Selective breeding: The future of TB management in African buffalo?

Acta Tropica 149 (2015) 38–44

Contents lists available at ScienceDirect

Acta Tropica journal homepage: www.elsevier.com/locate/actatropica

Selective breeding: The future of TB management in African buffalo? N. le Roex ∗ , C.M. Berrington, E.G. Hoal, P.D. van Helden Stellenbosch University, DST/NRF Centre of Excellence for Biomedical Tuberculosis Research/ Medical Research Council (MRC) Centre for TB Research, Division of Molecular Biology and Human Genetics, Faculty of Medicine and Health Sciences, Tygerberg, South Africa

a r t i c l e

i n f o

Article history: Received 3 February 2015 Received in revised form 11 May 2015 Accepted 14 May 2015 Available online 15 May 2015 Keywords: Breeding BTB Buffalo Genetic improvement Mycobacterium bovis Resistance Selection

a b s t r a c t The high prevalence of bovine tuberculosis (BTB) in African buffalo (Syncerus caffer) in regions of southern African has a negative economic impact on the trade of animals and animal products, represents an ecological threat to biodiversity, and poses a health risk to local communities through the wildlifecattle-human interface. Test and cull methods may not be logistically feasible in many free-range wildlife systems, and with the presence of co-existing BTB hosts and the limited effectiveness of the BCG vaccine in buffalo, there is a need for alternative methods of BTB management. Selective breeding for increased resistance to BTB in buffalo may be a viable method of BTB management in the future, particularly if genetic information can be incorporated into these schemes. To explore this possibility, we discuss the different strategies that can be employed in selective breeding programmes, and consider the implementation of genetic improvement schemes. We reflect on the suitability of applying this strategy for enhanced BTB resistance in African buffalo, and address the challenges of this approach that must be taken into account. Conclusions and the implications for management are presented. © 2015 Elsevier B.V. All rights reserved.

Contents 1. 2.

3. 4. 5. 6. 7.

Resistance vs tolerance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 Types of selective breeding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39 2.1. Phenotypic selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.2. Marker-assisted selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 2.3. Genomic selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Implementing genetic improvement strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 Application of selective breeding in African buffalo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Challenges and considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 41 Conclusions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Management implications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 42 Conflict of interest . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

The African buffalo is an ecologically important species in the savannah ecosystem; their physical size and large numbers makes them a considerable proportion of the prey biomass of lions, and their role as coarse grazers has implications for other fauna and flora (Prins, 1996). Designated one of the ‘Big Five’ most dangerous African animals, buffalo are a species of great economic significance in South Africa. They play an important role in the tourism indus-

∗ Corresponding author at: Stellenbosch University, PO Box 19063, Tygerberg 7505 South Africa. Tel.: +27 21 938 9694. E-mail address: [email protected] (N. le Roex). http://dx.doi.org/10.1016/j.actatropica.2015.05.012 0001-706X/© 2015 Elsevier B.V. All rights reserved.

try, and are one of the most sought-after game trophies (Lindsey et al., 2007). However, buffalo act as maintenance hosts for a number of diseases, such as corridor disease, foot and mouth disease, brucellosis, and bovine tuberculosis, maintaining infection through horizontal transfer within the population in the absence of other sources of infection (Renwick et al., 2007). The high prevalence of BTB in buffalo herds in southern African game reserves represents not only an ecological threat to biodiversity, but also a health risk through the wildlife-cattle-human interface (Tanner et al., 2014). In a study of cattle from 27 villages in Tanzania, Cleaveland et al. (2007) reported that cattle in contact with wildlife showed a significantly higher prevalence of BTB

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

infection than those without contact. Sufficient contact also makes Mycobacterium bovis a public health risk, as infection in humans has been known to occur through the consumption of unpasteurized milk, and could also result from the ingestion of undercooked meat of infected livestock (Cressey et al., 2006). High HIV prevalence within the rural communities that surround South African game reserves puts people at an increased risk of acquiring TB due to compromised immune systems (Amanfu, 2006; Michel et al., 2006). Economic losses may also occur through restrictions on the trade or sale of animals, both livestock and wildlife, as well as animal products (Ayele et al., 2004; Ramirez-Villaescusa et al., 2009). Varying environmental conditions, such as drought, can cause a dramatic decrease in buffalo populations, as a result of reduced body condition, starvation, predation and susceptibility to infectious diseases (Reardon, 2012). Under severe drought conditions, both the spread and severity of bovine tuberculosis may be altered, due to changes in herd density and the toll that environmental stress takes on the immune system (Cross et al., 2009; O’Brien et al., 2011). Thus, although stable at present, the presence of this new, introduced disease may have profound implications for the longterm stability of buffalo populations. In addition, a recent study demonstrated that BTB infection in buffalo can impact both individual and population health by affecting the outcome of endemic infections such as Rift Valley Fever (Beechler et al., 2015). Within South Africa, buffalo populations are found in both state-run and private game reserves, as well as breeding farms throughout the country, and animal translocations occur within similar reserves at least, and in some specific occasions, between these types of reserve/farm. The first BTB diagnoses in buffalo in two of South Africa’s largest conservation areas, Hluhluwe iMfolozi Park and the Kruger National Park, occurred in 1986 and 1990, respectively. Since that time, the disease has spread throughout both parks, and prevalence estimates within areas of KNP and HiP have reached 38% (Rodwell et al., 2001) and 54% (Le Roex et al., 2015), respectively. While prevention may be better than cure, the BCG vaccine, widely used in human populations, has limited effectiveness in buffalo. Experimental challenge with M. bovis in buffalo under captive conditions showed no significant protection conferred by the BCG vaccine, and consequently this does not represent a viable option for this species (Cross et al., 2009; De Klerk et al., 2010). The establishment of an effective BTB control strategy is also complicated by the presence of co-existing/spillover BTB hosts, such as lion (Panthera leo), chacma baboon (Papio ursinus), warthog (Phacochoerus africanus) and greater kudu (Tragelaphus strepsiceros). Some species, such as kudu, may also function as additional maintenance hosts in certain populations (Michel et al., 2006). Many developed countries have succeeded in eradicating or drastically reducing BTB prevalence in cattle using regular test and slaughter policies, but the logistic demands of operating such a programme in large, free-ranging wildlife populations render this option, as well as vaccination, impractical and financially unfeasible (Michel et al., 2010). This is particularly true if park-wide testing must be routine in order to achieve the success observed in agricultural practices. Furthermore, large-scale culling within conservation areas and national parks creates extremely negative publicity, and may have additional unintended ecosystem effects. For the above reasons, selective breeding for increased BTB resistance may be seriously considered as a potential control strategy in the management of BTB in African buffalo.

