Breeding strategies for the development of a disease-resistant stock of livestock

C H A P T E R

20 Breeding strategies for the development of a disease-resistant stock of livestock O U T L I N E Breeding strategy- factors effecting

298

Evaluation of available genetic resource Breed description Major disease traits and their inheritance (heritability, repeatability and genetic correlation) Repeatability Gene action (additive or nonadditive); nonadditive gene action includes heterosis, complementary effect, and maternal effect Heterosis Implications

299 299

Maternal effects

302

Existing constraints in improvement Other factors Economic relevance of the disease

303 303 303

Breeding objective

303

Breeding plan Choice of selection method A practical example of selection index developed in cattle farm with beef production unit at the international level Basis of selection or criteria of selection Individual selection or performance testing

303 303

Selection based on pedigree records Selection based on progeny testingdfarm or field

Choice of mating system Breeding organization Open nucleus breeding system

306 306 307 307

Multiple ovulation and embryo transfer

308

Case study for ONBS for Garole sheep Flow chart to achieve the target Work plan for genetic improvement of Garole sheep A classical example of disease resistance with epidemiological model: a case study

308 308 309

Breeding strategy for infectious disease Bacterial disease Mastitis Case study on genetic analysis on pathogen specific mastitis Disease traits in German Holstein cows Interpreting and analyzing field data Appropriate experimental designs

311 311 311

References

313

Further reading

314

299 299

300 300 301

304 304 305

305

310

311 311 312 312

In earlier chapters, we described in general the importance of breeding for disease resistance, genetic parameters for disease resistance, and selection for disease resistance. In this chapter, we shall describe the breeding strategy for disease resistance for different diseases as described in Chapter 2 with suitable examples. In earlier chapters, we had already discussed the issues as how disease-resistant traits are different from other phenotypic traits. These are difficult to measure, and challenge study has certain ethical concerns. One major problem is that frequency for disease occurrence is low. Diseases eliminate a relatively large proportion of parents to contribute to the gene pool. This leads to a series of complications associated with small population size.

Genetics and Breeding for Disease Resistance of Livestock https://doi.org/10.1016/B978-0-12-816406-8.00020-6

297

© 2020 Elsevier Inc. All rights reserved.

298

20. Breeding strategies for the development of a disease-resistant stock of livestock

Demographic stochasticity has adverse effects. Often unequal sex ratio leads to lower effective population size (Ne). This in turn causes inbreeding and genetic drift. Inbreeding is defined as the breeding of close animals (up to four to six generations) as opposed to those that would have been chosen at random from the mating population. Inbreeding leads to increased homozygosity and decreased heterozygosity in the population. Due to increased homozygosity, the frequency of harmful homozygotes also increases leading to incidences of Mendelian diseases and disorders. It also leads to reduced fitness in the population causing inbreeding depression. Reduced performance is observed in traits related to fertility and survival. For the standard breeding practice, the breeding population should have an effective population size of at least 100, and the rate of inbreeding should be less than 0.5% per generation. The development of mating schemes that maximize the response to selection while simultaneously minimizing the rate of inbreeding is an active area of research in animal breeding. Apart from inbreeding, genetic drift in small populations is important when formulating breeding strategies. Genetic drift is the variation in gene frequencies due to chance, causing random changes in gene frequency, particularly in cases of low gene frequency or small population size. As a result of genetic drift, there exists the loss of favorable alleles. As a result, genetic gain decreases, and selection is decreased and reached earlier. It is difficult to avoid the problems of genetic drift compared with those of inbreeding. An effective way of avoiding or overcoming the problem of genetic drift and inbreeding is recognizing that selection is usually more effective in larger populations. Many sheep farmers have joined sire reference schemes, which cause increase in effective population size and variation in breeding values. As a result, response to selection increases. Sire reference schemes require the use of the same sires in different flocks to create genetic links. Cattle breeders usually form part of a national breeding scheme to evaluate potential bulls.

Breeding strategy- factors effecting The first step in developing selection strategies to increase defensive response traits is to assess the presence of genetic variation for such traits and quantify the proportion of phenotypic variance explained by genetics (heritability). If heritability is large, selective breeding can improve the trait rapidly. It is also important to know whether there is any trade-off between the traits. For example, resistance and tolerance are negatively correlated on the genetic level, so improving one by selection will decrease the other unless both traits are included in the breeding goal and selection index. Existence of genetic variation in resistance to infection has been reported in farm animals such as dairy cattle, sheep, pigs, poultry, and fish. Heritability ranged from 0.04 to 0.33. The existence of genetic variation in resilience, without distinguishing between resistance and tolerance, has also been reported in sheep, pigs, and fish. Heritability of resilience, being the heritability of performance traits during disease periods, ranged from 0.09 to 0.46. The existence of genetic variation in tolerance is greatly overlooked. To date, there are very few studies on the genetic variation of tolerance to infection in farm animals. No genetic variance in tolerance of Soay sheep to strongyle nematode infection was observed, and no genetic correlation were observed between resistance and tolerance to strongyle nematode infection in Soay sheep. Genetic variation in tolerance, and a negative genetic correlation (
1.0) between resistance and to rodent malaria (Plasmodium chabaudi), were observed among five different inbred mouse strains. Genetic variation in tolerance to Listeria among four genetically diverse inbred mouse strains was observed. Differences in postinfection mortality as an indicator of tolerance for 11 lines (6 inbred and 5 wild) of Drosophila melanogaster infected by a strain of P. aeruginosa were also observed. In humans, there is evidence for variability in tolerance to human malaria. For instance, a monogenic disorder called aþ -thalassemia, causing formation of abnormal hemoglobin molecules, tends to reduce the incidence of severe disease, causing variability among individuals for disease tolerance (Williams et al., 2005). The basic points that need to be considered for formulating a breeding strategy for developing disease-resistant stock are as follows: ☛ ☛ ☛ ☛ ☛ ☛ ☛

available animal genetic resources, production potential, and utilization feed and fodder availability agroclimatic conditions agriculture and livestock production system in vogue marketing facility, available infrastructure, and development facilities socioeconomic status of farmer Considering these factors, let us discuss these points in detail.

Evaluation of available genetic resource

299

Evaluation of available genetic resource Breed description Breed-wise variation exists in disease resistance or susceptibility. Reports are available in which some breeds of livestock are reported to be resistant to a particular disease. The N’Dama breed of cattle was reported to be resistant to Trypanosomiasis in West Africa. Species-wise susceptibility or contrary resistance to a particular disease has already been observed. Comparatively better resistance in sheep has been observed in comparison to goats for PPR infection. Ducks were observed to be asymptotic to avian influenza in comparison with chickens. A list of diseases caused by different etiological agents is described in Chapter 2. Contrary to complete resistance, breeds for particular species have been observed to be less susceptible to certain diseases. Birbhum sheep were observed to be resistant to a wide range of diseases, parasitic diseases in particular (Pal et al., 2017).

Major disease traits and their inheritance (heritability, repeatability and genetic correlation) Major diseases have already been described in Chapter 2. It is very important to understand the mode of inheritance for disease traits, which vary from disease to disease, population structure, variability, and incidence of disease occurrence. There are difficulties in measuring the inheritance of rare diseases. In other words, the frequency for occurrence is much lower for such cases. Disease traits are mostly threshold traits, phenotypically expressed as all-or-none traits. It is difficult to assess the underlying continuous variable, which is normally distributed. It is very important to understand the mode of inheritance for disease traits. The ultimate goal for the breeding program includes selection of resistant stock of livestock. For the process of selection for a disease-resistance trait, heritability is very important. The ultimate genetic gain directly depends on the heritability of the trait concerned. Genetic gain would be better if we select the traits based on indicator traits. For example, if we need to select dairy cattle against mastitis, it is always better to select the animals based on an indicator trait such as somatic cell count, for a few reasons. Firstly, the trait as somatic cell count is a quantitative trait (not a threshold trait), easy to measure, and has high heritability, and hence the ultimate genetic gain would be greater. Heritability of somatic cell score is around 0.1, and genetic correlation between somatic cell score and clinical mastitis is around 0.6 to 0.8 (Shook et al., 1994) and 0.7 (Emanuelson et al., 1988). In any breeding program, selection is applied for lower somatic cell score, which will ultimately lead to better genetic improvement for total economic merit. Electrical conductivity for udder is another indicator trait for udder health. Heritability for electrical conductivity is estimated at 0.53  0.09 (Santos et al., 2018). But high correlation was not observed for electric conductivity with mastitis. Heritability of mastitis was observed to be low. In a study conducted on case of mastitis, heritabilities for mastitis and somatic cell count were observed to be 0.02 and 0.08, respectively (Emanuelson et al., 1988). Studies on resistance to Haemonchus contortus in Garole sheep reported it to be heritable. Through selective breeding or assortative mating of resistant animals, genetic resistance may be improved. The ability of resistance to natural infection can be detected as early as 6 months of age. Accordingly, the young males and females, resistant or rather less susceptible, may be selected at this stage based on the record of single FEC. Heritability for fecal count (measured as EPGdi.e., eggs per gram) at 3, 6, and 8 months of age was observed to be 0.610  0.103 versus 0.696  0.152, 0.538  0.094 versus 0.731  0.60, and 0.615  0.107 versus 0.579  0.173, respectively. A heritability of 0.23 for FEC was observed in the case of an artificial H. contortus infection in 5- to 6-month-old Merino lambs. Heritability for individual FEC was observed to range from 0.29 to 0.42 in Romneys facing natural mixed challenges on pasture, with the highest heritability recorded at 7e8 months of age. In, the current study, heritability of fecal egg count EPG was observed to be 0.107 at 8 months of age. Considerable high heritability of FEC indicates more additive genetic variance, which means considerable genetic improvement (in terms of genetic resistance against Haemonchus infection) is possible by applying higher selection pressure in resistant animals through mass selection. The details on heritability for disease resistance traits have already been discussed in Chapter 17. Repeatability Repeatability was measured for different generations and at different seasons for sheep. The repeatability (r  S.E.) of EPG among different seasons in first-, second-, and third-generation adult sheep has been estimated as 0.463  0.003, 0.50  0.003, and 0.260  0.019.

