Journal of Heredity Advance Access originally published online on July 13, 2005
Journal of Heredity 2005 96(7):739-744; doi:10.1093/jhered/esi068
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Evaluation of Canine COL4A3 and COL4A4 as Candidates for Familial Renal Disease in the Norwegian Elkhound
From the Department of Clinical Sciences of Companion Animals, Faculty of Veterinary Medicine, Utrecht University, 3508 TD Utrecht, The Netherlands (Wiersma and van Dongen); Veterinary Genetics Laboratory, University of California, Davis, CA 95616 (Millon); Department of Animals, Science and Society, Faculty of Veterinary Medicine, Utrecht University, 3508 TD Utrecht, The Netherlands (van Oost); and Department of Population Health and Reproduction, School of Veterinary Medicine, University of California, Davis, CA 95616 (Bannasch)
Address correspondence to Bernard A. van Oost at the address above, or e-mail: b.vanoost{at}vet.uu.nl.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The collagen type IV
3 and 4 chains (COL4A3 and COL4A4) are part of the specialized glomerular basement membrane in the kidney. In human these genes are responsible for Alport syndrome (a type of hereditary nephritis). Histopathological similarities between kidneys of Norwegian elkhound dogs affected with familial renal disease and human Alport syndrome were the basis for a candidate gene approach in Norwegian elkhounds. Three microsatellites—tightly linked to canine COL4A3 and COL4A4—were developed. The microsatellites were used to analyze linkage between COL4A3 and COL4A4 and familial renal disease in a Norwegian elkhound pedigree segregating this disease. Presence of one recombinant between familial renal disease and COL4A3/COL4A4 suggests that these genes are not likely candidates for familial renal disease in this breed.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The collagen type IV
3 and 4 chains (COL4A3 and COL4A4) are expressed in the renal glomerular basement membrane, where they provide a critical structural and functional matrix for other basement membrane components (Leinonen et al. 1994; Lemmink et al. 1997). In human these genes have shown to cause autosomal Alport syndrome, a form of hereditary nephritis (Lemmink et al. 1997; Van der Loop et al. 2000). In dogs, these genes have been suggested as candidates for similar hereditary nephropathies in several breeds, for example, in the English cocker spaniel (Lees et al. 1997) and bull terrier (Hood et al. 1995, 2002).
Hereditary renal diseases have been described in many different dog breeds. Hereditary nephritis is a structural (cf. functional) kidney anomaly that has been described in more than 10 breeds and has been most extensively studied in the Samoyed and a Navasota mongrel family (X-linked Alport syndrome), English cocker spaniel, and bull terrier. A single base pair mutation (Zheng et al. 1994) in exon 35 of COL4A5, which results in a premature stop codon, causes X-linked Alport syndrome in the Samoyed. This causes early onset renal disease that rapidly progresses to renal failure and death (at 8–10 months). In a Navasota mongrel dog family, X-linked Alport syndrome is caused by a 10-bp deletion in exon 9 of COL4A5 that results in a frameshift and premature stopcodon in exon 10 (Cox et al. 2003). Familial nephropathy in the English cocker spaniel is inherited as an autosomal recessive disease. Immunohistochemistry showed the type IV collagen chains 3 and 4 to be completely absent in the kidney (Lees et al. 1998). Lees et al. (1997) hypothesized that this renal disease might be caused by mutations in COL4A3 or COL4A4. Hereditary nephritis in the bull terrier is an autosomal dominant disease that causes chronic renal failure (Hood et al. 1991). Although the 3 and 4 type IV collagen chains seemed to be normally present in the kidneys (Hood et al. 2000), involvement of COL4A3 and COL4A4 has been suggested (Hood et al. 1995, 2002).
In addition to these well-studied canine models of Alport syndrome, inherited renal diseases are also found in other dog breeds, with clinical signs and some histopathological changes similar to the ones found in human Alport syndrome. In the Norwegian elkhound, familial renal disease was described in the 1970s by Finco (Finco 1973, 1976; Finco et al. 1970, 1977). It causes chronic renal failure with an age of onset of 3 months to 4 years. It was detected in both male and female dogs, but no mode of inheritance was established. On kidney biopsy, periglomerular fibrosis was a key finding. Collagen type IV presence in the kidney has never been studied in familial renal disease. Similarities in clinical appearance and histology between Norwegian elkhounds with familial renal disease and Samoyed X-linked Alport syndrome, English cocker spaniel familial nephropathy, and bull terrier hereditary nephritis indicated a common etiology for these canine renal diseases. In this study we investigated COL4A3 and COL4A4 as candidates for familial renal disease in the Norwegian elkhound.
