An Analysis to Visualize Influence of Global Warming on Wheat Grain Quality with High Molecular Weight Glutenin Subunits at Glu D1 Locus  

Devinder Mohan , Raj Kumar Gupta
Quality Section, Directorate of Wheat Research, Karnal 132001, India
Author    Correspondence author
Molecular Plant Breeding, 2014, Vol. 5, No. 7   doi: 10.5376/mpb.2014.05.0007
Received: 30 Apr., 2014    Accepted: 12 May, 2014    Published: 10 Jun., 2014
© 2014 BioPublisher Publishing Platform
This is an open access article published under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Preferred citation for this article:

Mohan and Gupta, An Analysis to Visualize Influence of Global Warming on Wheat Grain Quality with High Molecular Weight Glutenin Subunits at Glu D1 Locus, Molecular Plant Breeding, 2014, Vol.5, No. 7 36-42 (doi: 10.5376/mpb.2014.05.0007)

Abstract

Frequency of high yielding bread wheat genotypes with 5+10 and 2+12 banding patterns at Glu D1 locus was observed in 558 bread wheat entries evaluated during 2003-2013 in national wheat programme of India under different agro-climatic conditions. Frequency of 2+12 type wheats was much higher in the stressed environments suggesting preponderance of this category wheat under global warming. It could lower strength and quality of gluten but wet gluten per unit protein might improve, consequently, bread quality might decline. Chapati and biscuit qualities, protein content and flour yield might remain unaffected. More variations in end-products, flour recovery, protein and gluten contents and grain hardness are anticipated but variability in sedimentation volume and gluten index might squeeze. Shift in the route to product quality and flour yield is suggested under elevated temperature conditions with need to devise selection criteria for identifying product superior varieties of 2+12 category wheats.    

Keywords
End-product quality; Glu D1 locus; Grain quality components; Indian wheat; HMWGS

Dynamics ofclimate changes has drawn attention of wheat researcher’s world over as it is envisaged to affect yield and quality. Elevated temperatures under its influence can shrink the grain ripening period and hinder the grain development processes resulting in lower protein and higher starch contents (Wrigley, 2006). Decrease in wheat grain and flour protein has been demonstrated in greenhouse, growth chamber and field experiments (Ziska etal., 2004; Kimball, 2010). It is perceived that global climate change will result in weakened dough strength and diminished nutritional and processing quality of flour (Uthayakumaran et al., 2009; Erbs et al., 2010). Information about changes in composition and quality of protein under stressed environments is yet to be worked out. Certain reports from Australia suggest that gluten proteins from heat-stressed tolerant varieties have a higher molecular-weight-distribution than proteins from susceptible wheats (Uthayakumaran et al., 2009). Importance of high molecular weight glutenin subunits (HMWGS) is crucial in bread quality as it represents quality of gluten needed to make good dough (Pena, 2008). Genes controlling HMWGS are located on the Glu-A1, Glu-B1 and Glu-D1 loci, in the long arm of the group one chromosomes and can easily be detected through SDS–PAGE. There is differential quality effect linked to glutenin subunit composition as 1, 2* (Glu A1); 7+8, 17+18 (Glu B1); and 5+10 (Glu D1), generally contribute positively to high dough strength. Relevance of 5+10 in wheat dough quality, especially to bake high quality breads, has been strongly supported in wheat (Payne et al., 1981; Pena, 2008). Since effect of global warming has been envisagedin India also (Ortiz et al., 2008), it is important to examine whether proportion of protein fractions especially 5+10 and 2+12 type’s will remain same under the changed field growth conditions. In this investigation, only Glu D1 has been focussed and data generated in All India Coordinated Wheat and Barley Improvement Project (AICW&BIP) has been examined to understand whether frequency of 5+10 glutenin subunits among high yielders matches with 2+12 types in stressed environments. Study focussed on grain quality differences, inter-relationship between important grain properties and the route to end-product quality in the two categories of wheat. Information generated on these aspects is crucial to mitigate the fallout on end-product quality and the prospects of developing product specific varieties under global environmental changes.

