Genomic insights into the population history and adaptive traits of Latin American Criollo cattle

Abstract

Criollo cattle, the descendants of animals brought by Iberian colonists to the Americas, have been the subject of natural and human-mediated selection in novel tropical agroecological zones for centuries. Consequently, these breeds have evolved distinct characteristics such as resistance to diseases and exceptional heat tolerance. In addition to European taurine (Bos taurus) ancestry, it has been proposed that gene flow from African taurine and Asian indicine (Bos indicus) cattle has shaped the ancestry of Criollo cattle. In this study, we analysed Criollo breeds from Colombia and Venezuela using whole-genome sequencing (WGS) and single-nucleotide polymorphism (SNP) array data to examine population structure and admixture at high resolution. Analysis of genetic structure and ancestry components provided evidence for African taurine and Asian indicine admixture in Criollo cattle. In addition, using WGS data, we detected selection signatures associated with a myriad of adaptive traits, revealing genes linked to thermotolerance, reproduction, fertility, immunity and distinct coat and skin coloration traits. This study underscores the remarkable adaptability of Criollo cattle and highlights the genetic richness and potential of these breeds in the face of climate change, habitat flux and disease challenges. Further research is warranted to leverage these findings for more effective and sustainable cattle breeding programmes.

Headnote

Criollo cattle, the descendants of animals brought by Iberian colonists to the Americas, have been the subject of natural and human-mediated selection in novel tropical agroecological zones for centuries. Consequently, these breeds have evolved distinct characteristics such as resistance to diseases and exceptional heat tolerance. In addition to European taurine (Bos taurus) ancestry, it has been proposed that gene flow from African taurine and Asian indicine (Bos indicus) cattle has shaped the ancestry of Criollo cattle. In this study, we analysed Criollo breeds from Colombia and Venezuela using whole-genome sequencing (WGS) and single-nucleotide polymorphism (SNP) array data to examine population structure and admixture at high resolution. Analysis of genetic structure and ancestry components provided evidence for African taurine and Asian indicine admixture in Criollo cattle. In addition, using WGS data, we detected selection signatures associated with a myriad of adaptive traits, revealing genes linked to thermotolerance, reproduction, fertility, immunity and distinct coat and skin coloration traits. This study underscores the remarkable adaptability of Criollo cattle and highlights the genetic richness and potential of these breeds in the face of climate change, habitat flux and disease challenges. Further research is warranted to leverage these findings for more effective and sustainable cattle breeding programmes.

Keywords:

cattle, Latin America, microevolution, population genomics, selection, thermotolerance

1. Introduction

The term 'Criollo', with origins in the Portuguese 'Crioulo', historically distinguished people born in the New World from those native to Iberia and was subsequently extended to livestock (cattle, sheep, horses and goats). European livestock production in the Americas can be traced back to the second voyage of Columbus in 1493, when cattle, among other animals, sourced from La Gomera in the Canary Islands were brought to the island of Hispaniola [1-3]. The adoption and growth of cattle agriculture in Hispaniola was rapid; however, the spread of animals to the South American continent was more gradual. This was documented by early colonial accounts, illustrating a transformative ecological and economic process that started in the Caribbean, then Mexico, and eventually reached the Llanos of Colombia and Venezuela where, by 1600, cattle ranching had become a significant industry [4,5]. Over the ensuing five centuries, increasing numbers of cattle were exported to North and South America particularly as the process of European coloniation intensified during the eighteenth and nineteenth centuries [6]. Evidence for a similar Iberian origin narrative has been described for sheep [7], pigs [8] and horses. For example, recent work has shown that historical and extant North American horse populations exhibit pronounced genetic affinities with Iberian horse populations [9]. The ancestry of modern South American cattle has also been shaped by indicine (zebu) cattle (Bos indicus) that were directly introduced from South Asia-primarily present-day India and Pakistan [10].

The extent of contributions from cattle outside Europe and the Indian subcontinent to the South and Central American cattle gene pools remains uncertain. Modern Iberian cattle populations, whose ancestors probably contributed the bulk of Latin American Criollo cattle ancestry, have an African taurine cattle genomic component [11-13]. In parallel to this, it has also been posited that African cattle may have been directly introduced into South America alongside imports from Europe [14]. This hypothesis is supported by the presence of the Tl-c mtDNA haplotype at a frequency of 8% in Criollo cattle; Tl-c is prevalent in African cattle and found at low frequencies in Iberian populations [15]. It is also supported by recent findings of colonial-era mitogenomes, which reveal the presence of African haplogroups in seventeenth-century cattle from Mexico [16]. Recent work suggests that the degree of African ancestry in Criollo cattle depends on the breed. Notably, the Guadeloupe breed, indigenous to the Caribbean, traces almost 35% of its genetic ancestry back to African taurine cattle [17].

The cattle that were brought to the Americas from Iberia in the fifteenth and sixteenth centuries were adapted to Mediterranean agroecologies and, over the following centuries, these populations have evolved adaptations to tropical and arid environments, emerging as distinctive breeds with unique heat tolerance and disease resistance traits [12,18-21]. However, Criollo cattle are undervalued in modern production systems and face gradual replacement by more productive commercial breeds, which is eroding indigenous cattle genetic resources in Latin America [22]. Using genome-scale data, this study aims to evaluate B. taurus and B. indicus ancestry and detect historical gene flow from African cattle into the Blanco Orejinegro, Hartón del Valle and Criollo Limonero breeds. Additionally, we use high-resolution analyses of genetic structures, and selection signatures to provide new information about the microevolutionary histories of these populations.

2. Material and methods

2.1. Cattle samples, library preparation, high-throughput sequencing and variant calling

In total, 34 Criollo cattle were used to generate high-quality de novo whole-genome sequence (WGS) datasets. Coat hair follicles obtained from 24 Colombian and Venezuelan cattle and semen straws from 10 Colombian AI bulls were used to obtain genomic DNA using standard commercial procedures for livestock DNA extraction and purification (Weatherbys Scientific, Co. Kildare, Ireland). The WGS data generation was performed by sequencing at a mean depth of 20x using two commercial service providers (Novogene, Cambridge, UK, and Macrogen, Seoul, South Korea). Figure 1 shows the geographical origins of the three main Criollo breeds in this study.

Publicly available WGS data were also assembled for 37 European taurine, 13 American indicine and 13 American Criollo cattle. Sequence reads were mapped to the bovine reference genome (ARSUCD1.2) [24] using the default parameters for bwa-mem2 v. 2.2.1 [25]. Samtools v. 1.13 [26] was used to sort bam files and create index files. Potential PCR duplicates were identified using the 'MarkDuplicates' function of Picard v. 2.27.1 (http://broadinstitute.github.io/picard). The 'BaseRecalibrator' and 'ApplyBQSR' functions of the Genome Analysis Toolkit (GATK) v. 4.2.6.1 [27] were used to perform base quality score recalibration.

Candidate single-nucleotide polymorphisms (SNPs) were called from the bam files using the 'HapIotypeCaller' function in the GATK with the '-emit-ref-confidence GVCF' option. Individual GVCF files were merged using the 'GenomicsDBImport' function. SNPs from the genomics database were called and selected using 'GenotypeGVCFs' and the 'SelectVariants' function. The 'VariantFiltration' function of the GATK was used to avoid possible false-positive calls according to the GATK best practices: SNP clusters with '-cluster-size 3' and '-cluster-window-size 10', quality by depth 'QD < 2', read position rank sum test 'ReadPosRankSum < -8', phred-scaled variant quality score 'QUAL < 30', Fisher strand 'FS > 60', mapping quality 'MQ < 40', mapping quality rank sum test 'MQRankSum < -12.5' and strand odd ratio 'SOR > 3'. Triallelic SNPs and those with a minor allele frequency of less than 0.01 and/or missing genotype rates of greater than 0.1 were filtered out. The SNPs remaining in each WGS dataset were then annotated according to their positions using SnpEff [28]. The combined panel of 97 cattle samples with WGS data is detailed in the electronic supplementary material, table SI.

