Previous PageTable Of ContentsNext Page

Genotypic variations of δ 13C in rice (Oryza. sativa L. and Oryza. glaberrima Steud.) in relation to transpiration efficiency and biomass production as affected by soil water conditions and N

Motohiko Kondo1, Osamu Ideta1, Edward Barlaan2, Tokio Imbe1, Sumio Itoh3, Paquito P. Pablico4, Darryl V. Aragones4, Ruth Agbisit4 and D. Brar4

1National Institute of Crop Science, Kannondai, Tsukuba, Ibaraki, 305-8518, Japan, Email
Nagasaki Industrial Promotion Foundation
National Agricultural Research Center, Kannondai, Tsukuba, Ibaraki, 305-8518, Japan
International Rice Research Institute, Los Baos, Laguna, Philippines


Genotypic and environmental (soil water regime and N level) variations in δ13C in plant were investigated in relation to the gas exchange, transpiration efficiency (A/T), and biomass production in rice (O. sativa L. and O. glaberrima Steud.) genotypes. In addition, genetic control of δ13C was studied by QTL analysis using RIL derived from IR69093-41-3-2 and IR 72. The results showed that genotype factor was more dominant for variation in δ13C than in total biomass. Genotypic ranking in δ 13C was consistent across environments because of small genotype x environment interactions. Japonica genotypes tended to have higher δ13C than indica genotypes in O.sativa. Genotypes of O.glaberrima showed similar ranges of δ13C to indica genotypes. Higher soil water content significantly decreased δ13C. There was a positive correlation between δ13C and A/T among genotypes within water regimes. Genotypic variation in δ13C was associated mainly with variation in stomatal conductance under all soil water regimes and with photosynthetic capacity in late growth stages under aerobic soil conditions. The correlation between biomass and δ13C was not clear in aerobic soils, whereas it was negative in submerged soils in some cases, which indicated that the significance of lower or higher δ13C for improving biomass productivity may differ under different soil water regimes. QTLs inferring to δ13C were detected mainly on chromosome 1, 4, 8, 9, and 12 although the effects of these QTLs were relatively small, indicating that genetic control of δ13C was relatively complex.

Media Summary

The variation in δ13C in rice (O. sativa and O.glaberrima) genotypes was investigated in relation to biomass production. Genetic control of δ 13C was studied by QTL analysis.


Rice, Oryza sativa L. (Asian rice) and Oryza glaberrima Steud. (African rice), is cultivated under diverse soil water regimes from submerged soil under irrigation to aerobic soil under rainfed upland conditions in the world. Because of the magnifying water scarcity for rice culture, increasing yield under water-limited conditions such as uplands is an important target for rice research. Many studies focused on the improvement of water-capturing capacity through the modification of root system by genetic manipulation and agronomic management. On the other hand, considering the limitation of genetic variation in root capacity for water capture in rice, high transpiration efficiency (biomass/water transpired) is considered to be also a useful trait for breeding program, unless this trait has negative linkage with water uptake capacity and other traits for yield gain. In particular, the association between transpiration efficiency and biomass production should be carefully investigated, when increased transpiration efficiency is the objectives for improved productivity under water-limited conditions. After the finding that 13C/12C isotope discrimination (CID) or stable carbon isotope ratio (δ13C) in plant tissue is correlated with intercellular CO2/ambient CO2 (Ci/Ca) levels, many trials to use δ13C as a selection criterion for high transpiration efficiency have been reported (Farquhar et al. 1982, Farquhar et al. 1989). Genotypic variations for transpiration efficiency and δ 13C in O.sativa (Dingkuhn et al. 1991, Samejima 1985) implied an opportunity to select genotypes with high transpiration efficiency using high δ13C as criterion. On the other hand, it was suggested that selecting low δ13C is useful for improving yield and biomass under favorable growing conditions in other upland crops (Condon et al. 1987). However, there is limited information on the linkage among δ13C, transpiration efficiency, stomatal response, and biomass production under various water regimes and N levels in rice. Furthermore, genetic control of δ13C was not studied intensively.

This study aimed (1) to assess the relative magnitude of environmental and genotypic variation in δ13C among rice genotypes and (2) to analyze the relations among δ13C, gas exchange characteristics, single-leaf transpiration efficiency, and biomass production in O.sativa and O.glaberrima. Secondly, QTL analysis was conducted to clarify genetic control of δ13C.

