• View in gallery

    Box-and-whiskers plots of hip joint radiographic phenotypes of 90 Labrador Retrievers (LR) and 143 non–Labrador Retrievers (non-LR) used for sequencing of intron 30 of FBN2. LDI = Left DI. LDLS = Left DLS score. LEHR = Left extended hip joint radiographic score. LNA = Left NA (degrees). RDI = Right DI. RDLS = Right DLS score. REHR = Right extended hip joint radiographic score. RNA = Right NA (degrees). The y-axes are the span of measurements for that trait in the dogs tested. The dark horizontal line indicates the median, and the box indicates the interquartile range. The highest and lowest values are indicated by the horizontal lines at the ends of the vertical dotted lines; circles indicate outliers.

  • View in gallery

    Cumulative distribution (proportion) plots for Labrador Retrievers genotyped for the FBN2 haplotype illustrating the cumulative effect of the FBN2 haplotype on each dog's hip joint measurement (the left [L] and right [R] DLS score [n = 101 dogs], NA [102], DI [101], and EHR score [102]). AGC indicates homozygosity for the FBN2 deletion. GAT indicates homozygosity for the unaffected haplotype. Het = Heterozygote. See Figure 1 for remainder of key.

  • View in gallery

    Plots of the effect of FBN2 intron 30 haplotype on the left and right hip joint measurements for 110 Labrador Retrievers (Dl and DLS data from 109 dogs) and 132 non–Labrador Retrievers. Points represent the means and bars represent the 95% confidence intervals for each of the 3 haplotypes. See Figures 1 and 2 for remainder of key.

  • View in gallery

    Photomicrographic views of hip joint capsule specimens from dogs. A–-Histologic section of a hip joint capsule specimen from a dog without osteo-arthritis (immunohistochemical stain; negative control [IgG]). B–-Same section as in panel A, stained with anti-FBN2 antibody. Notice the moderately intense diffuse immunostaining (brown-red) of the subsynovial collagenous matrix. C–-Histologic section of a joint capsule specimen from a dog with hip joint osteoarthritis. The synovial lining is at the top of the image (immunohistochemical stain; negative control [IgG]). D–-Same section as in panel C, stained with anti-FBN2 antibody. Notice that immunostaining of the tissue is low, compared with that in panel B.

  • 1.

    Lust G. An overview of the pathogenesis of canine hip dysplasia. J Am Vet Med Assoc 1997; 210:14431445.

  • 2.

    Marschall Y, Distl O. Mapping quantitative trait loci for canine hip dysplasia in German Shepherd Dogs. Mamm Genome 2007; 18:861870.

  • 3.

    Kaneene JB, Mostosky UV, Padgett GA. Retrospective cohort study of changes in hip joint phenotype of dogs in the United States. J Am Vet Med Assoc 1997; 211:15421544.

    • Search Google Scholar
    • Export Citation
  • 4.

    Olsewski JM, Lust G, Rendano VT, et al. Degenerative joint disease: multiple joint involvement in young and mature dogs. Am J Vet Res 1983; 44:13001308.

    • Search Google Scholar
    • Export Citation
  • 5.

    Carr AJ, Jefferson RJ, Benson MK. Joint laxity and hip rotation in normal children and in those with congenital dislocation of the hip. J Bone Joint Surg Br 1993; 75:7678.

    • Search Google Scholar
    • Export Citation
  • 6.

    Lust G, Beilman WT, Rendano VT. A relationship between degree of laxity and synovial fluid volume in coxofemoral joints of dogs predisposed for hip dysplasia. Am J Vet Res 1980; 41:5560.

    • Search Google Scholar
    • Export Citation
  • 7.

    Farese JP, Todhunter RJ, Lust G, et al. Dorsolateral subluxation of hip joints in dogs measured in a weight-bearing position with radiography and computed tomography. Vet Surg 1998; 27:393405.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Smith GK. Advances in diagnosing canine hip dysplasia. J Am Vet Med Assoc 1997; 210:14511457.

  • 9.

    Zhu L, Zhang Z, Feng F, et al. Single nucleotide polymorphisms refine QTL intervals for hip joint laxity in dogs. Anim Genet 2008; 39:141146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.

    Ramirez F, Dietz HC. Fibrillin-rich microfibrils: structural determinants of morphogenetic and homeostatic events. J Cell Physiol 2007; 213:326330.

  • 11.

    Gupta PA, Putnam EA, Carmical SG, et al. Ten novel FBN2 mutations in congenital contractural arachnodactyly: delineation of the molecular pathogenesis and clinical phenotype. Hum Mutat 2002; 19:3948.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Todhunter RJ, Acland GM, Olivier M, et al. An outcrossed canine pedigree for linkage analysis of hip dysplasia. J Hered 1999; 90:8392.

  • 13.

    Zhang Z, Zhu L, Sandler J, et al. Estimation of heritabilities, genetic correlations, and breeding values of four traits that collectively define hip dysplasia in dogs. Am J Vet Res 2009; 70:483492.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Gustafsson PO, Olsson SE, Kasstrom H, et al. Skeletal development of Greyhounds, German Shepherd Dogs and their crossbreed offspring. Acta Radiol Suppl 1975; 344:81108.

    • Search Google Scholar
    • Export Citation
  • 15.

    Smith GK, LaFond E, Gregor TP, et al. Within- and between-examiner repeatability of distraction indices of the hip joints in dogs. Am J Vet Res 1997; 58:10761077.

    • Search Google Scholar
    • Export Citation
  • 16.

    Lust G, Williams AJ, Burton-Wurster N, et al. Joint laxity and its association with hip dysplasia in Labrador Retrievers. Am J Vet Res 1993; 54:19901999.

    • Search Google Scholar
    • Export Citation
  • 17.

    Lust G, Todhunter RJ, Erb HN, et al. Repeatability of dorsolateral subluxation scores in dogs and correlation with macroscopic appearance of hip osteoarthritis. Am J Vet Res 2001; 62:17111715.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Lust G, Todhunter RJ, Erb HN, et al. Comparison of three radio-graphic methods for diagnosis of hip dysplasia in eight-month-old dogs. J Am Vet Med Assoc 2001; 219:12421246.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19.

    Todhunter RJ, Mateescu R, Lust G, et al. Quantitative trait loci for hip dysplasia in a cross-breed canine pedigree. Mamm Genome 2005; 16:720730.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Clark LA, Tsai KL, Steiner JM, et al. Chromosome-specific microsatellite multiplex sets for linkage studies in the domestic dog. Genomics 2004; 84:550554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.

    Mateescu RG, Burton-Wurster NI, Tsai K, et al. Identification of quantitative trait loci for osteoarthritis of hip joints in dogs. Am J Vet Res 2008; 69:12941300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22.

    Phavaphutanon J, Mateescu RG, Tsai KL, et al. Evaluation of quantitative trait loci for hip dysplasia in Labrador Retrievers. Am J Vet Res 2009; 70:10941101.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Heath SC. Markov chain Monte Carlo segregation and linkage analysis for oligogenic models. Am J Hum Genet 1997; 61:748760.

  • 24.

    Rozen S, Skaletsky H. Primer 3 on the WWW for general users and for biologist programmers. Methods Mol Biol 2000; 132:365386.

  • 25.

    Zhu L, Zhang Z, Friedenberg S, et al. The long (and winding) road to gene discovery for canine hip dysplasia. Vet J 2009; 181:97110.

  • 26.

    Awano T, Katz ML, O'Brien DP, et al. A frame shift mutation in canine TPP1 (the ortholog of human CLN2) in a juvenile Dachshund with neuronal ceroid lipofuscinosis. Mol Genet Metab 2006; 89:254260.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Chase K, Lawler DF, Adler FR, et al. Bilaterally asymmetric effects of quantitative trait loci (QTLs): QTLs that affect laxity in the right versus left coxofemoral (hip) joints of the dog (Canis familiaris). Am J Med Genet A 2004; 124A:239247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28.

    Putnam EA, Zhang H, Ramirez F, et al. Fibrillin-2 (FBN2) mutations result in the Marfan-like disorder, congenital contractural arachnodactyly. Nat Genet 1995; 11:456458.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Kealy RD, Lawler DF, Ballam JM, et al. Evaluation of the effect of limited food consumption on radiographic evidence of osteoarthritis in dogs. J Am Vet Med Assoc 2000; 217:16781680.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Vanden Berg-Foels WS, Todhunter RJ, Schwager SJ, et al. Effect of early postnatal body weight on femoral head ossification onset and hip osteoarthritis in a canine model of developmental dysplasia of the hip. Pediatr Res 2006; 60:549554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Jansen RC, Nap JP, Mlynárová L. Errors in genomics and proteomics. Nat Biotechnol 2002; 20:19.

  • 32.