Recently, we published the complementary DNA (cDNA) sequence and the genomic organization of canine COL4A3 and COL4A4 and their map position on Canis familiaris chromosome (CFA) 25: REN05E03–C25.213–COL4A4–COL4A3–REN61G15 (Wiersma et al. in press). We present here three polymorphic microsatellites tightly linked to COL4A3 and COL4A4. These markers provide a powerful tool for the evaluation of these genes as candidates for canine hereditary nephritis (Alport syndrome). In this article, these polymorphisms are used in a linkage analysis study of a Norwegian elkhound family that segregates familial renal disease.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Isolation of COL4A3 and COL4A4 Positive BAC Clones
Probes for screening the canine RPCI-81 BAC (bacterial artificial chromosome) library (Li et al. 1999) for COL4A3 and COL4A4 were derived from polymerase chain reaction (PCR) amplification of cDNA of kidney tissue from a 6-week-old male Dalmatian puppy that was euthanized for other health reasons. Primer sets for clone A and B (Table 1) were used to amplify cDNA of COL4A3 and COL4A4, respectively (based on AY263362 [GenBank] for COL4A3 and AY263363 [GenBank] for COL4A4 from GenBank, at www.ncbi.nlm.nih.gov), on a Peltier Thermal Cycler (MJ Research). Eighty nanograms of purified PCR product (Qiaquick Gel Extraction Kit, Qiagen) was labeled with 50 µCi dCTP-32P (Klenow Multiprime DNA labeling systems, Amersham) and used as a probe for screening BAC filters. Positive BACs were identified and DNA was isolated (Qiagen Large Construct Kit). The approximate number of microsatellites in each BAC was estimated by hybridizing Southern blots with
ATP-32P-labeled (CA)18 and (GAAA)6 oligonucleotide probes (Polynucleotide Kinase reaction, New England Biolabs).
|
Development of Microsatellite Markers
BAC clones were subcloned by partial digestion with Sau 3AI to obtain inserts 300–700 bp in length and then cloned into the BamH1 site of pBS SKII (Stratagene). Transformants were screened by hybridization with the CA and GAAA repeat probes. Positive clones were sequenced. Primers were designed from sequence flanking the repeats and the forward primer was fluorescently (6FAM) labeled. Genotyping PCR reaction conditions were 94°C 12 min, 35x (94°C 10 s, Ta° 15 s, 72°C 30 s), 72°C 20 min (specific annealing temperatures [Ta] in Table 1). An ABI 3100 instrument was used for genotyping and allele sizes were determined by Strand software (Hughes 1998). For mapping purposes the three microsatellites—a3-CA1, a3-GAAA1, and a4-CA2 (Table 1)—were genotyped on the 3000 rad Canine/Hamster RH08 Radiation Hybrid (RH) panel (Thomas et al. 2001) from Invitrogen (Carlsbad, CA).
Linkage Analysis
Linkage analysis was performed using the MLINK (Terwillliger and Ott 1994) software (accessed through HGMP-RC online at www.rfcgr.mrc.ac.uk) with the disease modeled as an autosomal dominant trait with an age-dependent penetrance. Dogs with no signs of renal disease were classified into one of five liability classes; class 1 (age at which last tested 0–14 months) with a penetrance of familial renal disease of 20%, class 2 (15–36 months; 40%), class 3 (37–57 months; 60%), class 4 (58–79 months, 80%), and class 5 (
80 months, 99%).
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Segregation of the Disease
The reference family was developed from a Dutch pedigree of Norwegian elkhounds and contains four litters with one to four siblings each (Figure 1). Phenotypes were established by blood and urine screenings for signs of chronic renal failure (blood creatinin > 1.2 times the body weight [kg] + 50 µmol/L and specific gravity urine < 1.020). DNA and disease status were available for 15 dogs, and 1 remaining dog (A) died of old age without any clinical signs of renal failure. The dogs had been examined for signs of chronic renal failure at an average age of 55 months (13–151 months, Figure 1). Of the 15 available dogs (5 males and 10 females), 7 were diagnosed with chronic renal failure (3 males, 4 females). Familial renal disease seems to inherit as an autosomal trait in our Norwegian elkhound family. X-linked inheritance is unlikely because of male-to-male transmission of the disease (observed once in Figure 1). The hypothesized mode of inheritance that fits this pedigree best is an autosomal dominant (monogenic) one. This is firstly supported by the fact that every descendant (n = 6) in whom we diagnosed familial renal disease has at least one affected parent. Second, the disease does not skip any generations (observing three consecutive generations). Additional dogs of this family, of which no DNA sample was available, also supported the autosomal dominant hypothesis (data not shown).
