 |
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
|
||||
| 2007 JASN IMPACT FACTOR 7.111 | HOME AUTHOR INFO EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP | |||
| CURRENT ISSUE | ARCHIVES | JASN Express | ONLINE SUBMISSION | |
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Genetics and Development |
Motif of Novel Protein SamCystin is Associated with Polycystic Kidney Disease in (cy/+) Rat
* The Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford, United Kingdom; Medical Research Centre, Klinikum Mannheim, University of Heidelberg, Mannheim, Germany; Division of Nephrology, Medical Clinic III, University of Frankfurt, Frankfurt, Germany; Centre National de Génotypage, Evry, France; and || Institute for Molecular and Cellular Anatomy, University of Regensburg, Regensburg, Germany
Address correspondence to: Dr. Marie-Thérèse Bihoreau, The Wellcome Trust Centre for Human Genetics, University of Oxford, Oxford OX3 7BN, UK. Phone: +44-01-865-287-648; Fax: +44-01-865-287-533; E-mail: bihoreau{at}well.ox.ac.uk
Received for publication June 8, 2005. Accepted for publication August 13, 2005.
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
motif. The characterization of this protein, which does not share structural homologies with known polycystins, may give new insights into the pathophysiology of renal cyst development in patients. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Two genes that are responsible for ADPKD, PKD1 and PKD2, which encode polycystins, have already been characterized in humans. Mutations in these proteins account for the vast majority of ADPKD cases (1). However, a small number of families with ADPKD have been reported to develop PKD that seems to be unlinked to either PKD1 or PKD2, leading to the controversial suggestion of the involvement of other as-yet-unknown genes (1,2). Furthermore, several other, although less frequent, hereditary cystic kidney diseases that also lead to terminal renal failure (1) occur in childhood, adolescence, or adulthood (e.g., autosomal recessive polycystic kidney disease, nephronophthisis, medullary cystic kidney disease).
Rodent models of spontaneous renal cyst disorders are powerful systems to characterize the pathophysiologic mechanisms involved and potentially point to new candidate genes that contribute to human cystic diseases (3). Of the available rat models, altered physiologic and morphologic renal phenotypes of the inbred (cy/+) rat strain closely resemble human ADPKD (4,5). Renal cyst lesions progressively develop from the first few weeks of life, and gender differences in disease severity have been reported. In addition, this model is characterized by a slow progression of uremia, proteinuria, and hyperlipidemia. In three independent genetic studies, the gene that is responsible for PKD in (cy/+) rats has been localized to a large, poorly defined region on rat chromosome 5 (6–8) spanning between 15 and 5 cM according to the different published linkage maps (9,10). This locus, which is denoted Pkdr1, does not share homology relationships with the gene loci PKD1 or PKD2 (6).
To identify the PKD susceptibility gene in the (cy/+) rat, we combined traditional physical and genetic mapping methods with computational analyses of the emerging sequence of the rat genome at the refined Pkdr1 locus. We report that the susceptibility gene encodes a novel protein, SamCystin, which contains both ankyrin repeats and a sterile motif (SAM), in which we identified a missense mutation in the PKD affected rat strain. This protein, strongly expressed in kidneys, does not share structural similarities with known polycystins or other proteins that are known to be involved in cystic kidney diseases. Our findings provide new perspectives to elucidate the pathophysiology and genetic causes of cystic kidney disorders in human patients.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
A large cohort of 742 first-backcross (BC1) rats was derived from affected F1(cy/+) rats that were mated to BN rats as described previously (11). The BN rat was preferred to other inbred rat strains because of the high level of allele variation observed between BN and PKD/Mhm(cy/+) strains, which facilitates the detection of recombinant hybrids in the BC1 cohort and therefore increases efficiency of positional cloning the disease gene. All hybrids were killed at day 36. Kidneys were removed, weighed, fixed in 3% buffered formaldehyde, and then embedded in paraffin for histologic analyses. After hematoxylin/eosin staining of kidney sections, presence of renal cysts was assessed by microscopic examination (11).