Materials and methods

Experiment 1: Effect of genotype, soil water and N in O. sativa

The variation in δ13C among eleven rice (O. sativa) genotypes under different soil water regimes (W1: aerobic soil with water stress, W2: aerobic soil without water stress, and W3: submerged soil without water stress) and N levels (0 and 90 kg ha-1) were investigated in the field (Tropudalfs, pH 6.19) at International Rice Research Institute (IRRI), Los Baos in the Philippines in 1998. The relationships among δ13C of aboveground parts at maturity and stomatal conductance (gs), net photosynthetic rate (A), A/T, plant morphology, and biomass production were analyzed. In 2002, different set of thirty genotypes were grown in the field (Fluvaquents, pH 5.2) under submerged condition in Yawara, Ibaraki, Japan with four different N management (no N and 105 kg N ha-1 with (NH4)2SO4, compost + soybean cake, and green manure) for determination of δ13C at panicle initiation and maturity.

Experiment2: Comparison of δ13C between O. sativa and O. glaberrima

18 genotypes of O.glaberrima and 12 genotypes of O. sativa were grown in the pot (Andisol) under submerged and aerobic soil (60 %v/v) conditions for 67 days. The relationships among δ13C of aboveground parts at maturity and gas exchange characteristics, leaf development, and biomass production were analyzed.

Experiment 3: QTL analysis on δ13C

105 lines of recombinant inbred population derived from IR69093-41-3-2, a japonica/indica genotype and IR 72, an indica genotype, were grown in the field in submerged soil condition in Yawara, Ibaraki, Japan from 2001 to 2003. δ13C, C and N contents in the grain and other parts were determined at maturity. QTL analysis on these traits was conducted using QGENE software.


Experiment 1: Effects of soil water, N, and genotype on δ13C (Kondo et al. 2004) 

In the experiments across different soil water regimes and N, genotype accounted for the largest proportion of the variation of δ13C followed by water. In comparison with main effects, interactions were rather small, despite some interactions being significant. Compared with the biomass and yield, the variation in δ13C was more affected by genotype, and to lesser extent, by genotype x water interaction. Genotypic rankings was relatively consistent across plant parts and environments. Japonica genotypes tended to have a higher δ13C than indica genotypes. Higher soil water content decreased δ13C.

δ13C was negatively correlated with gs except at flowering stage in W1, where δ13C was positively correlated with A. Japonica genotypes tended to show higher A/T than indica genotypes in all environments. The results across environments showed that genotypic variation in δ13C had a positive correlation with A/T in most cases.

The relationships between δ13C and biomass were different between aerobic soils (W1 and W2) and submerged soils (W3). There were negative relationships between biomass and δ13C in W3 under submerged conditions. While there was no clear relation or a slightly negative correlation when one genotype was excluded in both W1 and W2 under aerobic soil conditions. There were significant negative correlations between δ13C and specific leaf area in all the treatments.

The results in 2002 under submerged soil conditions in Japan showed consistent tendency that indica genotypes had lowerδ13C than japonica genotypes as in results obtained in the Philippines. The difference between indica and japonica genotypes tended to be slightly larger at maturity than panicle initiation, and at no N than with N. Negative relationship between δ13C and biomass was observed in two of four N treatments.

Experiment 2: Comparison of δ13C between O. sativa and O. glaberrima

Among the genotypes of O. sativa, indica genotypes tended to have lower δ13C than japonica genotypes as observed in Experiment 1 under field conditions. δ13C of O. glaberrima was in the similar ranges to these of indica genotypes. Leaf area was larger in O. glaberrima than O. sativa genotypes on average due to a larger allocation of biomass to leaves and larger specific leaf area. Total aboveground biomass of O. glaberrima was relevant to these of O. sativa genotypes.

Experiment 3:QTL analysis on δ13C

Phenotypic variations in δ13C in the grain and straw were significantly correlated across years. In the grain in 2001, QTLs for δ13 C were detected on chromosome 4, 8, and 12, all of which having positive effect by IR69093-41-3-2. The QTLs on chromosome 4 and 12 were near to these of leaf N content. There were four QTLs detected located in chromosome 1, 9, and 12 in 2002. The most of these QTLs explained less than 20% of total variations. QTLs were detected in the similar region on chromosome 12 in all three years.


Genotypic and environmental variations in δ13C

The consistent difference in δ13C among genotypes across different water, N supplying conditions and locations suggested a strong effect of genetic factor on variation in δ13C. The small genotype x environment interaction for δ13C was also observed in other crops (for example, Condon and Richards 1992, Hall et al. 1990). The tendency for a higher δ13C in japonica genotypes than indica genotypes was consistent with other reports (Dingkuhn et al. 1991, Samejima 1985). δ13C increased with lower soil water conditions probably mainly due to lower gs induced by water stress.

Factors affecting genotypic variation in δ13C

A lower CID may result from either lower gs and/or greater photosynthetic capacity (Farquhar et al. 1989). The results in this study suggested that genotypic variation in δ13C under submerged conditions was mainly related to gs. The higher δ 13C in japonica genotypes was associated with lower gs and higher A/T than indica genotypes, which was consistent with previous studies (Maruyama et al. 1985). In addition, the genotypic variation in δ13C might be also related to the maintenance of photosynthetic capacity in part, under water-limited conditions, e.g. through leaf water status. The fact that japonica genotypes showed a larger increase of δ13C from panicle initiation to maturity than indica genotypes is possibly explained by the larger decrease of gs in japonica genotypes, which might be a constraint for assuring high gain filling and yield.

