- Email: [email protected]
Correlation of the feline PKD1 genetic mutation with cases of PKD diagnosed by pathological examination
Experimental and Molecular Pathology 83 (2007) 264 – 268 www.elsevier.com/locate/yexmp
Correlation of the feline PKD1 genetic mutation with cases of PKD diagnosed by pathological examination Chris Helps ⁎, Séverine Tasker, Ross Harley School of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Bristol, BS40 5DU, UK Received 10 January 2007, and in revised form 17 April 2007 Available online 4 May 2007
Abstract Autosomal-dominant polycystic kidney disease (AD-PKD) is the most prevalent inherited genetic disease of cats, particularly affecting Persians. Using archived tissue samples from 44 cats a genotype was successfully obtained by real-time PCR for 43 cats. Twenty-five cats (18 Persians, 4 domestic longhair cats and 3 domestic shorthair (DSH) cats) were found to carry the AD-PKD mutation and all of these cats had macroscopic and/or microscopic evidence of renal cysts consistent with PKD. Eighteen cats were found to be wild-type. Twelve of these (all Persians) had no pathological evidence of PKD, but the remaining 6 cats had evidence of renal cystic lesions. On pathological review the cystic lesions in 4 (2 Persians and 2 DSH) of these 6 cats were considered not to be consistent with a primary diagnosis of PKD. Histological evidence of polycystic kidneys was, however, confirmed in the remaining 2 cats (1 DSH and 1 Bengal) and may indicate that other PKD-causing mutations exist in the feline population. © 2007 Elsevier Inc. All rights reserved. Keywords: Polycystic kidney disease; Renal histopathology; Real-time PCR genotyping
Introduction Autosomal-dominant polycystic kidney disease (AD-PKD) is the most prevalent inherited genetic disease of cats. It occurs most commonly in Persians, in which the prevalence is approximately 40–50% worldwide (Barrs et al., 2001; Barthez et al., 2003; Beck and Lavelle, 2001; Cannon et al., 2001). Cases are also reported in Persian-related breeds, namely Exotic shorthairs, and occasionally in other breeds (Barrs et al., 2001; Barthez et al., 2003; Beck and Lavelle, 2001; Cannon et al., 2001). Recently, a genetic mutation has been identified in the PKD 1 gene of Persian cats which is linked to AD-PKD (Lyons et al., 2004). A single nucleotide polymorphism (SNP) was identified in exon 29 (C to A transversion) causing a stop codon to be introduced into the mRNA. Recent studies in America (Lyons et al., 2004) and Germany (Kappe et al., 2005) found that 100% and 95% respectively of Persians diagnosed with polycystic kidney disease ⁎ Corresponding author. Division of Veterinary Pathology, Infection and Immunity, School of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Bristol, BS40 5DU, UK. Fax: +44 117 928 9505. E-mail address: [email protected] (C. Helps). 0014-4800/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2007.04.002
(PKD) by ultrasound scanning had the mutation identified by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), whilst none of the unaffected cats in either study had the mutation. One limitation of these studies is that ultrasonographic diagnosis of PKD can occasionally yield falsepositive or false-negative results, and it is only considered reliable in cats over 10 months of age (Biller et al., 1996; Cannon et al., 2001). An alternative method of diagnosing polycystic kidneys is pathological examination of renal tissue collected at necropsy or following nephrectomy. The aims of this study were to determine whether AD-PKD genotyping could be performed on DNA isolated from archived formalin-fixed wax-embedded tissues and to correlate the detection of AD-PKD in cats with and without evidence of PKD confirmed by pathological examination. It is well documented that formalin fixation causes breakdown of genomic DNA into ∼300–400 base pair fragments (Lehmann and Kreipe, 2001). Hence, the reported PCR-RFLP assay (Lyons et al., 2004), which amplifies a product of 559 base pairs, is unlikely to work reliably on degraded DNA from fixed tissue. However, we have recently developed a rapid, accurate and sensitive multiplex real-time polymerase chain reaction (PCR) assay to detect the PKD1 SNP
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268 Table 1 Histological features of lesions observed in cases of PKD (based upon Eaton et al., 1997) Site
Histological features
Cyst lining
Single layer of squamous or (low) cuboidal epithelium. Rare foci of epithelial hyperplasia. Cyst contents Cysts may be empty or contain proteinaceous material, degenerate epithelial cells, fibrin or blood. Adjacent tissue Most cysts are surrounded by normal or minimally compressed tissue. Some cysts may be surrounded by fibrous connective tissue of variable thickness. Variable additional features Chronic tubulointerstitial nephritis (not necessarily directly Tubular epithelial atrophy or regeneration. associated with cysts) Interstitial fibrosis
in genomic DNA isolated from feline blood and buccal swabs (Helps et al., 2007). We compared this assay to both the conventional PCR-RFLP assay (Lyons et al., 2004) and ultrasound screening for PKD on 72 cats and have shown it to be extremely reliable. The use of a real-time PCR assay, which has been specifically designed to amplify a short PCR product (130 base pairs), should be ideally suited to the detection of the SNP in degraded genomic DNA recovered from formalin-fixed necropsy tissues. Materials and methods Cases Pathological reports available at the School of Clinical Veterinary Science, University of Bristol, were reviewed for cats with evidence of cystic renal lesions and Persian cats without cystic renal lesions. Over a period of 23 years (1982 to 2005), 44 cats from which formalin-fixed wax-embedded tissue(s) were available for genotyping were selected for retrospective analysis. In 43 of these cats, sections from kidney slices obtained at necropsy or following nephrectomy and ranging in size from 1.2 × 0.8 cm to 3.6 × 2.1 cm were also available for histological review. The age of 4 cats was unknown, however the remainder were aged from 9 weeks to 15 years.
Tissue samples Since at least 1984 the tissue samples within the Comparative Pathology Laboratory at the School of Clinical Veterinary Science have been fixed in neutral buffered formalin prior to wax embedding. In 44 cats, wax blocks containing portions of one or more of the following formalin-fixed tissues were selected for analysis; kidney, liver, lung, heart, spleen, lymph node and oesophagus. From each wax block, two 10 μm tissue sections were cut for subsequent AD-PKD genotyping. Between cutting each tissue block the microtome blade was thoroughly cleaned using alcohol and a blank wax block was also cut to prevent tissue carry-over between cases.