    Pascal LE, True LD, Campbell DS, et al. Correlation of mRNA and protein levels: cell type-specific gene expression of cluster designation antigens in the prostate. BMC Genomics 2008; 9:246.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Kielty CM, Baldock C, Lee D, et al. Fibrillin: from microfibril assembly to biomechanical function. Philos Trans R Soc Lond B Biol Sci 2002; 357:207217.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34.

    Chaudhry SS, Gazzard J, Baldock C, et al. Mutation of the gene encoding fibrillin-2 results in syndactyly in mice. Hum Mol Genet 2001; 10:835843.

  • 35.

    Arteaga-Solis E, Gayraud B, Lee SY, et al. Regulation of limb patterning by extracellular microfibrils. J Cell Biol 2001; 154:275281.

  • 36.

    Boregowda R, Paul E, White J, et al. Bone and soft connective tissue alterations result from loss of fibrillin-2 expression. Matrix Biol 2008; 27:661666.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 37.

    Hurle JM, Corson G, Daniels K, et al. Elastin exhibits a distinctive temporal and spatial pattern of distribution in the developing chick limb in association with the establishment of the cartilaginous skeleton. J Cell Sci 1994; 107:26232634.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 38.

    Yanagino T, Yuasa K, Nagahama M, et al. Transcriptional regulation of fibrillin-2 gene by E2F family members in chondrocyte differentiation. J Cell Biochem 2009; 106:580588.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 39.

    Toma DP, White KP, Hirsch J, et al. Identification of genes involved in Drosophila melanogaster geotaxis, a complex behavioral trait. Nat Genet 2002; 31:349353.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 40.

    Walji AH. Some histological aspects of the hip joint capsule in the vervet monkey. J Morphol 1988; 197:327335.

  • 41.

    Todhunter RJ, Lust G. Canine hip dysplasia: pathogenesis. In: Slatter D, ed. Textbook of small animal surgery. 3rd ed. St Louis: WB Saunders Co, 2003;20092019.

    • Search Google Scholar
    • Export Citation
  • 42.

    Greisen HA, Summers BA, Lust G. Ultrastructure of the articular cartilage and synovium in the early stages of degenerative joint disease in canine hip joints. Am J Vet Res 1982; 43:19631971.

    • Search Google Scholar
    • Export Citation
  • 43.

    Hui-Chou CS, Lust G. The type of collagen made by the articular cartilage in joints of dogs with degenerative joint disease. Coll Relat Res 1982; 2:245256.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 44.

    Wiltberger H, Lust G. Ultrastructure of canine articular cartilage: comparison of normal and degenerative (osteoarthritic) hip joints. Am J Vet Res 1975; 36:727740.

    • Search Google Scholar
    • Export Citation
  • 45.

    Oda H, Igarashi M, Hayashi Y, et al. Soft tissue collagen in congenital dislocation of the hip. Biochemical studies of the ligamentum teres of the femur and the hip joint capsule. Nippon Seikeigeka Gakkai Zasshi 1984; 58:331338.

    • Search Google Scholar
    • Export Citation
  • 46.

    Jensen BA, Reimann I, Fredensborg N. Collagen type III predominance in newborns with congenital dislocation of the hip. Acta Orthop Scand 1986; 57:362365.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 47.

    Wang EB, Zhao Q, Li LY, et al. Expression of COL1a1 and COL3a1 in the capsule of children with developmental dislocation of the hip [in Chinese]. Zhongguo Dang Dai Er Ke Za Zhi 2008; 10:493496.

    • Search Google Scholar
    • Export Citation
  • 48.

    Madsen JS. The joint capsule and joint laxity in dogs with hip dysplasia. J Am Vet Med Assoc 1997; 210:14631465.

  • 49.

    Steinetz BG, Williams AJ, Lust G, et al. Transmission of relaxin and estrogens to suckling canine pups via milk and possible association with hip joint laxity. Am J Vet Res 2008; 69:5967.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 50.

    Samuel CS, Sakai LY, Amento EP. Relaxin regulates fibrillin 2, but not fibrillin 1, mRNA and protein expression by human dermal fibroblasts and murine fetal skin. Arch Biochem Biophys 2003; 411:4755.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 51.

    Cardinet GH III, Guffy MM, Wallace LJ, et al. Canine hip dysplasia in German Shepherd Dog-Greyhound crossbreeds. J Am Vet Med Assoc 1983; 182:393395.

    • Search Google Scholar
    • Export Citation
  • 52.

    Matyas JR, Adams ME, Huang D, et al. Major role of collagen IIB in the elevation of total type II procollagen messenger RNA in the hypertrophic phase of experimental osteoarthritis. Arthritis Rheum 1997; 40:10461049.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 53.

    Hirokawa N, Noda Y. Intracellular transport and kinesin super-family proteins, KIFs: structure, function, and dynamics. Physiol Rev 2008; 88:10891118.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 54.

    Sutter NB, Bustamante CD, Chase K, et al. A single IGF1 allele is a major determinant of small size in dogs. Science 2007; 316:112115.

Advertisement

Evaluation of a fibrillin 2 gene haplotype associated with hip dysplasia and incipient osteoarthritis in dogs

View More View Less
  • 1 Department of Clinical Sciences
  • | 2 Department of Statistics, College of Arts and Sciences, Oklahoma State University, Stillwater, OK 74078
  • | 3 Institute for Genomic Diversity, College of Agriculture and Life Sciences
  • | 4 Department of Clinical Sciences
  • | 5 Biotechnology Resource Center Cornell University, Ithaca, NY 14853
  • | 6 Microarray Core Facility
  • | 7 Department of Biomedical Sciences
  • | 8 Department of Clinical Sciences
  • | 9 James A. Baker Institute for Animal Health College of Veterinary Medicine
  • | 10 Department of Clinical Sciences
  • | 11 Department of Clinical Sciences
  • | 12 Department of Biomedical Sciences
  • | 13 Department of Clinical Sciences
  • | 14 Department of Clinical Sciences
  • | 15 Department of Biomedical Sciences
  • | 16 James A. Baker Institute for Animal Health College of Veterinary Medicine
  • | 17 Department of Clinical Sciences
  • | 18 Department of Clinical Sciences
  • | 19 Department of Biomedical Sciences
  • | 20 Department of Clinical Sciences

Abstract

Objective—To determine whether a mutation in the fibrillin 2 gene (FBN2) is associated with canine hip dysplasia (CHD) and osteoarthritis in dogs.

Animals—-1,551 dogs.

Procedures—Hip conformation was measured radiographically. The FBN2 was sequenced from genomic DNA of 21 Labrador Retrievers and 2 Greyhounds, and a haplotype in intron 30 of FBN2 was sequenced in 90 additional Labrador Retrievers and 143 dogs of 6 other breeds. Steady-state values of FBN2 mRNA and control genes were measured in hip joint tissues of fourteen 8-month-old Labrador Retriever–Greyhound crossbreeds.

Results—The Labrador Retrievers homozygous for a 10-bp deletion haplotype in intron 30 of FBN2 had significantly worse CHD as measured via higher distraction index and extended-hip joint radiograph score and a lower Norberg angle and dorsolateral subluxation score. Among 143 dogs of 6 other breeds, those homozygous for the same deletion haplotype also had significantly worse radiographic CHD. Among the 14 crossbred dogs, as the dorsolateral subluxation score decreased, the capsular FBN2 mRNA increased significantly. Those dogs with incipient hip joint osteoarthritis had significantly increased capsular FBN2 mRNA, compared with those dogs without osteoarthritis. Dogs homozygous for the FBN2 deletion haplotype had significantly less FBN2 mRNA in their femoral head articular cartilage.

Conclusions and Clinical Relevance—The FBN2 deletion haplotype was associated with CHD. Capsular gene expression of FBN2 was confounded by incipient secondary osteoarthritis in dysplastic hip joints. Genes influencing complex traits in dogs can be identified by genome-wide screening, fine mapping, and candidate gene screening.

Abstract

Objective—To determine whether a mutation in the fibrillin 2 gene (FBN2) is associated with canine hip dysplasia (CHD) and osteoarthritis in dogs.

Animals—-1,551 dogs.

Procedures—Hip conformation was measured radiographically. The FBN2 was sequenced from genomic DNA of 21 Labrador Retrievers and 2 Greyhounds, and a haplotype in intron 30 of FBN2 was sequenced in 90 additional Labrador Retrievers and 143 dogs of 6 other breeds. Steady-state values of FBN2 mRNA and control genes were measured in hip joint tissues of fourteen 8-month-old Labrador Retriever–Greyhound crossbreeds.