1. Resistance vs tolerance Resistance and tolerance are the two main aspects of defence against pathogens, and together determine disease severity (Råberg et al., 2007). Whilst resistance involves limiting the

39

bacterial burden and has been shown to exhibit substantial genetic variation in animal models, tolerance is the restriction of the harmful consequences caused by the bacteria (Stear et al., 2001; Råberg et al., 2007). Resistance and tolerance have been shown to be negatively correlated in infectious disease, with a potential tradeoff occurring between them as they employ opposing strategies: the cost of increased immune control of infection is typically an increase in ‘collateral damage’ of infected tissue (Råberg et al., 2007). Thus, in a selective breeding programme, it would be important to establish whether resistance or tolerance is the desired goal. A possible negative outcome to breeding for increased resistance is the antagonistic co-evolution of the pathogen. Any alteration in the resistance of the host to the pathogen places selective pressure on the pathogen and corresponding counter-adaptations could occur (Berry et al., 2011). Fortunately, spoligotyping has shown that culling of infected cattle in the British Isles has resulted in a bottleneck for M. bovis evolution, thus the risk for selective pressure on pathogen evolution may be small (Smith et al., 2006). No such selective pressure is believed to be placed on the pathogen if selection of the host is tolerance-based, thus diminishing the antagonistic co-evolution of host and pathogen (Råberg et al., 2007; Berry et al., 2011). However, only highly susceptible individuals provide information on tolerance under most prevalence conditions, as a significant bacterial burden is required to accurately estimate effects. These individuals are thus the least desirable from a resistance perspective (Bishop and Woolliams, 2014). In the South African context, however, the presence of coexisting BTB hosts makes increasing the resistance of buffalo to BTB a more desirable goal, as it would be more effective at reducing interspecies spread by reducing BTB prevalence in the major host. Predation is typically viewed as a process that improves the health of a prey population by removing the weakest individuals. If infected individuals are more likely to be caught, the overall prevalence of a pathogen will be reduced, and disease transmission should decrease due to the removal of infectious hosts (Packer et al., 2003; Williams, 2008). African buffalo are heavily predated by lions within the national parks of South Africa, suggesting that there could be inadvertent selection for BTB tolerance within these areas. If lions remove the buffalo most severely affected by BTB, the prevalence of highly susceptible animals should decrease, and the prevalence of resistant and/or tolerant animals should increase. Even in the absence of selective predation, non-selective predation shortens the lifespan of infected individuals and thus may assist in reducing disease transmission (Williams, 2008). However, in a host-pathogen system where both the predator and prey species are affected by the same disease (as in the case with BTB), the situation may be a more complex interaction between mortality, population size and transmission in the two species (Packer et al., 2003; Roberts and Heesterbeek, 2013).

2. Types of selective breeding Selective breeding programmes seek to identify individuals with a particular trait of interest, and preferentially utilize those individuals for breeding. Over time, the trait variant will become more prevalent in a particular population. Selection for health and reproductive traits has occurred in breeding programs for centuries, but disease traits have been incorporated only recently. Despite relatively low heritability estimates, breeding programs have been successfully implemented for resistance to diseases such as mastitis and brucellosis in cattle (Morris, 2007). More recently, the selection of individuals for breeding programmes can be based on either phenotypic or genotypic merit, and can be categorized into three main

