Effect of the IGF-I gene polymorphism on growth, body size, carcass and meat quality traits in Coloured Polish Merino sheep

Insulin-like growth factor I, encoded by the IGF-I gene, plays a role in cell growth and differentiation, embryogenesis, metabolism regulation, skeletal growth, and protein synthesis. The aims of this study were to investigate the polymorphism in the 5 flanking region of the IGF-I gene and evaluate associations between the single-nucleotide polymorphism (SNP) in this gene and growth, body size, carcass and meat quality traits in Coloured Polish Merino sheep. In total 78 live and post mortem traits were investigated. Polymorphism in the IGF-I gene was identified with the use of the polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP) method in 305 Coloured Polish Merino sheep. In association studies, traits of interest were analysed with the use of the MIXED and GENMOD procedures of the SAS statistical package. Two alleles named A and B, and two IGF-I genotypes – AA and AB – were detected. The A allele and the AA genotype were predominant, with the frequencies of 91.6 and 83.3 %, respectively. The IGF-I genotype was found to have a highly significant effect on fore shank weight (P = 0.006), kidney fat class (P = 0.002) and EUROP fat class (P = 0.005). Furthermore, the IGF-I genotype significantly affected external fatness of carcass class (P = 0.038), drip loss (P = 0.049), and subjective assessment of meat colour (P = 0.043), and it tended to be associated with longissimus dorsi (LD) muscle width (P = 0.063) and flavour (0.067). Concluding, the IGF-I gene could be considered as a candidate gene of selected carcass and meat quality traits in sheep.


Introduction
Insulin-like growth factor I (IGF-I) is a polypeptide hormone that is encoded by the IGF-I gene and is similar in molecular structure to insulin.IGF-I participates in the somatotrophic axis together with growth-hormone-releasing hormone (GHRH), growth hormone (GH), insulin-like growth factor II (IGF-II), and their associated binding proteins (BPs) and receptors (GHRHR, GHR, IGF-IR and IGF-IIR).Moreover, it interacts with insulin-like growth factor I receptors (IGF-IR) in target tissues (Jones and Clemmons, 1995) and with proteins binding IGF in blood (IGF-BP), which can modulate action of this hormone (e.g.Lackey et al., 1999;Schams et al., 1999).IGF-I is also believed to mediate a wide spectrum of biological responses such as cell growth and differentiation, embryogenesis, metabolism regulation, skeletal growth, and protein synthesis (Baxter, 1986;Clemmons et al., 1987;Froesch et al., 1985).For these reasons, the IGF-I gene is considered as a functional candidate for a number of production traits, i.e. growth, carcass and meat quality traits in livestock.
Growth traits are important attributes in sheep breeding, as they, among other factors, affect breeder's profit.Lamb body weight is one of the factors that regulate incomes from meat Published by Copernicus Publications on behalf of the Leibniz Institute for Farm Animal Biology.
production.Therefore, the improvement of these traits should be one of the most important aims in sheep production.Growth traits, like other quantitative traits, are controlled by several genes, e.g. the IGF-I gene, and environmental factors.However, there have been only a few association analyses of the IGF-I gene polymorphism with growth traits in sheep.For example, Hajihosseinlo et al. (2013), Tahmoorespur et al. (2009), Negahdary et al. (2013) and Gholibeikifard et al. (2013) observed significant effects of the IGF-I gene polymorphism on several growth traits in Makooei and Baluchi sheep.Moreover, Hajihosseinlo et al. (2013) investigated the association of nucleotide substitution in the IGF-I gene with several body size traits in sheep.It should be pointed out that the abovementioned studies were conducted only in Iranian sheep breeds.Therefore, there is a lack of information about these associations in sheep breeds in other countries, especially in Europe.To our knowledge, until now only Proskura and Szwczuk (2014) had investigated effects of single-nucleotide polymorphism (SNP) in the third exon of the IGF-I gene on growth traits in Pomeranian Coarsewool sheep in Poland and there was no association study involving the IGF-I gene and body size traits in countries other than Iran, nor any subjective assessment of body muscle and fat class in sheep.Interestingly, in cattle, several authors have found significant associations between the IGF-I gene polymorphism and growth traits (e.g.De la Rosa Reyna et al., 2010;Ge et al., 2001;Li et al., 2004).Moreover, Zhang et al. (2008) reported significant effects of nucleotide substitution in the IGF-I gene on several body size traits in goats.Additionally, Gao et al. (2009) showed that the IGFBP-3 gene polymorphism was associated with chosen body measurements in Chinese beef cattle.