300

20. Breeding strategies for the development of a disease-resistant stock of livestock

Moderate to high repeatability for fecal egg count has been reported by most scientists. Repeatability (r  S.E.) of EPG in adult sheep of different generational parameters was observed to be 0.463  0.003, 0.501  0.003, and 0.260  0.019 at the first, second, and third generations. Thus the above result indicates that measuring single record is sufficient for selection. It was observed that the response to natural infection of H. contortus in lambs of the second and third generations was greater in the third month of life. Response to natural infection decreases subsequently with increased age and lowest in the eighth month of age. As age progresses, the immune mechanism becomes strong and sheep become more resistant to natural infection. Acquired resistance improves with age. A significant effect of sire as well as a dam was observed in lamb FEC. The progenies of resistant parents were found to have lower FEC compared with susceptible parents at all ages. EPG of lambs was recorded to be highly heritable and repeatable in all the generations. For the details of repeatability traits for disease resistance, please refer to Chapter 17.

Gene action (additive or nonadditive); nonadditive gene action includes heterosis, complementary effect, and maternal effect In the earlier section, we discussed the heritability of disease traits. Lower heritability was observed for disease traits, say for example mastitis, which indicates more nonadditive variance governing the traits. However, heritabilities for indicator traits were observed to be higher, indicative of more additive genetic variance.

Heterosis Nonadditive gene action is equally important for certain items that are commercially importance. We often come across the term “hybrid” or hybrid vigor as being of commercial importance. Heterosis, or hybrid vigor, was first proposed by Shull in 1914 to describe the phenomenon of a hybrid offspring with enhanced viability and developmental rates compared with its parents. Nowadays, commercially available poultry and fishes are claimed to be hybrid. There may be many causes for heterosis governed by nonadditive gene action. The first is the dominance model proposed by Bruce that emphasizes the relationship between dominant and recessive genes. The second is the overdominance model proposed by Shull and East that emphasizes heterozygosity. The third is epistasis. Additional mechanisms may explain heterosis in Arabidopsis and maize, including the expression of small RNA (siRNA) and DNA methylation. To examine these mechanisms, transcriptional or gene expression profiles are essential. The differential gene expression patterns in a cross combination and different sexes in hybrids and parents were assessed and correlated with economic traits, and were essential for evaluating the cause for heterosis. As heterosis is concerned with disease traits, it has been observed that heterozygotes are advantageous in increased fitness and less incidences for diseases. Conversely, in the case of inbreeding, genes responsible for causing diseases (or deleterious mutation) as present in homozygous condition cause disease. This may be the reason for inbreeding depression. Heterosis is based on the notion that genetic diversity increases the health (or fitness) and survival of an organism. Genetic diversity is defined by the Unified Medical Language System of the National Library of Medicine as “the phenotypic and genotypic differences among individuals in a population.” Genotype refers to the genes contained in a cell, while phenotype refers to how genes are expressed, such as in physical traits or abnormalities. The term “heterosis” is often used in genetics and selective breeding, in which desirable traits are bred into a species while undesirable traits are bred out. In breeding, heterosis refers to the idea that a hybrid (an animal or plant of mixed origin) has greater genetic strength than organisms of a homogenous (similar) background. It also refers to the potential to combine the positive traits of the parents into “better” offspring. In humans, heterosis refers to children of parents who do not share a blood line and thereby do not share genetic material. When offspring are considered better or more fit for survival than their parents, it is referred to as hybrid vigor. However, crossbred plants or animals are not always better than their parents. It is possible for a hybrid to be less fit for survival, which is called outbreeding depression. Inbreeding depression, on the other hand, represents the decrease in fitness as the result of the breeding of organisms of similar genetic backgrounds. As a result, rare genetic diseases or defects become more prevalent among populations as a result of increased heterozygosity. Inbreeding may lead to long-term effects on health. Inbreeding enhances the risk of late-stage disease states, such as obesity, heart disease, and type 2 diabetes. The opposite of a hybrid is a purebred. A purebred is the result of breeding by two organisms that have similar genetic material with no outbreeding (or breeding with those that have different genetic material) over many generations.

Evaluation of available genetic resource

301

There are two main hypotheses to explain the “fitness advantage” in heterosis: the overdominance hypothesis and the avoidance of deleterious recessive genes hypothesis. 1. Overdominance hypothesis: The overdominance hypothesis states that an organism that descends from parents of different genetic backgrounds will have greater resistance to a broader spectrum of potential dangers. On the other hand, an organism that descends from parents of similar genetic backgrounds will have resistance to a narrower spectrum of potential dangers. This concept is related to antibody diversity. Similar to genetic diversity, antibody diversity suggests that offspring of parents of different genetic backgrounds are more fit than those from parents of similar genetic backgrounds. This is because they have a greater ability to produce antibodies that can defend against a wider variety of pathogens or harmful substances, such as bacteria and viruses. Based on this hypothesis, these organisms are more fit due to their greater immunity. 2. Avoidance of deleterious recessive genes: In this hypothesis, an organism that descends from parents of different genetic backgrounds will have fewer harmful recessive genes. A person inherits genes from his or her parents. One copy of each gene is inherited from the mother and a second copy is inherited from the father. Each parent can only pass one copy of their genes on to the child, which is determined by chance. When both copies (alleles) of a gene are the same, a person is said to be homozygous for that gene. When different alleles of the gene are inherited from each male and female parent, the individual is said to be heterozygous for that particular gene. Recessive genetic conditions are caused by a mutation, or defect, in a gene. In order to inherit the condition, a person would have to receive two copies of the defective gene. If a person receives one copy of the defective gene and one normal copy, he or she will not have the condition and is known as a carrier. An organism with genetically dissimilar parents is more likely to have fewer recessive genes than an organism born from closely related parents. Therefore, its decreased number of recessive genes may lead to increased fitness. If overdominance is the main cause of greater fitness, then certain genes should be overly expressed in offspring of genetically dissimilar parents compared with offspring of closely related parents. If the main cause of fitness is the avoidance of recessive genes, fewer genes should be underexpressed in the offspring of genetically dissimilar parents when compared with their parents. In trying to understand the functional and evolutionary importance of heterosis, current research is exploring what determines the key steps in the creation of a hybrid organism. In the case of poultry and livestock, hybrids of two types of animals are crossed to bring about a better product. For instance, the process of heterosis is used to create poultry that only lays white eggs, animals that are more “meaty” at an earlier age, and animals that gain weight at a specific rate, making them marketable sooner. Heterosis may be applied to domestic animals, such as cats and dogs. It is generally well known that purebred dogs tend to have a higher incidence of specific health problems. On the other hand, hybrid or random-bred dogs (“mutts”) tend to suffer from fewer maladies. Humans: Some heterosis research in humans is focused on how inbreeding depression and the decreased fitness of offspring of parents from similar genetic backgrounds may contribute to long-term increases in chronic disease incidence. Heterosis research in humans is also examining how to improve disease resistance. Because heterozygotes (the offspring of parents with different genetic backgrounds) have a genetic advantage over homozygotes (offspring of parents with similar genetic backgrounds), researchers are looking at how this increased disease resistance can be applied to other populations. Heterozygosity is also responsible for the advantage that females have over males in terms of X-linked recessive disorders such as color blindness and hemophilia. If a condition is X-linked, the defective gene is located on the X chromosome, one of the sex chromosomes. In females who receive one defective copy and one normal copy of a gene for an X-linked recessive disorder, the normal copy can compensate for the defective copy. This is why many females who have one defective copy of a gene for an X-linked recessive disorder have mild or no symptoms of the disease. When males, who have one X chromosome, receive one defective copy of a gene for an X-linked recessive disorder, they do not have another copy to compensate. Implications In the past, heterosis has been manipulated for the purpose of eugenics. Eugenics is the science of improving the genetic composition of a population. It generally refers to humans and has been the source of much controversy and ethical debate. While the philosophical standpoint of eugenics is that it lessens human suffering by preventing the spread of negative genetic traits, it is generally regarding as violating human rights. Historically, proponents of eugenics went so far as to forcefully sterilize individuals thought to have such negative genetic traits.

