Rectovaginal Constriction in Jersey Cattle: . Genetics and Breed Dynamics1 H. W. LEIPOLD Department of Pathology Kansas State University Manhattan 66506
K. HUSTON Department of Dairy SCience Ohio Agricultural Research and Development Center The Ohio State University Wooster 44691 E. P. BARTON 2 American Jersey Cattle Club 2105 South Harrilton Road Columbus, OH 43227
J. R. HOLMAN Department of Statistics Ohio Agricultural Research and Development Center The Ohio State University Wooster 44691 D. L.TROYER Department of Anatomy and Physiology Kansas State University Manhattan 66506 ABSTRACT
males and 2,000 males annually are given for current conditions and after 500 generations of selection. (Key words: rectovaginal constriction, Jersey cattle, genetic model)
Data from fanner-owned herds and from experimental matings supported monofactorial recessive inheritance of rectovaginal constriction in US Jersey cattle. Kempthome' s population genetics model of a recessive trait involving only male selection was extended to include mutation and converted to selection of females only. Computer analyses with that model estimate slow decline in the frequency of the gene for rectovaginal constriction. Practical dynamics of the disorder in a breed registering 50,000 fe-
INTRODUCTION
Received October 2, 1989. Accepted March 14, 1990. IThis projoct was part of the Regional Project NC-2. CODtributioD number 360-1, Kansas Agricultural Experiment Station, and 10urnal Article Number 41-86, Ohio Agricultural Research and Development Center, 1be Ohio State University. 2Deceased . 1990 1 Dairy Sci 73:2516--2524
Rectovaginal constriction (RYC) is a congenital defect of Jersey cattle of both sexes. Its principal palpable and consequential characteristics are stenosis of the vestibular portion of the genital tract of females and of the anal sphincter of both sexes. Often it is accompanied by one or more of the constellation of defects that constitute the RYC syndrome including dystocia, udder edema, and excessive fibrous tissue in the anal sphincter muscle (1, 4, 5, 16, 21, 22). Comparisons of RYC homozygotes and heterozygotes with normal females revealed no differences in blood supply and innervation to the perineal viscera and musculature, in fibroblasts grown in vitro, in hydroxyproline content
2516
2517
GENETICS OF RECTOVAGINAL CONSlRICIlON
of tissues or urine, in duration of electromyographic action potentials, or in anal anatomical structures (anal canals). However, RVC homozygotes had unyielding connective tissue not only in the perineal region but also in the head region. Their external anal sphincter muscles also had more type I, m, and V collagen. Of possible significance in detecting heterozygotes are the discoveries that RVC heterozygotes, like RVC homozygotes, had greater amplitude in electromyographic action potentials than normals; they also had type n collagen, which normal cattle did not possess as reviewed recently (22). Rectovaginal constriction was fIrst reported to the Board of Directors, American Jersey Cattle Club (AlCC) in November 1972. Subsequently, in 1975, the AlCC listed RVC as the second abnormality in its Program of Action on Abnormalities based on information received since 1972 (7), of which some also was published in 1975 by Leipold and Saperstein (16), who suggested RVC was inherited as a simple recessive trait and may have been described in 1940 (13). In April 1980, the AlCC was sued by one of its members who sought $4.5 million for damage alleged to have resulted from the AlCC's actions (7). In October 1981, the court ruled in favor of the AlCC after hearing testimony from witnesses, some of whom were university scientists testifying as expert witnesses. That litigation revealed some evidentiary facets of the episode that may prove helpful to other scientists and that will be highlighted in the discussion (7, 9). In 1984, RVC was reported in Danish cattle sired by US bulls through AI (12, 15). Our primary purposes are to present analyses of data pertinent to the inheritance of RVC, a genetic model for it, and the dynamics of its occurrence in the Jersey breed. MATERIALS AND METHODS
Materials
Data were collected from: 1) Nevada herds identified by Leipold and Saperstein (16); 2) breeding trials in the Kansas State University experimental herd; and 3) field cases reported to the American Jersey Cattle Oub. During our 8-yr investigations of the disorder, we found that idiosyncracies of some persons performing examinations were a cause of