DNA extraction DNA was extracted from 10 μm tissue sections by one of two methods. Either the sections were placed in 2 ml tubes containing 1 ml phosphate buffer saline (PBS) and heated to approximately 90 °C for 5 min to melt the wax. The tubes were removed from the heating block and centrifuged at 20,000×g for 2 min. After removal of the wax plug and PBS, 270 μl of T1 buffer and 30 μl proteinase K were added (Nucleospin 8 Tissue kit, Macherey-Nagel, Germany). Alternatively, 300 μl of T1 buffer and 30 μl proteinase K were added directly to the tissue sections in 2 ml tubes. For both methods the samples were then incubated in a shaker/incubator (Vortemp 56, Appleton Woods, UK) at 56 °C
265
and 1000 rpm for 18–20 h. After addition of 300 μl buffer BQ1 the samples were incubated in a shaker/incubator at 70 °C and 1000 rpm for 15 min prior to addition of 300 μl 100% ethanol. Samples were transferred to Nucleospin 8 well strips mounted on a 96 deep well plate, sealed with AeraSeal tape (Camlab, UK), and centrifuged at 3000 rpm for 5 min in a Sigma 4-15C centrifuge equipped with 09100 rotor and deep well buckets. The flow-through was discarded and 800 μl buffer BW added to each well followed by centrifugation as above. The flow-through was discarded and 800 μl buffer B5 added to each well followed by centrifugation at 5000 rpm for 15 min. The Nucleospin 8 well strips were then transferred to a rack of 1.2 ml collection tubes. 100 μl BE buffer was added to each well and incubated at room temperature for 5 min prior to recovering the DNA by centrifugation at 3000 rpm for 5 min. The extracted genomic DNA was stored at − 20 °C.
Real-time PCR genotyping Real-time PCR for PKD1 SNP detection was performed as previously described (Helps et al., 2007) using a Bio-Rad IQ system. Briefly, genotyping was performed using 12.5 μl 2× ABsolute QPCR probe mix (ABgene, UK), 200 nM sense primer (5′ GACAAGCATCTCTGGCTCTCC 3′), 200 nM antisense primer (5′ ACGACCCCGTACCACACAG 3′) (both from Invitrogen, UK), 100 nM wild-type probe (5′ 6-carboxyfluorescein (FAM)-tgTtgCgtCctcBHQ1 3′, uppercase = LNA, lowercase = DNA), 200 nM AD-PKD probe (5′ hexachloro-fluorescein (HEX)-ctgTtgAgtCctc-BHQ1 3′) (both from Proligo, France), 5 μl genomic DNA and water to 25 μl. Reaction conditions using an iCycler IQ (Bio-Rad, UK) were 95 °C for 15 min then 40 cycles of 95 °C for 10 s and 62 °C for 30 s. Fluorescence was detected at 530 and 575 nm at 62 °C and the data analysed using the iCycler software version 3. Threshold cycle values (Ct) were calculated using a threshold of 100 and 50 relative fluorescence units for the FAM and HEX channels respectively. PCR controls consisted of a water negative control, samples of genomic DNA from both wild-type and AD-PKD positive cats and a diluted PCR product made by amplifying a synthetic template DNA molecule containing only the mutant AD-PKD sequence. All samples were run in duplicate.
Histology and classification groups Histological examination was performed on 4–6 μm sections of kidney stained with haematoxylin and eosin (H&E) available for 43 cats. Sections were examined for evidence of renal cysts with microscopic features consistent with AD-PKD as described previously (Eaton et al., 1997) and summarised in Table 1. The presence of additional pathological lesions, which could be
Table 2 AD-PKD genotyping and pathological results from 44 cats Renal microscopic and macroscopic features
Genotype AD-PKD (heterozygous)
Wild-type
Not determined
Cysts consistent with PKD
17 Persian 3 DLH 1 Blue longhair 3 DSH (n = 24) 0
1 Bengal 1 DSH (n = 2)
1 Persian
0
0
1 Persian 2 DSH (n = 3) 1 Persian
0 1 Persian
12 Persian 0
0 0
25
18
1
Cystic lesions secondary to non-PKD lesions Cystic lesions of uncertain aetiology No cysts observed Macroscopic cysts observed only (no histology) Total number of cats
0
AD-PKD = Autosomal-dominant polycystic kidney disease, DLH = domestic long hair, DSH = domestic short hair.
266
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
associated with secondary renal cyst formation, was also evaluated. Following a review of the histological features and macroscopic descriptions from the archived pathological reports from all cases, each cat was assigned to one of 5 groups; (1) Renal cysts consistent with PKD, (2) Renal cystic lesions most likely secondary to non-PKD lesions, (3) Renal cystic lesions of uncertain aetiology, (4) No renal cysts observed, (5) No histology available but multiple renal cysts described on macroscopic examination. Group 3 was established because 1 cat was found to have minor pathological changes, which could not be unequivocally assigned to groups 1 or 2.
Statistical analysis Spearmans Correlations were performed using year of tissue sample preservation against the wild-type allele threshold cycle value for wild-type or AD-PKD cats (SPSS ver. 12.0 for Windows, SPSS Inc. Chicago, USA). The results were considered significant where P b 0.05.
Results The results of the AD-PKD genotyping PCR assay performed on formalin-fixed waxed-embedded tissue sections from 44 cats are shown in Table 2. In 6 cats, no genotype could be obtained using DNA extracted from the first tissue blocks
Fig. 2. (A and B) Histological appearance of a section of kidney from a DSH cat aged 4 years and 6 months and positive for the AD-PKD mutation by PCR. (A) Shows multiple, sometimes multiloculated, cysts up to several mm diameter present in the renal cortex. (B) Shows two cysts (C and C′) lined by cuboidal (large arrow) or squamous (arrow-head) epithelium and separated by a band of fibrous connective tissue (F). Cyst C contains plentiful proteinaceous material and multiple detached epithelial cells (small arrows). H&E stain.