40

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

types: phenotypic, marker-assisted and genomic selection strategies. 2.1. Phenotypic selection The selection of animals for breeding programmes has traditionally been based on the estimated breeding values (EBVs) of individuals, calculated using phenotypic records, pedigrees and the heritability of a particular trait (Goddard and Hayes, 2009; Hayes et al., 2009). Many of the traits that are important in agriculture are complex or quantitative traits; the variation seen in these traits is influenced by many genes and produces a continuous distribution of phenotypes among individuals (Goddard and Hayes, 2009). Beef and dairy cattle stud farms routinely measure quantitative traits, and in conjunction with family structure data, EBVs are estimated across herds in order to enhance traits such as total milk yield, body composition and health and fitness traits (Hayes et al., 2013). While selection of complex traits based on phenotypic records has been successful in the past, it also can be difficult, and costly. Often the phenotype of interest, or indicator trait, can only be measured at a particular stage of life, after death, in one sex, or is very expensive to test (Goddard and Hayes, 2009; Hayes et al., 2013). Phenotypic studies have investigated the resistance of cattle to M. bovis by the measurement of the individual’s inflammatory response and by trial exposures in vivo and in vitro. The continuous distribution of resistance as a trait (measured by lesion scores or inflammatory responses), and the likelihood of this trait being polygenic and multi-factorial, suggests that resistance to M. bovis can be regarded as a quantitative trait (Hill and Mackay, 2004). Thus the same process should be applicable to breeding for BTB resistance as it is for a quantitative trait. The breeding value of an animal as a function of its resistance to BTB can be calculated using the relative phenotypic resistance of closely related animals, and will be a more robust measure than the animal’s own phenotypic measure of resistance (Meuwissen et al., 2001; Hill and Mackay, 2004). 2.2. Marker-assisted selection Marker-assisted selection incorporates the information obtained from a small number of genetic markers into the traditional EBV calculations – that is, breeding values are calculated by combining pedigree, phenotype and marker information (Goddard and Hayes, 2009). Markers are typically identified by candidate gene association studies in the trait of interest. In these studies, genes may be selected based on prior knowledge of the physiology of a particular trait or their position relative to a quantitative trait locus (QTL), and polymorphisms/variants within the gene are tested for statistical association with that trait. The success of marker-assisted selection depends on a number of factors, such as the number of genes contributing to variation in a trait, the relative contribution of each gene, and what has previously been attained in terms of genetic improvement using phenotypic selection (Goddard and Hayes, 2007). For example, if phenotypic selection has been implemented for many years and substantial improvement has already been achieved, it is unlikely that the addition of a single significant genetic marker to the EBV calculation will have a considerable effect. However, in relatively new breeding programmes, the incorporation of any genetic information into the calculation of breeding values could substantially improve the EBV calculations and thus the success of the programme (Goddard and Hayes, 2007). 2.3. Genomic selection Genomic selection refers to the selection of individuals based on marker or haplotype information obtained from a dense array

of markers that covers the entire genome, thus theoretically incorporating all the loci that contribute to a particular trait (Meuwissen et al., 2001). This information is compiled to calculate a genomicestimated breeding value (GEBV), which can then be used to predict the breeding value of individuals in the absence of phenotype data (Goddard and Hayes, 2009). In order to generate a reliable prediction equation for genomic selection, a large reference population with both phenotype and genotype data is required. Individuals in the reference population are genotyped across a genome-wide set of markers, and the effect of each of the markers on the trait of interest is quantified. Breeding values in individuals without phenotype data can then be predicted by the sum of the effects of their markers across the genome (Berry et al., 2014; Tsairidou et al., 2014). This allows the selection of the best individuals for breeding purposes. However, because the effect of most complex trait markers is small, the size of the reference population must be sufficiently large in order to estimate the effects accurately (Berry et al., 2014). This is one of the biggest challenges facing genomic selection, but if such a population can be established, a wide variety of traits can be phenotyped, and thus GEBVs for all these phenotypes can be calculated. This would provide an enormous amount of data, and could dramatically alter future breeding programmes (Goddard and Hayes, 2009).

3. Implementing genetic improvement strategies One of the most crucial parameters in any selective breeding programmes is heritability, as it provides an indication of the potential success of this strategy for a particular trait. The higher the heritability score, the greater the role that genetics plays in determining the phenotype and the more potentially enhanceable the trait is through selective breeding (Falconer and Mackay, 1996). The heritability of resistance to BTB infection in cattle and wildlife species has been calculated in a number of studies, and a summary of these estimates can be seen in Table 1. The most recent studies in cattle utilized genome-wide data to estimate the heritability of BTB resistance, and calculated values of approximately 0.21 and 0.23 (Bermingham et al., 2014; Tsairidou et al., 2014). While any heritability could be interesting in a research context, the substantial proportions determined indicate that it is worth investigating the genetics underlying such heritability in the practical context of selective breeding. Increased understanding of the genetic architecture of complex phenotypes and improvements in genetic technologies have resulted in the potential to apply genetic information to animal breeding programmes for many traits (Berry et al., 2011). Initially, microsatellite markers provided new opportunities to identify quantitative trait loci (QTL) or genes influencing desirable traits. However, these highly polymorphic markers are expensive to genotype and consequently many studies lacked the power to detect any but the largest effects (Garrick, 2011). The utilization of single nucleotide polymorphisms (SNPs) in candidate gene association studies typically provides a small number of causal variants for a particular trait, and more recently, the development of nextgeneration sequencing technologies has enabled large-scale SNP discovery, and a faster and more cost-effective solution to genotyping (Garvin et al., 2010; Everett et al., 2011). Marker-assisted selection has been used in a number of livestock species, to improve traits such as reproductive rate, body composition, growth rate and meat quality in pigs, sheep and cattle (Dekkers, 2004). The identification of a QTL for infectious pancreatic necrosis (IPN) in salmon enabled marker-assisted breeding, and resulted in a significant change in the frequency of the resistant allele in just one generation (Bishop and Woolliams, 2014). The effect of a marker-assisted selection strategy is difficult

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

41

Table 1 BTB heritability estimates in cattle and wildlife (adapted from Le Roex et al., 2013a) Animal Red deer Cattle Cattle Cattle Cattle

