Emerging DNA Technologies for Breed Improvement

Contributed by Paul Charteris, Dorian Garrick & Grant Montgomery

A bigger toolbox is becoming available for animal and plant improvement with the emergence of new DNA-based genetic technologies. This issue of Breeding Matters gives you a preview of some applications of technologies in that toolbox. Knowledge of the structure and function of DNA is not a prerequisite for direct selection to result in breed improvement. Breeders have been manipulating DNA for many centuries by selecting superior individuals to parent the next generation and by culling inferior individuals. A major drawback of direct selection is its reliance on assessment of the actual characteristics of the individual or its relatives. Emerging DNA technologies allow characterisation of an individual from inspection of its DNA sequences. This may allow selection to be undertaken at younger ages or with greater accuracy than is currently achievable by estimating breeding values from measurements on the individual and its relatives. Some features of the variation in DNA between individuals that can be exploited by new technologies are described in this issue. Three applications of this knowledge are outlined including: identification of parentage; testing for carriers of recessive genes; and selection using markers to major genes (known as quantitative trait loci, or QTLs).

DNA Structure & Function

Figure 1: Schematic representation of DNA

The genetic makeup of animals, plants, birds and most other organisms is determined by DNA (Deoxyribose Nucleic Acid). The structure of DNA is a double-helix (see Figure 1) represented by two strands, each strand consisting of a sequence of four nucleotide compounds (Adenine, Cytosine, Guanine and Thymine), known as bases. The sequence on one strand reflects the sequence on the complementary strand because Adenine on one strand always pairs with Thymine on the other strand and Cytosine pairs with Guanine. DNA is replicated by cleaving between the strands to form two sides, each side manufacturing a new complementary strand, resulting in two identical copies of the original double helix. A paired strand of DNA is known as a chromosome. Cattle normally have 60 chromosomes, grouped into 30 pairs. Sheep have 54 chromosomes or 27 pairs. One member of each pair of chromosomes is inherited from the dam (maternal chromosome) and the other member of the pair comes from the sire (paternal chromosome). The DNA strands store instructions for the manufacture of proteins. Following predetermined start regions, the order of bases, up to a stop region, dictates a sequence of organic compounds known as amino acids. Each so-called triplet of three bases (a codon) identifies the relevant amino acid (e.g. AAG corresponds to Lysine). Different proteins are described by their sequence of amino acids. The location of the instructions for a particular protein is known as a locus. At each of these localities (known as loci in plural), there is information carried for a particular function within the animal (such as coat colour). This information is known as a gene. For some traits such as coat colour or the presence of horns a single gene may affect the trait, however most productive traits such as number of lambs born, fleece weight, milk yield or calf weaning weight are controlled by many genes. Each chromosome contains thousands of genes. Genes for different traits may be located close together or far apart on the same chromosome, or on different chromosomes.

Genetic variation

Within the unit of DNA that codes for a protein are regions (called exons) which contain the instructions (or code) for the protein and supposedly unimportant regions (known as introns) that separate the exons. A change to an exon sequence will usually affect a cell process whereas changes in an intron sequence will usually be of little or no consequence. Accordingly, when DNA is compared between individuals within a population, far more variation is found for intron sequences than for exon sequences. One form of this variation involves monotonous replication of small units of DNA known as microsatellites. For example, an intron might contain an alternating sequence of two bases such as Adenine and Cytosine. One parent might have, on one chromosome in a pair, the sequence -T-G-(A-C)-(A-C)-(A-C)-G- which contains three AC repeats whereas the other chromosome in the pair may contain -T-G-(A-C)-(A-C)-(A-C)-(A-C)-(A-C)-G- which has five AC repeats. Half the progeny of this parent will receive the version with three repeats, and the other half of the progeny will receive the version with five repeats. An individual that contains neither three repeats nor five repeats at this locus would be excluded from being a progeny of this parent. The number of such repeats can vary widely among parents in the population. A laboratory method that enables the number of repeats at certain loci (known as marker loci) to be readily counted provides a method for identifying or verifying parentage directly from a small tissue sample containing DNA, such as can be obtained from milk, semen, blood or fibre.

Parentage identification

Parentage testing has been available for some time, exploiting variation in blood groups. Testing requires a blood sample which precludes the use of this technique for identification of embryos or semen. Some blood analyses lead to inconclusive results regarding parentage. DNA analysis allows a far greater accuracy of parent identification through comparison of microsatellite sequences of an individual and its candidate parents. A DNA-based technique can be used to identify parentage in situations with multiple sire matings.

Figure 2. Using a Marker to Determine Parentage



The technique can be demonstrated by example. Suppose a calf has its DNA typed for one particular microsatellite locus. This analysis might show the calf inherited 6 repeats and 4 repeats of the microsatellite sequence, as shown in Figure 2. Assuming all the possible sires and dams have also been DNA tested as shown in Figure 2, Bull A would be immediately excluded from being the sire of the calf, as this bull does not carry the 4 or 6 repeat sequence. All progeny of Bull A must have either 3 or 5 repeats. Bull B would therefore be the likely sire. If so, the calf must have inherited 4 repeats from Bull B and the 6 repeat sequence carried by the calf must then have come from the dam. Both cows carry the 6 repeats and either could therefore be the dam. This process of elimination can be repeated for microsatellite markers at other locations and unique identification of parents can usually be achieved through the use of five to ten different markers. A possible use of the technology would be to identify the sires of exceptional carcasses processed through meat plants. Commercial parentage services for deer and cattle are available in New Zealand through GenomNZ, a company formed by AgResearch.