302

20. Breeding strategies for the development of a disease-resistant stock of livestock

By applying heterosis, scientists can bring about more favorable traits in plants and animals used for human consumption while repressing less favorable ones. The other side of this is that breeding in desirable characteristics and breeding out undesirable ones contribute to a less diverse population. There are also implications for recessive genetic disorders that occur in closely related populations. Tay-Sachs disease, for example, is a recessive genetic condition common among certain groups, such as French Canadian and Ashkenazi Jewish populations. Couples from populations in which there is an increased risk of certain recessive genetic disorders may work with genetic counselors to evaluate the probability of having children with the disorder based on known risk factors. Counselors may also help prospective parents decide which testing methods are appropriate, how to interpret results, and whether to terminate an affected fetus. It is important that patients realize that genetic tests cannot guarantee accuracy; there is always a risk of terminating a healthy unborn child. Also, clinical studies have shown that women who have had abortions suffer an increased risk of anxiety, depression, and suicide and breast cancer.

Maternal effects In some cases, when the phenotypic effect of the offspring is also determined by the genotype of the mother or dam, apart from individual’s own genotype and the environmental effect. This is known as maternal effect. This may be due to supplying messenger RNA or proteins to the egg. Maternal effects can also be caused by the maternal environment independent of genotype, sometimes controlling the size, sex, or behavior of the offspring. Maternal environment may be in terms of providing uterine environment to the offspring. These adaptive maternal effects cause phenotypes of offspring that increase their fitness. Further, it introduces the concept of phenotypic plasticity, important for evolution. In this current topic we need to discuss how maternal effect affects the disease traits, its inheritance and how to formulate the breeding plan with the disease trait being governed or influenced by maternal effects. As we had already discussed earlier that disease traits are polygenic in inheritance, follow continuity or normal distribution, but phenotypically have discrete phenotype. For any formulation of breeding strategy, it is important to understand the mode of inheritance. This is essential in turn to determine the basis or method of selection accordingly. Maternal effect was observed to have important effect on population dynamics. Inheritance of maternal effects can be adaptive. The immunity of the offspring against parasitic infestation was observed to be influenced by maternal parasite exposure. Indirect maternal effects were observed on immunity mediated by other components of the maternal environment. The maternal effect was studied in an insectdvirus system on the immunity of offspring. Five different maternal resource levels and immunity in the offspring were studied against viral infestation. The studies were conducted by direct challenge with a virus, and indirectly by estimating a major component of the immune system. In both the cases, it was observed that offspring from mothers in poor environments are more resistant to parasites. This supports the importance of maternal effects on disease resistance. This maternal effect was observed to be mediated through indirect environmental factors with implications to both the ecological and evolutionary dynamics of hosteparasite interactions. It was observed that immune responses to microbial infections may cause neurodegenerative diseases. In this case study, Pseudomonas aeruginosa infection of Caenorhabditis elegans was observed to cause several neural changes that are hallmarks of neurodegeneration. Genetic screening was employed to identify genes controlling P. aeruginosa-induced neurodegeneration. mes-1 gene was identified, encoding a receptor tyrosine kinase-like protein. This protein is needed for unequal cell divisions in the early embryonic germ line. It was reported that only sterile mes-1 animals were observed to be resistant to neurodegeneration induced by P. aeruginosa infection. On the contrary, in the fertile mes-1 animals, resistance was not observed. Similar results were observed using animals carrying a mutation in the maternal effect gene pgl-1. This gene is essential for postembryonic germ line development and the germ line-deficient strains glp-1 and glp-4. FOXO transcription factor DAF-16 was observed to be responsible for resistance to P. aeruginosa-induced neurodegeneration in germ line-deficient strains. It was demonstrated that P. aeruginosa infection results in neurodegeneration phenotypes in C. elegans, controlled by the germ line in a cell-nonautonomous manner. Thus it was concluded that mutation in the maternal effect gene pgl-1 has a role in causing resistance to P. aeruginosa-induced neurodegenerative diseases to Caenorhabditis elegans, studied as a model organism in this case.

Breeding plan

303

Existing constraints in improvement The most important constraint in disease traits is to measure the trait. Since the traits are threshold traits, phenotypic expression is all or none. In the case of rare diseases, when the frequency of occurrence is rare, it becomes extremely difficult to measure the trait. Accordingly, other constraints pose difficulties in assessing inheritance, such as the heritability and repeatability of the trait. Most disease traits have low heritability, and hence response to selection or genetic gain becomes less. For any breeder, it is very important to select the animal for its disease resistance. Genetic selection becomes exceedingly difficult for traits with low frequency and those expressed later in life. Indicator traits with high genetic correlation serve as better alternative trait for selection or disease resistance.

Other factors Economic relevance of the disease It is important to assess the relative economic importance for the disease concerned. Mastitis is a disease with high relative economic importance. We need not to consider the disease traits with rare occurrence for selection. Similarly, it is also not necessary to consider the disease with less economic importance under selection criteria, say minor depletion of vision for dairy cattle in one eye. In dairy cattle, the trait has less economic importance, whereas in human it is of utmost importance.

Breeding objective Breeding objective gives an indication for improving disease resistance against a particular disease, which can be a single disease or more than one, and production criteria may be included or excluded. This is the second important point to be considered for formulating a breeding strategy after evaluation of animal genetic resource. In economic terms, breeding objective refers to the criteria that one attempts to improve genetically because of their influence on returns and the cost of production. Breeding objective considers the following points: 1. 2. 3. 4. 5. 6. 7.

Specification of the nature of diseasedinfectious/noninfectious How to measure the disease traits-natural infection/challenge study Whether the selection will be followed through direct selection or through indirect selection Determination of traits influencing economic return Economic value of each economic trait Choice of selection criteria depending on economic requirements and regional climatic conditions Choice of traits to be improved; whether to include a single disease or more than one disease; whether production parameters are to be included along with disease traits. In most cases, it has been observed that there exists a negative genetic correlation with production and disease traits. It is equally true for any breeding program that production traits cannot be compromised.

Breeding plan Choice of selection method To advocate the breeding plan for any disease trait, choice of selection method is important and depends on following points: (a) Number of important traits to be considered (b) Availability of the records (c) Economic value of the traits The selection methods are Independent culling level, tandem selection, and index selection. Index selection is the most effective method particularly for breeding for improvement of better disease resistance. In this case production and fitness traits need to be included along with the disease traits. We cannot compromise production in lieu of better disease-resistant animal. Low-producing animal, but resistant to diseases, are of no