Breed

Factor

Method

Heritability

Authors

Irish Holstein-Friesian British Holstein-Friesian Holstein-Friesian Holstein

Lesion severity Skin test response; M. bovis culture Skin test response; M. bovis culture Skin test response + evidence of pathology Skin test response + TB lesions at necropsy + M. bovis culture

Phenotypic Phenotypic Phenotypic Genomic Genomic

0.48 0.14; 0.18 0.16; 0.18 0.21 0.23

Mackintosh et al. (2000) Bermingham et al. (2009) Brotherstone et al. (2010) Bermingham et al. (2014) Tsairidou et al. (2014)

to quantify, particularly when the traits are complex or the breeding scheme has a multiple-trait breeding objective. However, changes in gene or marker frequencies, or changes in the genetic merit of the population can be used to provide an estimate of success (Dekkers, 2004). However, this method is not always successful for complex traits, due to the large number of markers of small effect (Hayes et al., 2013). Although, BTB resistance is a polygenic trait (Allen et al., 2010; Le Roex et al., 2013a), if the effect of a particular gene variant is large enough to cause a discontinuity of the phenotype (despite the effects of other loci and non-genetic factors), it may be detected in an association study and thus could have possible utility in breeding programmes (Hill and Mackay, 2004). Greater genetic gain for complex traits is expected through genomic selection. The successful incorporation of genome-wide genetic information into selective breeding programmes has already been achieved in a number of species – studies in HolsteinFriesian cattle have shown up to a 71% concordance between GEBVs and true breeding values, averaged across a number of traits (VanRaden et al., 2009), and studies in broiler chickens showed a fourfold increase in the accuracy of prediction of a food conversion rate phenotype (Gonzalez-Recio et al., 2009).

4. Application of selective breeding in African buffalo To date, selective buffalo breeding in South Africa has been based on phenotypic merit. In order to increase the improvements seen thus far, genetic information could be incorporated into breeding values, using either a marker-assisted or genomic selection approach. Buffalo genetic research in the past has predominantly focused on population genetics, although studies in disease genetics have become more frequent. A recent association study tested 69 SNPs in immune-related genes in buffalo for association with BTB infection status, and found three positive associations, in the SLC7A13, DMBT1 and IL1˛ genes (Le Roex et al., 2013b). SLC7A13 is a heteromeric amino acid transporter (HAT), and although relatively uncharacterized, other transporters, such as SLC11A1 (NRAMP1), have been associated with TB in humans (Hoal et al., 2004) and the chromosomal region containing SLC6A6 has be associated with BTB in cattle (Finlay et al., 2012). DMBT1 is regarded as a pattern recognition receptor (PRR), which forms part of the human host defence system against invading organisms such as M. tuberculosis (Ligtenberg et al., 2010). The IL1˛ protein is a pro-inflammatory cytokine, and plays a role in the activation of T-lymphocytes, and the recruitment of leucocytes to sites of inflammation (Schmidt and Kao, 2007). These variants could be promising candidates for marker-assisted selective breeding of buffalo with enhanced BTB resistance. To illustrate, a schematic diagram of the basic steps of selective breeding for enhanced BTB resistance in African buffalo is shown in Fig. 1. The best results of genetic improvement schemes are typically achieved from the genomic selection approach, but this approach is difficult in a species without an assembled genome. The African buffalo genome has not yet been assembled, and no commercial SNP chip for the buffalo is available. While some next-generation sequence information has been published for the African buffalo

(Le Roex et al., 2012), additional data would need to be generated in order to achieve the evenly-spaced, genome-wide coverage required. In order to achieve the maximum genetic improvement, this may be the best approach, provided that the expense of this strategy is not prohibitive. Investigating differential RNA expression in BTB-infected and uninfected individuals, using technologies such as RNA-Seq to generate genome-wide markers may also be an excellent source of candidate markers for BTB resistance, although the same challenges apply as with whole-genome sequencing. Mapping data to the reference genome of another species is complex, and typically results in a higher error rate and less usable data (Everett et al., 2011). 5. Challenges and considerations Selective breeding programmes operate optimally with additional management practices, such as frequent testing and status monitoring, quarantine and treatment of sick animals with antibiotics, but these practices are not possible in a free-ranging environment, such as national parks. In order to implement these practices, wildlife populations such as buffalo would need to be maintained in a fenced environment under controlled conditions, similar to that of cattle. This practice has other advantages in African buffalo, such as increased feed quality and availability, which results in a younger age of first calving (pers. comm Bernard Wooding). These additional management practices could, however, be implemented in smaller private reserves and game farms. Another factor that needs to be considered is the suitability of the phenotypic test. The more accurate the diagnostic test to determine the phenotype, the more reliable the association between genotype and phenotype, and thus the more accurate the predicted breeding values of individuals will be. The single intradermal comparative tuberculin test (SICTT) is the most commonly used diagnostic test for BTB infection, but sensitivity and specificity estimates

Fig. 1. Schematic overview of breeding for enhanced BTB resistance in African buffalo.

