Table of contents. Campinas (SP), Brazil

Index Table of contents Construction of a Genetic Map Based on an Interspecific F 2 Population between Coffea arabica and Coffea canephora and its Usefulness for Quality Related Traits R.H.G PRIOLLI, L.C.S. RAMOS 2, D. POT,, M. MOLLER 2, P.B. GALLO 2, M.M. PASTINA, A.A.F. GARCIA, P.Y. YAMAMOTO 2, S.D. LANNES, L.P. FERREIRA, M.B.S. SCHOLZ, P. MAZZAFERA, L.F.P. PEREIRA, C.A. COLOMBO 2 ESALQ/USP, Departamento de Genética, Piracicaba (SP), Brazil 2 IAC (Instituto Agronômico de Campinas), CPD de Recursos Genéticos Vegetais, Campinas (SP), Brazil IAPAR (Instituto Agronômico do Paraná), Londrina (PR) Brazil Cirad, UMR DAP, Montpellier, France UNICAMP, Instituto de Biologia, Campinas (SP), Brazil Embrapa Café, Brasília, DF, Brazil SUMMARY Genetic maps based on molecular markers have been developed in a large set of plants and this strategy has proven its efficiency towards the identification of tools for marker assisted selection. In the present study, AFLP and SSR markers were used to build a genetic map of an interspecific F 2 population between Coffea arabica and Coffea canephora. It was identified AFLP markers and SSR alleles segregating in F 2 plants from forty four AFLP combinations and SSR loci. For the map construction, only single dose markers segregating : in the F 2 were considered (2 AFLP markers and 2 SSR alleles, standing for to.% of the polymorphic markers). The genetic map was build and one hundred and sixty nine markers were mapped, corresponding to AFLP markers and SSR loci. Thirty seven linkage groups corresponding to a total map length of cm were obtained, with an average distance between the markers of. cm and an average of. markers per linkage group. Forty-six marker trait associations were found; of which, nineteen were associated with sugar content, eight for caffeine, eight for CGA, one for caffeine and CGA and ten for total production per plant. Only four single markers associations were detected at both years of determinations. The single markers analysis for QTL detection allowed us to obtain previous information of putative QTL association for coffee quality and productivity. Additional markers are being added to this working linkage map for more complete coverage of the coffee genome. INTRODUCTION C. arabica L. the only self-fertile tetraploid species (2n = x = ) of the Coffea genus, is characterized by low genetic diversity which has been attributed to its allotetraploid origin, its reproductive biology and its domestication history. In contrast, diploid species from the coffea genus (2n = 2X = ) are alogamous and are highly diverse at the phenotypic and molecular levels. These species form valuable gene reservoirs for different breeding purposes (Carvalho ). 2

Transfer of desirable genes from C. canephora to C. arabica varieties through interspecific crosses is one of the breeding strategies used for coffee improvement. C.arabica x C.canephora hybrids, resulting from the hybridization between C. arabica and colchicine doubled C. canephora have been cited as reasonably fertile (Berthaud, ; Owuor and Van der Wossen, ). They are also particularly favorable to intergenomic recombination and gene introgressions (Lashermes et al., 2; Herrera et al., ; Priolli et al., ). Molecular markers are being used successfully in many crops to assist directed germplasm improvement. Marker-assisted selection allows screening of large numbers of trees for a gene of interest at early stage and reduces the number of backcrosses required to select elite genotypes (Lashermes et al., ). In coffee, a saturated Coffea canephora genetic was built by Lashermes et al. () and others partial genetic maps were obtained for interspecific crosses involving diploid species (C. pseudozanguebarie x C. liberica (Ky et al., 2) e C. canephora x C. heterocalyx (Coulibaly et al.2)). Identification of quantitative trait loci allowed the localization of two markers that flanked a fructification time genomic region (Akaffou et al. 2) and three markers associated with pollen viability (Coulibaly et al., 2), which could be used for early marker-assisted selection. In C. arabica, according to its low polymorphism level associated with its tetraploid, the strategy consists on the construction of partial genetic maps (Pearl et al., 2; Teixeira- Cabral et al., 2) with posterior integration of the partial genetic maps. In the present study, AFLP and SSR markers were used to build a genetic map of an F 2 interspecic population between C. arabica and C.canephora. In addition, association between segregating markers and