Microsatellite and Gene Markers
All available markers that are known to map in the vicinity of the locus Pkdr1 were selected from public rat genetic databases (http://www.well.ox.ac.uk/rat_mapping_resources; http://rgd.mcw.edu) and tested for allele variation between the PKD/Mhm(cy/+) and BN rat strains. A total of 22 markers that already are genetically mapped (9,10) and/or localized on the T55 rat radiation hybrid panel (9,12) were selected to initiate this study.
We used data from existing comparative maps between rat chromosome 5 and mouse and/or human chromosomes (10,12) to search human (http://www.ncbi.nlm.nih.gov) and mouse (http://www.informatics.jax.org) public genome databases to establish a gene map of the original Pkdr1 locus (6), which was designed to facilitate physical mapping of the locus. Rat gene sequences that are homologous to human and mouse genes that were predicted to map at the Pkdr1 locus were identified by BLAST query. A list of human orthologous genes that mapped to the Pkdr1 interval was established using both the Human Genome Browser (http://genome.ucsc.edu) and Ensembl (http://www.ensembl.org). Rat sequences (mRNA and expressed sequence tag [EST]) homologous to each human sequence were accessed, when possible, via the data mining tools provided by the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov). Rat sequences were also identified through other public rat genome databases, the UIowa Rat EST Project (http://ratest.uiowa.edu) and RATMAP (http://ratmap.gen.gu.se). Since 2003, the release of the rat genome sequence (http://www.ensembl.org/Rattus_norvegicus; http://genome.ucsc.edu) has accelerated rat gene annotation and comparative genome analysis (10) and increased the number of new informative microsatellite markers for the positional cloning of Pkdr1.
Oligonucleotides for new rat microsatellites were designed using the OLIGO 4.0 computer program (National Biosciences Inc., Plymouth, MN). PCR primer pairs were synthesized commercially by Sigma Genosys (Pampisford, UK).
Genotype Determination
Microsatellite markers that showed evidence of allele variation between the PKD/Mhm(cy/+) and BN strains were typed on either the complete BC1 cohort or, after progress in the positional cloning of Pkdr1, a selection of recombinant hybrids. PCR amplification and genotyping procedures were carried out as described previously (6).
Physical Mapping and Sequence Tagged Site Screening
Both rat microsatellites and gene/EST markers that mapped to the original Pkdr1 locus were used for screening rat yeast artificial chromosome (YAC), bacterial artificial chromosome (BAC), and P1-derived artificial chromosome (PAC) libraries. Standard PCR screening methods were used to isolate YAC clones from the rat LPY library (https://www.resgen.com). YAC ends were obtained by use of vectorette bubble PCR (13) and sequenced directly using the BigDye Terminator kit (Perkin-Elmer/Applied Biosystems, Warrington, UK) to generate new markers and allow chromosome walking. Rat BAC clones were isolated from the RPCI-32 library (http://bacpac.chori.org) and PAC clones from the RPCI-31 library (14) by hybridization of radioactively labeled probes to high-density gridded nylon filters as described previously (14). BAC/PAC end sequences from strategically selected BAC and PAC clones were obtained using the same method as for YAC ends. Sequence tagged site content maps were assembled by PCR amplification of DNA minipreparations from the isolated clones.
RNA Isolation and cDNA Preparation
Total RNA was extracted from whole kidneys of 2-wk-old PKD/Mhm(cy/cy), PKD/Mhm(cy/+), PKD/Mhm(+/+), BN, and Lewis rat strains using the TRIZOL reagent (Life Sciences, Paisley, UK), followed by a clean-up with the RNeasy kit (Qiagen, Crawley, UK). First-strand cDNA was synthesized from 5 µg of total RNA using Superscript II RNase HReverse Transcriptase (Invitrogen, Paisley, UK).