Despite the fact that the IGF-I gene plays a role in the growth of an organism, it could also be directly or indirectly associated with other traits, i.e. carcass and meat quality traits, as SNPs in this gene could be markers in linkage with causative mutations in other genes.For example, Behzadi et al. (2015) investigated effects of the IGF-I gene polymorphism on carcass traits in sheep.Moreover, Sun et al. (2014) studied correlations between the IGF-I gene expression and carcass and net meat weight in Hu sheep.Siadkowska et al. (2006) and Curi et al. (2005) observed associations between the IGF-I gene polymorphism and several carcass traits in cattle.In the case of meat quality traits, there have only been a few association analyses in sheep.Behzadi et al. (2015) investigated effects of the IGF-I gene on the level of triglycerides and cholesterol in blood.Su et al. (2014) analysed relationships between the IGF-I gene expression and diameter and density of muscle fibre as well as muscle tenderness in Hu sheep.In pigs, Wang et al. (2009Wang et al. ( , 2010) ) observed effects of the IGFBP-3 and IGFBP-5 gene polymorphisms on meat colour and pH of various meat cuts, respectively.
Among many factors affecting growth, body size, carcass and meat quality traits, effects of many candidate genes still need to be investigated, especially in sheep.Identification of associations between candidate gene polymorphisms and the abovementioned attributes can provide useful markers for selection of economically favourable traits.However, little is known about the effect of the IGF-I gene polymorphism on growth, body size, carcass and meat quality traits in sheep, especially in European breeds.Moreover, to our knowledge no previous association studies between IGF-I genotypes and many important carcass and meat quality traits in sheep have been conducted.Therefore, the aims of the present study were to investigate the polymorphism in the 5 flanking region of the IGF-I gene and to evaluate the associations between SNP in this gene and a wide spectrum of growth, body size, carcass and meat quality traits in Coloured Polish Merino sheep.

Materials and methods
In total, 305 purebred Coloured Polish Merino lambs of both sexes were investigated.The Coloured Polish Merino is a sheep breed used for its wool and meat.It is included in the Programme of Farm Animal Genetic Resources Conservation supervised by the National Research Institute of Animal Production (NRIAP) in Poland.Sheep were kept indoors, but they also grazed on a pasture six times a week at the NRIAP Experimental Station Kołuda Wielka.Lambs were sired by nine different rams and were produced during a period of 3 years -from 2011 to 2013.Suckling lambs were fed dry granulated mash and meadow hay ad libitum.Procedures involving animals were approved by the local animal research ethics committee and the local veterinary service.
Growth data (body weight on the second, 30th, 56th and 78th days of life) were collected for 277 male and female lambs.Moreover, the average daily gains between the second and 30th, 30th and 56th, and 56th and 78th days were calculated.
In total, 305 male and female lambs were ultrasonically (USG) scanned according to the procedure described by Krupiński et al. (2009) using an Aloka SSD-900 device with a UST-5818-5 transducer (B-5MHz).In brief, USG examination of a cross section of the longissimus dorsi (LD) muscle was performed on the right body side perpendicularly to the spine axis behind the last thoracic vertebrae.Measurements of width, depth and area of the LD muscle and fat depth over LD muscle were performed on USG images with the use of MultiScan ver.18.03 software.
In total, 106 ram lambs were chosen for slaughter during the 3 experimental years.Before slaughter, the following body measurements were carried out on lambs: withers height, chest depth, body length, shoulder width, hip width and leg depth.The measurements were taken while the lambs were in a standing position, according to the methodology of Kawęcka (2013).Subjective evaluations of muscling as well as fatness with the use of a five-point system (ranges from one to five with a 0.5 precision, with a value of five being the meatiest or fattiest class) were undertaken in the lumbar part of the spine behind the last rib near the kidney according to the methodology of Jarrige (1988).Slaughters, carcass cutting and leg dissection took place three or four times a year from March to April according to the methodology of Nawara et al. (1963) and Krupiński et al. (2009) in the abattoir of the NRIAP Experimental Station Kołuda Wielka.Mean age at slaughter was 105 days (SD 4.2, range 92-119 days).In brief, lambs were weighed twice: 1 day before slaughter (final weight) and on the day of slaughter (pre-slaughter weight).Then animals were electrically stunned, exsanguinated and skinned.Subsequently, carcasses were weighed (hot carcass weight) and maintained in a chilling room, where the temperature was held at approximately 4 • C for 18 h.After chilling carcasses were weighed again (cold carcass weight) and the cold carcass dressing out was calculated by dividing the cold carcass weight by the preslaughter weight and multiplying by 100.In the next step, carcasses were visually graded for muscle and fat class on a nine-point scale (with the value of nine being the meatiest or fattiest class) according to the methodology implemented at ram evaluation stations in Germany as described by Krupiński et al. (2009).Conformation and muscling were assessed on each of the three parts of carcass separately: on the fore part of the carcass consisting of middle neck and shoulder, on the full loin part including rib and loin, and on the leg part.The evaluation of fatness involved two traits: kidney fat class and external fatness of carcass class.Carcasses were also categorised in terms of muscle and fat class according to the EUROP grading system as described by Krupiński et al. (2009).Subsequently, the following dimensions of carcasses were measured according to the methodology of Nawara et al. (1963): carcass length, breast depth, breast width, rump width, rump girth, leg depth, leg length index and loin width.Then carcasses were halved, and the right carcass side was weighed and taken for further examination.The kidney, kidney fat and the rest of the diaphragm were removed.At first, the right carcass side was divided into three parts: the fore part of the carcass, the full loin part, and the leg part, and each part was weighed.Subsequently, these three parts were divided into the following cuts: scrag, middle neck, shoulder, breast and flank, rib, loin, tenderloin, leg, fore shank, and hind shank, according to the methodology of Nawara et al. (1963).All cuts were weighed.On the surface, where a rib was cut from a loin, width, depth and the area of the eye of loin and fat depth over the eye of loin were measured.Moreover, fat depth over the ribs in the thickest point of this layer was recorded.The leg was dissected for three tissues: muscles, bones and fat, which were weighed separately and a yield of each tissue in the leg was calculated.