304

20. Breeding strategies for the development of a disease-resistant stock of livestock

use in animal production. Fitness is equally important. Reproduction and viability should not be effected in lieu of better disease resistance. A practical example of selection index developed in cattle farm with beef production unit at the international level A selection index should be based on a balanced breeding objective. Selection Indexes allow us to make balanced selection decisions. In the selection index, individual economic weights are to be assigned to individual traits, based on their relative economic importance. Selection Indexes are calculated using the Breed Object software that has been developed by the Animal Genetics and Breeding Unit (AGBU) at the University of New England. BreedObject combines the BREEDPLAN estimated breeding values (EBVs) of an animal with an economic weighting on each individual trait (based on costs of production and returns on output) to produce a single selection index value for each animal. An animal’s selection index value can effectively be interpreted as its EBV for profitability in a particular commercial production scenario and market. Ranking seed stock animals (for example bulls in a catalog) on their selection index value sorts them based on their progeny’s expected profitability for the targeted production system. Selection index values are expressed as differences in “net profit per cow mated” and reflect differences in profitability across the entire production chain - from joining to slaughter. In indexes designed for self-replacing production systems (maternal), the long-term profit generated by the sire’s daughters is also included. A number of breed societies have generic, market-based breeding objectives and selection indexes available on their respective websites, allowing commercial breeders to search for bulls that fit within their target index specifications (see Tool 4.01). These indexes are a very good guide as the objectives for many enterprises will be similar and rank animals similarly. The breed-based indexes have been calculated from very good industry feedback on the costs, returns and trait performance levels of the production system and market being supplied. Most commercial producers would be well advised to start with one of the standard breed society indexes and modify their selection procedures using a process described in Procedure 4. When selecting bulls using dollar index values it is important to also consider the individual EBVs and your herd situation. This concept is covered in more detail in Procedure 3. If your production system and target market are substantially different to those for which breed society selection indexes are based there is scope to develop your own specific breeding objective and associated selection index using BreedObject software. An example of a breed society selection index* In this example for a Bos taurus breed, the genetic differences between animals in net profitability per cow joined are estimated. The breeding goal is a high fertility, self-replacing commercial herd selling feeder steers and heifers for the short fed domestic feedlot trade. Steers are assumed to be marketed at 445 kg live weight (245 kg HSCW and 10 mm P8 fat depth) at 15 months of age. Emphasis is placed on growth to 400 days and high carcass yield while maintaining fertility and marbling. The key economic traits that are important in this selection index are shown in Fig. 20.1. The different trait emphases reflect the underlying profit drivers in a commercial operation targeting the short fed domestic markets. Another practical example of selection index developed at National level by NDDB, which included mastitis as a trait. NDDB has initiated certain pilot project for disease control particularly for some diseases as FMD Pilot Scheme, Ooty (1982e85), Animal Disease Control Project (ADCP) for FMD ControldKerala (2004e09), Pilot Project on Brucella Control (2013e2018), Pilot Project on Mastitis Control (2014e16).

Basis of selection or criteria of selection The selection may be based on (a) (b) (c) (d)

individual record, pedigree record, progeny testing collateral relatives

Some of the diseases may be sex limited, like mastitis, which one of the most important disease worldwide. Hence the selection criteria will be different for both the sexes-male and female. Selection criteria also depend on the heritability of the disease trait. Individual selection will be most effective for the traits with high heritability. For any disease trait, it has been observed that the disease trait directly lowers

Breeding plan

20 bulls X

305

600 cows 70% conception rate

420 cows conceived 10% embryonic mortality 380 progenies born

Considering 1:1 sex ratio 190 male

190 female

Mortality upto 1 yr (6%) 179 female progeny Mortality 1 yr to completion of 1st lactation (3%)

174 female progeny 5% infertility rate 165 female progeny 70% conception rate 115 female progeny 10% embryonic mortality 104 female progeny in lactation, whose yield can be recorded

FIGURE 20.1 A schematic representation of progeny testing with 20 bulls in a herd.

heritability and is even difficult to measure. But the indicator traits have better heritability compared with direct traits. Let us understand the situation with a suitable example of Mastitis. If the directly quantify the incidences of mastitis, say clinical or subclinical, the heritability is usually low, but when we quantify the cases of mastitis with indicator trait as somatic cell score, heritability is much higher. The fact has already been described in Chapter 17. Now, mastitis being a sex-limited trait, individual selection is not possible in case of the bull. The alternative is pedigree selection (through EPD calculation) and Progeny testing. Basis of selection also depend on the stage of life when the disease is expressed. Selection would become less efficient and costly if it is expressed late in its life time. Individual selection or performance testing Selection is based on the individual performance. This criteria for selection are effective only when the trait is highly heritable. Selection based on pedigree records Selection based on pedigree records are done through Expected predicted difference (EPD) calculation from the record of dam, maternal grand dam, paternal grand dam. With the help of pedigree record, the main advantage is that the selection can be employed at day old stage of the livestock. It has a practical usefulness for sex-limited traits. EPD can be easily employed for assessing the breeding value for sex limited trait in case of male animals. Let us discuss with suitable examples from mastitis: The expected predicted difference (EPD) of the animals were collected from the records maintained. EPD is calculated based on the record of the dam and paternal grand dam, and is calculated as follows:

EPD ¼ 1=2 h2 M1 M
m þ 1=4 h2 P1 P
m

306

20. Breeding strategies for the development of a disease-resistant stock of livestock

Where. M1 ¼

n 1 þ ðn
1Þr

P1 ¼

p 1 þ ðp
1Þr

n ¼ No. of lactation. r ¼ Repeatability. M ¼ Average mastitis incidence of dams. m ¼ Population average.

p ¼ Number of lactations of paternal grand dam, and. P ¼ Average lactation yield of paternal grand dam. Selection based on progeny testingdfarm or field Progeny testing is the process of sire evaluation, when the genetic merit of the sire is judged based on the milk yield of female progeny. When the genetic merit of the progeny is judged based on the available records of daughter in the farm, it is known as Farm progeny testing. The major limitation is that the required numbers of daughters are not available. The major constraints are conception rate, embryonic mortality, sex ratio, mortality of the born female calf at different stages of life, as discussed in Fig. 20.1. In order to overcome the above constraints, Associated herd progeny testing was employed. Here, the dams are located at different herds situated at different farms. In order to obtain a large number of female offspring to assess the genetic merit of the sire, the best option is to inseminate the dams reared under farmers’ herd. This method is known as Field progeny testing. The only limitation is the recording of data at field level. As in case of farm, it is not possible to record data daily. Instead, test day data recording may be advocated. Since we are discussing about disease resistance, say in case of mastitis, somatic cell score may be assessed on test day basis. In case of production data, say milk production, data recording in field level at farmers herd may be advocated as “Test day milk production”. Suitable statistical techniques as Random regression model may be employed for assessing the total milk production per lactation. • Progeny testing program are existing for Frieswal (HF x Sahiwal, PDC, Meerut), Sahiwal (NDRI, Karnal), Murrah (NDRI, CIRB, IVRI etc). Field Progeny Testing may be summarized as follows: • • • • • •

In order to increase the no. of female animals to be inseminated and to obtain more no. of female progeny Accuracy of selection increases Test day milk yield per month Proper awareness, inceptives in terms of medicine, vaccine to be provided to farmers In field progeny testing, minimum 30e40 daughters per bull Murrah under network project on Buffalo in NDRI, Frieswal project in Project Directorate on cattle, Meerut and KLDB, Kerala. Last but not least, data availability is the rate-limiting factor affecting the basis of selection.

Choice of mating system The choice of mating system will be based on according to the use of genetic variability: (a) Methods for maximum use of additive genetic variance or resemblance between parents and offspring Selective breeding Inbreeding Line breeding Pure breeding Outcrossing with a different breed

307

Breeding plan

Synthetic line breeding Upgrading (b) Methods for maximizing the use of additive and nonadditive genetic variance or from heterosis and position effect: Crossing among inbred lines Synthetic line breeding Recurrent selection Reciprocal recurrent selection Simple commercial crossing Intercrossing among two or more breeds In case of some genetic diseases that occur due to increased homozygosity, leading to gene fixation with deleterious effect. The situation can be overcome by crossbreeding with nonadditive gene action.

Breeding organization Group breeding schemes as Open nucleus breeding system or closed nuclear breeding system are available. Open nucleus breeding system is preferable technique at village level. The advantages of this technique are that it is equally applicable for any herd size, population of livestock can remain with the farmers, small holding size. It can work efficiently without proper recording of disease traits at farmer’s herd. Although genetic gain is not at par with cross breeding, but this technique leads to genetic improvement in each and every livestock present in the village and ultimately leads to a sustainable growth (Fig. 20.2). Open nucleus breeding system Open nucleus breeding system (ONBS) is a highly sustainable breeding system in which each animal in the village comes under the development scheme. ONBS is applicable when there is ☛ no infrastructure for field PT ☛ an improper milk recording system ONBS in the nucleus breeding scheme when every individual has the opportunity for genetic improvement. Increased intensity of selection is recorded in the ONBS system. It decreases the generation interval. There is very reliable recording of data in nucleus herd. Difficult traits like feed conversion efficiency, reproductive efficiency and disease-resistant traits can also be assessed. Proper quarantine and disease prevention measures is to be advocated. The best germplasm can be disseminated through AI in this technique. Selections of males are done based on performance of collateral relatives. It may be summarized as • increased intensity of selection • decreased generation interval • very reliable recording of data in nucleus herd

Open Nucleus Breeding System Village 1

Village 2 Best female

Best male

Best male

Village 3

FIGURE 20.2

Best female

Best male

Nucleus herdbest female & best male

Best female

Best female Best male

Village 4

Schematic flow chart for open nucleus breeding system.

