased frm abut ne mg per berry t abut 3.8 mg per berry when "Brix increased frm 7 f 25 (lland and Cmbe 1988). If apprximately 4 days are required fr the ripening prcess in Shiraz berries then, the ptassium increase rate will be; 1.8 rmleslberryl day. Thus the mean value f ptassium accumulatin wuld be 1.5 pmles/ berry/ day. Frm data presented in Fig. 6.1, (series 1) fr Muscat Grd phlem exudate, K was exuded int the buffer slutin at the rate f:.376 pmles/ berry/ day r 36/ f the calculated accumulatin rate. Frm data presented in Fig. 6.1, (series 2) K was exuded at the rate f:.339 pmles/ berry/ day r 32/ f the calculated accumulatin rate.

Using mæ<imum values fr ptassium exuded duing 24h,Fig.6.8 (series 1); the value was.937 pmles/ berry/ day r 89% f calculated accumulatin rate and fr Fig. 6.8 (series 2) l.

112 Using mæ<imum values fr ptassium exuded duing 24h,Fig.6.8 (series 1); the value was.937 pmles/ berry/ day r 89% f calculated accumulatin rate and fr Fig. 6.8 (series 2) l.l pmles/ berrylday r 14% f calculated accumulatin rate' Thus fr sucrse the amunts exuded int the buffer slutin by cut laterals wuld nly accunt fr abut l-3y and fr amin nitrgen abut l2% f the knwn rates f accumulatin f these cmpunds in the berry during ripening. Fr K, the levels exuded were higher, amunting t 3-1/ f the calculated rate f accumulatin in the berry' Hwever, there is a pssibility f xylem r ther surces f K cntributin in the levels f K exuded int the cllecting slutin. Girdling reduced efflux f sucrse by 78/ and amin acids by 86% but K efflux nly decreased by 39% (Table 6.5). This suggest that K efflux frm the cut laterals is frm ther surces as well as frm phlem sap' These results suggest that exudatin f phlem sap frm the cut lateral is nt a quantitative measure f material flwing int the berries. There appeils t be a significant decrease in flw f phlem sap nce the lateral is cut and the berries remved' This further suggests that there are active mechanisms t unlad cmpunds frm the phlem int the berry. Once the berry is remved, these cause cease perating and the flw f phlem sap int the lateral wuld decline Seasnal changes in exudate flux Cnsiderable amunts f sucrse, amin acids, and ptassium were exuded int the cllectin buffer slutin frm different grape cultivars and at different stages f grwth and develpment during the seasn (Tables 6.1). The amunt f sucrse exuded during 4 hurs f sampling varied with ripening and shwed lwer levels f sucrse in the samples cllected at veraisn than bth befre and after veraisn fr all three cultivars (Table 6.1). After fruit set the develping berries are pwerful sinks with high demands fr metablites while they are rapidly increasing in size (Hale and Weaver, 1962). The strng sink activities f berries declines during early veraisn and rises again during sugar accumulatin (Cmbe, 1989). There are a number f indicatins that at the nset f ripening in grapes a breakdwn f xylem cnnectins t the berry ccurs and after this stage, the majrity f water that accumulates in benies derives frm phlem sap (Cmbe and Bishp, 198; Lang et al., 1986; During et al ; Findlay et al., 1987; Lang and Thrpe, 1989; Creasy et al., 1993; Creas 'and Lmbard, 1993; Greenspan et al., 1994). The increased sink f activity f berries after the inceptin f ripening, i.e. after verisn (Hunter et al., 1995) suggests an increased phlem sap flux int the berries. The seasnal variatin in the phlem sap cmpsitin seems t be a natural endgenus phenmenn in grapevine and has als been reprted in the ther plants Q'{bel, l99l).

96 6.4.5 Diurnal changes in exudatin Diurnal patterns f metablites were fund in the grapevine exudates (Fig. 6.5-6.11).

113 Diurnal changes in exudatin Diurnal patterns f metablites were fund in the grapevine exudates (Fig ). This may result frm a native diurnal pattern in phlem sap r it may have been the result f wunding when the lateral is cut. T investigate this further tw experiments were cnducted t determine the effect f the time f excising the bunch laterals frm the bunch fr exudate cllectin. The time f the beginning f the sample cllectin (cutting the bunch lateral ff) had n effect n the sucrse exudatin pattern (Figs. 6.7, 6.8) suggesting that the amunt f metablites exuded are mst likely related t the surce activities and is nt a wund respnse. It is ntewrthy that the diurnal patterns were different at the different grwth stage f grapes (cmpare Figs 6.5, 6.7, 6.8). The xylem is the surce f water inflw fr grape berries befre veraisn with phlem playing a substantial but secndary rle. This is reversed within nly tw days after veraisn, when the xylem supply appears t diminish greatly, and phlem becmes the majr vascular pathway fr inflw (Cmbe and Bishp, 198; Lang and During, 1991; Greenspan et al., 1994). It is pssible that changes in the berry develpment and the lag phase befre veraisn culd be the cause f alteratin f the diurnal pattern between day and night. Hwever, the actual cause f this change is nt knwn. It is knwn that during the day phtsynthetic carbn assimilatin usually exceeds the exprt f the carbn frm the leaf s carbhydrates accumulate in leaf chlrplasts as starch r in the cytsl and vacule as sucrse. During the night, sucrse is mbilised frm the vacule and is catablised frm starch and exprted frm the leaves t maintain the translcatin rate in the absence f phtsynthesis (Candlfi-Vascncels et al., 1994; Hunter et al., 1994). There is n infrmatin n diurnal changes in the grapevine phlem sap cmpsitin but there are sme reprts abut diurnal patterns f carbhydrates in grapevine leaves. Carbhydrate cncentratins f grapevine leaves were reprted t be highest at l3: h.. then decreased during the afternn and reached lw levels at l8: h. (Hunter et al., 1994). Carbhydrates translcated t sink rgans f a grapevine culd riginate fim newly fixed carbn, r be derived frm starch r frm carbhydrate reserves in wd (Cmbe. 1989; Alni et al., l99l ; Williams and Biscay, 1991 ; Glad et al., 1992b; Candlfi-Vascncels et al., 1994)). It may als take sme time fr newly fixed carbhydrates t reach the sink rgans such as berries. Fr example, Hale and Weaver, (1962) reprted that the labelled CO2 fed t grapevine leaves was detected in the berries after 6 hurs. Chaumnt et al. ( 1994) shwed that bth starch and sucrse accumulate in grapevine leaves during the phtsvnthesis perid, but while sucrse levels increased at early mrning and remained almst cnstant during the rest f the day with a slight decrease during the night. starch cntent f the leaves increased at mid afternn and decreased during the dark perid. Jhnsn et al. (1982), shwed that carbn dixide enrichment

114 97 resulted in accumulatin f high levels f starch in the leaves f treated grapevines cmpared t the vines grwn in ambient carbn dixide. Accumulatin f carbhydrates in excess f exprt, prvides a mechanism by which, in spite f remarkable variatins in the availability f light and carbn dixide, leaves tend t ensure a cntinuus day and night supply f assimilated carbn fr plant metablism and exprt f assimilate t the varius sink tissues (Geiger, 1987; H, 1992; Chaumnt et a1.,1994). Althugh the data presented here are nt sufficient fr exact determinatin f the bichemical mechanisms respnsible fr the changes, they prvide an imprtant fundatin fr future bichemical investigatins f assimilate translcatin in the grapevine. Further studies are needed t check seasnal patterns and the mechanism respnsible fr diurnal changes between day and night and during the grwing perid, and t quantiff C and N inflw int fruits Grapevine leaf exudate The results f analysis f leaf exudates shwed that the phlem sap depafing frm the leaves twards the sink rgans cntains sucrse and als amin acids and ptassium (Table 6.9). There are reprts indicating that hexses are present at high levels in grapevine leaf tissue (Kliewer, 1965; Kliewer and Nassar, 1966; Sepúlveda and Keliwer, 1986; Hunter et al., lgg4) but the results here indicated that the levels f hexses in phlem exudates are very lw (Table 6.9) suggesting that sugars are nt translcating ut fthe leaves as hexses. In the samples cllected frm leaves at the berry ripening stage sucrse, ptassium and amin acids were exuded at higher levels in samples cllected in 1995 (in the dark) cmpared t thse cllected in 1994 (in the light). The pririty f carbn exprt frm the leaves under lw light r in darkness and reductin f the rati f exprlstarch synthesis under high light has been reprted fr sme plant species (King and Zeevaart 1974; Maillard and Prcher, l99l). There are sme reprts suggesting selective exprt f amin acids frm leaves via the phlem (Weiner et al., 1991; Riens et al., ). The prfile f amin acids in the exudates f grapevine leaves was slightly different fim what was detected in peduncle exudates, shwing higher levels f alanine, serine and prline. Thus there may be sme selectivity in the exprt f amin acids via the phlem.

98 6.5 Cnclusin The research presented in this chapter demnstrates that the EDTA-facilitating methd is apprpriate fr cllecting grapevine phlem exudates.

115 Cnclusin The research presented in this chapter demnstrates that the EDTA-facilitating methd is apprpriate fr cllecting grapevine phlem exudates. The analysis f samples cllected using this methd revealed that sucrse is the main carbhydrate translcating in the grapevine phlem sap. Ptassium is the majr mineral in in grapevine phlem exudate cllected frm grapevines. Amin acids are imprtant cnstituents f grapevine phlem exudates with glutamine being the principal amin acid. The amunts f these principale cmpnents cllected frm the cut peduncle were smaller than amunts calculated t flw int each berry in situ. The cmpsitin f grapevine phlem sap (sugars, amin acids' and ptassium) is subject t seasnal changes. There is clear diurnal pattern f translcatin f sucrse and ther metablites in grapevines with higher rates f assimilate transprt at night.

99 Chapter 7 Assimilate transprt in the grapevine ; secndary metablites Cntents Page 7.1 Intrductin...-... -----.11 T.2Watenals and methds.....--'.---.--- 13 7.2.lPlartmaterials... -..-.-----.. 13 7.2.2 Other materials.

116 99 Chapter 7 Assimilate transprt in the grapevine ; secndary metablites Cntents Page 7.1 Intrductin T.2Watenals and methds ' lPlartmaterials Other materials Phlem sap cllectin Standard methd I 3 Exclusin f EDTA Girdting Vineyard cntrl samples Washing slutin Glycsyl glucse determinatin The secndary metablite aglycne analysis Results G-G in phlem exudates frm peduncle laterals and leaves and cmparisn with the G-G f whle berry hmgenates , Secndary metablite aglycnes in phlem exudates f grapevines In phlem exudates frm peduncle laterals lnphlemexudatesfrmpetiles Glycside analysis f peduncle The effect n the cmpsitin f secndary metablites f variatins in the exudate cllectin technique Omitting EDTAfrm the cllectin slutin 18 Girdling lretment Secndary metablites present in grapevine tissues 19 Plant tissue surfce cmpunds In the wash slutin Effect f the diurnal pattem f phlem flux n the mnterpenes cllected.._ Discussin l l l

1 7.4.1 Glycsylated secndary metablites.. - -... -... - -. -. 7.4.2 Seasnal variatins and varietal differences. - - - - i- - - - - - - - - - - - - 7.4.3 Treatments t interfere with the phlem sap flux---.

117 Glycsylated secndary metablites Seasnal variatins and varietal differences i Treatments t interfere with the phlem sap flux ,4 Cmpunds washed frm plant tissue surface and wund Peduncle tissue analysis The cncentratin f secndary metablite glycsides in the phlem exudate samples with high and lw sucrse cncentratins """""" Cnclusin -...

118 11., 1. i,,1,,i''i lj i,,,,, l',",i,{.,. 1 l'ij -ir,- n. 7.1 Intrductin Phlem transprt in vascular tissues f plants enables the transfer f phtsynthetic assimilates frm mature leaves t sink tissues in the rest f the plant. Exprt f phtsynthetic prducts frm surce leaves is dependent n many cmplex metablic events that cntrl the prductin f slutes which are translcated in phlem (Stitt and euick, 1989; Geiger and Fndy, 1991). Fruits are strng sinks fr assimilates translcating via the phlem (Cmbe, 19S9). The assimilates at the first stage f fruit develpment are used fr tissue develpment, and then accumulate as strage prducts. Sucrse is reprted as a majr cnstituent f phlem sap translcating in almst all plants studied s far, including grapevines (Glad et al.,i992a). In the ripening phase, especially after the lag phase at veraisn, grape berries accumulate a wide range f secndary metablites, including flavur cmpunds, alng with sugars and ther metablites (Wilsn et al., 1984; Strauss et al., 1986; Williams et al., 1989; Shure and Acree, 1994). Berry sftening, sugar and secndary metablite accumulatin are cm.mn ripening changes that are accmpanied by an increase in the respiratry qutient f the berries. This is due t a change in respiratry substrates as, during this perid, the berry underges a cnsiderable bichemical differentiatin (Brady, 1987; Pandey and Farmahan, 1e77). Bisynthesis and accumulatin f flavur cmpunds in grapes have been the subject f extensive studies (Williams et al ; Park et al., 1991;V/illiams, 1993). The mst cmmn secndarv metablites that accumulate in grapes include flavur cmpunds in flral grapes (Wilsn et al : Williams et al., 1985; Park et al., 1991), plyphenlic cmpunds (Cnsidine. 1979: Hard and O'Brien, 1988; Hawker et al., 1972) and anthcyanins in clured cultivars (Pirie and Mullins. 7976, 7977, 198). Flavur cmpunds (e.g. mnterpenes) are trace cnstituents f grape berries, and even in juice f flral cultivars such as muscats mnterpene flavur cmpunds are present at cncentratins f nly 1-2 mgll (Gunata et al., 1985a). Since leaves are the primary surce f phtsynthetic metablites translcating t fruits. grapevine leaf cmpsitin has been studied; secndary metablites are reprted as imprtant cnstituents f leaves f several plant species including the grapevine (Wildenradt et al : Gunata et al , 1985a; Otsuka et al., 1989; Schulz and Stahl- Biskup, 1991; Humpf and Schreier,lggl: Tazaki et al., 1993; Skurumunis and Winterhalter, 7994 ). Terpenls and armatic alchls in free and glycsidically bund