1 Results
Distribution pattern of 5+10 and 2+12 Assortment on the basis glutenin subunits at Glu Di locus revealed that 5+10 dominated in the most congenial wheat growth region of the country i.e. NWPZ (78%) and its frequency dropped down to just 18% in harsh environment of CZ (Figure 1). In NEPZ, where growth conditions are not as favourable as that of NWPZ, proportion of these two types was around fifty-fifty. In the zone where environment was harsh due to cold stress i.e. NHZ, 5+10 proportion dropped to 37%. In PZ also, only 37% material showed 5+10 banding pattern. It clearly demonstrates that when environment becomes harsh, proportion of 2+12 increases whereas under soothing growth conditions, 5+10 types are more successful. If two zones falling in the Indo-Gangetic Plains (ME1 region as per CIMMYT classification) are compared, NEPZ has harsher environment in comparison to NWPZ, and the proportion of 2+12 type wheats in that zone is also larger (51% in comparison to 21% of NWPZ). Similarly within the central-peninsular India, PZ climate is more congenial in comparison to harsh CZ, correspondingly 5+10 types increased from 18 to 37%.


Figure 1 Frequency distribution of Glu D1 locus in different environments

In AICW&BIP, there are two trial series i.e. timely sown and late sown. The late sown materials face more heat stress as temperature is relatively higher especially during grain ripening. When timely sown and late sown entries were sorted zone-wise to observe heat influence within a zone, it was noted that in timely sown condition of NWPZ, 95% material belonged to 5+10 group only (60 entries out of 63) but under late sowing; this proportion dropped to 64%. It shows that even under wheat favouring productive land of the country, proportion of 5+10 among the high yielders decline when crop is cultivated under heat stress conditions. Trend was same in each zone as frequency of 5+10 declined in short duration wheats of late sowing as the reduction was 70 to 23% in NEPZ, 64 to 5% in NHZ, 30 to 11% in CZ and 39 to 35% in PZ. It clearly illustrates that irrespective of agro-climatic conditions, frequency of 2+12 type’s increases when wheat is grown under heat stressed condition.
1.1 Difference in grain quality
Material under study had good number of 5+10 (260) and 2+12 (298) entries. When grain and product quality of two groups was compared, 5+10 had clear edge over 2+12 in bread loaf volume and bread quality score but no difference could be noticed quality of chapati and cookies (Table 1). Sedimentation volume, wet gluten content, gluten index, grain hardness index and Glu 1 score had been rated high among quality determinant of Indian wheats (Mohan and Gupta, 2013a) and the 5+10 group expressed significant advantage over 2+12 for all such attributes. Grain hardness index was reduced in 2+12 wheats because genotypes of grain hardness ≤50 could only be observed in this category. Indian wheats characteristically lack in soft grain texture and the study material had only eight genotypes with grain hardness index ≤50 and they all belonged to 2+12 group. Differences between two wheat groups were insignificant in grain weight, grain protein percentage, dry gluten content and flour extraction rate. Even though confounding effect of regional and crop season variations cannot be separated in this comparison, differences were too glaring (P≤0.001) to ignore and can be rated as characteristic features of two different category wheats.


Table 1 Overall performance of 5+10 and 2+12 entries at all India level

It was noticed that bread quality in 2+12 wheats was lower because genotypes of loaf volume ≤475 cc and bread quality score ≤5 could be noticed only in this group, otherwise genotypes of good loaf volume (≥600 cc) and quality score (≥8.5) were equally available. Genotypes of good chapati score (≥8) could also be noted in both categories. Even though overall mean of biscuit spread factor was matching, genotypes for good cookies (spread factor ≥10) could only be noted in 2+12 wheats as soft grains desired for good biscuit making were available in such materials only. It shows that genotypes of good end-product quality (except cookies) can be noted in both categories of wheat. It was interesting to note that genotypes of sedimentation volume ≥60 could only be noted 5+10 wheats. This group also had an edge in gluten index as entries of very low gluten index (≤40%) were noted only in 2+12 wheats.
1.2 Variations in grain properties
Besides mean performance, coefficient of variation also varied among 5+10 and 2+12 wheats and differences in variance were confirmed by F test. Phenotypic variability in 2+12 wheats was statistically higher in end-products quality, protein and gluten contents, grain hardness and flour recovery. Variability for sedimentation volume, gluten index and GLU 1 score was found better in 5+10 wheats. Under the light of such observations, it is anticipated that global warming might bring more variations in traits like protein/ gluten contents, grain hardness and flour recovery but in dough quality related parameters like sedimentation volume and gluten index, variability might squeeze.
1.3 Alterations in grain quality relationships
It was interesting to note that with no disparity in grain weight, protein and dry gluten; wet gluten levels in 2+12 wheats were significantly better than 5+10. High wet gluten content in 2+12 group can be attributed to better water absorption capacity of such wheats. To investigate this peculiar behaviour, gluten levels were plotted against protein content (Figure 2). There was no peculiar difference in the dry gluten content but pattern changed in wet gluten. There was linear response in 2+12 wheats as wet gluten content continued to increase with enhanced grain protein. However, no increase in wet gluten content was visible in the 5+10 wheats once protein content exceeded 13% level. It means that very high protein levels in 5+10 wheats are no guarantee of enhanced wet gluten levels.