quality related traits were analyzed. MATERIAL AND METHODS Plant material The F tetraploid hybrid between Coffea arabica L. var. Bourbon Vermelho and Coffea canephora var. Robusta x, an artificial tetraploid obtained by Mendes (), has, since, been advanced to F 2 by selfing three F clones from the same plant. The F 2 segregating population was grown in a field trial at a site near the municipality of Mococa (latitude º ' S, longitude º ' W and altitude m) in São Paulo State Brazilian state received treatment with inorganic fertilizer, and weed and pest control and all other treatments recommended for growing coffee under Brazilian conditions (Thomaziello et al., ). Sample preparation and field data Leaves and fruits from each F2 were harvested two successive years (2 and ). Leaves were collected from the third and fourth leaf pairs from different sides of the tree canopy. Fruits were harvested at mature stage for analysis of caffeine, chlorogenic acids and sugar contents. Seeds were manually removed from the pericarp, dried at C for two weeks and then finely ground with a blade grinder or with a pestle and mortar. Caffeine and chlorogenic acids were extracted according to Priolli et al. (). Total and reducing sugars were extracted according to Rogers et al. () and quantified using Somogyi and Nelson reagent. Sucrose

content was estimated by the substraction of reducing sugars content from total sugar contents. Production was also evaluated (Kg fruits / Plant). Molecular marker assay Total genomic DNA was extracted from freeze-dried leaves of the parental, F hybrid and F 2 genotypes as described by Ky et al. (2). Nineteen microsatellite loci, previously identified as polymorphic between C. arabica and C. canephora, were analyzed using PCR. Some of these microsatellite loci (Table ) have been mapped in C. canephora (Lashermes et al, ) and other were obtained by Combes et al (2). The specific primer pairs, amplification conditions, radioactive labelling and polyacrylamide gel electrophoresis were as reported elsewhere (Priolli et al., ). The amplified fragment length polymorphism (AFLP) procedure was performed as previously reported (Vos et al., ). Briefly, ng of genomic DNA was digested with the restriction enzymes EcoRI and MseI. Restriction fragments were then ligated with double-strand EcoRI and MseI adapters. A selective preamplification was performed using the appropriate primers (named E and M, respectively) without selective nucleotide at the end (ie E+/M+). The reaction mixture was diluted / and µl was used for the final amplification with two primers, each containing three selective nucleotides (Table ). Data analysis Only markers polymorphic between the parents and present in the F were considered. This strategy was applied whatever the marker used. Consequently, both SSRs and AFLPs were considered here as dominant markers. Segregation distortion from the : expected ratio for single dose markers was analyzed by the chi-square test. Bonferoni correction was applied to control type I error for multiple tests. Map construction was carried out using Joinmap version. (Stam ). Linkage groups were established using two-point analysis with LOD threshold values of and recombination fraction of.. The Kosambi function was used for converting recombination fractions into map distances. Associations between markers and the analyzed traits (total sugar, reducing sugar, sucrose, caffeine, chlorogenic acids contents and production) were analyzed by one-way ANOVA. Significant association were considered for P value lower than. and suggestive associations were considered for P value between. and.. RESULTS AND DISCUSSION Linkage map Fifty alleles of SSR markers and AFLP markers were polymorphic in F 2 populations. Chi-square analysis revealed 2 markers (%) fitted a : ratio. The remaining (%) showed segregation distortion within the population. One hundred and sixty nine markers were mapped, corresponding to AFLP markers and SSR alleles. % of the mapped markers came from the Coffea arabica parent, % from the Coffea canephora x and % from both parent. These results suggest that the divergence between the two ancestral genomes of Coffea arabica (i.e C. canephora and Coffea eugenioides) is two times larger than the one between the two haplotypes of the C. canephora genotype used as a parent to create the F hybrid. Similar results were observed by Pearl et al (2) who found in a pseudo F 2 population derived from a cross between the cultivars of C. arabica, % of AFLP markers were from cv. Catimor, % from cv. Mokka hybrid, and 2% were codominant.