Sequencing and In Silico Sequence Analysis
Cycle sequencing was carried out using BigDye terminators (Perkin-Elmer/Applied Biosystems), followed by either ethanol precipitation or Sephadex G50 filter plate purification, and sequences were run on ABI 377 or 3700 automated sequencers. cDNA were sequenced in both directions using PCR primer pairs that amplified overlapping cDNA fragments designed along the full-length coding sequence. Raw sequence data were analyzed and assembled using the University of Washington’s suite of programs PHRED, PHRAP, and CONSED (www.phrap.org). For clone sequencing, vector and repeat sequences were masked and high-quality sequence was compared with available sequence databases using an in-house search program. The AceDB genome database program was used to display and manage this information (http://www.acedb.org). Mutation screening was performed on 20 different inbred rat strains and substrains, including BN, Lewis, GK, SHR, SHRSP, WKY, BB, GAERS, DA, ACI, Wistar-Furth, Zucker, and Fisher.
PCR-RFLP Analysis
The R823W C/T mutation was detected using PCR-RFLP. PCR with primers (CTAGAAGCCTCAGTGACCCC, CAGCGTGTGAACAAGGTAGG) designed from corresponding genomic DNA sequence was carried out as described previously (6). PCR products of length 242 bp were digested with Msp I restriction endonuclease (Boehringer Mannheim, Mannheim, Germany) that recognizes and cuts the wild-type (into two fragments of 157 and 85 bp) but not the mutated sequence site of PCR products. Reactions were incubated at 37°C for 1 h. The digested PCR products were resolved on an agarose gel.
Gene Transcription Analysis
A commercial panel of cDNA that was prepared from multiple rat tissues (Rat MTC Panel I; Clontech, Oxford, UK) was used to test the expression of the Pkdr1 gene in eight different tissues. We designed the WC4 primer pair (CTTCGAGCCAAAAACGGAT, CAGCATGGTCGTCAGTAACAC) to amplify a 462-bp cDNA segment between the ankyrin (ANK) and SAM domains of Pkdr1 to ensure maximum specificity. Transcript levels for the B2 microglobin housekeeping gene served as a cDNA loading control.
Northern blots were performed with 20 µg of total RNA that was prepared from kidneys of 21-d-old PKD/Mhm(cy/cy), PKD/Mhm(cy/+), and PKD/Mhm(+/+) rats. RNA was size-fractionated on a 0.8% formaldehyde-agarose gel and transferred to nylon membranes (Hybond N; Amersham, Braunschweig, Germany). A 32P random primer-labeled 1.6-kb fragment of Pkdr1 cDNA was prepared as gene-specific probe. Hybridization was carried out with a 50-mM sodium phosphate buffer (pH 7.0) that contained 50% formamide, 5x SSC, 5x Denhardt’s solution, 1% SDS, and 250 µg/ml denatured salmon sperm DNA, followed by washes in SSC/SDS solutions. Detection of the rat glyceraldehyde-3-phosphate dehydrogenase mRNA levels was used as loading standard. Transcripts were visualized on a FUJI BAS 2500 PhosphorImager (FUJI, Dusseldorf, Germany).
In Situ Hybridization
For kidney in situ hybridization, rats were killed by retrograde perfusion through the distal abdominal aorta with 0.9% sodium chloride for 3 min at pressure level of 220 mmHg. Kidney slices were immersed in 4% paraformaldehyde in PBS for 2 to 4 h, then in 18% sucrose for 2 h before being slowly frozen in isopentane at –30°C and placed directly into liquid nitrogen. Five- to 7-µm cryosections were postfixed in 4% paraformaldehyde in PBS, acetylated in 0.1 M triethanolamine (pH 8.0) that contained 0.25% acetic anhydride. The slides were dehydrated in graded ethanol up to 95% and air-dried. Sense and antisense Pkdr1 cRNA probes spanning the whole cDNA were generated by in vitro transcription using digoxigenin-11-UTP (Roche, Mannheim, Germany) according to the manufacturer’s instructions. Transcripts were subjected to partial alkaline hydrolysis to improve penetration. In situ hybridization was carried out as described previously (15). Briefly, sections were hybridized with a solution that contained 50% formamide and 5 to 8 ng/µl RNA probe overnight at 48°C and stringently washed. Sections then were incubated with a polyclonal alkaline phosphatase-coupled sheep anti-digoxigenin antibody. Signal detection was performed with NBT/X-phosphate, and the color reaction was controlled under the microscope. Sense probes did not produce any detectable signal. On consecutive cryosections, postfixed in 4% paraformaldehyde in PBS, staining for endogenous alkaline phosphatase was performed using NBT/X-phosphate as a substrate.
