Analysis of the sheep genome



The identification of genes controlling several traits of interest in sheep has been accomplished by positional candidate cloning. In these studies, the trait is first mapped to a specific chromosomal region by linkage analysis, which requires families that are segregating for the trait and for polymorphic markers. Microsatellite markers are usually used for these analyses because of their extensive genetic variability. Once the location of a trait is determined by linkage to the markers, possible candidate genes controlling the trait can be inferred because of their proximity to linked markers. It is not necessary to map all possible genes in sheep for this strategy to be effective. Rather, a subset of genes that are mapped in humans and mice have also been mapped in sheep; these genes serve as “anchors” across the comparative maps of the different species. Further study of these positional candidates has revealed naturally occurring mutations that produce phenotypes that are unique to sheep. Thus the genetic analysis of sheep traits advances knowledge not only in this species but provides critical information for understanding biological pathways in mammalian species.

  • ovine
  • comparative gene mapping
  • positional candidate cloning

The Sheep Genome Map

The diploid number in domestic sheep (Ovis aries) is 54, with 26 pairs of autosomes and two sex chromosomes. Ovine autosomes are telocentric with the exception of three pairs of large metacentric chromosomes (1, 2, and 3). The X is the largest acrocentric chromosome, and the Y is a very small metacentric chromosome, which usually looks like a small, square dot. The X chromosome can be recognized in unbanded preparations because of the presence of small short arms, but the telocentric autosomes cannot be distinguished without differential staining techniques (7). In 1995, a standard G-band karyotype for the sheep was agreed upon at the 9th North American Colloquium on Domestic Animal Cytogenetics and Gene Mapping (7). The ovine and bovine standards were correlated based on similar band patterns and mapped gene loci. G-, Q-, and R-banded karyotypes and idiograms for sheep chromosomes have also been presented (1). This report included chromosome-specific molecular markers that had been mapped by in situ hybridization to 24 of the 26 ovine autosomes.

The three metacentric chromosomes in sheep, OOV1, OOV2, and OOV3, are equivalent to centromerically fused bovine chromosomes 1 and 3 (BTA1 and BTA3), BTA2 and BTA9, and BTA5 and BTA11, respectively. These fusions at the centromere are supported by marker order on the sheep and bovine linkage maps (22, 24). Sheep chromosome 9 (OOV9) appears to be the product of a translocation of the telomeric end of BTA9 to BTA14 (22). The remainder of BTA9 corresponds to OOV8.

Development of the linkage map for sheep has been critical for genomic studies in this species. Prior to 1994, only 17 markers were assigned to 7 syntenic groups (7). In 1994, 19 linkage groups containing 52 markers including microsatellites and candidate gene restriction fragment length polymorphisms (RFLPs) were identified (21). These assignments were a consequence of a genome scan initiated to map the Booroola fecundity gene (64) using 12 pedigrees segregating for the Booroola gene. A more extensive ovine linkage map was published by this group in 1995 (22) and contained 246 markers (86 ovine microsatellites, 126 bovine microsatellites, 1 deer microsatellite, and 33 known genes), with marker spacing between 10 and 30 cM across the chromosomes. Total coverage of the map was 2,070 cM (∼75% of the genome), and markers were assigned to all 26 sheep autosomes. This map was constructed using the AgResearch International Mapping Flock (IMF). Three years later, a second-generation ovine genetic map was published by the United States Department of Agriculture (USDA) Agricultural Research Service (ARS) group in Nebraska (24). The map contained 519 markers (402 bovine microsatellites, 101 ovine microsatellites, and 16 known genes) and spanned 3,063 cM across the autosomes, with an average marker spacing of 6.5 cM ( A third generation map has been recently developed (58). This map contains 1,062 loci (941 anonymous loci and 121 gene) and is a compilation of genotype data generated by 15 laboratories using the IMF population (∼jillm/jill.htm). The map spans 3,400 cM (sex-averaged) for the autosomes and 132 cM (female) on the X chromosome.