diagnostic confusion. Although there is variation in the size and elasticity of the rectum of RVC animals of the same age and relative size, variation among examiners is greater, primarily because of their lack of experience. When experienced examiners were permitted to examine both RVC and normal animals in a series of independent tests at Kansas State, they had little trouble in distinguishing RVC animals (17). Data from the Nevada herds were based on diagnoses made by the herd owner or his brother, both of whom had extensive experience in AI and pregnancy examination. Some were reconfirmed by us. Data from the Kansas State University herd were based on diagnoses made by the authors and confirmed by numerous other professionals. Data from field cases were included only when they conformed to a set of criteria designed to eliminate false RVC reports. Criteria included descriptions of the calving process, udder symptoms, and calf survival. Where possible, blood typing, veterinary report, and necropsy were secured. Methods
Segregation ratios in full-sib families were analyzed by the postproband method of Lush and Hazel (20). Segregation ratios in threequarter-sib families were analyzed by methods of Chase (8). In his appendix Table 3, Otase computed maximum likelihood estimators for segregation frequencies and their variances in sib families containing one and two or more affected individuals and ranging in size from 3 to 50 individuals. Complete ascertainment was assumed. These estimators correspond to means and variances of truncated binomial distributions, the use of which is illustrated for human families by Li (18). The genetic model is an extension of Kempthorne's model (14), which was based on selection in males rather than females and did not include mutation. The RVC gene frequency estimates by Allaire et al. (2) provided a base for computations with the newly derived model. Statistical methods are according to Fisher (11). RESULTS
Genetics
Nevada Herds. Three full-sib families, including at least one RVC female, were noted. Journal of Dairy
SciCIICC
Vol. 73.
No.9. 1990
2518
LEIPOLD ET AL.
TABLE I. Rectovaginal constriction (RVe) segregation ratios in tbree-quarter sib families in the Nevada herds•. Number RVe Animals Sibship
Expected
Size
Observed E(b) (8)
Group
Internal proband, true tbree-quarter sib family A 7 1 1.44 B 6 2 1.36 e 10 2 1.70 D 21 3 2.82 Internal proband, mock three-quarter sib family E 7 1 1.44 F 5 2 1.28 G 9 1 1.61 Alien proband, mock tbree-quarter H 1 I 3 Totals 69
sib family 0 .12 0 .38 12 12.15
In two, the RVe female was the elder; in these, two nonnal females were observed and 1.75 were expected, an insignificant deviation from expectation (P>.05). Four true three-quarter sib female families and three mock three-quarter sib female families, each with at least one RVe female, were identified. The mock sib families were paternal half-sib families from dams that had a paternal RVe sister. We also included two nonsegregating families that could be identified as potential segregants from their relationships to RVe animals not included in the family (alien proband). We included data from the mock sib families in an effort to extract all relatively easily evalu-
ated infonnation. It is instructive to summarize the kinds of family data from these herds. Segregating families include two full sib, four true three-quarter sib, and three mock threequarter sib; nonsegregating, two mock threequarter sib. The pooled segregation ratio b = .122, with Sb = .0448, confonns closely with the expected ratio of .125 for recessive inheritance (Table 1). Kansas State University Experimental Herd. Segregation in progeny groups from several mating types is shown in Table 2. No significant deviation from expectation under simple recessive inheritance was found (P>.05). The breeding herd included two RVe males, father and son; 16 RVe females; and 4 nonnal females. Among the 20 females, 4 mother and daughter pairs included 1 RVe pair and 3 mixed pairs with mothers RVe and daughters nonnal. Two of the latter mothers had no other progeny in the herd. Four paternal half sib pairs included three female pairs and one mixed pair including the older male. If RVe were to be inherited as a recessive trait, the breeding herd would have had to have 36 Rve genes. From pedigrees extended four ancestral generations, we determined that 19 of the RVe genes, including all 4 possessed by the males, could have been identical by descent from ancestor A; 7 could have been identical by descent from ancestor B, including 2 already counted as descending from A. Ancestor A and B have been identified as two probable major sources of the Rve gene. Of the 12 remaining RVe genes, 5 could have descended from ancestor e, the maternal grandfather of B. Ances-