2.2. BovineHD 777 к array data

For broader comparative population genomics analyses, publicly available BovineHD 777 k SNP array data were used. These were 91 European taurine cattle representing two breeds, 82 Asian and American indicine cattle representing four breeds [29,30], 45 Iberian cattle representing 13 breeds [31] and 31 African taurine cattle representing two breeds [29,32]. In addition, 12 Criollo cattle representing the Senepol breed were obtained from the WIDDE database [33]. SNP positions were updated from the UMD3.1 assembly [34] to ARS-UCD1.2 following Riggio et al. [35]. Subsequent processing of these data was conducted using PLINK v. 1.90b6.25 [36] and R v. 4.1.3 [37]. The combined panel of 261 cattle samples with BovineHD 777 k SNP data is detailed in the electronic supplementary material, table S2.

2.3. Population differentiation, genetic structure and demographic analyses

Principal component analysis (PGA) for all animals (WGS and BovineHD genotype data) was performed using PLINK v. 1.90b6.25 and the results were plotted using ggplot2 v. 3.4.0 [38] in the R v. 4.1.3 environment. The genetic structure of each population was visualized using ADMIXTURE v. 1.3.0 [39] with К = 3-6 unsupervised modelled ancestries and ggplot2 v. 3.4.0. Demographic modelling of current and historical effective population size (Ne) trends was conducted using the SNeP software tool [40] with WGS data for the Blanco Orejinegro, Hartón del Valle, Limonero, N'Dama and Holstein populations.

2.4. Detection of historical admixture

Using ADMIXTOOLS [41] with default parameters, we calculated/3 statistics using the combined WGS and BovineHD genotype dataset to test for evidence that three Criollo breeds (Blanco Orejinegro, Hartón del Valle and Limonero) and three Iberian breeds (Limia, Pajuna and Sayaguesa) are derived from the admixture of two sets of populations (European taurine populations and indicine cattle, and European taurine populations and African taurine cattle). A significant negative/3 value is considered evidence of historical admixture in the target population (Criollo or Iberian). For visualization of the/3 statistics, we plotted the /3 values against the population trios with error bars indicating the standard errors using ggplot2 v. 3.4.0 [38].

2.5. Test statistics for the detection of selection signatures

For the selection signature analysis, only the WGS dataset was used, which included approximately 9.7 million SNPs genotyped in 97 cattle. SNP-by-SNP FgT values were calculated for each Criollo breed versus a panel of Holstein cattle using VCFtools v. 0.1.16 [42]. XP-EHH statistics were calculated using the program seiscan v. 1.2.0 [43] with default parameters, where each Criollo breed was also compared with the Holstein panel. The unstandardized XP-EHH values were then normalized using the -norm function in seiscan. We calculated the ASAF (directional change in the selected allele frequency) using the -freq function of VCFtools to generate allele frequencies, and then using a custom R script, we calculated the ASAF for each Criollo breed versus the Holstein panel. Similarly to the XPEHH values, ASAF values were standardized to Z ~ N(0,l) as described by Randhawa el al. [44].

2.6. Implementing the composite selection signal methodology

To obtain composite selection signal (CSS) statistics [44] for each Criollo breed, the FgT/ XP-EHH and ASAF statistics were combined using the SNP ID (chromosome:position), which produced the hnal sets of SNPs for the CSS analysis of 9 344 833, 9 465 007 and 9 660 647 million SNPs, for the Blanco Orejinegro, Hartón del Valle and Limonero breeds, respectively. Following this, the MINOTAUR R package [45] was used to calculate CSS statistics, which were then averaged over 20 kb windows to reduce spurious signals. In addition, a signal was only considered significant if at least one SNP in 0.1% of CSS scores was flanked by at least five other SNPs in the top 1% of CSS values [44,46].

2.7. Identification of genomic regions under selection

From the top 0.1% of SNPs identified, genomic intervals in the format chromosome:start-end were extracted and the BovineMine resource (v. 1.6) [47] was then used to find genes of ±500 kb from the start and end of these regions. The functional relevance of these CSS-derived gene sets was then evaluated using the scientific literature and the g:Profiler software tool [48], which was used to identify over-represented gene ontology (GO) terms for each of the CSS-derived gene sets by breed. Following this, BovineMine v. 1.6 was used to systematically identify previously described quantitative trait loci (QTLs) for production, health and welfare traits that are located within the CSS cluster regions for each breed.

3. Results

3.1. Population differentiation, genetic structure and admixture in Criollo cattle

The merged WGS and remapped BovineHD 777K BeadChip datasets yielded 302 012 shared SNPs for 308 individual animals. A PCA of these SNPs demonstrated that PCI and PC2 accounted for 44.7% and 12.9% of the variation observed across the first 20 PCs, with PCI representing the split between the B. taurus and the B. indicus lineages and PC2 representing the split between the European and African taurine lineages (figure 2a). The genetic structure was analysed using ADMIXTURE, with cluster numbers (K) ranging from 3 to 6 (figure 2b). At К = 3, we recovered the three clusters observed in the PCA, which delineated the European B. taurus and African B. taurus and B. indicus ancestry components, respectively. The subsequent split at К = 4 captured an additional cluster within the European taurine populations that differentiated Jersey cattle. At К = 5, a Holstein cattle ancestry component was evident, and at К = 6, a split was observed that divided the B. indicus ancestry into two clusters.

Further examination of the PCA and structure plots in figure 2a,b showed that the Criollo cattle samples cluster with the Iberian taurine cattle samples, all of which show relatively uniform levels of African taurine admixture at К = 3 and К = 4. In addition, the Senepol and Hartón del Valle Criollo breeds exhibit a consistent and relatively uniform ancestry component from the B. indicus lineage, while the Blanco Orejinegro, Costeño con Cuernos, Limonero and Sanmartinero breeds exhibit more heterogeneous patterns of indicine admixture. Conversely, /3 statistics [41] indicate that there is no evidence of admixture from either African taurine or indicine cattle in the Blanco Orejinegro and Limonero breeds when using any of the European taurine reference populations and African taurine or indicine breeds as the second reference (electronic supplementary material, figure SI# and c). However, there is some evidence of indicine admixture in the Hartón del Valle as the /3 statistics for all population trios that used any European taurine reference population with an indicine breed as the second reference population produced negative /3 values for this breed (electronic supplementary material, figure Sib). This is also evident from the PCA (figure 2#) and the structure plot (figure 2b).

3.2. Selection signals across Criollo cattle breed genomes

The CSS test identified selection signals for the three Criollo cattle breeds that had WGS data and sufficient sample sizes (Blanco Orejinegro, Hartón del Valle and Limonero). For each breed, the test highlighted several selection signals spanning multiple chromosomes. For the Blanco Orejinegro breed, the CSS test detected 31 cluster regions spanning 17 chromosomes, with 442 genes located within these regions and extended ±500 kb (figure 3#). Similarly, for the Hartón del Valle breed, the CSS test revealed 31 distinct clusters distributed over 17 chromosomes, encompassing 410 genes (figure 3b). Lastly, for the Limonero breed, the CSS test detected 38 cluster regions, also extending across 17 chromosomes and that contained 451 genes in total (figure 3c). When these results were consolidated, 76 genes within selection signature regions were shared across the three breeds, and BTA18 was the chromosome with the largest number of these genes (28 genes; 36.8% of the total). The CSS cluster regions and the genes within these regions are detailed by breed in the electronic supplementary material, tables S3 and S4.