It should be noted that these ovine linkage maps contain relatively few expressed genes (∼120) because of the difficulty in identifying allelic variation needed for linkage analyses. Additional assignments for ∼250 genes have been made using somatic cell hybrids (8, 82) and in situ hybridization.

There are some estimates of gene order for sheep and cattle, two ruminant species within Bovidae. In one comparison (24), marker order was inverted eight times (>2-cM intervals) when the sheep map was compared with the cattle linkage map (48). Length of the inversions ranged from 3 to 57 cM. This group also found six microsatellites that mapped to sheep linkage groups other than what would be expected based on the bovine homolog. The authors suggest that these may be the result of amplification of a nonhomologous sequence by heterologous primers.

Conserved chromosomal segments between sheep and humans have been determined using human chromosome painting probes on R-banded sheep chromosomes (44) and through the use of human and sheep chromosome painting probes on the Indian muntjac deer (9). A total of 48 human segments were found in sheep chromosomes.

An informational database that includes mapped loci in sheep is now available through the Roslin Institute (United Kingdom) at or Texas A&M University (USA) at The current version of SheepBase contains about 1,492 loci gathered from 507 publications. There are about 369 designated genes, 1,098 type II loci, and 25 other loci (such as blood group polymorphisms) in the database. Of the 2,257 map assignments, 1,413 were based on linkage studies, and 844 were assigned using cytogenetic methods.

Identification of QTLs for Wool, Meat, and Milk in Sheep

Genome scanning projects have been initiated for identification of genes controlling many quantitative traits in sheep. Preliminary reports of the identification of sheep quantitative trait loci (QTLs) include parasite resistance (4, 20), wool production traits (47, 73), milk traits (27), and fecal soiling of the wool or dagginess (56). Development of additional resource flocks is ongoing for traits in sheep such as parasite resistance, out-of-season breeding, carcass traits, and wool production.

Parasite resistance.

Genetic variation between breeds of sheep in levels of resistance to internal parasites has been well documented (5). Thus the development of genetic markers for parasite resistance has received major attention in the last 5 years. Resource families suitable for the identification of genetic markers for resistance to Trichostrongylus colubriformis have been established by Commonwealth Scientific and Industrial Research Organisation (CSIRO, Australia), AgResearch (New Zealand), and the University of Melbourne and for Haemonchus contortus by Louisiana State University, CSIRO, and International Livestock Research Institute (ILRI, Kenya). The initiation of whole genome scans has been reported by CSIRO (4) and AgResearch (20), but information on specific chromosomal regions that contain QTLs for parasite resistance are not yet available.


Wool fiber in sheep is a highly organized structure, with about 90% of the fiber made up of keratin intermediate filament (IF) and keratin-associated proteins (KAP). The keratin IF proteins form 8- to 10-nm diameter filaments that are embedded in a matrix of KAPs. The two types of keratin IF, type I and type II, are paired to form the basic unit of the filament. Wool keratin IF type I genes are 4–5 kb with 6 introns, whereas the type II genes are 7–9 kb with 8 introns (76). In sheep, the type I (KRT1.n) and type II (KRT2.n) genes map to 11q25-q29 and 3q14-q22, respectively (39). These physical map assignments have been confirmed through linkage analysis (63), with KRT1.2 mapping to OOV11 and KRT2.10 and KRT2.13 closely linked on OOV3. A 100-kb DNA fragment containing the keratin type II region of sheep has been characterized (77, 78). Three wool and three wool-related type II genes are found within the region in a tight cluster, which is a common feature of the keratin genes.

The KAPs have been grouped (76) into high-glycine-tyrosine KAPs (encoded for by the KAP6.n multigene family, KAP7, and KAP8), the high-sulfur KAPs (encoded for by the KAP1.n, KAP2.n, and KAP3.n multigene families), and the ultra-high-sulfur KAPs (encoded for by the KAP4.n and KAP5.n multigene families). The KAP genes are small, between 0.6 and 1.5 kb in size, and do not contain introns. The AgResearch IMF families have been used in a linkage analysis to map KAP6.1, KAP7, and KAP8 to OOV1; KAP1.1, KAP1.2, KAP1.3, and KAP3.2 to OOV11; and KAP5.1 to OOV21 (63). KRT1.2 was also linked to the KAP genes on OOV11. Another wool follicle gene, trichohyalin (THH), was assigned to OOV1 as well.

