Tropical Animal Health and Production (2023) 55:357

https://doi.org/10.1007/s11250-023-03688-z

REGULAR ARTICLES

Genome‑wide association study reveals candidate genes for traits

related to meat quality in Colombian Creole hair sheep
Herman Alberto Revelo1,2 · Diana López‑Alvarez1,3 · Yineth Alexandra Palacios1 · Oscar David Vergara4 ·
Moris Bustamante Yánez4 · Manuel Fernando Ariza5 · Susan Lorena Castro Molina5 · Yurany Ortiz Sanchez5 ·
Luz Ángela Alvarez1

Received: 2 October 2022 / Accepted: 6 July 2023 / Published online: 12 October 2023
© The Author(s) 2023

Abstract
Genome-wide association studies (GWAS) allow identifying genomic regions related to traits of economic importance
in animals of zootechnical interest. The objective of this research was to conduct a genome-wide association study on
meat quality traits using the Illumina OvineSNPs50 BeadChip array. The animals were sampled in the departments of
Córdoba, Cesar, and Valle del Cauca. The genotypes obtained with the Illumina OvineSNP50 BeadChip microarray were
analyzed SNP (single-nucleotide polymorphism) data to conduct a GWAS for pH and water-holding capacity (WHC)
traits measured after 7 days of maturation, in the Longissimus dorsi (LD) muscle, in 167 Creole hair sheep of 12 months
old belonging to Pelibuey (CHSP, n = 60), Ethiopian (CHSE, n = 44), and Sudan (CHSS, n = 63) breeds. The GWAS was
done using a mixed linear model (MLMA) and based on the Ovis aries v3.1 genome. The CHSE showed the lowest meat
juice release and, consequently, the highest water-holding capacity (WHC = 30.6 ± 0.1), suggesting that this breed has
better performance in the meat industry compared with CHSS (WHC = 41.7 ± 0.1) and CHSP (WHC = 36.8 ± 0.1), since
there is a relationship between WHC and juiciness. For the character pH, it was not possible to annotate genes related to
meat quality, while, for the WHC, they have obtained 11 candidate genes associated (ELOVL2, ARAP2, LOC101102527,
SHOC2, AIPL1, CSRNP3, IFRD, KDM8, NANS, DAPK1, IBN2, TPM2). Particularly, ELOVL2, ARAP2, IBN2, and
TPM2 genes are involved in muscle contraction and fatty acid composition in sheep. In this study, we generated a baseline
for GWAS related to meat quality traits in Colombian Creole hair sheep that can be used for future genomic selection plans.

1
* Herman Alberto Revelo Grupo de Investigación de Recursos Zoogenéticos,
[email protected] Departamento de Ciencia Animal, Facultad de Ciencias
Agropecuarias, Universidad Nacional de Colombia,
* Diana López‑Alvarez
763533 Palmira, Colombia
[email protected]
2
Present Address: Facultad de Medicina Veterinaria y
Yineth Alexandra Palacios
Zootecnia, Universidad San Martin Cali Colombia, Carrera
[email protected]
122 #23‑395 del, Vía Cali‑Puerto Tejada, 760022 Cali,
Oscar David Vergara Colombia
[email protected] 3
Grupo de Investigación en Diversidad Biológica,
Moris Bustamante Yánez Departamento de Ciencias Biológicas, Facultad de Ciencias
[email protected] Agropecuarias, Universidad Nacional de Colombia,
763533 Palmira, Colombia
Manuel Fernando Ariza
4
[email protected] Grupo de Investigación en Producción Animal Tropical,
Universidad de Córdoba, 14014 Córdoba, Colombia
Susan Lorena Castro Molina
5
[email protected] Department of Animal Production, Universidad Nacional de
Colombia, 111321 Bogotá D.C, Colombia
Yurany Ortiz Sanchez
[email protected]
Luz Ángela Alvarez
[email protected]

13
Vol.:(0123456789)
357 Page 2 of 13 Tropical Animal Health and Production (2023) 55:357

Keywords Growth · GWAS · Creole sheep · SNPs meat production chain since the consumer is not only the
final recipient but also the breaking point by deciding on
the final product (Hervé 2013; Leal-Gutiérrez 2013).
Introduction Currently, WHC and pH traits are used to estimate meat
quality (Pearce et al. 2011; Leal-Gutiérrez 2013). According
In Colombia, Creole hair sheep (CHS) are considered a to Zhang et al. (2009), WHC defines the ability of meat to
socially important genetic resource for rural economy- retain water during the exertion of external forces such as
based communities due to the multiple benefits that these cutting, heating, and grounding. WHC is a highly relevant
animals provide, especially, to indigenous communities trait since it affects product performance and has important
(Revelo et al. 2020). Furthermore, CHS exhibit high meat economic implications on the industry and the consumer
production potential given the diversity of agroecosys- (Leal-Gutiérrez et al. 2014; Navarro et al. 2015). On the
tems and favorable conditions in the country (Revelo et al. other hand, pH is one of the main parameters used to verify
2020; Flórez et al. 2020). meat quality since it correlates with other traits including
Hair sheep are mainly raised in the tropical Caribbean WHC and relates to color, tenderness, juiciness, and texture
region under traditional production systems that lack tech- (Warner et al. 2010; Pearce et al. 2011; Navarro et al. 2015).
nology and reproductive control (Flórez et al. 2020). Addi- A genome-wide association study (GWAS) is a feasible
tionally, a group of hair sheep called “Pelibuey” (literally, and powerful tool to discover candidate genes and loci
sheep with cow hair) are mainly raised in southwestern associated with quantitative traits (Yang et al. 2011; Zhao
Colombia (Valle del Cauca) in association with sugar- et al. 2016; Hernández-Montiel et al. 2020). It allows not
cane plantations, where they provide biological control of only identifying controlled phenotypes for single genes but
weeds in the rows of sugarcane grass paddocks.1 Accord- also complex traits produced by the interaction of many
ing to the sheep slaughter survey reported by the National genes. This methodology has been used to study diseases
Administrative Department of Statistics (Departamento in sheep (White et al. 2012; Leymaster et al. 2013), charac-
Administrativo Nacional de Estadística) (DANE 2020), teristics associated with the carcass (Palacios 2018), vari-
31,978 head were slaughtered, producing 1,006,355 kg of ations in the number of lambs per birth in Pelibuey sheep
pie and 498,679 kg of carcass meat. In the same year, the (Hernández-Montiel et al. 2020), productivity (AI-Mamun
national live weight at slaughter totaled 33.1 kg ± 2.5 kg, et al. 2015), growth and meat production traits (Zhao et al.
while the weight of the carcass was 18.5 kg ± 1.2 kg with 2016), resistance to parasites (Sallé et al. 2012), among
a performance of 49.5%. Sheep meat consumption is inter- others. However, to our knowledge, there are no GWAS
nal and regionalized. The largest volume of sheep carcass for meat quality traits such as WHC and pH. Therefore,
meat is commercialized in market plazas (94.5%) and has the objective of this research was to use SNPs to conduct
grown up to 75.0%. The consumption per capita ranges a GWAS on pH and water-holding capacity (WHC) in the
from 310 to 500 g per year (Arevalo and Correa Assmus Longissimus dorsi muscle of Pelibuey ­(CHSP), Ethiopian
2013). ­(CHSE), and Sudan ­(CHSS) Creole hair sheep.
The increasing demand of sheep meat in the country
(Mestra-Vargas et al. 2019) is generating new perspectives
for livestock production. However, there are limited studies
on meat quality-related traits (Navarro et al. 2015) and the Materials and methods
parameters that define these traits (e.g., water-holding capac-
ity and pH), causing farmers to obtain low prices that are not Ethics statement
differentiated by the quality of the meat (Arevalo and Correa
Assmus 2013). Furthermore, quality-related parameters are The sample collection procedure was approved by ethics
not implemented in genetic improvement plans since WHC committee of Universidad Nacional de Colombia author-
and pH measurement are expensive and difficult to obtain by ized the project by which this research was conducted (in
requiring the slaughter of the selection candidates. minutes No.005/2019). All procedures followed the guide-
Warner et al. (2010) suggest that meat quality is influ- lines and regulations established in the Colombian code
enced by genetic factors such as breed and type of muscle of bioethics (resolution 8430 of 1993) and Law 84/1989
fiber. Environmental factors such as management system, regarding the protection of animals. Blood samples were
diet, age, weight at sacrifice, and antemortem and post- collected by qualified veterinarians during their routine
mortem procedures also affect quality (Pearce et al. 2011; practice within the framework of official programs aimed
Joo et al. 2013; de Lima Júnior et al. 2016; Hervé 2013). at the identification, health monitoring, and parentage con-
These aspects are usually considered at each stage of the firmation of the breeds and populations included in this