Results—The Labrador Retrievers homozygous for a 10-bp deletion haplotype in intron 30 of FBN2 had significantly worse CHD as measured via higher distraction index and extended-hip joint radiograph score and a lower Norberg angle and dorsolateral subluxation score. Among 143 dogs of 6 other breeds, those homozygous for the same deletion haplotype also had significantly worse radiographic CHD. Among the 14 crossbred dogs, as the dorsolateral subluxation score decreased, the capsular FBN2 mRNA increased significantly. Those dogs with incipient hip joint osteoarthritis had significantly increased capsular FBN2 mRNA, compared with those dogs without osteoarthritis. Dogs homozygous for the FBN2 deletion haplotype had significantly less FBN2 mRNA in their femoral head articular cartilage.

Conclusions and Clinical Relevance—The FBN2 deletion haplotype was associated with CHD. Capsular gene expression of FBN2 was confounded by incipient secondary osteoarthritis in dysplastic hip joints. Genes influencing complex traits in dogs can be identified by genome-wide screening, fine mapping, and candidate gene screening.

Canine hip dysplasia has been the subject of intensive study1,2 because some large dog breeds have <50% incidence.3 The trait has a systemic component because multiple joints can be involved,4 yet its most conspicuous manifestation is in the hip joint. Debilitating osteoarthritis, which occurs in affected dogs, adversely affects hip function and gait.

Several investigators have explored the origins of hip joint laxity and instability in the supporting structures of the hip joint, especially the hip joint capsule and round ligament of the femoral head, because interruption of such support may be a critical initiating feature of dysplastic hip development.5 Lust et al,6 Farese et al,7 and Smith8 reported that hip joint laxity, accompanied by synovial effusion and hypertrophy of the round ligament of the femoral head that signal incipient osteoarthritis, contributed to the pathogenesis of CHD.

On the basis of previous genome-wide linkage analysis and fine mapping,9 we hypothesized that a candidate gene in the QTL interval on CFA11 at 18.5 to 21 cM influenced hip joint laxity, as measured by the DI, in dogs. One gene at 20.3 to 20.5 megabases on CFA11, FBN2, was an attractive positional candidate gene because it encodes for a microfibrillary component of extracellular matrix that is present in fibrous joint capsule and articular cartilage.10 Mutations in FBN2 are associated with congenital contractural arachnodactyly in humans,11 and some affected patients have had joint laxity. The purpose of the study reported here was to determine whether a mutation in FBN2 is associated with CHD and osteoarthritis in dogs.

Materials and Methods

Cornell DNA Hip Archive–-Study dogs were from closed breeding colonies at the Baker Institute for Animal Health at Cornell University and Guiding Eyes for the Blind (Yorktown Heights, NY), and dogs admitted to the Cornell University Hospital for Animals. There was DNA available from 210 Labrador Retrievers bred from 1989 over 8 generations. More recently, the authors developed a crossbreed pedigree for QTL mapping of CHD founded with 7 unaffected Greyhounds (2 males and 5 females) and 8 Labrador Retrievers (4 males and 4 females) with CHD and secondary osteoarthritis.12 The pelves of dogs from the Baker Institute for Animal Health underwent radiography between 8 and 12 months of age, whereas the dogs from Guiding Eyes for the Blind underwent radiography between 14 and 18 months of age. The pedigrees of all available dogs were used to derive breeding values for hip trait mapping. The final data set contained Labrador Retrievers, Greyhounds, their crossbred offspring, and 14 other breeds, totaling 1,551 dogs that had at least 1 of the 4 radiographic hip measurements. Ancestry of these 1,551 dogs was traced back as far as 17 generations, resulting in a total of 2,716 dogs, including 1,498 dogs from Guiding Eyes for the Blind, 571 from the Baker Institute for Animal Health, 425 from the Cornell University Hospital for Animals, and 222 ancestors traced from the Orthopedic Foundation for Animals semiopen database.13 The study was approved by the Cornell University Institutional Animal Care and Use Committee.

Phenotyping–-The severity of CHD was measured from pelvic radiographs by use of the NA,14 DI,15,16 DLS score,17 and EHR projection score adopted by the Orthopedic Foundation for Animals and by evaluation of hip osteoarthritis (new bone formation on the femoral head and neck and acetabular osteophytes) on the EHR. Although these radiographic measurements were correlated,18 none of the measurements completely replaced the other so that at least 2 of them are required to represent a dysplastic hip joint.

Narrowing the QTL interval on CFA11–-A QTL for CHD on 12 chromosomes in a Greyhound–Labrador Retriever pedigree was previously identified.19 To improve genome coverage, these dogs were later genotyped at 225 additional microsatellites of minimal screening set 2 on 16 chromosomes.20–22 Linkage analysis confirmed the presence of a QTL for CHD segregating in the crossbreed pedigree with a maximum logarithm of odds peak score of 3.45 on CFA11 at 36 cM for the second principal component of the DLS score and DI. Genotyping was performed on 257 dogs, including Labrador Retrievers, Greyhounds, and their crossbred offspring, at 111 SNPs in this QTL region on CFA11. Multipoint linkage analysis23 revealed substantial evidence for 2 QTLs contributing to the DI between 15.7 to 17 cM and 18.5 to 21 cM (95% posterior probability interval), which explained approximately 15% to 18% of the total variance in the DI.9 FBN2, a candidate gene at 20.3 to 20.5 megabases, was within this interval.

FBN2 screening–-The exon structure of FBN2 was identified by use of an online genome browsera and National Center for Biotechnology Informationb and University of California-Santa Cruzc genome databases of the second release of the Canis familiaris genome sequence. Primer sequences were designed by use of software24 and ordered from a commercial sourced and are available from the primary author (SGF). Exons were amplified from genomic DNA via PCR under standard conditions. Amplicons were purified.e,f Each exon was sequenced initially in the forward direction only, then in the reverse direction if necessary. Sequencing was performed on a commercial sequencer.g In addition, introns 1 and 2 and 372 bp upstream of exon 1 were sequenced from genomic DNA to identify regulatory regions that might affect gene expression. Sequence was analyzed by use of commercial software.h

Sequencing and association study–-FBN2 was first sequenced in 21 Labrador Retrievers and 2 Greyhounds. Intron 30 of FBN2 was then sequenced in a larger number of dogs selected from the dogs in the CHD DNA archive. Dogs were chosen by use of an algorithm to maximize genetic diversity and minimize relatedness.25 Dog selection based on phenotype was performed by use of the principal components of the DI, DLS score, EHR score, and NA, weighted by their eigenvalues. Intron 30 of FBN2 was sequenced in 90 additional Labrador Retrievers, 18 Border Collies, 47 German Shepherd Dogs, 29 Golden Retrievers, 29 Newfoundlands, 19 Rottweilers, and 1 Great Dane. The PCR primers (forward, 5′-ggccaatgtaccaacattcc; reverse, 5′-tgaaggatcccttggtgttc) were designed to amplify intron 30 of FBN2 and included at least 50 bp in either direction by use of commercial software.24 M13 universal primers were added to the 5′ end of each primer to facilitate sequencing.

Characterization of FBN2 expression–-Specimens of the dorsal fibrous joint capsule, round ligament of the femoral head, and full-thickness noncalcified articular cartilage from the dorsal surface of the femoral head but not including the perifoveal area were collected from fourteen 8-month-old Labrador Retriever–Greyhound crossbred dogs at necropsy. Six of the dogs were from one litter (three-quarter Labrador Retriever and one-quarter Greyhound [litter 1]), and the remaining 8 dogs were from another litter (seven-eighth Labrador Retriever and one-eighth Greyhound [litter 2]).12,19 The joints were evaluated grossly for evidence of osteoarthritis as the traditional indication of hip dysplasia.17

Tissue was harvested within 2 hours of euthanasia. Samples from the right hip joint of each dog were used for mRNA isolation. Samples from litter 1 were flash frozen in liquid nitrogen and stored at −80°C, and samples from litter 2 were preserved for later RNA extraction1 and stored at −20°C. The RNA was isolated by use of a commercial kit designed to extract RNA from fibrous tissues.26 Briefly approximately 150 mg of each frozen sample was placed in a stabilizing buffer and homogenized 2 or 3 times for 30 to 45 seconds on ice. The homogenate was incubated at 55°C for 20 minutes with proteinase K and was then pelleted by centrifugation at 4,500 × g to remove cellular debris. The supernatant solution was collected, a 0.5 volume of 100% ethanol was added, and the solution was filtered via centrifugation at 4,500 × g through a silica column designed to bind RNA. DNAse I was added to the bound RNA to remove any contaminating DNA. The sample was washed and eluted in nuclease-free water. Concentrations and purity were assessed spectrophotometrically.