Fig. 1. (A and B) Histological appearance of a section of kidney from a Persian kitten aged 4 months and positive for the AD-PKD mutation by PCR. (A) Shows multiple cysts of varying size (up to approximately 1.5 mm diameter) within the renal cortex. (B) Shows two cysts (C and C′) separated by normal renal cortical parenchyma. Both cysts contain proteinaceous material and are lined by cuboidal epithelium (arrows). H&E stain.
Fig. 3. Histological appearance of a section of polycystic kidney from a Bengal kitten (age unknown) negative for the AD-PKD mutation by PCR. Multiple cysts of varying size (up to approximately 0.75 mm diameter) are apparent within the renal cortex. H&E stain.
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
selected. However, in 5 of these cats the subsequent use of DNA extracted from sections of additional tissues permitted successful genotyping. For one cat, no additional tissues were available, thus a genotype was not obtained. All control samples (wild type, heterozygote, homozygote and water) gave Ct values of approximately 24–25 (positive) or N 40 (negative) in the appropriate fluorescent channel. When the threshold cycle value of the wild-type allele was compared to the year of tissue preservation a significant negative correlation was observed for both wild-type cats (rs = − 0.690, P b 0.001) and cats with AD-PKD (rs = − 0.637, P = 0.004). In the 6 cats where no genotype was obtained from the initial tissue sections analysed the tissues had been formalinfixed and wax-embedded for between 10 and 22 years. Twenty-five cats were found to be heterozygous for the ADPKD mutation (Table 2). In one Persian no kidney tissue was available for histological review, however the macroscopic description of the kidneys from this case reported bilateral cystic changes. The remaining 24 cats all had documented evidence of macroscopic renal cysts, and microscopic features consistent with a primary diagnosis of PKD were apparent upon histological review (Table 2; Figs. 1 and 2). Thus all cats identified with the AD-PKD mutation had phenotypic evidence of PKD. Eighteen cats were found to be wild-type (Table 2). In 12 of these cats (all Persians) no evidence of renal cysts was documented macroscopically or observed on histological examination. The 6 remaining wild-type cats all had microscopic evidence of renal cysts or dilated renal tubular structures (Table 2), and macroscopic evidence of renal cysts had been documented in 3 of these cases. Two of the DSH cats in this group had renal cysts that were considered likely to have arisen secondary to renal scarring and chronic nephritis. One Persian had segmental, severe, chronic tubulointerstitial nephritis with multiple, mildly dilated tubules. The tubular changes were considered to be secondary to the chronic inflammation. In
Fig. 4. Histological appearance of a focal lesion at the corticomedullary junction in a Persian aged 3 years and 9 months and negative for the AD-PKD mutation by PCR. There is a chain of structures (arrows) resembling dilated renal tubules (approximately 0.1–0.25 mm diameter) containing proteinaceous material and associated with a mild, lymphoplasmacytic, chronic inflammatory cell infiltrate (arrow heads). This solitary lesion was not considered consistent with PKD. RC, renal corpuscle. Ar, artery. H&E stain.
267
another Persian, aged 3 years 9 months, the kidney section contained a small, focal area with multiple, mildly dilated, structures resembling dilated renal tubules, and mild, chronic interstitial nephritis (Fig. 3). It was considered that these features most likely represented an incidental focus of tubulointerstitial nephritis with localised tubular dilatation, however, based upon these histological features alone, it was not possible to fully exclude the possibility that the dilated tubules might represent an early or mild PKD lesion. Consequently, the cause of the lesions in this case was classified as uncertain. In the final two cats, a Bengal kitten and an adult DSH, microscopic features of the renal cysts were considered consistent with a primary diagnosis of PKD (Fig. 4), although macroscopic cysts affecting the kidney (and pancreas and liver) had only been documented in the DSH. Discussion The results show that the real-time PCR for feline AD-PKD genotyping (Helps et al., 2007) can be used on genomic DNA isolated from archived, formalin-fixed wax-embedded tissues. Of the two DNA extraction methods used, both gave comparable results in terms of Ct values, however, the second method was less labour intensive and is therefore regarded as the method of choice for future samples. Using this technique 98% (43/44) of the cats tested were successfully genotyped. In 6 cases, no genotype could be obtained from DNA isolated from the initial tissues analysed, presumably because insufficient DNA was recovered or extensive DNA degradation had occurred. In 5 of these cats, the subsequent use of additional tissue blocks did, however, permit genotyping. It is notable that in each of these cases the tissues had been stored for 10 or more years. Significant negative correlations were found between the age of the preserved tissue and the Ct values of the wild-type allele for both wild-type and AD-PKD cats, and high levels of genomic DNA were never obtained from formalin-fixed tissues over 10 years of age. These findings suggest that degradation of DNA continues to occur in formalin-fixed wax-embedded tissues during storage (Goelz et al., 1985). This is the first reported study to compare the presence of the AD-PKD mutation in cats with evidence of PKD confirmed by pathological examination at necropsy or following nephrectomy. All 25 cats (18 Persian, 4 DLH and 3 DSH) in which the AD-PKD mutation was detected were heterozygous and all had phenotypic evidence of PKD. These findings confirm the dominant nature of this trait (Biller et al., 1996) and support the conclusion that the mutation is embryonic lethal when homozygous (Lyons et al., 2004). Eighteen cats (14 Persians, 3 DSH and 1 Bengal) were found not to have the AD-PKD mutation, although six of these cats did have evidence of renal cystic lesions or mildly dilated tubules. In three of these cats (2 DSH and 1 Persian) these lesions were considered to be secondary to renal scarring or nephritis. One Persian, aged 3 years 9 months, in which no macroscopic evidence of renal cysts was documented, was found to have a microscopic focus of chronic inflammation associated with
268
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
mildly dilated renal tubules. Based upon the histological features alone it was not possible to conclude if this lesion represented an early PKD lesion or an incidental finding, hence this cat was classified as having renal cystic lesions of uncertain etiology. However, in view of the solitary nature of this microscopic lesion, and the fact that macroscopic cysts are normally readily apparent in cats over 3 years of age with ADPKD (Eaton et al., 1997), it is considered most likely that this lesion represents an incidental focus of chronic tubulointerstitial nephritis. The absence of the AD-PKD mutation in this cat would further support the opinion that this lesion was an incidental finding. Overall, therefore, none of 14 wild-type Persians were found to have lesions indicative of AD-PKD, which is consistent with previous ultrasound-based studies (Kappe et al., 2005; Lyons et al., 2004). The two remaining wild-type cats, a DSH and a Bengal, represent a rare and interesting group that had evidence of PKD at necropsy but did not carry the AD-PKD mutation. It is notable that the DSH cat also had extrarenal polycystic lesions affecting the liver and pancreas. Extrarenal polycystic lesions have been observed in some cases of feline AD-PKD (Biller et al., 1996; Eaton et al., 1997), and are reported in cases of PKD in other species including man (Igarashi and Somlo, 2002; Johnstone et al., 2005; Krotec et al., 1996; McAloose et al., 1998; McKenna and Carpenter, 1980; Tahvanainen et al., 2005; Ward et al., 2002). The cause of the polycystic lesions in these two cats is not known, but these findings suggest that more than one PKDcausing mutation may be present in cats, as is the case in humans (Igarashi and Somlo, 2002). In summary, we have demonstrated that a real-time PCR assay for the detection of the feline PKD1 SNP linked to feline AD-PKD can be successfully applied for genotyping DNA recovered from formalin fixed wax-embedded tissues. This may be of use in retrospective studies examining the pathology and prevalence of feline AD-PKD, and in cases where only formalin-fixed tissues are available for genotyping. Acknowledgments This work was supported by a grant from the Pet Plan Charitable Trust. We thank Sheila Jones for her expertise in preparing the tissue sections, and the Veterinary Pathologists at the University of Bristol for their contributions to the case material used in this study.