3.3. Gene ontology term enrichment analysis

The functional significance of the genes within the CSS cluster regions for each breed was evaluated using GO term over-representation analysis with the g:Profiler software tool. After adjusting for multiple tests and using a p-value adjusted for the false discovery rate (FDR-padj. < 0.05), over-represented GO terms were catalogued for each breed. The results of the GO term over-representation analysis are shown in the electronic supplementary material, tables S5-S7 and figure 4. For both the Blanco Orejinegro (figure 4#) and Limonero (figure 4b) breeds, the GO terms related primarily to immunobiology, encompassing functions such as immunoglobulin receptor binding, phagocytosis, complement activation and antigen binding. Conversely, for the Hartón del Valle breed (figure 4c), the predominant GO terms were associated with skin epidermis and keratinization processes.

3.4. Identification of quantitative trait loci within composite selection signal cluster regions

QTLs within the CSS cluster regions were systematically identified using BovineMine v. 1.6. The number of QTLs identified varied across the breeds: 519 in Blanco Orejinegro, 332 in Hartón del Valle and 370 in Limonero. Interestingly only 11 QTLs were common to all three breeds and a QTL associated with tick resistance [49] was particularly noteworthy. In the case of Blanco Orejinegro, most of the identified QTLs (291 out of 519) in the CSS cluster regions were associated with calving ease. Additionally, 46 QTLs were linked to milk protein percentage, 40 to milk yield and 24 to coat texture. For the Limonero breed, there were 292 QTLs located in the CSS cluster regions related to calving ease; however, no other traits had more than five SNP markers among the QTLs. In the case of the Hartón del Valle breed, the most notable traits represented among the CSS cluster region QTLs were milk-related. Specifically, 86 QTLs were associated with milk protein percentage, 45 with milk yield and 35 with milk fat percentage. Additionally, 27 QTLs in Hartón del Valle were associated with coat texture. The results of the QTL search using BovineMine v. 1.6 are detailed in electronic supplementary material, table S8.

4. Discussion

4.1. Population structure and admixture

The differentiation of cattle samples across PCI of the PCA plot (figure 2a) reflects the evolutionary divergence between the B. taurus and the B. indicus lineages, which occurred 150-500 ka [50,51]. The second division evident for PC2 in the PCA plot (figure 2a) differentiates the European and African taurine groups, a separation that has been well documented in several previous studies using genomewide SNPs and other autosomal genetic markers [52-55]. These PCs are mirrored by the clusters observed at К = 3 in the structure analysis (figure 2b), which differentiate the European taurine, African taurine and indicine groups and reveal the patterns of shared ancestry among the major groups. Extending these observations further, the complex patterns of admixture owing to human-mediated and unmanaged cross-breeding between taurine and indicine cattle have also been described in cattle populations across several geographical regions, particularly in sub-Saharan Africa [53-55].

Previous studies have reported African taurine autosomal admixture in Criollo cattle breeds [12,17]. Results from the present study, which are captured in the PCA plot (figure 2a) and structure analysis (figure 2b) suggest this is also the case for the Criollo breed population samples analysed here (Blanco Orejinegro, Costeño con Cuernos, Hartón del Valle, Limonero, Sanmartinero and Senepol), albeit at lower levels compared with that observed for Guadeloupe cattle [17]. Conversely, the positive /3 statistics observed for the three Criollo breeds examined using this method (Blanco Orejinegro, Hartón del Valle and Limonero) do not support this (electronic supplementary material, figure Sla-c). It should be noted, however, that a positive/3 value does not necessarily imply a lack of admixture, since excessive genetic drift in the target population can obscure an admixture signal [41]. In this regard, the demographic modelling of historical effective population sizes (Ne values) and a comparison with Holstein, a commercial dairy population with a markedly low Ne [56], indicate that founder effects and subsequent genetic drift may have impacted the Blanco Orejinegro, Hartón del Valle and Limonero breeds (electronic supplementary material, figure S2).

For B. indicus ancestry, all the Latin American Criollo cattle populations examined here exhibit varying degrees of indicine admixture (figure 2b). Interestingly, however, the Blanco Orejinegro and Hartón del Valle both originate in Colombia but do not cluster together, which is presumably a consequence of the relatively modest indicine ancestry in the Blanco Orejinegro. The Limonero breed samples in this study largely cluster together except for four animals that exhibit a significant proportion of indicine ancestry. These results, taken together with the /3 statistics for the Blanco Orejinegro, Hartón del Valle and Limonero (electronic supplementary material, figure Sla-c), and the histories of these breeds (the Blanco Orejinegro breed has been in a conservation programme since 1940 [57]), suggest that indicine gene flow only had a long-term impact on breeds such as the Hartón del Valle, which exhibits relatively uniform levels of indicine admixture compared with the Limonero, where it is more heterogeneous (figure 2b). This has implications for understanding the selection signatures that are described in the following sections: in some cases, the selection signals detected may be caused by the adaptive introgression of B. indicus haplotypes.

Despite the inclusion of a diverse panel of European taurine, African taurine, Asian indicine and admixed cattle in the design and validation of the BovineHD SNP array [58], we acknowledge that ascertainment bias could have influenced the results from some of the analyses of population differentiation and genetic ancestry. However, we think that this is unlikely, because ascertainment bias typically affects analyses such as selection signal detection that use individual SNP locus frequencybased statistics (e.g. Fst) substantially more than genome-wide multi-locus dimension reduction tools like ADMIXTURE and PCA [59-62]. Importantly, in this regard, we used only WGS data for the CSS analyses (figure 3 and electronic supplementary material, table S3).

4.2. Genomic signatures of human-driven selection in Criollo cattle

Blanco Orejinegro are distinctive among Criollo cattle for their white coats, often speckled with black spots; they also have highly pigmented skin as exemplified by their black ears. This coat-skin pattern also occurs in British White and White Park cattle [63], and for White Park, it has been demonstrated that this is caused by a duplication in the KIT gene and an aberrantly inserted KIT gene on BTA29 [64]. Additional genes that harbour cattle coat colour polymorphisms include MC1R, TWIST2, MITT, PAX3, SLC45A2, COPA, TYR, TYRPI, KITLG and ASIP. For the Blanco Orejinegro breed, the genes PAX3, SLC45A2 and KITLG are in the genome regions with evidence for directional selection based on the CSS results. Mutations in PAX3 have been shown to result in a spectrum of phenotypic outcomes in horses from white spotting to a uniform white coat trait [65]. Similarly mutations in the bovine SLC45A2 gene have been shown to cause oculocutaneous albinism [66], while the KITLG gene has been associated with the roan coat type in cattle [67].

Interestingly for the Hartón del Valle and Limonero breeds, MC1R was part of a ETAIS CSS cluster region. The protein receptor encoded by MC1R regulates tyrosinase levels in melanocytes, an enzyme critical for melanin synthesis. High concentrations of tyrosinase facilitate the production of eumelanin, contributing to darker brown or black shades, whereas lower concentrations drive the synthesis of phaeomelanin, yielding lighter red or yellow tones [68]. It is therefore noteworthy that Limonero and Hartón del Valle cattle generally possess a 'bayo' or bay coat, which is a yellow to light brown/red colour [69,70]. Though not in a CSS cluster region in the Blanco Orejinegro, a notable peak close to MC1R is present on BTA18. This gene has been observed to be under positive selection in B. indiens cattle resulting in a light grey to white coat in the Brahman and Nelore breeds [71]. The adaptive significance of functional polymorphisms at MC1R is that lighter coat colours reflect a significant portion of incident solar radiation, which makes these cattle more suitable for tropical environments [72].

