Crop Science Grow Your Career with CSSA
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Agricola
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Related Collections
Right arrow Crop Genetics
Right arrow Other Grain Crops
Crop Science 41:1021-1026 (2001)
© 2001 Crop Science Society of America

CROP BREEDING, GENETICS & CYTOLOGY

Genetic Analysis of a Hulless x Covered Cross of Barley Using Doubled-Haploid Lines

Thin-Meiw Choo*,a, Keh Ming Hoa and Richard A. Martinb

a Eastern Cereal and Oilseed Research Centre, Agriculture and Agri-Food Canada, Ottawa, Ontario K1A 0C6, Canada
b Crops and Livestock Research Centre, Agriculture and Agri-Food Canada, P.O. Box 1210, Charlottetown, Prince Edward Island, C1A 7M8, Canada

* Corresponding author (ChooTM{at}em.agr.ca)


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
The effects of the hulless (nud) and rough-awned (Raw1) genes are not fully understood in hulless barley (Hordeum vulgare L.). A study was initiated to (i) determine the potential of hulless lines in a hulless x covered cross, (ii) detect additive x additive epistasis and estimate genetic correlations, and (iii) determine the effects of hulless and rough-awned genes on 11 agronomic traits of barley. Fifty covered lines and 48 hulless lines derived from a ‘Kunlun no. 1’ x ‘CIMMYT no. 6’ cross were evaluated for grain yield, test weight, seed weight, height, heading date, and maturity at two locations in Eastern Canada (Charlottetown and Ottawa) in 1998. Plant density, smut resistance, and scald resistance were also recorded at Charlottetown, while spike density was estimated at Ottawa. The 48 hulless lines contained 82 to 100% hulless kernels. At least one hulless line yielded similar to the highest yielding line if it was adjusted by the weight loss of the hull. This suggests that it is possible to breed for high-yielding hulless barley cultivars. Additive x additive epistasis was detected for some of the traits. Yield was significantly correlated with test weight, seed weight, and height. In Eastern Canada, hullessness was associated with 17% lower plant density, 11 to 18% shorter plant height, 15 to 19% lower seed weight, 20 to 21% higher test weight, and 21 to 36% yield reduction. Hullessness, however, was not associated with heading date, maturity, smut resistance, scald resistance, and spike density. Since hulless progeny could have lower emergence rates and shorter plant heights, hulless barley breeding programs should avoid propagating segregating materials from hulless x covered crosses in bulk populations, as many hulless plants may be eliminated by competition. Rough-awned hulless barley had more hulless kernels than smooth-awned. Therefore, selection for rough-awned plants could improve the threshability of hulless barley.

Abbreviations: DH, doubled haploid • QTL, quantitative trait loci


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 
HULLESS BARLEY contains high protein and high digestible energy, and is increasingly popular for use in swine diets in Canada (Bhatty, 1999). Hullessness is controlled by the recessive allele nud on chromosome 7H. Like the two-rowed/six-rowed (Vrs1/vrs1) gene, the nud gene also has remarkable effects on many other traits of barley. The effect of the nud gene, however, has not been fully investigated. More studies are needed to better understand its effects on other traits and to develop efficient selection strategies for hulless barley breeding programs. Hulless barley was reported to yield less than covered barley (Harlan et al., 1940; McGuire and Hockett, 1981). Takahashi et al. (1961), on the other hand, found a hulless mutant which yielded as well as its covered parent. They also found that a change in a single gene to hulless kernels by radiation was consistently associated with shorter stems and longer rachis internode. In an isogenic study, McGuire and Hockett (1981) noted that the hulless trait was associated with many malting quality traits. They also detected interactions between the nud and Lks2/lk2 (long-awned/short-awned) loci for some malting quality traits. In contrast, other workers found no differences between the two types of barley for diastatic activity (Day et al., 1955) and N percentage (Day and Dickson, 1957).

Awn barbing is controlled by the dominant allele Raw1 on chromosome 5H. Previously, Harlan et al. (1940) and Middleton and Chapman (1941) observed an association between rough awns and high yield. On the other hand, Everson and Schaller (1955) found an association between smooth awns and high yield. Suneson and Ramage (1962) reported that smooth-awned types tended to produce heavier but fewer kernels. Jui et al. (1997) noticed that rough awns were associated with high yield, high test weight, low seed weight, short plant height, poor lodging resistance, and earliness. Awn barbing could facilitate the separation of the hull from caryopsis in hulless barley because it creates more opportunities for friction during the threshing process.