13
Tropical Animal Health and Production (2023) 55:357 Page 3 of 13 357

study. The use of animals and private land in this study and stored at 4 °C. After 7 days of maturation (aging),
was approved by the respective owners. we measured the pH using a pH meter, with three read-
ings per sample. Additionally, the WHC was determined
Animal sampling and study area according to the protocol described by Leal-Gutiérrez
et al. (2014). Briefly, 20 g of the LD samples was ground
For ­CHSE, we evaluated 44 individuals (25 males and 19 for 30 s, then, approximately 0.3 g of the sample was
females) from four farms located in the Department of Cor- placed on a Walkman No. 5 filter paper (Figs. s1a) and a
doba and one farm in the Department of Cesar (Table 1). compression force of 2.5 kg was applied for 5 min. Then,
The sheep were raised in extensive production systems and the paper was removed, revealing two areas, one formed
fed with Angleton (Dichantium aristatum), Brachiaria (Bra- by the pressed meat (M) and the second corresponding to
chiaria sp.), Colosuana (Bothriochloa pertusa), and Guinea the water released by the meat (T) (Fig. 1. Fig.s1b). The
(Megathyrsus maximus) pastures. For C ­ HSS, we analyzed papers were photographed using Adobe Acrobat Profes-
63 animals (42 males and 21 females) from five farms sional and the two areas were measured using the measure
located in Cesar and one farm in Cordoba (Table 1). These function. The WHC is expressed as the ratio between the
animals were also raised in extensive production systems
with Colosuana (Bothriochloa pertusa), Angleton (Dichan-
tium aristatum), and Guinea (Megathyrsus maximus) pas-
tures. For these farms, we were not able to obtain a family
structure due to low technological levels. For C ­ HSP, we
evaluated 60 males from El Hatico Natural Reserve, located
in the municipality of Cerrito (Valle del Cauca) (Table 1).
The animals were raised in a silvopasture environment inte-
grated with sugarcane plantations. Furthermore, the farm
applies a directed breeding so we were able to obtain a fam-
ily structure of 60 offspring from six fathers.

Phenotyping for pH and WHC

In total, 167 animals were slaughtered at 12 months old

at different slaughterhouses according to their origin:
the animals from Cordoba were slaughtered at Frigocer
Expocol S.A.S. (Cereté); those from Cesar at Echeverry
Gutiérrez & CIA SC (San Juan del Cesar, La Guajira), Fig. 1  (a) Paper labeled with the number of the animal and the cor-
responding meat slice, (b) ground meat sample (0.3 g) on the paper
and those from Valle del Cauca at Carnes y Derivados
after applying 2.5 kg, and (c) paper after the applied weight, showing
de Occidente S.A. (Yumbo). After slaughter and at 24-h two different areas; the inner ring is formed by the pressed meat (M)
postmortem, we extracted the Longissimus dorsi (LD) and the outer ring is formed by the water released by the meat (T), (d)
muscle from the left carcass, which was vacuum sealed delimitation of the perimeter of the two areas

Table 1  List of Colombian Breed Acronym Department Municipality Farm N

Creole sheep breeds,
geographical positions Ethiopian CHSE Córdoba Chima Galilea (1G) 10
(Department, Municipality, and
Los Córdobas Israel (1I) 10
farm), and number of sampled
animals (N) San Andrés de Sotavento Bankoovino (1B) 10
San Pelayo Cañabraval (1C) 10
Cesar Valledupar Caracas (2C) 4
Sudan CHSS Cesar Valledupar San José (2SJ) 12
Cesar Valledupar Villa Clara (2VC) 10
Cesar Valledupar Villa Liana (2VL) 14
Cesar Valledupar La Playa (2P) 8
Cesar Chimichanga Los Ángeles (2A) 9
Córdoba Ciénaga de Oro Un Córdoba (2U) 10
Pelibuey CHSP Valle del Cauca Cerrito Hatico (3H) 60

13
357 Page 4 of 13 Tropical Animal Health and Production (2023) 55:357

areas M and T, and is given by the formula: Area-M*100/ is the fixed effects of the j-nth farm, E­ Pk is the fixed effects
Area-T. of the k-nth lambing season, (SF)ij is the fixed effects of the
interaction between sex and farm, (FP)jk is the fixed effects
DNA extraction, genotyping, and quality control of the interaction between sex and lambing season, (SFP)ijk
is the fixed effects of the interaction between farm and lamb-
A total of 10 mL of blood from each animal was collected ing season, and Ɛijkl is the error.
in BD vacuum tubes (Becton Dickinson, Franklin Lakes, Second, model (M2) included only two individuals for
NJ, USA) with k3-EDTA. Genomic DNA was extracted CHSP and the fixed effects were as follows: father sire (1, 2,
from 167 blood samples using the QIAamp® DNA Mini 3, 4, 5, and 6) and lambing season (drought 1 and rainy 2).
Kit (QIAGEN, Hilden, Germany). DNA quantity and qual- (M2)
ity were assessed by spectrophotometry using a ColibriTM
Yikl = μ + Pi + EPk + (PEP)ij + εikl
equipment (Titertek Berthold, Neulingen, Germany) and
agarose gel electrophoresis using a standard protocol to where Yijkl is the vector of traits WHC and pH, μ corresponds
evaluate sample degradation. to the overall mean, Pi is the fixed effects of the i-nth father,
All animals were genotyped with the Illumina EPk is the fixed effects of the k-nth lambing season, (SF)ij is
OvineSNP50 SNP chip, which allows the simultaneous char- the fixed effects of the interaction between father and lamb-
acterization of up to 54,241 independent SNPs, according to ing season, and Ɛikl is the error.
the Infinium® Assay Super II Illumina® protocol for use on Duncan’s pairwise comparisons test was used to deter-
the HiScan®SQ System and BeadChips scanning. Quality mine significant differences, using S.A.S. 9.1.3.
control (QC) was performed using PLINK v1.9b5.2 (Purcell
et al. 2007) to remove SNPs with a minor allele frequency
(MAF) less than 0.1 (7038), a call rate less than 0.90 (3806), GWAS analysis
and a deviation from the Hardy–Weinberg equilibrium deter-
mined at p < 0.0001 (696). Furthermore, SNPs on sex chro- Before conducting the GWAS, linear models were fitted
mosomes and the mitochondrial contig were also excluded. to each phenotype to identify the fixed effects models that
Finally, 42,701 SNPs remained for further analyses. The provided the best fit (Supplementary material). A mixed
genetic structure of the population of CHS was evaluated by linear model (MLMA) was used to control for popula-
a multidimensional scaling analysis (MDS) based on pair- tion structure and environmental effects. Specifically, the
wise identical-by-state (IBS) distances that were calculated genetics (through the sire) of the sheep and environmental
and plotted using Bite (Milanesi et al. 2017) package from effects such as, sex, farm (M1), and lambing season (M2)
Bioconductor in R v3.6.0 (http://​www.r-​proje​ct.​org). were controlled. Also, the model included a genomic rela-
tionship matrix (GRM) and the variance explained by the
SNPs.
Statistical analyses The model was:
y = μ + Xβ + Zu + e
To evaluate the phenotypes for WHC and pH values, a
descriptive analysis and an ANOVA of least squares or where y is the vector of phenotypes (WHC, pH), μ is the
GLM were performed using the statistical package SAS overall mean for the trait, β is the vector of fixed effects, u is
9.1.3 (SAS, Inc. Cary N.C). For the CHSE and CHSS the vector of random effects, X and Z are the design matri-
breeds, the effects of different farms were considered, while ces mapping vectors of fixed or random effects, G is the
for CHSP, controlled breeding was conducted on a single genomic relationship matrix, and Ɛ is the vector of residual
farm where 10 male offspring were produced per sire. For effects.
that reason, it was used two fixed models. First, model (M1) To facilitate this calculation, the genetic variance, var (G),
was used for CHSE and CHSS, where the following fixed is estimated based on the null model, y = μ + u + e, and is fixed
effects were included: sex (male and female), farm (1B, 1C, while the association between each SNP and the trait is tested.
1G, 1I, 1U, 2A, 2C, 2P, 2SJ, 2VC, and 2VL), and lambing In the null model, the variables indicate the following: u =  ~ N
season (drought and rainy). (0, Gσ2), where G is the genomic relationships matrix (GRM)
(M1) and σ2u = polygenic variance estimated by the null model.
Table S3 shows the genomic association models (GWAS)
Yijkl = μ + Si + Fj + EPk + (SF)ij + (SP)ik + (FP)jk + (SFP)ijk + εijkl
using a MLMA for WHC and pH in CHSE, CHSS, and CHSP.
where Yijkl is the vector of traits WHC and pH, μ corresponds QQ-plots (Quantile–Quantile plots) were utilized to visu-
to the overall mean, Si is the fixed effects of the i-nth sex, Fj alize the test statistics under the hypothesis of no association
(null hypothesis) to identify if there was inflation.