4.3. Selection for heat tolerance in Criollo cattle

Criollo cattle breeds such as the Blanco Orejinegro, Hartón del Valle and Limonero have evolved over five centuries to thrive in challenging tropical agroecologies. In the context of this environmental adaptation, a key finding for all three breeds is the presence of a CSS peak on BTA20 that encompasses the PRLR gene, most notably in the Blanco Orejinegro and Hartón del Valle breeds, where it was statistically significant. Mutations in this gene have been shown to cause the 'slick' phenotype in Criollo breeds such as the Blanco Orejinegro, Hartón del Valle, Limonero, Romosinuano, Costeño con Cuernos, Mexican Criollo Lechero Tropical and Criollo-influenced composite breeds such as the Carora and Senepol [73]. Of particular interest for understanding how convergent evolution can act in livestock populations is the existence of several different PRLR mutations that can produce the 'slick' phenotype in Criollo cattle [73]. Mutations in the bovine PRLR gene can have major effects on the length and the structure of hair coats providing improved thermotolerance and concomitant increases in fertility and milk yields in cattle populations that inhabit dry and tropical conditions [18,69,74]. In addition, it has been shown that these mutations can act pleiotropically and cause other physiological changes [75].

Variation in other genes may also confer enhanced thermotolerance in Criollo cattle; for example, MVD, previously implicated as being under selection in North African cattle [76], was found in a BTA18 CSS cluster region for all three breeds. In addition, CCM2, which regulates heart and blood vessel formation and integrity, is located in a Limonero BTA4 CSS cluster region and was also highlighted by Ben-Jemaa and colleagues in North African cattle [76]. This may represent an adaptation that increases heat transfer from the interior to the skin, which is also supported by the enhanced vascularization of the dermis that has been documented in Limonero cattle [69]. Another gene, SESN2, located in a BTA2 CSS cluster detected in the Hartón del Valle and the Blanco Orejinegro breeds, may also be associated with adaptation to tropical environments. This gene was observed to be downregulated in a functional genomics study of heat-stress responses in dermal fibroblast cells from indicine and indicine x taurine cross-bred cattle [77].

The BTA11 Limonero CSS cluster region contains FBXO4 that encodes a protein which regulates body temperature through interactions with the heat shock protein HSPB6 (previously HSP20) [72,78]. We also observed that DNAJA2 was in a Blanco Orejinegro BTA18 CSS cluster region. This gene encodes a member of the DNAJ/HSP40 family of proteins and shows increased expression in peripheral blood from Holstein calves exposed to heat stress [79]. Similarly, HIGD1A (BTA22 CSS cluster) and GBFA2T3 (a BTA18 CSS cluster distinct from that containing DNAJA2) encode proteins associated with cellular responses to hypoxia [80,81], which could be indicative of adaptation to oxidative stress caused by high temperatures [82].

4.4. Genomic signatures of selection for fertility and reproductive traits in Criollo cattle

Compared with taurine breeds from temperate zones, Criollo cattle are renowned for high fertility and excellent reproductive performance [21,83], and using the CSS method we detected several genes associated with these traits. For example, in the Blanco Orejinegro breed, a BTA7 CSS cluster region contains the CATSPER3 gene, which encodes a sperm-specific ion channel protein directly linked to sperm motility and male fertility in many species, including cattle [84]. Another CSS cluster region in the Blanco Orejinegro on BTA20 contained SPEF2, which is involved in the formation and functionality of sperm flagella [85]. Additionally, the CREM gene, implicated in spermatogenesis, was located within a Blanco Orejinegro CSS cluster region on BTA13 [86]. Other genes related to reproductive physiology were located within CSS cluster regions for the Hartón del Valle breed including HKDC1 on BTA28, which exhibits high expression in the testes [87], and MISP3 on BTA7, which is implicated in spermatogenesis [88]. For the Limonero breed, the CSS cluster regions included one on BTA10 that contains FSIP1, a gene crucial for normal spermiogenesis and flagella development [89]. In addition, the PTGES gene within a Limonero CSS cluster region on BTA11 encodes an enzyme involved in the synthesis of prostaglandin E2 (PGE2), which is a signalling molecule with a crucial role in reproductive processes [90]. In parallel to this, the QTL analysis (electronic supplementary material, table S8) for the Blanco Orejinegro breed identified QTLs predominantly associated with reproduction, notably calving ease and lactation. These findings corroborate previous results obtained using SNP array data for Blanco Orejinegro cattle [20,91]. The results of the QTL analysis in the CSS cluster regions for the Limonero breed were similar to those for the Blanco Orejinegro breed.

4.5. Genomic adaptations to infectious diseases in Criollo cattle

Comparable to reproductive and fertility traits, desirable disease resistance and tolerance traits have been highlighted as important features of Criollo cattle populations. However, empirical evidence for the genetic basis of these traits is often lacking; instead, much of the evidence base is anecdotal and derived from the practical knowledge and experiences of veterinarians, farmers and breeders [70]. In the present study, however, we have begun to address this knowledge gap through the identification of several immune genes in CSS cluster regions for the Blanco Orejinegro, Hartón del Valle and Limonero breeds. For example, the CBFA2T3 gene, located in the BTA18 CSS cluster region detected for all three breeds and discussed in the context of hypoxia, has a role in the bovine immune response to mammary gland inflammation [92]. Similarly, CXCL14, located in a BTA7 CSS cluster region for the Blanco Orejinegro and Limonero breeds, has also been implicated in the immune response to bovine mastitis caused by Staphylococcus aureus [93]. In addition, for the Limonero, a В TAXI CSS cluster region contains the LRRC8A gene, which is essential for the development and function of T lymphocytes [94].

In tropical and subtropical regions, there is a significant prevalence of vector-borne haemoparasitic infections [95] that induce anaemia, fever, wasting and reproductive dysfunction [96]. These diseases are primarily caused by eukaryotic parasites within the Babesia [97] and Trypanosoma [98] genera, and bacterial pathogens from the Anaplasma genus [99]. It is noteworthy, therefore, that the EPB41 gene located within a Blanco Orejinegro and Hartón del Valle BTA2 CSS cluster region is associated with resistance to anaemia and trypanotolerance in African cattle [100]. Also, for the Blanco Orejinegro and Limonero, RPS27 and RPS20 are located in CSS cluster regions on BTA3 and BTA14, respectively, and mutations in these genes can cause Diamond-Blackfan anaemia [101,102]. Furthermore, in the Blanco Orejinegro breed, the ZFPM1 gene within a BTA18 CSS cluster region has been shown to play a role in cardiac and haematopoietic development. In a murine Trypanosoma congolense infection model, mice exhibiting decreased ZFPM1 expression recovered more effectively from anaemia [103].

Gene GO term analysis in the Criollo CSS cluster regions revealed a marked over-representation of immunobiological processes for the Blanco Orejinegro and Limonero breeds, many of which overlapped between the two breeds (figure 4a, c and electronic supplementary material, tables S5 and S7). The statistically significant over-represented biological process GO terms in the Blanco Orejinegro and Limonero included several terms related to intracellular pathogens, including phagocytosis (G0:0006909), phagocytosis and engulfment (G0:0006911), and phagocytosis and recognition (G0:0006910). It is therefore noteworthy that Blanco Orejinegro cattle have a documented resistance to brucellosis [104], a disease characterized by abortion and retention of the placenta, which is principally caused by

Brucella abortus, an intracellular bacterium that subverts phagocytic pathways and processes to enter and manipulate host cells [105].

