Advances in molecular genetic techniques applied to selection for litter size in goats (Capra hircus): a review

ABSTRACT

Litter size, or prolificacy, in goats is defined as the number of kids born per doe kidding. Improving litter size through selection not only directly enhances producer profitability as more progeny can be marketed but can also increase genetic gains in other traits due to greater selection intensity. However, most traits associated with reproduction have low heritability, and genetic improvement will be slow if the selection is based on one or a few phenotypic records. In the absence of a genetic evaluation programme with extensive pedigrees and performance recording, phenotypic selection for litter size is not promising. Advances in molecular genetic techniques may serve as an alternative to increase genetic progress in prolificacy. Several techniques have been developed to elucidate the mechanisms involved in phenotypic expression at the DNA level. Although recent research has identified genomic regions associated with several production traits in goats, litter size has not been extensively researched. Nevertheless, recent advancements in molecular genetic have created new opportunities for the improvement of litter size in goats. The development of next generation molecular tools to identify genomic genetic variants has made it possible to apply whole-genome scanning techniques, genome-wide association studies, and genomic selection to improve goat prolificacy.

KEYWORDS:

• Capra hircus
• prolificacy
• genome-wide
• Single Nucleotide Polymorphisms
• molecular markers

1. Introduction

The world goat population has reached 1.06 billion placing goat production in a strategic position with regard to agribusiness and social programmes around the world. This livestock sector represents an important source of protein, which aids in combating poverty and food insecurity primarily in continent that hold the largest herds of goats such as Asia (58.5%) and Africa (34.5%) (FAO 2017).

Although dairy goats are concentrated in low-income, food-deficit countries, where their products are a key food source, they are also present in high-income, technologically developed countries (Pulina et al. 2018). The Mediterranean area is the main goat milk and goat cheese producer (18%) outside of India (22%), which has the greatest goat milk volume of all countries (Dubeuf et al. 2004). France, Greece, Italy, and Spain are countries that stand out in the goat dairy worldview as they lead the international market for goat dairy products. These countries have productive models based on the use of specific breeds and ‘Protected Denomination of Origin’ cheeses produced according to traditional recipes, which are currently appreciated as ingredients in healthy diets. The current production situation shows substantial room for improvement, and relevant increases in milk production (from 30 to 50%) are expected by 2030 (Pulina et al. 2018).

Despite its growth and the recognized socioeconomic value, most of the goat rearing in the world presents low productivity indexes due to the low reproductive and productive performance of the animals (Devendra 2013). Improvement of reproductive performance in livestock species increases the number of available offspring and total farm revenue. According to Ahlawat et al. (2015c), improving the reproductive rate of low-producing goat breeds could potentially overcome the gap between the demand and production of meat. Thus, prolificacy deserves special attention for selection programmes because it can lead to improved profitability. This trait refers to the average number of kids born per doe kidding and has a positive impact on total animals marketed and replacement rate. In general, small ruminants have a greater frequency of multiple births than other domestic livestock species (Hafez and Hafez 2004). Past research has shown that the average litter size in goats is greater than 1, varying from 1.30 to 2.37 (Table 1). This characteristic of goats constitutes an excellent opportunity to increase the income of the productive system by increasing the number of progenies per reproductive cycle.

Table 1. Average litter size of goats according to breed and country.

Breed	Country	Litter size	Authors
Alpina	Itália	1.60	Gonçalves et al. (2008)
Anglo-Nubiana	Brasil	1.90	Cardoso et al. (2018)
Black Bengal	Índia	1.75	Haldar et al. (2014)
Creole	Franch	2.10	Gunia et al. (2010)
Jining Grey	China	2.37	Feng et al. (2011)
Laoshan	China	1.70	Lai et al. (2016)
Crossbreed (Pardo Alpino × Gurgueia)	Brasil	1.49	Sarmento et al. (2010)
		1.40	Santos et al. (2013)
Saanem	Brasil	1.50	Leal et al. (2018)
Toggenburg	Brasil	1.75	Leal et al. (2018)
Zaraibi	Egito	1.60	Hamed et al. (2009)