Western Blot Analysis
cDNA were subcloned into the eukaryotic expression vector pEBG, which encodes fusion proteins with glutathione S-transferase at their NH2-terminus. COS-7 cells were transiently transfected using the DEAE-dextran method (16). Cells were lysed in 1x PBS/1% Triton X-100, and the protein concentration was determined using an improved Bradford assay (17). Equal amounts of protein were separated under reducing conditions on a denaturing SDS-polyacrylamide gel, transferred onto a polyvinylidene difluoride membrane, and then incubated with a mouse monoclonal anti-GST antibody (diluted 1:3000; Sigma, St. Louis, MO) as primary antibody. Specifically bound primary antibody was detected with a horseradish peroxidase–conjugated goat anti-mouse secondary antibody and subsequent chemiluminescence.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
The primary genetic screening of the BC1 cohort identified 45 hybrids that show recombinations within the locus Pkdr1. Subsequent extensive genotyping of these 45 rats led to the refinement of the disease locus to a critical region of approximately 1 cM between D5Rat224 and D5Rat137 (Figures 1 and 2). It also allowed the exclusion of some potential PKD candidate loci, such as Inv, Sntx17, Grhpr, and Tmeff1, which were identified by comparative genome analysis of the Pkdr1 locus. The BC1 rat panel was finally reduced to five recombinant animals that were genotyped with additional microsatellite markers generated from the emerging rat draft genomic sequence covering the region of interest (http://www.ensembl.org/Rattus_norvegicus; http://genome.ucsc.edu). Strong PKD candidate genes then could be excluded (LOC308281 similar to dynactin DCTN3, a dynein related to DNAI1, Cntr, Tbc1d2, Gpr51, and Col15a1). The critical interval was eventually reduced to 0.47 Mb between D5Wox207 and D5Rat227, where only two genes (Galnt12 and a novel gene) have been annotated by in silico mapping (Figures 1c and 2). D5Wox126, localized in the genomic sequence of Galnt12, was checked in the entire BC1 cohort for absence of any recombinant genotypes.
|
-d-galactosamine:polypeptide N-acetylgalactosaminyltransferase 12 (Galnt12) and an unknown gene (LOC362515), predicted to encode a protein that contained ANK repeats and a SAM domain, were evaluated for sequence mutations in their respective coding sequence. For that purpose, cDNA were prepared from kidneys of PKD/Mhm(cy/+), PKD/Mhm(cy/cy), PKD/Mhm(+/+), BN, and Lewis rats. No sequence variants between the four strains were found in Galnt12. Sequence analysis of the cDNA of the novel gene revealed a C
T transition in the cy rat that was absent in 20 different inbred rat strains, including the Sprague Dawley strain in which the spontaneous cy mutation initially arose (Figure 3a). The mutation was also present in the cy rat strain of the American colony (7) derived from the original Han:SPRD(cy/+) strain (Figure 3b). By BLAST alignment of our cDNA sequence with rat genomic clone AC132650, we deduced the exonic boundaries and established that the gene contains 16 exons spanning an open reading frame of 2.655 kb (Table 1) encoding a putative protein of 885 amino acids. The intron-exon structure, submitted to GenBank (AY661303), emerged as being slightly different from that predicted by various programs used for gene annotations (Figure 1d). The transcript prediction that was obtained with the Twinscan program omitted exon 16, whereas gene annotation by the Ensembl automatic analysis pipeline skipped exons 1 and 2 and altered exon 11. The Fgenesh++ program left out exons 1 and 16 and partially exon 2. Genscan computed a much longer gene with 19 exons but reduced the sequence of exons 11 and 12. The missense point mutation occurs in exon 14 and changes an arginine (CGG) to a tryptophan (TGG) at amino acid 823 within the SAM domain of the new protein, which we named SamCystin.