Furthermore, loin muscles from the right carcass side were vacuum-packed and transported to the Institute of Agricultural and Food Biotechnology (Poland), where meat quality analyses were undertaken.The pH measurements were performed on samples of longissimus lumborum (LL) muscle 24 h post mortem (pH24h) using a German Mettler Toledo 1140 pH meter with an integrated Mettler Toledo electrode (ISO 2917(ISO , 2001)).Results were averaged from the three measurements per one sample.For an instrumental evaluation of meat colour, steaks (thickness approximately 10 mm each) were cut crosswise in the direction of the LL muscle fibres and exposed to daylight or electric light for 15 min.The values of reflectance coordinates -L * (lightness), a * (redness) and b * (yellowness) -were gathered using a Konica Minolta chronometer CR-400.Drip loss value was measured on the LL muscle.Before, the test samples were weighed, individually packed in plastic bags, held at 4 • C for 48 h and weighed again.The value of drip loss was defined as a percentage weight loss calculated from the difference between initial and final weight of the sample.Water, intramuscular fat and total protein content percentages were determined on minced samples of the LL muscles according to the methods described in ISO 1442ISO (2000ISO ), 1444ISO (2000) ) and PN-75/A-04018 (2000), respectively, using a Kjeltec system 1002 distilling unit and a Soxtherm device manufactured by Gerhardt Analytical Systems.Water holding capacity was investigated in minced LL muscle samples according to the method devised by Grau and Hamm (1952) with later modifications made by Pohja and Ninivaara (1957).Cooking loss was analysed on samples of the longissimus dorsi (LD) muscle.First of all, the samples were weighed, then packed in plastic bags, heated in water until they reached 75 • C in the centre of the sample and were weighed again at the end of the analysis.The value of cooking loss was calculated from a difference in sample weights recorded before and after the process of heating and was expressed as a percentage.Furthermore, shear force was determined in cooked and cooled-down samples of lamb loin.For this purpose cylindershaped samples (2.5 cm diameter) of the LD muscle were extracted parallel to the direction of muscle fibres and were subjected to shear force (N cm −2 ) measurement with the use of a Zwick Roell ZO with a 0.5 kN head and Warner Bratzler device with a blade speed of 100 mm min −1 .Moreover, subjective visual evaluations of meat colour and marbling were performed on the LL muscle by a team of five people using an eight-point scale for colour, where one is related to the lightest and eight is related to the darkest colour, and a fourpoint scale for marbling, with a score of one relating to minor marbling and four relating to the greatest marbling.Subjective sensory evaluation of aroma, succulence, tenderness and flavour were performed on boiled LD muscle using a fivepoint scale for each trait according to the methodology of Baryłko-Pikielna (1975), where one was related to bad levels and five was related to very good levels of the trait.