Thirty seven linkage groups corresponding to a total map length of cm were obtained, with an average distance between the markers of. cm and. markers per linkage group (Figure ). Twenty linkage groups include only 2 markers. (A) EM_ EM_2 EM_ EM_ EM_ EM_2 C2_ EM_ EM_ E2M_ EM_ EM_ EM_ EM2_ E2M_2 2 E2M_ 2 EM_ C2_ 2 EM_ EM_ EM2_ EST_2 EM_2 EM2_2 EM_ EM_2 EM_ 2 EM2_ EM_2 EM_ 2 EM_ EM_ EM_ 2 E2M_22 EM2_ EM_ EM2_ EM_ E2M_2 EM_ EM_ EM_ EM_ EM_ EM_ EM_ EM_ EM2_2 EM2_2 EM2_ 2 M_ 2 EM_ EM2_ EST_2 EM_ EM_ EM_ EM_2 EM_ EM_2 EM_ EM_ EM_ EM_2 C2_ EM_ EM_ E2M_ EM_ EM_ EM_ EM2_ E2M_2 2 E2M_ 2 EM_ C2_ 2 EM_ EM_ EM2_ EST_2 EM_2 EM2_2 EM_ EM_2 EM_ 2 EM2_ EM_2 EM_ 2 EM_ EM_ EM_ 2 E2M_22 EM2_ EM_ EM2_ EM_ E2M_2 EM_ EM_ EM_ EM_ EM_ EM_ EM_ EM_ EM2_2 EM2_2 EM2_ 2 M_ 2 EM_ EM2_ EST_2 EM_ EM_ EM_ EM_2 EM_ EM2_ EM_ EM2_2 EM_ EM_2 EM_ 2 EM_2 EM_ EM_2 EM_ EM_ EM_ E_ EM_ E2M_ E2M_ 2 2 EM2_ EM2_2 EM_ EM_2 EM_ EM_ EM_ EM_ E2M_ 2 EM2_ EM_ EM_ EM_2 EM2_ M2_ EM_ EM2_ EM_ EM2_2 EM_ EM_2 EM_ 2 EM_2 EM_ EM_2 EM_ EM_ EM_ E_ EM_ E2M_ E2M_ 2 2 EM2_ EM2_2 EM_ EM_2 EM_ EM_ EM_ EM_ E2M_ 2 EM2_ EM_ EM_ EM_2 EM2_ M2_ E_ EM_ EM_ EM_ EM2_ 2 EM_ EM_ E2M_ EM_ 2 EM_ EM_ EM_ E2M_ EM_2 EM_ EM_ EM_2 EM_ EM_ EM_ E2M_ 2 EM_2 EM_ 2 EM_2 EM_ EM_ EM_2 EM_ EM2_ EM2_ EM_ 2 EM_2 EM_ EM_2 E2M_ EM_2 EM_ EM2_ EM2_2 2 EM_2 EM_ EM_ E_ EM_ EM_ EM_ EM2_ 2 EM_ EM_ E2M_ EM_ 2 EM_ EM_ EM_ E2M_ EM_2 EM_ EM_ EM_2 EM_ EM_ EM_ E2M_ 2 EM_2 EM_ 2 EM_2 EM_ EM_ EM_2 EM_ EM2_ EM2_ EM_ 2 EM_2 EM_ EM_2 E2M_ EM_2 EM_ EM2_ EM2_2 2 EM_2 EM_ EM_

2 2 2 2 EM_2 EM_ E2M_ EM_2 EM_ EST2_ EM_ EM_ EM_2 EM_ EM_ EM2_ EM_ EM_ 2 EM_ EM_ EM_ EM_2 EM_ 2 M_ EM_ M2_ EM_ M_ M2_ EM_ EM_ EM2_ EM_ EM_ EM_ EM_ 2 EM_2 EM_ E2M_ 2 C2_ EM_ (B) Figure.Genetic linkage map constructed from C.arabica x C.canephora containing AFLP markers and SSR markers. Mapping distances are represented in centimorgans (cm) and markers codes are provided on the right side of each linkage group. Segregation analyses of restriction fragment length polymorphism (RFLP) loci-markers have indicated tetrasomic inheritance resulting from the pairing of homologous chromosomes in meiosis of first generation C. arabica x C. canephora x hybrids (Lashermes et al., 2). In two BC F populations, (C. arabica C. canephora x) C. arabica, segregations and cosegregation of RFLP and microsatellite loci-markers conformed to the expected ratio assuming random chromosome segregation and the absence of selection (Herrera et al., ). In a study using SSR loci, the hybrid F showed that the ratios of the gametes genotype did not differ significantly from those expected assuming random associations and tetrasomic inheritance (Priolli et al., ). In C. arabica, according to its low polymorphism level associated with its tetraploid, the strategy consists on the construction of partial genetic maps with posterior integration of the partial genetic maps. A linkage map of arabica coffee was constructed from AFLP primer combinations resulting in major linkage groups containing - markers, and small linkage groups consisting of 2- linked markers. The total length of the map was,2. cm, with