Crop Science 41:1085-1091 (2001)
© 2001 Crop Science Society of America
CROP BREEDING, GENETICS & CYTOLOGY
Selection for Greater ß-Glucan Content in Oat Grain
C. T. Cervantes-Martineza,
K. J. Freyb,
P. J. Whitec,
D. M. Wesenbergd and
J. B. Holland*,e
a Universidad Autonoma Chapingo, Carretera Mexico-Texcoco km 38.5, Texcoco, Edo. de Mex., Mexico 56230
b Dep. of Agronomy, Iowa State Univ., Ames, IA 50011
c Dep. of Food Science and Human Nutrition, Iowa State Univ., Ames, IA 50011
d USDA-ARS, National Small Grains Germplasm Research Facility, P.O. Box 307, Aberdeen, ID 83210
e USDA-ARS Plant Science Research Unit, Dep. of Crop Science, Box 7620, North Carolina State Univ., Raleigh, NC 27695-7620
* Corresponding author (james_holland{at}ncsu.edu)
 |
ABSTRACT
|
|---|
Oat (Avena sativa L.) ß-glucan lowers serum cholesterol in humans. Development of oat cultivars with greater groat (caryopsis) ß-glucan content would increase the nutritional and economic value of the crop. The objectives of this experiment were to evaluate the response to phenotypic selection among individual S0 plants for greater groat ß-glucan content in two genetically broad-based populations; to compare selected experimental lines to standard check cultivars; and to estimate genetic variances and heritabilities and to test for nonadditive genetic variance for ß-glucan content. We measured groat ß-glucan contents of check cultivars and parental lines and random S0:1 lines from initial and selected generations of each population grown in field experiments in 1996 and 1997 at two Iowa locations. Mean ß-glucan content increased from 53.9 to 59.9 g kg-1 in one population, and from 63.5 to 66.0 g kg-1 in the other, following selection. Genetic variance of ß-glucan content decreased by 9 to 22% following selection, but heritability for ß-glucan content did not change significantly. Heritability estimates ranged from 0.80 to 0.85 on a line mean basis. Additive variance was the only substantial component of genetic variance. Some experimental lines had significantly greater ß-glucan content than the best check cultivars and lines. Phenotypic selection for greater groat ß-glucan content will be effective for developing cultivars with elevated ß-glucan contents.
Abbreviations: BLUP, best linear unbiased predictor • FIA, flow injection analysis • NIRS, near infrared reflectance spectroscopy • PI, plant introduction • REML, restricted maximum likelihood
|
INTRODUCTION
|
|---|
SERUM CHOLESTEROL LEVELS are strongly related to coronary heart disease risk in humans (Mayes, 1990; Shekelle et al., 1981). Greater levels of high-density lipoprotein cholesterol are related to lower incidence of coronary heart disease, whereas greater levels of low-density lipoprotein cholesterol are associated with greater levels of coronary heart disease (Castelli et al., 1986; Stamler, 1979; Miller et al., 1977). Soluble fiber has the potential to reduce serum cholesterol levels in humans (Shinnick et al., 1991; Van Horn et al., 1988; Jenkins et al., 1975). Oat bran and oatmeal have been intensively studied as sources of dietary fiber for human diets (Sacks, 1991). Oat soluble fiber, when added to the daily diet, tends to lower serum low-density lipoprotein cholesterol and total blood cholesterol, particularly in individuals with initially greater levels of serum cholesterol (Welch, 1995; Ripsin et al., 1992; Shinnick et al., 1991; Anderson et al., 1984). This results in a reduction of the risk of coronary heart disease. The (1
3),(14)-ß-D-glucan (ß-glucan) has been identified as the active component of soluble fiber that lowers serum cholesterol (Davidson et al., 1991; Klopfenstein and Hoseney, 1987).
ß-Glucan is a cell wall polysaccharide found in seeds of the Gramineae (Stinard and Nevins, 1980; Nevins et al., 1978). Among cereals, oat and barley (Hordeum vulgare L.) have the greatest concentrations of ß-glucan (Wood, 1994; Åman and Hesselman, 1985; Prentice et al., 1980), but oat generally has a larger proportion of soluble ß-glucan (Lee et al., 1997). In oat and barley grains, ß-glucan is found mainly in the endosperm and the subaleurone layer (Wood, 1993; Wood et al., 1983). Lim et al. (1992) reported that the groat ß-glucan content ranged from 38 to 61 g kg-1 among 102 oat lines, including commercial cultivars, experimental lines, and accessions of the wild, interfertile oat relative A. sterilis L.