119 frms were reprted in Muscat f Alexandria (syn. Muscat Grd Blanc) leaf tissues (Gunata et al., 1986). It was further shwn that the grapevine leaf was richer in free and bund armatic alchls than the berry, and terpenyl glycsides in the leaves increased prgressively with grape maturatin (Gunata et al., 1986). Wildenradt et al. (1975) reprted six-carbn cmpunds and als terpenes at high levels in grapevine leaves f the cultivar'chenin Blanc' amng the 84 different vlatile cmpunds identified in a leaf extract. The bservatins f several researchers that similar secndary metablites are present in grapevine leaves and in fruits has led t the suggestin that these cmpunds are prbably being translcated frm leaves t the fruits (Gunata et al., 1986; Skurumunis and Winterhalter, lg94). Hwever, apart frm this similarity in cmpsitin and ther circumstantial evidence, such as that arising frm the bservatins reprted in Chapter 3 f this thesis, there is n reliable evidence t supprt the cncept f translcatin f secndary metablites frm leaves t fruit. The increase in mnterpene flavur cmpunds and ther secndary metablites in grape berries during the ripening stage culd result frm the cntributin f phlem cmpnents t berry grwth and cmpsitin, r these cmpunds may be synthesised in the berry independently. The data in Chapter fur strngly suggest that secndary metablites are synthesised in the berry, als the pssibility remains that berries f particular cultivars may have mechanism t selectivly imprt and accumulate these cmpunds (see sectin 4.4). The imprtant questin is whether secndary metablites are actually translcating t grape berries via phlem sap. Furthennre, if translcatin f secndary metablites is ccurring in the grapevine, is this a mechanism fr the accumulatin f flavur cmpunds alternative t the direct bisynthesis f thse cmpunds in the berry itself? There is a lack f infrmatin n the exact site f bisynthesis f flavur cmpunds and the relatinship between leaves and berries regarding secndary metablite accumulatin in the fruit during berry maturatin and ripening. Althugh phlem sap is the mst imprtant surce f slute accumulatin in berries (Cmbe et al., 1987b). Phlem sap has nt been analysed fr flavur cmpunds in any plants, therefre cllectin and analysis f phlem sap culd help t answer these questins. An understanding f the bisynthesis and accumulatin f grape flavur cmpunds may have cnsiderable practical cnsequences by prviding valuable guidance in the applicatin f viticultural practices t ptimise flavur prductin in fruit. This knwledge als culd prvide a fundatin fr genetic engineering f grapevines. T prvide mre direct evidence f any pssible transprt f secndary metablites in phlem sap, the rle f assimilates translcating frm leaves t the berries has been investigated thrugh analysis f the phlem sap. This has been achieved by determining the prf,rle f the secndary metablites in grapevine (Vitis vinifera L.) phlem sap 12

13 exudates cllected frm bth leaf petiles and peduncle laterals at three intervals thrughut the grwing seasn, and at six hurly intervals thrughut a 48 hur perid. 7.2

120 13 exudates cllected frm bth leaf petiles and peduncle laterals at three intervals thrughut the grwing seasn, and at six hurly intervals thrughut a 48 hur perid. 7.2 Materials and methds Plant materials Described in chapter 6, (6.2.I) Other materials Described in chapter 6, (6.2.2) Phlem sap cllectin Standard methd Phlem sap was cllected as described in chapter six, (6.2.3). This prtcl is called the standard methd in this chapter t allw differentiatin when the prtcl was mdified t investigate treatments invlving samples cllected frm nn-phlem surces. Exclusin f EDTA T assess the effect f EDTA n the cmpsitin f phlem exudate, the EDTA was excluded frm the cllectin slutin used in the standa d methd and nly HEPES buffer (1 mm; ph 7.5) was used fr exudate cllectin as a cntrl treatment. Girdling The girdling methd was described in chapter 6 and illustrated in Figure Sampling nd analysis rtifcts Samples f the buffèred cllectin slutin, befre and after cntacting the blue-tack adhesive that was used t hld the cllectin tubes in place, were enzyme hydrlysed, extracted and subjected t GC-MS analvsis. The peaks bserved in the chrmatgrams f these samples were assigned as sampling and analysis artefacts. These artefact peaks were subtracted frm the chrmatgrams f the phlem exudate analysis. Vineyard cntrl sampl es T investigate cmpnents cntributed frm the plant surfaces a series f samples was cllected as vineyard cntrls. ln these samplings different tissues with n cut r wund were dipped int the buffered cllectin slutin fr three hurs. These slutins were then enzyme hydrlysed. extracted and subjected t GC-MS analysis.

121 ÍYashing slutin As described in chapter 6, (6.2.3) befre the main sampling tubes cntaining buffer slutin t be attached t the peduncle lateral, the cut end f the laterals were washed with the same buffer slutin by rinsing them fr 5-1 minutes. This slutin was taken fr analysis as washing slutin t determine the cmpunds that came ut f cut cells and wunds, dwing the washing prcedure Gly csyl glucse determinati Glycsyl glucse determinatin was dne as described in Chapter 3, (3.2'5)' The secndary metablite aglycne analysis The secndary metablite aglycne analysis was dne as described in Chapter 4, (4.2.4)and the cmpunds we e identified and quantified using the same methd as emplyed in chapter fur Results G-G in phlem exudates frm peduncle laterals and leaves and cmparisn with the G-G f whle berry hmgenates Phlem exudate was cllected frm fruits and leaves f grapevine cultivars f Muscat Grd Blanc, Sultana, and Shiraz as describedin The phlem sap exudates were analysed fr glycsidically bund secndary metablites (G-G) as described in chapter 3, (3.2.s). Results f G-G analvses f the phlem sap exudates are given in Table 7.l and the data are cmpared with G-G values f berry hmgenates made n fruit sampled and analysed at the same develpment intervals.

122 15 Table 7.1 Glycsyl glucse levels measured in phlem sap exuded int buffer slutin frm peduncle laterals and leaves, and als in berry hmgenates f Muscat Grd, Sultana and Shiraz at three berry develpment stages' nml G-G/ 1.5 ml f buffer slutin. * tml G-G per g fr.wt. Cultivars Peduncle exudate Leaf exudate Berry hmgenate BVVRBVVR BVVR Muscat Grd Sultana Shiraz * The exudatin duratin time fr fruit bunches was 4 hurs and fr leaves was l6 hurs - n data available; BV= befre veraisn, V= veraisn, and R: ripe' Each value represents mean f three replicates Secndary metablite aglycnes in phlem exudates f grapevines In phlem exudates frm peduncle laterals After the glycsidically bund fractin f the cllected phlem exudates were islated by retentin n Cl8 reverse phase Sep-pak cartridges, the islated fractins were enzymatically hydrlysed. liquid-liquid extracted, and the aglycnes analysed by GC-MS. Cmplex mixtures f glycsylated secndary metablites were present in the cllected slutins as can be seen frm the chrmatgrams f the aglycnes presented in Fig. 7.1' (ABC). Cmpunds riginating frm the enzyme and/r slvent were identified by GC-MS analysis f a HEPES buffer cntrl which was enzyme hydrlysed and liquid-liquid extracted in the same way as phlem exudate samples. After eliminating these artefacts frm the vlatiles recrded in the GC-MS analyses, the remaining cmpunds were assumed t be f plant rigin. Analysis values are shwn cllectively in Table 7.2 and the remaining discussin in this sectin f the chapter cncerns these cmpunds nly. Mnterpenes, including several with recgnised flavur prperties, made up a ntable prprtin f the vlatile aglycnes frm the bund secndary metablites in the phlem exudates f grape peduncle. The cncentratin f mnterpenes in l.5 ml aliquts f the phlem exudate cllectin slutins btained in 4 h frm fruit peduncle laterals f Muscat Grd, Sultana, and Shiraz at three ripening stages are prvided in Table 7.2.A The

16 cncentratins f C13-nrisprenids and benzene derivatives in the slutins are similarly shwn in Table 7.2.

123 16 cncentratins f C13-nrisprenids and benzene derivatives in the slutins are similarly shwn in Table Frm the results in Table 7.2.A it is evident that the exudates cllected frm Muscat Grd peduncle were richer in glycsides f mst f the mnterpenes bserved than were the exudates frm either Sultana r Shiraz. This bservatin parallels the findings in Chapter 4 in that Muscat Grd berries are richer in mnterpene glycsides than Sultana r Shiraz berries In phlem exudates frm petiles The phlem exudate f detached leaves was als cllected fr Muscat Grd, Sultana, and Shiraz grapevines. The glycsidically bund fractin f the samples were islated, enzymatically hydrlysed, and arralysed by GC-MS. Results f these analyses fr the mnterpenes and fr C13 nrisprenids and benzene derivatives bserved are given in Tables Of the three varieties, Muscat Grd leaf phlem exudate appeared t have the highest cncentratin f mnterpene glycsides. Only the glycside f geranil was crilnn t all three varieties.

124 A 17 c B MC 5a Fig. 7.1 GC-MS chrmatgrams f the aglycnes islated frm phlem exudates cllected frm ripe fruit bunch laterals f Muscat Grd (A), Suløna (B), and Shiraz (C) grapes.

18 7.3.2.3 Glycside analysis f peduncle The methd f phlem sap cllectin used in these experiments invlved the cllectin slutin being in cntact with cut and expsed vine tissue surfaces.

125 Glycside analysis f peduncle The methd f phlem sap cllectin used in these experiments invlved the cllectin slutin being in cntact with cut and expsed vine tissue surfaces. It was cnsidered necessary therefre t determine the extent, if any, f cntaminatin f the cllected phlem sap by nn-phlem vine tissue surces during the sampling prcesses. Peduncle tissue, with the vascular bundles therein, is a passage fr metablites translcating t the fruits, mainly frm leaves and als frm ther parts f the plant' The peduncle tissue f the three cultivars f this experiment was extracted, the glycsides islated, hydrlysed, and analysed in the same way that phlem exudates were prcessed. The results f these analyses a e shwn in Tables 7.2.C' The peduncle tissues fr all th ee cultivars were a rich surce f sme f mnterpene, C13-nrisprenids, and benzene derivatives aglycnes (Table 7.2.C). Va ietal differences were evident in the cncentratins f many f the cmpunds bserved The effect f variatins in the exudate cllectin technique n the cmpsitin f secndary metablites T investigate the ccurïence in the cllectin slutins f cmpunds f nn-phlem rigin, sme treatments were applied befre and during the cllectin. These treatments were designed t inhibit, r cut fi the phlem sap flux. The samples cllected in this way were analysed fr aglycnes f secndary metablites. Omitting EDTA frm the cllectin slutin As was shwn in Chapter 6 (6.3.3) excluding EDTA frm the cllectin slutin stpped phlem sap flux as indicated thrugh a terminatin f sucrse flux. The first treatment therefre was an missin f EDTA frm the buffer slutin used fr sap cllectin. Excluding the EDTA. resulted in a majr reductin in the number and cncentratin f aglycnes which were detectable in the hydrlysed exudate slutin, in cmparisn with the standard cllectin prtcl (see Table 7.2.D, Cntrl I). This reductin was evident fr samples cllected bth at veraisn and early ripening stages (data nt shwn). Girdling treslment The secnd treatment t stp phlem sap translcating in vascular bundles was girdling. Analysis f the exudates cllected frm girdled bunches (Figure 6.4) shwed that girdling n the cane abve and belw the fruit bunch, befre sampling eliminated phlem sap flux as demnstrated by the absence f sucrse flux. The girdling was dne abut l8-2 hurs befre sampling; the samples were cllected in the presence f EDTA and then analysed fr secndary metablite aglycnes. Data fr these analyses were cmpared with data fr

126 19 samples taken in parallel frm un-girdled bunches with EDTA mitted and by the standard prtcl. The results f analysis f exudates cllected frm girdled bunches revealed that the girdling treatment, despite stpping sucrse flux, had little r n effect n the number and cncentratin f the aglycnes cmpared t the samples cllected frm nn-girdled bunches (see Table 7.2.D, Cntrl II) Secndary metablites present in grapevine tissues Plant tissue surface cmpunds Anther set f samples was cllected t determine the pssible cntributin t the secndary metablites bserved in the cllectin slutins made by plant tissue surfaces. Fr this purpse the samples were cllected by putting grapevine tissues (rgans) such as tendrils, uncut bunch stems r the dried end f canes with ld cuts, in buffer slutin. In rder t get detectable levels f cmpunds in the sampling slutin frm these surces, the cntac t area between the tissue and the slutin was made as large as pssible. Therefre, the results f analysis f these samples (see Table 7.2.D, Cntrl III, vineyard) are imprtant because f the presence f the cmpunds bserved rather than their cncentratin levels. The analysis f these samples shwed that mst f the cmpunds detected were similar t thse f exudates cllected frm peduncle laterals. In the wash slutin As previusly described(7.2.3), fr phlem sap cllectin in this experiment, the cut end f a peduncle lateral was dipped int the buffer slutin immediately after cutting and the cut surface was rinsed fr 5-1 minutes t wash cmpunds frm the cut cells befre exudate cllectin. The slutins which were used fr washing the cut ends f the laterals were pled and analysed. The results f this analysis f wash slutin are given in Table 7.2.D. Cntrl III. (washing slutin). The results shwed that sme mnterpenes and als sme Cl3-nrisprenids and benzene derivatives were present in the washing slutin. The levels f snte cmpunds in the washing slutin were higher than had been bserved in samples f phlem exudates fr the same cmpunds in spite f the differences in time length fr washing and exudate cllecting (5-1 minutes and 3-4 hurs respectively) Effect f the diurnal pattern f phlem flux n the mnterpenes cllected The phlem sap exudates cllected during day and night perids shwed a diurnal pattern fr sucrse flux (see chapter 6). The analysis f secndary metablites in samples cntaining high and lw levels f sucrse cllected frm the same bunches during day and

127 night perids culd shw the cntributin f phlem sap flur in the levels f secndary metablites exuded int the cllectin slutin. A series f cllectins made during aday, a night and the fllwing day, yielded samples with lw levels f sucrse, cllected ver 12 h during the first daylight perid, samples with high levels f sucrse, cllected ver 12 h during the night perid, and samples with lw levels f sucrse, cllected ver 12 h during the secnd day perid. These were taken fr hydrlysis and GC-MS analysis f the majr aglycnes. Data fr the cncentratin f tw f the aglycnes detected in these samples, i.e geranil and 3,7-dimethylct-2-ene-1,7-dil are presented infig.7.2. The mnterpene aglycnes were present in all th ee cllected samples; thse with lw sucrse cncentratins had higher levels f the cmpunds detected in the buffer slutin than the samples with high sucrse levels. 11 I Geranil 14 c J a J E õ t I n 3,7-Dim"thylcÞ2ene.1,7 dil ! ET c c f N a N ô q N N N q N d Sampllng tlme cu se 6 (\l Fig. 7.2 The cncentratin f secndary metablite aglycnes shwn as ba s (right axis) in the phlem exudate samples cllected ver a 36 h perid. The sucrse cncentratin (left axis) in the same samples als shwn.