Studies of the expression and regulation of hair and wool keratin genes are ongoing. Several consensus sequences have been identified within the promoters of these genes, including binding sites for AP1 and AP2, two common transcription factor complexes, and for lymphoid enhancer factor 1 (LEF1) (76).

Many phenotypic traits associated with the color and quality of wool have been reported, such as albino, lustrous wool, silky wool, and fibre loss (reviewed in Ref. 89). The genetics of these phenotypes are well characterized, but the actual genes that are responsible are not yet known. It is likely that there is involvement of the keratin IF and KAP genes. Interestingly, two wool mutations, HH1 and HH2, affect not only the amount of medullated hair found within the fleece but also have a pleiotropic effect on the formation of horns. While horns and wool are visually different, both contain substantial amounts of keratin.

Limited reports of genes associated with wool production traits are available. Linkage between KAP6 and wool fiber diameters has been found (73). In addition, a resource population for wool production QTLs has been established (47) using a Merino × Romney backcross. While successful identification of QTLS for yield, standard deviation of follicle diameter in a staple, bulk, clean fleece weight, and greasy fleece weight was reported by these authors, no specific chromosomal regions were indicated.

Fecal soiling.

Dagginess is a trait unique to sheep in which there is excessive fecal soiling of the breech area. This soiling occurs because of scouring and too much wool in the anal region. There is an economic incentive to reduce the incidence or severity of dags, because soiled wool must be discarded or discounted and because fecal matter may contaminate the carcass at slaughter. Preliminary evidence for a QTL for dagginess has been reported (56).

Several QTL projects in sheep are now underway. However, production of suitable resource families for QTL searches has required a minimum of 3–5 years. Thus only preliminary results are available for many of these projects. Full reports of map assignments for these QTLs are forthcoming, but because of the paucity of sheep flocks suitable for QTL studies, only a few of these putative QTLs will be replicated in other studies. The identification of genes associated with ovine QTL will likely be dependent on the study of candidate genes revealed in other species.

Identification of Single Gene Traits in Sheep

Chromosomal assignment of many single-gene traits in sheep are known, including the Booroola fecundity gene (OOV6; 64, 65), the Inverdale fecundity gene (X; 23), callipyge muscle hypertrophy (OOV18; 15), Carwell muscle hypertrophy (OOV18; 61, 69), Belgium Texel double muscling (OOV2; 59), spider lamb syndrome (OOV6; 18), and horns (OOV10; 66).

Booroola fecundity.

The Booroola fecundity mutation, FecBB, is codominant for ovulation rate and partially dominant for litter size, with one copy of FecBB increasing litter size by approximately one to two extra lambs (75). In 1993, FecB was linked to markers from a region of human chromosome 4q (64). Subsequent mapping experiments assigned the locus to a position on ovine chromosome 6 (65), corresponding to human chromosome 4q21–25 (67, 96). Upon investigation of positional candidates from 4q21–25, these two groups independently identified a nonconservative mutation (Q249R) in the intracellular signaling domain of bone morphogenetic protein IB receptor (BMPR-IB), a member of the transforming growth factor-β (TGF-β) receptor family (67, 96). This substitution was in total linkage disequilibrium with FecB.

Growth differentiation factor-5 (GDF5) and BMP4, two natural ligands of BMPR-IB, normally inhibit the secretion of progesterone by sheep granulosa cells, which is necessary for differentiation of granulosa cells and maturation of ovulatory follicles. Granulosa cells from FecBB/FecBB females secreted significantly more progesterone than cells of FecB+/FecB+ females when cultured in vitro (67). These findings suggest that FecBB/FecBB cells are less responsive to the inhibitory effects of GDF5 and BMP4 because of altered functionality of the BMPR-IB receptor, leading to advanced differentiation of the granulose cells and enhanced follicular development in animals carrying the mutation.

