Crop Science Journal of Natural Resources and Life Sciences Education
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Published online 6 February 2007
Published in Crop Sci 47:278-284 (2007)
© 2007 Crop Science Society of America
677 S. Segoe Rd., Madison, WI 53711 USA
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Agricola
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Related Collections
Right arrow Water Stress
Right arrow Other Grain Crops
Right arrow Crop Genetics

PLANT GENETIC RESOURCES

Combining Ability of Waterlogging Tolerance in Barley

M. X. Zhoua,*, H. B. Lib and N. J. Mendhamb

a Tasmanian Inst. of Agricultural Research and School of Agricultural Science, Univ. of Tasmania, P.O. Box 46, Kings Meadows, Tasmania 7250, Australia
b Tasmanian Inst. of Agricultural Research and School of Agricultural Science, Univ. of Tasmania, P.O. Box 252-54, Hobart, Tasmania 7001, Australia

* Corresponding author (mzhou{at}utas.edu.au)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Waterlogging tolerance is one of the major objectives in barley (Hordeum vulgare L.) breeding programs. To make the selection more efficient, an understanding of the genetic behavior of waterlogging tolerance in barley is needed. For this purpose, a 6 by 6 half diallel analysis was conducted in barley from crosses of three waterlogging tolerant Chinese cultivars and three susceptible Australian or Japanese cultivars. The waterlogging treatment was imposed starting from the three-leaf stage. The percentage of yellow leaf was recorded after waterlogging treatment. The diallel analysis was conducted according to Griffing. Three Chinese cultivars showed significantly higher general combining ability (GCA) for waterlogging tolerance while the variance of specific combining ability (SCA) was not significant, indicating that the tolerance was mainly controlled by additive effects. High heritability (h2B = h2N = 0.73) of waterlogging tolerance indicated that selection in early generations could be very efficient. When selections are made in a segregating population, the most effective selection strategy is to discard the plants with severe leaf chlorosis.

Abbreviations: DH, doubled haploids • GCA, general combining ability • h2B, broad-sense heritability • h2N, narrow-sense heritability • SCA, specific combining ability


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
SOIL WATERLOGGING usually influences plant growth in a negative way. The inhibition of N uptake, and the consequent redistribution of N within the shoot, are important contributory factors in the early senescence of leaves and the retarded growth of shoots in flooded plants (Drew and Sisworo, 1977). A decrease in the N concentration in shoots of barley (Hordeum vulgare L.) seedlings can occur rapidly after the onset of flooding and precede leaf chlorosis (Drew and Sisworo, 1977; Wang et al., 1996) and consequently reduces shoot and root growth, dry matter accumulation, and final yield (Kozlowski, 1984; Drew, 1991; Huang et al., 1994a, 1994b; Malik et al., 2002). Roots are also injured by O2 deficiency and metabolic changes during acclimation to low concentrations of O2 (Drew, 1997). Crops differ in their tolerance to excess soil water condition (Tokimasa, 1951). Barley is relatively susceptible to waterlogging as shown by Wang et al. (1996), for example, where the most resistant group of barley cultivars nearly corresponds to a rather susceptible group of wheats (Triticum spp.) (Ikeda et al., 1955). Waterlogging is estimated to reduce yields on average by 20 to 25%, but the loss may exceed 50% depending on the stage of plant development (Setter et al., 1999). Bandyopadhyay and Sen (1992) reported more than 50% loss in yield after 2 d and 80% loss in yield after 3 d of super-saturation treatment after 6 wk normal growth in a coastal saline soil.

Bringing tolerance of waterlogging into barley cultivars is a very important breeding objective in high rainfall areas or where subsoils have low infiltration rates. To fulfil this aim, the most important step is to find waterlogging tolerance genes in barley germplasm. Takeda and Fukuyama (1986) tested 3457 cultivars (preserved at the Barley Germplasm Center, Okayama University) by submerging 50 sterilized grains of each in deionized water in a test tube for 4 d at 25°C and subsequently determining their germination percentage after 4 d on moistened filter paper at 25°C. The germination percentage ranged from 0 to 100. The collections from China, Japan, and Korea contained many tolerant cultivars (average indices 71.6, 66.3, and 60.5, respectively) while those from North Africa, Ethiopia, and Southwest Asia showed few tolerant cultivars (19.6, 13.8, and 13.2, respectively). The most tolerant cultivars retained complete germinability after 8 d soaking at 25°C. Qiu and Ke (1991), after testing 4572 barley cultivars, reported that some germplasm showed a very high level of waterlogging tolerance. Fufa and Assefa (1995) reported some variation among genotypes in their tolerance to waterlogging and suggested locally adapted landraces could be major sources of tolerance. Our previous studies also showed that cultivars showed differential tolerance to waterlogging with some Chinese cultivars showing much better tolerance than the Australian cultivars tested (Pang et al., 2005).