128 Discussin Glycsylated secndary metablites T investigate the site f bisynthesis f flavur cmpunds accumulating in grape berries, and the pssibility f translcatin f these cmpunds via the phlem sap, grapevine phlem exudate frm peduncle and leaves was cllected. These slutins were initially analysed fr ttal glycsidically bund cmpunds, i.e. G-G. Data fr these analyses (Table 7.1) shw that mst f the cllectin slutins cntained glycsides. The peduncle exudates f Muscat Grd and Sultana, cllected befre veraisn had higher levels f glycsides than the exudates cllected at later stages (Table 7.1). This trend was als seen in the G-G values btained fr the fruit hmgenates. Shiraz fruit bunch phlem exudates and berry hmgenates als shwed a parallel develpment f G-G cncentatin with berry develpment, but ppsite t that fund fr Muscat Grd and Sultana. The exudates cllected in 16 h frm leaves at veraisn and at berry ripeness had cncentratins f glycsides higher than were fund in the 4 h cllectin slutins f peduncle exudate. T investigate the individual vlatile secndary metablites that were bund as glycsides, larger vlumes f the phlem exudates were cllected, enzymatically hydrlysed, and the released aglycnes were analysed by GC-MS. The results f these chrmatgraphic analyses shwed mre than 6 cmpunds were bservable in the enzymic hydrlysates f the exudates cllected frm fruit bunches f the grapes (Data is nt shwn). Mst f the aglycnes detected and identified in these experiments were in ne f fllwing categries; mnterpenes, Cl3-nrisprenids, benzene derivatives, and fatty acid-derived cnstituents Seasnal variatins and varietal differences T study the seasnal variatins f the glycsidically bund vlatiles, peduncle exudate samples were analysed at three different stages f berry develpment. Althugh the cncentratins f mst bserved mnterpenes in the cllected exudates varied with the develpmental stage, n cnsistent patterns were evident. Thus, unlike the high G-G cncentratins seen befre veraisn in the peduncle exudate f Muscat Grd and Sultana, the cncentratins f nly a few individual mnterpene glycsides, e.g. thse f citrnelll, 3.7-dimethyl-1,7-ctandil and 3,7-dimethylcta-2-ene-1,7-dil in Muscat Grd shwed such a trend (Table 7.2.A). Several individual mnterpene glycsides were maximal at ripeness. Frm the data in Table 7.2.A it can be seen that glycsides f C13 nrisprenids and benzene derivatives were significant cnstituents f the phlem sap exudates cllected

frm all three varieties. As was the case with the mnterpenes, n bvius pattern f develpment was evident in the cncentratin f these aglycnes with ripening.

129 frm all three varieties. As was the case with the mnterpenes, n bvius pattern f develpment was evident in the cncentratin f these aglycnes with ripening. Such incnsistent patterns f secndary metablite ccurrence in the phlem sap exudates with berry develpment give little supprt t translcatin as the mechanism f secndary metablite accumulatin. The decreasing cncentratin f secndary metablite glycsides in the exudates during the grwing seasn is due pssibly t the increasing maturity f the peduncle tissue rather than changes in the cmpsitin f phlem sap as it has been recrded that yung plant tissues nrmally cntain mre f these cmpunds (Stahl-Biskup et al. 1993). There were als varietal differences in the prfiles f the secndary metablites detected in the peduncle exudates. Regarding mnterpene cmpunds, Muscat Grd peduncle exudate appeared t cntain a higher cncentratin f these cnstituents than did Sultana r Shiraz exudates (Table 7.2.A). The prfile f C13-nrisprenid and benzenederivative cnstituents f the exudates were similar fr all th ee cultiva s. Fewer mnterpene aglycnes were detected in the hydrlysates f leaf phlem exudates f the grapevines than were bserved in crrespnding hydrlysates f the phlem exudates f the fruit bunches (cmpare data in Table 7.2.8). Of the three varieties, Muscat Grd leaf phlem exudate appeared t have the highest cncentratin f mnterpene glycsides. Only glycsides f the furan linall xides and geranil were cmmn t all three varieties. N such differences in either cmpund type r cncentratin were evident fr the C 3 nrisprenid and benzene derivative glycsides Treatments t interfere with the phlem sap flux In chapter six it was shwn that sme treatments culd stp phlem sap flux. Phlem exudates were cllected fim Muscat Grd grape bunches after applicatin f these cntrl treatments. The samples u,ere analysed by GC-MS t bserve the effects f interfering with the phlem flux n the glycsylated secndary metablites and thus determine whether these cmpunds were translcated in the phlem sap. Excluding EDTA frm the buffer slutin stpped exudatin f sucrse int the buffer slutin ( Cntrl. I.). Tw samples cllected at different times with EDTA mitted frm the cllectin slutin shwed diminished levels f secndary metablite aglycnes (see Table 7.2.D). Samples were cllected frm girdled bunches, using buffer slutin cntaining EDTA, The results f the analyses f exudates cllected frm girdled bunches cmpared t thse f nn-girdled bunches. ie using the standard prtcl, shwed much smaller differences in bth the number and cncentratins f the cmpunds detected than betweetr the samples cllected frm nn-girdled bunches with and withut EDTA (seetable 7.2.D,

113 Cntrl II). These bservatins demnstrated that althugh girdling stpped the sucrse flux, the ccu 1ence f the secndary metablites in the cllectin slutins was largely unaffected.

130 113 Cntrl II). These bservatins demnstrated that althugh girdling stpped the sucrse flux, the ccu 1ence f the secndary metablites in the cllectin slutins was largely unaffected. This result, in cmbinatin with the absence f aglycnes in cllectins made with the missin f EDTA, indicated that the presence f EDTA in the cllectin slutins rather than phlem sap, was respnsible fr the ccurrence f the secndary metablites in the cllectins. EDTA can mve frm slutin int the vascular bundles f the vine tissue and exsert its effect inside the vascular cells (King and Zeevaart 1974) causing elutin f any secndary metablites resident in the tissue cells. The parenchymatic cells and the tissue surrunding vascular tissue are pssible surces f the secndary metablites Cmpunds washed frm plant tissue surfaces Cntrl samples fr determinatin f any alternative surces f secndary metablites were cllected by washing uncut Muscat Grd grapevine tissues surfaces int the buffer slutin. Data fr the analysis f these samples in Table 7.2.D, Cntrl III shw that the cllectin slutins made frm plant tissue surfaces cntained a number f the mnterpenes, C 3-nrisprenids and benzene derivatives seen in the exudate cllectins made under standard cnditins. It appears that these cmpunds in sme way have exuded t the surface f the tissues and culd be washed ff the external surface f the grapevines in the presence f the EDTA. This suggests that secndary metablites have exuded frm the plant cells, thrugh the epidermis, t the tissue surface. Such migratin has been shwn fr cmpunds that accumulate within grape berries as strage prducts, these can pass thrugh the cuticle and epicuticular layer and reach the surface f the berry (Padgett and Mrrisn 199). The results f the experiments reprted here shw that this type f secretin f cmpunds pssibly ccurs in all aerial parts f grapevines. As anther cntrl. the slutin used fr washing the wunded r cut end f the laterals frm which the exudates \À'ere cllected, was analysed (Table 7.2.D, Cntrl III ). The results shwed that a f'ew mnterpenes, ntably geranil and geranic acid, and als sme C13-nrisprenids and benzene derivatives were present in the washing slutin. The levels f sme cnipunds in the washing slutin were higher than had been bserved in samples f phlem exudates, despite the differences in the duratin f the washing and exudate cllectin regimens ( 5-1 minutes and 3-4 hurs respectively). The relatively high levels f tlie cmpunds detected suggests that they prbably are present in high cncentratins in the peduncle tissue cells as well as n the tissue surfaces.

114 7.4.5 Peduncle tissue analysis The analysis f peduncle tissue by GC'MS shwed that sme mnterpenes as well as nrisprenids and benzene derivatives were highly cncentrated in this tissue (Table 7.2.

131 Peduncle tissue analysis The analysis f peduncle tissue by GC'MS shwed that sme mnterpenes as well as nrisprenids and benzene derivatives were highly cncentrated in this tissue (Table 7.2.C). Althugh the cmpunds present in the peduncle tissue were cmmn t mst f the phlem exudate samples sme were clearly nt. Peduncle tissue was rich in mnterpene cmpunds such as geranil, nerl, citrnlll, linall xides, Z-2,6- dimethylc ta-2,7 -diene- 1,6-dil, 3,7-dimethy- I,7-ctandil, E-2,6-dimethyl cta'2,7 - diene-1,6-dil, and 3,7-dimethy-1,7-ctandil. Hwever, there were ther mnterpenes in samples cllected frm peduncle lateral exudates which were nt in the peduncle tissue extacts e.g. -terpinl and citrnelll. In light f the data discussed in and7.4.4, and the similarities amng the cmpunds in the phlem exudates and thse in the peduncle tissue, it seem likely that the latter tissue is the surce f these cmpunds The cncentratin f secndary metablite glycsides in the phlem exudate samples with high and lw sucrse cncentratins It was assumed that if secndary metablites are translcating in the phlem sap, the exudates cllected at perids f high sucrse flux shuld cntain higher levels f these cmpunds as well. Based n this assumptin the phlem exudates with high and lw levels f sucrse that were cllected at the peak and trugh times f phlem sap sucrse flux were analysed fr secndary metablite aglycnes. It was fund that the exudates cntaining higher sucrse had lwer amunts f detected aglycnes (Fig. 7.2). The absence f a psitive crrelatin between sucrse cncentratin and the levels f aglycnes detected in the same exudates, strngly reinfrce the view that the secndary metablites are nt translcated in the phlem.

115 7.5 Cnclusin The experiments described in this chapter were aimed at establishing the presence, and pssible translcatin in grapevine phlem sap, f glycsides f vlatile secndary metablites.

132 Cnclusin The experiments described in this chapter were aimed at establishing the presence, and pssible translcatin in grapevine phlem sap, f glycsides f vlatile secndary metablites. Whilst many vlatile secndary metablites were detected as aglycnes in phlem exudate cllectins, the ccurrence f these cmpunds apþears t be the result f their extractin frm vine tissues rather than their presence in the phlem sap. These deductins were drawn frm the results f aglycne analyses applied in three experimental appraches. In experiments designed t interfere with the phlem sap flux, the imprtance f EDTA in the cllectin slutins as an agent facilitating the extractin f the glycsides frm the vine tissues was shwn. Other experiments, invlving analysis f hmgenised peduncle tissues and f cllectins made frm plant tissue surfaces, cnfirmed the ubiquity f the secndary metablite glycsides in all tissues examined. Finally, the absence f a crrelatin between the amunts f tw majr secndary metablite aglycnes in exudate cllectins and phlem sap flux, the latter indicated by sucrse cncentratins in the cllectin slutins, shwed that the secndary metablites fllwed n diurnal pattern like the phlem flw. These results, tgether with the data in Chapter 4, shw that secndary metablites are nt transprted int fruit frm ther parts f the vine. It is cncluded that the berry itself is the site f bisynthesis f these cmpunds.

Table 7.2 The aglycnes determined in hydrlysates f phlem exudate cllectins and tissue extracts 116 Tal lc 7.2.4 Mnterpene, C 3-Nrisprenids and benzene Table 7.2,8 Mnterpene, C 3- Tal lc 7.2.C The mnterpene, Tnl le 7.

133 Table 7.2 The aglycnes determined in hydrlysates f phlem exudate cllectins and tissue extracts 116 Tal lc Mnterpene, C 3-Nrisprenids and benzene Table 7.2,8 Mnterpene, C 3- Tal lc 7.2.C The mnterpene, Tnl le 7.2 D The aglycnes detected in the samples frm cntrl T able Mnterpene, C t 3 derivative aglycnes determined in hydrlysates f phlem sap exuded int a buffered cllectin slutin frm fruit bunch laterals. Data are in pml/ I.5 ml f cllectin slutin/ 4h. / lateral, (BV= befre veraisn, V= veraisn, and R= ripe). Nrisprenids and benzene derivative aglycnes determined in hydrlysates f le f petile exudate exuded int a buffered cllectin slutin. Data are in pml/ 1.5 ml f cllectin slutin/ l6h. / Ieaf, (V= Cl 3-urisprenid and benzene derivative aglycnes determíned in peduncle (bunch stem tissue) hmgenates (pml/ g fr.wt.). lreîtrnents that interfere with phlem flux cnsisting EDTA, girdlíng and als in tissue surface and wund washings. prnl/ I.5 ml f buffer slutin/ 4h, nrisprenids and benzene derivative cmpunds detected in Muscat Grd, Sultana and Shiraz berry hmgenates, þmu g fr. vrt.). and R= Mnterpenes Furan linall xide-l Furan linall xide-2 Linall Htrienl Myrcenl cis-ociminl hans-ociminl þran linall xide-l a-terpinl þran linall xide-2 Cirnelll Nerl Geranil Geranic acid 2,6-Dimethylcta-3,7-dienc-2,6-dil 3,7-Dimethylcta-1,6-diene-3,5-dil 2,6-Dimethylcta- 1,7-diene-3,6-dil Z-2,6-Dimethylcta-2,7-diene- 3,7-Dimethy- 1,7{ctandil 1,6-dil E-2,6-Dimethylcta-2,7-diene-1,6-dil 3,?-Dimethylcta-2-ene- 1,7-dil p.menth- l -ene-6,8-dil p-menth- I -ene-7,8-dil Ttal mnterpencs Sucrse C 3-Nrisprenids 3-Hydrxy-7,8-dihydr-b.inl 3-Ox-b-inl 3-Ox- a-inl 3-Hydrxy-5,6-epxy-Linne Grasshpper ketne Benzene deriv tives Trimethxy benzen metha rl Trimethxy phenl - : nt detected Trace = less than.i pml Muscat BV.4 lt2 6 t5 t One analysis was dne fr each datum J 2 )) VR Peduncle exudate Sultana BVV RBVVR Leaf ptrlem exudate Peduncle tissue extract M. Grd Suløna Shiraz M.Grd Sult na Shi az VRVRVRRRR 4 J 2t face 62955tt745t26?6 7S 86 t4 )) l6 24 )) 5 I lt 27 t5,) E 5. 7 l4 4 l il t l7 7 J t8 56 te æ t2 5 9 II t tsac M 26 9t4 5 t t.4 a 6 3 a il a I 9 5 bacc trace tracc tracc 23 t9n t.m 4 N l9 il t il face 6t t2 I il.t trace il trace tace trac trac bacc tace tracc tsace 599 l3t8t t99 face tracc bace tsacc Cntrl I Std,r N EDTA trace trace trace 8 t a7 trace 366/. 46t t t7ø 6t2 Muscat Grd exudate Benies II M. Grd Sultana Shiraz srd. N Girdled Vinyard Washing EDTA slutinr' RR R 8l 3 t65 I 8l tace 6 )) t 342 t t t4 trace ,,, 3 traße bace l8? 3r 925 t969 2t 24 trâce S 6l 5 Eacc 37 Uacc ? l2t t bace 41 tracc I Std. - Stândsrd mcthd fphlcm sap cllectin in bulfcr + EDTA slutin. +r The time duratin fr washing thc wund was 5-1 minutc, t il5 7æ t4 t5r t I e t 7 3 2æ 31 l (E ntmcasu cd 39 t 32 il 16?s e bace tsace tacc 7 2æ t l,m

Chapter I 117 General discussin Althugh knwledge f flavur precursrs and bund vlatile arma cmpunds in grape berries has been extended ver the last decade, much mre remains t be clarified including the

134 Chapter I 117 General discussin Althugh knwledge f flavur precursrs and bund vlatile arma cmpunds in grape berries has been extended ver the last decade, much mre remains t be clarified including the site f bisynthesis f these cmpunds in different tissues f the vine. Secndary metablites and, mst imprtantly, flavur cmpunds accumulate in the grape berries in glycsylated frms during berry develpment but very little is knwn abut where they are made. Glycsylated secndary metablites can accumulate in almst all living tissues f the plant (Sugisawa et al., 1988; Lukner, 199), thugh in remarkably different cncentratins. The leaves f plants are the sites f active phtsynthesis and carbn assimilatin and leaves als accumulate glycsylated secndary metablites (Tazaki et al., 1993). The secndary metablites that accumulate in grape berries culd either be translcated frm phtsynthetically active sites such as leaves t ther sink rgans, r synthesised in the tissue itself. In this wrk and thrugh data presented in Chapter 3 it has been shwn that the grapevine leaves accumulate large quantities f glycsylated cmpunds including thse f vlatile secndary metablites. The immature berries f grapes accumulate glycsylated secndary metablites during the first phase f grwth. The berries f flral grape Muscat Grd had a different pattern f accumulatin f ttal glycsides t the nn-flral Sultana grape berries. Muscat Grd berries accumulated mre secndary metablite glycsides during ripening than Sultana berries. in fact accumulatin f these cmpunds in sultana almst stpped by the beginning f ripening. This ccurs despite the fact that Sultana berries had higher levels f glycsides than Muscat Grd berries at the early stages f berry grwth and develpment. Thus, metablically active green tissues f yung berries, and als leaves, are rich in secndarl, metablite glycsides and the accumulatin f vlatile secndary metablites that are specific t the flavur f the berry ccurs later. This phenmenn, which has nt been reprted befre, shws that differences exist between the metablic activities f berries f Muscat Grd and Sultana cultivars. This suggest, firstly, that glycside accumulatin in the berries is nt linked t sugar accumulatin and des nt ccur thrugh the same mechanism, althugh bth prcesses can ccur cincidentally. This is in agreement with bservatins reprted by Wilsn et al. (1984), Williams et al. (1985) and Park et al. (1991) wh cncluded that bisynthesis and accumulatin f mnterpene glycsides in grapes is independent f sugar translcatin, Secndly, the high levels f glycsides in the leaves dse nt necessarily increase the levels f glycsides in the berries.