TABLE 2. Segregation among progeny from Kansas State University experimental rectovaginal constriction (RVe) matings. Progeny Observed
Expected
Normal
RVe
Normal
RVe
Parental mating types
M1
F
M
F
M
F
M
F
RVe x Normal Normal Totals Totals Normal
0 3 0 3
0 3 0 3
8 0 1 9
11 0 1 12
0 3 0.5 3.5
0 3 0.5 3.5
9.5 0 0.5 10.0
9.5 0 0.5 10.0
RVe x RVe (RVe x N) x RVe (columns) (sexes combined) x uncertain
21
6
0
0
1M = Male, F = female. Jouma1 of Dairy Science Vol. 73,
No.9, 1990
7
20
GENETICS OF RECTOVAGINAL CONSTRICTION
tor A was in service 1943 to 1951; B, 1965 to 1972; and C, 1958 to 1968. Ancestors B and C have been declared RVC carriers by the AJCC. The remaining 7 can be identified only as identical in state. Field Cases. The AJCC encouraged reporting of suspected cases of RVC. From January 1973 to June 1980, the AJCC received 191 reports from 27 states, most of them of females. Knowledgeable AJCC officials sought additional infonnation. Confmnatory examinations of living animals by experienced persons were secured when possible, as were necropsies. Reproductive tracts of some animals were sent to the senior author for pathological examination. Blood types of about one-third of the cases helped confirm parentage. Pedigrees extended at least four ancestral generations were examined for common ancestry. One hundred fiftyone female case reports were determined to be of valid RVC females. These cases included the 16 RVC foundation females. The likely origins of the 302 RVC alleles posited under recessive inheritance were: ancestor A, 44.7%; ancestors B and C, 37.1 %, ancestor D, 2.3%; ancestor E, .7%, no common
2519
ancestor, 12.3%; and ancestry not recorded, i.e., one parent a grade animal, 3%. In a certain sense, each RVC animal represents the outcome of a test of allelism and that the two alleles that it must receive from its parents to become RVC are identical. In that connection, 42 animals likely received both alleles from ancestor A and 25 animals likely received both alleles from ancestors B plus C. Ancestors A and B plus C jointly contributed an allele in 35 animals. Alleles from ancestor A likely were tested also with 5 alleles from ancestor D and with 9 from unidentified sources. Alleles that were likely to be from ancestor B and C were tested also with one allele from ancestor D and E and with 19 from unidentified sources. Both alleles of four animals were from unidentified sources. A Genetic ModeJ for Rectovaglnal ConstrictIon: The Effects of selection Only In Females
Definitions. Let a = RVC allele, A = normal allele, and subscripts f and m denote alleles found in females and males, respectively. At the initial generation n, let frequency of allele ar, f(ar) = qr, f(anJ =
K. HUSTON Department of Dairy SCience Ohio Agricultural Research and Development Center The Ohio State University Wooster 44691 E. P. BARTON 2 American Jersey Cattle Club 2105 South Harrilton Road Columbus, OH 43227
J. R. HOLMAN Department of Statistics Ohio Agricultural Research and Development Center The Ohio State University Wooster 44691 D. L.TROYER Department of Anatomy and Physiology Kansas State University Manhattan 66506 ABSTRACT
males and 2,000 males annually are given for current conditions and after 500 generations of selection. (Key words: rectovaginal constriction, Jersey cattle, genetic model)
Data from fanner-owned herds and from experimental matings supported monofactorial recessive inheritance of rectovaginal constriction in US Jersey cattle. Kempthome' s population genetics model of a recessive trait involving only male selection was extended to include mutation and converted to selection of females only. Computer analyses with that model estimate slow decline in the frequency of the gene for rectovaginal constriction. Practical dynamics of the disorder in a breed registering 50,000 fe-
INTRODUCTION
Received October 2, 1989. Accepted March 14, 1990. IThis projoct was part of the Regional Project NC-2. CODtributioD number 360-1, Kansas Agricultural Experiment Station, and 10urnal Article Number 41-86, Ohio Agricultural Research and Development Center, 1be Ohio State University. 2Deceased . 1990 1 Dairy Sci 73:2516--2524
Rectovaginal constriction (RYC) is a congenital defect of Jersey cattle of both sexes. Its principal palpable and consequential characteristics are stenosis of the vestibular portion of the genital tract of females and of the anal sphincter of both sexes. Often it is accompanied by one or more of the constellation of defects that constitute the RYC syndrome including dystocia, udder edema, and excessive fibrous tissue in the anal sphincter muscle (1, 4, 5, 16, 21, 22). Comparisons of RYC homozygotes and heterozygotes with normal females revealed no differences in blood supply and innervation to the perineal viscera and musculature, in fibroblasts grown in vitro, in hydroxyproline content
