Materials and Methods

Sheep—All animal-handling procedures including blood collection and tissue processing were approved by the North Dakota State University Institutional Animal Care and Use Committee. Three years of lamb production data were collected from 2 locations representing commercial lamb producers in eastern and western North Dakota. Lambs (male, n = 380; female, 50) from flock 1 (North Dakota State University sheep flock, Fargo, ND) were primarily commercial western-whiteface Rambouillet and Katahdin influenced (n = 345), with the remaining lambs as follows: Rambouillet (24), Columbia (33), Suffolk (7), and Hampshire (21). Most lambs (male, n = 566) from flock 2 (Hettinger Research Extension Center, Hettinger, ND) were also commercial western-whiteface Rambouillet and Katahdin influenced (513), with the remainder Rambouillet (53). At both locations, lambs were fed a high-concentrate diet that contained 14% crude protein and met or exceeded other requirements¹⁶ for finishing lambs of a given age. All lambs were fed to achieve a common external fat cover and were slaughtered between 5 and 7 months of age.

PRNP genotyping—Jugular venous blood samples were collected in 1-mL evacuated tubes^a containing potassium EDTA to prevent coagulation. Genotyping was performed either by a commercial laboratory^b or by use of an SNP assay. For the SNP assay, DNA was extracted from whole blood by use of a commercially available column-based extraction kit^c or from muscle samples homogenized and processed with a DNA extraction solution^d and processed according to the manufacturer's recommendations. The purified genomic DNA was stored at 4°C prior to the SNP assay.

Genotyping was performed with a real-time PCR machine^e as described.¹⁷ Probes and primers for codon 171 were designed by use of primer design software,^e and the cDNA template sequence from bp 351 through 509 was sent to a commercially available service^e that uses special algorithms to design primers and probes for codon 136. Fluorescent probes and unlabeled primers for codon 171 were purchased^e separately, but for codon 136, the fluorescent probes and unlabeled primers were included in a 40× master mix.^e The sequences of probes and primers were disclosed with the order, but their exact concentration in the 40× master mix was not (Appendix).

Carcass evaluation—Live animal and carcass evaluations were determined subjectively by a professional evaluator who had > 30 years of research and teaching experience in meat animal and carcass evaluations. Lambs were sent to slaughter on the basis of results of visual appraisal of fat at the level of the 12th rib and carcass weight. Lambs were slaughtered at a commercial abattoir, and routine carcass data were collected approximately 24 hours after slaughter by an experienced carcass grader who had no knowledge of individual sheep genotype. Hot carcass weight was recorded at the time of slaughter. Longissimus dorsi muscle area, 12th rib fat, and body wall thickness were measured. Carcasses were assigned a limb score and conformation score, scaled from 10 (low choice) to 13 (low prime); a flank streaking score; and a quality grade.¹⁸ Yield grade¹⁸ and values for percentage boneless, closely trimmed retail cuts^f were calculated.

Final weight was calculated by use of the hot carcass weight and a standard dressing percentage of 53%.¹⁹ The initial weight, or weight upon beginning a feeding period, was used to calculate an ADG for the set feeding period. If initial weight was not available, a weaning ADG was calculated by use of the weight gain from weaning date to slaughter date.

Statistical analysis—Individual within-year data were analyzed by use of a statistical mixed model.²⁰,^g Fixed and covariate variables were selected with a stepwise selection procedure.²⁰,^g Variables were included in the model at the P ≤ 0.15 level.

For genotype analysis, the full model always included genotype or haplotype (AQ, VQ, or AR), breed, and year as fixed effects and days on feed as a covariate when they explained a significant portion of the variation. Means were separated by use of least significant difference (P ≤ 0.05).