References Barrs, V.R., Gunew, M., Foster, S.F., Beatty, J.A., Malik, R., 2001. Prevalence of autosomal dominant polycystic kidney disease in Persian cats and relatedbreeds in Sydney and Brisbane. Aust. Vet. J. 79, 257–259. Barthez, P.Y., Rivier, P., Begon, D., 2003. Prevalence of polycystic kidney disease in Persian and Persian related cats in France. J. Feline Med. Surg. 5, 345–347. Beck, C., Lavelle, R.B., 2001. Feline polycystic kidney disease in Persian and other cats: a prospective study using ultrasonography. Aust. Vet. J. 79, 181–184. Biller, D.S., DiBartola, S.P., Eaton, K.A., Pflueger, S., Wellman, M.L., Radin, M.J., 1996. Inheritance of polycystic kidney disease in Persian cats. J. Heredity 87, 1–5. Cannon, M.J., MacKay, A.D., Barr, F.J., Rudorf, H., Bradley, K.J., GruffyddJones, T.J., 2001. Prevalence of polycystic kidney disease in Persian cats in the United Kingdom. Vet. Rec. 149, 409–411. Eaton, K.A., Biller, D.S., DiBartola, S.P., Radin, M.J., Wellman, M.L., 1997. Autosomal dominant polycystic kidney disease in Persian and Persian-cross cats. Vet. Pathol. 34, 117–126. Goelz, S.E., Hamilton, S.R., Vogelstein, B., 1985. Purification of DNA from formaldehyde fixed and paraffin embedded human tissue. Biochem. Biophys. Res. Commun. 130, 118–126. Helps, C.R., Tasker, S., Barr, F.J., Wills, S.J., Gruffydd-Jones, T.J., 2007. Detection of the single nucleotide polymorphism causing feline autosomaldominant polycystic kidney disease in Persians from the UK using a novel real-time PCR assay. Mol. Cell. Probes 21, 31–34. Igarashi, P., Somlo, S., 2002. Genetics and pathogenesis of polycystic kidney disease. J. Am. Soc. Nephrol. 13, 2384–2398. Johnstone, A.C., Davidson, B.I., Roe, A.R., Eccles, M.R., Jolly, R.D., 2005. Congenital polycystic kidney disease in lambs. N. Z. Vet. J. 53, 307–314. Kappe, E.C., Hecht, W., Gerwing, M., Michele, U., Reinacher, M., 2005. Polycystic kidney disease in the German population of Persian cats. A comparative study of ultrasonographical examination and genetic testing. Tierärztl. Prax., Ausg. Kleintiere Heimtiere 33, 413–418. Krotec, K., Meyer, B.S., Freeman, W., Hamir, A.N., 1996. Congenital cystic disease of the liver, pancreas, and kidney in a nubian goat (Capra hircus). Vet. Pathol. 33, 708–710. Lehmann, U., Kreipe, H., 2001. Real-time PCR analysis of DNA and RNA extracted from formalin-fixed and paraffin-embedded biopsies. Methods 25, 409–418. Lyons, L.A., Biller, D.S., Erdman, C.A., Lipinski, M.J., Young, A.E., Roe, B.A., Qin, B., Grahn, R.A., 2004. Feline polycystic kidney disease mutation identified in PKD1. J. Am. Soc. Nephrol. 15, 2548–2555. McAloose, D., Casal, M., Patterson, D.F., Dambach, D.M., 1998. Polycystic kidney and liver disease in two related West Highland White Terrier litters. Vet. Pathol. 35, 77–81. McKenna, S.C., Carpenter, J.L., 1980. Polycystic disease of the kidney and liver in the Cairn Terrier. Vet. Pathol. 17, 436–442. Tahvanainen, E., Tahvanainen, P., Kaariainen, H., Hockerstedt, K., 2005. Polycystic liver and kidney diseases. Ann. Med. 37, 546–555. Ward, C.J., Hogan, M.C., Rossetti, S., Walker, D., Sneddon, T., Wang, X., Kubly, V., Cunningham, J.M., Bacallao, R., Ishibashi, M., Milliner, D.S., Torres, V.E., Harris, P.C., 2002. The gene mutated in autosomal recessive polycystic kidney disease encodes a large, receptor-like protein. Nat. Genet. 30, 259–269.
