What is DNA?

DNA (deoxyribonucleic acid) is a molecule of heredity. It was first shown to be so in 1953. Since then, molecular genetics has boomed tremendously and continues to do so. Today, I heard that the cholera bacteria genome has been completely sequenced.

DNA is a molecule consisting of two intertwined polynucleotide strands that form a double helix.  DNA contains only the four bases adenine (A), guanine (G), cytosine (C) and thymine (T).  A always pairs with T, and G always pairs with C.  Nuclear DNA and the genes are located within the nucleus of a cell.  One hundred thousand is probably a lower estimate of the number of genes located on a typical chromosome of any mammalian organism

What is Mitochondrial DNA?

The cell also contains mitochondria which are energy producing organelles that are found within the cell cytoplasm.  Mitochondria contain their own DNA – mitochondrial DNA (mtDNA), which is a circular molecule.  In contrast to nuclear DNA, the mitochondrial genome has only 13 genes.  When people work on the mitochondrial genome, they are working on a very small number of genes.   

Inheritance of mitochondrial DNA is different than the inheritance of nuclear DNA and is shown here (slide not shown). Inheritance of mitochondrial DNA is strictly maternal so it goes only from mother to offspring. The male contributes no mitochondrial DNA to the next generation.

The main application of mitochondrial DNA in conservation genetics.  In bison genetics, mtDNA is used to determine the evolutionary relationships between species, sub-species and populations. This is very useful because the mutation rate of mitochondrial DNA is faster than nuclear DNA. There are more changes over less time, therefore you can see greater differences. There is more opportunity to see variance between sub-species and different populations. It is also good for constructing evolutionary relationships because there is no recombination in mitochondrial DNA. Recombination mixes things up between what the male contributes and what the female contributes. The genes that can mix up can have different historical lineage. Recombination will mix up the lineage but with mitochondrial DNA, the lineage represents a true evolutionary lineage.

It is useful to look at  mitochondrial DNA in bison (slide not shown).  Here you can see the bison mitochondrial DNA (shows 2 samples, C-2 and C-35 from 1000 base pairs) and cattle mitochondrial DNA. You can see that the bison C-35 has cattle mitochondrial DNA in it, which means that at some time in the past, a bison bull mated with a beef cow and the offspring was taken into the bison herd. It is only passed on in lineage so if you do have a male with cattle mitochondrial DNA he does not pass it on to his offspring. Just the females do this.

What is Nuclear DNA?

The applications of nuclear DNA are much more broad, mainly because they contain most of the DNA, i.e. most of the gene’s coding for function in an organism. The inheritance of nuclear DNA comes from both the mother and the father. They each contribute 1/2 of the genes. This information can be used to determine paternity analysis.

Locus means some region of DNA (slide not shown).  Usually it is a gene that produces a protein and has a function, but it can also refer to a location on the chromosome. If you have two variants of the same locus, they are referred as different alleles.  Alleles occupy the same region of the chromosome, but they affect the same trait in different or an alternative manner.  You may have two different forms of DNA at a locus, that is, some sequence difference in them.

What is DNA Fingerprinting?

DNA fingerprinting is one of the techniques that is useful for gene mapping, finding genes that cause genetic diseases, identification of individuals and doing paternity analysis. DNA fingerprinting was first developed in 1985 but the type that is used at present was developed in 1989. It is now called DNA profiling.

DNA profiling (or fingerprinting) is gene typing of  an individual at a large number of loci.  The best way to do that, is to use loci that are highly variable.  Remember, your goal is to identify one individual from every other individual. By using several loci that are highly variable you can likely produce a genotype that does not occur in any other individual. In human forensics, they are now using only 13 different loci to type all individuals.

The type of loci we use to do this are called VNTR loci (variable number of tandem repeats) (slide not shown). VNTR loci are regions of DNA which contain tandem repeats of a short nuclear sequence. They have many alleles because you can change the size by changing the number of the repeats. The repeat that is most commonly used is very short, not 60 base pair repeats, but very short ones. The repeats are like AC, AC, AC, AC, AC, AC, AC, AC. The number of repeats will vary.  Each time they vary, and each number that varies, will represent a different allele because it has a different size. The type of variation you see here (slide not shown) shows 3 repeats in one and 5 repeats in the other. That makes them different sizes. When they are put on a gel and an electric current is run through the gel, the larger pieces of DNA get entangled in the gel so the two sizes can easily be separated.

A micro satellite locus is shown here (slide not shown).  An example of unique DNA is shown here as AC,AC,AC,AC,AC,AC,AC,AC.  The second double strand of DNA is the complement, so it is TG,TG,TG,TG,TG,TG,TG.  When DNA fingerprinting is done, whether it be for human forensics or bison paternity, they use loci like this with the different repeats. This example has 31 repeats but some can have 29 or 27 repeats. The size would then be slightly different.  You can tell them apart, because they run different on a gel.

Micro-satellites have become very popular because only a small amount of DNA is needed.  A paternity analysis can be done on a sample of hair as long as it has follicles on it. This is something a producer can collect himself. The quality does not even have to be good as the DNA can be amplified very easily. This amplification technique was developed in 1987. For cattle there are about 3000 different markers and those markers will work on bison. There are a lot of micro satellites to work with.

Measurements of Genetic Variation

There are  two ways to determine variability in a herd. The first is to examine 10 micro satellite loci, look at the average number of alleles at each of the loci and record that. The other is heterozygousity, which is the probability that the two copies of the gene that are found at a locus are different alleles.