|
Microsatellite Development
Microsatellites closely linked to the two collagen type IV genes were obtained by subcloning BAC clones that were identified using partial cDNA probes. Screening one filter of the canine BAC library per gene identified two positive BACs for each gene. The BAC coordinates for COL4A3 are RP81-8P2 and -19N22 and for COL4A4-62P11 and -83C14 from the RPCI canine BAC library 81. Genotyping primers were developed for one CA and one GAAA repeat for COL4A3 and one CA repeat for COL4A4. Primer sequences for the microsatellites markers, the BAC clones that these microsatellites were derived from and the annealing temperatures used for genotyping are listed in Table 1. RH mapping located all three microsatellites on CFA25 next to gene-specific PCRs (Table 2). The LOD scores and the estimated distances between the canine COL4A3 and COL4A4 microsatellites and gene-specific sequence tagged sites (STSs) indicated that both of these genes and their microsatellites are tightly clustered (Table 2). To characterize the informativeness of our microsatellites, polymorphism information content (PIC) and heterozygosity values were obtained based on the genotypes of 23 individuals from different breeds (Table 1).
|
Linkage Analysis
The reference family of Norwegian elkhounds in which renal disease segregates was typed for the polymorphic COL4A3 and COL4A4 microsatellites are shown in Figure 1. Microsatellites a3-CA1 and a3-GAAA1 were informative in this pedigree, and a4-CA2 was partially informative. No recombination was seen between these microsatellites. To estimate the penetrance of familial renal disease in this Norwegian elkhound family, a diagram was made (Figure 2) of the age at which chronic renal failure was detected in nine members (seven of which are included in Figure 1). With a linear trend line, five liability classes were established, based on the age at which each dog was last tested. Of the dogs without signs of renal disease, dog C, E, and F were assigned to liability class 5, dog K to class 3, dog P and Q to class 2, dog L to class 1, and dog M to class 4. Haplotypes were composed based on all three used markers (a3-CA1/a3-GAAA1/a4-CA2). Five unique chromosomes could be distinguished in this family (Figure 1). If the initial hypothesis is correct that familial renal disease is linked to COL4A3 or COL4A4, all Norwegian elkhounds with familial renal disease would be expected to have haplotype "151/199/162," and it is highly unlikely that unaffected Norwegian elkhounds should have this. However, one dog in this small reference family did not follow this model: dog K had familial renal disease but not the expected haplotype. This suggests that neither COL4A3 nor COL4A4 is closely linked to familial renal disease. The lowest LOD score that was obtained was –0.069 (at
= 0.0).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Available Pedigree
The Norwegian elkhound pedigree (Figure 1) available for this study was small; the complete litters are actually larger, but DNA samples were not available of all dogs. Furthermore, a missing parent (dog A) in the first generation limits the number of informative meiosis. Although an autosomal dominant transmission of familial renal disease still needs to be proved statistically, linkage analysis was performed with the disease treated as an autosomal dominant trait.
Tightly Linked Microsatellites
The three microsatellites for COL4A3 and COL4A4 (a3-CA1, a3-GAAA1, and a4-CA2) were obtained from BACs containing these canine genes. RH mapping indicated correct choice of the BACs by mapping the microsatellites on CFA25 next to the gene-specific PCRs. Blast analysis of all used primer sequences against the dog genome (at www.ensembl.org) showed the following order: REN05E03–microsat a4-CA2–COL4A4–COL4A3–microsat a3-CA1–microsat a3-GAAA1–C25.213–REN61G15.
PIC and heterozygosity values were determined to describe the informativeness of these microsatellites. A PIC larger than 0.50 is considered highly informative (Botstein et al. 1980). Tested in individuals from multiple breeds, the three developed microsatellites were highly informative and highly polymorphic with PIC values greater than 0.67. Within a breed, these markers have been shown to be informative in at least Norwegian elkhounds.
When checking parentage in our family, a new allele (187 bp) was seen in one dog (F) for microsatellite a3-GAAA1, being 4 bp smaller than parent dog B (191 bp). This was explained by the high polymorphism rate of this microsatellite (PIC = 0.94), because parentage of dog F was confirmed based on genotypes of other microsatellites from this study and several unlinked microsatellites (data not shown).