Due to its variability, which depends on genetic (Hamed et al. 2009) and non-genetic effects (Sarmento et al. 2010; Haldar et al. 2014), selection for litter size is feasible. Improving litter size in goats also increases the number or animals available for replacements and consequently increases selection intensity. Therefore, improving litter size can also increase annual genetic progress in other economically important goat traits (Sarmento et al. 2010; Santos et al. 2013). However, the improvement of reproductive traits by traditional selective breeding methods has proven to be difficult due to their sex- limited nature and low heritability, which varies from 0.08 to 0.18 (Hamed et al. 2009; Gunia et al. 2010; Santos et al. 2013; Menezes et al. 2016). While rapid rates of genetic progress for growth-related or milk traits have been achieved in small ruminants, a relatively lower rate of progress is possible for traits that are measured later in the life of females, such as reproductive ability (Mrode et al. 2018).

Molecular genetic techniques are promising because they have the ability to analyse genetic variability at the DNA level by detecting causal genes for reproductive characteristics (Ahlawat et al. 2015c). These technologies have progressed rapidly in recent years and have been successfully adapted to animal selection methods. They are used to identify loci responsible for genetic variation in quantitative traits. Major genes have a pronounced phenotype expression and they can be identified from candidate gene studies or genome-wide association studies (GWAS). The candidate approach quantifies the association between a trait’s phenotypic variation and one or more genes.

Genomic scanning is based on genotyping the whole genome with markers scattered throughout. The association between the marker alleles and the trait phenotypic variation has been established. According to Georges (1999), fine-mapping methods are being devised to refine the initial map positions of the quantitative trait loci to the point required for marker-assisted selection. Marker-assisted selection (MAS) uses genetic markers, which are jointly applied with traditional selection methods, to obtain more accurate predictions of genetic merit.

Advances in molecular biology have made the discovery of unprecedented levels of genomic variation possible, and consequently, the development of various single nucleotide polymorphism (SNP) chips for genotyping purposes. As a result of these advances, molecular information has been incorporated into small ruminant breeding programmes in developed countries to accelerate the rate of genetic progress (Mrode et al. 2018). One of the first studies that utilized Illumina’s 50 K SNP Beadchip in dairy goats revealed many genomic regions associated with economically important traits such as protein and fat yield (Maroteau et al. 2013).

In addition, other approaches have been developed aiming to accelerate genetic progress in goat breeding, such as genomic selection (Mucha et al. 2015; Mucha et al. 2017) and signatures of selection (Brito et al. 2017). This review aims to present advances in molecular-based methods applied to the improvement of litter size in goats.

2. Dense maps of single nucleotide polymorphisms and validation studies

Early research used restriction fragment length polymorphism (RFLP) markers associated with economically important traits in domestic animals. Later, the discovery of simple sequence repeats (SSR) and the development of DNA sequencers made the detection of allelic variations possible. After that, single nucleotide polymorphism (SNP) genotyping decreased costs and increased the accuracy of discovery. The SNP markers are based on changes in nitrogenous single chain bases (adenine, cytosine, thymine, and guanine). After identifying the genomes of economically important livestock species and mapping their haplotypes, technologies were developed for the mass genotyping of hundreds of thousands of SNP markers (Caetano 2009).

These advances in molecular genetic techniques provide the development of dense SNP and high-throughput sequencing. Although microsatellite markers have greater polymorphism on a per-marker basis, they are not as abundant as SNPs because they lack sufficient coverage across the genome and have limited automation options (Vignal et al. 2002). Research that investigated genes with a greater effect on the mechanisms involved in goat prolificacy (Ahlawat et al. 2016; Thomas et al. 2016; Ariyarathne et al. 2017) obtained minimal results due to limited coverage and automation options.

Therefore, the use of SNP genotyping through high-processing platforms allows the generation of information in a single test of millions of base pairs, making this process more advantageous in terms of cost per base and time saving when compared to other marker classes. These techniques have favoured studies on association and genetic mapping, genomic selection, diagnostic tests for confirming paternity, and individual identification.