42

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

for this test in buffalo have not been established. Selection based on the ability to pass the SICTT test, rather than direct selection for resistance phenotypes, may also result in the diminished ability of buffalo to develop a measureable response to the SICTT. This would result in the reduced ability to detect BTB infection, thus undermining the very usefulness of the test (Amos et al., 2013). Other diagnostic assays, such as interferon gamma release assays (IGRAs) and gene expression assays (GEAs), have been optimised for use in African buffalo (Michel et al., 2011; Parsons et al., 2012). These assays would not result in the reduced ability to show a measurable response, but they may have other limitations, such as the occurrence of false positives due to sensitization from environmental mycobacteria in IGRAs, which must then be considered. (Michel et al., 2011). As mentioned, heritability estimates should be calculated prior to the implementation of a selective breeding programme, in order to determine the extent to which the observed phenotypic variation is caused by genetic variance and is not a result of environmental variance. Studies have shown different heritability estimates for BTB in cattle, depending on breed and the diagnostic method utilized, and consideration must be given to these factors. The degree to which BTB resistance is caused by genetic variance in African buffalo has not been established as of yet, and this needs to be done before the true value of selective breeding for enhanced BTB resistance can be determined. Selection for individual immune traits can affect different components of the immune system, which in turn may compromise individual ability to fight other infections. For example, selecting for resistance to M. bovis is likely to select for strong Th1 and Th17 responses, which act against intracellular pathogens, viruses and fungal infections, rather than the Th2 response pathway which acts against extracellular bacterial and parasitic infections (Ardia et al., 2011). This trade-off between different aspects of the immune system was demonstrated in a study that showed that immune suppression resulting from nematode infection facilitates BTB infection in African buffalo (Ezenwa et al., 2010). Thus while genomic selection for enhanced BTB resistance may also increase resistance to other intracellular bacterial pathogens, such as Brucella abortus and Mycobacterium avium paratuberculosis, it may increase susceptibility to other extracellular pathogens, particularly if co-infection occurs (Allen et al., 2010; Ardia et al., 2011). This would be of significant concern in African buffalo, as they are endemic hosts of a vast array of bacterial, viral and parasitic infections. Variation within the immune response at the population level can, however, make the population as a whole more resilient against disease, as diverse immune responses to a pathogen provide a greater chance of survival and increased adaptability. Thus consideration must be given to possible changes in immunity when selecting for BTB resistance (Berry et al., 2011; Van Helden, 2011). This also applies to genetic diversity in general. With animals as valuable as buffalo, it is important that other traits are not negatively influenced as a result of selective breeding for enhanced BTB resistance. While artificial selection practices have been used to develop economically important traits in many domestic and livestock species (as mentioned), these processes may also result in unintended genetic and phenotypic changes within these populations. For example, selective breeding in salmon has demonstrated many undesirable traits, including increased aggression (Einum and Fleming, 1997; Houde et al., 2010), reduced stress responsiveness (Solberg et al., 2013) and lower survival rates in the wild (Fleming et al., 2000). Long term selection for body weight in chickens has resulted in numerous disorders of the skeletal, muscular and reproductive systems, among others (Hocking, 2014). A possible solution to this problem would be to incorporate a number of different desirable traits simultaneously, so that animals are not selected for one trait to the detriment of others.

Selection for multiple traits simultaneously is common practice in animal breeding, using individual indices that weights different traits according to relative desirability (Miglior et al., 2005). This approach would mitigate the detrimental effects of reduced genetic diversity at characterised loci, but would not safe-guard against this problem at unknown loci or in uncharacterized traits. As a species of immense economic value with a long generation time, it would therefore be particularly important to closely monitor selective breeding practices in African buffalo in order to ensure that concomitant selection of undesirable or harmful traits does not occur. Finally, ethical consideration must be given to the application of selective breeding in different populations of buffalo. For example, this strategy may not be suitable for some of the larger, state-run national parks, where intervention is kept to a minimum, and animals are managed as free-range, “wild” populations. However, in smaller game reserves and farms, active disease and population management strategies already exist, including hunting based on specific phenotypic characteristics, and thus the relevant managers may choose to implement selective breeding practices on existing herds, or in the establishment of new herds or farms. 6. Conclusions The African buffalo is a species of great importance in southern Africa, particularly in the tourism, hunting and commercial game farming industries. This creates a constant demand for buffalo, and prices have continued to increase accordingly. The market value of buffalo within South Africa varies according to disease status, with ‘disease-free’ buffalo valued at approximately 10 times that of their counterparts at auction; in 2013, the Cape buffalo ‘Mystery’ was sold at a game auction in South Africa for a record $4.1 million, highlighting the enormous value placed on exceptional animals. However, bovine TB is prevalent in a large proportion of the country, and the maintenance of the disease by buffalo has potential implications for biodiversity, ecotourism and public health. The BCG vaccine has no significant effect on BTB infection in buffalo, and as such, it would be enormously beneficial to breed buffalo with enhanced resistance to BTB. These buffalo could then be used to re-stock reserves where BTB is endemic, and establish new herds in private game farms. Selective breeding programmes have been successfully applied to improve numerous traits in livestock species. Advances in DNA technologies have allowed the incorporation of genetic information into the estimated breeding values of individuals, thus greatly increasing the scope for improvement and the success of these strategies. To the best of our knowledge, no genetic selective breeding programmes for enhanced BTB resistance in buffalo have been implemented, and as such, the incorporation of any genetic information into a marker-assisted breeding strategy could be enormously advantageous. While genomic selection may ultimately provide a more effective route to genetic improvement, until such time as there is a sufficiently large reference population and commercial SNP chip or genome-wide marker information available for the African buffalo, marker-assisted selection may be the best strategy. As the cost of genotyping and custom SNP arrays continues to decline, genomic selection may become more feasible. Future plans should look to the accurate phenotyping of a large reference population of buffalo for multiple traits, including BTB resistance. 7. Management implications While selective breeding for enhanced BTB resistance in African buffalo may offer increased protection, this strategy is unlikely to