an average distance of.2 cm between adjacent markers (Pearl et al., 2). In a backcross population of the C. arabica, a partial genetic map was constructed with 2 RAPD loci and covered the estimated length of.cm, average distance of. cm between adjacent markers in a total of eight linkage groups (Teixeira-Cabral et al 2). In C. canephora, a genetic map of 2 cm with linkage groups was reported using RFLP and RAPD markers (Paillard et al., ). Eleven linkage groups that putatively correspond to the gametic chromosomes of C. canephora were identified from 2 loci ( AFLP, RAPD, microsatellite, and RFLP) in a total map length of cm and

average distance of. cm. (Lashermes et al., ). Genetic maps for interspecific diploid crosses were also obtained for C. pseudozanguebarie x C. liberica (Ky et al., 2) and C. canephora x C. heterocalyx (Coulibaly et al., 2) leading to the identification of linkage groups covering, cm and linkage groups with, cm respectively. Table. List of SSR loci and selective AFLP primers (EcoRI+ and MseI+) with their codes presents in C.arabica X C.canephora genetic map. Loci SSR Code Primer AFLP Code Primer AFLP Code -2CTG M ACC E CAA M 2-2CTG C2 ACT E2 CAG M C2-2CATC C2 AAC E CGA M E-CTG E ACA E CGT M E-CTG E AAG E CGG M EST EST AGC E CCC M2 EST2 EST2 AGG E CGC M EST EST ACG E CCG M M M CAC M CCA M M M CTG M2 M2 M2 CTT M M2 M2 CTC M M2 M2 CAT M M M CTA M Single marker analysis Single marker analysis to detect associations between phenotypic traits including total sugar, reducing sugar, sucrose, caffeine, chlorogenic acids contents and production and molecular markers were performed. Two threshold levels corresponding to significant association (P value <.) and suggestive level (. < P value <.) were considered (Table 2). Overall, marker trait associations were found corresponding to seven from SSR makers and from AFLP markers. The percentage of the phenotypic variance explained by each marker ranged from.2 (EM_ marker) to 2.% (EM_ marker). In relation to biochemical contents, ten marker-trait associations were found for total sugar, seven to reducing sugar, eleven to sucrose, nine to caffeine and nine to CGA. For field production data, ten markers presenting suggestive and significant effect were detected. Nine markers were associated to total sugar and sucrose but only two of these markers presented significant or suggestive effects in the two years analysed (EM_ and EM_). Marker EM_ also presented association with caffeine content in the two successive years, the same effect stability was observed for EM_ in relation with the production level. These markers can be viewed as consistent markers, with potential to be applied in a further marker assisted selection. In a few cases the same markers presented association with two distinct traits. Such results were observed for marker E2M_2 for caffeine and GCA contents. The same observation was done for markers that presented associations with total sugars and sucrose contents. However such association was expected as sucrose content was derived from the total sugar and reducing sugar contents. Some QTLs in Coffea were detected using segregating population, derived from interspecific crosses. Three significant QTLs (LOD > and p <. by ANOVA) were detected for

