Genetic relationship among body condition score, type, fertility and production traits in Swiss Holstein cattle.

Abstract

AAABG Vol 15 GENETIC RELATIONSHIP AMONG BODY CONDITION SCORE, TYPE, FERTILITY AND PRODUCTION TRAITS IN SWISS HOLSTEIN CATTLE Haja N. Kadarmideen1 and Silvia Wegmann2 Statistical Animal Genetics Group, Institute of Animal Science, Swiss Federal Institute of Technology, ETH Zentrum, Zurich CH 8092, Switzerland 2 Holstein Association of Switzerland, Grangeneuve, 1725 Posieux, Switzerland 1 SUMMARY Genetic parameters for body condition score (BCS) , 27 linear type, 5 milk production and 2 fertility traits were estimated for Swiss Holstein cattle. Data set consisted of 25126 records and 80329 animals in pedigree. Heritabilities (h2), permanent environmental variances (c2) and genetic correlations (rg) were estimated via repeatability animal models. Estimates of h2 and c2 for BCS were 0.23 and 0.21, respectively. Estimated h2 ranged from 0.09 to 0.50 for type traits and 0.21 to 0.57 for production traits. The range of estimated rg of BCS with type traits was -0.69 to 0.58, with production traits was -0.27 to 0.17, and with fert ility traits was 0.002 to 0.289. Results showed that cows with lower body condition scores have genetically poor fertility . Type and production traits are favourably and unfavourably related to BCS, respectively. Based on the results from this study it could be concluded that BCS could be used as a potential indicator of functional and fertility traits. Key words: Body condition score, type and fertility traits, genetic analysis, dairy cattle INTRODUCTION Functional traits (e.g. health or fertility) in breeding goals are increasingly becoming an integral part of livestock breeding strategy and have been shown to maximize profit , by reducing costs and improving efficiency of production (Kadarmideen et al. 2002). Recent studies have shown that Body Condition Score (BCS) can be used in selecting robust and profitable animals, due to its strong genetic relationship with other functional traits such as body weight and feed efficiency (Coffey et al. 2001), type traits (e.g. Veerkamp et al. 1997), energy balance or metabolic stability (Coffey et al. 2001) and fertility (Pryce et al. 2000). BCS is routinely recorded from 2001 by Holstein Switzerland. T he main objective of this work was to estimate heritability for BCS and its genetic and other correlations with 27 type, 5 production and 2 fertility traits. MATERIALS AND METHODS Data. BCS is recorded on 1-5 scale (1=very thin; 5=very fat) with an increment of 0.25. Heifers are assessed once during the lactation. For animals with a BCS record, data on all 27 linear type (and composite) traits and 5 milk production records were also obtained. Also, Estimated Breeding Values (EBV) of sires for daughter's non-return rates 56 day post insemination (NRR) and days to first service (DFS) were obtained from the Holstein Switzerland. These estimates of sire breeding values for daughters NRR and DFS are based on the individual animal model (AM) as reported by Schnyder and Stricker (2002). There were 25126 records (5483 herd-year-season, 7516 herd-year-visits) and were 5 lactation classes . Pedigrees were traced as far back as possible which included 80329 animals. 77 Cattle Reproduction Statistical models and analyses. Estimation of genetic and environmental parameters was accomplished by defining three types of models / analysis as given below: Univariate repeatability animal models. To estimate heritability and permanent environmental variance, the f ollowing repeatability animal models [1] and [2] were used for type + BCS and production traits, respectively. 