Waterlogging tolerance is likely to be a complex trait which is related to many morphological and physiological traits that are under strong environmental influence. Direct selection on grain yield is less effective since the heritability of the yield after waterlogging was very low (Collaku and Harrison, 2005). Leaf chlorosis after waterlogging is one of the major indices used by researchers in different crops such as wheat (Triticum aestivum L.) (Ikeda et al., 1954; Cao et al., 1995; Cai et al., 1996; Boru et al., 2001), soybean [Glycine max (L.) Merr.] (Reyna et al., 2003), and barley (Hamachi et al., 1990). Understanding the genetic behavior of waterlogging tolerance is also important for breeding cultivars with waterlogging tolerance. Boru et al. (2001) studied the genetic behavior of waterlogging tolerance in wheat using leaf chlorosis as the indicator of the tolerance and found that the tolerance was mainly controlled by additive gene effects. The pregermination flooding tolerance of sorghum (Sorghum vulgare Pers.) seed was found to be controlled mainly by additive genes and the heritability was high in both broad (0.97) and narrow (0.75) sense, indicating that selection for tolerance could be effective in early generations (Thseng and Hou, 1993). Different results were found in wheat (Cao et al., 1995; Fang et al., 1997) and maize (Zea mays L. ssp. mays) (Sachs, 1993), indicating that waterlogging tolerance was controlled by one dominant gene. The broad-sense heritability of waterlogging tolerance of wheat was from as high as 0.71, based on the number of green leaves per main stem (Cai et al., 1996), and 0.74, based on plant grain yield (Bao, 1997), to as low as 0.25, based on grain yield (Collaku and Harrison, 2005). There are few reports on the heritability of and gene effects on barley waterlogging tolerance. The realized heritability for flooding tolerance of barley in four F4–F6 populations ranged from 0.12 to 0.48, based on percentage of dead leaf. Realized heritability estimates for three of the crosses ranged from –0.02 to 1.06 and from –0.12 to 0.32 on the basis of the tolerance index of culm length and grain yield, respectively (Hamachi et al., 1990). In this experiment, six cultivars with different waterlogging tolerance were selected to make crosses in a half diallel pattern to identify general and specific combining abilities for waterlogging tolerance and heritability of the tolerance.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Cultivars (or Crosses) and Waterlogging Treatment
Experiment 1
Six selected cultivars from the Tasmanian Institute of Agricultural Research, TX9425, YYXT, DYSYH (Chinese cultivars), Franklin and Gairdner (both Australian cultivars), and Naso Nijo (Japanese cultivar) were used to make all the possible crosses between them. The three Chinese cultivars had much greater waterlogging tolerance than the other cultivars (Zhou et al., 2003; Pang et al., 2005). The Chinese and Japanese cultivars have earlier maturity than the Australian cultivars. The six parents and the 15 F2 populations were sown in stainless steel tanks (200 by 100 by 85 cm) filled with soil from Cressy Research Station, Tasmania, Australia, where waterlogging occurs regularly, during the 2003–2004 summer at Mt. Pleasant Laboratories in Launceston, Tasmania. Each cultivar or F2 population contained 10 to 12 plants. Two replications (tanks) were used. Starting from the three-leaf stage, all the cultivars and F2s were subjected to waterlogging (keeping the water level just above the soil surface) for 10 d until severe damage occurred in susceptible plants. The percentage of yellow leaf area of each plant was recorded immediately after the termination of waterlogging. The average value of the 10 to 12 plants of each cultivar or F2 population was used in the final analysis.

Experiment 2
Similar waterlogging treatments were conducted in 1.5-L pots (filled with similar soil) for all the parents and F2 populations during the 2005–2006 summer in a glasshouse. Each pot contained 15 plants and three replications were applied. The pots were placed in 40-L bins to obtain similar waterlogging conditions. The waterlogging treatment was the same as that described above, starting from three-leaf stage and ending when severe damage was shown in susceptible cultivars.