Expression of FBN2 was measured via quantitative reverse transcription PCR assay. First, cDNA was prepared from each RNA sample. One microgram of RNA was added to a standard mixture of deoxynucleotide triphosphates (4×), random primers (1×), recombinant Moloney murine leukemia virus reverse transcriptase (1×), and reaction buffer (1×) in a 1:1 (vol/vol) RNA:mixture ratio. Reactions were incubated in a thermal cycler, annealment was performed at 25°C for 10 minutes, copying was performed at 37°C for 2 hours, and inactivation was performed at 85°C for 5 seconds. Custom primers and probes were designed to amplify portions of FBN2 cDNA at the junctions of exons 4–5, 30–31, and 59–60 (Appendix 1). Glucose phosphate isomerase was used as an endogenous control with a single probe at the exon 2–3 junction. Probes were labeled with 6-carboxy-fluorescein and a nonfluorescent quencher and synthesized by the manufacturerg at a 20× concentration. Second-step reactions were prepared as follows: approximately 60 ng of cDNA, each primer-probe mix (1×), and commercial master mix (1×)g in a final volume of 20 μL. Each PCR reaction was carried out in quadruplicateg and activated at 95°C for 10 minutes, followed by 40 repeat cycles of 15-second denaturation at 95°C, followed by annealing-extension at 60°C for 1 minute. Relative quantification was performed by use of the mean difference in threshold cycle values of all the samples for a particular probe as the calibrator; difference in threshold cycle values were read midway into the linear phase of the amplification curve. Expression of control genes FBN1, kinesin 11 (KIN11), and topoisomerase 1 (TOP1) was similarly measured.g

Alternative splicing–-Two or 3 mRNA samples isolated from fibrous joint capsule, round ligament, and articular cartilage from dogs of each intron 30 FBN2 genotype (unaffected homozygous, heterozygous, and affected or deletion homozygous) were selected for further analysis. One-step reverse transcription PCR amplification of these samples was performed in regions adjacent to intron 30 in 2 experiments. In the first experiment, 1 upstream primer was selected within exon 26 and downstream primers were chosen within exons 28, 29, 30, 31, and 33 (Appendix 2). In the second experiment, 1 upstream primer was chosen within exon 34 and downstream primers were chosen within exons 36, 37, 39, 40, 41, 42, and 43. The reverse transcription PCR amplification was performed by use of a 1-step reverse transcription PCR system.j Briefly, 250 ng of template RNA was mixed with 2 U of reverse transcriptase and 0.2 μM of each primer and reaction buffer and incubated. Thermal cycler temperatures were 55°C for 30 minutes, 94°C for 2 minutes, 40 cycles of 94°C for 15 seconds, 55°C for 30 seconds, 68°C for 1 minute, and 68°C for 5 minutes. The PCR reaction products were analyzed on a 2% agarose gel to detect amplicons shorter than expected.

Immunohistochemical analyses–-Formalin-fixed sections of dorsal hip capsule from affected and unaffected crossbred dogs from which the mRNA was isolated and amplified were also examined by use of H&E stain; Vierhoff von Giesen stain for elastin fibers was also used. The elastin stain revealed sparse numbers of short fibrils concentrated primarily beneath the synovial lining in the loosely arranged connective tissue. For immunohistochemical analyses, sections were deparaffinized, blocked with 0.5% H2O2 in methanol, and washed with distilled H2O and Tris-buffered saline solutionk (pH 7.6). Sections were blocked with 10% normal goat serum with 2× caseinl for 20 minutes, and incubated for 2 hours at 37°C with rabbit anti-FBN2m monoclonal antibody diluted 1:40 in Tris-buffered saline solution with 1× casein. After washing with Tris-buffered saline solution, biotinylated goat anti-rabbit IgGn was applied at 1:200 for 20 minutes at 20°C, followed by Tris-buffered saline solution washes and incubation for 20 minutes with streptavidin peroxidase.l The substrate was prepared as directedl and applied for 10 minutes. Slides were counterstained with hematoxylin and mounted.o Controls were included, substituting normal rabbit IgGk in place of primary antibody. Immunohistochemical staining for FBN2 expression was compared with serial sections treated with normal rabbit serum. The intensity of staining was subjectively graded.

Statistical analysis–-The effect of the FBN2 haplotype on each hip joint measurement was evaluated in cumulative distribution plots. The cumulative distribution plots were compared with results of the Kolmogorov-Smirnov test. The linear model used to test the effect of the FBN2 haplotype on hip dysplasia was as follows:

article image

where the interaction term included all possible pair-wise interactions. Variable selection strategy was performed to find the final linear model with significant (P >0.05) terms only. This model was applied to each of the 4 CHD traits. Given any randomly sampled dog of known breed and body weight and knowing its FBN2 haplotype, a prediction interval for the hip trait was calculated.

Values for expression of mRNA were log transformed and tested as a function of the presence or absence of secondary osteoarthritis or CHD phenotypes (DLS score, NA, and EHR score) in a mixed linear model that included the covariates of the dog's genotype at the FBN2 locus, sex, age, and body weight, with a random effect for litter. To test the effect of hip joint phenotype on the expression of the control genes, FBN1, TOP1, and KIN11 in fibrous joint capsule, the FBN2 genotype was not included in the model. The effect of the hip joint phenotype on the expression of FBN1 and FBN2 only was tested in the femoral head articular cartilage. For all comparisons, values of P >0.05 were considered significant.

Results

Dogs and phenotypes–-Twenty-three dogs were first selected for FBN2 sequencing of genomic DNA. One group of 12 affected Labrador Retrievers had EHR scores ≥ 5, NAs >100°, and DLS scores >45° for both hip joints. The other group of 11 unaffected dogs included 9 Labrador Retrievers and 2 Greyhounds with EHR scores ≤ 2, NAs <107°, and DLS scores <55° in both hip joints. The hip joint phenotypes of an additional 90 Labrador Retrievers and 143 non–Labrador Retrievers used to sequence intron 30 of FBN2 were determined (Figure 1). Of the 257 dogs originally used in SNP fine mapping that localized the QTL to 18 to 21 cM on CFA11,14 only 8 Labrador Retrievers from the Baker Institute for Animal Health and 11 German Shepherd Dogs satisfied the criteria for FBN2 sequencing of genetic and phenotypic diversity. Of the 14 Labrador Retriever crossbred dogs used for quantitative reverse transcription PCR analysis of FBN2, FBN1, TOP1, and KIN11, 6 dogs had early hip joint osteoarthritis at necropsy characterized by perifoveal articular cartilage fibrillation, and 8 dogs had no gross signs of hip joint osteoarthritis. Only 1 dog had unilateral (right) hip joint osteoarthritis and no signs of osteoarthritis in the opposite hip joint. For these dogs, mean ± SD right hip joint DLS score was 48 ± 6.3 (range, 41% to 62%), mean ± SD EHR score was 2.4 ± 1.2 (range, 1 to 5), and mean ±SD NA was 99.3 ± 5 (range, 90° to 105°).

Figure 1—
Figure 1—

Box-and-whiskers plots of hip joint radiographic phenotypes of 90 Labrador Retrievers (LR) and 143 non–Labrador Retrievers (non-LR) used for sequencing of intron 30 of FBN2. LDI = Left DI. LDLS = Left DLS score. LEHR = Left extended hip joint radiographic score. LNA = Left NA (degrees). RDI = Right DI. RDLS = Right DLS score. REHR = Right extended hip joint radiographic score. RNA = Right NA (degrees). The y-axes are the span of measurements for that trait in the dogs tested. The dark horizontal line indicates the median, and the box indicates the interquartile range. The highest and lowest values are indicated by the horizontal lines at the ends of the vertical dotted lines; circles indicate outliers.

Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.530

Sequencing–-A haplotype was identified within the first 100 bp of intron 30 in FBN2 that segregated in 12 Labrador Retrievers affected with CHD (AGC haplotype) and 9 Labrador Retrievers and 2 Greyhounds unaffected with CHD (GAT haplotype; Appendix 3). The GAT haplotype extended downstream (TACTTTA between bp 189 and 195 where the 7-bp deletion for the AGC haplotype occurred) within intron 30. Additional sequencing revealed that the GATTACTTTA haplotype in intron 30 extended further downstream, with A at intron 37 base 27, C at intron 40 base 32, and T at exon 41 base 143; the latter caused no change in amino acid sequence. In these same positions, the haplotype continued downstream as follows: G-A-C for the AGC (deletion) haplotype, and R-M-Y for the heterozygous haplotype. Via sequencing of 372 bp upstream of FBN2 exon 1, an SNP was found that was not associated with the 2 groups that were either affected or unaffected with CHD. It was not possible to sequence farther upstream. In the same group of dogs, intron 2 was also sequenced and none of the 3 SNPs discovered was associated with CHD in the same dogs.