Total genomic DNA was purified from the blood using a MasterPure TM DNA Purification Kit for Blood Version II (Epicentre, USA).Polymorphism in the 5 flanking region of the ovine IGF-I gene was detected with the use of the polymerase chain reaction restriction fragment length poly-morphism (PCR-RFLP) method.A fragment of the ovine IGF-I gene was amplified using primers reported by Ge et al. (1997).DNA amplifications were performed in a Mastercycler Pro (Eppendorf AG, Germany) in 20 µL reaction volume containing 50 ng of genomic DNA, 5 pmol of each primer (forward and reverse), 200 µM of each dNTP and 1 U DreamTaq DNA polymerase (Thermo Scientific, USA) in a one-fold DreamTaq reaction buffer.The temperature profile of the reaction consisted of denaturation at 94 • C for 2 min, followed by 30 amplification cycles, including denaturation at 94 • C for 30 s, annealing at 66 • C for 30 s and extension at 72 • C for 30 s, with a final 5 min extension step at 72 • C. The PCR reaction products were digested with 4 U of HaeII (NEB, UK) restriction enzyme (Yilmaz et al., 2005) for 2 h at 37 • C. Digested PCR products were separated on 2 % agarose gels in 1 × TBE (Tris-borate-EDTA) buffer for 90 min at 120 V and visualised by Midori Green (Nippon Genetics, Germany) staining.PCR products representing the AA and AB genotypes were cleaned up with the use of an ExoSAP-IT ® Affymetrix (USA) and sequenced in both directions in Genomed, Poland.Allele and genotype frequencies in the IGF-I locus were calculated.Furthermore, observed and expected heterozygosity as well as the Hardy-Weinberg equilibrium test calculations were made in the Arlequin 3.5.1.2(Excoffier and Lischer, 2010).
An association analysis was performed between IGF-I genotypes and growth traits using the MIXED procedure of the SAS software package (SAS, 2008).The following mixed effect model I was applied: where Y ij klm is the performance of the nth individual lamb for each trait of interest, µ is the general mean for each trait of interest, a i is the fixed effect of the ith IGF-I genotype (i = AA, AB), b j is the fixed effect of the j th sex (j = male, female), c k is the fixed effect of the kth litter size (k = 1, single; 2, twin and only one lamb born with triplet litter size was also included), d l is the fixed effect of the lth year of observation (l = 2011, 2012, 2013), f m is the random effect of the mth sire (m = sire 1, 2, . . .9) and e ij klm is the random error.For all variables two-way interactions between fixed effects were tested in the model; however, they did not have a significant effect on investigated traits and were therefore excluded from the final model.An association analysis between IGF-I genotypes and body size traits as well as IGF-I genotypes and slaughter traits was undertaken with the use of the model I described above without the effect of the sex because only male lambs were investigated.Moreover, final weight and cold carcass weight were found to be significant and were included in the final models as a covariate for live and post mortem attributes, respectively.Weight was not included as the covariate in the model where the Y variable was the final weight, pre-slaughter weight, a tissue proportion or dressing out.In the case of final and pre-slaughter weight, slaughter age was found to be significant and was included as the covariate in the model.Furthermore, final weight was fitted in the model as the covariate for hot and cold carcass weight.For all variables, two-way interactions between fixed effects were tested in the model but were not significant for almost all of the traits, with the exception of withers height, chest depth, leg depth (body measurement), scrag weight, and hot and cold carcass weight; therefore, they were included in the models only for the abovementioned traits, but not for all other characteristics.
Moreover, an association of IGF-I genotypes with USG measurements of the LD muscle was also estimated with the use of the model I described above; however, slaughter age was included as the covariate.For all variables, two-way interactions between fixed effects were tested in the model but did not have a significant effect on investigated traits, with the exception of fat depth over the LD muscle; therefore, it was fitted in the model only for this trait.An association analysis between IGF-I genotypes and meat quality traits was performed with the use of the model I described above without the effect of sex and litter size.For all variables, two-way interactions between fixed effects were tested in the model but did not show any significant effect on investigated traits; therefore, they were not included in the final model.In the case of each model, when a genotype was shown to be statistically significant, the significance of deviations was verified with the Tukey-Kramer test.
Associations of IGF-I genotypes with subjective assessment of live animal and carcass traits were analysed with the use of the GENMOD procedure of the SAS software package (SAS, 2008).The generalised linear model included the fixed effects of IGF-I genotype, litter size, year of observation and sire.Moreover, final weight and cold carcass weight were found to be significant and were fitted in the model as the covariates for live and post mortem attributes, respectively.4. None of the analyses showed any significant effect of the IGF-I genotype on the abovementioned characteristics; however, the IGF-I gene polymorphism tended to be associated with LD muscle width (P = 0.063).AB heterozygous lambs had wider LD muscle (0.14 cm) than in AA homozygotes (Table 4).Results of the generalised linear model analyses of the IGF-I genotype effect on body muscle and fat class are presented in Table 7.No associations between the IGF-I gene polymorphism and these traits were found.

Effect of the IGF-I gene polymorphism on carcass traits
Tables 5 and 7 show the results of association analyses between the IGF-I gene polymorphism and carcass traits in Coloured Polish Merino sheep.IGF-I genotype was associated with fore shank weight (P = 0.006).Heterozygous lambs had heavier fore shank (4.3 %) than AA homozygous animals.Furthermore, the IGF-I genotype was found to have highly significant effects on kidney fat class (P = 0.002) and EUROP fat class (P = 0.005) and a significant effect on external fatness of carcass class (P = 0.038).No significant associations between IGF-I genotype and other investigated carcass traits were detected.