Figure 2 Trend in protein-gluten and protein-bread quality relationship

Study points that variations in wet gluten derived from per unit protein exist in wheat. Significant genotypic differences in wet gluten-protein ratio had been reported by Zimic et al. (2006). In this investigation, mean wet gluten protein ratio of 2.54 in 5+10 wheats was significantly lower than 2.64 observed in 2+12 (P≤0.001). Besides mean difference, the pattern of association was also different in these two categories of wheat (Figure 3). In 5+10, wet gluten per unit protein increased when protein varied from 8-11%, remain static at 2.5 in the genotypes of protein range 11%~12% but decreased sharply when protein exceeded 12%. R2 value of this polynomial order 2 trend was 0.12. Such pattern was not strong in 2+12 group as R2 reduced from 0.12 to 0.03.


Figure 3 Relationship pattern between wet gluten-protein ration and protein content

This peculiar relationship of protein-gluten association could also have altered certain other inter-related grain components like gluten strength (sedimentation volume), gluten quality (gluten index) and such reports are available in literature (Drikvand et al., 2013). To investigate it deeply, the pattern of association among two traits was examined by scatter diagram and polymer trend of order 2 was noted. Only those relationships are presented where R2 value was higher ≥0.10. In wheat, there is hardly any quality parameter which is strongly influenced by just one component (Mohan et al., 2013a); therefore R2 value of ≥0.10 in big populations (≥200 genotypes) cannot be ignored. Gluten strength and gluten index are strongly correlated. For high HMWGS, the relationships with quality have been reported less strong (Weegels et al., 1996).
In this study, relationship between wet gluten and sedimentation volume was not very strong and the R2 value was 0.05 and 0.01 in 5+10 and 2 12 wheats, respectively. However, variation in the pattern could be noticed between gluten strength with gluten index, two strongly correlated grain qualities in wheat (Manthey et al., 2000). Gluten index and sedimentation volume were inter-related in this study as well but difference lied in the magnitude of association (Figure 4). Association between sedimentation volume and gluten index was very strong in 5+10 group (R2: 0.49) whereas in 2+12, this relationship was diluted (R2: 0.27). It was observed that incremental change brought in gluten index through sedimentation volume was sharp but started tapering at the higher levels. In 2+12 wheats, even though rate of increase in gluten index was slow but it stayed consistent at different levels of sedimentation volume. It illustrates that contribution of gluten strength in flour much higher in 5+10 in comparison to 2+12 wheats.


Figure 4 Relationship between gluten strength and gluten index

1.4 Influence on flour recovery
Flour extraction rate in Indian wheats is highly influenced by grain hardness index, protein/ gluten content and sedimentation volume (Mohan et al., 2013c; Mohan and Gupta, 2014). Even though there was hardly any difference in average flour yield of +10 and 2+12 wheats (68.8%) in this investigation, variations were obvious in the response of wet gluten and hardness of the grains (Figure 5). Wet gluten and grain hardness index were found strongly associated with flour yield in 2+12 wheats. In 5+10, contribution of such traits was very limited and there was hardly any trend. In 2+12 also, the response was good only up-to certain threshold value which was 35ml in sedimentation volume and 75 in grain hardness index. Study demonstrated that flour recovery in 2+12 wheats can be effectively enhanced through wet gluten and grain hardness up-to certain limit. Such gains in flour recovery cannot be harnessed in 2+12 wheats when wet gluten content goes very high (≥35%) or the grain becomes extremely hard (≥75).