The DNA from an additional 90 Labrador Retrievers and 143 non–Labrador Retrievers was sequenced for the same FBN2 haplotype in intron 30. The cumulative trait distributions for the effect of the 3 genotypes on each hip joint trait on the Labrador Retrievers were determined (Figure 2). The cumulative distribution plots of the Labrador Retrievers homozygous for the AGC haplotype were significantly different, compared with either of the other 2 haplotypes for all traits except the left NA (P = 0.08), the right DI (P = 0.08), and both left and right EHR scores (largest P = 0.29). However, via analysis in a linear model with covariates, dogs homozygous for the AGC (deletion) haplotype had worse hip joint conformation characterized by a lower DLS score, higher DI, lower NA, and higher EHR score than did those with the GAT or heterozygous haplotype (Figure 3).

Figure 2—
Figure 2—

Cumulative distribution (proportion) plots for Labrador Retrievers genotyped for the FBN2 haplotype illustrating the cumulative effect of the FBN2 haplotype on each dog's hip joint measurement (the left [L] and right [R] DLS score [n = 101 dogs], NA [102], DI [101], and EHR score [102]). AGC indicates homozygosity for the FBN2 deletion. GAT indicates homozygosity for the unaffected haplotype. Het = Heterozygote. See Figure 1 for remainder of key.

Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.530

Figure 3—
Figure 3—

Plots of the effect of FBN2 intron 30 haplotype on the left and right hip joint measurements for 110 Labrador Retrievers (Dl and DLS data from 109 dogs) and 132 non–Labrador Retrievers. Points represent the means and bars represent the 95% confidence intervals for each of the 3 haplotypes. See Figures 1 and 2 for remainder of key.

Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.530

An example of the effect of the FBN2 genotype on a CHD trait (the ANOVA table for the effect of haplotype, body weight, and breed on the left NA) was developed (Table 1). Other hip joint traits were similarly affected by the haplotype. Because the dog's sex, age when the radiograph was taken, and all other interaction terms except body weight by breed were not significant, these variables were dropped from the model. This linear model explained approximately 14% of the total variation in the left NA.

Table 1—

Estimated coefficients for the effects of the FBN2 haplotype, body weight, and breed on the left NA in dogs.

CoefficientEstimateSEP value
Intercept76.159.506.72
Haplotype GAT6.842.290.003
Haplotype het5.902.310.01
Body weight0.790.290.001
NLR breed31.7310.130.001
NLR breed body weight–1.150.31>0.001

Each genotype and osteoarthritis class of the fixed-effect variables is compared with the class with the lowest estimate.

het= Heterozygote. NLR = Non–Labrador Retriever.

Multiple R2= 0.143; adjusted R2= 0.124.

For any randomly sampled Labrador Retriever with body weight of 30 kg that was homozygous for the deletion AGC haplotype, the equation for the estimated left NA was as follows:

article image

where LR is Labrador Retriever breed. For a similar dog homozygous for the GAT haplotype, the equation for the estimated left NA was as follows:

article image

For a similar but heterozygous Labrador Retriever, the equation for the estimated left NA was as follows:

article image

where het is heterozygous. However, the interaction of breed and body weight contributed significantly to the variation of the NA. Heavy Labrador Retrievers had higher NAs than did heavy non–Labrador Retrievers. Similar interactions of body weight occurred with the other hip joint traits (data not shown).

FBN1, FBN2, KIN11, and TOP1 expression and immunohistochemical analysis–-In the mixed linear model, the effects of sex, age, and body weight on gene expression were not significant and these terms were dropped from the model. As the DLS score decreased by 1%, FBN2 expression in the fibrous joint capsule increased by approximately 5% (Tables 2–5). The model including the DLS score and hip joint osteoarthritis explained approximately 40% of FBN2 expression. Dogs that had incipient osteoarthritis at necropsy as the traditional indicator of prior CHD had approximately 50% greater FBN2 mRNA in their hip joint capsule than did dogs with nonaffected hip joints. Dogs with the AGC homozygous genotype had less FBN2 expression in hip joint capsules, although the difference was not significant. In the femoral head articular cartilage, the GAT homozygous dogs had significantly higher FBN2 expression, compared with the heterozygotes and dogs homozygous for the AGC (deletion) haplotype. There was no evidence of alternatively spliced FBN2 transcripts resulting from the deletion in intron 30 in mRNA transcript isolated from the joint capsule. Most immunostaining for FBN2 was seen in the sub-synovial region of the hip joint capsule (Figure 4). No qualitative difference was observed in immunostaining of FBN2 in the hip joint capsule of affected hip joints, compared with unaffected hip joints.

Figure 4—
Figure 4—

Photomicrographic views of hip joint capsule specimens from dogs. A–-Histologic section of a hip joint capsule specimen from a dog without osteo-arthritis (immunohistochemical stain; negative control [IgG]). B–-Same section as in panel A, stained with anti-FBN2 antibody. Notice the moderately intense diffuse immunostaining (brown-red) of the subsynovial collagenous matrix. C–-Histologic section of a joint capsule specimen from a dog with hip joint osteoarthritis. The synovial lining is at the top of the image (immunohistochemical stain; negative control [IgG]). D–-Same section as in panel C, stained with anti-FBN2 antibody. Notice that immunostaining of the tissue is low, compared with that in panel B.

Citation: American Journal of Veterinary Research 72, 4; 10.2460/ajvr.72.4.530

Table 2—

Results of ANOVA for FBN2 expression in the fibrous joint capsule of 14 Labrador Retriever–Greyhound crossbred dogs with and without hip joint dysplasia as a function of the presence (aff) or absence of osteoarthritis after adjusting for each dog's genotype at intron 30 of FBN2.

CoefficientEstimateSEPvalue
Intercept–0.160.410.76
Genotype aff–0.020.280.49
Genotype het0.040.180.83
Osteoarthritis aff0.520.220.03

aff = AGC homozygote.

See Table 1 for remainder of key.

Table 3—

Results of ANOVA for FBN2 expression in the fibrous joint capsule of 14 Labrador Retriever–Greyhound crossbred dogs with and without hip joint dysplasia as a function of the DLS score after adjusting for each dog's genotype at intron 30 of FBN2.

CoefficientEstimateSEP value
Intercept1.940.930.28
Genotype aff0.590.330.10
Genotype het0.520.260.06
DLS score–0.050.020.03

See Tables 1 and 2 for key.

Table 4—

Results of ANOVA for KIN11 expression in the fibrous joint capsule of 14 Labrador Retriever–Greyhound crossbred dogs with and without hip joint dysplasia as a function of the EHR score after adjusting for each dog's genotype at intron 30 FBN2.

CoefficientEstimateSEP value
Intercept–0.490.310.36
Genotype aff–0.480.260.11
Genotype het–0.070.200.71
EHR score0.340.130.04

See Tables 1 and 2 for key.

Table 5—

Results of ANOVA for FBN2 expression in the femoral head articular cartilage of 14 Labrador Retriever–Greyhound crossbred dogs with and without hip joint dysplasia as a function of the presence or absence of osteoarthritis after adjusting for each dog's genotype at intron 30 of FBN2.

CoefficientEstimateSEP value
Intercept0.431.340.02
Genotype unaff10.352.120.01
Genotype het1.691.700.35
Osteoarthritis unaff0.801.850.73

unaff = GAT homozygote. See Tables 1 and 2 for remainder of key.

There was no significant difference in mRNA expression in the fibrous joint capsule of FBN1 and TOP1 of dogs with osteoarthritis, compared with those without osteoarthritis. However, KIN11 mRNA expression was significantly greater in the joint capsules of osteoarthritic hip joints (Tables 2–5). For each unit increase (worsening) of the EHR score, the KIN11 expression increased by 34%. There was no effect of litter or tissue preparation on gene expression.

Discussion

The candidate gene, FBN2, was identified through genome-wide, microsatellite-based screening of a Labrador Retriever–Greyhound pedigree, followed by narrowing of the QTL region on CFA11 through multipoint linkage analysis with SNP genotypes. An association between a mutation-deletion haplotype in FBN2 and CHD was detected. This is the first gene reported to be associated with all 4 CHD radiographic measurements, an important orthopedic trait that leads to debilitating hip joint osteoarthritis. The FBN2 mutation was identified in an initial set of 23 dogs and in additional Labrador Retrievers and dogs of other breeds. Only 19 of the 257 dogs in which the QTL for the DI was originally identified14 were later subjected to FBN2 sequencing. The authors made the a priori decision not to include crossbred dogs from our colony in the analysis. We reasoned that testing the mutation in unrelated purebred dogs, most of which were not included in prior linkage analyses, would strengthen support for any association discovered. In hindsight, it is probably wise to screen for segregation of a candidate gene mutation in the subset of the individuals originally used in the linkage study prior to validation of a mutation discovery in a new population. The AGC allele had recessive characteristics because the heterozygous and homozygous GAT individuals were similar phenotypically. Clearly, other genes must contribute to CHD because the FBN2 locus does not explain all the genetic trait variation in CHD. This is consistent with the results of QTL mapping in Labrador Retriever–Greyhound crossbred dogs,19 Labrador Retrievers,22 German Shepherd Dogs,2 and Portuguese Water Dogs,27 in which no locus for CHD was identified that contributed <18% of trait variance.