308

20. Breeding strategies for the development of a disease-resistant stock of livestock

FIGURE 20.3

Schematic representation of Open nucleus breeding system with MOET (multiple ovulation and embryo transfer).

• difficult traits like feed conversion efficiency, reproductive efficiency and disease-resistant traits can also be assessed • proper quarantine and disease prevention measures • the best germplasm can be disseminated through AI • selection of males is done based on the performance of collateral relatives

Multiple ovulation and embryo transfer In this technique, the female is superovulated from elite animals and transferred to foster mother. MOET when combined with selection can reduce the generation interval. Increased genetic gain is recorded with MOET. ONBS-MOET is a very popular group breeding scheme. It is regularly employed in NDDB (National Dairy Development Board) (Fig. 20.3).

Case study for ONBS for Garole sheep Flow chart to achieve the target Data collection for the mature ewe and ram regarding body weight and litter size  Maintaining pedigree record in selected blocks at the breeding tract of Garole sheep  Screening of top 20% of the best performing ram (based on own body weight and litter size of dam and daughter) and ewe (based on own body weight and litter size) and subjected to genotypic screening for FecB homozygous (FecBBB)  Maintaining the screened FecBBB rams and ewes in ‘Nucleus Herd’ and rest will be culled  Dissemination of superior germplasm (semen of the best ram) to the adjacent village; and incorporation of the best ewe (containing FecBBB) from the farmer’s herd as a continuous process  In the ‘Nucleus Herd’, best breeding ram/germplasm for dissemination will be produced by selective breeding of ‘elite ewes’ and ‘elite rams’

Case study for ONBS for Garole sheep

309

Work plan for genetic improvement of Garole sheep 1. Survey, data collection, and identification ➢ Door-to-door surveys for the collection of data regarding socioeconomic status of the farmers, number of Garole sheep reared, their sex, age, litter size, body weight, etc. (proforma for data sheet attached in Annexure). ➢ Detailed information must be collected regarding socioeconomic profiles of Garole farmers in all 19 blocks of SBR. The survey would record the community of Garole farmers, landholdings of farmers, primary professions of farmers in different zones, monthly income pattern, literacy level, etc. ➢ Screening of the top 20% of the ewe based on litter size and body weight ➢ Screening of the top 20% of the ram based on dam’s and daughter’s performances ➢ Identification/tagging of the best 20% ram and ewe (with proper recording) 2. Genotyping of the identified sheep ➢ Collection of blood sample from the identified rams and ewes ➢ DNA isolation by phenol chloroform extraction method ➢ PCR-RFLP study for FecB gene (Booroola fecundity gene) ➢ Homozygous FecBBB rams and ewes for Booroola fecundity gene will be identified from the farmer’s herd 3. Maintaining “nucleus herd” and strategic breeding (ONBS) ➢ Nucleus herds will be maintained in different pockets/locations (based on micro-agroclimatic area) for easy accessibility to the adjacent villages ➢ Establishment of one such “Nucleus Breeding Farm” at Institute level and managed by ARD Dept, Govt. of West Bengal and University (West Bengal University of Animal and Fishery Sciences, Kolkata) as per the case. They will scatter the superior germplasm throughout the country as per demand. ➢ Animals will be properly identified; management, health checkup, disease investigation, and primary treatment will be carried out under supervision of experts/Veterinary Officers ➢ Training and workshops for the SHGs with the help of National/State Animal Husbandry Experts, regarding scientific inputs in “Backyard Sheep Management” ➢ Maintaining the screened homozygous (FecBBB) rams and ewes for Booroola fecundity gene in nucleus herd and rest are advised for culling ➢ Lucrative price should be offered to the farmers for forced selling their rams and ewes screened for having inferior genotype. Incentives (as per practical situation) must be advocated for the early adopters. Later on, after one generation they may be supplied with a female lamb produced from “Nucleus Herd” in subsidized rate or free of cost as per as permissible. Incentives in other forms based on practicality may be advocated for smooth operations to achieve the target. ➢ In the “Nucleus Herd,” breeding of best rams and best ewes will be carried out. ➢ The male and female lambs produced in the “Nucleus Herd” will be reared, vaccinated, offered balanced nutrition including supplementation of area specific vitamin-mineral mixture. ➢ “Flushing” of ewes will be adopted just prior to their parturition to obtain expression of their highest genetic merit. ➢ In the first year of the project, identified best rams will be supplied to the adjacent village areas for breeding purpose @ 1:8 ➢ Other existing rams in the village will be castrated and used for mutton purpose. ➢ Farmers will be encouraged for selling those castrated ram in the market, otherwise the castrated rams after marketable age may be procured by ARD, Govt. of West Bengal for selling through LPDC. ➢ The rams born and brought up in the “Nucleus Herd” will be supplied later on, to the adjacent villages for breeding and other rams prevailing in the area will be castrated/slaughtered before they arrive at a breedable age. ➢ Some of the surplus ewes will be supplied to the farmers/beneficiaries in subsidized rate/free of cost. Preference will be given to them who had donated the animals while constructing nucleus herd. 4. Establishment of elite “Nucleus Herd” (comprising of elite rams and ewes) near to University (W.B.U.A.F.S., Kolkata) and regulated by ARD, Govt. of West Bengal Main motto of this farm will be to maintain the parent stock of elite rams and ewes to meet the demand throughout the country. University will carry out regular research activities in different aspects of this important genetic resource including disease resistance, formation of different variants/strains of economic importance, production of value-added products, etc.

310

20. Breeding strategies for the development of a disease-resistant stock of livestock

5. Integrated survey in the village ➢ Another survey needs to be conducted for a second phase as before and top 20% of the best rams and ewes will be screened. Homozygous FecBBB rams and ewes for Booroola fecundity gene will be identified from the farmer’s herd. ➢ Identified and genotyped rams and ewes will be incorporated in the “Nucleus Herd” ➢ Present survey will judge cost-economics including socio-economic upliftment of farmers/beneficiaries • It is a continuous process for 5 years. • Castration of ram in farmer’s herd will be carried out and best rams will be supplied from Nucleus herd. • The rams produced from farmer’s herd with the ewe supplied from the “Nucleus Herd” will be screened for Homozygous FecBBB. It is expected that they will be homozygous and accordingly incorporated in the “Nucleus Herd.”

Annexure: Tentative Parameters for Survey Address Village/ Socioeconomic Sl. Name of block/ condition of No. the owner district the farmer

Litter size No. of Garole of dam Litter size sheep in Litter size in (mother) of daughter Body weight Heart girth the family Sex Age case of ewe of ram of ram of the sheep of the sheep

6. Use of biotechnological approaches: AI, MOET, MAS, Embryo transfer, andembryo splitting, transgenesis These topics will be dealt in detail in the subsequent chapters. Actual breeding strategy followed: 1. 2. 3. 4. 5. 6. 7. 8. 9.

Survey and population size Choice of breed Source of germplasm-AI (frozen semen)and natural service Availability of trained manpower/technical power for disease detection/quantification for indicator traits for diseases Record maintenance Criteria for selecting male and female and method of selection Mating system Evaluation of EPA, EPD Use of MAS, ETT, MOET, ONBS, with or without MOET, field PT, transgenesis. This is a classical example in sheep where genetic improvement for litter size is considered. But in the current context, we are discussing about the disease resistance traits. Gastrointestinal verminosis as Haemonchosis is very common in sheep. In case the genetic improvement of sheep for disease resistance against Haemonchosis is desired, instead of litter size, Faecal egg count will be the important trait under consideration. Genomic trait as immune response genes will be considered instead of Booroola fecundity gene.