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

confer complete resistance to BTB. As such, management strategies should be designed to incorporate selective breeding as a publicly acceptable and feasible method of control, in conjunction with other disease management efforts, such as test and cull programmes. In addition, as M. bovis is capable of infecting numerous species other than buffalo, decisions must be made regarding whether the desired outcome in a particular ecosystem is resistance or tolerance-based. An increase in tolerance to BTB in buffalo may not be the optimal solution to disease control in an ecosystem with co-existing BTB hosts, and the economic value of tolerant animals is unlikely to increase. However, if tolerance is associated with a reduction in M. bovis shedding and thus transmission, and buffalo sales are not a desired outcome, then increased tolerance may be an acceptable target. If a dual selection strategy that selects for both resistant and tolerant genotypes is not possible, it would be important to consider the resulting effect on tolerant genotypes within both the breeding programme population, as well as the neighboring population of the same and/or other susceptible species. Conflict of interest The authors have no conflict of interest to declare. References Allen, A.R., Minozzi, G., Glass, E.J., Skuce, R.A., McDowell, S.W.J., Woolliams, J.A., Bishop, S.C., 2010. Bovine tuberculosis: the genetic basis of host susceptibility. Proc. R. Soc. B: Biol. Sci. 277, 2737–2745. Amanfu, W., 2006. The situation of tuberculosis and tuberculosis control in animals of economic interest. Tuberculosis 86, 330–335. Amos, W., Brooks-Pollock, E., Blackwell, R., Driscoll, E., Nelson-Flower, M., Conlan, A.J.K., 2013. Genetic predisposition to pass the standard SICCT test for bovine tuberculosis in British cattle. PLoS One 8, e58245. Ardia, D.R., Parmentier, H.K., Vogel, L.A., 2011. The role of constraints and limitation in driving individual variation in immune response. Funct. Ecol. 25, 61–73. Ayele, W.Y., Neill, S.D., Zinsstag, J., Weiss, M.G., Pavlik, I., 2004. Bovine tuberculosis: an old disease but a new threat to Africa. Int. J. Tuberculosis Lung Dis. 8, 924–937. Beechler, B.R., Manore, C.A., Reininghaus, B., O’Neal, D., Gorsich, E.E., Ezenwa, V.O., Jolles, A.E., 2015. Enemies and turncoats: bovine tuberculosis exposes pathogenic potential of Rift Valley Fever virus in a common host, African buffalo (Syncerus caffer). Proc. Biol. Sci. R. Soc. 282, http://dx.doi.org/10.1098/ rspb.2014.2942 Bermingham, M.L., Bishop, S.C., Woolliams, J.A., Pong-Wong, R., Allen, A.R., McBride, S.H., Ryder, J.J., Wright, D.M., Skuce, R.A., McDowell, S.W., Glass, E.J., 2014. Genome-wide association study identifies novel loci associated with resistance to bovine tuberculosis. Heredity 112, 543–551. Bermingham, M.L., More, S.J., Good, M., Cromie, A.R., Higgins, I.M., Brotherstone, S., Berry, D.P., 2009. Genetics of tuberculosis in Irish Holstein-Friesian dairy herds. J. Dairy Sci. 92, 3447–3456. Berry, D.P., Bermingham, M.L., Good, M., More, S.J., 2011. Genetics of animal health and disease in cattle. Irish Vet. J. 64, 5. Berry, D.P., Wall, E., Pryce, J.E., 2014. Genetics and genomics of reproductive performance in dairy and beef cattle. Animal 8, 105–121. Bishop, S.C., Woolliams, J.A., 2014. Genomics and disease resistance studies in livestock. Livestock Sci., In press (accessed 17.06.14.) http://www. sciencedirect.com/science/article/pii/S1871141314002352 Brotherstone, S., White, I.M.S., Coffey, M., Downs, S.H., Mitchell, A.P., Clifton-Hadley, R.S., More, S.J., Good, M., Woolliams, J.A., 2010. Evidence of genetic resistance of cattle to infection with Mycobacterium bovis. J. Dairy Sci. 93, 1234–1242. Cleaveland, S., Shaw, D.J., Mfinanga, S.G., Shirima, G., Kazwala, R.R., Eblate, E., Sharp, M., 2007. Mycobacterium bovis in rural Tanzania: risk factors for infection in human and cattle populations. Tuberculosis 87, 30–43. Cressey, P., R., Lake, A. Hudson, 2006. Risk profile: Mycobacterium bovis in red meat. New Zealand Food Safety Authority. . Cross, P.C., Heisey, D.M., Bowers, J.A., Hay, C.T., Wolhuter, J., Buss, P., Hofmeyr, M., Michel, A.L., Bengis, R.G., Bird, T.L.F., Du Toit, J.T., Getz, W.M., 2009. Disease, predation and demography: assessing the impacts of bovine tuberculosis on African buffalo by monitoring at individual and population levels. J. Appl. Ecol. 46, 467–475. De Klerk, L.-M., Michel, A.L., Bengis, R.G., Kriek, N.P.J., Godfroid, J., 2010. BCG vaccination failed to protect yearling African buffaloes (Syncerus caffer) against experimental intratonsilar challenge with Mycobacterium bovis. Vet. Immunol. Immunopathol. 137, 84–92. Dekkers, J.C.M., 2004. Commercial application of marker- and gene-assisted selection in livestock: strategies and lessons. J. Anim. Sci. 82, E313–328.