Expression of FBN2 in the joint capsule was related inversely with the DLS score; that is, as the right DLS score decreased (a worse hip joint conformation), the FBN2 expression increased in the right hip joint capsule. It was expected that FBN2 expression would be lower in dogs with lower DLS scores because the dogs that were homozygous for the deletion (AGC) haplotype had significantly lower FBN2 expression in their femoral head articular cartilage. One explanation is that the experiments were confounded by the presence of secondary osteoarthritis in dysplastic hip joints. The presence of osteoarthritis was used as the definitive criterion for a dysplastic hip joint. The perifoveal area is the location where the incipient gross alterations in femoral head articular cartilage are observed. This tissue was deliberately excluded from analysis. The expression of the genes studied may not change as rapidly with incipient osteoarthritis in the chondrocytes in dorsal articular cartilage as in the fibroblasts in joint capsules.

FBN2 mRNA expression was examined in the hip joint tissues of the dogs when they reached approximate skeletal maturity (8 months of age). It is not known whether FBN2 is differentially expressed during canine joint development. In mice, FBN2 expression occurs early in development in the perichondrium, periosteum, and acetabulum; however, the joint capsule and round ligament were not included in that analysis.28 In mice, FBN1 expression occurs later in development but its expression remains high. Most of the tissue alterations that result in a dysplastic hip joint occur during the period of rapid growth, which for a large-breed dog like a Labrador Retriever slows at approximately 6 months of age, and more rapidly growing dogs develop a more severe form of CHD.29,30 Therefore, examination of FBN2 expression during the growth period is warranted. This would require biopsy of hip joint tissues of growing dogs of differing susceptibility to CHD and with known FBN2 genotype, then confirming the diagnosis at maturity. Finally, mRNA concentrations from a gene and the concentration of the expressed protein are not always strongly correlated.31,32 Protein studies may be required to reach any firm conclusions about the role of FBN2 in hip joint development and CHD. The analysis of tissues from only 14 dogs and the crossbreed genetic background of these dogs, which contained up to one-quarter Greyhound genes, could have influenced the results.

The fibrillins belong to a family of structurally related glycoproteins that include the latent transforming growth factor-β–binding proteins and fibulins.10 Fibrillins are the major structural components of extracellular microfibrils, which are either associated with elastin or act as elastin-free assemblies.33 Mutations in the gene encoding FBN2 have been associated with human congenital contractural arachnodactyly, a disorder characterized by arachnodactyly (elongated distal long bones), dolichosternomelia (elongated limbs), scoliosis, multiple congenital contractures, and abnormalities of the external portion of the ears.30 A similar phenotype was produced in mice through a deletion of the FBN2 gene.34 Most of the mutations discovered in FBN1 and FBN2 occur in the same region (ie, upstream or downstream of exon 30) as for the mutation discovered in dogs with CHD in the present study. The CHD phenotype is genetically complex, and the effect of any FBN2 haplotype would not be expected to be as severe as the FBN2 knockout phenotype in the mouse. The joint contractures in humans with congenital contractural arachnodactyly disappear with aging.10 Mice heterozygous for the FBN2 mutation do not have an abnormal phenotype unless they are also heterozygous for a mutation in bone morphogenetic protein 7.35 Some human patients with congenital contractural arachnodactyly have joint laxity including hypermobile thumbs.11 However, the mechanism by which such an intron deletion in FBN2 may contribute to the CHD phenotype remains unclear.

Fibrillins are present in the chondroepiphysis of the developing joint.36,37 Because congenital contractural arachnodactyly is characterized by deformation of the pinnae, Yanagino et al38 examined the regulation of FBN2 in chondrocytic differentiation of murine ATDC5 cells. The FBN2 was highly expressed in the early stage of differentiation and later declined. The 5′ flanking region of FBN2 contains potential binding sites for the E2 factor (F1) family, Runx, AP-2, and Sox transcription factors. Overexpression of E2F1 results in a marked increase in FBN2 promoter activity. Because the dogs homozygous for the FBN2 deletion mutation had lower FBN2 expression in femoral head cartilage, examination of FBN2 expression in chondroepiphyseal cartilage of the developing hip joint might further elucidate the functional importance of the FBN2 mutation. However, biopsy of femoral chondroepiphyseal tissue during development would damage the hip joint and result in osteoarthritis.

The mechanism by which altered concentrations of FBN2 protein might interrupt the mechanical integrity of the hip joint capsule or femoral head articular cartilage contributing to hip joint laxity and dysplasia remains unknown. Even small changes in gene expression may influence complex behavioral traits, such as geotaxis in Drosophila spp.39 The compliance of the joint capsule and round ligament of the femoral head relies partly on its elastic tissue content and distribution, and fibrillins are a component of the elastic tissue. In the fibrous hip joint capsule of monkeys, elastic tissue predominates caudodorsally and ventrally, compared with craniodorsally.40 So-called ball and socket joints such as the hip joint and shoulder joint require pliable and extensible capsular support devoid of strong collateral ligaments. Electron microscopic images of the hip joint capsule of an 8-month-old Labrador Retriever with a lax (high DI) joint capsule revealed that the cross-sectional diameter of the collagen fibrils was heterogeneous.41 The cross-sectional collagen fibrillar diameter in the hip joint capsule of a Labrador Retriever without CHD and with tight hip joints was uniform. Flexor digitorum longus tendons from mice null for FBN2 have the same collagen content as control mice but contain fewer intermolecular collagen cross-links of hydroxylysyl and lysyl pyridinoline.36 The collagen of femoral heads of dysplastic Labrador Retrievers is more soluble than the collagen of femoral heads of unaffected puppies. Electron micrographs of articular cartilage of the same femoral heads reveal disruption of collagen fibrils in young dogs with CHD.42–44 The etiology of this disruption, which was discovered through an electron microscopic and biochemical evaluation decades ago, has never been elucidated, but a plausible explanation may now be possible to achieve through positional cloning of contributing genes.

Understanding the genetic basis of CHD will help identify similar biochemical pathways that lead to developmental dysplasia of the human hip joint, for which there also is evidence of a genetic basis. Previous studies45,46 that examined the underlying biochemical mechanism for hip joint laxity in humans examined the capsular collagens. Increased collagen type III-to-type I ratios in hip joint capsules of dysplastic human neonates have been reported.46,47 Similar findings of increased ratios of type III to type I collagen propeptides in synovial fluid have been reported in the hip joints of dysplastic dogs, compared with unaffected dogs.48 The action of the maternal hormones estrogen, relaxin, and testosterone on genetically susceptible hip joint tissues may contribute to hip joint laxity in the perinatal period,49 but whether this effect persists is unknown. Fibrillin 2 is the preferred target for relaxin activity and expression, compared with fibrillin 1, as measured in human dermal fibroblasts and murine fetal skin.50 Because extracellular matrix proteins, such as the fibrillins, sequester TGF-β-signaling molecules, the effect of fibrillin concentration may be indirect. The location of TGF-β molecules affects the function of surrounding cells and the entire extracellular matrix.10 Impaired microfibril assembly can result in excess local TGF-β signaling,10 which participates in the pathogenesis of ruptured aortic aneurysm in Marfan syndrome, a result of an FBN1 or TGF-β signaling mutation.

Other genes in linkage disequilibrium with FBN2 may contribute to the QTL on CFA11 at 18.5 to 21 cM. We excluded a chondroitin sulfate synthase gene (CSS3) and a disintegrin and metalloproteinase with thrombospondin motifs gene (ADAMTS19, a member of the aggrecanase family in this interval) as candidates for CHD via direct sequencing in affected and unaffected Labrador Retrievers and found no mutations. Another gene warranting sequencing in the interval includes solute carrier family 12 member 2, a gene associated with loin muscle mass in pigs. Greyhounds, a breed resistant to CHD, have higher muscle-to-bone mass ratios than do Labrador Retrievers, possibly implicating the loin muscle mass gene in the pathogenesis of CHD.51

Dysplastic hip joints as measured by the DLS score and osteoarthritis had significantly higher FBN2 expression in hip joint capsules, compared with unaffected joints. The KIN11 expression was also increased in the joints with osteoarthritis, compared with expression in the unaffected joints, but expression of TOP1 and FBN1 was unaffected. The most dysplastic joints were in the incipient stages of osteoarthritis with peri-foveal articular cartilage fibrillation and synovitis. Gene expression is globally increased in the early stages of canine osteoarthritis,52 so it was not surprising that at least one of the genes chosen as a control gene had high mRNA expression in joints with osteoarthritis. KIN11 protein is a major component of intracellular machinery.53 Gross evidence of osteoarthritis was chosen in the present study as the traditional indicator of concomitant CHD because there is no precisely accurate method to assess functional or dynamic subluxation of a dysplastic hip joint that results in osteoarthritis. The DLS score alone has approximately 80% accuracy for predicting osteoarthritis of the hip joint as measured via combined sensitivity and specificity18 All the radiographic methods of hip joint assessment are performed with the dog passively positioned and recumbent, and radiographs do not reveal the incipient stages of osteoarthritis in the hip joint. Besides those genes that influence the expression of CHD, other genes likely influence the onset and progression of hip joint osteoarthritis in dysplastic hip joints.