Doubled haploids (DHs) are suited for determining relationships between marker genes and quantitative traits (Choo, 1983) and to study the inheritance of quantitative traits (Choo, 1981). Therefore, a study was initiated to use DH lines to (i) determine the potential of hulless lines in a hulless x covered cross, (ii) detect additive x additive epistasis (homozygote x homozygote interactions) and estimate genetic correlations, and (iii) determine the effects of the hulless and rough-awned genes on 11 agronomic traits of barley.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS AND DISCUSSION
 REFERENCES
 
Two barley cultivars, Kunlun no. 1 and CIMMYT no. 6, were used for this study. Kunlun no. 1 is a hulless, rough-awned barley developed at the Qinghai Academy of Agricultural Science, China (Feng, 1986), while CIMMYT no. 6 is a covered, smooth-awned accession from the International Maize and Wheat Improvement Center, Mexico. Approximately 400 DHs were derived from F1 hybrids of the Kunlun no. 1 x CIMMYT no. 6 cross by the bulbosum method (Kasha and Kao, 1970). Seeds of these DHs were first increased in the greenhouse to form DH lines and then these DH lines along with the two parents were further multiplied for seeds in the field at Ottawa (Ontario).

Fifty covered lines and 48 hulless lines were taken randomly from those with a sufficient quantity of seed for testing at two locations. These 98 DH lines and the two parents were seeded on 1 May 1998 at Charlottetown (Prince Edward Island) and 7 May 1998 at Ottawa, in a randomized complete block design with three replicates at each location. The two parents were both entered five times in each replicate. The seeding rate for Charlottetown and Ottawa was 350 and 300 seeds m-2, respectively. Each experimental plot at Charlottetown consisted of five 4-m rows with a row spacing of 15.6 cm. Each plot at Ottawa consisted of four 2.25-m rows with a row spacing of 23 cm. All rows were harvested at both locations for yield determination. Standard cultural practices were followed at each location. Grain yield, test weight, seed weight, plant height, heading date, and maturity date were recorded for each plot at both locations. Heading date was recorded when approximately 50% of the shoots had a fully emerged head, while maturity date was recorded when approximately 50% of the heads lost all green color. Plant density (i.e., number of plants m-2) and spike density (i.e., number of spikes m-2) were estimated at Charlottetown and Ottawa, respectively. Smut resistance was determined at Charlottetown by counting the number of spikes infected by loose smut (caused by Ustilago nigra Tapke). Resistance to scald, caused by Rhynchosporium secalis (Oud.) J. J. Davis, was rated at Charlottetown on a scale of 1 to 9 with 1 being resistant and 9 being highly susceptible. Some hulless lines contained a small portion of the caryopses that were still enclosed by a non-adhering hull (i.e., lemma and palea). Therefore, percent hullessness was also determined among the 48 hulless lines at both locations.

Four SAS procedures, UNIVARIATE, GLM, CORR, and VARCOMP (SAS Institute Inc., 1997), were used to analyze the data from the two locations. To detect additive x additive epistasis, the mean of the 98 DH was compared with the parental mean by an F-test (Choo et al., 1986), and the frequency distribution of the 98 DH lines was tested for normality by the W-test (Shapiro and Wilk, 1965; Choo and Reinbergs, 1982). Any DH line showing a mean value outside the parental range was considered to be a transgressive line. A superior line was defined as a transgressive line with higher grain yield, higher plant or spike density, or better scald resistance than CIMMYT no. 6, or as greater test weight or seed weight, taller, later heading date or maturity, or better smut resistance than the other parent Kunlun no. 1, on the basis of a t-test. An inferior line was defined as a transgressive line with lower grain yield, lower plant or spike density, or less scald resistance than Kunlun no. 1; or as lighter test weight or seed weight, shorter, earlier heading date or maturity, or less smut resistance than the other parent CIMMYT no. 6, on the basis of a t-test. Variance components were estimated by equating the expected mean square to its appropriate mean square.