Effect of the IGF-I gene polymorphism on meat quality traits
Results of association analyses of the IGF-I gene polymorphism with meat quality traits in Coloured Polish Merino sheep are presented in Table 6.The IGF-I gene polymorphism was significantly associated with drip loss (P = 0.049).Meat from AA homozygous lambs was characterised by lower average drip loss (−0.86 %) than meat from heterozygotes AB.Moreover, the IGF-I gene polymorphism had a significant effect on the subjective assessment of meat colour (P = 0.043).AA homozygous ram lambs had meat characterised by the higher average value of this trait (0.34 point) when compared to meat form heterozygotes AB.Furthermore, the IGF-I genotype tended to be associated with the flavour (P = 0.067).

Discussion
In the present study, polymorphism in the 5 flanking region of the IGF-I gene and its association with growth, body size, carcass and meat quality traits in Coloured Polish Merino sheep were analysed.Two alleles, named A and B, and only two genotypes, AA and AB, were detected.The A allele was predominant (91.6 %), which consequently resulted in the higher frequency of the AA homozygotes (83.3 %) than the AB heterozygotes.These results are consistent with the study of Yilmaz et al. (2005)   The IGF-I locus was characterised by a low genetic diversity in the investigated Coloured Polish Merino sheep population.The value of observed heterozygosity (0.167) was slightly higher than the value of expected heterozygosity (0.153).Moradian et al. (2013) also observed two alleles in the IGF-I locus in Makooei sheep; nevertheless, values of H o and H e were higher: 0.42 and 0.40, respectively.The Coloured Polish Merino sheep population was in the Hardy-Weinberg equilibrium (P = 0.247), indicating no selection for investigated IGF-I locus.Similarly, Negahdary et al. (2013), who investigated polymorphism of the 5 flanking region of the IGF-I gene, reported that the population of Makooei sheep was in the Hardy-Weinberg equilibrium (P > 0.05).Conversely, Nazari et al. (2016), who studied the IGF-I locus in Zandi sheep, showed that this population was not in the Hardy-Weinberg equilibrium (P < 0.01).
The IGF-I gene affects many important processes in an organism, growth being among them (Akers, 2006;Burgos and Cant, 2010).Circulating IGF-I concentration has an impact on fetal and neonatal size and postnatal growth in several species (Baker et al., 1993;Breier et al., 1988;Duclos et al., 1999;Yakar et al., 2002;Zapf and Froesch, 1999).However, association analysis did not show significant effects of the IGF-I gene polymorphism on growth traits in Coloured Polish Merino sheep.Similarly, Nazari et al. (2016) did not find significant associations between SNP in the 5 flanking region of the IGF-I gene and growth traits in Zandi sheep.Also, Proskura and Szewczuk (2014) did not show any relationships between the C/T substitution (g.271C > T) in the IGF-I gene and growth traits in Pomeranian Coarsewool sheep in Poland.Moreover, Gholibeikifard et al. (2013) did not observe the effects of SNP, located in the third exon of the IGF-I gene, on growth traits in Baluchi sheep.Conversely, Tahmoorespur et al. (2009) showed significant associations of the IGF-I gene polymorphism in the 5 flanking region with average daily gain (ADG) from birth to weaning in Baluchi sheep.Heterozygous lambs had higher ADG from birth to weaning than homozygotes AA and BB.Furthermore, Hajihosseinlo et al. ( 2013) observed significant effects of nucleotide variation in the 5 flanking region of the IGF-I gene with several growth traits in Makooei sheep: birth weight (BW), weaning weight (WW), 6-month weight (SW) and average daily gains from birth to weaning (GBW).Additionally, a significant association of polymorphism in the 5 flanking region of the IGF-I gene with BW, WW, SW, breeding value estimated for body weight in the sixth month of life (EBV 6MW), GBW and average daily gains from the sixth month to the ninth month of life (GSN) were found in Makooei sheep (Negahdary et al., 2013).AG heterozygous Makooei sheep had higher values of the traits EBV 6MW, 6MW and WW than homozygotes.Conversely, AA homozygous sheep had higher BW and GBW (Negahdary et al., 2013).Furthermore, Sun et al. (2014), who investigated the effects of different factors on the level of the IGF-I gene expression and its associations with growth traits, observed a positive correlation of this gene's expression with body weight in Hu sheep.Other authors analysed the effects of the IGF-I gene polymorphisms with growth traits in other livestock species.For example, Ge et al. (2001) showed associations between variation in the IGF-I gene and growth traits in Angus cattle and suggested that this polymorphism could have an impact on gene transcription and consequently on animal phenotype.Szewczuk et al. (2013) reported effects of two SNPs in the IGF-I gene on body weight in the second month of life in calves and average daily gains in the periods from the first to the second months and from the second to the third months of life and for the whole rearing period.Zhang et al. (2008) investigated polymorphism of the IGF-I gene and its association with growth and body size traits in Nanjiang Huang goats.They found a significant effect of G/C substitution on birth weight, body weight at 6 months, body weight at 12 months, heart girth at 2 months, body length at 6 months, wither height at 6 months, wither height at 12 months and heart girth at 12 months in the goats.Concluding, there are analyses that confirm associations of the IGF-I gene polymorphisms with growth traits in sheep and other livestock species; however, there are other reports that are opposite to these findings.The growth of animals is a complex process that is under the control of several genes and environmental factors.Moreover, the effects of the IGF-I gene polymorphism on growth traits seem to depend on sheep breed.Therefore, before introducing DNA tests in sheep breeding, additional analyses of this gene's polymorphism and its association with growth traits in different sheep breeds should be undertaken.