Inverdale fecundity.

The Inverdale fecundity gene, FecX, was identified in a prolific New Zealand sheep family descended from a Romney female that produced 33 lambs in 11 lambings (23). Progeny testing suggested that a major gene for prolificacy was located on the X chromosome. Females that are carriers for the Inverdale allele, FecXI, have an increase in ovulation by about one extra egg and give birth to ∼0.6 extra lambs. Females that are homozygous for the gene are infertile and have small nonfunctional “streak” ovaries that are about one-eighth the size of normal ovaries. Males with the Inverdale allele have normal fertility.

To more precisely map the location of the Inverdale locus, 14 microsatellite markers and 7 genes were incorporated into a linkage map of the ovine X chromosome (34), with a recombination length of 141.9 cM. The Inverdale gene was then localized to a 10-cM region on this chromosome (33), corresponding to human chromosome Xp11.2–11.4. A member of the TGF-β superfamily, bone morphogenetic protein 15 (BMP15), also known as GDF9B, is located in this region. The BMP15 protein is specifically expressed in oocytes, but its function is not yet known. Mutations in this gene are in linkage disequilibrium with FecXI, as well as another X-linked hyperprolific trait, Hanna or FecXH (35). The FecXI mutation is associated with a V31D substitution, resulting in a nonconservative change in a highly conserved region of the protein. A single C-to-T transition at nucleotide position 67 of the mature peptide introduces a premature stop codon in the FecXH allele. Both changes would be expected to produce a biologically inactive form of BMP15. It should be noted that these two X-linked fecundity traits as well as Booroola prolificacy are due to mutations in genes from the TGF-β pathway.


The callipyge gene is a mutation in sheep responsible for pronounced muscle hypertrophy of the fast twitch muscle fibers (Fig. 1), primarily in muscles of the pelvic limb (Fig. 2; 10, 50). Interestingly, the hypertrophy is absent at birth and develops only after ∼3 wk of age. Callipyge animals produce leaner, higher yielding carcasses (46, 50), but there is some concern with decreased tenderness of the loin (49, 50). Several postmortem tenderization methods are effective in improving the tenderness of callipyge lamb meat (17).

Fig. 1.

Micrographs of longissimus muscle of normal (A) and callipyge (B) lambs stained for ATPase and NADH tetrazolium reductase activity. Cellular hypertrophy is evident in the callipyge muscle, especially in the fast-twitch, glycolytic fibers, which appear clear to lightly stippled.

Fig. 2.

Conformation of a normal-appearing homozygous callipyge ram (far left) and four callipyge offspring.

Genetic characterization of the locus has demonstrated a unique mode of inheritance termed “polar overdominance” (16, 32), in which only heterozygous offspring inheriting the mutation from their sire express the phenotype. The three other genotypes are normal in appearance. Progeny data indicate that reactivation of the maternal callipyge allele occurs after passage through the male germ line, although this reactivation is not absolute (16).

Our earlier work (15) localized callipyge to the distal end of ovine chromosome 18, based on linkage to markers TGLA122, CSSM18, and GMBT16. The position of callipyge was then more precisely mapped to a 4.6-cM interval between IDVGA30 and OY3 by breakpoint mapping using lambs produced in crosses of the four callipyge genotypes and assuming a polar overdominance model (87). Isolation of eight new microsatellites from bovine (87) and ovine (86) contigs allowed us to more precisely localize callipyge within a 450-kb chromosome segment (6).

To characterize this region, two ovine bacterial artificial chromosome (BAC) clones and a long-range PCR product connecting these two BACs were completely sequenced. The resulting sequence traces of 248,602 bp were aligned with the corresponding human genomic sequence obtained from GenBank (accession numbers AL132711, AL117190, and AL132709), with an overall similarity of 44.7%. Six genes were predicted to reside within the analyzed segment. Two of these genes, DLK1 and GTL2, have been previously reported in humans (83, 97), mice (84, 8, 88), and cattle (30, 31), whereas the remaining genes, DAT, PEG11, antiPEG11, and MEG8, are previously undescribed (12).