Although the consistent genotypic difference across N supply conditions was obtained, the effect of N on δ 13C slightly varied among soil water regimes and genotypes. The reason for larger difference between japonica and indica genotypes with N than with no N under submerged conditions should be further studied in terms of stomatal response to varying leaf N content and other related aspects.

Comparison between O. sativa and O.glaberrima

O. glaberrima genotypes had similar δ13C to indica genotypes and not far different from the variations of O. sativa although only limited genotypes were tested. O. glaberrima genotypes tended to have larger capacity of leaf development at early growth stages, which was advantageous for weed competition. The results indicated that the faster leaf area development of O. glaberrima than O. sativa was not directly associated with the large biomass production nor the difference in δ13C.

Possible value of δ13C for improving rice production under diverse water supplies

Variable relationships between δ13C and biomass suggested that the impact of lower and higher δ13C and transpiration efficiency on biomass productivity may differ across water supply conditions. Consistent genotypic difference in δ13C across environments and its association with A/T implied a possible use of δ13C for improving water use efficiency. The significance of increased transpiration efficiency for improving productivity should be evaluated under different target stress conditions, considering the limits to the genetic diversity of root traits for water capture and the genetic linkage between shoot and root traits.

On the other hand, negative correlations between δ13C and biomass production and yield under favorable water supplying conditions leads to the strategy to select for higher δ13C as a criterion for yield improvement (Kondo et al. 1994) as suggested for wheat (Condon et al. 1987) and barley (Craufurd et al. 1991). It is possible that the association between higher δ13C and lower biomass under favorable water supply was partly because of limited stomatal conductance. Conversely, higher stomatal conductance, especially at late growth stages, will lead to higher biomass and yield by supporting a larger photosynthetic rate (Kuroda and Kumura 1990), mainly in indica genotypes under submerged soil conditions.

The results implied that manipulation of δ13C could be beneficial for improving productivity both under water-limited and irrigated conditions. On the other hand, the results on QTLs implied that genetic control of δ13C was relatively complex. The negative association between δ13C and harvest index and specific leaf area were observed here. The results on QTLs suggested the genetic association between δ13C and N status in leaf. It would be useful to further study the genetic linkage and physiological association between δ13C and these plant traits which are related to biomass production to clarify the value of δ13C.


Condon AG and Richards RA (1992) Broad-sense heritability and genotype x environment interaction for carbon isotope discrimination in field-grown wheat. Aust. J. Agric. Res. 43:921-934.

Condon AG, Richards RA and Farquhar GD (1987) Carbon isotope discrimination is positively correlated with grain yield and dry matter production in field-grown wheat. Crop Sci. 27:996-1001.

Craufurd PQ, Austin RB, Acevedo E, and Hall MA (1991) Carbon isotope discrimination and grain-yield in barley. Field Crops Res. 27:301-313.

Dingkuhn M, Farquhar GD, De Datta SK and O’Toole JC (1991) Discrimination of 13C among upland rices having different water use efficiency. Aust. J. Agric. Res. 42:1123-1131.

Farquhar GD, Ehleringer JR, and Hubick K (1989) Carbon isotope discrimination and photosynthesis. Ann. Rev. Crops Res. 40:67-86.

Farquhar GD, Oleary MH, and Berry JA (1982) On the relationship between carbon isotope discrimination and the intercellular carbon dioxide concentration in leaves. Aust. J. Plant Physiol. 9:121-202.

Hall AE, Mutters RG, Hubick KT, and Farquahr GD (1990) Genotypic differences in carbon isotope discrimination by cowpea under wet and dry field conditions. Crop Sci. 30:300-305.

Kondo M, Pablico PP, Aragones DV, and Agbisit R (2004) Genotypic variations in carbon isotope discrimination, transpiration efficiency, and biomass production in rice as affected by soil water conditions and N. Plant and soil (in press).

Kuroda E and Kumura A (1990) Difference in Single Leaf Photosynthesis between old and New Rice; I. Single-leaf photosynthesis and its dependence on stomatal conductance. Jpn. J. Crop Sci. 59:83-292.

Maruyama S, Kabaki N, and Tajima K 1985 Water consumption in japonica and indica rice varieties. Jpn. J. Crop Sci. 54:32-38.

Samejima M (1985) Intraspecific variations of 13C discrimination in Oryza sativa L. Bull. Natl. Inst. Agrobiol. Resource 1:63-84. (Japanese with English summary)

Previous PageTop Of PageNext Page