Correlation of the feline PKD1 genetic mutation with cases of PKD diagnosed by pathological examination Chris Helps ⁎, Séverine Tasker, Ross Harley School of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Bristol, BS40 5DU, UK Received 10 January 2007, and in revised form 17 April 2007 Available online 4 May 2007
Abstract Autosomal-dominant polycystic kidney disease (AD-PKD) is the most prevalent inherited genetic disease of cats, particularly affecting Persians. Using archived tissue samples from 44 cats a genotype was successfully obtained by real-time PCR for 43 cats. Twenty-five cats (18 Persians, 4 domestic longhair cats and 3 domestic shorthair (DSH) cats) were found to carry the AD-PKD mutation and all of these cats had macroscopic and/or microscopic evidence of renal cysts consistent with PKD. Eighteen cats were found to be wild-type. Twelve of these (all Persians) had no pathological evidence of PKD, but the remaining 6 cats had evidence of renal cystic lesions. On pathological review the cystic lesions in 4 (2 Persians and 2 DSH) of these 6 cats were considered not to be consistent with a primary diagnosis of PKD. Histological evidence of polycystic kidneys was, however, confirmed in the remaining 2 cats (1 DSH and 1 Bengal) and may indicate that other PKD-causing mutations exist in the feline population. © 2007 Elsevier Inc. All rights reserved. Keywords: Polycystic kidney disease; Renal histopathology; Real-time PCR genotyping
Introduction Autosomal-dominant polycystic kidney disease (AD-PKD) is the most prevalent inherited genetic disease of cats. It occurs most commonly in Persians, in which the prevalence is approximately 40–50% worldwide (Barrs et al., 2001; Barthez et al., 2003; Beck and Lavelle, 2001; Cannon et al., 2001). Cases are also reported in Persian-related breeds, namely Exotic shorthairs, and occasionally in other breeds (Barrs et al., 2001; Barthez et al., 2003; Beck and Lavelle, 2001; Cannon et al., 2001). Recently, a genetic mutation has been identified in the PKD 1 gene of Persian cats which is linked to AD-PKD (Lyons et al., 2004). A single nucleotide polymorphism (SNP) was identified in exon 29 (C to A transversion) causing a stop codon to be introduced into the mRNA. Recent studies in America (Lyons et al., 2004) and Germany (Kappe et al., 2005) found that 100% and 95% respectively of Persians diagnosed with polycystic kidney disease ⁎ Corresponding author. Division of Veterinary Pathology, Infection and Immunity, School of Clinical Veterinary Science, University of Bristol, Langford House, Langford, Bristol, BS40 5DU, UK. Fax: +44 117 928 9505. E-mail address: [email protected] (C. Helps). 0014-4800/$ - see front matter © 2007 Elsevier Inc. All rights reserved. doi:10.1016/j.yexmp.2007.04.002
(PKD) by ultrasound scanning had the mutation identified by polymerase chain reaction restriction fragment length polymorphism (PCR-RFLP), whilst none of the unaffected cats in either study had the mutation. One limitation of these studies is that ultrasonographic diagnosis of PKD can occasionally yield falsepositive or false-negative results, and it is only considered reliable in cats over 10 months of age (Biller et al., 1996; Cannon et al., 2001). An alternative method of diagnosing polycystic kidneys is pathological examination of renal tissue collected at necropsy or following nephrectomy. The aims of this study were to determine whether AD-PKD genotyping could be performed on DNA isolated from archived formalin-fixed wax-embedded tissues and to correlate the detection of AD-PKD in cats with and without evidence of PKD confirmed by pathological examination. It is well documented that formalin fixation causes breakdown of genomic DNA into ∼300–400 base pair fragments (Lehmann and Kreipe, 2001). Hence, the reported PCR-RFLP assay (Lyons et al., 2004), which amplifies a product of 559 base pairs, is unlikely to work reliably on degraded DNA from fixed tissue. However, we have recently developed a rapid, accurate and sensitive multiplex real-time polymerase chain reaction (PCR) assay to detect the PKD1 SNP
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268 Table 1 Histological features of lesions observed in cases of PKD (based upon Eaton et al., 1997) Site
Histological features
Cyst lining
Single layer of squamous or (low) cuboidal epithelium. Rare foci of epithelial hyperplasia. Cyst contents Cysts may be empty or contain proteinaceous material, degenerate epithelial cells, fibrin or blood. Adjacent tissue Most cysts are surrounded by normal or minimally compressed tissue. Some cysts may be surrounded by fibrous connective tissue of variable thickness. Variable additional features Chronic tubulointerstitial nephritis (not necessarily directly Tubular epithelial atrophy or regeneration. associated with cysts) Interstitial fibrosis
in genomic DNA isolated from feline blood and buccal swabs (Helps et al., 2007). We compared this assay to both the conventional PCR-RFLP assay (Lyons et al., 2004) and ultrasound screening for PKD on 72 cats and have shown it to be extremely reliable. The use of a real-time PCR assay, which has been specifically designed to amplify a short PCR product (130 base pairs), should be ideally suited to the detection of the SNP in degraded genomic DNA recovered from formalin-fixed necropsy tissues. Materials and methods Cases Pathological reports available at the School of Clinical Veterinary Science, University of Bristol, were reviewed for cats with evidence of cystic renal lesions and Persian cats without cystic renal lesions. Over a period of 23 years (1982 to 2005), 44 cats from which formalin-fixed wax-embedded tissue(s) were available for genotyping were selected for retrospective analysis. In 43 of these cats, sections from kidney slices obtained at necropsy or following nephrectomy and ranging in size from 1.2 × 0.8 cm to 3.6 × 2.1 cm were also available for histological review. The age of 4 cats was unknown, however the remainder were aged from 9 weeks to 15 years.
Tissue samples Since at least 1984 the tissue samples within the Comparative Pathology Laboratory at the School of Clinical Veterinary Science have been fixed in neutral buffered formalin prior to wax embedding. In 44 cats, wax blocks containing portions of one or more of the following formalin-fixed tissues were selected for analysis; kidney, liver, lung, heart, spleen, lymph node and oesophagus. From each wax block, two 10 μm tissue sections were cut for subsequent AD-PKD genotyping. Between cutting each tissue block the microtome blade was thoroughly cleaned using alcohol and a blank wax block was also cut to prevent tissue carry-over between cases.