Figure 5 Association of wet gluten and grain hardness with flour recover

1.5 Relationship with product quality
Traditionally, protein and hardness are considered the two most desirable attributes linked with dough quality. When variations are found in relationship of grain properties, product quality can also be affected. Since wet gluten derived from per unit protein was higher in 2+12 wheats and the increase was also linear (Figure 2), similar impact was also noticed in the bread quality (Figure 6). Response to protein content in bread quality was also linear in 2+12 wheats (R2: 0.145) whereas the relationship was not only weakened in 5+10 (R2: 0.06). It states that very high protein levels in 5+10 wheats are no guarantee of enhanced wet gluten levels and improved bread quality.


Figure 6 Relationship of grain protein content with bread quality

Grain hardness also exhibited peculiar pattern in quality of chapati and cookies. For good chapati, wheats of hard grain texture are preferred (Pena, 2011; Mohan and Gupta, 2013a). Interestingly, such association in chapati could only be observed in 2+12 wheats (Figure 7). Trend was reverse in 5+10 category but in the light of low R2 value, it could not be confirmed. It shows that capacity of absorbing more water in harder grains prevail only in 2+12 wheats which brings more puffing and elevates chapati quality. In biscuit spread factor, undoubtedly grain hardness lowered the biscuit spread factor. Since soft grain wheats could only be noticed in 2+12 wheats, the trend could also be validated only in 2+12 group.


