Phenotypic variations of orpk mutation and chromosomal localization of modifiers influencing kidney phenotype
CARLA SOMMARDAHL1,
MARILYN COTTRELL1,
J. ERBY WILKINSON1,
RICHARD P. WOYCHIK2 and
DABNEY K. JOHNSON3
1 Department of Pathology and Department of Large Animal Clinical Sciences, University of Tennessee College of Veterinary Medicine, Knoxville, Tennessee 37901-1071
2 Lynx Therapeutics, Hayward, California 94545
3 Life Sciences Division, Oak Ridge National Laboratory, Oak Ridge, Tennessee 37831
 |
ABSTRACT
|
---|
The Oak Ridge polycystic kidney (orpk) mutant mouse model resulted from a transgene insertion into the Tg737 gene and exhibits a pleiotropic syndrome with lesions in the kidney, liver, and pancreas. We found marked differences in the phenotypic expression of the orpk mutation when bred on different genetic backgrounds. In the FVB/N background, the phenotype is very severe for kidney, pancreas, and liver lesions. To evaluate better how genetic background might influence the expressivity of the orpk phenotype, we bred the transgene into the C3HeB/FeJLe (C3H) genetic background. We performed a genome-wide scan using backcross and intercross populations with more than 150 markers to map the chromosomal location of the modifier genes that differ in the FVB/N and C3H genetic backgrounds that affect the severity of kidney disease in the orpk mouse. Low-resolution interval mapping was performed using the Map Manager QTb program, with the interval explaining a significant portion of the variance being the distal end of chromosome 4.
polycystic kidney disease; biliary hyperplasia; pancreatic cysts
 |
INTRODUCTION
|
---|
IN BOTH DOMINANT AND RECESSIVE polycystic kidney disease (PKD) in humans, there is extreme phenotypic variability in disease progression and in extrarenal manifestations. This variability is, in part, a result of genetic heterogeneity due to mutations in distinct genes giving rise to PKD. However, marked variability in clinical disease has been reported among kindred that probably have the same mutation predisposing them to PKD; thus other genetic and environmental factors are likely to affect the manifestation of the phenotype (7). Effects of genetic background on disease severity have also been reported in mouse models of PKD (1, 5, 18). Potential modifiers affecting the severity of the cystic progression have been detected in several of these mouse models (3, 5, 6, 19, 20).
The recessive Oak Ridge polycystic kidney (orpk) mouse mutation causes a form of polycystic kidney disease (PKD) that closely resembles human autosomal recessive PKD (ARPKD)(12). The orpk mutation arose as a result of a transgene insertion on the FVB/N genetic background. The abnormal phenotype is the result of disruption of an endogenous gene (Tg737) and is not associated with the nonexpressed transgene product. Therefore, the mutant phenotype in the orpk line occurs only among transgenic homozygotes derived from crosses between animals heterozygous for the transgene (12).
To evaluate how genetic background might influence the expressivity of the orpk phenotype, we backcrossed the FVB/N orpk (orpk-F) mutation for multiple generations to the C3HeB/FeJLe (C3H) strain to make the mutant locus congenic on the C3H genetic background (orpk-C). The orpk-C mutant mice have a much less severe phenotype, and initial experiments show that these animals live much longer (weeks to months vs. days to weeks), develop renal cysts at a slower rate, and have a less aggressive liver lesion (12). Significant differences in the lesions of orpk mutant mice on the FVB/N and C3H genetic backgrounds suggest the presence of important modifier genes that vary in these strains. We report here a genome-wide scan using backcross and intercross populations with more than 150 markers to map the chromosomal location of the modifier genes that affect the severity of the kidney phenotype.
 |
MATERIALS AND METHODS
|
---|
Mice.