DNA extraction DNA was extracted from 10 μm tissue sections by one of two methods. Either the sections were placed in 2 ml tubes containing 1 ml phosphate buffer saline (PBS) and heated to approximately 90 °C for 5 min to melt the wax. The tubes were removed from the heating block and centrifuged at 20,000×g for 2 min. After removal of the wax plug and PBS, 270 μl of T1 buffer and 30 μl proteinase K were added (Nucleospin 8 Tissue kit, Macherey-Nagel, Germany). Alternatively, 300 μl of T1 buffer and 30 μl proteinase K were added directly to the tissue sections in 2 ml tubes. For both methods the samples were then incubated in a shaker/incubator (Vortemp 56, Appleton Woods, UK) at 56 °C
265
and 1000 rpm for 18–20 h. After addition of 300 μl buffer BQ1 the samples were incubated in a shaker/incubator at 70 °C and 1000 rpm for 15 min prior to addition of 300 μl 100% ethanol. Samples were transferred to Nucleospin 8 well strips mounted on a 96 deep well plate, sealed with AeraSeal tape (Camlab, UK), and centrifuged at 3000 rpm for 5 min in a Sigma 4-15C centrifuge equipped with 09100 rotor and deep well buckets. The flow-through was discarded and 800 μl buffer BW added to each well followed by centrifugation as above. The flow-through was discarded and 800 μl buffer B5 added to each well followed by centrifugation at 5000 rpm for 15 min. The Nucleospin 8 well strips were then transferred to a rack of 1.2 ml collection tubes. 100 μl BE buffer was added to each well and incubated at room temperature for 5 min prior to recovering the DNA by centrifugation at 3000 rpm for 5 min. The extracted genomic DNA was stored at − 20 °C.
Real-time PCR genotyping Real-time PCR for PKD1 SNP detection was performed as previously described (Helps et al., 2007) using a Bio-Rad IQ system. Briefly, genotyping was performed using 12.5 μl 2× ABsolute QPCR probe mix (ABgene, UK), 200 nM sense primer (5′ GACAAGCATCTCTGGCTCTCC 3′), 200 nM antisense primer (5′ ACGACCCCGTACCACACAG 3′) (both from Invitrogen, UK), 100 nM wild-type probe (5′ 6-carboxyfluorescein (FAM)-tgTtgCgtCctcBHQ1 3′, uppercase = LNA, lowercase = DNA), 200 nM AD-PKD probe (5′ hexachloro-fluorescein (HEX)-ctgTtgAgtCctc-BHQ1 3′) (both from Proligo, France), 5 μl genomic DNA and water to 25 μl. Reaction conditions using an iCycler IQ (Bio-Rad, UK) were 95 °C for 15 min then 40 cycles of 95 °C for 10 s and 62 °C for 30 s. Fluorescence was detected at 530 and 575 nm at 62 °C and the data analysed using the iCycler software version 3. Threshold cycle values (Ct) were calculated using a threshold of 100 and 50 relative fluorescence units for the FAM and HEX channels respectively. PCR controls consisted of a water negative control, samples of genomic DNA from both wild-type and AD-PKD positive cats and a diluted PCR product made by amplifying a synthetic template DNA molecule containing only the mutant AD-PKD sequence. All samples were run in duplicate.
Histology and classification groups Histological examination was performed on 4–6 μm sections of kidney stained with haematoxylin and eosin (H&E) available for 43 cats. Sections were examined for evidence of renal cysts with microscopic features consistent with AD-PKD as described previously (Eaton et al., 1997) and summarised in Table 1. The presence of additional pathological lesions, which could be
Table 2 AD-PKD genotyping and pathological results from 44 cats Renal microscopic and macroscopic features
Genotype AD-PKD (heterozygous)
Wild-type
Not determined
Cysts consistent with PKD
17 Persian 3 DLH 1 Blue longhair 3 DSH (n = 24) 0
1 Bengal 1 DSH (n = 2)
1 Persian
0
0
1 Persian 2 DSH (n = 3) 1 Persian
0 1 Persian
12 Persian 0
0 0
25
18
1
Cystic lesions secondary to non-PKD lesions Cystic lesions of uncertain aetiology No cysts observed Macroscopic cysts observed only (no histology) Total number of cats
0
AD-PKD = Autosomal-dominant polycystic kidney disease, DLH = domestic long hair, DSH = domestic short hair.
266
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
associated with secondary renal cyst formation, was also evaluated. Following a review of the histological features and macroscopic descriptions from the archived pathological reports from all cases, each cat was assigned to one of 5 groups; (1) Renal cysts consistent with PKD, (2) Renal cystic lesions most likely secondary to non-PKD lesions, (3) Renal cystic lesions of uncertain aetiology, (4) No renal cysts observed, (5) No histology available but multiple renal cysts described on macroscopic examination. Group 3 was established because 1 cat was found to have minor pathological changes, which could not be unequivocally assigned to groups 1 or 2.
Statistical analysis Spearmans Correlations were performed using year of tissue sample preservation against the wild-type allele threshold cycle value for wild-type or AD-PKD cats (SPSS ver. 12.0 for Windows, SPSS Inc. Chicago, USA). The results were considered significant where P b 0.05.
Results The results of the AD-PKD genotyping PCR assay performed on formalin-fixed waxed-embedded tissue sections from 44 cats are shown in Table 2. In 6 cats, no genotype could be obtained using DNA extracted from the first tissue blocks
Fig. 2. (A and B) Histological appearance of a section of kidney from a DSH cat aged 4 years and 6 months and positive for the AD-PKD mutation by PCR. (A) Shows multiple, sometimes multiloculated, cysts up to several mm diameter present in the renal cortex. (B) Shows two cysts (C and C′) lined by cuboidal (large arrow) or squamous (arrow-head) epithelium and separated by a band of fibrous connective tissue (F). Cyst C contains plentiful proteinaceous material and multiple detached epithelial cells (small arrows). H&E stain.