Linkage Analysis
No recombination was seen between the used COL4A3 and COL4A4 microsatellites in our pedigree. When using haplotypes of all three microsatellites combined an almost-disease-specific haplotype was observed, "151/199/162." Haplotype analysis of the Norwegian elkhounds revealed one recombinant (affected dog K) between this COL4A3/COL4A4 haplotype-at-risk "151/199/162" and familial renal disease. The phenotype of this recombinant is beyond reasonable doubt; the dog had been diagnosed with familial renal disease (including presence of the typical histological picture for this disease of periglomerular fibrosis in a kidney biopsy). In case any dogs of our pedigree with no signs of renal failure will be diagnosed as affected when retested at a later age, the number of recombinants can only increase. Genotyping of additional polymorphic microsatellites and single nucleotide polymorphisms in dog K (data not shown) did not indicate incorrect parentage/sample mix-up.
In general, the hypothesis for presence of linkage is supported by a minimum LOD score (Z) of 3, which means that linkage is 10Z or 1,000 times more likely than the null hypothesis for no linkage. Conversely, a LOD of –2 is considered to be the maximum threshold to exclude linkage. The single recombinant that we detected seems to indicate that COL4A3/COL4A4 are not responsible for familial renal disease. However, these genes can not be definitely excluded because the lowest LOD score reached, based on our tightly linked microsatellites, was –0.069.
Like dogs of any breed, Norwegian elkhounds can get renal failure due to many different causes. So as not to overlook the possibility of phenocopies, a second linkage analysis was performed allowing presence of these (with age-dependent penetrance up to 0.05). The lowest LOD score obtained was 0.94 (at = 0). In our family all dogs with renal failure are suspected of being affected with familial renal disease. We could not establish risk factors for renal failure in the dog's surroundings for any affected individual. To obtain more information about affected dogs, histological examination of a kidney biopsy was used. Presence of periglomerular fibrosis (light microscopy) was used as a key feature to support the diagnosis of familial renal disease. Of the seven dogs with renal failure, kidney biopsies have been taken from six (dogs B, D, G, recombinant K, N, and O) and in all periglomerular fibrosis was detected. It seems unlikely that this typical histological picture would also be found in a phenocopy of chronic renal failure.
To obtain a definite indication about involvement of COL4A3 and/or COL4A4 with the currently available DNA samples, sequencing of these genes in both affected dogs and dogs without signs of renal failure would be needed. But then also expression of both genes in affected kidney tissue should be studied. The statistical ambiguity of our results can be attributed to the limited sample size; samples of additional informative dogs should be collected. This will allow us to take a genome scan (Cargill et al. 2002; Eggleston et al. 2002) approach to map the disease.
|
Acknowledgments |
|---|
We would like to thank: all Norwegian elkhound owners who contributed their dogs' DNA; Rob Tryon, Cathy Rinaldo, and Ruud Jalving for their help and support; the Canine Research Group of the Veterinary Genetics Lab for the CFA25 microsatellite primers (University of California, Davis). This project was supported by a grant from the Fulbright program; Jo Kolk study fund (VVAO), The Netherlands; Commissie Bevordering Diergeneeskundig en Vergelijkend Ziektekundig Onderzoek, The Netherlands; Graduate School of Animal Health, University of Utrecht, The Netherlands; the Center for Companion Animal Health, School of Veterinary Medicine, University of California, Davis. This paper was delivered at the 2nd International Conference on the "Advances in Canine and Feline Genomics: Comparative Genome Anatomy and Genetic Disease," Universiteit Utrecht, Utrecht, The Netherlands, October 14–16, 2004.
|
Footnotes |
|---|
Corresponding Editor: Francis Galibert
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-
Botstein D, White RL, Skolnick M, and Davis RW, 1980. Construction of a genetic linkage map in man using restriction fragment length polymorphisms. Am J Hum Genet 32:314–331.[ISI][Medline]
Cargill EJ, Clark LA, Steiner JM, and Murphy KE, 2002. Multiplexing of canine microsatellite markers for whole-genome screens. Genomics 80:250–253.[CrossRef][ISI][Medline]
Cox ML, Lees GE, Kashtan CE, and Murphy KE, 2003. Genetic cause of X-linked Alport syndrome in a family of domestic dogs. Mamm Genome 14:369–403.
Eggleston ML, Irion DN, Schaffer AL, Hughes SS, Draper JE, Robertson KR, Millon LV, and Pedersen NC, 2002. PCR multiplexed microsatellite panels to expedite canine genetic disease linkage analysis. Anim Biotechnol 13:223–235.[CrossRef][ISI][Medline]
Finco DR, 1973. Congenital and inherited renal disease. J Am Anim Hosp Assoc 9:301–303.