DNA tests to identify carriers of some known recessive genes

Most production traits are influenced by a large number of genes. Some characteristics including coat colour and many congenital diseases are controlled by one or few genes. For example, consider the gene that controls red colour in Angus cattle. True-breeding black Angus animals carry two copies of the gene for black. True-breeding red cattle carry two copies of the red gene. Crosses between black and red Angus will produce black animals that carry the red gene, but the effect of the red gene is masked by the presence of the black gene. Without the use of DNA tests, it is impossible to distinguish a black animal carrying the red gene (known as a heterozygous individual) from a true-breeding black animal (known as homozygous), except by progeny testing. Given knowledge of the biochemical pathways involved in the expression of coat colour, it is possible to isolate the proteins responsible for black as compared to red colour, and therefore to develop a test that detects the presence of the DNA base sequence that produces either colour. This allows a carrier to be identified using a DNA test, perhaps from semen or from a sample taken off an embryo. Identification of carriers facilitates selection to change coat colour or to limit the occurrence of some congenital diseases. The dairy industry routinely tests young bulls for a number of such conditions (including BLAD, citrullinemia, Factor XI and DUMPS). Research is ongoing to identify tests for the presence of horned/polled genes, spider syndrome in sheep and AgResearch has flanking marker tests (see next section) for the presence of major genes for sheep fertility, such as the Booroola and Inverdale genes.

Markers for major gene effects

Suppose a sire is heterozygous for a particular microsatellite marker. For example, the sire might have 5 repeats on one member of a chromosome pair and 7 repeats on the other member. Offspring of this sire can be partitioned according to which version they inherited from their sire. One would expect about half of the progeny to inherit the 5 repeats and the rest of the progeny would inherit 7 repeats. Sometimes these two groups of progeny appear or perform differently for a particular trait. This suggests that a gene that affects this trait is physically located on the same chromosome, near the microsatellite marker. In the case of production traits, such a gene is known as a Quantitative Trait Loci (QTL).

Figure 3. Microsatellite markers flanking a QTL


The use of microsatellite markers located on every chromosome would ensure every QTL would be close to a marker and many would be flanked by markers on either side. This situation offers at least two opportunities for new selection strategies. Suppose an animal is heterozygous for a QTL. That is, the animal carries two forms of the QTL, one form being more desirable than the other form. The genetic makeup of such an individual is shown in Figure 3. Only those progeny that inherited the flanking version of the markers (i.e. 5 repeats and 9 repeats) would be used for breeding as these are likely to carry that section of the paternal chromosome that includes the desirable form of the QTL. Offspring with 7 and 3 repeats would be culled as these would have the section of the maternal chromosome which includes the undesirable form of the QTL. A few progeny may have new combinations of the flanking markers such as 5 and 3 repeats, or 7 and 9 repeats, and these would likely be culled. The proportion of such progeny is likely to be low but depends upon the physical distance between the microsatellites that flank the QTL. This technique of selecting individuals on the basis of flanking markers would be particularly useful to introduce a desirable QTL from one breed into another breed without carrying along other undesirable characteristics. An alternative application of this technology would involve routine marker assessment of many animals, these data being included in evaluation procedures to obtain BVs, along with ancestral and individual performance records.

Finding QTLs

One experimental approach to detect the presence of QTLs involves the establishment and use of crosses between genetically distant strains such as between Romneys and Merinos, lean and fat sheep, or Jerseys and Limousins. Jerseys and Limousins are quite different in appearance and will be homozygous for different sets of genes. When a first-cross (F1) animal is produced it will be heterozygous for all these genes. That is, the F1 will inherit one copy of the Jersey form of the marker and one copy of the Limousin form of the marker. When an F1 bull is used as a sire half of its offspring will inherit the Limousin form of the marker and half will inherit the Jersey form of the marker. If these two offspring groups differ in their appearance or performance then the QTL affecting performance may be located close to the region where the microsatellite marker is located. A systematic search using 200 or more markers spaced along all the chromosomes can enable discovery of any large or moderately sized genes which make Limousins and Jerseys different. An alternative approach to identify QTL's involves study within sire families. Suppose a sire is heterozygous for a marker. The offspring can be categorised into two groups according to the marker variant inherited from the sire and the performance of these two groups can be compared. If these progeny groups differ in performance, this indicates that the sire may be heterozygous for a QTL located near the marker. Such an exercise will require large progeny groups with offspring typed for a range of markers. A slight variation of this approach involves only DNA testing sons of a sire and relating the merits of the sons to the marker variants of their sire. The merits of the sons are determined by measurements on their offspring. That is, three generations are involved: the sire; his sons; and progeny of the sons. In this grandsire design, the genes that are detected could be exploited immediately, for example by selecting those young offspring that inherited the markers associated with the favourable QTL. Identification and exploitation of these QTLs will rely on the collection and analysis of pedigree and performance records. Strategic alliances between sire breeders, molecular and quantitative geneticists will be essential to the cost-effective application of these new technologies in livestock breeding programmes. Progress in the search for QTLs is currently limited by the availability of well-recorded pedigree and performance information. Summary On a world-wide basis, considerable resources have and are being invested in the development of these technologies. These developments promise to be of enormous benefit in expanding our knowledge of the genetic basis of some characteristics. However, the practical application of these technologies to animal improvement is very much in its infancy. Breeders need to remain informed as to the availability of these technologies and to the potential use of them in their own breeding programmes.