The mapping and identification of FBN2 as the first gene reported to be associated with CHD indicate that complex traits such as CHD can be evaluated in dogs at the molecular genetic level. The strategy of coarse genome screening with microsatellites or SNPs followed by fine mapping with SNPs can resolve a QTL to a small interval suitable for candidate gene screening.54 The association between the FBN2 deletion haplotype and CHD was detected in Labrador Retrievers and non–Labrador Retriever breeds, indicating that some of the genes that contributed to CHD were conserved across breeds. Further studies are needed to elucidate the biochemical pathway through which mutations in FBN2 might influence in the developmental pathogenesis of CHD.

ABBREVIATIONS

CHD

Canine hip dysplasia

DI

Distraction index

DLS

Dorsolateral subluxation

EHR

Extended-hip radiograph

NA

Norberg angle

QTL

Quantitative trait locus

SNP

Single-nucleotide polymorphism

a.

Ensembl [database online]. Cambridge, England: European Bioinformatics Institute and Wellcome Trust Sanger Institute. Available at: www.ensembl.org/Canis_familiaris/index.html. Accessed May 24, 2009.

b.

Entrez [database online]. Bethesda, Md: National Center for Biotechnology Information. Available at: www.ncbi.nlm.nih.gov/mapview/map_search.cgi?taxid#9615. Accessed May 24, 2009.

c.

UCSC Genome Bioinformatics [database online]. Santa Cruz, Calif: University of California-Santa Cruz. Available at: genome.ucsc.edu. Accessed May 24, 2009.

d.

Integrated DNA Technologies, Coralville, Iowa.

e.

Wizard SV 96, Promega Corp, Madison, Wis.

f.

United States Biochemical Corp, Cleveland, Ohio.

g.

Applied Biosystems, Foster City, Calif.

h.

Gene Codes Corp, Ann Arbor, Mich.

i.

Ambion, Austin, Tex.

j.

Superscript III, Invitrogen, Carlsbad, Calif.

k.

Dako, Carpinteria, Calif.

l.

Zymed Laboratories, Carlsbad, Calif.

m.

Abcam, Cambridge, Mass.

n.

Vector Laboratories Inc, Burlingame, Calif.

o.

Fisher Scientific Research, Waltham, Mass.

References

  • 1.

    Lust G. An overview of the pathogenesis of canine hip dysplasia. J Am Vet Med Assoc 1997; 210:14431445.

  • 2.

    Marschall Y, Distl O. Mapping quantitative trait loci for canine hip dysplasia in German Shepherd Dogs. Mamm Genome 2007; 18:861870.

  • 3.

    Kaneene JB, Mostosky UV, Padgett GA. Retrospective cohort study of changes in hip joint phenotype of dogs in the United States. J Am Vet Med Assoc 1997; 211:15421544.

    • Search Google Scholar
    • Export Citation
  • 4.

    Olsewski JM, Lust G, Rendano VT, et al. Degenerative joint disease: multiple joint involvement in young and mature dogs. Am J Vet Res 1983; 44:13001308.

    • Search Google Scholar
    • Export Citation
  • 5.

    Carr AJ, Jefferson RJ, Benson MK. Joint laxity and hip rotation in normal children and in those with congenital dislocation of the hip. J Bone Joint Surg Br 1993; 75:7678.

    • Search Google Scholar
    • Export Citation
  • 6.

    Lust G, Beilman WT, Rendano VT. A relationship between degree of laxity and synovial fluid volume in coxofemoral joints of dogs predisposed for hip dysplasia. Am J Vet Res 1980; 41:5560.

    • Search Google Scholar
    • Export Citation
  • 7.

    Farese JP, Todhunter RJ, Lust G, et al. Dorsolateral subluxation of hip joints in dogs measured in a weight-bearing position with radiography and computed tomography. Vet Surg 1998; 27:393405.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 8.

    Smith GK. Advances in diagnosing canine hip dysplasia. J Am Vet Med Assoc 1997; 210:14511457.

  • 9.

    Zhu L, Zhang Z, Feng F, et al. Single nucleotide polymorphisms refine QTL intervals for hip joint laxity in dogs. Anim Genet 2008; 39:141146.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 10.

    Ramirez F, Dietz HC. Fibrillin-rich microfibrils: structural determinants of morphogenetic and homeostatic events. J Cell Physiol 2007; 213:326330.

  • 11.

    Gupta PA, Putnam EA, Carmical SG, et al. Ten novel FBN2 mutations in congenital contractural arachnodactyly: delineation of the molecular pathogenesis and clinical phenotype. Hum Mutat 2002; 19:3948.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 12.

    Todhunter RJ, Acland GM, Olivier M, et al. An outcrossed canine pedigree for linkage analysis of hip dysplasia. J Hered 1999; 90:8392.

  • 13.

    Zhang Z, Zhu L, Sandler J, et al. Estimation of heritabilities, genetic correlations, and breeding values of four traits that collectively define hip dysplasia in dogs. Am J Vet Res 2009; 70:483492.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 14.

    Gustafsson PO, Olsson SE, Kasstrom H, et al. Skeletal development of Greyhounds, German Shepherd Dogs and their crossbreed offspring. Acta Radiol Suppl 1975; 344:81108.

    • Search Google Scholar
    • Export Citation
  • 15.

    Smith GK, LaFond E, Gregor TP, et al. Within- and between-examiner repeatability of distraction indices of the hip joints in dogs. Am J Vet Res 1997; 58:10761077.

    • Search Google Scholar
    • Export Citation
  • 16.

    Lust G, Williams AJ, Burton-Wurster N, et al. Joint laxity and its association with hip dysplasia in Labrador Retrievers. Am J Vet Res 1993; 54:19901999.

    • Search Google Scholar
    • Export Citation
  • 17.

    Lust G, Todhunter RJ, Erb HN, et al. Repeatability of dorsolateral subluxation scores in dogs and correlation with macroscopic appearance of hip osteoarthritis. Am J Vet Res 2001; 62:17111715.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 18.

    Lust G, Todhunter RJ, Erb HN, et al. Comparison of three radio-graphic methods for diagnosis of hip dysplasia in eight-month-old dogs. J Am Vet Med Assoc 2001; 219:12421246.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 19.

    Todhunter RJ, Mateescu R, Lust G, et al. Quantitative trait loci for hip dysplasia in a cross-breed canine pedigree. Mamm Genome 2005; 16:720730.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 20.

    Clark LA, Tsai KL, Steiner JM, et al. Chromosome-specific microsatellite multiplex sets for linkage studies in the domestic dog. Genomics 2004; 84:550554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 21.

    Mateescu RG, Burton-Wurster NI, Tsai K, et al. Identification of quantitative trait loci for osteoarthritis of hip joints in dogs. Am J Vet Res 2008; 69:12941300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 22.

    Phavaphutanon J, Mateescu RG, Tsai KL, et al. Evaluation of quantitative trait loci for hip dysplasia in Labrador Retrievers. Am J Vet Res 2009; 70:10941101.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 23.

    Heath SC. Markov chain Monte Carlo segregation and linkage analysis for oligogenic models. Am J Hum Genet 1997; 61:748760.

  • 24.

    Rozen S, Skaletsky H. Primer 3 on the WWW for general users and for biologist programmers. Methods Mol Biol 2000; 132:365386.

  • 25.

    Zhu L, Zhang Z, Friedenberg S, et al. The long (and winding) road to gene discovery for canine hip dysplasia. Vet J 2009; 181:97110.

  • 26.

    Awano T, Katz ML, O'Brien DP, et al. A frame shift mutation in canine TPP1 (the ortholog of human CLN2) in a juvenile Dachshund with neuronal ceroid lipofuscinosis. Mol Genet Metab 2006; 89:254260.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 27.

    Chase K, Lawler DF, Adler FR, et al. Bilaterally asymmetric effects of quantitative trait loci (QTLs): QTLs that affect laxity in the right versus left coxofemoral (hip) joints of the dog (Canis familiaris). Am J Med Genet A 2004; 124A:239247.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 28.

    Putnam EA, Zhang H, Ramirez F, et al. Fibrillin-2 (FBN2) mutations result in the Marfan-like disorder, congenital contractural arachnodactyly. Nat Genet 1995; 11:456458.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 29.