13
Tropical Animal Health and Production (2023) 55:357 Page 5 of 13 357

The Bonferroni correction, which is a statistical tech- Table 2  Comparison of mean values of meat quality-related traits,
nique to determine the threshold, was used. This technique WHC and pH, according to sex, farm, and lambing season in Ethio-
pian Creole hair sheep ­CHSE
divides the 0.05 significance level by the number of mark-
ers used in the GWAS (42,701 SNPs). Therefore, the sig- Sources of variance Factor N WHC (%) pH
nificance threshold after applying the Bonferroni correc-
Sex Female 19 28.91 ± 8.2a 5.72 ± 0.1a
tion (1.17 × 10^ − 6) is − log10(p) ≥ 5.93, for genome-wide
Male 25 31.93 ± 12.5a 5.78 ± 0.1a
significance threshold. The chromosome-wide significance
Farm 1G 10 28.86 ± 8.8b* 5.79 ± 0.07a
threshold had a value of 9.97 × 10^ − 4, which in the Man-
1B 10 25.52 ± 6.2b* 5.70 ± 0.1a
hattan plots corresponds to a value of 3.01. The Manhattan
1C 10 27.75 ± 6.1b* 5.72 ± 0.07a
plots were generated using GWAS-tools package (Gogarten
1I 10 34.10 ± 14.9b* 5.81 ± 0.15a
et al. 2012) in R v3.6.0 (http://​www.r-​proje​ct.​org) to visual-
2C 4 46.32 ± 8.76a* 5.71 ± 0.05a
ize the level of significance of the GWAS, the chromosomal
Lambing season Rainy 19 32.16 ± 12.22a 5.8 ± 0.1a*
locations, and SNPs.
Drought 25 29.46 ± 9.90a 5.71 ± 0.1b*
Total mean 44 30.63 ± 10.9 5.75 ± 0.1
Functional analysis
Mean + s.d. of variables. Superscripts denote Duncan pairwise com-
The functional information and biological processes of parisons between sources of variance; means with the same letter do
not differ significantly (**p < 0.001; *p < 0.05); N, number of obser-
the mapped genes were retrieved using AnnotationHub, vations
Mygene, and ReactomePA libraries (Yu 2012) of Biocon-
ductor in R v.3.6.0, (http://​www.r-​proje​ct.​org) by assign-
ing functions and GO terms reported in Gene Ontology
Table 3  Comparison of mean values of meat quality-related traits,
(GO), Kyoto Encyclopedia of Genes (KEGG) https://​www.​ WHC and pH, according to sex, farm, and lambing season in Sudan
genome.​jp/​kegg/), and PATHWAY. Additionally, gene net- Creole hair sheep C
­ HSS
works and gene enrichment analyses were done using the
Sources of variance Factor N WHC (%) pH
Database for Annotation, Visualization and Integrated Dis-
covery (DAVID) and its integrated databases (INTERPRO, Sex Female 21 38.65 ± ­15a 5.73 ± 0.1a
UP-Keywords, Uniprot, SMART, and Pir-Superfamily). Male 42 43.21 ± 16.5a 5,75 ± 0.1a
Farm 1U 10 25.06 ± 6.1b* 5.69 ± 0.13b*
2SJ 12 49.27 ± ­27a* 5.71 ± 0.1b*
Results 2VC 10 43.88 ± ­16a* 5.81 ± 0.05a*
2A 9 44.24 ± 8.7a* 5.71 ± 0.1b*
Meat quality traits — WHC and pH — in three 2VL 14 40.90 ± 4.9a* 5.71 ± 0.1b*
populations of Creole hair sheep 2P 8 46.90 ± 8.1a* 5.85 ± 0.09a*
Lambing season Rainy 17 46.6 ± ­25a 5.77 ± 0.1a
In sheep, a normal pH ranges from 5.5 to 5.8 and is related to Drought 46 39.8 ± ­11a 5.73 ± 0.01a
desirable characteristics in meat quality, such as color, shear Total mean 63 41.69 ± 16.3 5.74 ± 0.1
force, and water holding capacity (Arce-Recinos et al. 2021).
Mean + s.d. of variables. Superscripts denote Duncan pairwise com-
The average pH value of the LD in CHSE was 5.75 ± 0.1 parisons between sources of variance; means with the same letter do
(Table 2). We did not find significant differences between not differ significantly (**p < 0.001; *p < 0.05); N, number of obser-
sexes or farms (Table S1a) but rather for the lambing season vations
(p < 0.01). The average WHC value of the LD in CHSE was
30.63% ± 10.9 and no significant differences were observed farm*lambing season (p < 0.05, Table S1c). Moreo-
between sexes (Table 2, Table S1b). However, we found ver, for WHC of the LD in CHSS, the average value was
significant differences between farms (p < 0.05), specifi- 41.69% ± 16.3 and we determined significant differences
cally, for farm 2C. This farm is the only one located in the between farms (p < 0.01) and the interaction farm*lambing
Department of Cesar (Table 1) in a Tropical Very Dry Forest season (p < 0.01, Table S1d). Farm 1U showed the lowest
according to Holdridge’s life zones (Palacios 2018). Further- WHC (Table 3) and significantly differed from the other
more, we could not evaluate 10 animals in farm 2C since it farms. This farm is located in the Department of Cordoba
applies a low-technology production system that limited the (Table 1) in a Tropical Dry Forest according to Holdridge’s
possibility of conducting breeding control. life zones (Palacios 2018).
The average pH value of the LD in CHSS was 5.7 ± 0.1 Finally, the average pH value of the LD in CHSP was
(Table 3). Furthermore, we observed significant differ- 5.74 ± 0.1 (Table 4). We determined significant differences
ences between farms (p < 0.01) and for the interaction between fathers (p < 0.001, Table S1e); accordingly, the