5. Conclusions

This study provides valuable insights into the genomic basis of microevolutionary change in Criollo cattle as they have adapted to the tropical environments of Latin America. Evidence of strong selective pressure was apparent, particularly for the distinct coat and skin coloration traits observed in these breeds, which are advantageous in cattle populations exposed to significant levels of incident solar radiation. Notably, we also discovered genomic selection signatures that may be associated with thermotolerance, again underscoring adaptation to hot climates. In addition, some of the selection signatures we identified align with the documented fertility traits in Criollo cattle. Moreover, functional over-representation analysis revealed many genes related to immune function, which could reflect resilience to multiple infectious disease challenges. Taken together, our results show the remarkable adaptability of Criollo cattle, which has been driven by natural and human-mediated selection, and underscore the genetic richness and value of these breeds for future breeding programmes.

Ethics. This work did not require ethical approval from a human subject or animal welfare committee.

Data accessibility. New genome sequence data is available at [106].

Electronic supplementary material is available online [107].

Declaration of Al use. We have not used Ai-assisted technologies in creating this article.

Authors7 contributions. J.A.W.: conceptualization, data curation, formal analysis, methodology, software, visualization, writing-original draft, writing-review and editing; S.I.N.: data curation, formal analysis, methodology writing -review and editing; I.A.S.R.: formal analysis, methodology, software, writing-review and editing; G.P.M.: formal analysis, methodology, software, writing-review and editing; J.F.O.: formal analysis, methodology, software, writing-review and editing; J.M.F.: conceptualization, resources, writing-review and editing; J.A.B.: methodology, writing-review and editing; A.M.P.O.: conceptualization, resources, writing-review and editing; A.J.L.-H.: conceptualization, resources, writing-review and editing; J.F.G.: conceptualization, resources, supervision, writing-review and editing; T.S.S.: conceptualization, funding acquisition, project administration, supervision, writing-review and editing; L.A.F.F.: conceptualization, data curation, methodology, resources, writing-review and editing; M.S.-Т.: conceptualization, methodology, supervision, writing-review and editing; D.E.M.: conceptualization, funding acquisition, project administration, supervision, writing-original draft, writing -review and editing.

All authors gave final approval for publication and agreed to be held accountable for the work performed therein.

Conflict of interest declaration. We declare we have no competing interests.

Funding. J.A.W. was supported by Science Foundation Ireland (SFI) and Acceligen/Recombinetics Inc. through the SFI Centre for Research Training in Genomics Data Science under grant no. 18/CRT/6214. G.P.M. and D.E.M. were supported by an SFI Investigator Award (SFI/15/IA/3154). T.S.S. and D.E.M were supported by a United States Department of Agriculture (USDA) and Department of Agriculture, Food, and the Marine (DAFM) US-Ireland Research and Development Partnership grant (TARGET-TB, 17/RD/USROI/52).

Acknowledgements. We thank Thomas Hall for advice on bioinformatics and computational genomics methodologies.

Sidebar

Cite this article: Ward JA etai. 2024 Genomic insights into the population history and adaptive traits of Latin American Criollo cattle. R. Soc. Open SciAV.im^.

Received: 14 September 2023

Accepted: 31 January 2024

Subject Category: Genetics and genomics

Subject Areas: evolution, ecology, genomics

Author for correspondence: David E. MacHugh e-mail: [email protected]

References

References

1. RouseJE. 1977 The (Hollo: Spanish cattle in the Americas. Norman, OK: University of Oklahoma Press.

2. Armenteros Martínez I. 2018 The Canary Islands as an area of interconnectivity between the Mediterranean and the Atlantic (fourteenthsixteenth centuries). In Entre mers-Outre-mer: spaces, modes and agents of Indo-Mediterranean connectivity (eds N Jaspert, S Kolditz), pp. 201-217, Heidelberg, Germany: Heidelberg University Publishing.

3. Rodero A. 1992 Primitive Andalusian livestock and their implications in the discovery of America. Arch. Zootec. 41,383-400.

4. Dunmire WW. 2004 Gardens of new Spain: how Mediterranean plants and foods changed America, 1st edn. Austin, TX: University of Texas Press.

5. Crosby AW. 2003 The Columbian exchange: biological and cultural consequences of 1492,30th anniversary edn. Westport, CT: Praeger.

6. Willham RL. 1982 Genetic improvement of beef cattle in the United States: cattle, people and their interaction J. Anim. Sci. 54,659-666. (doi: 10.2527/jas1982.543659x)

7. Tobar KMC, Âlvarez DC, Franco LA. 2020 Genome-wide association studies in sheep from Latin America. Rev. Mex. Gene. Pecu. 11. (doi:10. 22319/rmcp.v1li3.5372sander)

8. Revidatti MA etai. 2021 On the origins of American Criollo pigs: a common genetic background with a lasting Iberian signature. PloS One 16, e0251879. (doi:10.1371/journal.pone.0251879)

9. Taylor WTT et al. 2023 Early dispersal of domestic horses into the Great Plains and Northern Rockies. Science 379,1316-1323. (doi:10.1126/ science.adc9691)

10. Sanders JO. 1980 History and development of Zebu cattle in the United States. J. Anim. Sci. 50,1188-1200. (d oi : 10.25 2 7/jas 1980.5061188x)

11. da Fonseca RR, Ureña I, Afonso S, Pires AE, Jørsboe E, Chikhi L, Ginja C. 2019 Consequences of breed formation on patterns of genomic diversity and differentiation: the case of highly diverse peripheral Iberian cattle. BMC Genom. 20,334. (doi:10.1186/s 12864-019-5685-2)

12. Pitt D, Bruford MW, Barbato M, Orozco-terWengel P, Martinez R, Sevane N. 2019 Demography and rapid local adaptation shape Creole cattle genome diversity in the tropics. Evol. Appi. 12,105-122. (d o i : 10.1111 /eva.12641 )

13. Upadhyay M et al. 2019 Deciphering the patterns of genetic admixture and diversity in southern European cattle using genome-wide SNPs. Evol.Appl. 12,951-963. (doi:10.111l/eva.12770)

14. Magee DA, Meghen C, Harrison S, Troy CS, Cymbron T, Gaillard C, Morrow A, Maillard JC, Bradley DG. 2002 A partial African ancestry for the Creole cattle populations of the Caribbean. J. Hered. 93,429-432. (doi:10.1093/jhered/93.6.429)

15. Ginja C etai. 2019 The genetic ancestry of American Creole cattle inferred from uniparental and autosomal genetic markers. Sci. Rep. 9,11486. (doi:10.1038/s41598-019-47636-0)

16. Delsol N, Stucky BJ, Oswald JA, Cobb CR, Emery KF, Guralnick R. 2023 Ancient DNA confirms diverse origins of early post-Columbian cattle in the Americas. Sci. Rep. 13,12444. (doi:10.1038/s41598-023-39518-3)

17. Ben-Jemaa S, Adam G, Boussaha M, Bardou P, Klopp C, Mandonnet N, Naves M. 2023 Whole genome sequencing reveals signals of adaptive admixture in Creole cattle. Sci. Rep. 13,12155. (doi:10.1038/s41598-023-38774-7)

18. Landaeta-Hernández AJ etai. 2021 Heat stress response in slick vs normal-haired Criollo Limonero heifers in a tropical environment. Trop. Anim. Health Prod. 53,445. (doi:10.1007/sl 1250-021 -02856-3)

19. Martinez AM etai. 2012 Genetic footprints of Iberian cattle in America 500 years after the arrival of Columbus. PloS ONE. 7, e49066. (doi:10. 1371/journal.pone.0049066)

20. Caivio-Nasner S, López-Herrera A, González-Herrera LG, Rincón JC. 2021 Frequency of genotypic markers for genetic disorders, colour, polledness, and major genes in Blanco Orejinegro cattle. Trop. Anim. Health Prod. 53,546. (doi:10.1007/s11250-021 -02990-y)

21. Anderson DM, Esteli RE, Gonzalez AL, Cibils AF, Torell LA. 2015 Criollo cattle: heritage genetics for arid landscapes. Rangelands 37,62-67. (doi: 10.1016/j.rala.2015.01.006)

22. Food and Agriculture Organization. 2015 The second report on the state of the world's animal genetic resources for food and agriculture, (eds BD Scherf, D Pilling), Rome, Italy: FAO Commission on Genetic Resources for Food and Agriculture Assessments.