Phenotypic correlation was used to detect genetic linkages and/or pleiotropy (Frégeau-Reid et al., 1996; Jui et al., 1997). Assuming characters X and Y with heritabilities h2X and h2Y, respectively, Falconer (1960) in his Equation 19.1 showed that the phenotypic correlation (rP) between X and Y is a linear combination of genetic correlation (rA) and environmental correlation (rE). Mathematically, rP = hXhyrA + eXeYrE, where e2X = 1 - h2X and e2Y = 1 - h2Y. When an identical set of DH lines is grown at locations as diverse as Charlottetown (C) and Ottawa (O), then the environmental correlation for X and Y becomes negligible or zero (i.e., rE = 0). Consequently, the resulting phenotypic correlation provides an estimate of genetic correlation. Using this method, two phenotypic correlation coefficients between X and Y can be obtained: one is between X at C and Y at O and the other is between X at O and Y at C. The two phenotypic correlation coefficients may vary depending on the heritability values. The two correlation coefficients were compared by the z-test (Snedecor and Cochran, 1967) to determine if they were homogeneous. The two correlation coefficients were expected to be homogeneous in the absence of line x location interactions.

The effects of the two marker genes (i.e., hulless vs. covered and rough- vs. smooth-awned) on the agronomic traits were studied by comparing the two marker classes for each morphological trait (Choo, 1983). The between-classes mean square was tested against the within-classes mean square by an F-test. Differences between marker classes would suggest the presence of either linkage between the marker locus and quantitative trait loci (QTL), or a pleiotropic effect of the marker gene. The magnitude of the marker-gene effect was determined by the ratio of the between-classes variance and the total genotypic variance (sum of between-classes and within-classes variance components).


   RESULTS AND DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
REFERENCES
 
Potential of Lines
Kunlun no. 1 differed from CIMMYT no. 6 for all traits tested (Table 1). Kunlun no. 1 yielded 14 and 34% less than CIMMYT no. 6 at Ottawa and Charlottetown, respectively. At both locations, Kunlun no. 1 was taller, later in heading date and maturity, greater in test and seed weights, and produced fewer plants m-2 at Charlottetown or spikes m-2 at Ottawa. Kunlun no. 1 was resistant to smut, but susceptible to scald at Charlottetown.


View this table:
[in this window]
[in a new window]
  		Table 1. Agronomic traits for 98 doubled-haploid (DH) lines derived from a Kunlun no. 1 x CIMMYT no. 6 cross of barley evaluated at two eastern Canada locations in 1998.

 
Across the two locations, none of the 98 DH lines had higher yield than CIMMYT no. 6. One hulless line (H462) did not significantly differ from CIMMYT no. 6 (3.66 t ha-1) for yield, and it yielded 24% higher than Kunlun no. 1 (3.35 vs. 2.71 t ha-1). Two hulless lines (H115 and H412) had an average test weight of 79 kg hL-1, which was significantly higher than that of Kunlun no. 1 (76 kg hL-1). One hulless line (H849) had greater seed weight than Kunlun no. 1 (40 vs. 36 mg). Eighteen covered lines were taller than Kunlun no. 1, but only three hulless lines were taller. Twelve lines headed and matured earlier than Kunlun no. 1. Seven hulless lines had better emergence rate than the hulless parent Kunlun no. 1 and 13 hulless lines had more spikes than Kunlun no. 1. Two of the hulless lines (H385 and H462) had a high plant density (297–346 plants m-2) as well as a high spike density (333–355 spikes m-2). Five lines were more susceptible to smut than CIMMYT no. 6. The two lines most severely infected by loose smut were covered barleys. The covered line H656 produced 58 infected spikes out of 312 plants m-2, and the covered line H262 produced 51 infected spikes out of 276 plants m-2. Forty-two lines showed a scald rating outside the parental range.

The above results indicated this hulless x covered cross generated a tremendous amount of variation for many traits. Hulless DH lines with a good emergence rate, good tillering ability, high test weight, high seed weight, or good disease resistance can be selected from this cross for further testing and for use as parents for further crossing. In fact, one of the hulless lines (H462) has achieved that status, with 100% emergence rate, 333 spikes m-2, and a yield level of 3.35 t ha-1. The yield level of H462 is comparable to that of the highest yielding line in this study (4.05 t ha-1) after adjusting for hull weight of H462. Therefore, it is still possible to breed for high-yielding hulless barley cultivars.