    Kealy RD, Lawler DF, Ballam JM, et al. Evaluation of the effect of limited food consumption on radiographic evidence of osteoarthritis in dogs. J Am Vet Med Assoc 2000; 217:16781680.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 30.

    Vanden Berg-Foels WS, Todhunter RJ, Schwager SJ, et al. Effect of early postnatal body weight on femoral head ossification onset and hip osteoarthritis in a canine model of developmental dysplasia of the hip. Pediatr Res 2006; 60:549554.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 31.

    Jansen RC, Nap JP, Mlynárová L. Errors in genomics and proteomics. Nat Biotechnol 2002; 20:19.

  • 32.

    Pascal LE, True LD, Campbell DS, et al. Correlation of mRNA and protein levels: cell type-specific gene expression of cluster designation antigens in the prostate. BMC Genomics 2008; 9:246.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 33.

    Kielty CM, Baldock C, Lee D, et al. Fibrillin: from microfibril assembly to biomechanical function. Philos Trans R Soc Lond B Biol Sci 2002; 357:207217.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 34.

    Chaudhry SS, Gazzard J, Baldock C, et al. Mutation of the gene encoding fibrillin-2 results in syndactyly in mice. Hum Mol Genet 2001; 10:835843.

  • 35.

    Arteaga-Solis E, Gayraud B, Lee SY, et al. Regulation of limb patterning by extracellular microfibrils. J Cell Biol 2001; 154:275281.

  • 36.

    Boregowda R, Paul E, White J, et al. Bone and soft connective tissue alterations result from loss of fibrillin-2 expression. Matrix Biol 2008; 27:661666.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 37.

    Hurle JM, Corson G, Daniels K, et al. Elastin exhibits a distinctive temporal and spatial pattern of distribution in the developing chick limb in association with the establishment of the cartilaginous skeleton. J Cell Sci 1994; 107:26232634.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 38.

    Yanagino T, Yuasa K, Nagahama M, et al. Transcriptional regulation of fibrillin-2 gene by E2F family members in chondrocyte differentiation. J Cell Biochem 2009; 106:580588.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 39.

    Toma DP, White KP, Hirsch J, et al. Identification of genes involved in Drosophila melanogaster geotaxis, a complex behavioral trait. Nat Genet 2002; 31:349353.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 40.

    Walji AH. Some histological aspects of the hip joint capsule in the vervet monkey. J Morphol 1988; 197:327335.

  • 41.

    Todhunter RJ, Lust G. Canine hip dysplasia: pathogenesis. In: Slatter D, ed. Textbook of small animal surgery. 3rd ed. St Louis: WB Saunders Co, 2003;20092019.

    • Search Google Scholar
    • Export Citation
  • 42.

    Greisen HA, Summers BA, Lust G. Ultrastructure of the articular cartilage and synovium in the early stages of degenerative joint disease in canine hip joints. Am J Vet Res 1982; 43:19631971.

    • Search Google Scholar
    • Export Citation
  • 43.

    Hui-Chou CS, Lust G. The type of collagen made by the articular cartilage in joints of dogs with degenerative joint disease. Coll Relat Res 1982; 2:245256.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 44.

    Wiltberger H, Lust G. Ultrastructure of canine articular cartilage: comparison of normal and degenerative (osteoarthritic) hip joints. Am J Vet Res 1975; 36:727740.

    • Search Google Scholar
    • Export Citation
  • 45.

    Oda H, Igarashi M, Hayashi Y, et al. Soft tissue collagen in congenital dislocation of the hip. Biochemical studies of the ligamentum teres of the femur and the hip joint capsule. Nippon Seikeigeka Gakkai Zasshi 1984; 58:331338.

    • Search Google Scholar
    • Export Citation
  • 46.

    Jensen BA, Reimann I, Fredensborg N. Collagen type III predominance in newborns with congenital dislocation of the hip. Acta Orthop Scand 1986; 57:362365.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 47.

    Wang EB, Zhao Q, Li LY, et al. Expression of COL1a1 and COL3a1 in the capsule of children with developmental dislocation of the hip [in Chinese]. Zhongguo Dang Dai Er Ke Za Zhi 2008; 10:493496.

    • Search Google Scholar
    • Export Citation
  • 48.

    Madsen JS. The joint capsule and joint laxity in dogs with hip dysplasia. J Am Vet Med Assoc 1997; 210:14631465.

  • 49.

    Steinetz BG, Williams AJ, Lust G, et al. Transmission of relaxin and estrogens to suckling canine pups via milk and possible association with hip joint laxity. Am J Vet Res 2008; 69:5967.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 50.

    Samuel CS, Sakai LY, Amento EP. Relaxin regulates fibrillin 2, but not fibrillin 1, mRNA and protein expression by human dermal fibroblasts and murine fetal skin. Arch Biochem Biophys 2003; 411:4755.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 51.

    Cardinet GH III, Guffy MM, Wallace LJ, et al. Canine hip dysplasia in German Shepherd Dog-Greyhound crossbreeds. J Am Vet Med Assoc 1983; 182:393395.

    • Search Google Scholar
    • Export Citation
  • 52.

    Matyas JR, Adams ME, Huang D, et al. Major role of collagen IIB in the elevation of total type II procollagen messenger RNA in the hypertrophic phase of experimental osteoarthritis. Arthritis Rheum 1997; 40:10461049.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 53.

    Hirokawa N, Noda Y. Intracellular transport and kinesin super-family proteins, KIFs: structure, function, and dynamics. Physiol Rev 2008; 88:10891118.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • 54.

    Sutter NB, Bustamante CD, Chase K, et al. A single IGF1 allele is a major determinant of small size in dogs. Science 2007; 316:112115.

Appendix 1

Primer sequences for amplification of FBN2, FBN1, and GPI (glucose phosphate isomerase) mRNA from hip joint capsule tissue of dogs.

GeneLocationForward primersReverse primers
FBN2Exon 4–5CAGTGCCAGAAGGGATATATTGGAACGTCCACCATTCTGACATCCATTT
FBN2Exon 30–31TGGCTTCATGGCTTCAATGGACTCACATTCCCCAAACATACAGATGT
FBN2Exon 59–60GCTATGTCCTACAAGAGGATGCAAAGCAGTTGTGTTGTTTGGTTTGACA
GPIExon 2–3ACACCGTGATGCAGATGCTTGATCTTCTCGCCACTGAACATG
FBN1Exon 6–7GCCGAGGCTTCATTCCAAATATCGGGATGGCCTGGCATTCTA

Appendix 2

FBN2 primer sequences for exon amplification to test for alternative transcripts in a study of hip dysplasia in dogs.

Primer locationDirectionSequenceExpected amplicon size (bp)
Exon 26Forwardaggctacgaaagtgggttca
Exon 28Reversettcatagctgccttcggaat364
Exon 29Reversegtacattggccaccatcaca468
Exon 30Reversegttctcacattccccaaaca598
Exon 31Reversegtgggcaccaatttcacact697
Exon 33Reversettgatgttttctgcgcactc942
Exon 34Forwardgtgcatatcgttgcgagtgt
Exon 36Reverseacacagaggccattcacaca235
Exon 37Reversecgaggtccgaacttcaggta349
Exon 39Reversecctggtaactcctggcattc568
Exon 40Reverseatacaccagggtgtgcgaaa699
Exon 41Reversecgttctcacaagtggttccat846
Exon 42Reverseggatgtcaaaggtgaatccag996
Exon 43Reverseaccaggagcaggtcgttgta1,132

= Not applicable.

Appendix 3

Haplotype structure of intron 30 of FBN2 where a deletion mutation occurred, in dogs with CHD.

Haplotype structureSequence SequenceSequence Sequence SequenceSequenceSequenceSequence
Unaffected homozygous… CATCTGTGCAC …GGGAA …AGAGGCTTTTTCAGT …AGATTATACTTAACTTTTA …
Affected homozygous… CATCTATGCAC …GGGAA …GGAGGCTTTCTCAGT …AGATTA……ACTTTTA …
Heterozygous… CATCTRTGCAC …TGCAC …RGAGGCTTTYTCAGT …AGATTANNNNNACTTTTA …
Base position 41  68 77  189–195 

N = Unknown nucleotide sequence.

Contributor Notes

Address correspondence to Dr. Todhunter (rjt2@cornell.edu).

Supported by Collaborative Research Grant program in the College of Veterinary Medicine, Cornell Biotechnology, the Morris Animal Foundation, the NHLBI Mammalian Genotyping Service, and NIH grant No. 1 R21 AR055228–01A1.

Presented as a poster at the Annual Meeting of the Morris Animal Foundation, Boulder, Colo, June 2007.

The authors thank Alma Jo Williams for technical assistance and Dr. Jody Sandler for contribution of hip joint radiographs and blood samples.