Using primers designed from each of these genes, Charlier et al. (12) performed RT-PCR on several tissues sampled from an 8-wk-old lamb, demonstrating that all six transcripts are preferentially expressed in skeletal muscle. Because the corresponding regions in humans and mice contain genetically imprinted loci, the imprinting status of these six genes was investigated using single nucleotide polymorphisms (SNPs). Genomic DNA of lambs contained both alleles of the SNP; however, because all six genes were imprinted, only one allele of each gene was detected in skeletal muscle mRNA. By tracing the alleles back to the lambs’ parents, maternal or paternal imprinting status of each gene was determined. These results demonstrated that DLK1, DAT, and PEG11 are paternally expressed, whereas GTL2, antiPEG11, and MEG8 are maternally expressed (12).

In a subsequent study (13), poly(A)+ RNA was isolated from the longissimus dorsi of 8-wk-old animals of the four callipyge genotypes. Northern blotting and/or RT-PCR were used to examine expression of DLK1, GTL2, PEG11, and MEG8 with G3PDH expression serving as a control for similar amounts of mRNA. These experiments demonstrated that the callipyge mutation does not alter the imprinting status of the four genes but dramatically enhances their expression level in cis (Fig. 3). The paternally expressed genes (DLK1 and PEG11) had increased expression in the C/C and CP/NM genotypes (where P and M superscripts refer to the paternal and maternal chromosomes, respectively), those genotypes that have the paternally expressed callipyge chromosome. For the maternally expressed genes (GTL2 and MEG8), increased expression occurred in the C/C and NP/CM genotypes, which have the callipyge mutation on their maternal chromosome. Thus the callipyge mutation appears to modify the activity of a common regulatory element, which could be either an enhancer (boosted by the callipyge mutation) or a silencer (inhibited by the callipyge mutation). This altered regulatory element may be responsible for the enhanced expression of DLK1, GTL2, PEG11, and MEG8, and possibly other untested genes as well. It is likely that increased expression of one or more of these genes initiates the manifestation of the callipyge phenotype. However, the actual callipyge mutation is not yet known.

Fig. 3.

A model for the callipyge locus.

Rib-eye muscle.

Characterization of another gene responsible for an increase in the rib-eye muscle of lambs has been reported (2). Effects of the rib-eye muscle (REM) locus are less dramatic than for callipyge, with an 11% increase in muscle mass limited to the longissimus dorsi. The REM locus appears to act as a dominant gene with no evidence of imprinting and has only minor effects on meat tenderness (62). Interestingly, the REM locus has been localized to the distal end of ovine chromosome 18 (61, 69). However, recent information (McLaren R, personal communication) indicates that the REM locus lies centromeric to callipyge, suggesting that these traits are not allelic to each other but rather may be controlled by two closely linked genes.

Muscle hypertrophy.

Texel sheep are characterized by generalized muscular hypertrophy (11). Based on the association between mutations in the myostatin gene and double muscling in cattle, a study was recently initiated to investigate the possible involvement of myostatin in the hypertrophy of Belgian Texel sheep (59, 60). In this study, the entire coding sequence of myostatin was determined for the Texel and normally muscled Romanov controls; however, no sequence differences were identified. Interestingly, analysis of microsatellite markers flanking myostatin on chromosome 2 in a (Texel × Romanov) × Romanov backcross pedigree revealed significant linkage (Marcq F, personal communication). Thus the causative mutation responsible for hypertrophy found in Texel sheep may reside outside the coding segments of the myostatin gene.

Spider lamb syndrome.