23. Felius M. 1995 Cattle breeds: an encyclopedia. Doetinchem, The Netherlands: Misset.

24. Rosen BD etai. 2020 De novo assembly of the cattle reference genome with single-molecule sequencing. Gigascience 9, giaa021. (doi:10.1093/ gigascience/giaa021)

25. Vasimuddin M, Misra S, Li H, Aluru S. 2019 Efficient architecture-aware acceleration of BWA-MEM for multicore systems. In 2019 IEEE Int. Parallel and Distributed Processing Symp. (IPDPS), Rio de Janeiro, Brazil, pp. 314-324.

26. Danecek P etai. 2021 Twelve years of SAMtools and BCFtools. Gigascience 10, giab008. (doi:10.1093/gigascience/giab008)

27. Van der Auwera GA, O'Connor BD. 2020 Genomics in the cloud: using Docker, GATK, and WDL in Terra, 1 st edn. Sebastopol, CA: O'Reilly Media.

28. Ci n gola n i P, Platts A, Wang LL, Coon M, Nguyen T, Wang L, Land SJ, Lu X, Ruden DM. 2012 A program for annotating and predicting the effects of single nucleotide polymorphisms, SnpEff. Fly 6,80-92. (doi:10.4161/fly.19695)

29. Bahbahani H etai. 2017 Signatures of selection for environmental adaptation and Zebu x taurine hybrid fitness in East African Shorthorn Zebu. Front. Genet. 8,68. (doi:10.3389/fgene.2017.00068)

30. Verdugo MP et al. 2019 Ancient cattle genomics, origins, and rapid turnover in the Fertile Crescent. Science 365,173-176. (doi:10.1126/ science.aav1002)

31. Upadhyay MR etai. 2017 Genetic origin, admixture and population history of aurochs {Bos primigenius) and primitive European cattle. Heredity (Edinb.) 118,169-176. (doi:10.1038/hdy.2016.79)

32. Ward JA et al. 2022 Genome-wide local ancestry and evidence for mitonuclear coadaptation in African hybrid cattle populations. ¡Science 25, 104672. (doi:10.1016/j.isci.2022.104672)

33. Sempéré G, Moazami-Goudarzi К, Eggen A, Laloë D, Gautier M, Flori L. 2015 WIDDE: a Web-Interfaced next generation database for genetic diversity exploration, with a first application in cattle. BMC Genom. 16,940. (doi:10.1186/s 12864-015-2181 -1 )

34. Zimin AV etai. 2009 A whole-genome assembly of the domestic cow, Bos taurus. Genom. Biol. 10, R42. (doi:10.1186/gb-2009-10-4-r42)

35. Riggio V etai. 2022 Assessment of genotyping array performance for genome-wide association studies and imputation in African cattle. Genet. Sei. Evol. 54,58. (doi:10.1186/s12711-022-00751-5)

36. Chang CC, Chow CC, Tellier LC, Vattikuti S, Purcell SM, Lee JJ. 2015 Second-generation PLINK: rising to the challenge of larger and richer datasets. Gigascience 4,7. (doi:10.1186/s13742-015-0047-8)

37. R Core Team. 2022 R: A language and environment for statistical computing. Vienna, Austria: R Foundation for Statistical Computing. See http:// www.R-project.org

38. Wickham H. 2016 ggplot2: elegant graphics for data analysis. New York, USA: Springer Publishing Company.

39. Alexander DH, Novembre J, Lange K. 2009 Fast model-based estimation of ancestry in unrelated individuals. Genom. Res. 19,1655-1664. (doi: 10.1101/gr.094052.109)

40. Barbato M, Orozco-terWengel P, Tapio M, Bruford MW. 2015 SNeP: a tool to estimate trends in recent effective population size trajectories using genome-wide SNP data. Front. Genet. 6,109. (doi:10.3389/fgene.2015.00109)

41. Patterson N, Moorjani P, Luo Y, Mallick S, Rohland N, Zhan Y, Genschoreck T, Webster T, Reich D. 2012 Ancient admixture in human history. Genetics'll, 1065-1093. (doi:10.1534/genetics.112.!45037)

42. Danecek Petal. 2011 The variant call format and VCFtools. Bioinformatics 27,2156-2158. (doi:10.1093/bioinformatics/btr330)

43. Szpiech ZA, Hernandez RD. 2014 seiscan: an efficient multithreaded program to perform EHH-based scans for positive selection. Mol. Biol. Evol. 31,2824-2827. (doi:10.1093/molbev/msu211)

44. Randhawa IAS, Khatkar MS, Thomson PC, Raadsma HW. 2014 Composite selection signals can localize the trait specific genomic regions in multi-breed populations of cattle and sheep. BMC Genet. 15,34. (doi:10.1186/1471-2156-15-34)

45. Verity R, Collins C, Card DC, Schaal SM, Wang L, Lotterhos KE. 2017 minotaur: a platform for the analysis and visualization of multivariate results from genome scans with R Shiny. Mol. EcoL Resour. 17,33-43. (doi:10.1111/1755-0998.12579)

46. Han H, Randhawa IAS, MacHugh DE, McGivney BA, Katz LM, Dugarjaviin M, Hill EW. 2023 Selection signatures for local and regional adaptation in Chinese Mongolian horse breeds reveal candidate genes for hoof health. BMC Genomics 24,35. (doi:10.1186/s12864-023-09116-8)

47. Elsik CG, Unni DR, Diesh CM, Tayal A, Emery ML, Nguyen HN, Hagen DE. 2016 Bovine genome database: new tools for gleaning function from the Bos taurus genome. Nucleic Acids Res. 44, D834-D839. (doi:10.1093/nar/gkv1077)

48. Raudvere U, Kolberg L, Kuzmin I, Arak T, Adler P, Peterson H, Vilo J. 2019 g:Profiler: a web server for functional enrichment analysis and conversions of gene lists (2019 update). Nucleic Acids Res. 47, W191-W198. (doi:10.1093/nar/gkz369)

49. Sollero BP, Junqueira VS, Gomes CCG, Caetano AR, Cardoso FF. 2017 Tag SNP selection for prediction of tick resistance in Brazilian Braford and Hereford cattle breeds using Bayesian methods. Genet. Sei. Evol. 49,49. (doi:10.1186/s12711-017-0325-2)

50. Chen N etai. 2018 Whole-genome resequencing reveals world-wide ancestry and adaptive introgression events of domesticated cattle in East Asia. Nat. Commun. 9,2337. (doi:10.1038/s41467-018-04737-0)