135 Frm the data presented in Chapter 3 and als that reprted by Gunata et al' (1986), Bravd et al. (1989), Skurumunis and Winterhalter (1994 ) it is knwn that many secndary cmpunds, including glycsides f flavur cmpunds are present in grapevine leaves. There are tw pssibilities; synthesis f these cmpunds in the leaves and transprtatin t the berries, r their frmatin in the berries themselves frm precursr mlecules imprted via the phlem sap. In this wrk the pssible translcatin f vlatile flavur cmpunds frm grapevine leaves t the berries r, altematively, their synthesis in the berries, were investigated using tw different methds i.e. bunch grafting between cultivars described in chapter 4, and phlem sap analysis' Mnterpene cmpunds are imprtant cnstituents f flral grape varieties such as Muscat Grd (Williams et al., 1982a, b, Wilsn et al., 1984; Gunata et al., 1985a), s they were the main cmpunds that were analysed in the fruit. The results btained fr mnterpene prfiles in the grape bunches after being grafted between cultivars indicated the exclusive rle f the berries in accumulatin f mnterpene cmpunds. The same spectrum f mnterpene glycsides accumulated in grape bunches, irrespective f whether the fruit develped n its wn vine r fllwing bunch grafting nt genetically different vines. Thus the ability t accumulate mnterpene glycsides resides within grape berries and is independent f the rest f the vine. Hwever, there still remained the pssibility that the mnterpene glycsides \ryere imprted int the fruit s that Muscat Grd berries, fr example, might have mechanisms t selectively imprt and accumulate these cmpunds, while these mechanisms were nt active in grapes f the nn-flral varieties. The grafting experiments d indicate that the inability f nn-flral grapes t accumulate mnterpene glycsides des nt result frm a lack f these cmpunds r their precursrs in the phlem sap circulating arund the vine, since fruit f Muscat Grd grafted nt these varieties still prduced the nrmal levels f mnterpene glycsides. ln Chapter 5 it was discvered that Muscat Grd berries accumulate anthcyanin pigments at the ver-ripe stage (Ghlami and Cmbe, 1995). Cnstantly similar prfìles f anthcyanin in the Muscat Grd berries frm grafted and nn-grafted bunches and als frm vines with different rt systems, supprt the view that berries are independent f leaves and rtstck fr synthesis and accumulatin f anthcyanins as well as ther secndary metablites, which was cncluded earlier frm the grafting experiment (Ghlami et al., 1995). It seems that the berry itself synthesises mnterpene flavur cmpunds, and anthcyanins, althugh precursrs f these cmpunds may be prduced in ther parts f the vine. The hypdermal cells f the berries are the mst prbable sites f bisynthesis and als strage f these secndary metablite cmpunds as was als suggested by Wilsn et al. (1986). The grape berries have a cntrlling rle n metablism f flavur cmpunds prduced and stred within them. 118

136 The phlem sap analysis methd was used t investigate any pssibility f glycsylated secndary metablites being translcated in grapevines. The samples cllected using the EDTA-facilitated methd shwed that the main carbhydrate translcating in the grapevine phlem sap was sucrse and ptassium was the majr mineral in. Amin acids were als imprtant cnstituents f grapevine peduncle exudates with glutamine being the principal amin acid in the sap. The cmpsitin f grapevine phlem sap exudates (sugars, amin acids, and ptassium) were subject t diurnal and seasnal changes. There was a clear and cnstant diurnal pattern f metablite translcatin in grapevines with high rates f assimilate transprt at night under nrmal cnditins. Diurnal variatins in the cncentratin f sucrse in phlem sap translcating t the berries, supprts the suggestins f Brwn and Cmbe (1985) that va iable phlem sap unlading int the applast has the primary cntrl ver sugar accumulatin in the berry rather than cmpartmentatin within pericarp cells. The cmpsitin f the phlem exudates and the experiments in which the canes were girdled and EDTA was mitted frm the buffered cllectin slutin, supprt the idea that the technique prvides a sample f phlem sap. Hwever, the calculatins presented in Chapter 6 indicate that the amunt f sucrse and amin acids cllected frm the peduncle laterals is insufficient t accunt fr the predicted rates f accumulatin f these cmpunds in the berry. This suggests that cutting the lateral and remving the fruit has a significant effect n the phlem sap flw frm the lateral. Fr this reasn, the data may nt prvide accurate quantitative data n the transprt f metablites int the fruit via the phlem. The results f GC-MS analysis shwed that the phlem exudates cllected frm bth fruits and leaves cntain substantial levels and cmplex mixtures f glycsylated secndary metablites. Hwever, the results btained in Chapter seven indicate that these cmpunds are nt translcating via phlem sap. Since, frm the beginning f grape berry ripening, mst f the fln, t the berry is prvided by the phlem (During et al., 1987; Findlay et al., 1987; Creasl' et al., 1993; Greespan et al.,1994), the increasing levels f sme amin acids such as arginine, alanine and prline, (Kliewer, 1969,197b) and als f the tripeptide glutathine in ripening grape benies (Adams and Liyanage, 1993) supprts the idea that these cmpunds are synthesised in the berries themselves frm precursrs imprted via phlem sap. and implies that the grape berries are actively cntrlling their strage pl f carbh -drates and amin acids as well as prteins (Tensniére et al., 1994). A cnsideratin f all the pathways fr amin acid metablism and secndary metablite synthesis indicates that grape berries are metablically active rgans and shuld nt be cnsidered nly as strage sites. Althugh it is difficult t determine whether secndary metablite mlecules in a phlem exudate slutin are actually in transit frm leaves t sink tissues, the data presented here, althugh insufficient fr exact determinatin f the bichemical mechanisms respnsible fr the changes, des prvide an imprtant 119

137 fundatin fr future bichemical investigatins f assimilate translcatin in the grapevine. tr'urther studies: Substantial advances have been made in research n flavur cmpunds in grapes, but bisynthetic studies n whle fruit have been limited. Hwever, lw levels f flavur cmpunds is a majr limitatin in studies n the bisynthetic pathways f secndary metablite s in Vitis vinifera species. Understanding the metablic prcesses and bisynthetic pathways f flavur cmpunds in grapes and the mechanisms cntrlling them is an imprtant aspect. Future research activities n flavur cmpunds shuld pay mre attentin t research at the cellular level particularly n the site f bisynthesis f imprtant flavur cmpunds in the grapevine. V/hen the site f bisynthesis f flavur cmpunds (i.e. mnterpenes) becmes knwn it may be pssible t alter varietal characteristics f grapes by applicatin f apprpriate techniques such as genetic engineering. Then it shuld becme pssible t adjust the vines t prduce grapes with desired flavur characteristics. The results f this thesis suggest further wrk shuld cncentrate n the bisynthesis f these cmpunds in the berries. Often prducts are synthesised at particular develpmental stages r in a specific rgan, suggesting that sphisticated cntrl mechanisms exist t regulate their synthesis (Lukner, 199). Plant cell culture is f ptential value fr studies f bisynthesis f secndary metablites (Charlwd et al., 1988; Cnstable, 1988; Sugisawa et a1.,1988 and references therein). The use f this technique and grape berry cell suspensin wuld seems t be anther methd available t investigate the site f bisynthesis f flavur cmpunds in grapevine and the ptential abilities f different plant rgans and cells in synthesising thse cmpunds. Further understanding f the mechanisms by which flavur cmpunds accumulate in grapes will require experimentatin t identify the actual sites f synthesis f these cmpunds and their cmpartmentatin within the berry and the rle f the bern' itself. The study f enzymes invlved in mnterpene bisynthesis pathways and the rle geranil plays will be extremely helpful in elucidating the bisynthesis f these secndary metablites. Further studies are needed t check seasnal patterns and diurnal changes at different times f the seasn and the effect f envirnmental factrs n these patterns especially sucrse diurnal pattern and quantitative measurement f C an N inflw int fruits. The technique used in this wrk was sufficient t cllect samples hpefully representative f grapevine phlem sap. nevertheless the rate f exudatin was much less than the nrmal flw f phlem sap ìnt the berry in situ. Therefre, further studies are required t imprve the phlem sap cllectin technique, testing buffers f differing cmpsitin and the effect f ther facilitating agents. 12

121 References Ackerman, LE., D.V. Banthrpe, W.D. Frdham, J.P. Kinder, and I. Pts, 1989, P-Dglucsides f arma cmpnents frm petals f Rsa species: Assay, ccurrence and bisynthetic implicatins. J. Plant Physil.

138 121 References Ackerman, LE., D.V. Banthrpe, W.D. Frdham, J.P. Kinder, and I. Pts, 1989, P-Dglucsides f arma cmpnents frm petals f Rsa species: Assay, ccurrence and bisynthetic implicatins. J. Plant Physil. 134: Adams, D.O., and C. Lyanage,1993, Glutathin increases in grape berries at the nset f ripening. Am. J. Enl. Vitic Alni, R., A. Raviv, and C.A. Petersn, 199I, The rle f auxin in the remval f drmancy callse and resumptin f phlem activity in Vitis vinifera. Can. J. Bt Andersn, P.C., and B.V. Brdbeck, 1989a, Chemical cmpsitin f xylem exudate frm bleeding spurs f Vitis rtundiflia Nble and Vitis hybrid Suwannee in relatin t pruning date. Am. J. Enl. Vitic. 4: Andersn, P.C., and B.V. Brdbeck, 1989b, Diurnal and tempral changes in the chemical prfile f xylem exudate frm Vitis rtundiflia. Physil. Plant. 75:63-7. Andersn, P.C., B.V. Brdbeck, and R.F. Mizell, III, 1995, Diurnal variatin in tensin, smlarity, and the cmpsitin f nitrgen and carbn assimilates in xylem fluid f Prunus persica, Vitis hybrid, and Pyrus cmmunis. J. Amer. Sc. Hrt. Sci. 12:6-66. Arnff. S., Translcatin frm sybean leaves. Plant Physilgy, 3: Baker, J., and C.F. Timberlake, (1985) The distributin f anthcyanins in grape skin extracts f prt wine cultivars as determined by high perfrmance liquid chrmatgraph,. Jurnal f the Science f Fd and Agriculttue 36, Banthrpe, D.V., 1991, Classificatin f terpenids and general prcedures fr their characterisatin. In: " Methds in plant bichmistry" series editrs; Dey, P.M. and J.B. Harbrne, Vlume 7 " Terpenids" Charlwd 8.V., and D.V. Banthepe, (Eds.) Academic Press, Ha curt Brace Jvanvich, Publishers Lndn. pp. l-42. Barlw C.4.. and P.A. Randlph, 1978, Quality and quantity f phlem sap available t the pea aphid. Ann. Entml. Sc. Am. 7l:46-48.

Barz, W. and J. Kster, 1981, Turnver and degradatin f secndary (natural) prducts. In: Bichemistry f plants. Secndary plant prducts. Cnn, E. E., ed.; Academic: Lndn, vl. 7, PP. 35-84. Bath, G.I., C.J. Bell, and H.

139 Barz, W. and J. Kster, 1981, Turnver and degradatin f secndary (natural) prducts. In: Bichemistry f plants. Secndary plant prducts. Cnn, E. E., ed.; Academic: Lndn, vl. 7, PP Bath, G.I., C.J. Bell, and H.L. Llyd, 1991, Arginine as an indicatr f the nitrgen status f wine grapes. Prceedings f the Internatinal Sympsium n Nitrgen in Grapes and Wine. J.M. Rantz (Ed.) pp , American Sciety fr Enlgy and Viticulture. Bernfeild, P. 1967, Bigenesis f Natural Cmpunds. Pergamn Press Ltd. Bidwell, R.G.S. lgtg,plantphysilgy. Macmillan Publishing, New Yrk, pp.643. Bledse, 4.M., W.M. Kliewer, and J.J. Maris, 1988, Effect f timing and severity f leaf remval n yield and fruit cmpsitin f Sauvignn blanc grapevines. Am. J. Enl. Vitic.39: Bselli, M., B. Vlpe, and C. Divai, 1995, Effect f seed number per berry n mineral cmpsitin f grapevine (Vitis vinifera L.) benies. J. Hrt. Sci. 7: ' Brady, C.J. 1987, Fruit ripening. Annual Review f Plant Physilgy, 38; Bravd. B., O. Shseyv, R. Ikan, and A. Altman, 1989, Free and bund mnterpene cntent f leaves and berries and their bisynthesis by in vitr grwn berries. In: Prceedings f the first Internatinal Sympsium n the A matic Substances in Grapes and Wines. Scienza, A.; Versini, G., Eds., Michele all'adige, Italy, Institut Agrari Prvenciale: Trent, Italy, pp Bray. C.M., Nitrgen Metablism in Plants. 214 pp. Lngman Grup Limited, Essex, England. Brwn, S.C., 1981, Sugar accumulatin in the develping grape berry, Ph.D. thesis, University f Adelaide. Brwn, S.C. and B.G. Cmbe, 1985, Slute accumulatin by grape pericarp cells. III. Sugar change in viv and the effect f shading. Bichem. Physil. Pflanzen. 18: Buttrse, M.S., and C.R. Hale. 1971, Effects f temperature n accumulatin f starch r lipid in chlrplasts f grapevine. Planta, (Berl.) 11:

140 Candlfi-Vascncels, M.C., M.P. Candlfi, and W. Kblet,1994, Retranslcatin f carbn reserves frm the wdy strage tissues int the fruit as a respnse t defliatin sfress during ripening perid nvitis vinifera L. Planta Candlfi-Vascncels, M.C., and W. Kblet, 199, Yeild, fruit quality, bud fertility and sta ch reserves f the wdas a functin f leaf remvalin Vitis vinifera, evidence f cmpensatin and stress recvering. Vitis' 29:199'221' Carrll, D.E., and J.E. Marcy,1982, Chemical and physical changes during maturatin f muscadine grapes (vitis rtundifulia). Am. J. Enl. vitic.33: Cash, J.N.; V/.4. Sistrunk, and C.A. Stutte, 1976, Characteristics f Cncrd grape plyphenlxidase invlved in juice clurlss. J. Fd Sci. 41: L ' Charlwd, 8.V., and D.V. Banthrpe, 1987, The bisynthesis f mnterpenes, In,.Prgress in Phytcemistry", vl.5, Eds. Reinhld, L. ; J. Harbrbe and T. Swain, Perganm press, Oxfrd, UK, pp 65'125. Charlwd, B.V., J.T. Brwn, C. Mustu, and G.S. Mnis, 1988, The accumulatin f isprenid flavur cmpunds in plant cell cultures. In: Biflavur '87, analysis, bichemistry, bitechnlgy. Sch eier (Ed.), Walter de Gruyter & C., Berlin, pp Chaumnt, M., J.F. Mrt-Guadry, and C.H. Fyer, 1994, Seasnal and diurnal changes in phtsynthesis and carbn partitiningin Vitis vinifera leaves in vines with and withut fruit. J. Exp. Bt.,45: 1235-'1243. Charlwd. 8.V.. and K.A. Charlwd, 1991, Mnterpenids. In: " Methds in plant bichmistry" series editrs; Dey, P.M. and J.B. Harbrne, Vlume 7 " Terpenids" Charlwd 8.V.. and D.V. Banthepe, (Eds.) Academic Press, Harcurt Brace Jvanvich, Publishers Lndn, PP Chin, M., H. Hayashi. S. Nakumura, T. Oshima, H. Turner, D. Sabnis, V. Brkvec, D. Baker, G. Girusse, J.L. Bnnemain, and S. Delrt, 1997, Phlem sap cmpsitin. ln: Recent advances in phlem transprt and assimilate cmpartmentatin. Bnnemain, J.L., W.J. Lucas and J. Dainty, (Eds'), Ouset Editins, pp Cnradie, Vy'. J., 198. Seasnal uptake f nutrients by Chenin blanc in sand culture nitrgen. S. Afr. J. Enl. Vitic., l:

124 Cnsidine, J.4., 1979, Biphysics and bichemistry f fruit grwth and develpment. Ph.D. Thesis, University f Melburne. Cnsidine, J.4., and R.B. Knx,lgTg, Develpment and histchemistry f the cells, cell walls and cuticle f the dermal system f fruit f the grape, Vitis vinifera.

141 124 Cnsidine, J.4., 1979, Biphysics and bichemistry f fruit grwth and develpment. Ph.D. Thesis, University f Melburne. Cnsidine, J.4., and R.B. Knx,lgTg, Develpment and histchemistry f the cells, cell walls and cuticle f the dermal system f fruit f the grape, Vitis vinifera. Prtplasm a 99 : Cmbe, 8.G., 1995, Adptin f a system fr identiffing grapevine grwth stages. Australian Jurnal f Grape and Wine Research 1: ' Cnstable, 1988 Cytdifferentiatin. In: Biflavur '87, analysis, bichemistry, bitechnlgy. Schreier (Ed.), Walter de Gruyter & C., Berlin, pp Cmbe,8.G., 1973, The regulatin f set and develpment f the grape berry. Acta Hrtic., 34: Cmbe, B.G., 1975, Develpment and maturatin f the grape berry. The Australian Grapegrwer and Winemaker, N: 136. Cmbe, 8.G., 198, Develpment f the grape berry. I. Effect f time f flwering and cmpetitin. Aust. J. Agric. Res.,3l: Cmbe, B.G.,1987a, Influence f temperature n cmpsitin and quality f grapes' Acta Hrticulture, 26: Cmbe, B.G.,1987b, Distributin f slutes within the develping grape berry in relatin t its mrphlgy. Am. J. Enl. Vitic.38: ' Cmbe, 8.G.,1989. The grape berr_v as a sink. Acta Hrticulture, 239: Cmbe,8.G., Research n develpment and ripening f the grape berr)'.4m. J Enl. Vitic. 43: ll -l 1. Cmbe, B.G., and G.R. Bishp, 198. Develpment f the glape berry. II. Changes in diameter and defrmabilir_v during veraisn. Austral. J. Agric. Res. 3 l: Cmbe, B.G., and Ph. Matile. 198, Slute accumulatin by grape pericarp cells. I. Sugar uptake by skin segments. Bichem. Physil. pflanzen. 175: Cmbe, 8.G., and P.E. Phillips, 1982, Develpment f the grape berry. III. Cmpsitinal changes during veraisn measured by sequential hypdermic sampling. Internatinal sympsium n Grapes and Wine, Jun, 198, Davis, Califrnia, pp

Cmbe, B.G., M. Bvi and A. Schneider, 1987, Slute accumulatin by grape pericarp cells. V. Relatinship t berry size and the effects f defliatin. Jurnal f Experimental Btany, 38: I 789-1 798.

142 Cmbe, B.G., M. Bvi and A. Schneider, 1987, Slute accumulatin by grape pericarp cells. V. Relatinship t berry size and the effects f defliatin. Jurnal f Experimental Btany, 38: I Crdnnier, R., and C. Baynv, 1974, Bichimie appliquée, Mise en évidence dans la baie de raisin, var. Muscat d'alexandrie, de mnterpenes liés, révélables par une u plusieurs enzymes du fruit. C.R. Acad. Sc. Paris, t.278, Série D, Craver, M.C., Guidni, S., Schneider, A. and Di Stefan, R. (1994) Caractérisatin variétal cépages musqués à raisin clré au myen de paramètres ampélgraphiques descriptifs et bichemiques. Vitis 33, Creasy, G.L., and P.B. Lmbard,1993, Vine water stress and peduncle girdling effects n pre- and pst-veraisn grape berry grwth and defrmability. Am. J. Enl. Vitic. 44: Creasy, G.L., S.F. Price, and P.B. Lmbard, t993, Evidence fr xylem discntinuity in Pint nir and Merlt grapes: Dye uptake and mineral cmpsitin during berry maturatin, Am. J. Enl. Vitic. 44: Crippen, D.D.Jr. and J.C. Mrrisn, 1986, The effects f sun expsure n the phenlic cntent f Cabemet sauvignn berries during develpment. Am. J. Enl. Vitic. 37: Crnshw, J., and D.D. Sabnis, 199, Phlem prteins. In: Sieve elements, Behnke, H.-D., Sjlund. R.D. (eds.), Springer. Berlin, pp Crteau. R., Bisynthesis and catablism f mnterpenes In " Ispentenids in plants. Bichemistry' and functin", W.D. Nes; G. Fuller and L.S. Tsai, eds. Marcel Dekker. lnc. Neu'Yrk. Crteau, R., 1987, Bisynthesis and catablism f mnterpenids. Chem. Rev. 87: Crteau, R., and C. Maninkus Metablism f Mnterpenes; Demnstratin f (+)-Nementhil-p-D-glucside as a majr metablite f (-)-Menthn in Pepperminf ( Mcnrha piperita ). Plant Physil. 64: Darné. G., 1993, Nuvelles hypthèses sur la synthèses des anthcyanes dans les baies et dans feuiles devigne. Olew hyptheses n anthcyanin bisynthesis in berries and leaves f the grapevine). Vitis, 32:

Delrt, S., and J.L. Bnnemain, 1981, invlvement f prtns as a substrate fr the sucrse carrier during phlem lading inviciafabø leaves. Plant Physil. 67: 56-564. Delrt, S.

143 Delrt, S., and J.L. Bnnemain, 1981, invlvement f prtns as a substrate fr the sucrse carrier during phlem lading inviciafabø leaves. Plant Physil. 67: Delrt, S., 1981, Prtn fluxes assciated with sugar uptake in Vicia faba Plant Physil. 68: leaf tissues. Dimitriadis, E., and P.J. Williams, 1984, The develpment and use f a rapid analitical technique fr estimatin f free and ptentially vlatile mnterpene flavrants f grapes. Am. J. Enl. Vitic.,32(2) : During, H., A. Lang, and F. Ogginni, 1987, Patterns f water flw in Reisling berries in relatin t develpmental changes in their xylem mrphlgy. Vitis 26:123-13I. Dzhakhua, M.J., E.S. Derbglav, and M.A.Dzhapariðze, 1978, Study f the vlatile phenl cmpunds in white wines. Prihl. Bikhim. Mikrbil. l4: Ehret, D.L. and L.C. H, 1986, Translcatin f calcium in relatin t tmat fruit grwth. Ann. Bt. 58: Esau, K., 1965, Plant Anatmy, Jhn Wiley and Sns, Inc, Tppan Cmpany, LTD., Tky, Japan. pp Eschrich, W., and W. Heyser, 1975, Bichemistry f phlem cnstituents. In: Encyclupedia f plant physilgy. N.S. Vl. 1 :Transprt in plants. 1. Phlem transprt. pp , Zimmerman, M.H., Milburn, J.A., (eds.), Springer, Berlin. Feller, U. and R. Helzer Transprt f amin acids via xylem and phlem in detached wheat shts. In: Recent advances in phlem transprt and assimilate cmpartmentatin. Bnnemain. J.L., W.J. Lucas and J. Dainty, (Eds.), Ouset Editins. pp.9l-95. Fellws. R.J., and M.J. Dalling. 1974, Structural and physilgical changes in sugar beet leaves during sink t surce cnversin. Plant Physil. 54: Fernández-Lpez. J.4.. Hidalg, V., Almela, L. and López-Rca, J.M. (1992) Quantitative changes in anthcyanin pigments f Vitis vinifera cv. Mnastrell during maturatin. J. Sci. Fd Agric., 58: Findlay, N.K., J. Oliver. N. Nii, and B.G. Cmbe, 1987, Slute accumulatin by grape pericarp cells. IV. Perfusin f pericarp applast via the pedicel and evidence fr xylem malfunctin in ripening berries. J. Exp. Bt. 38:

Fisher, D.B. and P. Mac-Nicl, 1986, Amin acids cmpsitin alng the pathway during grain filling in wheat. Plant Physil. 82:119-123' Flwers, T.J., and A.R. Ye, lgg2, Slute transprt in plants.

144 Fisher, D.B. and P. Mac-Nicl, 1986, Amin acids cmpsitin alng the pathway during grain filling in wheat. Plant Physil. 82: ' Flwers, T.J., and A.R. Ye, lgg2, Slute transprt in plants. Blackie Academic & Prfessinal ( An Imprint f Chapman & Hall), Glasgw' UK' Fndy, 8.R., and D.R. Geiger, 1977, Sugar selectivity and ther characteristicsf phlem lading in Bet vulgaris I. Plant Physil. 59:953-96' Fng, C.H., S. Hasegawa, Z. Herman, and P. Ou, 1991, Bisynthesis f limnid glycsides in lemn (citrus limn ). J. sci. Fd Agric.,54: Fraga, B.M., 1991, Sesquiterpenids, In: "Methds in plant bichmistry" series editrs; Dey, P.M. and J.B. Harbrne, vlume 7 " Terpenids" charlwod 8.v., and D.V. Banthepe, (Eds.) Academic Press, Harcurt Brace Jvanvich, Publishers Lndn, pp Fukumrita, T. and M. Chin, 1982, Sugar, amin acid and inrganic cntents in rice phlem sap. Plant Cell Physil.,23: Galet, P., (1958 and 1962) Cepages et Vignbles de France. Tme II et III. Les Cepages de Cuve, 986 et 1234.lere Partie. Dehan, Mntpellier. Gamalei, Y.V., 1985, Characteristics f phlem lading in wdy and herbaceus plants Sv. Plant Physil. 32: Ga. Y., and G.A. Cahn, 1994, Cluster shading effects n fruit quality, fruit skin clur, and anthcyanin cntent and cmpsitin in Reliance (Vitis hybrid). Vitis,33: Gieger, D.R., Understanding interactins f surce and sink regins f plants. Plant Physil. Bichem., 25 : Geiger, D.R., and B.R. Fndy, 1991, Regulatin f carbn allcatin and partitining: status and research agenda. In: J.L. Bnnemain, S. Delrt and J. Dainty, W. Lucas. (eds.) "Research advances in phlem transprt and assimilate cmpartmentatin. Ouest Editins, Nants, France, pp l-9. Ghlami, M., and B.G. Cmbe, 1995, Occurrence f anthcyanin pigments in berries f the white cultivar Muscat grd Blanc (Vitis vinifera L.). Australian Jurnal f Grape and Wine Research l:

145 Ghlami, M., Hayasaka Y., Cmbe, 8.G., Jacksn, J.F., Rbinsn, S.P. and W'illiams' P.J., 1995, Bisynthesis f flavur cmpunds in Muscat Grd Blanc grape berries. Australian Jurnal f Grape and Wine Research, l: Giaquinta, R.T., 1983, Phlem lading f sucrse, Ann. Rev. Plant Physil., 34: Giaquinta, R.T., and D.R. Geiger, 1977, Mechanism f cyanide inhibitin f phlem translcatin. Plant Physil. 59: Girusse, C., J.L. Bnnemain, S. Delrt, and R. Burnville,lggI, Sugar and amin acid cmpsitin f phlem sap f medicag sativa: a cmparative study f tw cllecting methds. Plant physil. Bichem.29: 4I'48. Glad, C., J.-L. Regnard, Y. Quer, O. Brun, and J.-F. Mrt-Gaudry,1992a, Phlem sap exudate as a crterin fr sink strength appreciatin in Vitis vinifera cv. Pint nir gapevines. Vitis, 3 1 : Glad, C., J.-L. Regnard, Y. Quer, O. Brun, and J.-F. Mrt-Gaudry, I992b, Flux and chemical cmpsitin f xylem exudatin f m Chrdnnay grapevines: tempral evlutin and effect f recut. Am. J. Enl. Vitic. 43: Glad, C., J. Farineau, J.L. Regnard, and J.F. Mrt-Gaudry, 1994, The relative cntributin f nitrgen riginating frm tw seasnal l5n supplies t the ttal nitrgen pl present in the bleeding sap and in whle Vitis vinifera cv. Pint nir grapevine at blm time. Am. J. Enl. Vitic., 45: Gldschmidt, E.E.. and S.C. Huber, 1992, Regulatin f phtsynthesis by end-prduct accumulatin in leaves f plants string starch, sucrse, and hexse sugars. Plant physil Gnrez, E., A. Martinez. and J. Leancina, lgg4,lcalizatin f free and bund armatic cmpunds amng skin, juice and pulp fractins f sme grape varieties. Vitis 33: t-4. Gmez, E., and A. Martinez. 1995, Changes in vlatile cmpunds during maturatin f sme grape varieties. J. Sci. Fd Agric., 67 : Gdwin, T.W.. and E.l. Mercer, 1983, Intrductin t plant bichemistry; Pergamn: Australia. p Grdn,4.J., G.J.A. Ryle, and G. Webb, 198, The relatinship between sucrse and starch during'dark'exprt frm leaves f uniculm barley. J. Exp. Bt.31: 845-8s.