43

Einum, S., Fleming, I.A., 1997. Genetic divergence and interactions in the wild among native, farmed and hybrid Atlantic salmon. J. Fish Biol. 50, 634–651. Everett, M.V., Grau, E.D., Seeb, J.E., 2011. Short reads and nonmodel species: exploring the complexities of next-generation sequence assembly and SNP discovery in the absence of a reference genome. Mol. Ecol. Resour. 11, 93–108. Ezenwa, V.O., Etienne, R.S., Luikart, G., Beja-Pereira, A., Jolles, A.E., 2010. Hidden consequences of living in a wormy world: nematode-induced immune suppression facilitates tuberculosis invasion in African buffalo. Am. Nat. 176, 613–624. Falconer, D.S., Mackay, T.F.C., 1996. Introduction to Quantitative Genetics. Longman Group, Essex, England. Finlay, E.K., Berry, D.P., Wickham, B., Gormley, E.P., Bradley, D.G., 2012. A genome wide association scan of bovine tuberculosis susceptibility in Holstein-Friesian dairy cattle. PloS One 7, e30545. Fleming, I.A., Hindar, K., Mjølnerød, I.B., Jonsson, B., Balstad, T., Lamberg, A., 2000. Lifetime success and interactions of farm salmon invading a native population. Proc. R. Soc. London B: Biol. Sci. 267, 1517–1523. Garrick, D.J., 2011. The nature, scope and impact of genomic prediction in beef cattle in the United States. Genet. Sel. Evol. 43, 17. Garvin, M.R., Saitoh, K., Gharrett, A.J., 2010. Application of single nucleotide polymorphisms to non-model species: a technical review. Mol. Ecol. Resour. 10, 915–934. Goddard, M.E., Hayes, B.J., 2007. Genomic selection. J. Anim. Breed. Genet. 124, 323–330. Goddard, M.E., Hayes, B.J., 2009. Mapping genes for complex traits in domestic animals and their use in breeding programmes. Nat. Rev. Genet. 10, 381–391. Gonzalez-Recio, O., Gianola, D., Rosa, G.J., Weigel, K.A., Kranis, A., 2009. Genome-assisted prediction of a quantitative trait measured in parents and progeny: application to food conversion rate in chickens. Genet. Sel. Evol. 41, 3. Hayes, B.J., Bowman, P.J., Chamberlain, A.J., Goddard, M.E., 2009. Invited review: genomic selection in dairy cattle: progress and challenges. J. Dairy Sci. 92, 433–443. Hayes, B.J., Lewin, H.A., Goddard, M.E., 2013. The future of livestock breeding: genomic selection for efficiency, reduced emissions intensity, and adaptation. Trends Genet. 29, 206–214. Hill, W.G., Mackay, T.F.C., 2004. D.S. Falconer and introduction to quantitative genetics. Genetics 167, 1529–1536. Hoal, E.G., Lewis, L.A., Jamieson, S.E., Tanzer, F., Rossouw, M., Victor, T., Hillerman, R., Beyers, N., Blackwell, J.M., van Helden, P.D., 2004. SLC11A1 (NRAMP1) but not SLC11A2 (NRAMP2) polymorphisms are associated with susceptibility to tuberculosis in a high-incidence community in South Africa. Int. J. Tuberculosis Lung Dis. 8, 1464–1471. Hocking, P.M., 2014. Unexpected consequences of genetic selection in broilers and turkeys: problems and solutions. Br. Poult. Sci. 55, 1–12. Houde, A.L.S., Fraser, D.J., Hutchings, J.A., 2010. Fitness-related consequences of competitive interactions between farmed and wild Atlantic salmon at different proportional representations of wild-farmed hybrids. ICES J. Mar. Sci.: J. Conseil 67, 657–667. Le Roex, N., Cooper, D., van Helden, P.D., Hoal, E.G., Jolles, A.E., 2015. Disease control in wildlife: evaluating a test and cull programme for bovine tuberculosis in African buffalo. Transboundary Emerging Dis. Le Roex, N., van Helden, P.D., Koets, A.P., Hoal, E.G., 2013a. Bovine TB in livestock and wildlife: what’s in the genes? Physiol. Genomics 45, 631–637. Le Roex, N., Koets, A.P., van Helden, P., Hoal, E.G., 2013b. Gene polymorphisms in African buffalo associated with susceptibility to bovine tuberculosis infection. PLoS One 8, e64494. Le Roex, N., Noyes, H., Brass, A., Bradley, D.G., Kemp, S.J., Kay, S., van Helden, P.D., Hoal, E.G., 2012. Novel SNP discovery in African buffalo, Syncerus caffer, using high-throughput sequencing. PLoS One 7, e48792. Ligtenberg, A.J.M., Karlsson, N.G., Veerman, E.C.I., 2010. Deleted in malignant brain tumors-1 protein (DMBT1): a pattern recognition receptor with multiple binding sites. Int. J. Mol. Sci. 11, 5212–5233. ˜ Lindsey, P.A., Roulet, P.A., Romanach, S.S., 2007. Economic and conservation significance of the trophy hunting industry in sub-Saharan Africa. Biol. Conserv. 134, 455–469. Mackintosh, C.G., Qureshi, T., Waldrup, K., Labes, R.E., Dodds, K.G., Griffin, J.F., 2000. Genetic resistance to experimental infection with Mycobacterium bovis in red deer (Cervus elaphus). Infect. Immun. 68, 1620–1625. Meuwissen, T.H., Hayes, B.J., Goddard, M.E., 2001. Prediction of total genetic value using genome-wide dense marker maps. Genetics 157, 1819–1829. Michel, A.L., Bengis, R.G., Keet, D.F., Hofmeyr, M., Klerk, L.M., Cross, P.C., Jolles, A.E., Cooper, D., Whyte, I.J., Buss, P., Godfroid, J., 2006. Wildlife tuberculosis in South African conservation areas: implications and challenges. Vet. Microbiol. 112, 91–100. Michel, A.L., Cooper, D., Jooste, J., de Klerk, L.M., Jolles, A., 2011. Approaches towards optimising the gamma interferon assay for diagnosing Mycobacterium bovis infection in African buffalo (Syncerus caffer). Prev. Vet. Med. 98, 142–151. Michel, A.L., Muller, B., van Helden, P.D., 2010. Mycobacterium bovis at the animal-human interface: a problem, or not? Vet. Microbiol. 140, 371–381. Miglior, F., Muir, B.L., Van Doormaal, B.J., 2005. Selection indices in Holstein cattle of various countries. J. Dairy Sci. 88, 1255–1263. Morris, C.A., 2007. A review of genetic resistance to disease in Bos taurus cattle. Vet. J. 174, 481–491. O’Brien, D.J., Schmitt, S.M., Rudolph, B.A., Nugent, G., 2011. Recent advances in the management of bovine tuberculosis in free-ranging wildlife. Vet. Microbiol. 151, 23–33.