pollen viability in a backcrossed progenies originating from a cross between Coffea canephora and Coffea heterocalyx (Coulibaly et al., 2). Table 2. Single marker-test for total and reducing sugar (TS and RS), sucrose, caffeine, chlorogenic acids (CGA), and production at years 2 and, with their effects in % of the phenotypic variability. Markers TS RS Sucrose Caffeine CGA Production 2 2 2 2 2 2 EM_*.2 2.. 2.2 EM_**.... EM2_**.. EM2_2* 2.. EM_*.. EM_*.2. EM_2* 2. 2. EM_**.. EST_*.. E_*. EM_2**. EM_** 2.2 EM_* 2. M_**. EM_**. EM_*. M_**. EM_2*. E_*. EM_*.. EM_2**. EM2_*. EM_**. EM_2**. EM_*. EM_2* 2. EM_* 2. E2M_2**.. EM_*. EM_2**. E2M_**. EM_*.2 EM_**. EM_**. EM_**. EM_**. E2M_* 2.2 E2M_**. EM_** 2. EM_**.. EM_2**. EM_**. EM_*.2 EST_2**.2 C2_** 2. EM_2*. *Significance at P <. **Significance at P <.. Bold letters: significance at 2 and. Red letters indicate one mark associated with two traits.

Only one QTL was identified for fructification time using a one-way ANOVA with a significance level of P <. in a cross between Coffea pseudozanguebariae X C. liberica var. Dewevrei. The QTL was located on linkage group E defined by Ky et al. (2), between the AFLP marker ACCCTT and the RFLP marker G The ACCCTT marker explained % of the fructification variance (Akaffou et al., 2). The single markers analysis allowed us to obtain preliminary information of putative QTL for biochemical components involved in the quality of coffee beverage an their relation with production. However, others QTL detection approaches such as interval mapping (Lander and Botstein, ) and composite interval mapping (Zeng, ), both, having more power to detect single QTL marker association are planned to be applied to our data set in a short-term. REFERENCES Akaffou D.S., Ky C.L., Barre P., Hamon S., Louarn J., Noirot M. (2) Identification and mapping of a major gene (Ft) involved in fructification time in the interspecific cross Coffea pseudozanguebariae X C. liberica var. Dewevrei: impact on caffeine content and seed weight Theor Appl Genet (2) :- Berthaud J. () L hybridation interspécifique entre Coffea arabica L. et Coffea canephora Pierre. Obtention et comparaison des hybrides triploïdes arabusta et hexaploïdes. Café Cacao Thé : - Carvalho A. () Principles and practices of coffee plant breeding for productivity and quality factors: Coffea arabica. In Coffee. Vol.. Agronomy. Edited by R.J. Clarke and R. Macrae. Elsevier Applied Science, London. pp. 2-. Combes M.C., Andrzejewski S., Anthony F., Bertrand B., Rovelli P., Graziosi G. and Lashermes P. (2) Characterization of microsatellite loci in Coffea arabica and related coffee species. Mol Ecol : -. Coulibaly I., Louarn J., Lorieux M., Charrier A., Hamon S., Noirot M. (2) Pollen viability restoration in a Coffea canephora P. and C. heterocalyx Stoffelen backcross. QTL identification for marker-assisted selection