$L. Baynve, R.L. Baumes, and R.E. Crdnnier, 1985a, I' Extractin and determinatin f free and glycsidically bund fractins f grape arma cmpnents. J. Chrmatgr. 331 : 83-9. Gunata, Y.2., C.L. Baynve, R.L. Baumes, and R.E. Crdnnier, 1985b, The arma f grapes.$

146 Greenspan, M.D., K.A. Shackel, and M.A. Matthews, 1994, Develpmental changes in the diurnal water budget f the grape berry expsed t water deficits. Plant Cell and Envirnment, l7: Gunata, Y.2., C.L. Baynve, R.L. Baumes, and R.E. Crdnnier, 1985a, I' Extractin and determinatin f free and glycsidically bund fractins f grape arma cmpnents. J. Chrmatgr. 331 : Gunata, Y.2., C.L. Baynve, R.L. Baumes, and R.E. Crdnnier, 1985b, The arma f grapes. Lcatin and evlutin f free and bund fractins f sme grape arma cmpnents cv.muscat During first develpment and maturatin. J. Sci. Fd Agric. 36: Gunafa,Y.Z;C.L. Baynve; R.L. Baumes, and R.E. Crdnnier, 1986, Changes in free and bund fractins f armatic cmpnents in vine leaves during develpment f Muscat grapes. Pytchemistry, 25: 943'946. Gutiérez-Granda, M.J., and J.C. Mnisn, 1992, Slute distributin and malic enzyme activity in develping grape berries. Am. J. Enl. Vitic. 43:323'328. Hahlbrck, K., and Scheele, D., 1989, Physilgy and mlecular bilgy f phenilprpanid metablism. Ann. Rev. Plant Physil. Plant Ml. Bil. 4: Hale, C.R., Relatin between ptassium and the malate and tartarate cntent f grape berries. Vitis. I 6:9-19. Hale, C.R., and R.J. 'ù/eaver, The effect f develpmental stage n directin f phtsynthate in Vitis vinifera. Hilgardia,33: Hall, S.M., and D.A. Baker, The chemical cmpsitin f Ricinus phlem exudate. Planta, 16: Hamiltn, D.4.. and P.J. Davis, Mechanism f exprt f rganic material frm the develping fruits f pea. Plant Physil. 86: Hansn, S., and J. Chen. 1985, A techniquefr cllectin f exudate frm pea seedlings Plant Physil. 43 : Hardy, P.J. 1967, Sucrse breakdwn and synthesis in the ripening grape berry.aust. J. Bil. Sci

Hardy, p-j. 1968, Metablism f sugars and rganic acids in immature grape berries. Plant Physil. 43:224-228. Hardy, W.J., and T. P. O'Brien, 1988, Cnsideratins f the bilgical significance f sme vlatile cnstituents f grape (Vitis spp.

147 Hardy, p-j. 1968, Metablism f sugars and rganic acids in immature grape berries. Plant Physil. 43: Hardy, W.J., and T. P. O'Brien, 1988, Cnsideratins f the bilgical significance f sme vlatile cnstituents f grape (Vitis spp.). Aust. J. Bt., 36: I7-ll7. Harris, J.N., P.E. Kriedemann, and J.V. Pssingham, 1968, Anatmical aspects f grape berry develpment, Vitis, 7 : Haslam, E.IgT5,Natural pranthcyanidins In: Flavnids. Harbrne, J.8., T.H' Marby, and H. Marby (Eds.) pp Hawker, J.S., M.S. Buttrse, A. Seffky and J.V. Pssingham,1972, A simple methd fr demnstrating macrscpically the lcatin f plyphenlic cmpunds in grape berries. Vitis, 1 1: Hayashi, H., and M. Chin, 1986, Cllectin f pure phlem sap frm wheat and its chemical cmpsitin. Plant Cell Physilgy, 27 : H, L.C., 1988, Metablism and cmpartmentatin f imprted sugars in sink rgans in relatin t sink strength. Ann. Rev. Plant Physil. Plant Ml. Bil '378 H, L.C., lgg2, The pssible effects f sink demand fr assimilate n phtsynthesis. In: Murata,N. (ed.) Research in phtsynthesis, vl. IV, Drdrecgt: Kluwer, pp Hfläker, W., 1978, lnvestigatin n the phtsynthesis f vines influence f defliatin, tpping. girdling and remval f grapes. Vitis, l7: Hking, P.J., 198, The cmpsitin f phlem exudate and xylem sap frm tree tbacc (Nictiana glauca Grah). Ann. Bt., 45: Hsel, W., 1981, Glycsilatin and glycsides. In "The Bichemistry f Plants" Vl. 7, Chapter 23, pp , Academic Press, Inc. Hrazdina, G., and A. Mskwitz. 198, Subcellular status f anthcyanins in g ape skins. Centenary Symp.n Grape and Wine, Univ. Calif. Davis, July 198. Hrazdina, G., G.F. Parsns, and L.R. Mattick, 1984, Physilgical and bilgical events during develpment and maturatin f grape berries. Am. J. Enl. Vitic.35:

Huber, S.C., J.L.A. Huber and R.W. McMichael Jr, 1992, The regulatin f sucrse synthesi in leaves. In: Carbn partitining within and between rganisms. C.J' pllck, J.F. Farra and A.J. Grdn (Eds.

148 Huber, S.C., J.L.A. Huber and R.W. McMichael Jr, 1992, The regulatin f sucrse synthesi in leaves. In: Carbn partitining within and between rganisms. C.J' pllck, J.F. Farra and A.J. Grdn (Eds.) BIOS Scientific Publishers Limited, UK. pp. l-26. Humpf, H-V., and P. Schreier, lggl. Bund Arma cmpunds frm the Fruit and the Leaves f Blackberry (Rubus laciniata L.), J.Agric.Fd Chem. 39: Hunter, J.J., and J.H. Visser, 1988, Distributin d l4c- phtsynthetate in the sht f Vitis vinifera L. cv. Cabernet Sauvignn. I. the effect f leaf psitin and develpmental stage f vine. S.4fr. J. Enl' Vitic' 9: Hunter, J.J., and J.H. Visser, 1989, The effect f partial defliatin, leaf psitin and develpmental stage f the vine n leaf chlrphyll cncentratin in relatin t the phtsynthetic activity and light intensity in the canpy f Vitis vinifera L. cv. Cabernet Sauvignn. S. Afr' J. Enl. Vitic. l: Hunter, J.J., and J.H. Visser, 199, The effect f partial defliatin n grwth characte istic f Vitis viniferal. cv. Cabernet Sauvignn. II. Reprductive grwth' s. Afr. J. Enl. Vitic. ll: Hunter, J.J., R. Skrivan, and H.P. Ruffner, 1994, Diurnal and seasnal physilgical changes in leaves f Vitis vinifera L.: CO2 assimilatin rate, sugar level and sucrlytic enzyme activity. Vitis, 33: Hunter, J.J., H.P. Ruffner. C.G. Vlschenli, and D.J. Le Rux, 1995, Partial defliatin f Viti.s t,inifþra L. cr,. Cabernet Sauvignn 99 Richter: Effect n rt grwth, canpy efficiencl'. grape cmpsitin, and wine quality. Am.J. Enl. Vitic. 46: 36-3 I 4. Iland. P.G. and B.G.Cmbe, Malate, tartarate, ptassium, and sdium in flesh and skin f Shiraz grapes during ripening: cncentratin and cmpartmentatin. Am. J. Enl. Vitic. 39: 7 l Ishiwatari, Y., C. Hnda. l. Kawashima, S. Nakamura. H. Hiran, S. Mri, T. Fujiwara, H. Hayashi. and M. Chin, Thiredxin h is ne f the majr prteins in rice phlem sap. Pianta. 195: Jeschke, W.D. and S. Pate. 1995, Mineral nutritin and transprt in xylem and phlem f Banksia printes (Prteaceae), a tree with dimrphic rt mrphlgy. J. Exp. Bt. 46:

Jhnsn, J.O., R.J. Weaver, and D.F. Paige, 1982, Differences in the metablizatin f assimilates f Vitis vinifera L. grapevines as influenced by an increased surce strength. Am. J. Enl. Vitic.33:27'213.

149 Jhnsn, J.O., R.J. Weaver, and D.F. Paige, 1982, Differences in the metablizatin f assimilates f Vitis vinifera L. grapevines as influenced by an increased surce strength. Am. J. Enl. Vitic.33:27'213. Kanasawud, P., and J.C. Cruzet, 199, Mechanism f frmatin f vlatile cmpunds by thermal degradatin f cartenids in aquus medium. l. b- cartenedegradatin. J. Agric. Fd Chem. 38: Kawabe, S., T. Fucumrita, and M. Chin, 198, Cllectin f rice phlem sap frm stylets f hmpterus insects by YAG laser. Plant Cell Physilgy, 2I: l3l King, R.W., and J.A.D. Zeevaart, 1974, Enhancement f phlem exudatin frm cut petiles by chelating agents. Plant Physilgy. 53: Kliewer, W.M. 1965, The sugars f grapevines. II. Identificatin and seasnal changes in the cncentratin f several trace sugars in Vitis vinifera. Am. J. Enl. Vitic. l6: I Kliewer, W.M., 1967a, The glucse-fructse rati f Vitis vinifera grapes. Am. J. Enl Vitic. l8:33-41 Kliewer, W.M., 1967b, Annual cyclic changes in the cncentratin f free amin acids in grapevines. Am. J. Enl. Vitic. 18: Kliewer, W.M., Free amin acids and ther nitrgenus substances f table grape varieties. L. Fd Sci. 34: Kliewer, W.M.. 197a. Effect f time and severity f defliatin n grwth and cmpsitin f Thmpsn Seedless grapes. Am. J. Enl. Vitic.21: Kliewer, W.M., 197b. Free amin acids and ther nitrgenus fractins in wine grapes. J Fd Sci. 35: Kliewer, W.M.. 197c. Effect f da,temperature and light intensity n clratinf Vilis vinifera L. grapes. Jurnal f the. American Sciety fr Hrticultural Science 95, Kliewer, W.M.. and A.J. Antcliff. 197, Influence f defliatin, leaf darkening, and cluster shading n the grwth and cmpsitin f sultana grapes. Am. Enl. Vitic. 2l:26-36.

Kliewer, W.M.,and A.L.Y.R. Nassar, 1966, Changes in cncentratin f rganic acids, sugats, and amin acids in grape leaves. Amer. J. Enl. Vitic. 17:48-57. Kblet, W.

150 Kliewer, W.M.,and A.L.Y.R. Nassar, 1966, Changes in cncentratin f rganic acids, sugats, and amin acids in grape leaves. Amer. J. Enl. Vitic. 17: Kblet, W., Ig77, Translcatin f phtsynthate in grapevines. Prc. Intern. Symp' euality Vintage, Feb. 1977, Cape Twn, Republic f Suth Africa. K iedemann, P.E., 1977, Vine leaf phtsynthesis. Prc. Intern. Symp. Quality Vintage, Feb. 1977, Cape Twn, Republic f Suth Africa' Kursanv, 4.L., 1984, Assimilate transprt in plants. Elsevier Science Publishers B.V., Amesterdam. pp. 66. Kurt, B. and G. Trssell, 1983, Natural prduct chemistry. Jhn Wilry and sns limited, N.Y. pp.4l. Lang, A. and H. During, 1991, Partitining cntrl by water ptential gradient: Evidence fr cmpartmentatin breakdwn in grape berries. J. Exp. BL 42 (242): llltrt22. Lang,4., and M.R. Thrpe, 1989, Xylem, phlem and transpiratin flws in a grape: Applicatin f a technique fr measuring the vlume f attached fruits t high reslutin using Archimedes principle. J.Exp. Bt. Vl. 4 (219): Lang, 4., M.R. Thrpe, and W.R.N. Edwards, 1986, Plant water ptential and translcatin. In: J.L. Bnnemain, S. Delrt and J. Dainty, W. Lucas, (eds.) "Research advances in phlem transprt and assimilate cmpartmentatin. Ouest Editins, Nants. France, pp Leckstein, P.M., and M. Llewellyn, 1975, Quantitative analysis f seasnal variatin in the amin acids in phlem sap f Salix alba L., Planta, 124: Lukner, M., 1972, Secndary metablism in plants and animals. Translated by T.N Vasudevan, Chapman and Hall Ltd.Lndn, pp. 44. Lukner, M., 199, Secndary metablism in micr-rganisms, plants and animals.3rd editin.springer Werlag, Berlin. Lunn. J.E., and M.D. Hatch, 1995, Primary partitining and strage f phtsynthate in sucrse and starch in leaves f Ca plants. Planta, 197: Madre, M.4., 1991, Diurnal ligsaccharide exprt patterns in Cucurbit su ce leaves. In: Recent advances in phlem transprt and assimilate cmpartmentatin. Bnnemain, J.L., W.J. Lucas and J. Dainty, (Eds.), Ouset Editins, pp

Mailla d, P., J.P. Prcher, 1991, Effect f light n the regulatin f carbn partitining in mature leaves f Zea mays L.In: Recent advances in phlem transprt and assimilate cmpartmentatin. Bnnemain, J.L., W.

151 Mailla d, P., J.P. Prcher, 1991, Effect f light n the regulatin f carbn partitining in mature leaves f Zea mays L.In: Recent advances in phlem transprt and assimilate cmpartmentatin. Bnnemain, J.L., W.J. Lucas and J. Dainty, (Eds.), Ouset Editins, pp. 27'32. Mann, J., 1987, Secndary metablism. Atkins, P.W., J.S.E' Hlker and A.K. Hlliday, (Eds.), Oxfrd Chemistry Series, Oxfrd University Press, Oxfrd England. Marins, V.4., 1992, Applicatin f Fast Atm Bmbardment Mass Spectrmetry t the examinatin f glyccnjugates in grape juice. Ph. D. Thesis, The University f Adelaide, S.A. Maris, J. 1983, Terpenes in the arma f grapes and wines. S. 4fr. J. Enl. 4:49-6. Marschner, H., 1983, General intrductin t the mineral nutritin f plants. In: "Encyclpedia f plant physilgy" Lauchli, 4., and R.L. Bieleski (eds.), New series, Vlume l2,lnrganic plant nutritin. Berlin: Springer-verlag, pp Matsui, H., E. Yuda, and S. Nakagawa,1979, Physilgical studies n the ripening f Delaware grapes. I. Effect f the number f leaves and changes in plysaccharides r rganic acids n sugar accumulatin in the berries. J. Japan. Sc. Hrt. Sci. 48: Mattivi, F., Scienza, 4., Failla, O., Villa, P., Anzani, R., Tedsc, G', Gianazza, E. and Righetti. P. (1989) Vitis vinifera - a chemtaxanmic apprach: anthcyanins in the skin. Prceedings f the 5th Internatinal Sympsium n Grape Breeding, Vitis 199 special issue, pp I l Milburn, J.A Phlem exudatin frm castr bean: inductin by message. Planta, Mrrisn, J.C., and M. Idi. 199, The influence f water berry n the develpment and cmpsitin f thmpsn seedless grapes. Am. J. Enl. Vitic.4l: Mss, D.N., and H.P. Rasmussen, 1969, Cellular lcalizatin d CO2fixatin and translcatin f metablites. Plant Physi l. 44 I Mtmra, Y., 1993, lac-assimilate partitining in grapevine shts: effect f sht pinching, girdling f sht, and leaf-halving n assimilate partitining frm leaves int clusters. Am.J. Enl.Vitic. 44:

135 Mullins, M. G., and Rajasekaran K. (1981). Fruiting cuttings: Revised methd prducing test plants f grapevine cultiva s. Am. J. Enl. Vitic.32: 35-4' fr Murphey, J.M., S.E. Spayd, and J.R. Pwers, 1989, Effect f grape maturatin n sluble prtein characteristics f Gewürztraminer and White Riesling juice and wine.