13
357 Page 6 of 13 Tropical Animal Health and Production (2023) 55:357

Table 4  Comparison of mean values of meat quality-related traits, vs ­CHSE (red) and ­CHSS (yellow) along the first axis, and
WHC and pH, according to father and lambing season in Pelibuey a secondary split of ­CHSS vs ­CHSE along the second axis
Creole hair sheep C
­ HSP
(Fig. 2). However, a slight overlap was observed between
Sources of variance Factor N WHC (%) pH individuals of ­CHSS and ­CHSE, caused by a group of Sudan
Father 1 10 28.47 ± 3.96 b**
5.68 ± 0.05b**
Blanco sheep (n = 8) that was sampled at La Playa farm in
2 10 28.49 ± 3.38 b**
5.62 ± 0.1b**
Valledupar. This group of sheep shares a common genetic
3 10 38.47 ± 9.13a** 5.79 ± 0.1a** background with ­CHSE (n = 4), which were sampled at Cara-
4 10 41.64 ± ­7a** 5.78 ± 0.1a** cas farm in Valledupar. Consequently, the overlap indicates
5 10 41.17 ± 7.25a** 5.79 ± 0.1a** that Sudan Blanco sheep are a mix between ­CHSE and ­CHSS.
6 10 42.54 ± 7.4a** 5.78 ± 0.1a** We observed population stratification due to the presence
Lambing season Rainy 18 38.06 ± 7.7a 5.74 ± 0.1a of the three breeds from the different farms with contrast-
Drought 42 36.26 ± 9.37a 5.7 ± 0.1a ing genetic backgrounds. Individual kinship and population
Total mean 60 36.8 ± 8.8 5.74 ± 0.1 stratification in GWAS are the main causes of false positive
Mean + s.d. of variables. Superscripts denote Duncan pairwise com-
correlations; therefore, we performed a GWAS per breed to
parisons between sources of variance; means with the same letter do identify candidate SNPs and genes.
not differ significantly (**p < 0.001; *p < 0.05); N, number of obser-
vations Genome‑wide association study (GWAS) of meat
quality traits — WHC and pH

A GWAS for each of the three breeds was tested using

mixed linear models (MLMA) with fixed effects. The cri-
terion applied to select the genomic association model was
based on previous results of an ANOVA on phenotypic
data (Table S1a, b, c, d, e, f). The GWAS of pH did not
show significantly associated SNPs. On the other hand,
the GWAS of WHC for CHSE showed 12 significant SNPs
located on eight chromosomes (3, 4, 6, 14, 16, 19, 20, and
21) and eight genes (Table 5). Furthermore, 25% of these
SNPs were found in intergenic regions, while 66.6% were in
introns and 8.3% in exons (Table 5). The Manhattan plot of
WHC for CHSE is shown in Fig. 3A. For CHSS, we found
33 SNPs significantly associated with WHC (p < 0.001,
Table 6), located on 16 chromosomes (1–7, 9, 11, 13, 15,
17–20, 22, and 24) and 28 genes (Table 6). The SNPs were
distributed throughout intergenic regions (33.3%), introns
(60.6%), promotors (3%), and exons (3%). The Manhattan
plot of WHC for CHSS is shown in Fig. 3B. Finally, for
Fig. 2  Two-dimensional PCA plot of 44 individuals of C
­ HSE, 63 of CHSP, we identified nine SNPs significantly associated with
­CHSS, and 60 of CHSP based on 42,701 SNPs WHC (p < 0.001, Table 7), located on five chromosomes
(2, 5, 11, 13, and 20) and eight genes (Table 7). Further-
more, 44.44% of the SNPs were found in intergenic regions,
lowest values were observed for individuals from fathers 33.33% in introns, and 11.11% in promoters. The Manhattan
1 and 2. Moreover, for WHC, ram lambs showed an aver- plot of WHC for CHSP is shown in Fig. 3C.
age value of 36.8% ± 8.8 (Table 4). We found significant
differences between fathers (p < 0.001, Table S1f), and the Gene ontology (GO) and KEGG gene enrichment
highest values were obtained for fathers 3 (38.47% ± 9.13), 4 analysis for WHC in Colombian Creole hair sheep
(41.64% ± 7), 5 (41.17% ± 7.25), and 6 (42.54% ± 7.4). breeds
Population structure For ­CHSE, 20 GO terms were assigned to the ARAP2 gene
(Table S2a). Additionally, we identified three terms for the
The first two components of the bidimensional PCA plot ELOVL2 gene based on KEGG functional assignments
explained 11.8% and 6.1% of the total variance, respectively. (Table S2a). Regarding ­CHSS, 110 GO terms were assigned
Furthermore, we observed a separation between C ­ HSP (blue) to five genes (Table S2b). Most terms were assigned to the

13
Tropical Animal Health and Production (2023) 55:357 Page 7 of 13 357

Table 5  Details of SNPs significantly associated with WHC for ­CHSE using MLMA and nearest candidate genes
SNP Chromosome Position − log10(p-value) Gene Gene description Annotation

OAR16_57770340.1 16 53058133 1.55E + 01 CDH18 cadherin 18 Intron

OAR16_57803053.1 16 53091017 1.55E + 01 CDH18 cadherin 18 Intron
s03399.1 16 674786 1.73E + 01 SLIT3 slit guidance ligand 3 Intron
OAR21_54114533.1 21 48838644 1.98E + 01 NA – Distal Intergenic
s36343.1 6 55461883 2.86E + 01 ARAP2 ArfGAP with RhoGAP domain, Exon
ankyrin repeat and PH domain 2
s35737.1 19 30792252 4.08E + 01 NA – Distal Intergenic
OAR3_96465036.1 3 90809192 6.35E + 01 LTBP1 latent transforming growth factor Intron
beta binding protein 1
OAR20_48109414.1 20 44188831 6.40E + 01 ELOVL2 ELOVL fatty acid elongase 2 Intron
OAR14_29971688.1 14 28809538 7.16E + 01 CDH8 cadherin 8 Intron
OAR4_15640071.1 4 15399962 7.41E + 01 COL28A1 collagen type XXVIII alpha 1 chain Intron
OAR16_7661219.1 16 7207579 9.46E + 01 NA – Distal Intergenic
s67263.1 3 75480558 9.52E + 01 LOC105606945 – Intron

Genes within candidate regions were annotated based on the Ovis aries v.3.1 genome assembly; those in bold are the most promising candidate
genes based on their effect on meat quality or identified as harboring significant SNP associated with sheep meat quality

PRKCQ gene (Table S2b), followed by AIPL1, SHOC2,

and LOC101102527. The functional KEGG annotation
revealed seven terms related to PRKCQ and one term to
LOC101102527 (Table S2b). Additionally, a pathway
enrichment analysis using DAVID significantly associated
the term “nucleus” (p = 5.7E − 2) to four genes, namely
AIPL1, CSRNP3, IFRD1, and KDM8. Finally, for ­CHSP,
50 GO terms were associated with two genes (Table S2c),
mostly FBN2, followed by TPM2. The KEGG annotation
showed six terms associated with DAPK1, TPM2, and
NANS.

Discussion

The water-holding capacity is possibly one of the most

important quality traits of raw meat products (Huff-Lonergan
and Lonergan 2005). Therefore, new methodologies should
be implemented to improve economically important traits
in the meat industry (Leal-Gutiérrez et al. 2014). We also
evaluated the pH of the LD muscle since it is directly related
to WHC (Pearce et al. 2011; Torrescano et al. 2009). Most
water in muscle is retained in the myofibrils (Huff-Lonergan
and Lonergan 2005); therefore, an accelerated decrease in
pH and consequent reduction in WHC generate the Pale,
Soft, and Exudative (PSE) condition of meat (Warner et al.
2010). On the other hand, high pH values result from a slow
death since chronic stress for a long period before slaugh-
ter reduces glycogen reserves in the muscle and causes a
Fig. 3  Manhattan plots of WHC for three Colombian Creole-hair low lactic acid content. For this reason, a higher pH at 24-h
sheep breeds: (A) ­CHSE; (B); ­CHSS; (C) ­CHSP. The grey horizon- postmortem (6.0 to 6.8) compared to normal meat (5.4 to
tal lines in the Manhattan plots indicate the significance thresholds
(p < 0.001). The x-axis shows the physical positions of the SNPs per 5.9) turns the meat dark, firm, and dry (DFD) with a high
chromosome based on the Ovis aries v3.1. genome assembly WHC (Pearce et al. 2011), These characteristics cause dark

13
357 Page 8 of 13 Tropical Animal Health and Production (2023) 55:357

Table 6  Details of SNPs significantly associated with WHC for ­CHSS using MLMA and nearest candidate genes
SNP Chromosome Position − log10(p-value) Gene Gene description Annotation