51. Wu DD etai. 2018 Pervasive introgression facilitated domestication and adaptation in the Bos species complex. Nat. EcoL Evol. 2,1139-1145. (doi:10.1038/s41559-018-0562-y)

52. MacHugh DE, Shriver MD, Loftus RT, Cunningham P, Bradley DG. 1997 Microsatellite DNA variation and the evolution, domestication and phylogeography of taurine and zebu cattle (Bos taurus and Bosindicus). Genetics 146,1071-1086. (doi:10.1093/genetics/146.3.1071)

53. Kim K etai. 2020 The mosaic genome of indigenous African cattle as a unique genetic resource for African pastoralism. Nat. Genet. 52,1099ИЮ. (doi:10.1038/S41588-020-0694-2)

54. Hanotte 0, Bradley DG, Ochieng JW, Verjee Y, Hill EW, Rege JEO. 2002 African pastoralism: genetic imprints of origins and migrations. Science 296,336-339. (doi:10.1126/science.!069878)

55. Decker JE et al. 2014 Worldwide patterns of ancestry, divergence, and admixture in domesticated cattle. PLoS Genet. 10, e1004254. (doi:10. 1371/journal.pgen.1004254)

56. Makanjuola BO, Miglior F, Abdalla EA, Maltecca C, Schenkel FS, Baes CF. 2020 Effect of genomic selection on rate of inbreeding and coancestry and effective population size of Holstein and Jersey cattle populations. J. Dairy Sci. 103,5183-5199. (doi:10.3168/jds.2O19-18013)

57. Toro RJE. 2020 Evaluación de Herramientas para Desarrollar UN Programa de Mejoramiento Genético en La Raza Blanco Orejinegro (BON). PhD thesis, University of Antioquia, Medellín, Colombia.

58. Illumina. 2015 Data sheet: BovineHD Genotyping BeadChip. Seewww.illumina.com/Documents/products/datasheets/datasheet_bovineHD.pdf

59. Albrechtsen A, Nielsen FC, Nielsen R. 2010 Ascertainment biases in SNP chips affect measures of population divergence. Mol. Biol. Evol. 27, 2534-2547. (doi:10.1093/molbev/msq148)

60. Malomane DK, Reimer C, Weigend S, Weigend A, Sharifi AR, Simianer H. 2018 Efficiency of different strategies to mitigate ascertainment bias when using SNP panels in diversity studies. BMC Genomics 19,22. (doi:10.1186/s12864-017-4416-9)

61. McTavish EJ, Hillis DM. 2015 How do SNP ascertainment schemes and population demographics affect inferences about population history? BMC Genomics 16,266. (d o i : 10.1186/s12864-015-1469-5)

62. McVean G. 2009 A genealogical interpretation of principal components analysis. PLoS Genet. 5, e1000686. (doi:10.1371 /journal.pgen. 1000686)

63. Pearson L. 1968 A note on the history of black-eared white cattle. Agric. Hist. Rev. 16,159-160.

64. Brenig В, Beck J, Floren C, Bornemann-Kolatzki K, Wiedemann I, Hennecke S, Swalve H, Schütz E. 2013 Molecular genetics of coat colour variations in White Galloway and White Park cattle. Anim. Genet. 44,450-453. (doi:10.1111/age.12029)

65. Druml T, Grilz-Seger G, Neuditschko M, Horna M, Ricard A, Pausch H, Brem G. 2018 Novel insights into Sabinol and splashed white coat color patterns in horses. Anim. Genet. 49,249-253. (doi:10.1111/age.12657)

66. Rothammer S, Kunz E, Seichter D, Krebs S, Wassertheurer M, Fries R, Brem G, Medugorac I. 2017 Detection of two non-synonymous SNPs in SLC45A2 on BTA20 as candidate causal mutations for oculocutaneous albinism in Braunvieh cattle. Genet. Sei. Evol. 49, 73. (doi:10.1186/ s12711-017-0349-7)

67. Voß К, Blaj I, Tetens JL, Thaller G, Becker D. 2022 Roan coat color in livestock. Anim. Genet. 53,549-556. (doi:10.1111/age.1324O)

68. Goud TS, Upadhyay RC, Onteru SK, Pichili VBR, Chadipiralla K. 2020 Identification and sequence characterization of melanocortin 1 receptor gene (MCIR) in Bosindicusversus (Bos taurus^Bosindicus). Anim. Biotechnol. 31,283-294. (doi:10.1080/10495398.2019.1585866)

69. Landaeta-Hernández A et al. 2011 Variability of hair coat and skin traits as related to adaptation in Criollo Limonero cattle. Trop. Anim. Health Prod. 43,657-663. (doi:10.1007/S11250-010-9749-1)

70. Landaeta-Hernández A, Zambrano S, Marin N, Moreno Y. 2011 Criollos Latinoamericanos. In Workshop Jornadas de Actualización en Ganadería Tropical pp. 52-62, Asociación de Ganaderos Alberto Adriani.

71. Mei C et al. 2018 Genetic architecture and selection of Chinese cattle revealed by whole genome resequencing. Mol. Biol. Evol. 35,688-699. (doi:10.1093/molbev/msx322)

72. Hansen PJ. 2020 Prospects for gene introgression or gene editing as a strategy for reduction of the impact of heat stress on production and reproduction in cattle. TheriogenologyASĄ, 190-202. (doi:10.1016/j.theriogenology.2020.05.010)

73. Florez Murillo JM, Landaeta-Hernández AJ, Kim E, Bostrom JR, Larson SA, Pérez O'Brien AM, Montero-Urdaneta MA, Garcia JF, Sonstegard TS. 2021 Three novel nonsense mutations of prolactin receptor found in heat-tolerant Bos taurus breeds of the Caribbean Basin. Anim. Genet. 52, 132-134. (doi:10.1111/age.13027)

74. Dikmen S, Khan FA, Huson HJ, Sonstegard TS, Moss JI, Dahl GE, Hansen PJ. 2014 The SLICK hair locus derived from Senepol cattle confers thermotolerance to intensively managed lactating Holstein cowsJ. Dairy. Sei. 97,5508-5520. (doi:10.3168/jds.2014-8087)

75. Sosa F et al. 2022 Effects of the SLICK1 mutation in PRLR on regulation of core body temperature and global gene expression in liver in cattle. Animal 16,100523. (doi:10.1016/j.animal.2022.100523)

76. Ben-Jemaa S, Mastrangelo S, Lee SH, Lee JH, Boussaha M. 2020 Genome-wide scan for selection signatures reveals novel insights into the adaptive capacity in local North African cattle. Sci. Rep. 10,19466. (doi:10.1038/s41598-020-76576-3)

77. Singh AK, Upadhyay RC, Chandra G, Kumar S, Malakar D, Singh SV, Singh MK. 2020 Genome-wide expression analysis of the heat stress response in dermal fibroblasts of Tharparkar (zebu) and Karan-Fries (zebu x taurine) cattle. Cell Stress Chaperones. 25,327-344. (doi:10.1007/ s12192-020-01076-2)

78. Pengj etai. 2018 An Hsp20-FBX04 axis regulates adipocyte function through modulating PPARy ubiquitination. Cell Rep. 23,3607-3620. (doi: 10.1016/j.celrep.2O18.05.065)

79. Srikanth K, Kwon A, Lee E, Chung H. 2017 Characterization of genes and pathways that respond to heat stress in Holstein calves through transcriptome analysis. Cell Stress Chaperones. 22,29-42. (doi:10.1007/s12192-016-0739-8)