5 .......[1] = � + HYV + L + S + M + at + hp + a + w + e y ijklmnopq i j k l Where: y. = BCS or type traits; � = the overall mean; HYV = herd-year-visit of classifier; L= lactation number, S= stage of lactation (in months from calving date) at the time of classification; M = month of calving; atm = age (in days) at condition scoring or type classification nested within mth lactation and m is the regression coefficient for atm for m=1 to 5; hp = percentage of Holstein genes and n is the regression coefficient for hp; ao = random genetic effect of animal; wp = random permanent environmental effect of animal; and e . = residual error term. 5 ..................................[2] = � + HYS +L + ac + a + w + e y ijklmn i j m =1 m m. n o p ijklmnopq k =1 k k. l m ijklmn Where: y. = milk production traits; �= the overall mean; HYSi = herd-year-season of calving; ack = age (in days) at calving nested within k th lactation and k is the regr ession coefficient for ack for k =1 to 5 and all other terms are as in Model [1]. Bivariate repeatability animal models. T wo-trait genetic models were used to estimate variances, covariances and correlations for genetic, permanent environmental and residual effects specified under the univariate models [1] and [2]. T erms in model [1] and [2] were used jointly in the bivariate model but applying only corresponding model terms for each trait. Genetic regression models. The statistical model used for all type and BCS traits and was y ijklmnopq = � + HYV i + L j + S k + M l + at + hp + EBV m =1 m m. n o 5 NRR + p EBV DFS +e .[3] ijklmnopq Genetic regressions for all milk production traits were the same and was y ijklmn = � + HYS i + L j + k ac k =1 5 k. + l EBV NRR + m EBV DFS +e ijklmn ......[4] Where: EBVNRR is the estimated breeding value of the sire for daughters' non-return rate 56 day p ost insemination and �o or �l is the corresponding regression coefficient. Similarly, EBV is the DFS estimated breeding value of the sire for daughters' interval (in days) between calving and first service and �p or �m is the corresponding regression coefficient. All other terms are as in models (1 and 2). Implementation and Software: All p arameters for models [1] to [4] were estimated using the software package, ASREML (Gilmour et al. 2001). It was not our aim to compute correlations among type traits or between type and milk production traits, as theywere already estimated for Swiss Holsteins. 78 AAABG Vol 15 RESULTS AND DISCUSSION Heritabilities and permanent environmental variance s. Heritability (h2) and permanent environmental variance (c2 ) from univariate repeat ability AM are given in Table 1. Heritabilities and c2 for BCS was 0.23 and 0.21, respectively. Among type traits, heel depth had the lowest h2 (0.09) and rump width had the highest h2 (0.50). Udder traits had h2 of 0.20 to 0.3 1, `feet &leg' traits had h2 of 0.15 to 0.18 and `rump' traits were highly heritable with a range of 0.25 to 0.50. The c2 estimates were generally higher than h2 estimates and were the highest for stature (0.87) and lowest for rump width (0.26). The h2 estimates were significant for all 27 type traits and BCS, with their standard error being small (= 0.03). Similarly, the c2 estimates were significant for all 27 type traits and BCS, with their standard error being small (= 0.04). For milk production traits, estimates of h2 were the highest for fat percentage (0.57) and lowest for fat yield (0.21). Estimates of h2 and c2 production were generally lower than those for type traits. Estimates of h2 for BCS and type trait are similar to literature estimates (e.g. Veerkamp et al. 1997, Berry et al. 2002). As for permanent environmental variances, literature estimates based on multiple lactation records for BCS are scarce. 