A classical example of disease resistance with epidemiological model: a case study Bovine tuberculosis (bTB) is a challenging disease caused by Mycobacterium tuberculosis. There exists genetic variation in host resistance to tuberculosis. Hence, there is an ample scope for the genetic improvement of host resistance to Mycobacterium bovis infection. In the United Kingdom (UK), genetic evaluation for resistance to infection in dairy cattle have been included in breeding program. Programs have been undertaken for genetic selection against bTB. The objective of this study was to assess the impact of genetic selection for bovine tuberculosis resistance on cattleto-cattle disease transmission dynamics and prevalence by developing a stochastic genetic epidemiological model. It was employed for genetic selection in a simulated cattle population. Various levels of selection intensity over 20 generations were considered. Assumptions were genetic heterogeneity in host resistance to infection. The current model considers the dairy cattle population structure and present bTB control strategies in the UK, and was informed by genetic and epidemiological parameters inferred from data collected from UK bTB infected dairy herds. The risk of a bTB breakdown was modeled as the percentage of herds where initially infected cows (index cases) caused secondary cases by infecting herd-mates. The model predicted that the risk would be reduced by 50% after 4, 6, 9, and 15 generations for selection intensities. This corresponds to the genetic selection of the 10%, 25%, 50%, and 70% most resistant sires, respectively. In the herds with the occurrence of bTB breakdowns, genetic

Breeding strategy for infectious disease

311

selection reduced the severity of breakdowns over generations as assessed by the reduction of both the percentage of secondary cases and the duration over which new secondary cases were detected. Selection of the 10%, 25%, 50%, and 70% of most resistant sires caused the reduction of the percentage of secondary cases to <1% in 4, 5, 7, and 11 generations, respectively. Similarly, the proportion of long breakdowns (breakdowns in which secondary cases were detected for more than 365 days) were observed to be reduced by half in 2, 3, and 4 generations, respectively. From the following observations, it was suggested that genetic selection could be a viable tool to complement existing management and surveillance methods for the control and finally the eradication of bTB. A stochastic within-herd bTB transmission model was developed to simulate bTB spread in each herd and give estimates of severity and duration of bTB breakdowns.

Breeding strategy for infectious disease Bacterial disease Mastitis Mastitis is defined as a disease involved in inflammation of the udder. Mastitis in cows is reflected as a binary trait, reflecting presence or absence of clinical mastitis (CM), or as a count variable, number of mastitis cases (NCM), within a defined time interval. Different models have been proposed for genetic analyses of mastitis. The objective was to evaluate the predictive ability and sire predictions of a set of models for genetic evaluation of clinical mastitis or number of mastitis cases. Linear and threshold liability models for clinical mastitis, and linear, censored ordinal threshold, and zero-inflated Poisson (ZIP) models for number of mastitis cases were compared in a crossvalidation study. To assess the ability of these models to predict future data, records from 620,492 first-lactation Norwegian Red cows, which were daughters of 3064 sires, were evaluated in a fourfold cross-validation scheme. The mean squared error of prediction was used for model comparison. Ordinal threshold model was observed to be promising when comparing the overall predictive ability. This result was on average, across sick and healthy cows; however, the models have different value for each category of animals. Healthy cows were predicted better by the threshold and linear models for binary data and ZIP model, whereas for mastitic cows, the ordinal threshold model was by far the best model. Predicted sire effects and rankings of sires were highly correlated across all models. For practical purposes, the linear models are very competitive with the nonlinear models. Case study on genetic analysis on pathogen specific mastitis The present study was aimed to investigate the presence of genetic variation for susceptibility to pathogenspecific mastitis and to detect haplotypes of an identified quantitative trait locus with effect on unspecific mastitis resistance, with effects on specific mastitis pathogens. Bacteriological data on mastitis pathogens were obtained from the diagnostic laboratory at the Swedish National Veterinary Institute. The data mostly consists of cases of subclinical mastitis but cases on clinical mastitis were also included. Variance components were estimated for incidence of the six most frequent pathogens using Markov Chain Monte Carlo methodology via Gibbs sampling. Genetic variation for susceptibility to pathogen-specific mastitis was observed to be higher in comparison to estimates of general resistance to clinical mastitis in most other studies. Care should be undertaken to handle the nonrandom nature of data, while comparing with other studies. The effect of haplotype on the risk of being infected by a given mastitis pathogen, in relation to other pathogens, was examined using an allele substitution model. Although there were no significant haplotype substitution effects on the resistance to any of the six mastitis pathogens, there was a significant difference between the effects of two of the haplotypes regarding the risk of acquiring a Streptococcus dysgalactiae infection. Disease traits in German Holstein cows Various health problems in dairy cows have been related to the magnitude and duration of the energy deficit post partum. Disease frequencies were observed to be 24.6% for mastitis, 9.7% for metabolic disorders and 28.2% for claw and leg diseases. Heritabilities were estimated to be 0.06, 0.30, and 0.34 for energy balance, fat/protein ratio, and body condition score, respectively. For the disease traits, heritability ranged between 0.04 and 0.15. It is suggested from the results that an improvement of overall health is possible if energy balance traits are included into future breeding programs. A low fat/protein ratio might serve is indicative for metabolic stability and health of claw and legs. Between body condition and mastitis, a significant negative correlation of
0.40 was observed.

312

20. Breeding strategies for the development of a disease-resistant stock of livestock

The outcome is a new insight into the role energy balance traits can play as auxiliary traits for robustness of dairy cows. It was concluded that both, fat/protein ratio and body condition score, are potential variables to explain the adaptability of cows to the challenge of early lactation. The genetic parameters need to be estimated on a more comprehensive data set as “days in milk.” Average daily energy balance, milk fat/protein ratio, and body condition score were observed to be important traits. Energy balance indicator traits like fat/protein ratio in milk and body condition score are equally useful traits to be used in selection programs to help predicting breeding values for health traits. Therefore, genetic correlations among energy balance, fat/protein ratio, and body condition score, and mastitis, claw and leg diseases, and metabolic disorders were estimated using linear and threshold models on data, which serve as very useful indicator. Interpreting and analyzing field data In order to obtain sufficient data to quantify genetic variation in resistance and to perform genomic studies, it is often necessary to use field data. While such data can be extremely informative, and natural disease outbreaks can provide data cost-effectively, there are a number of sources of environmental noise that potentially mask the genetic signal. These include incomplete exposure to infection, imperfect diagnostic tests and variable infection pressures over time and between environments. These influences will all tend to reduce heritability and the power to detect SNP associations. A broad summary of the main issues is given here. Incomplete exposure to infection results in some animals not having the opportunity to express their resistance genotype. Therefore, uninfected animals will comprise individuals that may truly be resistant (at the level of challenge they have encountered) or animals that have yet to be exposed to an infectious dose of pathogen. Assuming that it is not possible to distinguish between these two categories of animals, incomplete exposure biases both estimated SNP effects and heritability downwards, with the former reduced by a factor ε, where ε is the proportion of the population exposed to the infection, assuming that exposure is an all or none event. Furthermore, ε will change continuously during an epidemic, and accounting for epidemic dynamics while estimating quantitative genetic parameters remains computationally challenging. Collection of field data requires diagnosis of the infection (or disease) state of an animal, with all diagnostic tests being described by the concepts of specificity and sensitivity. Specificity (Sp) is the probability that a truly healthy individual is confirmed as healthy through suitable diagnostic procedures and sensitivity (Se) is the probability that a truly diseased individual is concfirmed and classified as diseased by suitable diagnostic procedures. This parameterization of the 2  2 classification of true status and diagnosed status is universal in epidemiological theory, rather than the alternative classification of true and false test outcomes. If either Sp or Se is less than one, then observed prevalence (p0 ) will differ from true prevalence according to the following regression: p0 ¼ (1
Sp) þ (Sp þ Se
1)p. As with incomplete exposure, imperfect diagnosis will reduce both heritability and estimated SNP effects, with the SNP effect biased downwards by the factor (Sp þ Se
1). The consequences of incomplete exposure to infection and imperfect diagnosis are simply that genetic signals get diluted and the power to quantify genetic effects is reduced. These factors probably lie behind the commonly held belief that disease-resistance traits are lowly heritable, an observation that flies in the face of the near-ubiquitous variation seen in immune-related genes and in immune responses. Therefore, the identification of a genetic signal for resistance under field conditions most likely indicates an underlying genetic control that is much stronger. Consequently, the opportunities for studying genetic resistance to disease, and even selecting for increased resistance, may be somewhat greater than is apparent from low observed heritability. The considerations so far have considered only “static” datadi.e., they have ignored the time-dependent changes in infection pressure. The impacts of variable infection pressures on genetic parameter estimation are complex and have yet to be fully elucidated. They will vary according to whether infection pressure is presumed to be “constant” but different in different circumstances/environments or whether it varies dynamically during an epidemic. An example of the former is discussed below in the salmon disease case study. The latter case is addressed with analytical properties of the estimates of genetic effects. Appropriate experimental designs The ideal experimental design for detecting genetic variation in resistance and for identifying SNP associations or developing genomic predictors of resistance would exploit continuously varying phenotypes measured on animals subjected to identical environmental and challenge conditions. Such circumstances, where challenge conditions are deliberate and standardized, are rare and are most likely to be feasible for studies in fish, where several examples do exist or chickens. Under some circumstances it has been possible to achieve this situation