Ovine hereditary chondrodysplasia or “spider lamb syndrome” (SLS) is a recessive genetic disorder (90) causing skeletal deformities in newborn lambs. Structural abnormalities include abnormally long, spider-like legs, humped and twisted spines, deformed ribs and sternebra, facial deformities, lack of body fat and underdevelopment of muscle (Fig. 4). Multiple irregular islands of ossification are found in shoulders, elbows, and sternum of spider animals upon radiological evaluation (81, 93). It is believed this disorder arose as a mutation in a Suffolk genetic line that was used heavily during the late 1960s because of desirable production and show-ring characteristics. The syndrome has since surfaced in several sheep breeds within the last two decades, including North American Suffolks and Hampshires, and US Southdowns, Shropshires, and Oxfords. In addition, there are reported cases of SLS in New Zealand and Australia, after the importation of several US Suffolks into Australia in 1987 (74).

Fig. 4.

Conformation of a spider lamb (left) compared with a normal flock mate (right), both 1 mo of age. Note the abnormally long legs of the spider lamb.

A project to identify a genetic marker for SLS was initiated at Utah State University in 1991. Genetic mapping of the SLS locus was performed by linkage analysis between microsatellite loci in two large pedigrees produced from matings of two carrier rams to 65 carrier and normal females, who produced 223 normal and 80 spider lambs. After screening 174 markers, results from microsatellite marker OarJMP8, located on the distal end of ovine chromosome 6 (55), revealed significant linkage with the SLS trait (18). A combined LOD (logarithm of the likelihood ratio) of 2.85 at 28.1% recombination was obtained across both pedigrees. Three additional markers that had been previously mapped to this region were also linked to SLS. Final order of the loci in this region was determined as OarJMP8-McM214-OarJMP12-BL1038-SLS. A maximum LOD score of 12.89 was obtained between the SLS locus and BL1038 at 3% recombination.

A comparative mapping approach was initiated to identify candidate genes that are located on ovine chromosome 6 (OOV6). Prior mapping of structural genes on OOV6 indicated that it is orthologous to human chromosome 4 (HSA4; 53) and that genes mapping within HSA4p16.3 are distal to microsatellite markers OarJMP8 and OarJMP12 (Lord et al., Ref. 53). Contained within HSA4p16.3 is the gene for fibroblast growth factor receptor 3 (FGFR3; 91). Mutations in the FGFR3 gene of humans result in varying forms of dwarfism (94). Two groups have produced mice in which the FGFR3 gene is disrupted by homologous recombination (19, 25). Remarkably, mice homozygous for the FGFR3 knockout display skeletal abnormalities that are similar to SLS. From these knockout studies, the function of FGFR3 can be defined as a negative regulator of bone growth. To investigate the possible role of FGFR3 in SLS, a single-stranded confirmational polymorphism (SSCP) analysis of the gene was undertaken. An SSCP spanning from the 3′ end of exon 14 through the 5′ end of exon 18 was predictive of SLS genotype; linkage analysis of this polymorphism in a subset of the pedigrees resulted in a LOD score of 18.58 at 0% recombination. To further investigate FGFR3, genomic DNA from one homozygous normal animal, two carriers, and two spider lambs was amplified and sequenced. Analysis of the sequence revealed a T-to-A transversion in exon 17, resulting in a nonconservative amino acid substitution of valine to glutamine at position 700 within the second tyrosine kinase domain (3). Using a PCR-RFLP designed to detect this polymorphism, researchers have performed population studies including more than 2,000 sheep of differing SLS genotypes suggesting that this is the causative mutation in SLS. It is most likely that the mutation leads to loss of receptor function in homozygotes, resulting in poorly controlled chondrocyte differentiation. There is also a suggestion that heterozygous animals have increased bone growth, making them taller in stature than their homozygous normal counterparts. This might be an explanation for the high frequency of the SLS allele in show-ring animals (Beever JE, unpublished data). Further studies of in vitro receptor function are being conducted.


There are at least three genetic systems controlling development of horns in sheep. In some sheep strains, the halo hair (HH1) locus controls the sex-limited expression of horns and has marked pleiotropic effects on fleece characteristics (28), with the expression of horns associated with highly medullated hairlike fleece. Sheep carrying the HH1N allele produce hair in addition to wool on their backs as young lambs. A second genetic system for horns is expressed in Merino sheep and controlled by the Ho locus, with three alleles: HoP, Ho+, and Hoh1 (52). Polledness (absence of horns) is produced by the allele HoP, which is incompletely dominant in rams and almost completely dominant in females. Ho+ produces horns in both sexes, and Hoh1 produces sex-limited horns. There can also be modification of the growth of horns in intact males by the autosomal scurs locus (52).