152 135 Mullins, M. G., and Rajasekaran K. (1981). Fruiting cuttings: Revised methd prducing test plants f grapevine cultiva s. Am. J. Enl. Vitic.32: 35-4' fr Murphey, J.M., S.E. Spayd, and J.R. Pwers, 1989, Effect f grape maturatin n sluble prtein characteristics f Gewürztraminer and White Riesling juice and wine. Am. J. Enl. Vitic. 4: Nakamura, S., H. Hayashi, S. Mri, and M. Chin, 1995, Detectin and characterizatin f prtein kinases in rice phlem sap. Plant Cell Physil.36; Nassar, 4.L., and W.M. Kliewer, l966,free amin acids in varius parts f Vits vinifera at different stages f develpment. Prc. Am. Sc. Hrtic Sci., 89: Nii, N., and B.G. Cmbe, 1983, Structure and develpment f the berry and pedicel f the grape Vitis viniferal' ActaHrtic' 139:129-14' Nbel, p.s.,1991, Physicchemical and envirnmental plant physilgy. Academic press' Inc. Harcurt Brace Jvanvich, publishers. San Dieg Calfmia. Nble, A.C., 1994, Wine flavur. In: understanding natural flavurs. Piggtt, J.R. and patersn, A. (Eds.), Blackie Accademic and prfessinal, an imprit f Chapman & Hall. Glasgw. PP Nble, A.C.; C.R. Strauss; P.J. Williams and B. Wilsn, 1988, Cntributin f terpene glycsides t bitterness in Muscat wines. Am. J. Enl. Vitic Ohlff. G Naturally ccurring arma cmpunds. In: 'Prgress in the chemistry f rganic natural prducts'. Vl.35, W. Herz, H. Grizebach and G.W. Kirby (eds.), pp Springer-Verlag. Wien. Otsuka, H., K. Yamasaki. and T. Yamaguchi. 1989, Alangifliside, a diphenylmethylene derivative. and ther phenlics frm the leaves f Alangium plantiflium var. Trilbum. Phytchem. 28( I I ): Padgett, M., and J.C. Mrrisn, 199, Changes in grape berry exudates during fruit develpment and their effect n mycelial grwth f Btrytis cinera.j. Am. Sc' Hrt. Sci. 115: Pandey, R.M., and H.L. Farmahan, 1977, Changes in the rate f phtsynthesis and respiratin in leaves and berries f Vitis vinifera grapevines at varius stages f berry dcvelpment. Vitis, l6: 16- I 1 1.

park, S.K., J.C. Mnisn, D.O. Adams, and A.C. Nble, 1991, Distributin f Free and Glycsidically Bund Mniterpenes in the Skin and Mescarp f Muscat f Alexandria Grapes During Develpment. J. Agric.

153 park, S.K., J.C. Mnisn, D.O. Adams, and A.C. Nble, 1991, Distributin f Free and Glycsidically Bund Mniterpenes in the Skin and Mescarp f Muscat f Alexandria Grapes During Develpment. J. Agric. Fd Chem. 39: park, S.K., and A.C. Nble, 1993, Mnterpenes and mnterpen glycsides in wine armas; analysis, characterizatin, and technlgical advances. In: "Beer and Wine Prductin". Gump, 8.H., (ed), ACS Sympsium Series 536' Pate, J.S., P.J. Sharkey, and O.A.M. Lewis, 1974, Phlem bleeding frm legume fruits; a technique fr study f fruit nutritin. Planta, 12: ' Pate, J.S., 1975, Exchange f slutes between phlem and xylem and circulatin in the whle plant. In: Zimmernann and J. Bilburn (Eds.) Transprt in plants. Springer, Berlin. pp.45i-473. Pate, J.S. 198, Transprt and partitining f nitrgenus slutes. Ann. Rev. Plant Physilgy, 3 I : Pate, J.S., 1983, Patterns f nitrgen metablism in higher plants and their eclgical significanc e. P , In: "Nitrgen as Eclgical factr", J.A. Lee, S. Mc Neill, and I.H. Rrisn (Eds.), Blackwell, Bstn. Peynaud, E., and P. Ribéreau-Gayn, 1971, The grape. In: "The bichemistry f fruits and their prducts". Vl. 2, Hulme, A.C.(Ed.), Academic press. Pirie,4., and M.G. Mullins 1976, Changes in anthcyanin and phenlics cntent f grapevine leaf and fruit tissues treated with sucrse, nitrate, and abscisic acid. Plant Physil. 58: Pirie. 4., and M.G. Mullins 1977 In enelatinships f sugars anthcyanins, ttal phenlsand dry weight in the skin f grape berries during ripening. Amr. J. Enl. Vitic. 28; Pirie, 4., and M.G. Mullins, 198, Cncent atin f phenlics in the skin f grape berries during final develpment and ripening. Am. J. Enl. Vitic., 3l: Pssner, D., and W.M. Kliewer. 1985, The lcalisatin f acids, sugars, ptassium and calcium in develping grape berries. Vitis, 24: Pssner, D., H.P. Ruffner, and D.M. Rast, 1983, Regulatin f malic acid metablism in berries f Vitis vinifera. Acta Hrtic. 139: ll

$puey, E., M. Dizy, and M.C. Pl, 1993, Varietal differentiatin f must and wine by means f prtein fractin. Am. J. Enl' Vitic. 44:255'26. euinlan, L.D., and R.J. Weaver,!$

154 puey, E., M. Dizy, and M.C. Pl, 1993, Varietal differentiatin f must and wine by means f prtein fractin. Am. J. Enl' Vitic. 44:255'26. euinlan, L.D., and R.J. Weaver,!97, Mdificatin f pattern f the phtsynthate mvement within and between shts f Vitis vinifera I. Plant Physil. 46: Ramaswami, S.K., P. Briscese, R.J. Garguilll, and T. vn Gelder, 1986, Sesquiterpene hydrcarbns: frm mass cnfusin t rderly line-up. In: Flavurs and Fragrants: A wrld perspective. Prceedings f the loth Internatinal; Cngress f Essential Oils, Fragrances and Flavurs, Lwtence, B.M., B.D. Mkherjee, and B.J. V/illis, (Eds.), 16-2 Nvember 1986, Washingtn, DC, USA, Rapp, 4., H. Mandery, and H. Ullemeyer, 1983, 3,7-Dimethyl-1,7-ctan-dil-eine neue Terpen-Kmpnente des Trauben-und wine-armas. vitis, 22: Rapp, 4., 1988a, Wine arma substances frm gas chrmatgraphic analysis. In: Mdern methds f plant analysis. Vl. 6, Wine Analysis; Linskens, H.F. and J.F. Jacksn, (Eds.) Springer Verlag, Berlin, pp.29-66' Rupp,4., 1988b, Studies n terpene cmpunds in wines. In; "Frntiers f flavur", prceedings f the Fifth Intematinal Flavur cnference. Charalambus, G. Ed., l-3 July, 1987, prt Karras, Calkidiki, Greece; Elsevier Science, Amsterdam, pp Razungles, 4.. C.L. Baynv, R.E. Crdnier, and J.C. Sapis, 1988, Grape cartenids: changes during maturatin perid and lcalizatin in mature berries. Am. J. Enl. Vitic. 39: Ribéreau-Gayn. P The anthcyanins f grapes and wines. In: Anthcyanins as Fd Clrs. Markakis.P.. Ed.. Academic Press, New Yrk,29. Ribéreau-Gayn. P., J.N. Bidrn, and A. Terrier, I 975, Arma f Muscat grape varieties J. Agric. Fd Chem.23(6): Richardsn, P.T.. D.A. Baker. and L.C. H, 1982, The chemical cmpsitinf cucurbit vascular exudates. J.Exp. Bt.. 33: Ridge, 1., Plant physilgy. Bilgy: Frm and Functin, Hdder and Stughtn, The Open University.

Riens,8., G. Lhaus, D. Heineke, and H.W. Heldt, 1991, Amin acid and sucrse cntent determined in the cytslic, chlrplastic, and vacular cmpartments and in the phlem sap f spinach leaves. Plant Physil.

155 Riens,8., G. Lhaus, D. Heineke, and H.W. Heldt, 1991, Amin acid and sucrse cntent determined in the cytslic, chlrplastic, and vacular cmpartments and in the phlem sap f spinach leaves. Plant Physil. 97: 227'233 ' Rgger, J.P., Cen, S. and Ragnnet, 8., 1986, High perfrrnance liquid chrmatgraphy suvey n changes in pigment cntent in ripening grapes f Syrah. An apprach t anthcyanin metablism. Am. J. Enl. Vitic. 37:77'83. Rubelakis-Angelakis, K.4., 1991, Amin acid and prtein metablism in Vitis spp', Prceedings f the Internatinal Sympsium n Nitrgen in Grapes and 'Wine' J.M. Rantz (Ed.) pp , American Sciety fr Enlgy and Viticulture. Rubelakis-Angelakis, K.4., and'w. M. Kliewer,lgg2,Nitrgen Metablism in Grapevine' Hrticultural Reivew, Vl. 14: Ruffner, H.P., and J.S. Hawker, 1977, Cntrl f glyclysis in ripening berries f Vitis vinifera. Phytchemistry, 73 : Ruffne, H.P., and W.M. Kliewer, 1975 Phsphenlpyruvate carxykinase EC activity in grape berries. Plant Physil.56:67-71' Ruffner, H.P., W. Kblet, and D. Rast, 1975, Glucnegenesis in the ripening fruit f Vitis vinifera. Vitis, I 3 : Salles, C., H. Essaied, P. Chalier, J.C. Jallageas, and J. Cruzet, 1988, Evidence and characterisatin f glycsidically bund vlatile cmpnents in fruits. In: Biflavur 87. analysis. bichemistry, bitechnlgy. Schreier (Ed.), Walter de Gruyter & C.. Berlin, pp Schreier, P., F. Drawert aand A. Junker, 1976, Identificatin f vlatile cnstituents frm grapes. J. Agric. Fd Chem.?4: Schuls, G. and E. Stahl-Biskup, l99l. Essential Oils and Glycsidic Bund Vlatiles frm Leaves, Stems. Flwers and Rts f Hysspus fficinalis L. (Lamiaceae), Flavur Fragr. J. 6: Seftn, M.4., G.K. Skurumunis. R.A. Massey-V/estrpp, and P.J. Williams, 1989, Nrisprenids in Vitis vinifera white wine grapes and the identificatin f a precursr f damascenne in these fruits. Aust. J. Chem. 42: Seftn, M.4., Francis, I.L., and Williams P.J., 1993, The vlatile cmpsitin f Chardnnay juices: a study by flavur precusr analysis. Vitis, 44:

Seftn, M.4., I.L. Francis, and P.J. Williams, 1994, Free and bund vlatile secndary metablites f Vitis vinifera grape cv. Sauvignn Blanc. Jurnal f Fd Science, 59: 142-147. Sepúlveda, G. and W.M. Kliewer, 1986, Effect f high temperature n grapevines (Vitis vinifera L.

156 Seftn, M.4., I.L. Francis, and P.J. Williams, 1994, Free and bund vlatile secndary metablites f Vitis vinifera grape cv. Sauvignn Blanc. Jurnal f Fd Science, 59: Sepúlveda, G. and W.M. Kliewer, 1986, Effect f high temperature n grapevines (Vitis vinifera L.).il. Sistibutin f sluble sugars. Am. J. Enl. Vitic.37:2-25' Shseyv, O., B.A. Bravd, D. Seigel, A. Gldman, S. Chen, L. Shseyv, and R. Ikan, 199, Imbilised end-ß-glucsidase enriches flavur f wine and passin fruit juice. J. Agric. Fd Chem. 38: Shure, K.B., T.E. Acree, 1994, Changes in the dr-active cmpunds in Vitis labruscana cv. Cncrd during grwth and develpment. J. Agric. Fd Chem. 42: Sidahmed, O.4., and W.M. Kliewer, 198, Effect f defliatin, gibberellic acid and 4- chlrphenxyacetic acid n grwth and cmpsitin f Thmpsn Seedless grape berries. Am. J. Enl. Vitic.31: Simpsn, R.J., and M.J. Dalling,l98l, Nitrgen redistributin during grain grwth in wheat (Triticum aestivum L.) ili. Enzymlgy and transprt f amin acids frm senescing flag leaves. Planta, l5l: Skurumunis, G.K., and P. Winterhalter, 1994, Glycsidically bund nrisprenids frm Vitis vinifera cv. Reisling leaves. J. Agric. Fd Chem.,42: Smers, T.C. l975,ln search f quality fr red wines. Fd technlgy in Australia,2T; Smers, T.C., and G. Ziemelis,1973, Direct determinatin f wine prteins. Am. J. Enl Vitic. 24:47-5. Smers, T.C., 1976, Pigment develpment during ripening f the grape. Vitis 14, Stahl-Biskup E., F. Intret, J. Hlthuijzen, M. Stengele, and G. Schulz, 1993, Glycsidically bund vlatiles- a review Flavur and Fragrance Jurnal. 8: 6l -8. Stitt, M., and W.P. Quick, 1989, Phtsynthetic carbn partitining: its regulatin and pssibilities fr manipulatin. Physil. Plant. 77:

157 Strauss, C.R., B. Wilsn, P.R. Gdley, P.J. Williams, 1986, Rle f mnterpenes in grape and wine flavur. In: "Bigeneratin f armas", sympsium spnsred by the divisin f Agriculnre and Fd Chemistry at 19th meeting f the American Chemical Sciety; Parliment T.H. and R. Crteau, (Eds.) 8-13 Septamber 1985, Chicag, Illinis, ACS sympsium series 317, American Chemical Sciety, Washingtn, DC, pp. 222'242. Strauss, C.R., R.P. Gdley, B. Wilsn, and P.J. Williams, 1987, Applicatin f drplet cunter current chrmatgraphy t the analysis f cnjugated frms f terpenids, phenls, and ther cnstituents f grape juice. J. Agric. Fd Chem. 35: Strauss, C.R.; B.Wilsn and P.J.Williams, 1988, Nvel mnterpene dils and dil glycsides in Vitis vinifera grapes. J. Agric. Fd Chem. 36: ' Sugisawa, H., K. Miwa, T. Matsu, and h. Tamura, 1988, Vlatile cmpunds prduced frm the cultured cells f Thyme (Thymus vulgaris L.). In: Biflavur '87, analysis, bichemistry, bitechnlgy. Schreier (Ed.), 'Walter de Gruyter & C., Berlin. pp Swansn, C.4., and E.D.H. El-Shishini, 1958, Translcatin f sugars in the Cncrd grapes. Plant Physilgy, 33: Taiz. L. and E. Zeiger. 1991, Plant Physilgy, the Benjamin/cummings publishing cmpany. Inc. pp 565. Tazaki.H., N. Ohta. R. Hri. H. Okuyama, M. Okumura, T. Fujimri, and, K. Nabeta, 1993, The glycsides f xygenated lwer terpenes in plant leaves. In: Fd flavurs. ingredients and cmpsitin. Charalambs, G. (ed.), Elsevier Science Publishers B.V. Tesniére C.M., C. Rmieu. and J.-P. Rbin, 1994, Grape inviv prtein synthesis changes in respnse t anaerbisis. Am. J. Enl. Vitic. 45: Turgen, R., J.A. Webb, Leaf develpment and phlem transprt in Cucurbita pep: transitin frm imprt t exprt. Planta I l3: Turgen, R., J.A. Webb. R.F. Every, 1975, Ultrastructure f minr veins in Cucurbita pep leaves. Prtplasma 83 : Urquhart, A.A. and K.S/. Ly. 1981, Use f phlem exudate technique in the study f amin acid transprt in pea plants. Plant Physilgy, 68:

141 Van Bel, 4.J., 1993, Strategies f phlem lading. Ann. Rev. Plant Physil. 44:253-281' Versini, G., A. DalaSeea, M. Dell'Eva, A, Scienza, A.