The sequencing of the goat genome by Dong et al. (2013) was a significant advancement in molecular genetics in this field. The Illumina’s 50 K SNP Beadchip development is an example of this advancement. It has 53,347 SNPs with different spaces between them, and segregates with high to moderate frequency in the Alpine, Boer, Crioulo, Katjang, Saanen and Savana breeds (Tosser-Klopp et al. 2014). This high-processing platform was developed by the International Goat Genome Consortium and gives support to GWAS, the prediction of genomic breeding values (GBV), and the incorporation of this information into goat selection programmes. During the development of the genotyping array, only six goat breeds (Alpine, Boer, Crioula, Katjang, Saanen, and Savana) were included. Hence, it is necessary to apply validation studies to the different breeds.

3. Major genes affecting prolificacy in goats

The most common methods used to identify genetic markers are candidate genes and genomic scanning. Loci with moderate to large effects on the expression of quantitative traits are called quantitative trait loci (QTL, Geldermann 1975). The candidate gene approach characterizes the relationship between a sequenced gene of known biological action and the development or physiology of a given trait (Rothschild and Soller 1997). Many regions of the goat genome that are associated with economically important traits were detected after the development of a dense microsatellite linkage map for Capra hircus (Vaiman et al. 1996) and linear regression methods to detect QTL in relatives (Koning et al. 1998). Marrube et al. (2007) used microsatellite markers to detect chromosomal segments associated with conformation traits in Angora goats. Both Marrube et al. (2007) and Caño et al. (2009) also utilized microsatellite markers and found QTLs affecting Angora fleece traits. The segregation of QTL influencing mohair production and pre-weaning growth traits in South African Angora goats was investigated by Visser et al. (2011 and 2013, respectively).

More recently, major genes have also been identified from GWAS using the Illumina’s 50 K SNP Beadchip. With the decrease of sequencing and genotyping costs and the increase of genomic studies in small ruminants, it is expected that many more major genes and causal mutations will be available in the near future (Rupp et al. 2016). GWAS is a technique that uses high density markers located throughout the genome to identify regions associated with a trait without prior knowledge of mutations responsible for the variation in phenotype. The markers for this type of genomic analysis are SNPs because they are easily genotyped on high-definition platforms, although they have a high mutation rate and Mendelian inheritance (Guimarães et al. 2013).

Genome-wide association studies are based on the assumption that a causative mutation for a given phenotype is in linkage disequilibrium (LD) with adjacent markers in the various families of a population. This LD remains even after several generations of genome crosses and recombination, allowing for the use of these markers in a wide scan of the genome regardless of the design of segregating families. In order to capture all the chromosomal sequences in weak LD with small effects, a high-density map is necessary. The reliability provided by molecular information is lower in sheep and goats than in cattle, which is probably due to a lower linkage disequilibrium rate in small ruminants. This can be due to higher effective population sizes and the inclusion of crossbreeds in sheep and goats during data analysis. Lower LD indicates that, for some breeds, the addition of new genotypes is mandatory and that a denser SNP panel than the current Illumina’s 50 K SNP Beadchip could be beneficial (Rupp et al. 2016).

Several studies of GWAS have already been carried out on some species of domestic animals, such as dairy cattle (Ismail et al. 2017), beef cattle (Santiago et al. 2017), swine (Van Son et al. 2017), and sheep (Berton et al. 2017). Currently, there are various GWAS in goats. For instance, Martin et al. (2016) conducted a GWAS in French Alpine and Saanen goats using logistic polygenic models and estimated genetic parameters of the presence of supernumerary teats, and Martin et al. (2017) identified two new mutations in Dgat1 associated with reduced milk fat content. Mucha et al. (2017) investigated the SNPs associated with milk yield and udder, teat, feet, and legs in crossbred dairy goats.

The litter size trait is a complex quantitative trait that involves multiple genes and loci incorporating environmental factors, maternal effects such as maternal age, and intrauterine environment. So, the identification of whole candidate genes and loci associated with prolificacy becomes a challenge in modern genetics. Due to the multiple contributions of genes in the complex quantitative litter size trait, previous studies concentrated on the association of one or a few genotype polymorphism(s) in reproductive-related genes such as KISS, GDF9, POU1F1, and others that may not lead to an obvious improvement. High throughput screening for multiple genotype polymorphism combinations become more and more important. However, due to the limitation of cost and experimental design, further verification of the newly found variations in large groups is necessary (Lai et al. 2016).

Several genomic regions with major effects have been associated with litter size in goats using different methods such as PCR, RFLP, SSR, and whole-genome scanning (Table 2).