80. Hayashi T et al. 2015 Higdla is a positive regulator of cytochrome c oxidase. Proc. Natl Acad. Sei. USA. 112,1553-1558. (doi:10.1073/pnas. 1419767112)

81. Zhang J etai. 2022 Yak miR-2285o-3p attenuates hypoxia-induced apoptosis by targeting caspase-3. Anim. Genet. 53,49-57. (doi:10.1111/ age.13153)

82. Zhang X, Wang D, Liu J. 2023 Hypoxia-inducible factor-1 a is involved in the response to heat stress in lactating dairy cowsJ. Therm. Biol. 112, 103460. (doi:10.1016/j.jtherbio.2023.103460)

83. Armstrong E, Rodriguez Almeida FA, McIntosh MM, Poli M, Cibils AF, Martinez-Quintana JA, Félix-Portillo M, Esteli RE. 2022 Genetic and productive background of Criollo cattle in Argentina, Mexico, Uruguay and the United States. J. Arid Environ. 200,104722. (doi:10.1016/j. jaridenv.2022.104722)

84. Johnson GP, English AM, Cronin S, Hoey DA, Meade KG, Fair S. 2017 Genomic identification, expression profiling, and functional characterization of CatSper channels in the bovine. Biol. Reprod. 97,302-312. (doi:10.1093/biolre/iox082)

85. Guo F etai. 2014 Alternative splicing, promoter methylation, and functional SNPs of sperm flagella 2 gene in testis and mature spermatozoa of Holstein bulls. Reproduction 147,241-252. (doi: 10.1530/REP-13-0343)

86. Mantel F, Monaco L, Foulkes NS, Masquilier D, LeMeur M, Henriksen K, Dierich A, Parvinen M, Sassone-Corsi P. 1996 Spermiogenesis deficiency and germ-cell apoptosis in CREM-mutant mice. Nature 380,159-162. (doi:10.1038/380159a0)

87. Zapater JL, Lednovich KR, Khan MW, Pusec CM, Layden ВТ. 2022 Hexokinase domain-containing protein-1 in metabolic diseases and beyond. Trends Endocrinol. Metab. 33,72-84. (doi:10.1016/j.tem.2O21.10.006)

88. Ren F, Xi H, Qiao P, Li Y, Xian M, Zhu D, Hu J. 2022 Single-cell transcriptomics reveals male germ cells and Sertoli cells developmental patterns in dairy goats. Front. Cell Dev. Biol. 10,944325. (doi:10.3389/fcell.2022.944325)

89. Gamallat Y etai. 2021 Bi-allelic mutation in Esip! impairs acrosome vesicle formation and attenuates flagellogenesis in mice. Redox Biol. 43, 101969. (doi:10.1016/j.redox.2021.101969)

90. Berisha B, Rodler D, Schams D, Sinowatz F, Pfaffl MW. 2019 Prostaglandins in Superovulation Induced Bovine Follicles During the Preovulatory Period and Early Corpus Luteum. Front. Endocrinol. (Lausanne) 10,467. (doi:10.3389/fendo.2019.00467)

91. Caivio-Nasner S, López-Herrera A, González-Herrera LG, Rincón JC. 2021 Diversity analysis, runs of homozygosity and genomic inbreeding reveal recent selection in Blanco Orejinegro cattle. J. Anim. Breed Genet. 138,613-627. (doi:10.1111/jbg. 12549)

92. Alshawi A, Essa A, Al-Bayatti S, Hanotte 0. 2019 Genome analysis reveals genetic admixture and signature of selection for productivity and environmental traits in Iraqi cattle. Front. Genet. 10,609. (doi:10.3389/fgene.2019.00609)

93. Fang L et al. 2016 Genome-wide transcriptional and post-transcriptional regulation of innate immune and defense responses of bovine mammary gland to Staphylococcus aureus. Front. Cell Infect. Microbiol. 6,193. (doi:10.3389/fcimb.2016.00193)

94. Kumar L et al. 2014 Leucine-rich repeat containing 8A (LRRC8A) is essential for T lymphocyte development and function. J. Exp. Med. 211,929- 942.(doi:10.1084/jem.20131379)

95. Rodriguez-Morales AJ, Bonilla-Aldana DK, Idarraga-Bedoya SE, Garcia-Bustos JJ, Cardona-Ospina JA, Faccini-Martínez ÄA. 2018 Epidemiology of zoonotic tick-borne diseases in Latin America: are we just seeing the tip of the iceberg? FlOOORes. 7,1988. (doi:10.12688/f1000research. 17649.2)

96. Ristie M, McIntyre 1.1981 Diseases of cattle in the tropics: economic and zoonotic relevance. The Hague, The Netherlands: Martinus Nijhoff Publishers.

97. Jaimes-Dueñez J, Triana-Chávez O, Holguín-Rocha A, Tobon-Castaño A, Mejia-Jaramillo AM. 2018 Molecular surveillance and phylogenetic traits of Babesia bigemina and Babesia bovis in cattle (Bos taurus) and water buffaloes (Bubalus bubalis) from Colombia. Parasit. Vectors 11, 510. (doi:10.1186/s13071-018-3091-2)

98. Fetene E, Leta S, Regassa F, Büscher P. 2021 Global distribution, host range and prevalence of Trypanosoma vivax; a systematic review and meta-analysis. Parasit. Vectors. 14,80. (doi:10.1186/s13071-021-04584-x)

99. Jaimes-Dueñez J, Triana-Chávez 0, Mejia-Jaramillo AM. 2018 Genetic, host and environmental factors associated with a high prevalence of Anaplasma marginale. Ticks Tick Borne Dis. 9,1286-1295. (doi:10.1016/j.ttbdis.2O18.05.009)

100. Naval-Sánchez M, Porto-Neto LR, Cardoso DF, Hayes BJ, Daetwyler HD, Kijas J, Reverter A. 2020 Selection signatures in tropical cattle are enriched for promoter and coding regions and reveal missense mutations in the damage response gene HELB. Genet. Sei. Evol. 52,27. (doi:10. 1186/s12711-020-00546-6)

101. Bhar S et al. 2020 Expansion of germline RPS20 mutation phenotype to include Diamond-Blackfan anemia. Hum. Mutat. 41,1918-1930. (doi: 10.1002/humu.24092)

102. Wang R etai. 2015 Loss of function mutations in RPL27 and RPS27 identified by whole-exome sequencing in Diamond-Blackfan anaemia. Br.J. Haematol. 168,854-864. (doi:10.1111/bjh.13229)

103. Noyes HA et al. 2009 Mechanisms controlling anaemia in trypanosoma congolense infected mice. PLoS One 4, e5170. (doi:10.1371/journal. pone.0005170)

104. Martínez R, Dunner S, Toro R, Tobón J, Gallego J, Cañón J. 2010 Effect of polymorphisms in the Sid lai coding region on resistance to brucellosis by macrophages in vitro and after challenge in two Bos breeds (Blanco Orejinegro and Zebu). Genet. Mol. Biol. 33, 463-470. (doi:10.1590/ S1415-47572010000300014)

105. Celli J. 2019 The intracellular life cycle of Brucella spp. Microbiol. Spectr. 7. (doi:10.1128/microbiolspec.BAI-0006-2019)

106. ENA. 2023 Data from: Population genetics of Latin American Criollo Cattle. European Nucleotide Archive, https://www.ebi.ac.uk/ena/browser/ view/PRJEB65887

107. Ward JA, etai. 2024 Data from: Genomic insights into the population history and adaptive traits of Latin American Criollo Cattle. Figshare, (doi: 10.6084/m9.figshare.c.7146965)