2 Table 1: Phenotypic means (Mean), standard deviations (S.d.), h and c2 with their standard 2 2 1 errors (s.e (h ), s.e.(c )), genetic and phenotypic correlations with their standard errors, (rg (s.e.), rp (s.e.)) for body condition score (BCS)2, type 3 and production traits, based on 25126 records. 79 Cattle Reproduction Trait Mean S.d BCS 2.7 Stature 145.5 Heart girth 197.1 Strength 5.3 Body depth 6.2 Loin 5.9 Rump angle 4.6 Rump width 6.2 Dairy char 6.0 Rear leg side view 5.7 Pastern 4.5 Heel depth 5.2 Rear leg rear view 5.3 Fore udder attach 5.7 Rear udder height 5.2 Rear udder width 5.7 Udder cleft 5.9 Udder depth 5.3 Udder quality 5.9 T eat length 4.8 T eatposition front 4.8 T eat position rear 6.4 Capacity 80.4 Rump 81.8 Dairy 82.0 Feet & legs 81.2 Udder 80.7 Final class 80.9 Milk yield 7299 Fat yield 285 Protein yield 262 Fat percent 3.93 Protein percent 3.19 1 Permanent environmental shown; 2 BCS was on 1 to 5 scale; 3All type traits were on 1 to 9 scale except stature, heart girth and composite traits; h2 s.e. 0.4 0.23 0.03 5.9 0.09 0.01 8.8 0.28 0.03 1.2 0.26 0.03 1.0 0.33 0.03 1.0 0.28 0.03 0.9 0.32 0.03 1.2 0.50 0.03 0.9 0.21 0.02 0.7 0.18 0.02 0.8 0.18 0.02 0.8 0.09 0.02 1.1 0.17 0.02 1.2 0.20 0.02 1.3 0.28 0.03 1.0 0.25 0.03 1.0 0.20 0.02 1.3 0.31 0.03 1.0 0.29 0.03 0.9 0.37 0.03 1.0 0.28 0.03 0.9 0.24 0.03 5.9 0.41 0.03 4.6 0.24 0.03 3.5 0.28 0.03 3.5 0.14 0.02 3.5 0.17 0.02 3.3 0.30 0.03 1658 0.26 0.02 66 0.21 0.03 69 0.27 0.01 0.42 0.57 0.03 0.21 0.47 0.03 and residual correlations h2 c2 c2 rg (s.e.) of rp (s.e.) of s.e. BCS with BCS with 0.21 0.04 0.87 0.01 0.01 (0.09) 0.02 (0.01) 0.54 0.03 0.52 (0.06) 0.28 (0.01) 0.42 0.03 0.58 (0.06) 0.34 (0.01) 0.32 0.03 -0.05 (0.08) 0.04 (0.01) 0.37 0.03 -0.63 (0.05) -0.29 (0.01) 0.45 0.03 0.08 (0.07) 0.03 (0.01) 0.26 0.03 -0.15 (0.06) -0.06 (0.01) 0.41 0.03 -0.44 (0.07) -0.20 (0.01) 0.40 0.03 -0.24(0.08) -0.09 (0.01) 0.37 0.03 0.10 (0.08) 0.02 (0.01) 0.28 0.03 0.38 (0.09) 0.10 (0.01) 0.38 0.03 0.11 (0.08) 0.09 (0.01) 0.48 0.03 0.02 (0.08) 0.08 (0.01) 0.44 0.03 -0.07 (0.07) -0.03 (0.01) 0.32 0.03 -0.09 (0.07) -0.01 (0.01) 0.49 0.03 -0.09 (0.08) -0.09 (0.01) 0.40 0.03 0.08 (0.07) 0.01 (0.01) 0.33 0.03 -0.48 (0.06) -0.19 (0.01) 0.37 0.03 -0.12 (0.07) -0.01 (0.01) 0.47 0.03 -0.16 (0.07) -0.04 (0.01) 0.38 0.03 -0.41 (0.06) -0.10 (0.01) 0.48 0.03 0.23 (0.06) 0.17 (0.01) 0.49 0.03 0.06 (0.08) 0.02 (0.01) 0.42 0.03 -0.69 (0.04) -0.31 (0.01) 0.54 0.03 0.17 (0.08) 0.07 (0.01) 0.67 0.02 0.03 (0.08) -0.01 (0.01) 0.54 0.03 0.13 (0.07) 0.07 (0.01) 0.33 0.03 0.17 (0.00) 0.31 (0.00) 0.31 0.02 -0.27 (0.04) -0.08 (0.00) 0.29 0.03 -0.19 (0.03) -0.03 (0.00) 0.20 0.03 0.19 (0.06) 0.05 (0.01) 0.18 0.03 0.17 (0.05) 0.04 (0.01) between BCS, type and production traits are not Correlations of body condition score with type and production traits. Genetic correlations: Estimates of genetic correlations (rg) of BCS with type and production traits are given in Table 1, along with phenotypic correlations, rp. BCS had lower positive estimates of r with stature, pastern, g 80 AAABG Vol 15 right leg rear view, fore udder attachment, rump, rump angle, udder, udder depth, final class, feet & leg. Moderate to high positive