Results

In the study population of 996 sheep, 43.2% (n = 430) were the RR₁₇₁ genotype, 43.3% (431) were QR₁₇₁, and 13.6% (135) were QQ₁₇₁. At codon 136, 3.9% had at least 1 valine allele, with most (n = 957) homozygous for alanine. As expected, the distribution of the PRNP alleles varied between flocks, depending on the founder and breeding stock genetics (Table 1).

Table 1—

Genotype distribution* defined at codons 136 and 171 of the PRNP gene in 2 flocks of sheep.

View Table

Within flock 1, few growth or carcass traits were associated with genotype at codon 136 (Table 2) or 171 (Table 3). Weaning ADG was greater (P = 0.04) in AA₁₃₆ sheep (n = 371), compared with AV₁₃₆ sheep (20; 0.26 ± 0.01 kg/d vs 0.23 ± 0.02 kg/d; respectively). Weaning ADG was also greater in QR₁₇₁ (145; P = 0.06) and RR₁₇₁ (211; P = 0.05) sheep, compared with QQ₁₇₁ sheep (34; 0.27 ± 0.01 kg/d for QR₁₇₁ and RR₁₇₁ vs 0.24 ± 0.01 kg/d for QQ₁₇₁).

Table 2—

Mean flock 1 production traits associated with codon 136 PRNP genotype.

View Table

Table 3—

Mean flock 1 production traits associated with codon 171 PRNP genotype.

View Table

Flock 2 had few genotype associations with production traits at codon 136 (Table 4) or 171 (Table 5). Average daily gain was greater (P = 0.02) in AV₁₃₆ sheep (n = 13) than AA₁₃₆ sheep (495; 0.33 ± 0.02 kg/d vs 0.29 ± 0.003 kg/d, respectively) and also greater (P = 0.02) in QR₁₇₁ sheep (230), compared with QQ₁₇₁ sheep (186; 0.30 ± 0.004 kg/d vs 0.28 ± 0.01 kg/d, respectively). Final slaughter weight was greater (P = 0.02) in AV₁₃₆ (n = 15) sheep than AA₁₃₆ sheep (518; 59.99 ± 2.31 kg vs 55.29 ± 1.54 kg). AV₁₃₆ sheep also had greater hot carcass weight (P = 0.05; 32.81 ± 1.08 kg vs 30.68 ± 0.24 kg; n = 15 and 518, respectively) and body wall thickness (P = 0.02; 2.53 ± 0.18 cm vs 2.13 ± 0.12 cm; 13 and 440, respectively) than AA₁₃₆ sheep. Rib eye area was significantly greater in QQ₁₇₁ sheep (n = 67) than QR₁₇₁ sheep (194; P = 0.05) and RR₁₇₁ sheep (154; P = 0.02; 16.96 ± 0.25 cm² vs 16.28 ± 0.15 cm² and 16.18 ± 0.16 cm², respectively).

For comparison of the combined codon 136 and 171 genotypes (VVQQ, AVQQ, AAQQ, AVQR, AAQR, and AARR), ADG was the only significant association found. In flock 1, AAQR (n = 66; P = 0.03) and AARR (75; P = 0.05) sheep had significantly greater weaning ADG, compared with AAQQ (16) sheep (0.24 ± 0.04 vs 0.27 ± 0.01 kg/d and 0.26 ± 0.01 kg/d, respectively).

In flock 2, ADG was significantly associated with PRNP genotype (P = 0.03). AVQR sheep (n = 10) had greater ADG than AAQQ (89; P = 0.01) and AAQR sheep (220; P = 0.08; 0.33 ± 0.02 kg/d vs 0.28 ± 0.01 kg/d and 0.30 ± 0.004 kg/d, respectively). AAQR sheep (n = 220) had greater ADG than AAQQ (89; P = 0.01) and AARR sheep (186; P = 0.06; 0.30 ± 0.004 kg/d vs 0.28 ± 0.01 kg/d and 0.28 ± 0.01 kg/d, respectively).