2516
2517
GENETICS OF RECTOVAGINAL CONSlRICIlON
of tissues or urine, in duration of electromyographic action potentials, or in anal anatomical structures (anal canals). However, RVC homozygotes had unyielding connective tissue not only in the perineal region but also in the head region. Their external anal sphincter muscles also had more type I, m, and V collagen. Of possible significance in detecting heterozygotes are the discoveries that RVC heterozygotes, like RVC homozygotes, had greater amplitude in electromyographic action potentials than normals; they also had type n collagen, which normal cattle did not possess as reviewed recently (22). Rectovaginal constriction was fIrst reported to the Board of Directors, American Jersey Cattle Club (AlCC) in November 1972. Subsequently, in 1975, the AlCC listed RVC as the second abnormality in its Program of Action on Abnormalities based on information received since 1972 (7), of which some also was published in 1975 by Leipold and Saperstein (16), who suggested RVC was inherited as a simple recessive trait and may have been described in 1940 (13). In April 1980, the AlCC was sued by one of its members who sought $4.5 million for damage alleged to have resulted from the AlCC's actions (7). In October 1981, the court ruled in favor of the AlCC after hearing testimony from witnesses, some of whom were university scientists testifying as expert witnesses. That litigation revealed some evidentiary facets of the episode that may prove helpful to other scientists and that will be highlighted in the discussion (7, 9). In 1984, RVC was reported in Danish cattle sired by US bulls through AI (12, 15). Our primary purposes are to present analyses of data pertinent to the inheritance of RVC, a genetic model for it, and the dynamics of its occurrence in the Jersey breed. MATERIALS AND METHODS
Materials
Data were collected from: 1) Nevada herds identified by Leipold and Saperstein (16); 2) breeding trials in the Kansas State University experimental herd; and 3) field cases reported to the American Jersey Cattle Oub. During our 8-yr investigations of the disorder, we found that idiosyncracies of some persons performing examinations were a cause of
diagnostic confusion. Although there is variation in the size and elasticity of the rectum of RVC animals of the same age and relative size, variation among examiners is greater, primarily because of their lack of experience. When experienced examiners were permitted to examine both RVC and normal animals in a series of independent tests at Kansas State, they had little trouble in distinguishing RVC animals (17). Data from the Nevada herds were based on diagnoses made by the herd owner or his brother, both of whom had extensive experience in AI and pregnancy examination. Some were reconfirmed by us. Data from the Kansas State University herd were based on diagnoses made by the authors and confirmed by numerous other professionals. Data from field cases were included only when they conformed to a set of criteria designed to eliminate false RVC reports. Criteria included descriptions of the calving process, udder symptoms, and calf survival. Where possible, blood typing, veterinary report, and necropsy were secured. Methods
Segregation ratios in full-sib families were analyzed by the postproband method of Lush and Hazel (20). Segregation ratios in threequarter-sib families were analyzed by methods of Chase (8). In his appendix Table 3, Otase computed maximum likelihood estimators for segregation frequencies and their variances in sib families containing one and two or more affected individuals and ranging in size from 3 to 50 individuals. Complete ascertainment was assumed. These estimators correspond to means and variances of truncated binomial distributions, the use of which is illustrated for human families by Li (18). The genetic model is an extension of Kempthorne's model (14), which was based on selection in males rather than females and did not include mutation. The RVC gene frequency estimates by Allaire et al. (2) provided a base for computations with the newly derived model. Statistical methods are according to Fisher (11). RESULTS
Genetics
Nevada Herds. Three full-sib families, including at least one RVC female, were noted. Journal of Dairy
SciCIICC
Vol. 73.
No.9. 1990
2518
LEIPOLD ET AL.