OAR3_218139642.1 3 202690,985 2.50E − 02 ENSOARG00000026051 – Intron

s10070.1 9 19268400 2.50E − 02 SNORA25 – Distal Intergenic
s50062.1 11 16487540 2.50E − 02 ASIC2 acid sensing ion channel Intron
subunit 2
OAR17_39271762.1 17 36371262 2.50E − 02 FSTL5 follistatin like 5 Intron
s17049.1 18 28968415 2.50E − 02 OTUD7A OTU deubiquitinase 7A Intron
OAR24_27265846.1 24 24776080 2.50E − 02 KDM8 lysine demethylase 8 Intron
OAR5_68161647_X.1 5 62000366 1.38E − 01 NA – Distal Intergenic
s10696.1 1 52719709 1.28E + 01 ST6GALNAC5 – Distal Intergenic
OAR2_151938109.1 2 142963793 1.30E + 01 CSRNP3 cysteine and serine rich Distal Intergenic
nuclear protein 3
OAR7_100406183.1 7 92215284 1.30E + 01 NA – Distal Intergenic
OAR11_26036763.1 11 25104680 1.85E + 01 AIPL1 aryl hydrocarbon recep- Distal Intergenic
tor interacting protein
like 1
OAR20_39711404.1 20 36261532 1.86E + 01 CDKAL1 CDK5 regulatory subu- Intron
nit associated protein
1 like 1
OAR24_21411103.1 24 19579841 2.31E + 00 LOC101114079 ATP-binding cassette Intron
sub-family A member
3-like
OAR9_37599748.1 9 35709024 2.35E + 01 XKR4 – Intron
OAR17_49866586.1 17 45793784 3.12E + 01 ADGRD1 adhesion G protein-cou- Intron
pled receptor D1
s43108.1 17 47855868 4.18E + 01 GLT1D1 glycosyltransferase 1 Intron
domain containing 1
OAR4_59276586.1 4 56032346 4.42E + 00 ZNF277 zinc finger protein 277 Intron
OAR17_28971772.1 17 26404750 4.45E + 01 NA – Distal Intergenic
s65074.1 3 213086338 4.98E − 01 BCL2L13 – Intron
OAR22_34955585_X.1 22 30443185 5.71E + 01 SHOC2 SHOC2 leucine rich Intron
repeat scaffold protein
OAR1_147233859.1 1 136261672 5.72E + 01 NA – Distal Intergenic
s52527.1 20 39053345 6.60E + 01 LOC105603846 – Intron
OAR10_48194802.1 10 47420556 6.83E + 01 LOC101122287 – Intron
OAR19_39736257.1 19 37888856 7,56E + 01 SYNPR synaptoporin Intron
OAR15_5833133.1 15 6314820 7.73E + 01 CEP126 centrosomal protein 126 Distal Intergenic
s55275.1 13 11004130 8.41E + 01 PRKCQ protein kinase C theta Intron
OAR7_104475739.1 7 95912557 8.55E − 4 NA – Distal Intergenic
OAR2_204712172.1 2 193176,635 9.23E + 01 LOC105610065 – Intron
s10999.1 17 45411347 9.35E + 01 ENSOARG00000025678 – Exon
OAR1_85261298.1 1 79964281 9.52E + 01 RNPC3 RNA binding region Promoter
(RNP1, RRM) contain-
ing 3
s07011.1 13 11383479 9.91E + 01 SFMBT2 Scm like with four mbt Distal Intergenic
domains 2
s24860.1 18 13745532 9.93E + 01 ST8SIA2 ST8 alpha-N-acetyl- Intron
neuraminide alpha-
2,8-sialyltransferase 2
s69516.1 20 29281487 9.98E + 00 ZSCAN23 – Intron

Genes within candidate regions were annotated based on the Ovis aries v.3.1 genome assembly; those in bold are the most promising candidate
genes based on their effect on meat quality or identified as harboring significant SNP associated with sheep meat quality

13
Tropical Animal Health and Production (2023) 55:357 Page 9 of 13 357

Table 7  Details of SNPs significantly associated with WHC for ­CHSP using MLMA and nearest candidate genes
SNP Chromosome Position − log10(p-value) Gene Gene description Annotation

OAR2_56383269.1 2 52536647 3.38E + 01 TPM2 tropomyosin 2 Promoter (< = 1 kb)

OAR13_65478113.1 13 60250874 5.27E + 01 NA – Distal Intergenic
OAR13_69620023.1 13 64546634 5.44E + 01 PHF20 PHD finger protein 20 Intron
OAR13_63071015.1 13 57933199 6.42E + 01 LOC105606211 – Distal Intergenic
s33429.1 20 32872334 6.91E + 01 LOC105603826 – Distal Intergenic
s54820.1 2 49365215 7.42E + 01 NANS N-acetylneuraminate Promoter (2-3 kb)
synthase
OAR11_3378385.1 11 4045712 7.76E + 01 ENSOARG00000026349 – Distal Intergenic
s11926.1 2 31911076 9.294E + 01 DAPK1 Death Associated Protein Intron
Kinase 1
s16895.1 5 23139577 9.77E + 00 FBN2 fibrillin 2 Intron

Genes within candidate regions were annotated based on the Ovis aries v.3.1 genome assembly; those in bold are the most promising candidate
genes based on their effect on meat quality or identified as harboring significant SNP associated with sheep meat quality

cuts and the meat will likely be rejected by the consumer, We did not find significant differences in the WHC of ­CHSE
given the impression that is not appetible or comes from old and ­CHSS between sexes, probably because these animals were
animals (Leal-Gutiérrez et al. 2014). managed under extensive grazing. These results agree with
In this study, the pH values ­(CHSE = 5.75, ­CHSS = 5.7, Expósito et al. (2003), who did not report differences in the WHC
­CHSP = 5.74) indicated that the meats were red, firm, and of the Longissimus dorsi muscle between sexes in Segureña sheep
non-exudative (RFN), which are optimal meat quality and attributed these findings to similarities in intramuscular fat.
parameters (Gallo and Tadich 2008). This good meat qual- Moreover, we determined significant differences in WHC
ity was attributed to relevant environmental factors, such for ­CHSE and ­CHSS between farms, which can be attributed
as animal manipulation and carcass management for 24 h to the geographic location of the farms (Table 2, Table 3).
before and after slaughter according to the established pro- The farms in the Department of Cordoba are located in a
tocols. These findings demonstrate that an adequate manage- Tropical Dry Forest, while the farms in Cesar are found in a
ment before and after slaughter is essential to obtain good Tropical Very Dry Forest.
quality meat. For ­CHSP, we determined significant differences between
The WHC is expressed as a percentage and is related to fathers (Table 4) and these findings could be associated with
the amount of juice released by compression or manipulation; the genetic traits of the father. Conversely, no differences were
therefore, greater juice release indicates lower liquid retention observed between lambing season (drought and rainy), which
and, consequently, lower WHC (Leal-Gutiérrez et al. 2014; is possibly due to silvopasture at El Hatico natural reserve that
Navarro et al. 2015). The amount and location of water in meat guarantees the supply of high-quality forage year-round. In
can vary depending on several factors related to the muscle tis- this farm, the animals graze on the sugarcane paddocks and
sue Pearce et al. 2011). Generally, in sheep, the water-holding are supplemented with balanced diets. Similarly, Ramírez-
capacity decreases with age so adult sheep meat releases a Bribiesca et al. (2007) reported a WHC of 42.4 ± 2.3 (p ≤ 0.05)
higher percentage of juice than lamb meat. However, scien- in Pelibuey × Katadin × BlackBell sheep raised in an inten-
tific reports on the WHC of sheep meat are varied. For exam- sive system and slaughtered at an average weight of 40 kg at
ple, Pérez et al. (2006) indicated low values for lactating lamb 10 months old. Moreover, Frías et al. (2011) reported a WHC
of Suffolk Down × German Precocious Merino slaughtered of 12.8 ± 0.16 in Pelibuey with Katahdin and Dorper lambs fed
at live weights of 10 and 15 kg, reporting a WHC of 14.5% with grass and supplemented with fermented sugarcane, which
in females and 12.5% in males. On the other hand, a WHC were sacrificed at 20 kg of weight.
of 42.4 ± 2.3 (p ≤ 0.05) was reported for individuals of Peli- Regarding pH, we did not find significant differences
buey × Katadin × BlackBell that were slaughtered at an aver- between sexes for C ­ HSE and C ­ HSS. These results agree
age weight of 40 kg at 10 months old (Ramírez-Bribiesca et al. with Rodríguez et al. (2011), who reported a pH of 5.65
2007). We report a WHC at 7-day postmortem of 41.69% for and 5.61 at 24-h postmortem in the Longissimus lumborum
­CHSS, 30.63% for ­CHSE, and 36.8% for ­CHSP slaughtered at and semimembranosus muscles in Assaf and Merino × Assaf
12 months old. Furthermore, we found that C ­ HSE showed lower sheep. Moreover, the effect of farm on pH was significant
juice release and, consequently, higher WHC, suggesting that for ­CHSS and the highest values were observed in farms
sheep of this breed can exhibit better performance in the meat 2VC (5.81 ± 0.05) and 2P (5.85 ± 0.09). Conversely, we did
industry compared with ­CHSS and ­CHSP. not find significant effects of farm on pH for C ­ HSE. For