|
|
In situ hybridization on normal adult rat kidney sections using a Pkdr1-specific digoxigenin-labeled antisense riboprobe showed strong mRNA expression in proximal tubules of the cortex and in the outer stripe of the medulla (Figure 4a) but not in the medulla. Higher magnification of the kidney cortex (Figure 4b) showed the specific cytoplasmic localization of labeled Pkdr1 mRNA in proximal tubules, corresponding to the site of cyst formation in cy/+ rats. This was confirmed by specific labeling of the proximal tubules using a substrate for alkaline phosphatase (Figure 4, d and e). Similar findings were obtained on kidney sections of PKD/Mhm(cy/+) rats (Figure 4, c and f).
|
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Although the cellular function of SamCystin and the role of the mutation in renal cyst formation in the PKD/Mhm(cy/+) rat are unknown, we can anticipate an involvement of SamCystin in protein scaffolding from the presence of two key features: the ANK motifs and the SAM domain. The ANK repeats are tandem repeats of approximately 33 amino acids that are composed of a 
-hairpin and two -helices that provide a site for protein–protein interaction and are reported to be present in transcription factors and cytoskeleton proteins (21). The SAM domain is a protein–protein and protein–nucleic acid interaction module of approximately 70 amino acids that is found in a variety of proteins, including scaffolding proteins, transcriptional and translational regulators, tyrosine and serine/threonine kinases, and RNA-binding proteins (22–25). As the residues that are implicated in RNA binding in Smaug (26) are not conserved in the SamCystin SAM domain (Figure 5), it seems more likely that it participates in protein–protein interactions. The SAM domain functions in protein interactions as a result of its ability to self-associate (27), bind other SAM domains (28), and form interactions with non–SAM domain–containing proteins (29–32). The precise residues that are responsible for oligomer formation between SAM domains vary considerably depending on the protein (22) (Figure 5). Accordingly, it is not possible to infer directly that the R-W mutation disrupts a protein–protein interaction interface. The low sequence similarity of SamCystin to any SAM domains of known structure (29% at best, to that of Smaug) hampers accurate prediction of the precise local environment of the native arginine side chain. However, its position adjacent to a loop region that is involved in self-association of Ph SAM domains suggests that disruption that is caused by accommodation of the very different biochemical properties of the tryptophan residue may affect the ability of the SamCystin protein to form complexes.
The presence of both SAM and ANK motifs in SamCystin indicates that this novel protein binds simultaneously to multiple interaction partners and self-associates to form large high molecular mass complexes. There are several known proteins that contain a combination of these two motifs together, including SANS (33), the SHANK family (34), CASK-interacting proteins (35), TANK (36), and GASZ (37).
Functional mutations that alter SAM protein domains have already been described in human and mouse models. The SAM domain of the mouse homolog of the Drosophila Bicaudal C gene, Bicc1, which is expressed in cilia (3) and shares extensive similarities with that of SamCystin, is either altered or truncated in the jcpk and bpk mouse models of PKD (38). The role of the protein encoded by Bicc1 in developmental processes in Drosophila (39) suggests that it is required for establishing and maintaining normal tubule architecture, with its SAM domain playing a key role (38). Mutations in the SAM domain of the SANS protein have been associated with deafness in human (33) and in the Jackson shaker deaf mutant mouse (40). Disruptions of the hair bundles, which are mechanoreceptive structures of inner ear sensory cells, in these mice suggest that SANS may function as an anchoring/scaffolding protein in cochlea hair cells mediated by protein–protein interactions (40). These results may be particularly relevant to the function of SamCystin in PKD in the PKD/Mhm(cy/+) rat because SamCystin and SANS are currently the only known proteins that contain ANK and SAM domains exclusively and recent evidence suggests a causal link between ADPKD and dysfunction in renal ciliary structures (41), which may be the kidney mechanosensory equivalent of the auricular hair bundles.