Oat ß-glucan content is a polygenic trait under the control of genes with mainly additive effects (Holthaus et al., 1996; Kibite and Edney, 1998). Heritability estimates for ß-glucan content have ranged from 0.27 to 0.58 (Holthaus et al., 1996; Humphreys and Mather, 1996; Kibite and Edney, 1998). ß-Glucan content is affected by environmental factors, including soil nitrogen level and precipitation (Brunner and Freed, 1994; Humphreys et al., 1994; Peterson, 1991; Peterson et al., 1995; Welch et al., 1991). Although genotype x environment interaction sometimes is a significant source of variation for ß-glucan content, the ranking of genotypes is generally consistent over environments (Peterson et al., 1995; Brunner and Freed, 1994; Lim et al., 1992; Saastamoinen et al., 1992).
The inheritance of ß-glucan content and the availability of technology developed to measure the trait rapidly by flow injection analysis (FIA) (Jørgensen, 1988) or near infrared reflectance spectrophotometry (NIRS) (Osborne et al., 1983) make possible the improvement of oat ß-glucan content through phenotypic selection. The development of oat cultivars with greater groat ß-glucan contents should increase the nutritional and economic value of the oat crop.
The objectives of this study were (i) to develop two genetically broad-based oat populations from parents with greater levels of ß-glucan content and to conduct phenotypic selection for greater groat ß-glucan content in these populations, (ii) to determine the progress from phenotypic selection of individual S0 plants for greater ß-glucan content, and (iii) determine if selection resulted in changes in genetic variance and heritability for ß-glucan content, and to determine if additive variance is the predominant component of genetic variance for ß-glucan content.
|
MATERIALS AND METHODS
|
|---|
Two base populations, BG1C0 and BG2C0, were developed in which to test the effects of selection for greater ß-glucan content.
Development of the BG1C0 Base Population
The BG1C0 population was developed by intermating 23 oat breeding lines and commercial cultivars chosen for their greater ß-glucan content or good agronomic characteristics (Table 1), followed by an additional generation of intermating unrelated F1 plants. A third generation of random mating among F1 plants resulted in 97 crosses. The 1665 S0 plants obtained from these 97 crosses constituted BG1C0, the base population for selection.
Selection within BG1CO for Greater ß-Glucan Content
In 1992, at Aberdeen, ID, 1665 S0 plants (from two to 38 per cross) from BG1C0 were grown as spaced plants in the field. S0 plants from the same cross made to obtain the BG1C0 seed represented a full-sib family. The S1 seed of each S0 plant was harvested individually and analyzed for groat ß-glucan content via automated FIA as described by Lim et al. (1992). The groat ß-glucan content of each plant was estimated by the mean of three subsamples and three individual measurements per subsample. The 40 C0 S0:1 lines with greatest ß-glucan content represented 21 full-sib families and were selected to be parents of the BG1C1 population. The 40 parent lines were intermated in the greenhouse in spring of 1993 by crossing each line to approximately six others at random to obtain 116 single crosses, which together constituted the BG1C1 population. In 1993, 582 S0 plants representing the BG1C1 population were grown as spaced plants in the field at Aberdeen, ID. S0 plants from each cross made to form BG1C1 represented a full-sib family. Seed was harvested from each plant individually to form 582 S0:1 lines.
Development of the BG2C0 Base Population
The BG2C0 base population was developed primarily by crossing selected lines from the BG1C1 population with lines from a population (BGPI) possessing germplasm from unadapted plant introductions (PIs) with greater ß-glucan content. The BGPI population was used as a donor for potentially unique alleles for greater ß-glucan content. It was developed by mating S0 plants from 10 crosses with at least 25% PI parentage (Table 2) to the same 40 S0:1 lines selected from BG1C0 that were also used as parents of BG1C1.