Table 2. Candidate genes and the role of regulatory genes in the litter size of goats according to breed.

Gene	Mutation	Role of regulatory genes in littER size	Breed	Authors
AA-NAT	C825T, C1249T	A key enzyme associated with melatonin (MLT) biosynthesis critical for the animal reproductive system (Sharma et al. 2015a)	Beetal, Barbari, Black Bengal, Malabari, Osmanabadi, Sangamneri, Jakhrana, Ganjam, Sirohi	Sharma et al. (2015a)
BMP15	G751T, G752C, A757C, G760C, G762A, G762T, G769C, G769A	Regulates granulosa cell proliferation and differentiation by promoting granulosa cell mitosis, suppressing follicle-stimulating hormone receptor expression, and stimulating kit ligand expression, all of which play a pivotal role in female fertility in mammals (Juengel et al. 2002)	Zaraibi, Baladi, Damasco, Alpino	Heikal and Naby (2017)
	G735A, C808G		Barbari, Beetal, Black Bengal, Jakhrana, Osmanabadi, Sangamneri, Ganjam	Ahlawat et al. (2016)
BMPR1B	T242C	Plays an important role in increasing ovulation rate and litter size (Souza et al. 2001)	Barbari, Beetal, Black Bengal, Jakhrana, Osmanabadi, Sangamneri, Ganjam	Ahlawat et al. (2015b; 2016)
BMP4	G1534A	Modulates the granulosa cell (GC) function in the follicle of domestic animals (Rajesh et al. 2018)	Barbari, Beetal, Black Bengal, Malabari, Jakhrana, Osmanabadi, Sangamneri, Sirohi, Ganjam	Sharma et al. (2013)
CART	C539A	Important in the regulation of GnRH secretion and the onset of puberty (Boone et al. 2008)	Chuandong White, Guizhou white, Gulin ma	Wang et al. (2011)
CSNS1	CSN1S1	Expressed in goat ovaries and may produce several key hormonal changes related to production traits (Wang et al. 2018)	Shaanbei, Cashmere	Wang et al. (2018)
FSHβ	A2645G	Plays a key role in the regulation of gonadal function and follicle development in mammals (Aerts and Bols 2010)	Boer, Matou, Boer–Matou crossbred	Zhang et al. (2011b)
FSHR	T70A, G130C	Relays neuronal signals from the hypothalamus to the gonads and keeps the endocrine balance in the reproductive axis maintaining the follicle growth, development, differentiation, and maturation as well as spermatogenesis (George et al. 2011)	Jining Grey, Boer, and Inner Mongolia Cashmere	Guo et al. (2013)
GDF9	G3288A, G423A, A959C, G1189A	Plays a crucial role in early folliculogenesis, and its mutations can increase ovulation and infertility in female mammals	Jining Grey, Guizhou White, Boer, Lianoning Cashmere	Feng et al. (2011)
	C818T, A959C, G1189A	(Otsuka et al. 2011)	Barbari, Beetal, Black Bengal, Jakhrana, Osmanabadi, Sangamneri, Ganjam	Ahlawat et al. (2012, 2016)
	C183A, C336T		Jinning Grey, Boer, Wendeng Dairy, Liaoning Cashmere and Beijing native	Chu et al. (2011)
GH	A781G, A1575G	A peptide hormone that is produced from the anterior pituitary gland. It acts on the development of small antral follicles in gonadotrophin-dependent stages and stimulates oocyte maturation (Silva et al. 2009)	Boer, Matou	Zhang et al. (2011a)
GnRHR	G757A	A key regulator of the reproductive system. It triggers the synthesis and release of LH and FSH in the pituitary gland that in turn regulate the production of gametes and gonadal hormones (Naor 2009)	Jining Grey, Angora, Boer, Inner Mongolia, Cashmere	Chu et al. (2009)
	G891T, G757A		Boer	Yang et al. (2011)
	A730G, G757A, G891T		Local goats in Sri Lanka	Ariyarathne et al. (2017)
	G29T		West African Dwarf	Bemji et al. (2018)
GPR54	C1122T	G protein-coupled receptor (GPR54) acts as a cognate receptor for a family of neuropeptides called kisspeptins, which are the peptide products of the KiSS1. The KiSS1/GPR54 system is considered a fundamental gatekeeper of puberty onset in mammals (Ahlawat et al. 2015a)	Black Bengal	Ahlawat et al. (2015a)
IGF1	G224A, G227C	A member of the somatotropic axis and plays a key role in the proliferation of cells, mitosis, myogenesis, meiosis, differentiation in fetal development, and post-natal growth (Thomas et al. 2017)	Attappady Black Malabari	Thomas et al. (2016)
INHbb	G693A, C840T	A chemical signal of the number of growing follicles in the ovary to the pituitary gland to reduce the secretion of FSH to a level which maintains the species-specific number of ovulations (Taya et al. 1996)	Beetal, Osmanabadi, Sangamneri, Ganjam	Sharma et al. (2015b)
KDM6A	16 bp indel	Encodes a protein that demethylates tri and dimethylated lysine 27 of histone H3 and can affect gametophyte development (Cui et al. 2018)	Shaanbei white cashmere	Cui et al. (2018)
			Laoshan	Lai et al. (2016)
KITLG	G12654A, G12772A, T12829C	Stimulates primordial follicle activation and follicle growth (Celestino et al. 2010)	Xinong Saanen, Guanzhong	An et al. (2015)
KiSS-1	T0950C, C3416T, C3811T, T3963C	Kisspeptins stimulate the secretion of gonadotropins from the pituitary by stimulating the release of GnRH from the forebrain after the activation of GPR54, which is expressed by GnRH neurons (Dungan et al. 2006)	Gondar, Woyto Guji	Mekuriaw et al. (2017)
	G296C, G2510A, C2540T		Beetal, Barbari, Black Bengal, Malabari, Osmanabadi, Sangamneri, Jakhrana, Ganjam, Sirohi	Maitra et al. (2014)
LHβ	C59A, C543T	Confers its specific biological action and is responsible for the specificity of the interaction with the LH receptor (Long et al. 2009)	Boer	Li et al. (2011)
PRLR	C113T, C59T	Member of the growth hormone/prolactin receptor gene family, which is a mediator of prolactin (Ahlawat et al. 2015c)	Boer	Li et al. (2011)
SETDB2	C1540T	After ovulation, the expression of SETDB2 plays a potential role in the early stages of luteinization (Maekawa et al. 2016)	Laoshan	Lai et al. (2016)