Finco DR, 1976. Familial renal disease in Norwegian elkhound dogs: physiologic and biochemical examinations. Am J Vet Res 37:87–91.[ISI][Medline]
Finco DR, Duncan JD, Crowell WA, and Hulsey ML, 1977. Familial renal disease in Norwegian elkhound dogs: morphologic examinations. Am J Vet Res 38:941–947.[ISI][Medline]
Finco DR, Kurtz HJ, Low DG, and Perman V, 1970. Familial renal disease in Norwegian elkhound dogs. J Am Vet Med Assoc 156:747–760.[ISI][Medline]
Hood JC, Dowling J, Bertram JF, Young RJ, Huxtable C, Robinson W, and Savige J, 2002. Correlation of histopathological features and renal impairment in autosomal dominant Alport syndrome in bull terriers. Nephrol Dial Transplant 17:1897–1908.
Hood JC, Robinson WF, Clark WT, Sutherland RJ, James I, Thomas MAB, and Huxtable CR, 1991. Proteinuria as an indicator of early renal disease in bull terriers with hereditary nephritis. J Small Anim Pract 32:241–248.
Hood JC, Savige J, Hendtlass A, Kleppel MM, Huxtable CR, and Robinson WF, 1995. Bull terrier hereditary nephritis: a model for autosomal dominant Alport syndrome. Kidney Int 47:758–765.[ISI][Medline]
Hood JC, Savige J, Seymour AE, Dowling J, Martinello P, Colville D, Sinclair R, Naito I, Jennings G, and Huxtable C, 2000. Ultrastructural appearance of renal and other basement membranes in the bull terrier model of autosomal dominant hereditary nephritis. Am J Kidney Dis 36:378–391.[ISI][Medline]
Hughes SS, 1998. STRand nucleic acid analysis software. Available online at www.vgl.ucdavis.edu/strand Regents of the University of California, Davis.
Lees GE, Helman RG, Kashtan CE, Michael AF, Homco LD, Millichamp NJ, Ninomiya Y, Sado Y, Naito I, and Kim Y, 1998. A model of autosomal recessive Alport syndrome in English cocker spaniel dogs. Kidney Int 54:706–719.[CrossRef][ISI][Medline]
Lees GE, Wilson PD, Helman RG, Homco LD, and Frey MS, 1997. Glomerular ultrastructural findings similar to hereditary nephritis in 4 English cocker spaniels. J Vet Intern Med 11:80–85.[ISI][Medline]
Leinonen A, Mariyama M, Mochizuki T, Tryggvason K, and Reeders ST, 1994. Complete primary structure of the human type IV collagen alpha 4(IV) chain. Comparison with structure and expression of the other alpha (IV) chains. J Biol Chem 269:26172–26177.
Lemmink HH, Schroder CH, Monnens LA, and Smeets HJ, 1997. The clinical spectrum of type IV collagen mutations. Hum Mutat 9:477–499.[CrossRef][ISI][Medline]
Li R, Mignot E, Faraco J, Kadotani H, Cantanese J, Zhao B, Lin X, Hinton L, Ostrander E.A, Patterson DF, and de Jong PJ, 1999. Construction and characterization of an eightfold redundant dog genomic bacterial artificial chromosome library. Genomics 58:9–17.[CrossRef][ISI][Medline]
Terwillliger JD, and Ott J, 1994. Part I: Two-point linkage analysis In: Handbook of human genetic linkage. Baltimore: Johns Hopkins University Press; 13–104.
Thomas R, Breen M, Deloukas P, Holmes NG, and Binns MM, 2001. An integrated cytogenetic, radiation-hybrid, and comparative map of dog chromosome 5. Mamm Genome 12:371–375.[CrossRef][ISI][Medline]
Van der Loop FT, Heidet L, Timmer ED, Van den Bosch BJ, Leinonen A, Antignac C, Jefferson JA, Maxwell AP, Monnens LA, Schroder CH, and Smeets HJ, 2000. Autosomal dominant Alport syndrome caused by a COL4A3 splice site mutation. Kidney Int 58:1870–1875.[CrossRef][ISI][Medline]
Wiersma AC, Millon LV, Hestand MS, van Oost BA, and Bannasch DL, in press. Canine COL4A3 and COL4A4 sequencing, mapping and genomic organization. DNA Sequence.
Zheng K, Thorner PS, Marrano P, Baumal R, and McInnes RR, 1994. Canine X chromosome-linked hereditary nephritis: a genetic model for human X-linked hereditary nephritis resulting from a single base mutation in the gene encoding the alpha 5 chain of collagen type IV. Proc Natl Acad Sci USA 91:3989–3993.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||