The choice of lines from the BG1C1 and BGPI populations to use as parents for the second base population was made on the basis of ß-glucan contents of S0 plants grown at Aberdeen in 1993. A total of 951 S0 plants (from one to ten per full sib family) representing the BG1C1 and BGPI populations combined were evaluated in a common experiment. Grain was harvested from each plant individually and analyzed for ß-glucan content via FIA.
Thirty-eight S0:1 lines representing 31 full-sib families from the BG1C1 population and seven lines from the BGPI population were selected to be used as parents of BG2C0 because of their greater ß-glucan contents. In addition, 12 high ß-glucan experimental lines and the cultivar Marion also were selected to be parents for the BG2C0 population, giving a total of 50 parents. The 50 parent lines were intermated in the greenhouse in spring of 1994 by crossing each line to approximately 10 others at random to obtain 248 crosses. The S0 seed of these crosses constituted the BG2C0 population. The BG2 population is related to the BG1 population, but differs from it by the inclusion of PI germplasm and of germplasm from only selected lines of BG1C0 and BG1C1.
Selection within BG2C0 for Greater ß-Glucan Content
In 1994, at Aberdeen, ID, 916 BG2C0 S0 plants (from one to five per cross) were grown as spaced plants. S0 plants from the same cross made to form the BG2C0 seed represented a full-sib family. Each S0 plant was harvested individually and analyzed for groat ß-glucan content via FIA. Selection was first performed among 248 full-sib families by selecting the 50 families with greatest mean ß-glucan content. Then within-family selection was practiced by choosing the line with the greatest ß-glucan content within each of the selected families to be a parent of the BG2C1 population. The selected lines were intermated in the greenhouse in spring of 1995 by crossing each line to approximately 10 others at random. S0 seeds from 250 crosses among these parent lines constituted the BG2C1 population. The BG2C1 population was grown as spaced plants at Aberdeen, ID, in the summer of 1995. Each plant was harvested individually to form 1471 S0:1 lines.
Field Evaluation
Experimental and check oat lines were evaluated in field experiments in 1996 and 1997. The experimental design was a sets within replications design. Each replication had five sets. Each set of 100 entries was arranged as a randomized complete block with two replications at each environment. The experiment was grown at the Agronomy and Agricultural Engineering Field Research Center near Ames, IA, on a Nicollet loam soil (fine-loamy, mixed, mesic Aquic Hapludoll) and the Northeast Research Center near Nashua, IA, on a Readlyn loam soil (fine-loamy, mixed, mesic Aquic Hapludoll) in both years.
The treatment design consisted of five populations (BG1C0, BG1C1, BG2C0, BG2C1, and BGPI). Each population was represented by 100 S0:1 lines. The family structure created during population development was maintained; lines that were derived from the same cross were members of a common full-sib family. Two lines were randomly chosen from each of 50 randomly chosen full-sib families, resulting in a sample of 100 lines from each population. An exception to this was that only 33 families (66 lines) from the BG1C1 population and only 17 families (34 lines) from the BGPI population were evaluated. Families were randomly assigned to five sets such that each set received both lines from 10 families from each of the BG1C0, BG2C0, BG2C1 populations; 4 to 10 families (8–20 lines) from the BG1C1 population; and zero to six families (0–12 lines) from the BGPI population. There were a total of 80 experimental lines per set. Twenty check or parent line entries also were included in each set. Checks included seven commercial oat cultivars (Don, Marion, Hazel, Premier, Ogle, Starter, and Noble), which were included in each set in duplicate. Each set also included five of the original parental lines of BG1C0 (Table 1) or BGPI (PI412928, PI361884, PI361886, PI502955, PI504593, or PI504601), which were assigned to sets at random. Each set also included a high ß-glucan experimental line, IAN979-5-2, to make a total of 100 entries per set.
Field plots consisted of hills of 20 seeds planted on a grid 0.3 m apart. Each experiment was surrounded by two rows of hills of a common check cultivar to provide competition to peripheral plots. Weeds were controlled manually. Plots were sprayed with the systemic fungicide Bayleton [1-(4-chlorophenoxy)-3.3-dimethyl-1-(1H-1,2,4-triazol-1-yl)-2-butanone] to protect them from crown rust [Puccinia coronata Corda] infection. To have sufficient seed for spectrophotometry and chemical analysis, the grain from both replications of each entry within a location was bulked together, mixed thoroughly, and the ß-glucan content of a representative portion of each sample was estimated.