13
357 Page 10 of 13 Tropical Animal Health and Production (2023) 55:357

this breed, the pH values ranged from 5.7 to 5.8 and cor- a wide variety of protein targets and are involved in sev-
responded to the optimal values for meat quality. These eral cell signaling pathways (UniProt Consortium 2015).
findings are attributed to an adequate management of the We found that this gene is associated with several terms
animals before and after slaughter, according to standard- involving immune-related functions, as well as with the
ized protocols. term adipocytokine signaling pathway (oas04920) (see
Supplementary Table S2c). The IFRD1 gene encodes the
SNP markers associated with WHC, gene ontology, interferon-related developmental regulator 1 and can func-
and metabolic pathways tion as a transcriptional coactivator that controls the growth
and differentiation of specific cell types during embryonic
For ­CHSE, we found two SNPs potentially associated with development and tissue regeneration (Stelzer et al. 2016).
WHC. The first SNP, OAR20_48109414.1, is located in the In sheep, this gene is associated with myoblast growth
ELOVL2 (fatty acid elongation) gene, which encodes the and muscle differentiation (Cheng et al. 2020). Moreover,
enzyme that catalyzes the synthesis of long-chain polyun- KDM8 encodes a bifunctional enzyme that acts as an endo-
saturated fatty acids (C20- and C22-PUFA), acting specifi- peptidase and a 2-oxoglutarate-dependent monooxygenase
cally on polyunsaturated acyl-CoA and with greater activity (Liu et al. 2016). The photoreceptor/pineal-expressed gene,
towards acyl-CoA C20: 4 (n-6). This enzyme participates in AIPL1, encodes the arylhydrocarbon interacting protein-like
the production of polyunsaturated VLCFA of different chain 1 and is found in the candidate LCA4 region. This encoded
lengths that are involved in several biological processes as protein contains three tetratricopeptide motifs, compat-
membrane lipid precursors and lipid mediators (UniProt ible with nuclear transport or chaperone activity (Stelzer
Consortium 2015). Similarly, Bolormaa et al. (2016) found et al. 2016). Finally, SNP OAR24_21411103.1 located in
SNPs strongly associated with polyunsaturated fatty acid LOC101102527, SNP OAR22_34955585_X.1 in SHOC2,
concentrations, such as OAR20_44.1, located at 100 Kb and SNP OAR2_151938109.1 in CSRNP3 are not reported
from the ELOVL2 gene, in a GWAS of 56 traits (e.g., weight, in the literature so their functions related to WHC are
fat, muscling, tenderness, meat color, pH level, and fatty acid unknown.
profile) in 10,613 in nine sheep breeds (Merino, Poll Dor- For ­CHSP, we found nine SNPs, including four asso-
set, Border Leicester, Suffolk, White Suffolk, Texel, Cor- ciated with WHS. First, SNP s11926.1 is located in the
riedale, Coopworth, and various crosses). ELOVL2, along DAPK1 gene that encodes a calcium-dependent threo-
with PNPLA3 and FADS2 genes, participate in fatty acid nine kinase protein. This protein is involved in several
synthesis and metabolism. Moreover, the EVOVL6 gene was cellular signaling pathways that activate cell survival,
found as a candidate affecting the composition of palmitic apoptosis, and autophagy; specifically, regulating type
and palmitoleic fatty acids (Bolormaa et al. 2016). These I apoptotic and type II autophagic cell death signals,
two genes were also found in the dorsal fat of pigs Coromi- depending on the cellular setting (UniProt Consortium
nas et al. 2013). 2015). In this study, this gene was found associated with
The second candidate SNP found in C ­ HSE is OAR6_ the term autophagy — animal (oas04140) (Table S2c).
s36343.1, which is located in the ARAP2 gene and encodes Singh et al. (2016) found it related to programmed cell
the Arf-GAP with Rho-GAP domain, ANK repeat, and PH death or apoptosis. Second, SNP s16895.1 is found in
domain-containing protein 2. This protein modulates actin FBN2, which encodes the fibrilin-2 protein involved in
cytoskeleton remodeling by regulating ARF and RHO family the formation of a binding network with other proteins
members and is activated by binding to phosphatidylinositol to produce filiform filaments called microfibrils in the
3,4,5-trisphosphate (PtdIns 3,4,5 P3) (UniProt Consortium muscle. The microfibrils are a component of elastic fib-
2015). Bolormaa et al. (2016) also identified ARAP2 as a ers that allow the skin, ligaments, and blood vessels to
candidate gene associated with meat color and concentra- stretch (Frédéric et al. 2009). Researchers have suggested
tion of polyunsaturated fatty acids (omega-3 and omega-6). that fibrilin-2 plays a role in the direction of assembly of
For ­CHSS, we propose eight candidate SNPs, includ- elastic fibers during embryonic development. Microfibrils
ing five that are associated with WHC, namely s55275.1 also contribute to the establishment of more rigid tissues
located in PRKCQ; OAR11_26036763.1 in AIPL1; that support the crystalline lens, nerves, and muscles;
OAR2_151938109.1 in CSRNP3; OAR4_59276586.1 in furthermore, they contain the transforming growth fac-
IFRD1; and OAR24_27265846.1 in KDM8. Altogether, tor beta (TGF-β) that inactivates them. When the micro-
these genes participate in the metabolic pathway of “nucleus fibrils are released, TGF-β is activated and affects tissue
processes.” Particularly, PRKCQ encodes a protein kinase C growth and repairment throughout the organism (Frédéric
theta type, which belongs to the family of serine/threonine- et al. 2009; Ramirez & Dietz 2007). On the other hand,
specific kinase proteins that are activated by calcium and SNP OAR2_56383269.1 is located in the TPM2 gene that
diacylglycerol. These protein family members phosphorylate regulates the production of beta (β)-tropomyosin, which