Fig. 1. (A and B) Histological appearance of a section of kidney from a Persian kitten aged 4 months and positive for the AD-PKD mutation by PCR. (A) Shows multiple cysts of varying size (up to approximately 1.5 mm diameter) within the renal cortex. (B) Shows two cysts (C and C′) separated by normal renal cortical parenchyma. Both cysts contain proteinaceous material and are lined by cuboidal epithelium (arrows). H&E stain.
Fig. 3. Histological appearance of a section of polycystic kidney from a Bengal kitten (age unknown) negative for the AD-PKD mutation by PCR. Multiple cysts of varying size (up to approximately 0.75 mm diameter) are apparent within the renal cortex. H&E stain.
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
selected. However, in 5 of these cats the subsequent use of DNA extracted from sections of additional tissues permitted successful genotyping. For one cat, no additional tissues were available, thus a genotype was not obtained. All control samples (wild type, heterozygote, homozygote and water) gave Ct values of approximately 24–25 (positive) or N 40 (negative) in the appropriate fluorescent channel. When the threshold cycle value of the wild-type allele was compared to the year of tissue preservation a significant negative correlation was observed for both wild-type cats (rs = − 0.690, P b 0.001) and cats with AD-PKD (rs = − 0.637, P = 0.004). In the 6 cats where no genotype was obtained from the initial tissue sections analysed the tissues had been formalinfixed and wax-embedded for between 10 and 22 years. Twenty-five cats were found to be heterozygous for the ADPKD mutation (Table 2). In one Persian no kidney tissue was available for histological review, however the macroscopic description of the kidneys from this case reported bilateral cystic changes. The remaining 24 cats all had documented evidence of macroscopic renal cysts, and microscopic features consistent with a primary diagnosis of PKD were apparent upon histological review (Table 2; Figs. 1 and 2). Thus all cats identified with the AD-PKD mutation had phenotypic evidence of PKD. Eighteen cats were found to be wild-type (Table 2). In 12 of these cats (all Persians) no evidence of renal cysts was documented macroscopically or observed on histological examination. The 6 remaining wild-type cats all had microscopic evidence of renal cysts or dilated renal tubular structures (Table 2), and macroscopic evidence of renal cysts had been documented in 3 of these cases. Two of the DSH cats in this group had renal cysts that were considered likely to have arisen secondary to renal scarring and chronic nephritis. One Persian had segmental, severe, chronic tubulointerstitial nephritis with multiple, mildly dilated tubules. The tubular changes were considered to be secondary to the chronic inflammation. In
Fig. 4. Histological appearance of a focal lesion at the corticomedullary junction in a Persian aged 3 years and 9 months and negative for the AD-PKD mutation by PCR. There is a chain of structures (arrows) resembling dilated renal tubules (approximately 0.1–0.25 mm diameter) containing proteinaceous material and associated with a mild, lymphoplasmacytic, chronic inflammatory cell infiltrate (arrow heads). This solitary lesion was not considered consistent with PKD. RC, renal corpuscle. Ar, artery. H&E stain.
267
another Persian, aged 3 years 9 months, the kidney section contained a small, focal area with multiple, mildly dilated, structures resembling dilated renal tubules, and mild, chronic interstitial nephritis (Fig. 3). It was considered that these features most likely represented an incidental focus of tubulointerstitial nephritis with localised tubular dilatation, however, based upon these histological features alone, it was not possible to fully exclude the possibility that the dilated tubules might represent an early or mild PKD lesion. Consequently, the cause of the lesions in this case was classified as uncertain. In the final two cats, a Bengal kitten and an adult DSH, microscopic features of the renal cysts were considered consistent with a primary diagnosis of PKD (Fig. 4), although macroscopic cysts affecting the kidney (and pancreas and liver) had only been documented in the DSH. Discussion The results show that the real-time PCR for feline AD-PKD genotyping (Helps et al., 2007) can be used on genomic DNA isolated from archived, formalin-fixed wax-embedded tissues. Of the two DNA extraction methods used, both gave comparable results in terms of Ct values, however, the second method was less labour intensive and is therefore regarded as the method of choice for future samples. Using this technique 98% (43/44) of the cats tested were successfully genotyped. In 6 cases, no genotype could be obtained from DNA isolated from the initial tissues analysed, presumably because insufficient DNA was recovered or extensive DNA degradation had occurred. In 5 of these cats, the subsequent use of additional tissue blocks did, however, permit genotyping. It is notable that in each of these cases the tissues had been stored for 10 or more years. Significant negative correlations were found between the age of the preserved tissue and the Ct values of the wild-type allele for both wild-type and AD-PKD cats, and high levels of genomic DNA were never obtained from formalin-fixed tissues over 10 years of age. These findings suggest that degradation of DNA continues to occur in formalin-fixed wax-embedded tissues during storage (Goelz et al., 1985). This is the first reported study to compare the presence of the AD-PKD mutation in cats with evidence of PKD confirmed by pathological examination at necropsy or following nephrectomy. All 25 cats (18 Persian, 4 DLH and 3 DSH) in which the AD-PKD mutation was detected were heterozygous and all had phenotypic evidence of PKD. These findings confirm the dominant nature of this trait (Biller et al., 1996) and support the conclusion that the mutation is embryonic lethal when homozygous (Lyons et al., 2004). Eighteen cats (14 Persians, 3 DSH and 1 Bengal) were found not to have the AD-PKD mutation, although six of these cats did have evidence of renal cystic lesions or mildly dilated tubules. In three of these cats (2 DSH and 1 Persian) these lesions were considered to be secondary to renal scarring or nephritis. One Persian, aged 3 years 9 months, in which no macroscopic evidence of renal cysts was documented, was found to have a microscopic focus of chronic inflammation associated with