A direct role of the SAM domain of SamCystin in polycystin-1–related signaling pathways cannot be ruled out. After chromosomal rearrangement, the SAM domain of the TEL (translocation Ets leukemia) transcriptional repressor fuses to a variety of tyrosine kinases, such as Janus kinase 2 (JAK2), resulting in constitutive activation and various hematologic malignancies (42). JAK2 associates with and is activated by a complex of polycystin 1 and polycystin 2 in the JAK-STAT signaling pathway, mutations in which can result in deregulated cell proliferation and differentiation (43). SamCystin could also participate in renal morphogenesis as it has been shown that bifunctional apoptosis regulator protein mediates in vitro apoptosis through the interaction of its SAM domain with Bcl-2 (31).
One of the important features of SamCystin lies in that it is a novel protein that is structurally completely unrelated to the polycystins 1 and 2 as well as any other proteins associated with human cystic kidney disease. Recently, increasing evidence has been accumulated for the role of primary cilia in the pathogenesis of cyst formation as a result of the localization of Bicaudal C; cystin; fibrocystin; polaris; inversin; nephrocystins 1, 3, 4, and 5; polycystins 1 and 2; and kinesin-II 3a subunit (3,44,45) to this organelle. Therefore, an important development of our study will be to test whether the SamCystin protein is also part of the cilium despite its particular structure. In light of reports of cyst occurrence in other renal structures, including the collecting duct and distal nephron (46), further investigations into the biologic function of SamCystin and the effects of the mutation involved in the development of abnormal tubule architecture in the PKD rat may define new pathways and molecular mechanisms that contribute to renal cyst formation. Structural studies of SamCystin will help with resolving binding surfaces in SAM–domain interactions, identifying interacting proteins in the kidney, and characterizing the molecular pathways involving SamCystin to uncover important hitherto unknown contributing factors in normal renal development and in cyst formation.
The identification of the mutation that is likely to cause renal cyst development in the PKD/Mhm(cy/+) rat in a novel gene encoding SamCystin provides new opportunities for further dissecting mechanisms that control cystogenesis and modulate PKD severity (11). Both this novel gene and genes that encode proteins that interact with SamCystin call for large-scale genetic and functional studies in human PKD that may have significant implications for the clinical classification of cystic kidney diseases and the development of novel therapeutic strategies.
|
Acknowledgments |
|---|
We are grateful for the expert technical assistance of Vera Könecke and Larissa Osten.
|
Footnotes |
|---|
Supplemental information for this article is available online at http://www.jasn.org/
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  T. Weimbs Polycystic kidney disease and renal injury repair: common pathways, fluid flow, and the function of polycystin-1 Am J Physiol Renal Physiol, November 1, 2007; 293(5): F1423 - F1432. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
HOME
CURRENT ISSUE
ARCHIVES
JASN Express
ONLINE SUBMISSION
AUTHOR INFO
EDITORIAL BOARD SUBSCRIBE FEEDBACK ALERTS HELP |
if microsatellites, by 
if clone ends, and by
if expressed sequence tag. Bacterial artificial chromosome and P1-derived artificial chromosome shown beneath are denoted by solid black lines. (c) Annotation of the chromosomal interval defined by recombinant mapping for known and predicted genes (in green and orange, respectively). (d) Genomic structure of four Pkdr1 gene predictions and the final version derived from cDNA sequencing. Solid black lines represent the exons, and numbers indicate exon length (bp). In Pkdr1, the asterisk shows the location of the C
, recombinant genotype at the marker; 
, no recombination. The values to the right are the numbers of recombinant progeny for each marker.