References

313

for terrestrial mammals, e.g., the PRRSV challenge experiments described, however these studies generally require large-scale funding coordinated between several partners and hence are relatively uncommon. However, even in the context of challenge tests there is now a recognition that typical studies address genetic variation in susceptibility to infection but do not address the possible existence of genetic variation in infectiousness. Infectiousness has an indirect genetic effect on the population and, unlike competition effects among animals, it is dynamic. Its detection depends primarily on detecting variance in the speed of epidemic development among groups. This is both demanding of data, requiring “replicated” epidemics and computationally demanding. The novelty of this area means that optimum challenge designs for detecting such variation have yet to be defined. However, initial results are given by who present analytical solutions for prediction accuracy in the case of major gene effects and propose novel Bayesian inference approaches for estimating such effects. In summary, unless infectiousness can be measured directly, it is likely to pose estimation problems. Disease-resistance studies more commonly use opportunistic “harvesting” of data either from epidemics, such as bovine tuberculosis (bTB) outbreaks or from endemic diseases such as mastitis or nematode infections. Even in this situation distinct differences are seen between endemic and epidemic diseases. For the two endemic diseases mentioned, phenotypes can be captured by measurements which show continuous variation, whereas for the epidemic diseases the phenotype is more often a binary variable, i.e., infected/diseased or not. One of the issues faced when using data from an epidemic, particularly when the outcome is a binary variable (affected or not), is the choice of animals to include in the dataset, and hence to genotype. Ideally, one would sample all animals from a cohort, or take a random sample with affected and unaffected animals sampled in proportion to the disease prevalence. However, there are several factors to consider when making such decisions. Firstly, if prevalence is low, then sampling many unaffected or control animals can be perceived as wasteful of resources when compared with standard case-control designs which maximize the power of a contrast. However, case-control designs make estimation of, or correction for, nongenetic factors difficult as the sampling has been nonrandom and, hence, the effects of both genetic and nongenetic factors will be incorrectly estimated. Secondly, definition of control animals may be problematic, especially if exposure to infection is unknown or if diagnostic test sensitivity is low. In either case animals will be misclassified and, combining the two concepts, the downward bias in estimated SNP effects will be ε(Sp þ Se
1). The problem of control definition has often been avoided in human genetics studies using the so-called Wellcome Trust design, in which cases are compared against a reference population average sample. In cases where disease prevalence is low, or diagnostic test sensitivity (or ascertainment of cases) poor or exposure probabilities low, then true controls are unlikely to differ greatly from a random sample from the population, and the two experimental designs converge. The Wellcome Trust design may also be advantageous in situations where large numbers of “population average” animals (for the trait of interest) have already been genotyped, an obvious example being the large numbers of Holstein dairy animals genotyped as part of genomic selection programs. But it is appreciated that apart from the case of Holstein cattle, the subpopulation structure often seen in livestock will make it difficult to define appropriate “population average” animals. However, in cases where true controls (i.e., uninfected animals that have been exposed to an infectious dose of pathogen) can be defined with some accuracy, alternative experimental designs have been proposed that may have greater power. Quite simply, animals could be sampled to maximize their expected genetic differences in resistance to the disease. Therefore, cases could be preferentially sampled from cohorts with a low force of infection (therefore more susceptible) and controls preferentially sampled from cohorts with a high force of infection (therefore more resistant). However, the properties of this design are unknown, and potentially it creates risks in terms of unobserved risk factors and hidden genetic structure, and research is required to quantify the balance between extra power and greater risk of unknown factors biasing the results.

References Emanuelson, U., Danell, B., Philipsson, J., 1988. Genetic parameters for clinical mastitis, somatic cell counts, and milk production estimated by multiple-trait restricted maximum likelihood. J. Dairy Sci. 71 (2), 467e476. Pal, A., Chatterjee, P.N., Das, S., Batobyal, S., Biswas, P., Sharma, A., 2017. Biodiversity among sheep and goat reared under different agroclimatic regions of West Bengal, India. Indian J. Anim. Sci. 87 (1), 80e86. Santos, L.V., Brugemann, K., Bru¨gemann, K., Ebinghaus, A., Ko¨nig, S., 2018. Genetic parameters for longitudinal behavior and health indicator traits generated in automatic milking systems. Arch. Anim. Breed. 61, 161e171. https://doi.org/10.5194/aab-61-161-2018. Shook, G.E., Schutz, M.M., 1994. Selection on somatic cell score to improve resistance to mastitis in the United States. J. Dairy Sci. 77 (2), 648e658. Williams, Y.J., Doyle, P.T., Egan, A.R., Stockdale, C.R., 2005. Increasing the intake of highly digestible persian clover herbage reduces rumen fluid pH and the rate of degradation of neutral detergent fibre in grazing dairy cows. Aust. J. Exp. Agric. 45 (12), 1529e1537.