13
Tropical Animal Health and Production (2023) 55:357 Page 11 of 13 357

regulates muscle fiber tension (muscular contraction) by and small-scale producers of the departments of Cesar and Cordoba
controlling myosin and actin binding. In non-muscle cells, for their collaboration. We also would like to thank Andrea Gonzalez
for translating the manuscript.
tropomyosin proteins play a role in controlling cell shape
(Tajsharghi et al. 2007; Ochala et al. 2007). β-tropomyosin Author contribution Conceptualization: LAA, MFA, ODV. Data cura-
is mainly found in skeletal muscles, which are used for tion: HAR and DLA. Formal analysis: HAR and DLA. Funding acqui-
movement. This protein helps to regulate muscle contrac- sition: LAA, HAR, and YAP. Investigation methodology: HAR, YAP,
and DLA. Supervision: LAA and DLA. Visualization: HAR, LAA,
tion by interacting with other muscle proteins, particu- and DLA. Writing—original draft: HAR and DLA. Writing—review
larly, myosin and actin. These interactions are essential editing: all authors.
to stabilize and maintain sarcomeres within muscle cells
(Tajsharghi et al. 2007; Ochala et al. 2007). Sarcomeres Funding Open Access funding provided by Colombia Consor-
tium. This work was supported by COLCIENCIAS, Grant Number
are the basic units of muscle contraction and they are com- 757 Programa de Becas Doctorales Colombia 2016.
posed of proteins that exert the mechanical force needed
for muscle contraction. Noce et al. (2018) determined that Data availability The data sets generated or analyzed during the cur-
TPM2, as well as ACTA1, MYLPF, MYH2, MYH7, TPM2, rent study, or both, are available from the corresponding author upon
reasonable request.
and TTN, encode myofibril proteins involved in muscle
contraction in the Longissimus dorsi muscle in autoch- Declarations
thonous Spanish sheep breeds (Canaria de Pelo, Roja
Mallorquina, Gallega, Xisqueta, and Ripollesa). Regard- Ethics approval This work had the endorsement of the ethics commit-
ing metabolic pathways associated with striated muscle tee of Universida Nacional de Colombia sede Palmira. Collection of
blood samples was carried out by veterinarians adhering to the regula-
contraction, we found that gluconeogenesis, glycolysis, tions and guidelines on animal husbandry and welfare.
the citric acid cycle, and the electron transport chain were
enriched. Finally, SNP OAR2_56383269.1 is located in Consent for publication All the authors consent to publish the manu-
the NANS gene that encodes an enzyme involved in sialic script.
acid biosynthesis. This protein uses N-Acetyl-mannosa-
mine 6-phosphate and mannose 6-phosphate as substrates Conflict of interest The authors declare no competing interests.
to produce phosphorylated forms of N-acetylneuraminic Open Access This article is licensed under a Creative Commons Attri-
acid (Neu5Ac) and 2-keto-3-deoxy-D-glycero-D-galacto- bution 4.0 International License, which permits use, sharing, adapta-
nononic acid (KDN) (Stelzer et al. 2016). tion, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source,
provide a link to the Creative Commons licence, and indicate if changes
were made. The images or other third party material in this article are
Conclusion included in the article's Creative Commons licence, unless indicated
otherwise in a credit line to the material. If material is not included in
We did not find SNPs significantly associated with pH prob- the article's Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will
ably since the measurements were taken at 7-day postmor- need to obtain permission directly from the copyright holder. To view a
tem, when most biochemical and physical processes are copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.
stable. Therefore, this trait must be evaluated at the moment
of slaughter and within 3-h postmortem to determine the
velocity at which the pH decreases.
References
We generated a baseline for sheep farmers who seek to
improve meat and carcass quality based on a genome-wide AI-Mamun, HA, Kwan, P, Clark, SA, Ferdosi, MH, Tellam, R. 2015.
association study of meat quality traits in Colombian Creole Genome-wide association study of body weight in Australian
hair sheep. We found that the LD muscle of ­CHSE releases Merino sheep reveals an orthologous region on OAR6 to human
and bovine genomic regions affecting height and weight. Genet.
fewer juice and, consequently, has greater WHC, suggest-
Sel. Evol. 47:66. https://​doi.​org/​10.​1186/​s12711-​015-​0142-4.
ing that these animals will exhibit better performance in the Arce-Recinos, C., Chay-Canul, A. J., Alarcón-Zúñiga, B., Ramos-
meat industry. Overall, we identified candidate genes that are Juárez, J. A., Vargas-Villamil, L. M., Aranda-Ibáñez, E. M., ... &
promising for genomic selection (ELOVL2, ARAP2, FBN2, Lopes Dias da Costa, R. (2021). Índices de eficiencia alimenticia
en ovinos de pelo: calidad de la carne y genes asociados. Revisión.
and TPM2) and are involved in muscle contraction, metabo-
Revista mexicana de ciencias pecuarias, 12(2), 523–552
lism, and fatty acid composition. Arevalo, G. Á., & Correa Assmus, G. 2013. Tecnología en la ovino-
cultura colombiana: estado del arte. Revista ciencia animal, 1(6),
Supplementary Information The online version contains supplemen- 125-142.
tary material available at https://​doi.​org/​10.​1007/​s11250-​023-​03688-z. Bolormaa, S., Hayes, B. J., van der Werf, J. H., Pethick, D., Goddard,
M. E., & Daetwyler, H. D. 2016. Detailed phenotyping identifies
Acknowledgements The authors would like to thank the sheep farmers genes with pleiotropic effects on body composition. BMC genom-
who contributed to this work, to Asovinos, el Hatico Natural Reserve, ics, 17(1), 1-21.