Detection of Additive x Additive Epistasis
The average of the 98 DH lines did not differ from the mid-parental value for test weight, seed weight, height, and heading date at both locations; for maturity, plant density, smut resistance, and scald resistance at Charlottetown; and for yield at Ottawa (Table 1). They, however, yielded 10% less than the mid-parent value at Charlottetown; and they matured 3 d later and produced fewer spikes than the mid-parent value at Ottawa. Variation among the 98 DH lines was significant (P < 0.01) for all of the traits studied. These 98 DH lines had a normal distribution for yield, seed weight, height, maturity, and plant or spike density at both locations (W >= 0.97, P >= 0.10). The frequency distribution, however, was bimodal for test weight (W <= 0.94, P <= 0.002) and non-normal for heading date (W <= 0.95, P <= 0.005) at both locations. The frequency distribution was non-normal also for smut (W = 0.36, P = 0.001), scald (W = 0.90, P = 0.001), and percent hullessness (W <= 0.62, P <= 0.001). The means-comparison method should be more accurate than the non-normality method for detecting additive x additive epistasis, because the former theoretically has a smaller standard error (Kendall and Stuart, 1967). Therefore, these results suggest the presence of additive x additive epistasis for yield, maturity, and spike density, and perhaps, for heading date. The frequency distribution for test weight was non-normal also, but this was not due to the presence of additive x additive epistasis. Instead, it was due to the segregation of the nud gene.

Genetic Correlations
Yield was significantly correlated with test weight, seed weight, and height for the 98 DH lines; however, it was not correlated with test weight and seed weight for each of the two classes (Table 2). Yield was not correlated with smut resistance or scald resistance. Among the 48 hulless lines, yield was positively correlated with spike density. Two correlation coefficients were obtained for each pair of traits. For example, one correlation coefficient (r = -0.52) was obtained for all 98 lines between yield at Charlottetown and test weight at Ottawa, another (r = -0.48) was obtained between yield at Ottawa and test weight at Chalottetown. A majority of the correlation coefficients (i.e., 16 of 21 for all 98 lines, 17 of 21 for the covered class, and 19 of 21 for the hulless class) did not differ from their counterparts based on the z-test. Genetic correlations detected in this study were most likely due to linkages.


View this table:
[in this window]
[in a new window]
  		Table 2. Correlation coefficients among 98 doubled-haploid lines derived from a Kunlun no. 1 x CIMMYT no. 6 cross of barley for agronomic performance at Ottawa vs. Charlottetown.

 
Effects of Marker Genes
The nud Gene
On average, the hulless class yielded less than the covered class by 21% at Charlottetown and by 36% at Ottawa (Table 3). The nud gene contributed 47 to 57% of the total variation for yield. Though the hulless parent had 9 to 28% greater seed weight than the covered parent, their hulless progeny on average produced 15 to 19% lighter seed than their covered progeny. Most of the weight difference between these two classes of barley can be attributed to the weight loss of the hull in hulless barley. The nud gene seems to have no pleiotropic effect on seed weight because seed weight of some hulless lines was relatively high. In fact, the highest seed weight in the hulless class was only 7% lower than that in the covered class (40 vs. 43 mg). It is possible that the nud locus is linked with QTL for low seed weight, but this remains to be shown. If the yield of hulless barley is adjusted for the weight loss of the hull by adding 15 to 19% to its yield, the hulless class still yielded less than the covered class. Therefore, the low yield of hulless barley is not solely due to the weight loss.


View this table:
[in this window]
[in a new window]
  		Table 3. Means of four morphological classes for 98 doubled-haploid lines derived from a Kunlun no. 1 x CIMMYT no. 6 cross of barley (21 covered and rough: 29 covered and smooth: 25 hulless and rough: 23 hulless and smooth).§

 
The hulless class had 20 to 21% greater test weight than the covered class (Table 3). The two distributions of the two classes of barley for test weight did not overlap at Charlottetown except for one hulless outlier (data not shown). The nud locus contributed 85 to 95% of the total variation for test weight. The effect of the nud gene on test weight most likely is a reflection of the presence or absence of the hull. The test weight of some hulless lines was excellent, up to 79 kg hL-1.

The hulless parent was taller than the covered parent, but in contrast, the height of their hulless progeny was only 84 to 90% of that of the covered progeny (Table 3). The association was due to the pleiotropic effect of the nud gene, because the tallest hulless line in this study was still 9 cm shorter than the tallest covered line. Hulless mutants derived from radiation consistently produced shorter plant height than covered parents (Takahashi et al., 1961). Unlike hulless barley, covered barley produces a cementing substance from its pericarp epidemis that causes the hull-caryopsis adherence (Gaines et al., 1985). The chemical composition of this cementing substance has not been identified. Whether or not this cementing substance or its precursor is required for further stem elongation and plant growth is not known.