TABLE I. Rectovaginal constriction (RVe) segregation ratios in tbree-quarter sib families in the Nevada herds•. Number RVe Animals Sibship
Expected
Size
Observed E(b) (8)
Group
Internal proband, true tbree-quarter sib family A 7 1 1.44 B 6 2 1.36 e 10 2 1.70 D 21 3 2.82 Internal proband, mock three-quarter sib family E 7 1 1.44 F 5 2 1.28 G 9 1 1.61 Alien proband, mock tbree-quarter H 1 I 3 Totals 69
sib family 0 .12 0 .38 12 12.15
In two, the RVe female was the elder; in these, two nonnal females were observed and 1.75 were expected, an insignificant deviation from expectation (P>.05). Four true three-quarter sib female families and three mock three-quarter sib female families, each with at least one RVe female, were identified. The mock sib families were paternal half-sib families from dams that had a paternal RVe sister. We also included two nonsegregating families that could be identified as potential segregants from their relationships to RVe animals not included in the family (alien proband). We included data from the mock sib families in an effort to extract all relatively easily evalu-
ated infonnation. It is instructive to summarize the kinds of family data from these herds. Segregating families include two full sib, four true three-quarter sib, and three mock threequarter sib; nonsegregating, two mock threequarter sib. The pooled segregation ratio b = .122, with Sb = .0448, confonns closely with the expected ratio of .125 for recessive inheritance (Table 1). Kansas State University Experimental Herd. Segregation in progeny groups from several mating types is shown in Table 2. No significant deviation from expectation under simple recessive inheritance was found (P>.05). The breeding herd included two RVe males, father and son; 16 RVe females; and 4 nonnal females. Among the 20 females, 4 mother and daughter pairs included 1 RVe pair and 3 mixed pairs with mothers RVe and daughters nonnal. Two of the latter mothers had no other progeny in the herd. Four paternal half sib pairs included three female pairs and one mixed pair including the older male. If RVe were to be inherited as a recessive trait, the breeding herd would have had to have 36 Rve genes. From pedigrees extended four ancestral generations, we determined that 19 of the RVe genes, including all 4 possessed by the males, could have been identical by descent from ancestor A; 7 could have been identical by descent from ancestor B, including 2 already counted as descending from A. Ancestor A and B have been identified as two probable major sources of the Rve gene. Of the 12 remaining RVe genes, 5 could have descended from ancestor e, the maternal grandfather of B. Ances-
TABLE 2. Segregation among progeny from Kansas State University experimental rectovaginal constriction (RVe) matings. Progeny Observed
Expected
Normal
RVe
Normal
RVe
Parental mating types
M1
F
M
F
M
F
M
F
RVe x Normal Normal Totals Totals Normal
0 3 0 3
0 3 0 3
8 0 1 9
11 0 1 12
0 3 0.5 3.5
0 3 0.5 3.5
9.5 0 0.5 10.0
9.5 0 0.5 10.0
RVe x RVe (RVe x N) x RVe (columns) (sexes combined) x uncertain
21
6
0
0
1M = Male, F = female. Jouma1 of Dairy Science Vol. 73,
No.9, 1990
7
20
GENETICS OF RECTOVAGINAL CONSTRICTION
tor A was in service 1943 to 1951; B, 1965 to 1972; and C, 1958 to 1968. Ancestors B and C have been declared RVC carriers by the AJCC. The remaining 7 can be identified only as identical in state. Field Cases. The AJCC encouraged reporting of suspected cases of RVC. From January 1973 to June 1980, the AJCC received 191 reports from 27 states, most of them of females. Knowledgeable AJCC officials sought additional infonnation. Confmnatory examinations of living animals by experienced persons were secured when possible, as were necropsies. Reproductive tracts of some animals were sent to the senior author for pathological examination. Blood types of about one-third of the cases helped confirm parentage. Pedigrees extended at least four ancestral generations were examined for common ancestry. One hundred fiftyone female case reports were determined to be of valid RVC females. These cases included the 16 RVC foundation females. The likely origins of the 302 RVC alleles posited under recessive inheritance were: ancestor A, 44.7%; ancestors B and C, 37.1 %, ancestor D, 2.3%; ancestor E, .7%, no common
2519
ancestor, 12.3%; and ancestry not recorded, i.e., one parent a grade animal, 3%. In a certain sense, each RVC animal represents the outcome of a test of allelism and that the two alleles that it must receive from its parents to become RVC are identical. In that connection, 42 animals likely received both alleles from ancestor A and 25 animals likely received both alleles from ancestors B plus C. Ancestors A and B plus C jointly contributed an allele in 35 animals. Alleles from ancestor A likely were tested also with 5 alleles from ancestor D and with 9 from unidentified sources. Alleles that were likely to be from ancestor B and C were tested also with one allele from ancestor D and E and with 19 from unidentified sources. Both alleles of four animals were from unidentified sources. A Genetic ModeJ for Rectovaglnal ConstrictIon: The Effects of selection Only In Females
Definitions. Let a = RVC allele, A = normal allele, and subscripts f and m denote alleles found in females and males, respectively. At the initial generation n, let frequency of allele ar, f(ar) = qr, f(anJ =