Theor Appl Genet : - Coulibaly I., Revol B., Noirot M., Poncet V., Lorieux M., Carasco-Lacombe C., Minier J., Dufour M., Hamon P. (2) AFLP and SSR polymorphism in a Coffea interspecific backcross progeny [(C. heterocalyx X C. canephora) X C. canephora] Theor Appl Genet : - Herrera J.C., Combes M.C., Anthony F., Charrier A. and Lashermes P. () Introgression into the allotetraploid coffee (Coffea arabica L.) segregation and recombination of the C. canephora genome in the tetraploid interspecific hybrid (C. arabica X C. canephora). Theor Appl Genet : -. Ky C.L., Barre P., Lorieux M., Trouslot P., Akaffou S., Louarn J., Charrier A., Hamon S. and Noirot M. (2) Interspecific genetic linkage map segregation distortion and genetic conversion in coffee (Coffea sp.). Theor Appl Genet : -. Lander E.S., Botstein D. () Mapping Mendelian factors underlying quantitative traits using RFLP linkage maps Genetics. : -. Lashermes P., Combes M.C., Robert J., Trouslot P., D hont A.A.F. and Charrier A. () Molecular characterization and origin of the Coffea arabica L. genome. Mol. Gen. Genet. :.

Lashermes P., Paczek V., Trouslot P., Combes M.C., Couturon E. and Charrier A. (2) Single-locus inheritance in the allotetraploid Coffea arabica L. and interspecific hybrid C. arabica X C. canephora. J Heredity : - Lashermes P., Combes M.C., Prakash N.S., Trouslot P., Lorieux M. and Charrier A. () Genetic linkage map of Coffea canephora: effect of segregation distortion and analysis of recombination rate in male and female meioses. Genome : - Mendes A.J.T. () Observações citológicas em Coffea: Métodos de tratamento pela colchicina. Bragantia : 2-2. Owuor J.B.O. and Van der Vossen H.A.M. () Interspecific hybridization between Coffea arabica L. and tetraploid Coffea canephora Pierre: I: Fertility in F hybrids and backcrosses to C. arabica. Euphytica : -. Paillard M., Lashermes P. and Pétiard V. () Construction of a molecular linkage map in coffee. Theor Appl Genet : -. Pearl H.M., Nagai C., Moore P.H., Steiger D.L., Osgood R.V. and Ming R. (2) Construction of a genetic map for arabica coffee. Theor Appl Genet : 2-. Priolli R.H.G., Mazzafera P., Siqueira W.J.G., Möller M., Zucchi M.I., Ramos L.C., Gallo P.G., Colombo C.A. () Caffeine inheritance in interspecific hybrids of Coffea arabica X Coffea canephora (Gentianales, Rubiaceae). Genetics and Molecular Biology : - Stam P. () Construction of integrated genetic linkage maps by means of a new computer package: JoinMap. Plant J. : -. Teixeira-Cabral T.A., Sakiyama N.S., Zambolim L., Pereira A.A. and Schuster I. (2) Single-locus inheritance and partial linkage map of Coffea arabica L. Crop Breeding and Applied Biotechnology : -. Thomaziello R.A., Oliveira E.G., Toledo Filho J.A. and Costa T.E. () Cultura do café, Fundação Cargill, Campinas, pp. Vos P., Hogers R., Bleeker M., Reijans M., Van de Lee T., Hornes M., Frijters A., Pot J., Peleman J., Kuiper M., Zabeau M. () AFLP: a new technique for DNA fingerprinting. Nuclei Acid Res 2: -. Zeng Z.B. () Precision mapping of quantitative trait loci. Genetics : -.