Grain samples were dehulled with an air pressure dehuller to obtain approximately 8 g of groats. The ß-glucan content of each groat sample was determined with a near-infrared reflectance spectrophotometer (NIRS). The ß-glucan value for each sample was the mean of three measurements. To calibrate the prediction equation for ß-glucan content for each evaluation year, ß-glucan contents of 92 samples from the 1996 evaluation and 95 samples from the 1997 evaluation (representing approximately 10% of the total number of samples from each year) also were measured with automated FIA as described by Lim et al. (1992). The calibration samples were selected on the basis of NIRS spectral features to best represent the spectral variability of the whole set of samples. A ß-glucan determination for a sample chosen for the calibration equation was the mean of nine values: three subsamples were taken from each sample and three FIA measurements were obtained for each subsample. The prediction equations for each year of evaluation were developed using modified partial least squares regression (Benson, 1986; Aastveit and Martens, 1986). Briefly, 70 to 80% of the calibration samples (for which both NIRS and FIA measurements were available) were assigned at random to a training data set, whereas the remaining samples were assigned to a test data set. Partial least squares regression was applied to the training set. Models obtained from the training data set were applied to the test data set and the standard errors of cross validation were calculated for each model. The optimal number of latent variables extracted by the partial least squares regression method was determined by choosing the model that minimized the standard error of cross validation. Partial least squares regression was then applied to the entire calibration data set, retaining the optimal number of latent variables determined by test set validation. This model was then used to predict ß-glucan values for the remaining NIRS samples.
Statistical Analysis
To compare population means, an analysis was performed by Proc MIXED of SAS (Littell et al., 1996), considering population as a fixed effect factor and all other factors (environment, set, environment x set, population, environment x population, set x population, environment x set x population, genotype within set x population, environment x genotype within set x population, and residual) as random. The analysis was performed on values for ß-glucan content of each bulk of grain representing an entry-environment combination. The residual variance was due to check entries repeated within sets and environments. Best linear unbiased predictors (BLUPs) for genotypes were computed as linear functions of fixed and random effect estimates. The BLUP for Genotype k within Population j and Set i was calculated as:
where m is the overall mean; Si is the effect of Set i; 
.i is the mean of environment x set interaction effects involving Set i averaged over all environments; Pj is the effect of Population j; SPij is the effect of the interaction of Set i and Population j; 
.ij is the mean of environment x set x population interaction effects involving Set i and population j averaged over all environments; and G(SP)ijk is the effect of the Genotype k within Set i and Population j. The sum of population x environment interaction effects over all environments for a specific population is zero, therefore this term does not contribute to the genotypic BLUP. Standard errors for comparisons of BLUPs within the same set were obtained using "estimate" statements in Proc MIXED (Littell et al., 1996).
The components of variance within each population were estimated by the REML method (Searle, 1971) using Proc MIXED of SAS (Littell et al., 1996). Each population was analyzed independently, considering all effects (environment, set, environment x set, family within set, environment x family within set, line within family within set, and residual environment x line within family within set) except the overall mean to be random. Significance tests of the components of variance were obtained with the likelihood ratio test, which assumes that the difference between the –2 REML log-likelihood of the full model and the reduced model without the component of variance in question has a chi-square distribution with one degree of freedom (Self and Liang, 1987; Littell et al., 1996).