In an experiment to map the ovine Ho locus (66), pedigrees were established by crossing horned Merino rams with polled New Zealand Romney females. The resulting F1 rams were then backcrossed to Merino females, and horn data were collected on 183 offspring. It was expected that markers on ovine chromosome 1 (OOV1) would be linked to the Ho locus because of the previous assignment of the Bos taurus polled locus to bovine chromosome 1 (BTA1; 36), which is the evolutionary homolog to OOV1. Although 14 markers from OOV1 were tested, no linkage was detected, and OOV1 was excluded. Instead, significant linkage was found with markers on OOV10, and the Ho locus was mapped to the telomere of this chromosome (66). The authors suggest that the Ho locus may correspond to the african horn locus in cattle (36), which has not been mapped, whereas the ovine HH1 locus may be orthologous to the cattle polled locus on BTA1.

The genetic inheritance of many other traits in sheep has been characterized (reviewed in Ref. 52). However, the chromosomal assignment and/or causative mutation is not yet known.

Genes for Economically Important Traits In Sheep Are Isolated Through Functional Cloning

Coat color.

The control of coat color in mammals is based on the ability of melanocyte cells to produce melanin pigments and move the melanin into hair, skin, and wool. Melanin can be either eumelanin (black or chocolate brown, mostly composed of tyrosine) or phaeomelanin (reddish-brown or yellowish tan, mostly composed of cysteine). Melanocytes can produce both eumelanin and phaeomelanin but usually only one pigment type at a time. Binding of α-melanocyte-stimulating hormone (α-MSH) to MSH receptor (MC1-R) on melanocyte cell surfaces initiates production of eumelanin. Absence of the α-MSH signal results in phaeomelanin production. Any number of possible alterations to the normal functioning of this complex system will result in hair, skin, or wool without pigmentation or a dilution of pigmentation. For instance, white color can be caused by several reasons such as the lack of melanocytes or decreased effectiveness of melanin production (89). Another example is the dominant dark coat color in which a mutation in MC1-R causes activation of this receptor even in the absence of α-MSH (54). Recently, the association of a mutation in MC1-R and the dominant black coat color in sheep has been reported (92). In this study, the complete coding region of the MC1-R gene was sequenced from one white and one black sheep of the Norwegian Dala breed. Sequence analysis revealed two mutations in the black animal, resulting in amino acid substitutions M73K and D121N. Design of a PCR-RFLP allowed detection of the mutations in a Norwegian Dala pedigree. All animals (9 black and 13 white) were descendants of a single black female. Both mutations showed complete linkage with black coat color. Black animals were heterozygous for the mutations, whereas the white sheep were homozygous wild type, thereby supporting the dominant inheritance of the black coat color. To further characterize these mutations, they were introduced into the mouse MC1-R gene, and activity of the expressed receptor was assessed using pharmacological assays (92). The M73K mutation was able to constitutively activate the receptor, whereas the D121N mutation did not. When the mutations were introduced together, activation of the receptor still occurred but was less intense. Taken in total, these experiments suggest that the black coat color in sheep is due to the M73K mutation in the MC1-R gene. The mechanism by which M73K activates the receptor is distinct from that in mice and the Alaska silver fox.


Transmissible spongiform encephalopathies (TSE) are characterized as transmissible neurodegenerative disease in mammals, with susceptibility to infection modulated by host-genetic interactions (72). A unique characteristic of TSE, whether sporadic, dominantly inherited, or acquired by infection, is the accumulation of an abnormal, protease-resistant prion protein (PrPSc) in the brain of TSE-infected animals (80). The normal cellular form of the PrP protein (PrPC) is encoded by a host autosomal gene; it has been suggested that the abnormal form (PrPSc) serves as a template for posttranslational misfolding of the normal protein (79).