Figure 7 Response of grain hardness in quality of chapati and cookies

2 Discussion and Conclusion
Global warming is a global phenomenon and the study convincingly demonstrates that alterations in protein fractions in wheat, especially at Glu D1 locus, shall be a major concern in grain quality. At places where wheat has un-favouring growth conditions; frequency of 2+12 shall go higher among the high yielders and such changes can be expected all over the world. Since dough quality of 2+12 wheats is lower in comparison to 5+10 wheats (Payne et al., 1987; Horvat et al., 2002; Pena, 2011), erosion in product quality is most likely under changing climate. In counties like India where wheat cultivars are generally low in strength and quality of gluten (Mohan et al., 2013b), effect on dough quality can be even more severe. Even though decline in dough quality of 2+12 wheats enrooted through lower sedimentation volume and gluten index was obvious, reduction in genotypic variability of such traits was also illustrated in this study. However, better wet gluten content per unit protein higher variations for such components in 2+12 wheats do ensure opportunity of picking up good product making genotypes. Since genotypes of superior bread or chapati making were traced in both categories of wheat, it implies that varieties good for product making can be developed in the era of global warming but the route to develop such genotypes may undergo a change. Under such situations, it shall be easier to exploit gluten content in comparison to gluten strength and gluten index.
It is apparent that effect of global warming on wheat quality can assume different dimensions depending upon the genetic materials and the surrounding environment. Reduction in crop duration has been envisaged by Ortiz et al., (2008) but the late sown genotypes which have shorter duration had not shown any major disadvantage in AICW&BIP, except some reduction in grain size and yellow pigments content (Mohan et al., 2011). Instead, reduced grain size in short duration varieties resulted in better grain protein content or even bread quality. There are three major components affecting the dough quality i.e. gluten content, gluten strength and gluten quality; and contribution of these grain properties in end-product quality varying according to cultivars, climate and other growth conditions. Any impact of climate change depends upon the prominent component articulating product quality in a given condition. At places like India where cultivars generally have low gluten strength and gluten index; dough quality is derived mainly through wet gluten content, therefore no major loss in bread quality is observed in wheats grown in warmer areas or the cultivars of shorter duration. Alteration in protein content under high temperature was also not observed in Chile (Lizana and Calderini, 2013). However, the whole dynamics change when quality of protein fractions are taken into consideration.
Investigations revealed that varying weather conditions can usher more variations in end-products, flour recovery, protein/gluten contents and grain hardness. Since a shift in inter-related grain properties had been observed among two categories of wheat, it’s got be reflected in major parameters of industrial quality. Prospects of enhancing flour recovery through grain hardness and gluten content will increase. Opportunities to augment chapati quality through grain hardness might turn handy. Increased variability in grain hardness might also offer more chance of getting soft grains needed for biscuit spread factor. In anticipation of such changes, it shall be immensely useful if selection parameter to develop high end-use quality varieties is revisited to support product quality in 2+12 category wheats.
3 Materials and Methods
3.1 Study material and environment
Pipe-line and released varieties evaluated by AICW&BIP during 2003-13 in five mega zones of the country were taken as study material. Genotypes for each zone were different depending upon growth conditions prevailing in that agro-climatic region. Northern hills zone (NHZ) covers hills and foothills of Himalayas where crop is liable to face cold stress due to low temperature. North-western plains zone (NWPZ) labelled as most productive wheat land of India, has the most soothing wheat growth environment whereas adjoined north-eastern plains zone (NEPZ) has shorter winter and humid climate. Climate in the central India (CZ) is hot and dry and crop often faces soil moisture stress. Peninsular zone (PZ) has similar temperature and soil moisture conditions but humidity is relatively higher. 558 genotypes evaluated during the 11 year period (NHZ: 92, NWPZ: 132, NEPZ: 141, CZ: 98 and PZ: 95) were considered for distribution pattern of 5+10 and 2+12 and important grain quality attributes.
3.2 Observations recorded
Samples received from ­3~5 locations of each zone were analyzed at single laboratory located at the Directorate of Wheat Research, Karnal. Protocol described by Payne et al., (1981) was applied to assort study material for high molecular weight glutenin subunits (HMWGS) and SDS (PAGE) was run for fractionation of proteins. Observations were also recorded on important grain quality attributes (sedimentation volume, grain hardness index, grain protein content at 14% grain moisture, dry and wet gluten contents, gluten index, kernel weight, flour recovery) and quality of the end-products (bread loaf volume, bread quality score, chapati (flat bread)quality score and biscuit spread factor). AACC (2000) method was applied to examine processing and milling quality. Conventional approach was adopted to derive GLU 1 score (Payne et al., 1981) and chapati quality was analysed as per method suggested by Rao et al., (1986). Grain protein recoded on Infra-Tec 1225 instrument was converted to 14% grain moisture. Grain hardness index was measured by single kernel characterization system 4100 and Quadrumat Senior mill was used for recording flour extraction rate.