The IGF-I hormone plays a crucial role in the postnatal linear growth of animals (e.g.Yakar et al., 2002;Zapf and Froesch, 1999).In this study an association of the IGF-I gene polymorphism with body size traits and subjective assessment of body muscle and fat class in Coloured Polish Merino sheep was investigated; however, the effects of IGF-I genotypes on these traits were not significant.Conversely, Hajihosseinlo et al. (2013), who investigated associations of the IGF-I gene polymorphism in the 5 flanking region with such body size traits as height and length of body, wither height, chest width and rump length in Makooei sheep, found a significant effect of this gene's genotypes on sheep body length.Moreover, Chelongar et al. (2014) observed a significant effect of SNP in the first exon of the IGF-I gene on fat-tail fat thickness (the thick rump).AA homozygotes were superior in terms of this trait, whereas GG male lambs had the lowest fat thickness.The associations of nucleotide variation in the first exon of the IGF-I gene with tail length and width (rump length and width) in Makooei sheep was not significant (Chelongar et al., 2014).Associations between the IGF-I gene polymorphism and body size traits were also investigated in other livestock species.For example, Zhang et al. (2008) found the significant effects of G/C substitution in the fourth intron of the IGF-I gene on heart girth at 2 months, body length at 6 months, wither height at 6 months, wither height at 12 months and heart girth at 12 months in Nanjiang Huang goats.Gao et al. (2009) showed that the polymorphism in the IGFBP-3 locus was associated with rump width and heart girth at 24 and 36 months in Chinese beef cattle.Furthermore, Mullen et al. (2011) reported that SNP in the IGF-I gene (IGF1i6; G allele) was positively associated (P < 0.05) with body condition score in Holstein-Friesian dairy cattle, while Lynch et al. (2010) showed a relationship between IGF1i2 and body condition score at calving in a cohort of 241 dairy cows.IGF-I protein affects longitudinal bone growth by promoting osteoblast division and proliferation.Furthermore, it regulates muscle growth by enhancing myocyte differentiation and multiplication and has an impact on cartilage growth by increasing chondrocyte colony formation (Duclos et al., 1999;Yakar et al., 2002;Zapf and Froesch, 1999).It should be pointed out that there is limited information about effects of the IGF-I gene polymorphism on body size traits, especially in sheep.Moreover, to our knowledge it was the first association analysis of IGF-I genotypes and subjective assessment of body muscle and fat class in sheep.Therefore, similar association analyses should be undertaken in other sheep breeds.
IGFs, their receptors and BPs play a crucial role in muscle growth and differentiation (Oksbjerg et al., 2004).In this study effects of the IGF-I gene polymorphism with ul-  trasound measurements of LD muscle in Coloured Polish Merino sheep was not significant; however, it tended to be associated with the width of this muscle.Additionally, LD muscle dimensions were also measured post mortem, but no sig-nificant associations between the IGF-I gene polymorphism and these traits were showed.No other paper was found in the literature concerning association between IGF-I genotypes and ultrasound LD muscle measurements in sheep.In-terestingly, in cattle the SNP in the IGF-I gene tended to be associated with the ultrasound LD area (P = 0.06) (Curi et al., 2005).As the current study was the first association analysis of IGF-I genotypes with ultrasound LD muscle measurements in sheep, similar analysis in other sheep breeds should be undertaken in order to investigate possible breed effects on these traits.