We were able to classify lines according to which full-sib family (or cross) they belonged because of the family structure within populations. Therefore, we were able to estimate components of variance due to full-sib family and due to S0:1 line within family for each population. This partitioning of the total genetic variance into among- and within-full-sib family variation allowed us to test the hypothesis that additive genetic variance was the sole component of genetic variance, as follows. Genetic expectations of the family and line within family variance components were derived following Cockerham (1971)(1983). The genetic expectations for these variance components in the BG1C0 population were different from the other populations because the parents of the BG1C0 population were non-inbred S0 plants, whereas the parents of BG1C1, BG2C0, and BG2C1 were partially inbred S1 plants. The genetic expectancies of family and S0:1 line within family variance components of the BG1C0 population are:
 |
and
 |
where 
2A is the additive variance, 2D is the dominance variance, D1 is the covariance between additive effects and homozygous dominant effects, D2 is the variance of homozygous dominant effects, and 2AA, 2AD and 2DD are the additive x additive, additive x dominant and dominant x dominant epistatic variances, respectively (Nyquist, 1991). The expectations of the variance components of family and line within family of the BG1C1, BG2C0 and BG2C1 populations are:
Under the hypothesis that additive genetic variance is the only substantial component of the genetic variance, the following equalities are true: 2family – 2line
= 0 for BG1C0 and 2family – 32line = 0 for the BG1C1, BG2C0, and BG2C1 populations. These equations were tested assuming asymptotic normality of the variance component estimates (Self and Liang, 1987; Searle, 1971).
Heritability on a line mean basis was estimated as:
Heritability on a family mean basis was estimated as:
Approximate standard errors of heritability estimates were obtained by means of the delta method (Lynch and Walsh, 1997).
|
RESULTS AND DISCUSSION
|
|---|
Prediction Equation Validation
The coefficient of determination (R2) of the NIRS prediction equation for ß-glucan content was 0.76 for the 1996 evaluation and 0.80 for 1997. The standard error of calibration was 4.7 g kg-1 for 1996 and 3.8 g kg-1 for 1997. The standard error of FIA laboratory measurements was 3.3 g kg-1 in 1996 and 0.6 g kg-1 in 1997. The precision of FIA measurements was much greater in 1997 than in 1996, but this resulted in only a small improvement in the precision of the regression equation.
Response to Selection
Selection resulted in significant increases in mean ß-glucan content in both populations (Table 3). The mean ß-glucan content of the BG1 population changed from 53.9 g kg-1 in C0 to 59.9 g kg-1 in C1, an increase of 11% of the unselected population mean. A smaller increase of 4% was observed in the BG2 population, which changed from 63.5 g kg-1 in C0 to 66.0 g kg-1 in C1. The mean ß-glucan content of check cultivars was not significantly different from the BG1C0 or parental line means, and was significantly lower than the BG1C1, BG2C0, BG2C1, and BGPI means (Table 3).
View this table:
[in this window]
[in a new window]
|
Table 3. Mean groat ß-glucan contents of the initial (C0) and selected (C1) generations of two broad-based oat populations (BG1 and BG2); the wild germplasm donor population (BGPI); parental lines; and check cultivars averaged across environments.
|
|
Lines from the BG2C0 and BG2C1 populations generally had greater ß-glucan content than lines from BG1C0 and BG1C1. This was because the BG2 population was developed in part from lines selected for greater ß-glucan content from the BG1C1 population. In set one, for example, two lines from BG2C0 and six lines from BG2C1 had significantly greater ß-glucan content BLUPs than the check with greatest ß-glucan content, experimental line IAN979-5-2 (Table 4). The 10 lines with greatest ß-glucan content BLUPs in set one had significantly greater ß-glucan percentage than all of the commercial checks (Table 4). Similar trends were observed in the other four sets.
View this table:
[in this window]
[in a new window]
|
Table 4. Best linear unbiased predictors (BLUPs) of ß-glucan content (and rank) of experimental oat lines with greatest ß-glucan BLUPs and check cultivars from set one.
|
|
Selections were made on the basis of ß-glucan content of plants grown in Idaho (a favorable oat-growing region, Holland et al., 2000), whereas evaluation of response to selection was performed in Iowa (a more stressful environment for oat, Holland et al., 2000). Therefore, the observed gains in population means demonstrated that ß-glucan content was sufficiently stable as to permit substantial progress from breeding to be observed across diverse environments.
Variance Component and Heritability Estimates
The family, family x environment, and line within family components of variance were significant in each population (Table 5). The family variance component was larger than the family x environment interaction variance within each population. Heritability estimated on a line mean basis ranged from 0.80 to 0.85 and heritability estimated on a family mean basis ranged from 0.56 to 0.77 (Table 6).