3.3 Statistical analysis
Data pertaining to different classes is presented as means and Student t–test was applied to compare at 5% level of significance. Difference in variance among genotypes of two groups was examined by F test at P 0.05. Polynomial Order 2 trend was observed in the scattered diagrams to observe relationship between two component traits.
Acknowledgements
The authors express gratitude to the researchers conducting field trials of AICW&BIP and providing samples for quality analysis. This investigation is an outcome of a core research project funded by the Indian council of Agricultural Research for Directorate of Wheat Research, Karnal.
References
AACC, 2000, Approved methods of American Association of Cereal Chemists, American Association of Cereal Chemists, 15th Edn, Arlington, USA
Drikvand R., Bihamta M.R., Najafian G., and Ebrahimi A., 2013, Kernel quality association and path analysis in bread wheat, International J. Biol., (3): 73-79
Erbs M., Manderscheid R., Jansen G., Seddig S., Pacholski A., and Weigel H-J, 2010, Effects of free-air CO2 enrichment and nitrogen supply on grain quality parameters of wheat and barley grown in a crop rotation, Agric., Ecosys. and Envt., 136: 59-68
http://dx.doi.org/10.1016/j.agee.2009.11.009
Horvat D., Jurkovi? Z., Sudar R., Pavlini? D., and Šimi?, G., 2002, The relative amounts of HMW Glutenin Subunits of OS wheat cultivars in relation to bread-making quality, Cereal Res. Communication, 30: 415-422
Kimball B.A., 2010, Lessons from FACE: CO2 Effects and interactions with water, nitrogen, and temperature, In: Hillel D., Rosenzweig C., (eds.), Handbook of Climate Change and Agro-ecosystems: Impacts, Adaptation, and Mitigation, Imperial College Press, London, UK, pp 87-110
http://dx.doi.org/10.1142/9781848166561_0006
Lizana X.C., and Calderini D.F., 2013, Yield and grain quality of wheat in response to increased temperatures at key periods for grain number and grain weight determination: considerations for the climatic change scenarios of Chile, The J. Agric. Sci., 151: 209-221
http://dx.doi.org/10.1017/S0021859612000639
Manthey F.A., Kegode R.B., and Elias E.M., 2000, Relationship between gluten index and SDS micro-sedimentation and mixograph tests, AACC 2000 Annual Meeting, American Association of Cereal Chemists, Inc
Mohan D., Singh S.S., and Gupta R.K., 2011, Vibrancy of the Indian wheat in upholding Yield and Quality under Global Environmental Change, In: Singh S.S., Hanchinal R.R., Singh G., Sharma R.K., Tyagi B.S., Saharan M.S., and Sharma Indu (eds.), Wheat: Productivity Enhancement under Changing Climate, Narosa Publishing House, New Delhi, pp 87-94
Mohan D., and Gupta R.K., 2013a, Analysing grain properties of Indian bread-wheat cultivars for defining route to end-product quality and key attributes for selection, Indian J. Genet., 73(4): 1-8
Mohan D., Gupta R.K., and Verma A., 2013b, Characterization of Indian hallmark wheat germplasm for grain quality and the stable genetic resource, Indian J. Genet., 73(1): 14-22
Mohan D., Gupta R. .K, and Sharma Indu, 2013c, Harnessing flour recovery in the Indian wheats, Indian J. Agric. Sci., 83(7): 781-87
Mohan D., and Gupta R.K. 2014, Exploring crop morphology to supplement augmentation of wheat flour recovery – An analysis, Canadian J. Plant Breed. 2(1): 44-50
Ortiz R., Sayre K.D., Govaerts B., Gupta R., Subbarao G.V., Ban T., Hodson D., Dixon J.M., Ivan Ortiz-Monasterio J. and Reynolds M., 2008, Climate change: Can wheat beat the heat?, Agric., Ecosys. and Envt., 126: 46-58
http://dx.doi.org/10.1016/j.agee.2008.01.019
Payne P.I., Corfield K.G., Holt L.M., and Blackman J.A., 1981, Correlations between the inheritance of certain high-molecular weight subunits of glutenin and bread-making qualities in progenies of six cresses of bread wheat, J. Sci. Food and Agric., 32: 51-60
http://dx.doi.org/10.1002/jsfa.2740320109
Payne P.I., Nigtingale M.A., Krattiger A.F., and Holt L.M., 1987, The relationship between HMW glutenin subunit composition and the bread-making quality of British-grown wheat varieties, J. Sci. Food and Agric., 40: 51-65
http://dx.doi.org/10.1002/jsfa.2740400108
Pena R.J., 2008, Improving or preserving bread making quality while enhancing grain yield in wheat., In: Reynolds M.P., Pietragalla J., and Braun H.J., (eds.), International Symposium on Wheat Yield Potential: Challenges to International Wheat Breeding, Mexico, D.F.: CIMMYT, pp 171-74
Pena R.J., 2011, Breeding bread wheat for diverse food end-use. In: Singh S.S., Hanchinal R.R., Singh G., Sharma R.K., Tyagi B.S., Saharan M.S., and Sharma Indu (eds), Wheat: Productivity Enhancement under Changing Climate, Narosa Publishing House, New Delhi, pp 315-323
Rao P.H., Leelavathi K., and Shurpalekar S.R., 1986, Test baking of chapati –development of a method, Cereal Chem., 63: 297-303
Simic G., Horvat D., Jurkovic Z., drezner G., Novoselovic D., and Dvojkovic K., 2006, The genotype effect on the ratio of wet gluten content to total wheat grain protein, J. Central European Agric., 7(1): 13-18
Uthayakumaran S., Batey I., and Wrigley C., 2009, Implications of global climate change for wheat-grain quality, Online, AACC International Cereal Science Knowledge Database
Weegels P.L., Hamer R.J., and Schofield I.D., 1996, Functional properties of wheat glutenin, J. Cereal Sci., 23: 1-18
http://dx.doi.org/10.1006/jcrs.1996.0001
Wrigley C., 2006, Global warming and wheat quality, Cereal Foods World, 51: 34-36
Wrigley C.W. ,2009, Wheat: A Unique Grain. In: Khan A., and Shewry P.R. (eds.), Wheat Chemistry and Technology, 4th Edn, AACC International, Inc., St. Paul, MN. pp 3-19
http://dx.doi.org/10.1094/9781891127557.001

Ziska L.H., Morris C.F., and Goins E.W., 2004, Quantitative and qualitative evaluation of selected wheat varieties released since 1903 to increasing atmospheric carbon dioxide: Can yield sensitivity to carbon dioxide be a factor in wheat performance?, Global Change Biology, 10: 1810-1819
http://dx.doi.org/10.1111/j.1365-2486.2004.00840.x