Response of Three Brassica Species to High Temperature Stress During Reproductive Growth S. V. Angadi 1 *, H. W. Cutforth 1, P. R. Miller 2, B. G. McConkey 1, M. H. Entz 3, S. A. Brandt 4 and K. M. Volkmar 5 1 Semiarid Prairie Agricultural Research Centre, Swift Current; 2 Montana State University, Bozeman, Montana; 3 University of Manitoba, Winnipeg; 4 Scott Experimental Farm, Scott; 5 Brandon Research Centre, Brandon. ABSTRACT The effect of short periods of high temperature stress on the reproductive development and yield of three Brassica species were studied in a growth chamber experiment conducted for two years. Two genotypes from Brassica juncea L. and one each from B. napus L. and B. rapa L. were grown under day/night temperatures of 20/15 /C till early flowering or early pod development, then were subjected to high temperature stress of 28/15 /C or 35/15 /C for seven days and then were allowed to recover at 20/15 /C. Species differed in optimum temperatures, with B. juncea. and B. rapa. having higher optimum temperature than B. napus. Dry matter was unaffected by moderate temperature stress, while it reduced by high temperature stress. The 35/15 /C treatment was injurious to reproductive organs at different developmental stages of all 3 species. High temperatures at flowering affected yield formation more than high temperature at pod development. On the main stem, mean seed yield reduction due to heat stress was 89 %, but partial compensation by pods on the branches reduced mean per plant seed yield decrease to 52 %. Reduction in fertile pods (not total pod number), thousand seed weight and seeds per pod were responsible for the reduced seed yield. B. rapa L. was more sensitive to heat stress than B. napus and B. juncea. A direct temperature effect on reproductive organs appeared to be responsible for the reduction in yield. All genotypes began to recover from the stress by continuing flowering after returning to 20/15 /C. B. napus. was least able to recover from severe stress at flowering, as evidenced by the formation of many abnormal pods during recovery. Thus, heat stress effect depends on the growth stage in canola and mustard and Brassica species differ in heat stress response. INTRODUCTION High temperature stress is one of the most important but least studied abiotic stresses affecting plant productivity around the world. A field grown crop is under temperature stress for much of the season. The yield losses due to high temperature are large and are often combined with losses from other environmental stresses. The direct effects of high temperature stress depend on the crop species and its adaptability. There is a need to understand the relationship between heat stress and yield for major crops in the semiarid prairie. Canola (B. napus and B. rapa) and mustard (B. juncea) are important oilseed crops in Canada and the acreage of canola in the Brown and Dark Brown soil zones of the Canadian prairies, where it has not been traditionally grown, is increasing. Canola is a cool season crop and is believed to suffer from high temperature stress. Mustard is reported to be better adapted than
canola to semiarid prairie conditions. Research on the direct effect of high temperature stress on Brassica species is limited. Hence, there is an urgent need for evaluating high temperature stress characteristics of Brassica species to predict its suitability and proper management for the semiarid prairie cropping system. The aims of the current study were to determine the most sensitive crop growth stage for high temperature stress in three oilseed Brassica species, the variation in critical temperature among the Brassica species, and to assess whether canola quality Brassica juncea is more tolerant to high temperature than the other Brassica species. Finally, the Brassica species were evaluated for their ability to recover from heat stress. MATERIALS AND METHODS The experiment was conducted at the Semiarid Prairie Agricultural Research Centre (SPARC), AAFC, Swift Current, Canada during 1997-1998 and 1998-1999. All plants were grown in a controlled environment chamber at 20/15 /C day/night temperature until the high temperature treatments were imposed. High temperature treatments were imposed in a growth cabinet for 7 days either at early flower or at early pod stages. After heat stress treatment plants were returned to 20/15 /C growth chambers. Five temperature regimes were used: 1) control = continuous 20/15 /C; 2) 28/15 /C at early flower; 3) 28/15 /C at early pod; 4) 35/15 /C at early flower; and 5) 35/15 /C at early pod. Four Brassica spp. genotypes were used in this study: J90-4316, a canola-quality B. juncea breeding line developed at the AAFC Saskatoon Research Centre (pers. comm., Dr. G. Rakow, Research Scientist, Canola Genetics); Oriental mustard (B. juncea) cv. Cutlass; Argentine canola (B. napus) cv. Quantum; and Polish canola (B. rapa) cv. Maverick (1997) and cv. Parkland (1998). The experimental design was a split plot, with temperatures as main plots and genotypes as sub-plots. The experiment was repeated twice. Common observations (DM, seed yield and HI) from the first and the second run were pooled after testing the homogeneity of error variance using Bartlett s test and analyzed using a split-split-plot design (GLM procedure, SAS Institute Inc., Cary, NC). The separate observations made during each run were analyzed using a split-plot design. Fisher protected LSD was used for all mean separation analysis. RESULTS AND DISCUSSION WHOLE PLANT