Experiment 3
A further experiment compared 350 doubled haploids developed from F1 of the cross of TX9425/Naso Nijo using isolated microspore culture (Davies, 2003), sown in the stainless steel tanks described above with the parents as control, during the 2004–2005 summer. Each line contained five plants. The same waterlogging treatment was applied.

Statistical Analysis
Parental lines and F2s were subjected to an analysis of variance. Combining ability effects were analyzed according to Griffing (1956), method 2 [1/2p(p + 1)] with fixed model. Broad-sense heritability was calculated by dividing genotypic variance by total variance and narrow-sense heritability was calculated by dividing additive genetic variance by total variance. Average values of five plants of each doubled-haploid (DH) line were used to study the distribution pattern of waterlogging tolerance of the DH population. The broad-sense heritability of this population was also calculated.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Difference in Waterlogging Tolerance of Selected Parents
In this experiment, waterlogging caused significant chlorosis of the older leaves of all the cultivars. Cultivars showed significant differential tolerance to waterlogging (Table 1). Three Chinese cultivars, TX9425, DYSYH, and YYXT, showed significantly lower yellow leaf percentage than Franklin, Gairdner, and Naso Nijo. Figure 1 shows the differences between the tolerant cultivar YYXT and susceptible cultivar Franklin after waterlogging treatment. The tolerant cultivar not only had less yellow leaf and healthier plants but also developed a better root system.


View this table:
[in this window]
[in a new window]

  		Table 1. Half diallel data of yellow leaf percentage after waterlogging, including parents and F2s from the indicated crosses.{dagger}

 

Figure 1
View larger version (139K):
[in this window]
[in a new window]

  		Fig. 1. Leaf chlorosis after waterlogging in two barley cultivars (Franklin, left; YYXT, right).

 
Growing conditions also showed very significant effects on yellow leaf percentage after waterlogging treatment. Even though interactions between cultivar and environment were not significant, the relative differences between cultivars were much less in Experiment 2, the pot experiment. The yellow leaf percentages of the tolerant parent cultivars were from 5.0 to 12.7% in Experiment 1 and from 25.2% to 34.2% in Experiment 2 while for susceptible cultivars, the ranges were from 31.3 to 43.9% in Experiment 1 and from 43.3 to 53.0% in Experiment 2. Thus, to make an effective evaluation of waterlogging tolerance, it is important to provide suitable conditions where differences are highlighted.

Combining Ability
Table 1 lists the yellow leaf percentages of parents and their F2 populations after waterlogging treatment from both Experiments 1 and 2. ANOVA showed that even though growing conditions had very great effects, significant differences were found between different parent or F2 populations (P < 0.01). The interactions between cultivar or F2 and growing conditions were relatively small and the overall ranking of the cultivars or F2 populations in yellow leaf percentage changed little between experiments.

The variance of GCA was highly significant (P < 0.01) and that of SCA was not significant, indicating that waterlogging tolerance was mainly controlled by additive effects and that no significant dominance effect or nonallelic interaction could be detected. Of all the cultivars, DYSYH had the lowest negative GCA (–7.5, lowest yellow leaf percentage) (Table 2), and showed greater tolerance than the other two tolerant cultivars, YYXT (–4.1) and TX9425 (–5.4). The other three cultivars showed positive GCA (higher yellow leaf percentage after waterlogging). Franklin and Naso Nijo were the most susceptible parents to waterlogging (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2. General combining ability (GCA) of waterlogging tolerance (yellow leaf percentage) of parents and F2s.{dagger}

 
Significant correlations (R2 = 0.92 for the tank experiment and R2 = 0.75 for the pot experiment) were found between yellow leaf percentage of hybrids (F2s) and that of midparents (Fig. 2 ). The average yellow leaf percentages of all the crosses were similar to the midparent value, confirming that the tolerance was mainly controlled by additive effects. Since no significant dominant effect on waterlogging tolerance was found in this experiment, the estimated broad-sense heritability (h2B) was the same as narrow-sense heritability (h2N). From the variance of GCA, SCA, and experimental error, the estimated heritability (h2B = h2N) was 0.73. The estimation was based on the average value of different populations. When the estimations were based on individual experiments, the broad-sense heritability was 0.85 for Experiment 1 and 0.58 for Experiment 2. A lower value of heritability would be expected if the estimation had been based on single plants. For example, from Experiment 3, the broad-sense heritability was 0.88 based on average value of DH lines and 0.65 when based on individual plants.