View this table:
[in this window]
[in a new window]
|
Table 5. Variance component estimates (and their standard deviations) and test for non-additive genetic variance for ß-glucan content from each of four experimental oat populations.
|
|
View this table:
[in this window]
[in a new window]
|
Table 6. Heritability estimates (and their standard deviations) for ß-glucan content within each of four experimental oat populations.
|
|
Since genetic variance depends on gene frequency, selection may cause changes in the components of variance (Falconer and Mackay, 1996). We observed reductions of 9% and 22% in the genetic variance for ß-glucan content in BG1 and BG2, respectively, after selection for greater ß-glucan content (Table 5). Nevertheless, there was no evidence for a significant reduction in heritability due to selection in either population (Table 6).
The contrasts 2family vs 2line for BG1C0 and 2family vs 32line for the BG1C1, BG2C0, and BG2C1 populations were not significant (Table 5), indicating that we could not reject the null hypothesis that additive genetic variance was the only substantial component of the genetic variance. This result agrees with earlier reports that oat ß-glucan is a polygenic trait under the control of genes with mainly additive effects (Kibite and Edney, 1998; Holthaus et al., 1996). Assuming that additive genetic variance is the sole genetic component of variance, the total genetic variance in a random mated population is equal to 2family + 2line (Table 5).
|
CONCLUSION
|
|---|
Phenotypic selection of field-grown S0 plants for greater ß-glucan content increased the mean of this trait in the BG1 and BG2 populations by 11 and 4%, respectively. The intermediate heritability, the predominantly additive inheritance, the significant genetic variability of the trait, and the availability of instruments to rapidly measure groat ß-glucan concentration allowed substantial improvements to be made with single-plant phenotypic selection. Furthermore, ß-glucan content is sufficiently stable that selections made in favorable Idaho environments resulted in improvements realized in more stressful Iowa environments. The germplasm lines with elevated ß-glucan content developed from this program should prove useful as breeding parents for cultivar development programs.
|
ACKNOWLEDGMENTS
|
|---|
This research was supported by The Quaker Oats Co. and by a fellowship to C.T. Cervantes-Martinez by CONACYT. We thank Drs. D.D. Stuthman and V.D. Burrows for contributing germplasm and making some of the initial crosses for the BG1C0 population. We thank Drs. D.D. Stuthman, A.R. McElroy, and F.L. Kolb for supplying pedigree information. Journal paper No. J-19336 of the Iowa Agric. and Home Econ. Exp. Stn., Ames, IA, Project No. 3768, and supported by Hatch Act and State of Iowa funds.
Received for publication February 25, 2000.
|
REFERENCES
|
|---|
- Åman, P., and K. Hesselman. 1985. An enzymic method for analysis of total mixed-linkage ß-glucans in cereal grains. J. Cereal Sci. 3:231–237.
- Anderson, J.W., L. Story, B. Sieling, W. Lin Chen, M.S. Petro, and J. Story. 1984. Hypocholesterolemic effects of oat-bran or bean intake for hypercholesterolemic men. Am. J. Clin. Nutr. 40:1146–1155.[Abstract/Free Full Text]
- Aastveit, A.H., and H. Martens. 1986. ANOVA interactions interpreted by partial least squares regression. Biometrics 42:829–844.
- Benson, R.D. 1986. Statistical process control techniques using near-infrared reflectance spectroscopy. p. 257–260. In A.R. Baldwin (ed.) Proceedings of world conference on emerging technologies in the fats and oils industry, Cannes, France. 1985. American Oil Chemists' Society, Champaign, IL.
- Brunner, B.R., and R.D. Freed. 1994. Oat grain ß-glucan content as affected by nitrogen level, location, and year. Crop Sci. 34:473–476.[Abstract/Free Full Text]
- Castelli, W.P., R.J. Garrison, P.W.F. Wilson, R.D. Abbott, S. Kalousdian, and W.B. Kannel. 1986. Incidence of coronary heart disease and lipoprotein cholesterol levels. J. Am. Med. Assoc. 256:2835–2838.[Abstract]
- Cockerham, C.C. 1983. Covariances of relatives from self-fertilization. Crop Sci. 23:1177–1180.[Abstract/Free Full Text]
- Cockerham, C.C. 1971. Higher order probability functions of identity of alleles by descent. Genetics 69:235–246.[Free Full Text]
- Davidson, M.H., L.D. Dugan, J.H. Burns, J. Bova, K. Story, and K.S. Drennan. 1991. The hypocholesterolemic effects of ß-glucan in oatmeal and oat bran. J. Am. Med. Assoc. 265:1833–1839.[Abstract]
- Falconer, D.S., and T.F.C. Mackay. 1996. Introduction to quantitative genetics. 4th ed. Longman Technical and Scientific, Essex, UK.