RESPONSE. Both temperature and genotypes influenced seed yield significantly. Pooled over genotypes and runs, 35/15 /C temperature decreased seed yield per plant by 2.18 g (35 %) compared to the control (Table 1). The early flowering stage was more sensitive to 35/15 /C stress with 53 % yield decrease over the control than early pod stage with 18 % yield reduction. Highest seed yield was produced by B. napus, while the lowest was with B. rapa. Although, the seed yield was lowest with 35/15 /C @ Early Flower in all genotypes, interaction between genotype and temperature was significant. For example, 35/15 /C @ Early Flower reduced seed yield of Cutlass by 40 % while the same stress in B. rapa reduced yield by 80 %. Thus, irrespective of Brassica species, 35/15 /C temperature for one week during early flowering reduced seed yield drastically. Harvest indices (HI) corresponded closely with the seed yield data, consistent with our finding that reductions in seed yield were greater in magnitude than the corresponding reductions
in shoot dry matter. High temperature of 35/15 /C reduced HI significantly (Table 1). Early flowering stage was more sensitive than early pod stage. Two B. juncea genotypes, pooled over temperature treatments, had higher HI than B. napus which in turn had higher HI than B. rapa. The responses of different genotypes to high temperature stress were different. For example, 28/15 /C temperature had no influence on HI except in B. rapa, which recorded significantly higher HI with 28/15 /C @ Early Flower. Another exception was, lower HI in 35/15 /C @ Early Pod was observed only in B. napus and B. rapa. Thus, short period of heat stress during early flowering reduced HI in mustard and canola. MAIN SHOOT RESPONSE. There was a significant effect of one week of high temperature on seed yield and yield components of the main shoot (raceme) (Table 2). Seed yield of the main shoot of all genotypes was reduced by 35/15 /C @ Early Flowering relative to the control while yield increased with 28/15 /C @ Early Flowering for Cutlass and J90-4316. Except J90-4316, this response is similar to the pooled whole plant yield response of all genotypes. This suggests that the optimum daytime temperature for seed yield of the main shoot of Cutlass and J90-4316 is closer to 28 /C than to 20 /C or 35 /C. In contrast, the decrease in seed yield of the main shoot at 28/15 /C at flowering for Quantum (which was similar to whole plant yield response), although not statistically significant, suggest that the optimum for Quantum may be lower than 28 /C. All genotypes typically have a narrow optimum temperature range and temperatures either side of that optimum range reduces seed yield. The main stem yield indicates that Argentine canola may have a cooler optimum temperature range than B. rapa or B. juncea. Upon exposure to one week of high temperatures, all plants were returned to the control (20/15 /C) growth chambers where all three Brassica species recommenced flowering. However, B. napus subjected to 35/15 /C @ Early Flower produced abnormal pods, which were plump and short (Fig. 1). Previous results attributed this to failure of fertilization at high temperature, leading to parthenocarpy. Source-sink balance plays a significant role in the normal development of reproductive organs in plants. Heat stress alters the source sink balance. Heat shock initiates leaf senescence and remobilisation of photosynthates. When the temperature stress was relieved, most of the stored photosynthate may have surged to the pods (sink). Retarded development of reproductive organs could not properly assimilate the flush of photosynthate, leading to bulged, short pods. CONCLUSIONS All Brassica species were most adversely affected by heat stress of 35/15 /C at the early flowering stage. Optimum temperature for B. napus appears to be lower than optimum temperature for B. juncea and B. rapa. Generally, as temperature increased, the number of pods produced by the plants increased and seed weight decreased. Across genotypes, the number of seeds produced by the main stem (as well as the number of fertile pods) was dependent upon temperature and developmental stage. Dry matter yield was relatively unaffected by temperature. Harvest index was reduced by 35/15 o C @ Early Flower because the portion of dry matter used for reproductive development did not contribute to seed yield but rather to sterile pod formation. This observation further supports our hypothesis that high temperature had a direct effect on the formation of reproductive organs.
ACKNOWLEDGMENTS The authors thank Dean Klassen, Lee Poppy, Darren Steinley, Cal McDonald and Dennis Dyck for assistance in carrying out the experiments. The research was financed by Saskatchewan Wheat Pool, Saskatchewan Canola Development Commission and AAFC - Matching Investment Initiative. REFERENCES Hall, A.E. 1992. Breeding for heat tolerance. Plant breeding review. 10:129-168. Howarth, C. J. 1996. Growth and survival at extreme temperatures: Implications for crop improvements. p. 467-483. In V.L. Chopra, R.B. Singh and A. Verma (ed.) Crop Productivity and Sustainability-Shaping the Future. Proc. Second Int. Crop. Sci. Cong., New Delhi, India. Nov. 17-23, 1996. Morrison, M.J. 1993. Heat stress during reproduction in summer rape. Can. J. Bot. 71:303-308. Paulsen, G.M. 1994. High temperature responses of crop plants. p. 365-389. In K.J. Boote,, J.M. Bennett, T.R. Sinclair and G.M. Paulsen (ed) Physiology and determination of crop yield. ASA, CSSA, SSSA., Madison USA.