Figure 2
View larger version (17K):
[in this window]
[in a new window]

  		Fig. 2. Correlations between average yellow leaf percentages of F2 populations and those of the midparent values.

 
Segregation of DH Populations between Tolerant and Susceptible Cultivars
Figure 3 shows the distribution of waterlogging tolerance of the DH lines from the cross between TX9425 and Naso Nijo. The average yellow leaf percentages were 7.9 for TX9425, 26.7 for Naso Nijo, and 16.0 for the DH population. The tolerance of the DH lines showed continuous distribution, ranging from very tolerant to very susceptible. While there was a good proportion of lines in the tolerant class, and hence scope for further selection, there was no evidence of bimodal distribution and hence of single gene effects.


Figure 3
View larger version (26K):
[in this window]
[in a new window]

  		Fig. 3. Distribution of waterlogging tolerance of the doubled-haploid lines from the cross between TX9425 and Naso Nijo.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Waterlogging inhibits the uptake of N which leads to the decrease of N concentration in shoots of barley seedlings (Drew and Sisworo, 1977). Pang et al. (2005) found that both shoot and root growth were negatively affected by waterlogging. As waterlogging stress developed, chlorophyll content, CO2 assimilation rate, and maximal quantum efficiency of photosystem II (variable fluorescence/maximum fluorescence) decreased significantly, with cultivars showing less yellow leaf percentage having less adverse effects (Pang et al., 2005). Dead leaf percentage under excess soil moisture was thought to be the best criterion for selection for flooding tolerance in early generations because its heritability values are relatively constant and it is easy to measure (Hamachi et al., 1990), and it was correlated with reduction of grain yield and plant and culm length (Hamachi et al., 1989). Oxygen deficiency in the rooting zone occurs under waterlogging conditions. The lack of O2 can severely damage the root (Drew, 1997). Figure 1 shows that the tolerant cultivar not only had less yellow leaf and healthier plants but developed a better root system, which is consistent with our previous report (Pang et al., 2005). The three Chinese cultivars used in this experiment all showed very good waterlogging tolerance with significantly lower yellow leaf percentage than other cultivars. The tolerance may also be partly contributed by the formation of aerenchyma in roots under waterlogging conditions. For example, aerenchyma accounted for 23.9 and 7.1% of the root cross-section area for TX9425 and Naso Nijo, respectively, after 3 wk waterlogging (Pang et al., 2005). Our preliminary yield trials (data not shown) showed that under waterlogging conditions, the yield reductions of Franklin and TX9425 were 86 and 28% in a pot experiment and 61 and 39% in a controlled field experiment.

No significant dominance effects were found in this study. The mean yellow leaf percentages of all the F2s were similar to that of their midparent value. The results were similar to those previously reported in wheat (Boru et al., 2001) and sorghum (Thseng and Hou, 1993), both showing that the waterlogging tolerance was mainly controlled by additive gene effects. The continuous distribution of waterlogging tolerance in our doubled haploid barley population generated from a cross of TX9425 (tolerant) and Naso Nijo (susceptible) indicated that the tolerance was controlled by several genes, which is consistent with the earlier report by Hamachi et al. (1989) but different from results in wheat reported by Cao et al. (1992, 1995), in which a single gene was involved in waterlogging tolerance.

The high heritability and the presence of only additive effects for waterlogging tolerance indicated that selecting in early generations for this trait would be effective. High heritability of waterlogging tolerance was also reported in sorghum (Thseng and Hou, 1993) and wheat (Cai et al., 1996) but in each case, the existence of dominance effects was reported. Our estimation of the heritability was based on the average values of different populations. If the estimation of the heritability is based on single plants, the value could be much lower since greater variation was observed even within a homozygous population (parental cultivar). With the DH population from TX9425/Naso Nijo, the h2B estimated from the average values (0.88) was much higher than that from individual plants (0.65). Even with different experimental conditions the evaluated heritability may differ substantially. In this experiment, the h2B was 0.85 for Experiment 1 but only 0.58 for Experiment 2. Thus it is not surprising that some earlier studies on barley showed very low heritability of waterlogging tolerance (Hamachi et al., 1989, 1990).