arylalkylamine-N-acetyltransferase (AA-NAT), bone morphogenetic protein 15 (BMP15), bone morphogenetic protein receptor 1B (BMPR1B), bone morphogenetic proteins 4 (BMP4), cocaine-amphetamine-regulated transcript (CART), alpha-S1-casein (CSNS1), follicle stimulating hormone beta (FSHβ), follicle stimulating hormone receptor (FSHR), growth differentiation factor 9 (GDF9), growth hormone (GH), gonadotropin releasing hormone receptor (GnRHR), G protein-coupled receptors 54 (GPR54), insulin-like growth factor 1 (IGF1), inhibin (INHββ), lysine demethylase 6A (KDM6A), kit ligand (KITLG), kisspeptin (KiSS-1), luteinizing hormone beta-subunit (LHβ) and prolactin receptor (PRLR), SET domain, bifurcated 2 (SETDB2).

Currently, whole-genome sequencing and GWAS are used to explore genetic variants strongly associated with production traits. However, numerous potential genes identified by GWAS have not been fully verified. To address this problem, methods which combine GWAS analysis results and MAS to screen for critical genetic variations (Martin et al. 2017) in large livestock populations have been developed (Cui et al. 2018).

Besides the mentioned approaches, Miao et al. (2016) used RNA-Seq technology to perform genome-wide sequencing of mRNAs prepared from the ovaries of the Jining Grey goat (high prolificacy) and Laiwu Black goat (lower prolificacy). These authors predicted 3,493 candidates as new genes and 742 genes were differentially expressed between these two goats. Functional annotation of these genes identified various biological functions and signalling pathways that hold clues as to which genes might be involved in regulating goat fecundity and prolificacy.

4. Genomic selection for prolificacy in goats

Reproductive traits generally have low heritability and do not exhibit large responses to phenotypic selection. Therefore, the inclusion of genetic information on the genes associated with reproductive ability could enhance selection response. In this sense, genomic best linear unbiased predictions (GBLUP) of breeding values can be promising tools for the genetic improvement of goat reproductive performance.