Table 1: Mean plant dry matter, seed yield and harvest index for temperature treatments in different Brassica genotypes Temperature J90-4316 Cutlass Quantum Maverick z Mean Shoot Dry Matter (g plant -1 ) 20/15 /C Control 20.77 20.25 23.43 24.26 22.18 28/15 /C @ Early 20.30 20.53 24.46 20.79 21.52 28/15 /C @ Early 19.20 18.85 21.18 23.72 20.74 35/15 /C @ Early 15.34 17.03 21.23 16.58 17.55 35/15 /C @ Early 18.09 18.21 24.26 21.24 20.45 Mean 18.74 18.97 22.91 21.32 LSD (T) 1.55 LSD (G) 1.39 LSD (T X G) NS Seed Yield (g plant -1 ) 20/15 /C Control 7.43 6.24 7.75 3.16 6.19 28/15 /C @ Early 7.39 7.37 7.09 4.51 6.59 28/15 /C @ Early 6.89 6.32 6.32 2.22 5.44 35/15 /C @ Early 3.30 3.76 4.11 0.64 2.94 35/15 /C @ Early 6.05 5.80 6.84 1.62 5.08 Mean 6.21 5.88 6.46 2.43 LSD (T) 0.57 LSD (G) 0.59 LSD (T X G) 1.25 Harvest Index 20/15 /C Control 0.37 0.31 0.34 0.14 0.29 28/15 /C @ Early 0.37 0.36 0.29 0.22 0.31 28/15 /C @ Early 0.36 0.34 0.31 0.10 0.28 35/15 /C @ Early 0.22 0.23 0.19 0.05 0.17 35/15 /C @ Early 0.33 0.32 0.28 0.07 0.25 Mean 0.33 0.31 0.28 0.12 LSD (T) 0.02 LSD (G) 0.02 LSD (T X G) 0.05 z Maverick in the first run and Parkland in the second run.
Table 2. Main shoot seed yield, fertile pods, seed per pod and thousand kernel weight for temperature treatments in different Brassica genotypes in Experiment 2. Temperature Treatments J90-4316 Cutlass Quantum Parkland Mean Seed Yield (g) 20/15 /C Control 0.88 0.52 1.66 0.23 0.85 28/15 /C @ Early Flower 1.87 1.73 1.22 0.56 1.40 28/15 /C @ Early Pod 0.83 0.53 0.85 0.05 0.56 35/15 /C @ Early Flower 0.16 0.10 0.09 0.01 0.09 35/15 /C @ Early Pod 0.96 1.03 0.87 0.21 0.77 Mean 0.98 0.80 0.94 0.21 LSD (T) 0.26 LSD (G) 0.24 LSD (T X G) 0.53 Fertile Pods (main stem) -1 20/15 /C Control 33 16 36 35 31 28/15 /C @ Early Flower 69 51 37 31 48 28/15 /C @ Early Pod 35 23 28 14 25 35/15 /C @ Early Flower 12 13 14 3 11 35/15 /C @ Early Pod 60 40 30 42 43 Mean 43 29 29 26 LSD (T) 8 LSD (G) 7 LSD (T X G) 17 Seeds Pod -1 20/15 /C Control 8.3 9.4 11.1 2.0 7.6 28/15 /C @ Early Flower 9.6 10.4 9.8 7.9 9.5 28/15 /C @ Early Pod 7.5 5.8 7.1 2.1 5.6 35/15 /C @ Early Flower 6.1 4.3 2.8 6.9 4.8 35/15 /C @ Early Pod 6.2 12.5 8.8 5.6 8.2 Mean 7.6 8.4 7.9 4.6 LSD (T) 2.3 LSD (G) 2.1 LSD (T X G) 4.8 Seed Weight (g 1000 seeds -1 ) 20/15 /C Control 3.16 3.57 4.29 2.57 3.39 28/15 /C @ Early Flower 2.92 3.34 3.37 2.33 3.03 28/15 /C @ Early Pod 3.11 3.92 4.45 1.91 3.35 35/15 /C @ Early Flower 2.16 1.54 2.10 1.03 1.72 35/15 /C @ Early Pod 2.46 2.19 2.60 1.16 2.07 Mean 2.79 2.88 3.40 1.81 LSD (T) 0.52 LSD (G) 0.46 LSD (T X G) 1.01
a. b. Figure 1. Abnormal pod formation in Argentine canola (Quantum) in response to 35/15 /C heat stress at early flower. a. Main raceme showing aborted flowers and abnormal pods. Arrows indicate beginning and end of stress period. b. Close up pictures of abnormal pod, seed arrangement in abnormal pod and abnormal seeds (Top to Bottom).