44

N. le Roex et al. / Acta Tropica 149 (2015) 38–44

Packer, C., Holt, R.D., Hudson, P.J., Lafferty, K.D., Dobson, A.P., 2003. Keeping the herds healthy and alert: implications of predator control for infectious disease. Ecol. Lett. 6, 797–802. Parsons, S.D.C., Menezes, A.M., Cooper, D., Walzl, G., Warren, R.M., van Helden, P.D., 2012. Development of a diagnostic gene expression assay for tuberculosis and its use under field conditions in African buffaloes (Syncerus caffer). Vet. Immunol. Immunopathol. 148, 337–342. Prins, H.H.T., 1996. Ecology and Behaviour of the African Buffalo: Social Inequality and Decision Making. Chapman & Hall. Råberg, L., Sim, D., Read, A.F., 2007. Disentangling genetic variation for resistance and tolerance to infectious diseases in animals. Science 318, 812–814. Ramirez-Villaescusa, A.M., Medley, G.F., Mason, S., Green, L.E., 2009. Herd and individual animal risks associated with bovine tuberculosis skin test positivity in cattle in herds in south west England. Prevent. Vet. Med. 92, 188–198. Reardon, M., 2012. Shaping Kruger. Struik Nat. Renwick, A.R., White, P.C.L., Bengis, R.G., 2007. Bovine tuberculosis in southern African wildlife: a multi-species host-pathogen system. Epidemiol. Infect. 135, 529–540. Roberts, M.G., Heesterbeek, J.A.P., 2013. Characterizing the next-generation matrix and basic reproduction number in ecological epidemiology. J. Math. Biol. 66, 1045–1064. Rodwell, T.C., Kriek, N.P., Bengis, R.G., Whyte, I.J., Viljoen, P.C., de, V., Boyce, W.M., 2001. Prevalence of bovine tuberculosis in African buffalo at Kruger National Park. J. Wildl. Dis. 37, 258–264.

Schmidt, D.R., Kao, W.J., 2007. The interrelated role of fibronectin and interleukin-1 in biomaterial-modulated macrophage function. Biomaterials 28, 371–382. Solberg, M.F., Skaala, Ø., Nilsen, F., Glover, K.A., 2013. Does domestication cause changes in growth reaction norms? A study of farmed, wild and hybrid Atlantic Salmon families exposed to environmental stress. PLoS One 8, e54469. Stear, M.J., Bishop, S.C., Mallard, B.A., Raadsma, H., 2001. The sustainability, feasibility and desirability of breeding livestock for disease resistance. Res. Vet. Sci. 71, 1–7. Tanner, M., Inlameia, O., Michel, A., Maxlhuza, G., Pondja, A., Fafetine, J., Macucule, B., Zacarias, M., Manguele, J., Moiane, I.C., Marranangumbe, A.S., Mulandane, F., Schönfeld, C., Moser, I., van Helden, P., Machado, A., 2014. Bovine tuberculosis and brucellosis in cattle and African buffalo in the Limpopo National Park, Mozambique. Transboundary Emerging Dis. Tsairidou, S., Woolliams, J.A., Allen, A.R., Skuce, R.A., McBride, S.H., Wright, D.M., Bermingham, M.L., Pong-Wong, R., Matika, O., McDowell, S.W.J., Glass, E.J., Bishop, S.C., 2014. Genomic prediction for tuberculosis resistance in dairy cattle. PloS One 9, e96728. Van Helden, P., 2011. Adaptation, redundancy or resilience. EMBO Rep. 12, 872. Van Raden, P.M., Van Tassell, C.P., Wiggans, G.R., Sonstegard, T.S., Schnabel, R.D., Taylor, J.F., Schenkel, F.S., 2009. Invited review: reliability of genomic predictions for North American Holstein bulls. J. Dairy Sci. 92, 16–24. Williams, P.D., 2008. Unhealthy herds: some epidemiological consequences of host heterogeneity in predator–host–parasite systems. J. Theor. Biol. 253, 500–507.