13
357 Page 12 of 13 Tropical Animal Health and Production (2023) 55:357

Corominas, J., Ramayo-Caldas, Y., Puig-Oliveras, A., Pérez-Montar- 1 and 3 on susceptibility to ovine progressive pneumonia virus
elo, D., Noguera, J. L., Folch, J. M., & Ballester, M. (2013). Poly- following natural exposure in sheep. Journal of Animal Science,
morphism in the ELOVL6 gene is associated with a major QTL 91(11), 5114-5121.
effect on fatty acid composition in pigs. PloS one, 8(1), e53687. Liu, H., Wang, C., Lee, S., Deng, Y., Wither, M., Oh, S., ... & Zhang,
Cheng, S., Wang, X., Zhang, Q., He, Y., Zhang, X., Yang, L., & Shi, J. G. 2016. Clipping of arginine-methylated histone tails by JMJD5
2020. Comparative transcriptome analysis identifying the different and JMJD7. Proceedings of the National Academy of Sciences,
molecular genetic markers related to production performance and 114(37), E7717-E7726.
meat quality in Longissimus dorsi tissues of MG× STH and STH Mestra-Vargas, L. I., Martínez-Reina, A. M., & Santana-Rodríguez,
sheep. Genes, 11(2), 183. M. O. 2019. Technical and economic characterization of lamb
DANE. Boletín técnico Encuesta de sacrificio de ganado. (2020). http://​ meat production in Córdoba, Colombia. Agronomía Mesoameri-
www.​dane.​gov.​co/​files/​inves​tigac​iones/​bolet​ines/​sacri​ficio/​bol_​ cana, 30(3), 871-884. http://dx.doi.org/https://​doi.​org/​10.​15517/​
sacrif_​IVtri​m14.​pdf am.​v30i3.​36931.
de Lima Júnior, D. M., de Carvalho, F. F., da Silva, F. J., Rangel, Milanesi, M., Capomaccio, S., Vajana, E., Bomba, L., Garcia, J. F.,
A. H. D. N., Novaes, L. P., & Difante, G. D. S. 2016. Intrinsic Ajmone-Marsan, P., & Colli, L. (2017). BITE: an R package for
factors affecting sheep meat quality: a review. Revista Colom- biodiversity analyses. bioRxiv.: 181610. Preprint, 29.
biana de Ciencias Pecuarias, 29(1), 03-15.doi: https://​doi.​org/​ Navarro, Y.T., Mendoza, M.A., & Osorio, D. S. D. 2015. Influencia de
10.​17533/​udea.​rccp.​v29n1​a01 la edad y el tiempo de madurez en la capacidada de retencion de
Expósito, C., Blanco, T. P., Peinado, J. M., García, V. D., Aldea, M. agua (CRA) en la carne de ovino criollo. @ limentech, Ciencia y
A., Martínez, A. G., ... & De La Cruz, R. A. (2003). Calidad Tecnología Alimentaria, 13(1), 41–47.
de la canal y de la carne en corderos ligeros de raza Segureña. Noce, A., Cardoso, T. F., Manunza, A., Martínez, A., Cánovas, A.,
Archivos de zootecnia, 52(199), 315–326. Pons, A., ... & Amills, M. 2018. Expression patterns and genetic
Flórez, M. J., Hernández, P. M., Bustamante, Y. M., & Vergara, G. variation of the ovine skeletal muscle transcriptome of sheep from
O. 2020. Caracterización morfológica y faneróptica de hembras five Spanish meat breeds. Scientific reports, 8(1), 1–7.
Ovino de Pelo Criollo Colombiano “CHS” Sudán. MVZ Cor- Ochala, J., Li, M., Tajsharghi, H., Kimber, E., Tulinius, M., Oldfors, A.,
doba, 84–93. https://​doi.​org/​10.​21897/​r mvz.​125 (2020). & Larsson, L. 2007. Effects of a R133W β‐tropomyosin mutation
Frédéric, M. Y., Monino, C., Marschall, C., Hamroun, D., Faivre, on regulation of muscle contraction in single human muscle fibres.
L., Jondeau, G., & Collod‐Béroud, G. 2009. The FBN2 gene: The Journal of physiology, 581(3), 1283-1292.
new mutations, locus‐specific database (Universal Mutation Palacios, Y. 2018. Análisis genómico de características de crecimiento
Database FBN2), and genotype‐phenotype correlations. Human muscular y de calidad de la canal en poblaciones ovinas de pelo.
mutation, 30(2), 181-190. Universidad de Colombia.
Frías, J. C., Aranda, E. M., Ramos, J. A., Vázquez, C., & Díaz, P. Pearce, K. L., Rosenvold, K., Andersen, H. J., & Hopkins, D. L. 2011.
2011. Calidad y rendimiento en canal de corderos en pastoreo Water distribution and mobility in meat during the conversion of
suplementados con caña de azúcar fermentada. Avances en muscle to meat and ageing and the impacts on fresh meat quality
Investigación Agropecuaria, 15(3), 33-44. attributes A review. Meat science, 89(2), 111-124.
Gallo, C., & Tadich, N. 2008. Bienestar animal y calidad de carne Pérez, P., Maino, M., Tomic, G., Köbrich, C., Morales, M. S., &
durante los manejos previos al faenamiento en bovinos. RED- Pokniak, J. 2006. Calidad de carne de corderos lechales del cruce
VET. Revista electrónica de Veterinaria, 9(10B). Suffolk Down x Merino Precoz Alemán: Efecto del peso de sac-
Gogarten, S. M., Bhangale, T., Conomos, M. P., Laurie, C. rificio y sexo. Archivos de Zootecnia, 55(210), 171-182.
A., McHugh, C. P., Painter, I., ... & Laurie, C. C. 2012. Purcell, S., Neale, B., Todd-Brown, K., Thomas, L., Ferreira, M. A.,
GWASTools: an R/Bioconductor package for quality control Bender, D., ... & Sham, P. C. 2007. PLINK: a tool set for whole-
and analysis of genome-wide association studies. Bioinformat- genome association and population-based linkage analyses. The
ics, 28(24), 3329–3331. American journal of human genetics, 81(3), 559–575. https://​doi.​
Hernández-Montiel, W., Martínez-Núñez, M. A., Ramón-Ugalde, org/​10.​1086/​519795.
J. P., Román-Ponce, S. I., Calderón-Chagoya, R., & Zamora- Ramírez-Bribiesca, E., Hernández-Cruz. L., Guerrero-Legarreta, I., &
Bustillos, R. 2020. Genome-wide association study reveals Hernández-Calva, L.M. 2007. Calidad de la carne y análisis sen-
candidate genes for litter size traits in pelibuey sheep. Animals, sorial en ovinos de pelo y lana provenientes de engorda intensiva
10(3), 434. en méxico. Vº Congreso de Especialistas en Pequeños Rumiantes
Hervé, M. 2013. Carne ovina: Producción, características y oportu- y Camélidos Sudamericanos, Mendoza, Argentina.
nidades en lo que hoy demanda el consumidor nacional e inter- Ramirez, F., & Dietz, H. C. 2007. Fibrillin‐rich microfibrils: Structural
nacional. Oficina de Estudios y Políticas Agrarias (ODEPA), determinants of morphogenetic and homeostatic events. Journal
Santiago, Chile. of cellular physiology, 213(2), 326-330.
Huff-Lonergan, E., & Lonergan, S. M. 2005. Mechanisms of water- Revelo, H. A., López-Alvarez, D., Landi, V., Rizzo, L., & Alvarez, L. A.
holding capacity of meat: The role of postmortem biochemical 2020. Mitochondrial DNA Variations in Colombian Creole Sheep
and structural changes. Meat science, 71(1), 194-204. Confirm an Iberian Origin and Shed Light on the Dynamics of
Joo, S. T., Kim, G. D., Hwang, Y. H., & Ryu, Y. C. 2013. Control of Introduction Events of African Genotypes. Animals, 10(9), 1594.
fresh meat quality through manipulation of muscle fiber charac- Rodríguez, A. B., Bodas, R., Landa, R., López-Campos, Ó., Mantecón,
teristics. Meat science, 95(4), 828-836. A. R., & Giráldez, F. J. 2011. Animal performance, carcass traits
Leal-Gutiérrez, J. D. 2013. Marcadores moleculares asociados a la and meat characteristics of Assaf and Merino× Assaf growing
Capacidad de Retención de Agua (CRA) en carne de Bos indicus lambs. Livestock Science, 138(1-3), 13-19.
y sus cruces (Doctoral dissertation). Sallé, G., Jacquiet, P., Gruner, L., Cortet, J., Sauvé, C., Prévot, F., ... &
Leal-Gutiérrez, J. D., Jiménez-Robayo, L. M., Ariza, M., Manrique, Moreno, C. R. 2012. A genome scan for QTL affecting resistance to
C., López, J., Martínez, C., ... & Jiménez, A. 2014. Efecto del tipo Haemonchus contortus in sheep. Journal of animal science, 90(13),
genético y la maduración sobre la retención de agua en carne de 4690–4705.
toros castrados. Archivos de zootecnia, 63(243), 409–418. Singh, P., Ravanan, P., & Talwar, P. (2016). Death associated protein
Leymaster, K. A., Chitko-McKown, C. G., Clawson, M. L., Harhay, kinase 1 (DAPK1): a regulator of apoptosis and autophagy. Fron-
G. P., & Heaton, M. P. 2013. Effects of TMEM154 haplotypes tiers in molecular neuroscience, 9, 46.

13
Tropical Animal Health and Production (2023) 55:357 Page 13 of 13 357

Stelzer, G., Rosen, N., Plaschkes, I., Zimmerman, S., Twik, M., Fishi- Genome-wide association identifies multiple genomic regions
levich, S., ... & Lancet, D. (2016). The GeneCards suite: from associated with susceptibility to and control of ovine lentivirus.
gene data mining to disease genome sequence analyses. Current PLoS ONE 7(10): e47829.https://​doi.​org/​10.​1371/​journ​al.​pone.​
protocols in bioinformatics, 54(1), 1–30 00478​29
Tajsharghi, H., Ohlsson, M., Lindberg, C., & Oldfors, A. 2007. Con- Yang, J., Lee, S. H., Goddard, M. E., & Visscher, P. M. (2011). GCTA:
genital myopathy with nemaline rods and cap structures caused a tool for genome-wide complex trait analysis. The American
by a mutation in the β-tropomyosin gene (TPM2). Archives of Journal of Human Genetics, 88(1), 76-82.
neurology, 64(9), 1334-1338. Yu, G. 2012. Reactome pathway analysis. R package version 1.10.1.
Torrescano, G.R., Sánchez-Escalante, A., Peñúñuri-Molina, F. J., Zhang, L., Yue, H. Y., Zhang, H. J., Xu, L., Wu, S. G., Yan, H. J., ... &
Velázquez-Caudillo, J., & Tineo-Sierra, R. 2009. Características Qi, G. H. 2009. Transport stress in broilers: I. Blood metabolism,
de la canal y calidad de la carne de ovinos pelibuey, engordados glycolytic potential, and meat quality. Poultry science, 88(10),
en hermosillo, sonora. Biotecnia, 11(1), 41-50. 2033-2041.
UniProt Consortium. 2015. UniProt: a hub for protein information. Zhao, F. P., Wei, C. H., Zhang, L., Liu, J. S., Wang, G. K., Tao, Z. E. N.
Nucleic Acids Research, 43(D1), D204-D212. G., & Du, L. X. (2016). A genome scan of recent positive selec-
Warner, R.D., Greenwood, P. L., Pethick, D. W., & Ferguson, D. M. tion signatures in three sheep populations. Journal of Integrative
2010. Genetic and environmental effects on meat quality. Meat Agriculture, 15(1), 162-174.
science, 86(1), 171-183. doi:https://​doi.​org/​10.​1016/j.​meats​ci.​
2010.​04.​042. Publisher's note Springer Nature remains neutral with regard to
White, S. N., Mousel, M. R., Herrmann-Hoesing, L. M., Reynolds, J. jurisdictional claims in published maps and institutional affiliations.
O., Leymaster, K. A., Neibergs, H. L., ... & Knowles, D. P. 2012.

13