- Holthaus, J.F., J.B. Holland, P.J. White, and K.J. Frey. 1996. Inheritance of ß-glucan content of oat grain. Crop Sci. 36:567–572.[Abstract/Free Full Text]
- Holland, J.B., A. Bjørnstad, K.J. Frey, M. Gullord, D.M. Wesenberg, and T. Buraas. 2000. Recurrent selection in oat for adaptation to diverse environments. Euphytica 113:195–205.
- Humphreys, D.G., and D.E. Mather. 1996. Heritability of ß-glucan, groat-percentage, and crown rust resistance in two oat crosses. Euphytica 91:359–364.[ISI]
- Humphreys, D.G., D.L. Smith, and D.E. Mather. 1994. Nitrogen fertilizer and seeding date induced changes in protein, oil, and ß-glucan contents of four oat cultivars. J. Cereal Sci. 20:283–290.
- Jenkins, D.J., A.R. Leeds, C. Newton, and J.H. Cummings. 1975. Effect of pectin, guar gum, and wheat fibre on serum-cholesterol. Lancet 1:1116–1117.[ISI][Medline]
- Jørgensen, K.G. 1988. Quantification of high molecular weight (13),(14)-ß-D-glucan using calcofluor complex formation and flow injection analysis: I. Analytical principle and its standardization. Carlsberg Res. Commun. 53:277–285.
- Kibite, S., and M.J. Edney. 1998. The inheritance of ß-glucan concentration in three oat (Avena sativa L.) crosses. Can. J. Plant Sci. 78:245–250.
- Klopfenstein, C.F., and R.C. Hoseney. 1987. Cholesterol–lowering effect of beta-glucan enriched bread. Nutr. Rep. Int. 36:1091–1098.
- Lee, C.J., R.D. Horsley, F.A. Manthey, and P.B. Schwarz. 1997. Comparisons of ß-glucan of barley and oat. Cereal Chem. 74:571–575.
- Lim, H.S., P.J. White, and K.J. Frey. 1992. Genotypic effects on ß-glucan content of oat lines grown in two consecutive years. Cereal Chem. 69:262–265.
- Littell, R.C., G.A. Milliken, W.W. Stroup, and R.D. Wolfinger. 1996. SAS System for mixed models. SAS Institute Inc., Cary, NC.
- Lynch, M., and B. Walsh. 1997. Genetics and analysis of quantitative traits. Sinauer, Sunderland, MA.
- Mayes, P.A. 1990. Cholesterol synthesis, transport, and excretion. p. 249–260. In R.K. Murray et al. (ed.) Harper's biochemistry, 22nd ed. Appleton and Lange, Norwalk, CT.
- Miller, N.E., D.S. Thelle, O.H. Forde, and O.D. Mjos. 1977. The Tromso heart-study. High-density lipoprotein and coronary heart-disease: a prospective case-control study. Lancet 2:965–967.
- Nevins, D.J., R. Yamamoto, and D.J. Huber. 1978. Cell wall ß-D-glucans of five grass species. Phytochemistry 17:1503–1505.
- Nyquist, W.E. 1991. Estimation of heritability and prediction of selection response in plant populations. Crit. Rev. Plant Sci. 10:235–322.
- Osborne, B.G., T. Fearn, and P.G. Randall. 1983. Measurements of fat and sucrose in dry cake mixes by near infrared reflectance spectroscopy. J. Food Technol. 18:651–656.
- Peterson, D.M. 1991. Genotype and environment effects on oat beta-glucan concentration. Crop Sci. 31:1517–1520.[Abstract/Free Full Text]
- Peterson, D.M., D.M. Wesenberg, and D.E. Burrup. 1995. ß-Glucan content and its relationship to agronomic characteristics in elite oat germplasm. Crop Sci. 35:965–970.[Abstract/Free Full Text]
- Prentice, N., S. Babler, and S. Faber. 1980. Enzymatic analysis of beta-D-glucans in cereal grains. Cereal Chem. 57:198–202.