There have been no previous reports on the combining ability of waterlogging tolerance of barley. In wheat, Cao et al. (1994) found a significant effect of GCA for the number of green leaves per main shoot after waterlogging treatment at the booting stage. They also found significant SCA effects, indicating the existence of dominance effects. In our experiments, there were significant effects of GCA for yellow leaf percentage after waterlogging treatment, but the SCA effect was not significant, indicating that waterlogging tolerance was mainly controlled by additive gene effects. The selection of parent cultivars should therefore just be based on the parental data of waterlogging tolerance, without considering the SCA. All three Chinese cultivars used in this experiment showed negative GCA (low yellow leaf percentage). Of these three cultivars, DYSYH is a six-rowed barley with relatively poorer agronomic traits and both YYXT and TX9425 are two-rowed and showed better agronomic traits. Thus the latter two cultivars could be more suitable for a breeding program focused on producing two-row malting barley, even though they had slightly higher yellow leaf percentage than DYSYH.

Accurate phenotyping is one of the vital criteria required for the improvement of waterlogging tolerance (Setter and Waters, 2003). In these experiments, the environment used to induce waterlogging showed very significant effects on yellow leaf percentage. The differences between tolerant and susceptible cultivars were much less in the pots than in tanks, with yellow leaf percentage being consistently higher for all cultivars and F2s in the pots. So while the pot treatments apparently imposed greater stress on the plants, the tank treatment gave better discrimination between susceptible and tolerant cultivars. Thus, carefully chosen waterlogging conditions could make the selection much more effective. In the tank (Experiment 1), a very small variation in yellow leaf percentage was found among progeny of crosses between tolerant cultivars. The yellow leaf percentage ranges for individual plants of the three tolerant cultivars in Experiment 1 were from 5 to 20%. In contrast, the range of yellow leaf percentage for susceptible cultivars was much greater among individuals. For example, the yellow leaf in Franklin ranged from 25 to 90%, even though most individuals had 30 to 40%, which may be due to variation in plant development stage when the treatment was imposed, or possibly due to heterogeneity in the cultivar, although all were nominally pure lines. The variation in yellow leaf percentage of different cultivars indicated that tolerant cultivars will normally not have any very susceptible individuals, whereas susceptible cultivars could have a few individuals showing better tolerance, presumably due to environmental effects rather than genetic differences within homozygous lines. Thus, when selecting individuals from an F2 population, plants with severe leaf chlorosis can be discarded since they would almost always be susceptible. There may be a small number of apparently tolerant plants which may not contain tolerant genes. For these tolerant plants, further evaluation in F3 is necessary.

From the results obtained, it would seem likely that the same genes and/or alleles are mainly controlling waterlogging tolerance in the three tolerant cultivars. The F2 populations from the crosses of TX9425/DYSYH and YYXT/DYSYH both had means close to that of the midparent in both experiments (Table 1), and were much lower than any of the susceptible parents. The cross TX9425/YYXT, however, produced an F2 with a mean equal to or greater than the higher parent in both experiments, suggesting that there may be further genes or modifiers involved here.

In conclusion, GCA was very high for waterlogging tolerance while no significant SCA existed. Since heritability was relatively high for waterlogging tolerance, early generation selection could be efficient, especially when selections were based on the average value of the population. Well-controlled waterlogging conditions are crucial for the evaluation of this trait. Development of molecular markers could avoid environmental effects. We have now identified four QTLs on chromosomes 1, 2, 3, and 7 from a DH population of TX9425/Franklin (Li et al., unpublished data, 2006), which will soon be available for use in breeding programs.


   ACKNOWLEDGMENTS
 
This work has been supported by the Grains Research and Development Corporation (Project UT8) of Australia. Thanks also to Western Australian Department of Agriculture for the development of DH population and Dr Rene Vaillancourt for the assistance with the project.

Received for publication February 2, 2006.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 





This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF) Free
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Agricola
Right arrow Articles by Zhou, M. X.
Right arrow Articles by Mendham, N. J.
Related Collections
Right arrow Water Stress
Right arrow Other Grain Crops
Right arrow Crop Genetics

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
The SCI Journals Agronomy Journal Vadose Zone Journal
Journal of Natural Resources
and Life Sciences Education		 Soil Science Society of America Journal
Journal of Plant Registrations Journal of
Environmental Quality		 The Plant Genome