rg were found with heart girth, strength, heel depth and capacity. This indicates that cows with good BCS tend to have good heart girth, capacity etc., at the genetic level. BCS had lower negative estimates of rg with body depth, rump width, rear udder height, rear udder width, udder cleft, teat length, teat position front and moderate to high negative estimates of r with g dairy character, loin, udder quality, teat position rear, right leg side view and dairy. Genetic correlations with milk production traits were such that higher BCS increase milk yield but decrease fat and protein yield. Fat and protein percent also showed positive moderate genetic correlations. Generally, estimates of rg between BCS and type traits were favourable suggesting that selection for good BCS would increase a chance of having desired type. Phenotypic correlations Estimates of phenotypic correlations (rp) are also given in Table 1. In general, estimates of r were lower than p estimates of rg.. Absolute estimates of rp with type t raits ranged from 0.00 for teat length to 0.34 for strength. Estimated correlations agree with earlier estimates (e.g. Veerkampet al. 1997). Genetic relationship of body condition score with fertility. The estimated regression coefficients � for regressing BCS on the EBVNRR was 0.002 (s.e.=0.001). Similarly, � for regressing BCS on the EBVDFS was -0.004 (s.e. 0.001). Estimate of � indicates that for each unit increase in EBVNRR, BCS increas ed by 0.002 and that for each unit increase in EBVDFS, BCS decreased by 0.004. With �'s, approximate estimates of r could be derived using genetic standard deviations of traits involved. g Genetic regression methods for approximate rg were also used by others studies (e.g. Kadarmideen and Pryce, 2001). Letting genetic standard deviations of BCS, DFS and NRR as, s g_BCS, s g_DFS and s g_NRR, respectively , t he rg between BCS and DFS was rg_BCS,DFS = �BCS, DFS (s g _DFS / s g_BCS)= - 0.269. Similarly, rg between BCS and NRR was 0.002. Estimates for s g_DFS and s g_NRR for Swiss Holsteins were taken from Schnyder and Stricker (2002). For all type and production traits, �'s for NRR and DFS were available but are not reported here. Both � and rg for DFS show that cows with good body condition have genetically shorter time to recommence cycling after calving for first inseminat ion. This result could also be extrapolated to the genetic relationship of BCS with calving interval since the lat t er has very strong genetic relationship with DFS (rg of 0.90; Kadarmideen et al. 2000). In fact, Pryce et al. (2000) have reported rg of -0.41 between BCS and calving interval. In general, genetic relationships of BCS with other traits show that the cows with lower BCS tend to have poor fertility , BCS is favourably related to type and unfavourably to milk production traits. REFERENCES Berry, D.P., Buckley , F., Dillion, P., Evans, R.D., Rath, M. and Veerkamp, R.F. (2002) J.Dairy.Sci. 85: 2030. Coffey, M.P., Emmans, G.C. and Brotherstone, S. (2001) Animal Science 73:29. Gilmour, A.R., Cullis, B.R., Welham, S.J. and Thompson, R. (2001). ASREML Manual, Sept, 2001. Kadarmideen, H. N., Thompson, R. and Simm, G. (2000) Animal Science 71: 411. Kadarmideen, H. N., and Pryce, J.E. (2001) Animal Science 73: 19. Kadarmideen, H. N. and Simm, G. (2002) Proc. 7th WCGAL P. 29:119. Pryce, J.P., Coffey, M.P. and Brotherstone, S. (2000) J. Dairy Sci. 83: 2664. Schnyder,U. and Stricker, C. (2001 ) INTERBULL Bulletin.29:138. Veerkamp, R.F. and Brotherstone, S. (1997) Animal Science 64: 385. 81