314

20. Breeding strategies for the development of a disease-resistant stock of livestock

Further reading Albers, G.A.A., Gray, G.D., Piper, L.R., Barker, J.S.F., Le Jambre, L.F., Barger, I.A., 1987. The genetics of resistance and resilience to Haemonchus contortus infection in young Merino sheep. Int. J. Parasitol. 17, 1355e1363. Baker, R.L., Mwamachi, D.M., Audho, J.O., Aduda, E.O., Thorpe, W., 1999. Genetic resistance to gastro-intestinal nematode parasites in Red Massai, Dorper and Red Maasai  Droper ewes in sub-humid tropics. Anim. Sci. 69, 335e334. Baker, R.L., Watson, T.G., Bisset, S.A., Vlassoff, A., Douch, P.G.C., 1991. Breeding sheep in New Zealand for resistance to internal parasites: research results and commercial application. In: Gray, G.D., Woolaston, R.R. (Eds.), Breeding for Disease Resistance in Sheep. Australian Wool Corporation, Melbourne, pp. 228e241. Bermingham, M.L., Woolliams, J.A., Skuce, R.A., Allen, A.R., McDowell, S.W.J., McBride, S.H., Bishop, S.C., Glass, E.J., 2014. Genome-wide association study identifies novel loci associated with resistance to bovine tuberculosis. Heredity 112, 543e551. Bijma, P., 2010. Estimating indirect genetic effects: precision of estimates and optimum designs. Genetics 186, 1013e1028. Bishop, S.C., 2010. Disease resistance: genetics. In: Pond, W.G., Bell, A.W. (Eds.), Encyclopedia of Animal Science. Marcel Dekker, Inc., New York, pp. 288e290. Bishop, S.C., 2012. A consideration of resistance and tolerance for ruminant nematode infections. Front. Livest. Genomics. 3, 168 [PMC free article] [PubMed]. Bishop, S.C., Doeschl-Wilson, A.B., Woolliams, J.A., 2012. Uses and implications of field disease data for livestock genomic and genetics studies. Front. Livest. Genomics. 3, 114. Bishop, S.C., Stear, M.J., 2001. Inheritance of faecal egg counts during early lactation in Scottish Blackface ewes facing mixed, natural nematode infections. Anim. Sci. 73, 389e395. Bishop, S.C., Stear, M.J., 2003. Modeling of host genetics and resistance to infectious diseases: understanding and controlling nematode infections. Vet. Parasitol. 115, 147e166 [PubMed]. Bishop, S.C., Woolliams, J.A., 2010. On the genetic interpretation of disease data. PLoS One 5, e8940 [PubMed]. Bishop, S.C., Woolliams, J.A., 2010b. Understanding field disease data. In: Proceedings of the 9th World Congress on Genetics Applied to Livestock Production, Leipzig, Germany, August 1e6, 2010. Bishop, S.C., Woolliams, J.A., 2014. Genomics and disease resistance studies in livestock. Livest. Sci. 166, 190e198. https://doi.org/10.1016/ j.livsci.2014.04.034. Bisset, S.A., Vlassof, A., Morris, C.A., Sothey, B.R., Baker, R.L., Parker, A.G.H., 1992. Heritability of and genetic correlations among fecal egg counts and productivity traits in Romney sheep. N. J. Agric. Res 35, 51e58. Boddicker, N.J., Bjorkquist, A., Rowland, R.R., Lunney, J.K., Reecy, J.M., Dekkers, J.C.M., 2014. Genome-wide association and genomic prediction for host response to porcine reproductive and respiratory syndrome virus infection. Genet. Sel. Evol 46 (1), 18. Boddicker, N., Waide, E.H., Rowland, R.R.R., Lunney, J.K., Garrick, D.J., Reecy, J.M., Dekkers, J.C.M., 2012. Evidence for a major QTL associated with host response to porcine reproductive and respiratory syndrome virus challenge. J. Anim. Sci. 90, 1733e1746. Boots, M., Roberts, K.E., 2012 Oct 7. Maternal effects in disease resistance: poor maternal environment increases offspring resistance to an insect virus. Proc. Biol. Sci 279 (1744), 4009e4014. Browning, B.L., Browning, S.R., 2008. Haplotypic analysis of wellcome trust case control consortium data. Hum. Genet. 123, 273e280. Dunkelberger, J.R., 2017. The Role of Host Genetics in Susceptibility to Viral Disease in Pigs. Graduate Theses and Dissertations, p. 15297. https:// lib.dr.iastate.edu/etd/15297. Fievet, J.B., Nidelet, T., Dillmannand, de Vienne, D., 2018. Heterosis is a systemic property emerging from non-linear genotype-phenotype relationships: evidence from in Vitro genetics and computer simulations. Front. Genet. https://doi.org/10.3389/fgene.2018.00159. Gheyas, A.A., Houston, R.D., Mota-Velasco, J.C., Guy, D.R., Tinch, A.E., Haley, C.S., Woolliams, J.A., 2010. Segregation of infectious pancreatic necrosis resistance QTL in the early life cycle of Atlantic Salmon (Salmo salar) Anim. For. Genet. 41, 531e536. Groszmann, M., et al., 2011. Changes in 24-nt siRNA levels in Arabidopsis hybrids suggest an epigenetic contribution to hybrid vigor. P. Natl Acad. Sci. U.S.A 108, 2617e2622. Hayward, A.D., Nussey, D.H., Wilson, A.J., Berenos, C., Pilkington, J.G., Watt, K.A., et al., 2014. Natural selection on individual variation in tolerance of gastrointestinal nematode infection. PLoS Biol. 12 (7), e1001917. https://doi.org/10.1371/journal.pbio.1001917. Holmberg, M., Fikse, W.F., Andersson-Eklund, L., Artursson, K., Lunde´n, A., 2012. Genetic analyses of pathogen-specific mastitis. J. Anim. Breed. Genet. 129 (2), 129e137. Hu, L.P., Huang, X.T., Sun, Y., Mao, J.X., Wang, S., Wang, C.D., Bao, Z.M., 2015. Molecular genetic analysis of heterosis in interspecific hybrids of Argopecten purpuratus x A. irradians irradians. Genet. Mol. Res. 14 (3), 10692e10704. https://doi.org/10.4238/2015.September.9.9. Larruskain, A., Jugo, B.M., August 19, 2013. Retroviral infections in sheep and goats: small ruminant lentiviruses and host interaction. Viruses 5 (8), 2043e2061. https://doi.org/10.3390/v5082043. Lippman, Z.B., Zamir, D., 2007. Heterosis: revisiting the magic. Trends Genet. 23 (2), 60e66. Lipschutz-Powell, D., Woolliams, J.A., Bijma, P., Doeschl-Wilson, A.B., 2012. Indirect genetic effects and the spread of infectious disease: are we capturing the full heritable variation underlying disease prevalence? PLoS One 7, e39551. Lough, G., Kyriazakis, I., Bergmann, S., Lengeling, A., Doeschl-Wilson, A.B., November 22, 2015. Health trajectories reveal the dynamic contributions of host genetic resistance and tolerance to infection outcome. Proc. Biol. Sci. 282 (1819), 20152151. https://doi.org/10.1098/ rspb.2015.2151. Maisie, E., Dawes Lisa, M., Griggs Ellen, W., Collisson Worthie, E., Yvonne Drechsler, B., 2014. Dramatic differences in the response of macrophages from B2 and B19 MHC-defined haplotypes to interferon gamma and olyinosinic:polycytidylic acid stimulation. Poultry Sci. 93 (4), 830e838. https://doi.org/10.3382/ps.2013-03511. Melchinger, A.E., Utz, H.F., Piepho, H.P., et al., 2007. The role of epistasis in the manifestation of heterosis: a systems-oriented approach. Genetics 177 (3), 1815e1825. Mitchell, S.E., Read, A.F., 2005. Poor maternal environment enhances offspring disease resistance in an invertebrate. Proc. R. Soc. B . 272, 2601e2607. https://doi.org/10.1098/rspb.2005.3253.

Further reading

315

Morris, C.A., Vlassof, A., Bisset, S.A., Baker, R.L., West, C.J., Hurford, A.P., 1997. Responses of Romney sheep to selection for resistance or susceptibility to nematode infection. Anim. Sci. 64, 319e329. Moskwa, B., Doligalska, M., Cabaj, M., 1998. The repeatability of haematological and parasitological parameters in polish Wrzosowka hoggets naturally infected with trichostrogylid nematodes. Acta Parasitol. 43, 148e153. Pinard-van der Laan, M.H., Bed‫׳‬hom, B., Coville, J.L., Pitel, F., Feve, K., Leroux, S., Legros, H., Thomas, A., Gourichon, D., Repe´rant, J.M., Paul Rault, P., 2009. Microsatellite mapping of QTLs affecting resistance to coccidiosis (Eimeria tenella) in a FayoumiWhite Leghorn cross. BMC Genomics 10, 31. Pooley, C.M., Bishop, S.C., Marion, G., 2014. Estimation of Single Locus Effects on Susceptibility, Infectivity and Recovery Rates in an Epidemic Using Temporal Data. Proceedings of the 10th World Congress on Genetics Applied to Livestock Production. Vancouver, Canada. August 17e22, 2014. Raphaka, K., Sa´nchez-Molano, E., Tsairidou, S., Anacleto, O., Glass, E.J., Woolliams, J.A., Doeschl-Wilson, A., Banos, G., 2018. Impact of genetic selection for increased cattle resistance to bovine tuberculosis on disease transmission dynamics. Front. Vet. Sci. 5, 237. https://doi.org/ 10.3389/fvets.2018.00237. Roy, M., Senapati, P.K., Ch Pakhira, M., Sarkar, U., Datta, S., 2018. Studies of inheritance of resistance to Haemonchus contortus through diallel mating in Garole sheep. J. Entomol. Zool. Stud. 6 (1), 569e573. Shen, H.S., et al., 2012. Genome-wide analysis of DNA methylation and gene expression changes in two Arabidopsis ecotypes and their reciprocal hybrids. Plant Cell 24, 875e892. Shull, G.H., 1948. What is heterosis. Genetics 33, 439e446. Vanimisetti, H.B., 2003. Genetics of Resistance to Haemonchus contortus Infections in Sheep. Master of Science in Animal Science. Thesis submitted to the Faculty of the Virginia Polytechnic Institute and State University, Blacksburg, Virginia. Vazquez, A.I., Perez-Cabal, M.A., Heringstad, B., Rodrigues-Motta, M., Rosa, G.J.M., Gianola, D., Weigel, K.A., 2011. Predictive ability of alternative models for genetic analysis of clinical mastitis. J. Anim. Breed. Genet. 129 (2), 120e128. Wakchaure, R., Ganguly, S., Praveen, P.K., Sharma, S., Kumar, A., Mahajan, T., Qadri, K., 2015. Importance of heterosis in animals: a review. Int. J. Adv. Eng. Technol. Innov. Sci. 1 (1), 1e5. Wang, H., Fang, Y., Wang, L., Zhu, W., Ji, H., Wang, H., Xu, S., Sima, Y., 2015a. Heterosis and differential gene expression in hybrids and parents in Bombyx mori by digital gene expression profiling. Sci. Rep. 5, 8750. Wang, H., Fang, Y., Wang, L., Zhu, W., Ji, H., Wang, H., Xu, S., Sima, Y., 2015b. Heterosis and differential gene expression in hybrids and parents in Bombyx mori by digital gene expression profiling. Sci. Rep. 5, 8750. https://doi.org/10.1038/srep08750. Woolaston, R.R., Piper, L.R., 1996. Selection of Merino sheep for resistance to Haemonchus contortus: genetic variation. Anim. Sci. 62, 451e460. Woolaston, R.R., Windon, R.G., 2001. Selection of sheep for response to Trichostrongylus colubriformis larvae: genetic parameters. Anim. Sci. 73, 41e48. Wu, Q., Cao, X., Yan, D., Wang, D., Aballay, A., 2015. Genetic screen reveals link between the maternal effect sterile gene mes-1 and Pseudomonas aeruginosa-induced neurodegeneration in Caenorhabditis elegans. J. Biol. Chem. 290 (49), 29231e29239. Yu, S.B., et al., 1997. Importance of epistasis as the genetic basis of heterosis in an elite rice hybrid. P. Natl Acad. Sci. U.S.A 94, 9226e9231.