Regarding the litter size of goats, selection programmes have generally considered only the phenotypic and pedigree information of the animals (Shrestha and Fahmy 2007; Mohammadi et al. 2012; Castañeda-Bustos et al. 2014; Zhou et al. 2014). Due to the low heritability of litter size (0.08–0.18), its rate of genetic improvement is slow (Hamed et al. 2009; Gunia et al. 2010; Santos et al. 2013; Menezes et al. 2016). Therefore, the inclusion of genomic information in the genetic evaluation of prolificacy in goats would greatly increase the accuracy of selection. According to Amills et al. (2017), the implementation of these new molecular technologies, such as whole-genome sequencing and genome-wide genotyping, allows for the exploration of caprine diversity at an unprecedented scale, thus providing new insights into the evolutionary history of goats. Genomic selection and genome editing can also be applied to the improvement of traits that cannot be modified easily by traditional selection.

In the traditional methods of genetic evaluation, breeding values are predicted using only pedigree information and this results in lower accuracy. On the other hand, genomic evaluation methods can substantially improve the accuracies of GBV estimation when applied to small ruminants and, therefore, accelerate the response to selection. The basic building blocks of conventional breeding programmes for small ruminants in most developing countries are lacking. Thus, genomic information can help overcome some of the limitations through parent verification using the genomic matrix to estimate GBV (Mrode et al. 2018).

Indeed, the accuracy of methods that only use the phenotypes of the genotyped animals and ignore records of the non-genotyped part of the population (e.g. GBLUP and BLUP-SNP) is limited when the reference population is small (Rupp et al. 2016). Therefore, a single-step approach is the recommended method for such small reference populations. The single-step approach combines all of the available phenotypic, pedigree, and genomic information in a single-step procedure to calculate genomic breeding values (Legarra et al. 2009), thus avoiding bias in the estimation of GBV due to the pre-selection of candidates (Rupp et al. 2016).

Using Illumina’s 50 K SNP Beadchip caprine, Mucha et al. (2015) estimated the GBV for milk production in crossbred genotyped dairy goats. They compared two methods for the estimation of GBV, BLUP at the single nucleotide polymorphism level (BLUP-SNP) and single-step BLUP. Results showed that the single-step approach had a higher GBV accuracy in comparison with BLUP-SNP. In a genetic evaluation study in Spanish goats, Molina et al. (2018) related that the correlation between A (pedigree relationship matrix) and G (genomic relationship matrix) was 0.826 and the correlation between BV and GBV was 0.989. These authors showed an increase of 1.06% in the average reliability of estimates in the single-step genomic BLUP compared to the traditional BLUP evaluation.

5. Future perspectives

The advancement of molecular genetic techniques has made significant contributions to the progress of animal selection. Whole-genome scanning and genome-wide association studies have shown greater accuracy in the identification of major genes related to litter size in goats.

The use of a genomic matrix in the estimation of GBV strongly contributes to the acceleration of genetic gains by selection. These approaches can be used to impact the litter size of goats because they offer quick results in relationship verification. This can help overcome the problem of pedigree inconsistency; many small ruminant flocks have weak genetic connections due to unsuitable pedigrees. Thus, GBV prediction ⁣⁣provides greater accuracy in goat litter size selection. This not only results in a greater number of progenies but also allows higher selection intensity for all other traits considered in a selection programme.

Evidence has shown that the use of genomic information to select for goat litter size creates the opportunity to increase genetic gains. However, these technologies are underutilized in goats and more multidisciplinary research in this area must be carried out.

Disclosure statement

No potential conflict of interest was reported by the authors.

ORCID

Raisa Rodrigues Rios http://orcid.org/0000-0002-1012-1055

Breno Araújo de Melo http://orcid.org/0000-0002-2277-4125

Lays Thayse Alves dos Santos http://orcid.org/0000-0002-0653-1094

Kleibe de Moraes Silva http://orcid.org/0000-0001-5056-6828

Angelina Bossi Fraga http://orcid.org/0000-0001-6557-3000

Additional information

Funding

This work was supported by Coordenação de Aperfeiçoamento de Pessoal de Nível Superior: [grant number 001].