Greg Wilson, my graduate student, has looked at several different populations of public herds to see how variable they were. These are averaged over 11 loci (slide showed examples from different bison populations). This is Antelope Island,  Custer State Park, Elk Island National Park, National Park Plains Bison, Ft. Niobrara, the National Bison Range, Big Mountain (which came out of Elk Island National Park plains bison), Wichita Mountains, Yellowstone National Park, Elk Island National Park wood bison, Mackenzie Bison Sanctuary (north of Great Slave Lake) and the Wood Buffalo National Park bison.

As you can see, most of the heterozygousity is more or less the same. There is a little bit of difference between them, in particular the heterozygosities are a little lower in the Elk Island National Park wood bison, Mackenzie Bison Sanctuary and Wood Buffalo National Park.  The only one that is strongly deficient in heterozygousity (in bred) is Antelope Island.

How do you detect inbreeding in the public herds? - from a genetic perspective as opposed to simply observing bad calves or observing genetic diseases within the herd.  There are not a lot of ways to do it, however, because a reference point is needed from which to measure.  In order to calculate inbreeding you must understand that things are identical by descent.  For example, a brother and sister both have genes from their parents. If they mate, any given gene in question can be homozygous.  Therefore, in that individual the gene could have come from the mother or from the father.  That is what is meant by an inbred population, but you do need that reference point in order to calculate an inbreeding coefficient for a population or an individual. You really need an individual’s pedigree and one way to get a pedigree is to do a paternity analysis.

One reason to worry about inbreeding is inbreeding depression, which is the detrimental effects of inbreeding. Most out-crossing species that have been quickly inbred will show inbreeding depression.  Species that are inbred to begin with, and continue to be inbred,  will probably not show much deleterious effect because they have rid themselves of all bad genes.  If they had not they would have died along the way. In an out-crossing species, there is no mechanism to do that, so when an out-crossing species inbreeds, problems result.

Paternity Analysis

This female individual (slide not shown) is homozygous (lavender and lavender) and in this example, the male is heterozygous (chartreuse and yellow).  And if the resulting offspring is heterozygous (lavender and chartreuse), then the mother must have given the lavender gene and the male must have given the chartreuse gene. Next, go through all the males in the population and find every individual that has a chartreuse gene.  At that locus, any of them could be the father of that offspring.  On a larger scale take several loci. This is one loci represented as 194-194. This is another loci, so the pairs of loci are down and you have 2 copies of the gene each.  If you examine the second  locus of this individual offspring, 160-162, then the female must have given 160 and the male must have given 162, so the male that has 162 is the father. Search all the males for 162. Do the same thing for all other identifiable loci until you eventually eliminate all the males, except one – the father.

Quantitative Genetics

Most of the traits that producers are interested in breeding for are not traits that are used for DNA fingerprinting.  Traits such as birth weight, weaning weight or quantity of milk produced are called quantitative traits.  They are distinctly different from the discreet traits that we have been discussing up to now where there are two forms of alleles.  Quantitative traits describes when there are several forms of alleles.  Mendelian segregation occurs but not of discreet characters but of continuous characters. 

With such traits, the exact genotype of the individual will not be known because several genes interact to determine those quantitative traits. They are controlled by a large number of loci, each of which has a small effect. Also the environment affects quantitative traits. The study of how the environment interacts with quantitative traits when subject to selection is the study of quantitative genetics.

One of the concepts used in quantitative genetics is heritability.  The heritability of a trait can be estimated by comparing the mean (average) of the offspring size, or whatever trait is being measured, to the mean of the traits of the parents. As an example, a trait is heritable if two large animals are mated and they produce a large offspring. A trait is not heritable if two large animals produce an average sized offspring.

Take beans as an example. They are extremely inbred and essentially homozygous. If you try to select for size by selecting two beans that are long, then mate them and look at the distribution of sizes of beans in the offspring, you would find this distribution to be no different than if you had mated two short ones. There is no heritability for size in beans because they have no genetic variation. Another way to measure heritability is to look at the mean of the parents and do a regression analysis against the size of the offspring.

Response to Selection

Selection of quantitative traits will be measured with a response. When doing a selection, you take males of a certain size or higher; they will have a certain mean. Look at all their offspring and look at their mean. That is the response to selection; the differential from where the mean of the total population was to where it ended up after you have done selection. The greater the heritability of the trait the easier it is to select for that trait. Very few traits in very few populations are not heritable. You can select for almost anything in most species. Dogs are a prime example of what can be selected for. They are all wolves but you would not know it.

One of the problems of continuous selection is that sooner or later you will not be able to select any further because you run out of genetic variation for the character you are trying to select for.  Most of the time selection is done with a small number of individuals not the whole herd so you risk inbreeding your herd. The other concern is a correlated response, which means you select for one thing but end up with a trait you do not want. Usually that trait is a loss of reproductive ability or loss of growth.

Selection Index

One way around selecting for a single trait is to select for a series of traits - all at the same time (slide not shown). It is not advisable to do the first selection for weight gain at weaning and then select for size of the adult. Do them all together. It is very difficult to do each one separately. Determine the kind of total animal that you want and give a specific trait a weight of how important it is to the merit of the animal you ultimately want. Certainly one trait would be ease in calving.  Select on that weighted sum of the traits. That is called an selection index. You really do not want to select for one trait because if you continuously select for size, will end up producing a very big animal that does not do anything else well.

Editor’s note:  this informal Introduction to Genetics introduced subsequent presentations within the Breeding and Genetics Session.