The effects of AQ, VQ, and AR haplotype on growth and carcass traits were also examined. In flock 1, sheep with the AR haplotype had greater ADG (P = 0.01) from weaning to slaughter than sheep with the AQ haplotype (0.27 ± 0.1 vs 0.24 ± 0.01 kg/d; n = 141 and 16, respectively). In flock 2, body wall thickness was greater (P = 0.12) in the AQ haplotype sheep than the AR haplotype sheep, although not significantly (2.26 ± 0.13 cm vs 2.14 ± 0.12 cm; n = 75 and 368, respectively). Rib eye area was significantly greater (P = 0.01) in the AQ haplo-type sheep than the AR haplotype sheep (16.96 ± 0.25 cm² vs 16.23 ± 0.11 cm²; n = 67 and 338, respectively). Average daily gain was greater (P = 0.02) for the AQ haplotype sheep than for the VQ haplotype sheep (0.28 ± 0.01 vs 0.33 ± 0.02 kg/d; n = 92 and 11, respectively).

Table 4—

Mean flock 2 production traits associated with codon 136 PRNP genotype.

View Table

Table 5—

Mean flock 2 production traits associated with codon 171 PRNP genotype.

View Table

Discussion

Parry²¹ suggests that there is an increase in scrapie susceptibility when producers select for larger, earliermaturing, faster-growing, more muscular sheep and an association between phenotypically superior sheep and scrapie susceptibility. The current USDA APHIS eradication program²² for the United States includes selection for the resistant ARR/ARR genotype after scrapie exposure or as a preventive measure. Therefore, any real or perceived economic advantage of selection for scrapie-susceptible sheep must be addressed for success and acceptance of the eradication program. In the present study, the evidence of an association between PRNP genotype and any of the measured production traits in this population of commercial lambs was weak. However, because of the low natural occurrence of the valine allele in the United States,²³ comparisons based on codon 136 were difficult. This skewed genotype distribution likely resulted in less accurate estimates. However, our data agree with that of others in revealing limited associations between genotype and production traits or only associations that may be attributable to chance.

A number of studies^12–15,24 have examined associations between PRNP genotype and economic or production traits and found few. In 2004, Brandsma¹² did find a weakly significant association in which selection for the ARR/ARR genotype gave a small positive effect on litter size and small negative effect on 135-day weight. However, these data were in direct contrast to data reported in 2005 by Brandsma.²⁴

Economically viable traits in livestock are largely quantitative in nature, suggesting little effect of a single gene polymorphism. However, in the present study population, differences were detected in ADG among the genotypes. In flock 1, weaning ADG was greater in those genotypes considered genetically resistant. In lambs in flock 2, there was an advantage in ADG in those sheep with a valine allele. However, 13 sheep with the valine allele were compared with 495 sheep homozygous for alanine, and thus the results were likely affected by the low frequency of the valine allele. When haplotypes were compared, sheep with the AR haplotype also had better ADG. In flock 2, sheep with the AR haplotype also had less body wall thickness, an indicator of lean body mass. Sheep with the VQ haplo-type, considered the most susceptible, had a lower ADG than sheep with the AQ haplotype.

In this study, effects of breed on genotype were minimized by use of breed as a covariate in the statistical analyses. However, each sheep breed has a different fraction of PRNP alleles in their population, and each has been selectively bred for different criteria. The small, but significant, differences observed between the 2 flocks in this study may correlate with the presence of an increased number of large-breed lambs (Columbia, Suffolk, Hampshire) in flock 1, compared with flock 2, which had no large-breed lambs.

Findings from this study suggest an advantage for ADG in lambs with an arginine allele at codon 171, but no other genotype effects on production traits. Thus, selection for the resistant arginine allele at codon 171, to comply with USDA scrapie eradication guidelines, should not be detrimental to lamb production in commercial flocks. However, any economic value of selection for an arginine allele at codon 171 will likely be based on acquiring resistance to scrapie rather than on improved ADG.