The TSE of sheep is called scrapie (40). Typical clinical signs of scrapie in sheep begin with mildly impaired social behavior, followed by locomotor uncoordination or ataxia with trembling (26). Pruritis can result from the animal attempting to relieve what seems to be an intense itching by scratching against fence posts or by biting the affected area, and these clinical signs can last from 2 wk to 6 mo. Brain lesions develop late in the incubation period and include neuronal degeneration, vacuole formation, and astroglial cell proliferation.

It has been proposed (41, 43) that the genetic control of susceptibility to scrapie is either closely linked to, or identical to, the prion protein (PRPN) gene. The ovine PRPN gene contains three exons and is over 20 kb (95). The PRPN open-reading frame is contained in exon 3. Polymorphic variants in the coding region of the ovine PRPN gene have been associated with the incidence of both experimental and naturally occurring scrapie (14, 43). In many breeds, including the Cheviot, Swaledale, Ile de France, Shetland, Scottish Halfbred, and Bleu du Maine, the PRPN allele encoding valine at codon 136 (V136) is associated with an extremely high risk of scrapie, with most scrapie-positive sheep of these breeds, either naturally or experimentally infected, possessing the V136 codon (14, 38, 41, 42, 51, 57). However, the V136 variant is rare in British and US Suffolk (38, 41, 70, 95) and at a very low frequency in Japanese Suffolk (45). In contrast to other breeds, variation at codon 171 has been associated with scrapie susceptibility in Suffolk (95). Three amino acid variants at codon 171 [glutamine (Q171), arginine (R171), and histidine (H171)] have been identified (38, 95). In previous studies (70, 95), Suffolk sheep that were scrapie positive (n = 61) were all homozygous for glutamine at codon 171 (QQ171). Also, sheep homozygous for arginine (RR171) or heterozygous (QR171) did not develop scrapie from either natural or experimental challenge in several studies (38, 51, 70, 71, 95). These analyses suggest a relationship between the PRPN QQ171 genotype and scrapie susceptibility in Suffolk sheep. However, this association is not absolute, as 2 of 64 scrapie-positive animals were QR171 in one study (43) and a scrapie-positive Suffolk sheep with the RR171 genotype was found in Japan (45).

An extensive study of a scrapie epidemic in a closed flock of Romanov has been presented (29). Over a 4-year period, 1,015 animals were exposed to scrapie, resulting in 304 deaths. The most resistant PRPN allele was ARR (codons 136/154/171), with no ARR carriers developing scrapie during the epidemic. The VRQ allele was highly susceptible, with 76% of the scrapie-positive animals possessing at least one copy of this allele. Other alleles were ranked by the authors for scrapie resistance/susceptibility. The authors suggest that selection for ARR or AHQ carriers may be a method for increasing scrapie resistance within a flock. However, they note that these animals may still be susceptible to other scrapie strains, and it is possible that these healthy animals may transmit the scrapie agent.


For some ovine traits such as coat color, comparison to similar traits in other species has led directly to the appropriate gene in sheep. This candidate approach at the protein level has also allowed characterization of many other genetic traits in sheep, particularly disease disorders (68). In other cases, such as SLS, localization of the locus through a genome scan narrowed the search for a functional candidate to a specific chromosomal region. However, for traits unique to the sheep, such as Booroola fecundity, callipyge, REM, dagginess, and horns, the candidate gene approach has not been effective because the gene responsible for a similar trait in other species has not yet been identified or characterized. Researchers are seeking to identify these genes in sheep using positional cloning techniques, thereby leading to the discovery of genetic mechanisms not yet explored. Clearly, the analysis of traits unique to sheep has extended our understanding of the functional role and regulation of genes beyond what was known in mice or humans.


  • Article published online before print. See web site for date of publication (

    Address for reprint requests and other correspondence: N. E. Cockett, Utah State Univ., 4700 Old Main Hill, Logan, UT 84322-4700 (E-mail:


View Abstract