158 141 Van Bel, 4.J., 1993, Strategies f phlem lading. Ann. Rev. Plant Physil. 44: ' Versini, G., A. DalaSeea, M. Dell'Eva, A, Scienza, A. Rapp, 1988, Evedence f sme glycsidically bund mnterpenes and nrisprenids in grapes. In: Biflavur '87, analysis, bichemistry, bitechnlgy. Schreier (Ed.), Walter de Gruyter & C., Berlin, pp Warlaw, I.F., 199, The cntrl f carbn partitining in plants. New Phytl. 116: Watanabe, Y. and Shiriashi, S.I., 1991, Relatinship between skin clr and anthcyanin f grapes. Science Bulletin f The Faculty f Agriculture, Kyushu University,46:3' 38. Watanabe, Y. K., Sihara, I. and Shiriashi, S.I., 1992, Variatin f anthcyanin cmpsitin in fruit skin f F-l hybrid grapes. Science Bulletin f the Faculty f Agriculture, Kushu University, 46: Weibull, J.; F. Rnquist, and S. Brishammar, 199, Free amin acid cmpsitin f leaf exudates and phlem sap. Plant Physil., 92: Weiner, H., S. Blaechschmidt-Schneider, H. Mhme, W. Schrich, and H.W. Heldt, 1991, Phlem transprt f amin acids. Cmparisn f amin acid cntents f maize leaves and f the sieve tube exudate. Plant Physil. Bichem.29: Wenzel. H., H. Ditrich, and M. Heimfarth, 1987, Die Zusammensetzung der Anthcyane in den Beeren verschiedener Rebsrten. Vitis, 26: Wermelinger, Nitrgen dynamics in grapevine: physilgy and mdelling Prceedings f the Internatinal Sympsium n Nitrgen in Grapes and Wine J.M. Rantz (Ed.) pp American Sciety fr Enlgy and Viticulture. West, C.4., 199. Terpene bisynthesis and metablism. In Plant Physilgy, Bichemistry and Mlecular Bilgy, eds. D.T. Dennis and D.H. Turpin. Lngman Scientific Technical. Jhn Wiley & Sns, New Yrk. pp, Wijayanti, L., M. Kbayashi, S. Fujika, K. Yshizawa, and A. Sakurai, 1995, Identificatin and quantificatin f Abscisci acid, Indle-3-acetic acid and Gibberellins in phlem exudates f Pharbitis nil. Bisci. Bitech. Bichem. 59: 1533-l 535.

Wilden adt, H.L., E.N. Christensen, B. Stackler, A. Caputi, Jr. K. Slinkard, and K. Scutt, 1975, Vlatile cnstituents f grape leaves. I. Vitis vinifera variety 'Chenin Blanc'. Am. J. Enl. Vitic.

159 Wilden adt, H.L., E.N. Christensen, B. Stackler, A. Caputi, Jr. K. Slinkard, and K. Scutt, 1975, Vlatile cnstituents f grape leaves. I. Vitis vinifera variety 'Chenin Blanc'. Am. J. Enl. Vitic.26: Williams, L.E., and P.J. Biscay,IggI, Partitining f dry weight, nitrgen, and ptassium in Chabernet Sauvignn grapevines frm anthesis until harvest. Am' J. Vitic. Enl. 42: ll Williams, P.J. 1993, Hydrlytic flavur release in fruit and wines trugh hydrlysis f nn-vlatile precursrs. In: Flavur Science, sensible principal and Techniques; Acree,T.E., Teranishi, R., Eds.; American Chemical Sciety: V/ashingtn, DC. Williams, P.J., C.R. Strauss, and B. Wilsn, 198, New linall derivatives in Muscat f Alexandria grapes and wines, Phytchem. 19: I I37-1I39. Williams, P.J.; Strauss,C.R.; Wilsn, B. and Massy-V/estrpp, R.A., 1982a, Nvel mnterpen disaccharide glycsides f Vitis vinifera grapes and wines. Phytchemistry, 2 1, Williams, P.J.; Strauss, C. R.; Wilsn, B. and Massy-Westrpp, R.4., 1982b, Studies n the hydrlysis f Vitis vinifera mnterpen prcursr cmpunds and mdel mn terpen p-d- glucsides ratinalising the mn terpen cmpsitin f grapes. J. Agric. Fd Chem.3: Williams, P. J.; Strauss, C.R. and Wilsn,8., 1982c, Use f Cl8 reversed-phase liquid chrmatgraphy fr the islatin f mnterpen glycsides and nr-isprenid precr rsrs frm grape juice and wines. J. Ch matg Williams. P.J., C.R. Strauss. B. Wilsn, and R.A. Massy-Westrpp, 1983, Glycsides f 2-phenylethanl and benzyl alchl in Vitis vinifera grapes. Phytchem. 22: Williams, P.J., Strauss, C.R., V/ilsn, 8., and Dimitriadis, E.,1985, Origin f sme vlatile mnterpenes and nr-isprenids in grapes and wines - bisynthetic and bigenetic cnsideratins. Berger, R.G.; Nitz, S.; Schreier, P., eds. Tpics in flavur research: prceedings f the internatinal cnfernce; l-2 Aplil 1985; Freising-Wrihenstephan, FRG. marzling-hangenham, FRG: H. Eichrn; Williams, P.J., C.R. Strauss, A.P. Aryan, and B. Wilsn, 1987, Grape flavur - a review f sme pre- and pstharvest influences. In: Prceedings f the sixth Australian wine industry technical cnference, Lee, T.H., (Ed.),14-16 July 1986, Adelaide, S.A. Australian Industry Publishers, I l1-ll7. 142

Williams, P.J., M.A. Seftn, B. Wilsn, 1989, Nnvlatile cnjugates f secndary metablites as precurses f varietal grape flavur cmpnents. In Flacur Chemistry: Trends and develpments; Teranishi, R.; R.G.

160 Williams, P.J., M.A. Seftn, B. Wilsn, 1989, Nnvlatile cnjugates f secndary metablites as precurses f varietal grape flavur cmpnents. In Flacur Chemistry: Trends and develpments; Teranishi, R.; R.G. Buttery; F. Shahidi, eds.; ACS Sympsium Series N. 388; American Chemical Sciety: Washingtn DC, pp V/illiams, P.J., M.A. Seftn, and I.L. Francis, 1991, Glycsidic precursrs f varietal grape and wine flavur. In: Thermal and Enzymatic Cnversins f Precursrs t flavur cmpunds. ACS Sympsium series. V/illiams, P.J., M.A. Seftn, and V.A. Marins, 1993, Hydrlytic flavur release frm nn-vlatile precursrs in fruits, wines and sme ther plant derived fds. In: Recent develpments in flavur and fragrance chemistry. Prceedings f the Third Haarman and Reimer Sympsium; Hpp, R' Mri, K. Eds.; VCH Verlagsgesellschaft : Weinheim, pp Williams, P.J.; W. Cynkar; I.L. Francis; J.D. Gray; P.G. Iland, and B.G. Cmbe, 1995, Quantificatin f glycsides in grapes, juice, and wines thrugh a determinatin f glycsyl glucse. J. Agric. Fd Chem. 43 l2l-i28. Wilsn, 8., C.R. Strauss, and P.J. 'Williams, 1984, Changes in free and glycsidically bund mnterpenes in develping Muscat grapes. J. Agric. Fd Chem.32: 'Wilsn, 8., C.R. Strauss, and P.J. Williams, 1986., The distributin f free and glycsidically-bund mnterpenes amng skin, juice, and pulp fractins f sme white grapr varieties. Am. J. Enl. Vitic. 37:17-llI. Winkler. 4.J.. J.A. Ck. W.M. Kliewer, and L.A. Lider General Viticulture Univ.f Califrnia Press. Berkeley, Clifrnia. Winterhalter, P., and P. Schreier. 1988a, Free and bund C13 nrisprenids in Quince (Cydnia blangu Mill. ) fruit. J. Agric. Fd Chem. 36: 125l Winterhalter, P., and P. Schreier, 1988b, Studies n C13-nrisprenid precursrs. In: Biflavur'87, analysis, bichemistry, bitechnlgy. Sch eier (Ed.), Walter de Gruyter & C.. Berlin, pp Winterhalter, P. and P. Schreier,1994, C13-Nrisprenid glycsides in plant tissues: an verview n their ccurrence, cmpsitin and rle as flavu precursrs. Flavur and Fragrance Jurnal, 9 :

V/interhalter, P., M.A. Seftn, and P.J. Williams, 199a, Tw-dimensinal GC-DCCC analysis f the glyccnjugates f mnterpenes, nrisprenids, and shikimate derived metablites frm Riesling wine. J. Agric.

161 V/interhalter, P., M.A. Seftn, and P.J. Williams, 199a, Tw-dimensinal GC-DCCC analysis f the glyccnjugates f mnterpenes, nrisprenids, and shikimate derived metablites frm Riesling wine. J. Agric. Fd Chem. 38: Winterhalter P., M. A. Seftn, and P.J. rù/illiams, 199b, Vlatile C13-nrisprenid cmpunds in Riesling wine are generated frm multiple precursrs. Am. J. Enl. Vitic. 4I: Wding, F.B.P., 1978, Phlem. Carlina Bilgy Readers, J.J. Head (Ed.), Carlina Bilgical Supply Cmpany, Oxfrd university Press, 2nd ed.pp. 16. Wulf, L.V/. and C.V/. Nagel, 1978, High pressure liquid chrmatgraphy separatin f anthcyanins î Vitis vinifera. Am. J. Enl. Vtic. 29: Ziegler, H., 1975, Nature f transprted substances. In: Encyclpedia f plant physilgy, New Series. Vl. 1, Transprt in plants, pp , Zimmernann M.H. and J.A. Milburn, eds. Springer Verlag, Berlin. Ziegler, H., 1956, Utersuchungen iiber die leitung und sekretin der assimilate. Planta, Zimmermann, M.H., lg5t,translcatin f rganic substances in trees. I. The nature f the sugar in the sieve tube exudate f trees. Plant Physilty, 32:288'291. Zimmermann, M.H.. 196, Transprt in the phlem, Ann. Rev. Plant physil. ll:

145 Appendix Table 1 The seasnal changes in phlem sap amin acids f Muscat Grd, Sultana and Shiraz. Values given arey f tøl amin acids befre veraisn (B), at veraisn (V), and at ripening time (R).

162 145 Appendix Table 1 The seasnal changes in phlem sap amin acids f Muscat Grd, Sultana and Shiraz. Values given arey f tøl amin acids befre veraisn (B), at veraisn (V), and at ripening time (R). Muscat Grd Sultana Shiraz Amin acids (B) (V) (R) (B) ('v) (R) (B) CV) (R) ASP GLU ASN SER GLN HIS GLY THR ALA ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU LYS PRO Ttal * t , r I l.l r I t.l r , I t t t t l t J I I Brix tr Ttal values are in pmles amin acids/ 1.5 ml f buffer slutin/ 4 h t.7 r l.l 1.8 ) t The data are ÍÌm ( 1995).

163 146 Table 2Thedifferences in amin acid prfiles f leaf exudates in light and dark cllected at berry ripening time (% f ttal). Amin acids Muscat Grd Sultana Shiraz Muscat Grd Sultana Shiraz Dark (1995) Lisht (1994) ASP GLU r ASN.5.3 SER GLN HIS GLY THR ALA 2.4 JJ.Z J ARG TYR.7.3 CYS-CYS VAL r MET TRP PHE 1.5 I.6 t.7 ILE LEU r LYS PRO r Ttal * *Ttal values are in rmles amin acidsi 1.5 ml f buffer slutin exuded fim l5 leaves /16 h

164 147 Table 3 The seasnal changes in amin acid prhles (% f ttal) f Muscat Grd peduncle exudates. Amin acids ASP GLU ASN SER GLN HIS GLY THR ALA ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU LYS PRO Sampling date 3l1l9s rvv95 l9lrl95 t2l2l9s t5, /4/ t t t I Ttal * b "Brix t.r 23.8 t4.l t *Ttal values are in pmles anrin acidsi 1.5 ml f buffer slutin/ 4 h t I t I t.l I t I 1.1 r.2.1 t.2

148 Table 4 The seasnal changes in amin acid prfiles (% f ttal) f Sultana peduncle exudates. Sampling date Amin acids 29n2194 r2ly95 912195 tr/3t95 ASP 2.9 2.6 4.2 6.9 GLU 1.2 4.7 8.5 1.4 ASN.3.5.2.7 SER 3.

165 148 Table 4 The seasnal changes in amin acid prfiles (% f ttal) f Sultana peduncle exudates. Sampling date Amin acids 29n2194 r2ly tr/3t95 ASP GLU ASN SER J.J 2.4 GLN r 51.7 HIS.8 t GLY I THR ALA t t ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU.3 I.4.7 LYS PRO Ttal * Brix ì 7.7 l *Ttal values are in rmles amin acids,' I.5 ml f buffer slutin/ 4 h

LYS PRO 8.3 6.8.3 4.5 68.8 t.7 1.5 1.7 1.6.6.8.6.2.3.4.4.4 1.1 ll 12.2.7 6.2 49.9 1.3 J 2.4 3.2 4.t.8 I.4.2.9.6.7.6 1 t8 18.

166 149 Table 5 The seasnal changes in amin acid prfiles (% f ttal) f Shiraz peduncle exudates. Sampling date Amin acids 221t / 95 tv3lgs ASP GLU ASN SER GLN HIS GLY THR ALA ARG TYR CYS-CYS VAL MET TRP PHE ILE LEU LYS PRO t ll J t.8 I t Ttal * "Brix *Ttal values are in rmles anrin acidsi 1.5 ml f buffer slutin/ 4 h t t

167 15 Fig. 1 The inflrescence grwn frm drmant bud f Muscat Grd bud-grafted nt a Shiraz cane.

OF "THOMPSON SEEDLESS" GRAPEVINES

OF THOMPSON SEEDLESS GRAPEVINES EFFECT OF TIME AND SEVERITY OF DEFOLIATION ON GROWTH OF ROOTS, TRUNK, AND SHOOTS OF "THOMPSON SEEDLESS" GRAPEVINES W. MARK KLIEWER and R. D. FULLER Respectively Assciate Bichemist, Department f Viticulture