- Ripsin, C.M., J.M. Keenan, D.R. Jacobs, P.J. Elmer, R.R. Welch, L. Van Horn, K. Liu, W.H. Turnbull, F.W. Thye, M. Kestin, M. Hegsted, D.M. Davidson, M.H. Davidson, L.D. Dugan, W. Demark-Wahnefried, and S. Beling. 1992. Oat products and lipid lowering. J. Am. Med. Assoc. 267:3317–3325.[Abstract]
- Saastamoinen, M., S. Plaami, and J. Kumpulainen. 1992. Genetic and Environmental variation in ß-glucan content of oats cultivated or tested in Finland. J. Cereal Sci. 16:279–290.
- Sacks, F.M. 1991. The role of cereals, fats, and fibers in preventing coronary heart disease. Cereal Foods World 36:822–826.
- Searle, S.R. 1971. Linear models. John Wiley and Sons. New York.
- Self, S.G., and K. Liang. 1987. Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions. J. Am. Stat. Assoc. 82:605–610.[ISI]
- Shekelle, R.B., A.M. Shryock, O. Paul, M. Lepper, J. Stamler, S. Liu, and W.J. Raynor. 1981. Diet, serum cholesterol, and death from coronary heart disease. New Engl. J. Med. 304:65–70.[Abstract]
- Shinnick, F.L., R. Mathews, and S. Ink. 1991. Serum cholesterol reduction by oats and other fiber sources. Cereal Foods World 36:815–821.
- Stamler, J. 1979. Population Studies. p. 25–88. In R.I. Levy et al. (ed.) Nutrition, lipids, and coronary heart disease. Vol. 1. 1st ed. Raven Press, New York.
- Stinard, P.S., and D.J. Nevis. 1980. Distribution of noncellulosic ß-D-glucans in grasses and other monocots. Phytochemistry 19:1467–1468.[ISI]
- Van Horn, L., L.A. Emidy, K. Liu, Y. Liao, C. Ballew, J. King, and J. Stamler. 1988. Serum lipid response to a fat-modified, oatmeal-enhanced diet. Prev. Med. 17:377–386.[ISI][Medline]
- Welch, R.W. 1995. Oats in human nutrition and health. p. 433–471. In R.W. Welch (ed.) The oat crop: Production and utilization. 1st ed. Chapman and Hall, London.
- Welch, R.W., J.M. Leggett, and J.D. Lloyd. 1991. Variation in kernel (13),(14)-ß-D-glucan content of oat cultivars and wild Avena species and its relationship to other characteristics. J. Cereal Sci. 13:173–178.
- Wood, P.J. 1993. Physicochemical characteristics and physiological properties of oat (13),(14)-ß-D-glucan. p. 83–112. In P.J. Wood (ed.) Oat bran. American Association of Cereal Chemists, St. Paul, MN.
- Wood, P.J. 1994. Evaluation of oat bran as a soluble fibre source. Characterization of oat ß-glucan and its effect on glycaemic response. Carbohydr. Polym. 25:331–336.
- Wood, P.J., R.G. Fulcher, and B.A. Stone. 1983. Studies on the specificity of interaction of cereal cell wall components with congo red and calcofluor: Specific detection and histochemistry of (13), (14)-ß-D-glucan. J. Cereal Sci. 1:95–110.
This article has been cited by other articles:
|
|
|
|
 
A. A. Chernyshova, P. J. White, M. P. Scott, and J.-L. Jannink
Selection for Nutritional Function and Agronomic Performance in Oat
Crop Sci.,
November 7, 2007;
47(6):
2330 - 2339.
[Abstract]
[Full Text]
[PDF]
|
|
|
|
|
|
|
C. T. Cervantes-Martinez, K. J. Frey, P. J. White, D. M. Wesenberg, and J. B. Holland
Correlated Responses to Selection for Greater {beta}-Glucan Content in Two Oat Populations
Crop Sci.,
May 1, 2002;
42(3):
730 - 738.
[Abstract]
[Full Text]
[PDF]
|
|
|
TOP
ABSTRACT
INTRODUCTION