In this study effects of IGF-I genotypes on carcass traits in Coloured Polish Merino sheep were investigated.Among 40 traits, significant associations between IGF-I genotypes and fore shank weight, EUROP fat class, kidney fat class and external fatness carcass class in Coloured Polish Merino ram lambs were found.Similarly, Behzadi et al. (2015) investigated effects of the IGF-I gene polymorphism on fatrelated traits in Mehraban sheep.They observed a tendency for an association between SNP in this gene and dorsal fat thickness (P = 0.07); however, effects of this nucleotide variation in abdominal fat and fat-tail weight were not significant.Furthermore, Siadkowska et al. (2006) reported correlations between polymorphism in the 5 non-coding region of the IGF-I gene and lean and fat weight of valuable cuts and a percentage of fat of valuable cuts in Polish Holstein-Friesian cattle.Moreover, concerning the IGF2 gene polymorphism, CT heterozygous Polish Holstein-Friesian bulls had more fat in valuable cuts in the 15th month of life (Zwierzchowski et al., 2010).Also in cattle, Curi et al. (2005) observed a significant association of IGF-I/SnaBI polymorphism in the regulatory region of the IGF-I gene with subcutaneous back fat assessed by ultrasonography.In pigs, Wang et al. (2006) investigated effects of polymorphism of the IGF-I receptor gene (IGF-IR) on back-fat thickness and lean percentage estimated by ultrasonography, and they found less back-fat thickness (P < 0.05) and a greater lean percentage (P < 0.01) in AA homozygous Yorkshire pigs than BB individuals.To summarise, the results of the abovementioned research showed that polymorphisms in the IGF-I gene, as well as IGF2 and IGF-IR, were associated with several fat traits of sheep, cattle and pig carcasses.Indeed, the IGF-I gene was shown to be involved in fat cell development in transgenic mice (Rajkumar et al., 1999).Rajkumar et al. (1999) observed enhanced expression of this gene during differentiation of precursor cells into mature fat cells.Furthermore, Anderson et al. (1988) found that circulating IGF-I protein was negatively correlated with carcass fat percentage, fat accretion rate and fat thickness in Simmental crossbred bulls.Additionally, Davis and Simmen (1997) found associations between lower plasma IGF-I protein concentrations and higher marbling scores as well as dorsal fat thickness in Angus bulls.Concluding, the IGF-I gene should be considered as a potential candidate gene of fat-related carcass traits in sheep.
As mentioned before, among several carcass and carcass cut weights, IGF-I genotype was only associated with fore shank weight in Coloured Polish Merino lambs.Contrary to our results, other authors showed an effect of this gene's polymorphism on carcass weights.For example, Behzadi et al. (2015) reported a tendency of the IGF-I gene polymorphism to be associated with carcass weight (P = 0.07).Moreover, Sun et al. (2014) found that an expression of the IGF-I gene was positively correlated with carcass weight in Hu sheep.Siadkowska et al. (2006) observed a correlation between polymorphism in the 5 non-coding region of the IGF-I gene and cold carcass weight in Polish Holstein-Friesian cattle.Curi et al. (2005) found significant association of IGF-I/SnaBI polymorphism in the regulatory region of the IGF-I gene with hot carcass weight, but the effect of this SNP on carcass yield was insignificant in beef cattle.Zwierzchowski et al. (2010) reported effects of the IGF2 gene polymorphism on cold carcass weight and percentage meat content in valuable cuts at the age of 11 months in Polish Holstein-Friesian cattle.It should be pointed out that there is a dearth of information about the association of the IGF-I gene polymorphism with carcass traits in sheep.To our knowledge, until now only Behzadi et al. (2015) had investigated effects of SNP in this gene on carcass traits in Iranian sheep.In the current study we reported results of association analyses in Polish sheep breeds involving new carcass traits, which have not been investigated before.These results suggest that the IGF-I gene should be considered as a potential candidate gene of carcass traits in sheep, and similar studies involving other sheep breeds in different countries should be continued.
In this study, association analysis between the IGF-I gene polymorphism and meat quality traits was undertaken.A significant effect of IGF-I SNP on natural drip loss in Coloured Polish Merino sheep was found.Meat from AA male lambs had lower natural drip loss than meat from heterozygous individuals.Moreover, an association of this nucleotide variation in the IGF-I gene with subjective meat colour was observed.In this case, meat from AA homozygous ram lambs was slightly darker than meat from heterozygotes.Little is known about associations of the IGF-I gene polymorphism with meat quality traits in sheep.To our knowledge, this study is the first that presents the effect of IGF-I genotypes on several new meat quality traits in sheep.Other authors have shown association analyses of the IGF-I gene polymorphism with various meat quality traits in sheep, but with different traits than investigated in this study.For example, Behzadi et al. (2015) found significant effects of nucleotide variation in the first exon of the IGF-I gene with the level of triglycerides and cholesterol in blood.Su et al. (2014) analysed relationships of the IGF-I gene expression in LD muscle with meat quality traits in Hu sheep.They observed that the expression of this gene was positively and significantly (P<0.01)correlated with muscle fibre diameter and muscle fibre shear stress, and negatively and significantly (P < 0.01) correlated with muscle fibre density.Furthermore, Davis and Simmen (1997) showed that lower plasma IGF-I protein concentrations were associated with higher marbling scores in Angus cattle.Wang et al. (2009) observed an effect of the IGFBP-3 gene polymorphism on meat colour in pigs.More-over, Wang et al. (2010) found an association between the IGFBP-5 gene and pH of various meat cuts in pigs.SNPs in the IGF-I gene may have a direct or indirect effect on meat quality traits in sheep, cattle and pigs.There is also a possibility that SNPs detected in the IGF-I gene are not causative mutations for the investigated effects on meat quality traits, but are markers in linkage with the causative mutations in the other genes.As the results of this study constitute the first information gathered on the IGF-I gene polymorphism association with several meat quality parameters in sheep, further investigation should be carried out to evaluate potential breed-specific effects of SNPs in this gene on lamb quality.