Hulless lines were not only shorter than covered lines, they also had lower emergence rate (17% fewer plants m-2) (Table 3). Therefore, hulless barley breeding programs should avoid propagating segregating materials from hulless x covered crosses in bulk populations because many hulless plants may be eliminated by competition. The importance of height as a factor in competition was recognized by Clements et al. (1929). They noted that "The plants may be so nearly the same height that the difference is only a millimeter, yet this may be decisive since one leaf overlaps the other." One way to avoid the competition is to separate the F2 seeds from hulless x covered crosses into hulless and covered groups and to propagate only the hulless F2 seeds. Selection would be greatly improved if one could develop a machine that could quickly distinguish and separate the two types of barley kernels from the segregating materials.

Hulless barley was associated with poor emergence. Subsequent tillering might or might not compensate for the reduction in emergence rate for some hulless lines, which could lead to lower yield for hulless barley. The nud gene might also have a pleiotropic effect on yield, or might be linked to QTL for yield.

Some covered lines were more susceptible to smut than the susceptible parent (Table 3). Many DH lines had a scald rating outside the parental range, indicating two or more genes are involved with smut or scald resistance.

The Raw1 Gene
The rough-awned class did not differ from the smooth-awned class for all of the traits measured except percent hullessness (Table 3). The percent hullessness among the 48 hulless lines ranged from 82 to 100%. The rough-awned hulless class had more hulless kernels than the smooth-awned class at both locations. Previously, other workers reported an association between awn type and other traits (Harlan et al., 1940; Middleton and Chapman, 1941; Everson and Schaller, 1955; Suneson and Ramage, 1962; Jui et al., 1997). This study, however, failed to find any association, except with percent hullessness. Perhaps, QTL affecting these traits in the chromosomal region surrounding the Raw1 locus did not differ in this cross.

Besides high yield, complete hullessness is another important objective for many hulless barley breeding programs. Currently, a minimum of 85 and 97% hullessness are required to be eligible for Standard and Select Hulless Barley Grades, respectively, in Canada. Therefore, ways to improve the threshability of hulless barley would be very useful for hulless barley production and utilization. The results of this study showed that rough-awned hulless barley contained more hulless kernels than smooth-awned at both locations. Therefore, selection for rough-awned plants should increase the threshability of hulless barley. In this study, percent hullessness was found to be correlated negatively with yield (r = -0.38, n = 48, P < 0.01) and positively with test weight (r = 0.57, n = 48, P < 0.01) at Charlottetown. Therefore, selection for high test weight may also help improve the threshability of hulless barley. More studies are needed to find ways to improve the threshability of hulless barley. The negative correlation between percent hullessness and yield could be the result of the weight loss of the hull of the completely hulless lines. None of the correlation coefficients between percent hullessness and other traits were significant at Ottawa (data not shown).

In conclusion, hullessness was associated with poor emergence, short plant height, high test weight, low seed weight, and low grain yield. Hullessness, however, was not associated with heading date, maturity, spike density, smut resistance, and scald resistance. Rough-awned barley had better threshability than smooth-awned barley. In this study at least one hulless line yielded similar to the highest yielding line if it was adjusted by the weight loss of the hull. This suggests that it is possible to breed for high-yielding hulless barley through recombination and selection.


   ACKNOWLEDGMENTS
 
The authors were grateful to Mo Kuc for producing the DH lines, to Sharon ter Beek and Dan Murphy for their technical assistance in conducting the field trials, and to Linda Langille for technical assistance in statistical analysis and data collection.

Received for publication December 31, 1999.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS AND DISCUSSION
 REFERENCES
 


Related articles in Crop Science:

This issue in Crop science

Crop Science 2001 41: 991. [Full Text]  



This article has been cited by other articles:


Home page
 ANN BOT (LOND)Home page
M. Pourkheirandish and T. Komatsuda
The Importance of Barley Genetics and Domestication in a Global Perspective
Ann. Bot., October 1, 2007; 100(5): 999 - 1008.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
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 Related articles in Crop Science
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
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 HighWire
Right arrow Citing Articles via ISI Web of Science (4)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Right arrow Search for Related Content
PubMed
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Agricola
Right arrow Articles by Choo, T.-M.
Right arrow Articles by Martin, R. A.
Related Collections
Right arrow Crop Genetics
Right arrow Other Grain Crops

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