268
C. Helps et al. / Experimental and Molecular Pathology 83 (2007) 264–268
mildly dilated renal tubules. Based upon the histological features alone it was not possible to conclude if this lesion represented an early PKD lesion or an incidental finding, hence this cat was classified as having renal cystic lesions of uncertain etiology. However, in view of the solitary nature of this microscopic lesion, and the fact that macroscopic cysts are normally readily apparent in cats over 3 years of age with ADPKD (Eaton et al., 1997), it is considered most likely that this lesion represents an incidental focus of chronic tubulointerstitial nephritis. The absence of the AD-PKD mutation in this cat would further support the opinion that this lesion was an incidental finding. Overall, therefore, none of 14 wild-type Persians were found to have lesions indicative of AD-PKD, which is consistent with previous ultrasound-based studies (Kappe et al., 2005; Lyons et al., 2004). The two remaining wild-type cats, a DSH and a Bengal, represent a rare and interesting group that had evidence of PKD at necropsy but did not carry the AD-PKD mutation. It is notable that the DSH cat also had extrarenal polycystic lesions affecting the liver and pancreas. Extrarenal polycystic lesions have been observed in some cases of feline AD-PKD (Biller et al., 1996; Eaton et al., 1997), and are reported in cases of PKD in other species including man (Igarashi and Somlo, 2002; Johnstone et al., 2005; Krotec et al., 1996; McAloose et al., 1998; McKenna and Carpenter, 1980; Tahvanainen et al., 2005; Ward et al., 2002). The cause of the polycystic lesions in these two cats is not known, but these findings suggest that more than one PKDcausing mutation may be present in cats, as is the case in humans (Igarashi and Somlo, 2002). In summary, we have demonstrated that a real-time PCR assay for the detection of the feline PKD1 SNP linked to feline AD-PKD can be successfully applied for genotyping DNA recovered from formalin fixed wax-embedded tissues. This may be of use in retrospective studies examining the pathology and prevalence of feline AD-PKD, and in cases where only formalin-fixed tissues are available for genotyping. Acknowledgments This work was supported by a grant from the Pet Plan Charitable Trust. We thank Sheila Jones for her expertise in preparing the tissue sections, and the Veterinary Pathologists at the University of Bristol for their contributions to the case material used in this study.
References Barrs, V.R., Gunew, M., Foster, S.F., Beatty, J.A., Malik, R., 2001. Prevalence of autosomal dominant polycystic kidney disease in Persian cats and relatedbreeds in Sydney and Brisbane. Aust. Vet. J. 79, 257–259. Barthez, P.Y., Rivier, P., Begon, D., 2003. Prevalence of polycystic kidney disease in Persian and Persian related cats in France. J. Feline Med. Surg. 5, 345–347. Beck, C., Lavelle, R.B., 2001. Feline polycystic kidney disease in Persian and other cats: a prospective study using ultrasonography. Aust. Vet. J. 79, 181–184. Biller, D.S., DiBartola, S.P., Eaton, K.A., Pflueger, S., Wellman, M.L., Radin, M.J., 1996. Inheritance of polycystic kidney disease in Persian cats. J. Heredity 87, 1–5. Cannon, M.J., MacKay, A.D., Barr, F.J., Rudorf, H., Bradley, K.J., GruffyddJones, T.J., 2001. Prevalence of polycystic kidney disease in Persian cats in the United Kingdom. Vet. Rec. 149, 409–411. Eaton, K.A., Biller, D.S., DiBartola, S.P., Radin, M.J., Wellman, M.L., 1997. Autosomal dominant polycystic kidney disease in Persian and Persian-cross cats. Vet. Pathol. 34, 117–126. Goelz, S.E., Hamilton, S.R., Vogelstein, B., 1985. Purification of DNA from formaldehyde fixed and paraffin embedded human tissue. Biochem. Biophys. Res. Commun. 130, 118–126. Helps, C.R., Tasker, S., Barr, F.J., Wills, S.J., Gruffydd-Jones, T.J., 2007. Detection of the single nucleotide polymorphism causing feline autosomaldominant polycystic kidney disease in Persians from the UK using a novel real-time PCR assay. Mol. Cell. Probes 21, 31–34. Igarashi, P., Somlo, S., 2002. Genetics and pathogenesis of polycystic kidney disease. J. Am. Soc. Nephrol. 13, 2384–2398. Johnstone, A.C., Davidson, B.I., Roe, A.R., Eccles, M.R., Jolly, R.D., 2005. Congenital polycystic kidney disease in lambs. N. Z. Vet. J. 53, 307–314. Kappe, E.C., Hecht, W., Gerwing, M., Michele, U., Reinacher, M., 2005. Polycystic kidney disease in the German population of Persian cats. A comparative study of ultrasonographical examination and genetic testing. Tierärztl. Prax., Ausg. Kleintiere Heimtiere 33, 413–418. Krotec, K., Meyer, B.S., Freeman, W., Hamir, A.N., 1996. Congenital cystic disease of the liver, pancreas, and kidney in a nubian goat (Capra hircus). Vet. Pathol. 33, 708–710. Lehmann, U., Kreipe, H., 2001. Real-time PCR analysis of DNA and RNA extracted from formalin-fixed and paraffin-embedded biopsies. Methods 25, 409–418. Lyons, L.A., Biller, D.S., Erdman, C.A., Lipinski, M.J., Young, A.E., Roe, B.A., Qin, B., Grahn, R.A., 2004. Feline polycystic kidney disease mutation identified in PKD1. J. Am. Soc. Nephrol. 15, 2548–2555. McAloose, D., Casal, M., Patterson, D.F., Dambach, D.M., 1998. Polycystic kidney and liver disease in two related West Highland White Terrier litters. Vet. Pathol. 35, 77–81. McKenna, S.C., Carpenter, J.L., 1980. Polycystic disease of the kidney and liver in the Cairn Terrier. Vet. Pathol. 17, 436–442. Tahvanainen, E., Tahvanainen, P., Kaariainen, H., Hockerstedt, K., 2005. Polycystic liver and kidney diseases. Ann. Med. 37, 546–555. Ward, C.J., Hogan, M.C., Rossetti, S., Walker, D., Sneddon, T., Wang, X., Kubly, V., Cunningham, J.M., Bacallao, R., Ishibashi, M., Milliner, D.S., Torres, V.E., Harris, P.C., 2002. The gene mutated in autosomal recessive polycystic kidney disease encodes a large, receptor-like protein. Nat. Genet. 30, 259–269.