Conclusion
In the present study, information about the IGF-I gene polymorphism and its effect on a wide range of growth, body size, carcass and meat quality traits in Coloured Polish Merino sheep were reported.Associations between IGF-I genotype and fore shank weight (P = 0.006), kidney fat class (P = 0.002), EUROP fat class (P = 0.005), external fatness of carcass class (P = 0.038), natural drip loss (P = 0.049) and subjective assessment of meat colour (P = 0.043) were found.Therefore, the IGF-I gene polymorphism could be used as a marker for an improvement of these traits in Coloured Polish Merino sheep.Moreover, in the case of several investigated traits, the present study provided the first information about the effect of this marker in sheep.Concluding, the results suggest that the IGF-I gene could be considered as a candidate gene of certain carcass and meat quality traits in sheep; however, further investigations are recommended in other sheep breeds in the world in order to evaluate potential breed-and/or flock-specific effects of this gene's polymorphism on sheep production traits.Furthermore, the IGF-I gene is well known for its role in many processes in an organism, which is a result of its pleiotropic function; therefore, before using IGF-I polymorphism in a production system, association analyses regarding other traits, which were not investigated in this study, i.e. reproductive and milk production attributes, should be carried out to confirm that selection will not have negative consequences for the breed.
PCR-RFLP analysis of polymorphism in the 5 flanking region of the IGF-I gene in Coloured Polish Merino sheep revealed two alleles, named A and B, and only two genotypes: AA and AB.The HaeII digested allele A amplicon, while allele B remained undigested.Sequence analysis showed that two SNPs differed allele A from allele B. The first polymorphism covered G/C transversion, while the second covered G/A transition, at positions 85 and 87, respectively (nucleotide positions are relative to the first nucleotide in the sequence; GenBank no.LC151296.1).
e ), which amounted to 0.153.The population was in the Hardy-Weinberg equilibrium (P = 0.247).3.3Effect of the IGF-I gene polymorphism on traits measured in live animalsMixed-model association analyses of the IGF-I genotypes were performed for body weights, average daily gains, body measurements and ultrasound LD muscle measurements.Effects of the IGF-I gene polymorphism on these traits in Coloured Polish Merino sheep are presented in Tables 2-

Table 2 .
Results of association analysis between IGF-I genotypes and growth traits -LSM ± SE (standard error) -in Coloured Polish Merino sheep.
* LSM is least squares mean.

Table 3 .
Results of association analysis between IGF-I genotypes and body size traits (LSM ± SE) in Coloured Polish Merino sheep.
Niżnikowski et al. (2015)an.theTallele (5 %) in Baluchi sheep and they did not find TT homozygotes.Niżnikowski et al. (2015)investigated polymorphism in the third exon of the IGF-I gene in different sheep breeds in Poland and they detected the T allele only in Charollais ewes.To summarise, allele and genotype frequencies in the IGF-I locus vary between different sheep breeds; therefore, studies covering other sheep breeds should be continued.

Table 4 .
Results of association analysis between IGF-I genotypes and live ultrasound measurements of the LD muscle (LSM ± SE) in Coloured Polish Merino sheep.
* LSM is least squares mean.

Table 5 .
Results of association analysis between IGF-I genotypes and carcass traits (LSM ± SE) in Coloured Polish Merino sheep.LSM is least squares mean.Values with different superscripts a,b within a row are significantly different (P < 0.05).Values in bold were applied to highlight highly significant effect of the IGF-I gene on carcass traits. *

Table 6 .
Results of association analysis between IGF-I genotypes and meat quality traits (LSM ± SE) in Coloured Polish Merino sheep.LSM is least squares mean.Values with different superscripts a,b within a row are significantly different (P < 0.05).Values in bold were applied to highlight significant effects of the IGF-I gene on meat quality traits. *

Table 7 .
Results of association analysis between IGF-I genotypes and subjective assessment of live animal and carcass traits in Coloured Polish Merino sheep (n = 106).