A Database for Triticeae and Avena
The goal of the barley breeder is to develop new cultivars that are
superior to the "best" existing cultivars. This is accomplished usually
by combining large numbers of favorable genes for agronomic and malt
quality. Many of the gains made in plant breeding are relatively small,
but they accumulate over time and result in the frequent release of
improved cultivars. Thus, much money and time is invested to identify new
To increase the probability and efficiency of obtaining superior new
cultivars, it would be advantageous to identify genes that have a large
effect on agronomic traits and malt quality. Most of the economically
important traits of barley (e.g. yield, test weight, and many malt quality
parameters) are known as quantitative or polygenic traits. However, not
all traits of interest are controlled by this type of gene action. In
animals, many traits that are generally regarded as polygenic have been
associated with a single gene (Crawford and Smith, 1964; Hartman, 1972;
Merat, 1986; Pirchner, 1988). A major gene affecting a quantitative trait
that has been localized to a chromosome is called a quantitative trait
locus (QTL) by Gelderman (1975). If QTLs can be identified in barley,
they can be evaluated and accumulated in improved lines.
Two methods of identifying QTLs are available currently. The first
method involves utilization of restriction fragment length polymorphisms
(RFLPs), isozymes, and other molecular markers. Utilization of molecular
marker techniques requires special facilities, equipment, and personnel;
thus, these methods are expensive and unavailable to many barley breeders.
The second method is to identify QTL linked to a morphological marker
locus. Morphological markers are simply inherited genes that cause a
visual change in a morphological trait such as kernel color, surface
waxes, plant height, etc. Because a large number of morphological markers
have been mapped, this method is available to barley breeders and does not
require special equipment, facilities, or personnel. A proposal
utilizing morphological marker loci to identify QTLs is outlined and
After suitable morphological markers are selected and placed in a
common genetic background, identification of a QTL is a four step process.
First, a possible association between a QTL and morphological marker locus
must be identified. Secondly, it must be determined if the association
between the morphological marker locus and the proposed QTL is due to
linkage or pleiotropy. Next, if linkage is observed, the QTL needs to be
localized in relation to other morphological markers on that chromosome.
Finally, it must be determined if the QTL is omnipresent in elite breeding
BC1 lines can be used to identify possible associations between a
morphological marker locus and QTL. The recurrent parent should have
excellent agronomic and malt quality and the donor parent with the
morphological marker should have relatively poor agronomic and malt
quality as compared to the recurrent parent.
Agronomic quality of BC1F3 and BC1F lines can be evaluated in yield
trial experiments conducted over two summers at two locations. Growing
the experiments over two years will allow an evaluation of the genotype X
environment effects. An augmented block design (Federer, 1961) should be
the best experimental design to detect significant differences between BC1
lines and other entries in the experiment. Use of an augmented block
design would be advantageous over replicated designs for several reasons.
Replicated yield trials would require an extra generation to increase seed
for yield trials and would occupy more area in the field. Also, because
of the greater size of replicated yield trials, only five to ten
morphological marker loci could be evaluated each year for possible
associations to QTLs. With an augmented block design, up to 50
morphological marker loci could be screened each year. For midwestern
six-rowed breeding material, entries that could be included in each block
are the recurrent parent, Morex as the quality check, the donor parent,
and seven to ten BC1F3 lines having the same morphological marker.
Field notes should be taken on all desired agronomic traits (e.g.
heading date, plant height, foliar diseases, lodging, etc.) At maturity,
each plot would be harvested and individually threshed. Data to be
collected after harvest could include grain yield, test weight, 1000-
kernel weight, percent barley protein, kernel assortment and other
characters of interest. Malt quality can be evaluated on samples of grain
obtained from the yield trial experiments. Malt parameters to be studied
could include percent soluble wort protein, percent fine-grind extract,
percent coarse-grind extract, fine-coarse grind differences, diastatic
power, alpha-amylase activity and other parameters of interest.
Possible associations between a morphological marker locus and a QTL
can be determined statistically. If BC1 progeny are distributed normally
between the donor line and the recurrent parent, no association between
the marker locus and a QTL for a specific trait should be assumed. An
association between a marker locus and a QTL is possible if the values for
BC1 lines are skewed towards the donor line.
For those traits showing an association with a genetic marker, it will
be necessary to determine if the association is due to pleiotropy or due
to tight linkage with a QTL. To screen for linkage, a BC1F4 line that has
a low value for the trait in question and expresses the morphological
marker should be backcrossed to the recurrent parent. Two spikes from
1,000 randomly selected plants should be harvested from the BC2F2
population. The progeny should not be selected for expression or non-
expression of the morphological marker. BC2F2 derived BC2F3 families
should be grown in head rows and notes should be taken on the
morphological marker and the trait in question for each family. Presence
of linkage can be determined using chi-square analysis. Fisher's Maximum
Likelihood Method can be used to estimate linkage intensity between the
marker locus and the QTL.
Other elite breeding material should be evaluated for the presence of
the QTL. To screen the germplasm, a BC1F4 line that is a recombinant
(i.e. a BC1F4 line with the marker locus and quality resembling the
recurrent parent for the trait in question) can be crossed to several
elite lines. In the F2 generation, two spikes from at least 50 plants
expressing the marker phenotype should be selected. F2 derived F3
families should be grown in head rows and notes should be taken on the
trait(s) in question for each family row. The QTL will be deemed absent
in the elite line if transgressive segregates are found in the F2 derived
F3 families. If all families appear similar, then the same QTL will be
assumed to be present in both populations.
Ideally, morphological markers and molecular biology techniques would
be combined to develop a map of the barley genome which includes
morphological marker loci, enzymatic loci, RFLPs, and QTLs. With such a
map, even barley breeders having limited resources could manipulate
important QTLs. Several researchers (Tanksley et al., 1981; and Tanksley
et al., 1982) have shown that introgression of a favorable QTL from
unadapted material is possible. Selection for the marker along with the
linked QTL will reduce the generations required to complete backcrossing
and space required for evaluation of segregating progeny.
Possible associations between morphological marker loci and QTLs have
been studied at North Dakota State University (Horsley, 1988; Gonzalez,
1990). Horsley identified a possible association between the ant13 gene of
DM582 on chromosome 6 conditioning for low polyphenol content (Falk, 1985;
Hormis, 1988) and a major gene for diastatic power (DP). Gonzalez (1990)
identified a possible association between the v3 gene on chromosome 5 for
intermediate six-row type (Fukuyama, 1983) and a major gene conditioning
resistance to spot blotch, incited by Cochliobulus sativus (Ito and
Kurib.) Dreschs. ex Dast.
Continued research on the ant13 gene will include determining if the
association is due to linkage or pleiotropy. If the association is due to
linkage, linkage intensity between the QTL affecting DP and the ant13 gene
will be studied. Since the ant13 has been located on chromosome 6,
linkage intensity will be determined between the QTL for DP and other
morphological marker genes located on chromosome 6. If the association is
due to linkage, it will be determined if the QTL affecting DP is present
in barley lines having inherently low DP. If it is not, this QTL could be
introgressed into such lines to further increase DP.
Crawford, R.D. and J.R. Smyth. 1964. Studies on the relationship between
fertility and the gene for rose comb in the domestic fowl. Poul. Sci.
Falk, D.E. 1985. Genetic studies with proanthocyanidin-free barley.
Federer, W.T. 1961. Augmented designs with one-way elimination of
heterogeneity. Biometrics 17:447-473.
Fukuyama, T. 1983. Six-rowed 3. BGN 13:113.
Gelderman, H. 1975. Investigations on inheritance of quantitative
characters in animals by gene markers. Methods Theor. Appl. Genet.
Gonzales-Ceniceros, F. 1990. Assigning genes conferring resistance to
net and spot blotch in barley to a specific chromosome. Ph.D. Thesis.
North Dakota State Univ., Fargo.
Hartman, W. 1972. Relationship between genes at the pea and single comb
locus and economic traits in broiler chickens. Brit. Poul. Sci. 12:305-
Hormis, Y.A. 1988. Location of genes controlling proanthocyanidin
production in barley. Ph.D. Thesis, North Dakota State Univ., Fargo.
Horsley, R.D. 1988. Effects of the ant13 gene of DM582 on agronomic
traits and malt quality of barley. Ph.D. Thesis, North Dakota State
Merat, P. 1986. Potential usefulness of the NNa gene in poultry
production. World Poul. Sci. 42:124-143.
Pirchner, F. 1988. Finding genes affecting quantitative traits in
domestic animals. p. 243-249. In B.S. Weir, M.M. Goodman, E.J. Eisen,
and G. Namkoong (eds.) Proc. 2nd Int. Conf. on Quantitative Genetics,
Raleigh, North Carolina. 31 May to 5 June, 1987. sinauer Associates,
Inc. Sunderland, Massachusetts.
Tanksley, S.D., H. Medina-Filho, and H. Rick. 1981. The effect of
isozyme selection on metric characters in a interspecific backcross of
tomato - basis of an early screening procedure. Theor. Appl. Genet.
Tanksley, S.D., H. Medina-Filho, and H. Rick. 1982. Use of naturally
occurring enzyme variation to detect and map genes controlling
quantitative traits in a interspecific backcross of tomato. Heredity