The orpk-F transgenic mice were generated on the FVB/N genetic background following standard pronuclear injection protocols (4). The transgene was moved on the C3HeB/FeJLe genetic background by breeding orpk-F mice heterozygous for the orpk locus to wild-type C3H animals. The heterozygous offspring were subsequently backcrossed to wild-type C3H animals, and this was continued for 10 generations. All animals were handled in accordance with National Institutes of Health guidelines.
Genotyping of animals.
Genotyping of animals for the orpk locus was performed by polymerase chain reaction (PCR) analysis on DNA isolated from tail biopsies using previously described conditions (16). The sequences of the primers for orpk genotyping are RW450 (5'-ATGCACAAAGTTAGCTCTGC-3'), RW451 (5'-GGATCCAGCAATACCTCTCC-3'), and RW452 (5'-CAGGCCATGGATCCAACTCT-3'). RW450 is located in the genomic sequence just 5' to the transgene integration in the mutant animals and serves as a forward primer for both mutant and wild-type Tg737 alleles. RW451 is a reverse primer located within the transgene and produces a 340-bp fragment with RW450 corresponding to the mutant allele. RW452 is a reverse primer corresponding to a genomic sequence that is deleted in orpk mutant animals as a result of the transgene integration and produces a 270-bp PCR fragment with RW450 from the wild-type allele.
Histological analysis of renal, hepatic, and pancreatic tissues.
The kidney, liver, and pancreas from orpk-C and orpk-F mutant animals were fixed in 10% buffered formalin and then processed by paraffin wax embedding. Sections (5 to 8 µm) were cut and attached to glass slides and were then dewaxed and stained with trichrome and/or counterstained with hematoxylin and eosin.
Serum and urine analysis.
Blood was collected from Metofane-anesthetized mutant animals by retro-orbital sinus puncture or at necropsy via cardiac puncture. Urine was collected from mutant animals by a free-catch method on cellophane. Serum was separated in Microtainer tubes (Becton-Dickinson, Rutherford, NJ) following the manufacturers instructions. Serum concentrations of alkaline phosphatase (ALP) and bile acids were measured to assess liver damage and function, respectively. An automated biochemical analyzer (Spectrum, Irving, TX) was used to analyze the serum samples. Urine specific gravity and urine osmolality were measured, and the ratio of urine osmolality to serum osmolality was calculated to assess renal concentrating ability. Serum and urine osmolality were measured using a vapor-pressure osmometer (Wescor, Logan, UT).
Morphometric analysis.
Hematoxylin and eosin-stained histological sections of kidney from 10 orpk-F and 10 orpk-C mice were used for morphometric analysis of the kidney lesion. The percentage of the total area of kidney that is cystic was measured using a computerized imaging system connected to a video scope (Bioquant, Nashville, TN). A cross-section and a longitudinal section of kidney were measured from each animal, and the cystic percentages of these two sections were summed to obtain the total percentage of cystic kidney. This total percentage of cystic kidney was used as the quantitative trait for the QTL analysis of the kidney phenotype (kidney score).
Genome-wide scan design and phenotype scoring.
Male and female mice heterozygous for orpk (orpk/+) were selected from within the orpk-F line and mated to C3H mice to obtain F1 progeny. These progeny were genotyped at 2124 days of age, and the mice that carried the transgene were saved for the backcross and intercross matings. The backcross was established between FVB orpk/+ and F1 animals, and the intercross was established between (C3H x FVB) F1 orpk/+ animals. All progeny from these matings were genotyped at 2124 days of age for the transgene. For the initial genome scan, 100 backcross (BC) progeny and 200 F2 progeny that were homozygous for the transgene (orpk/orpk) were scored for the selected quantitative traits. These mice were killed at 2735 days of age for the phenotype assessment. Sections of kidney and liver were placed in 10% formalin for histological examination. The remainder of the liver and the spleen were collected and snap frozen for DNA isolation. Hematoxylin and eosin-stained histological sections of kidney were used for the morphometric analysis of the kidney lesion, which served as the kidney score for the QTL analysis of the kidney phenotype.
PCR-based genotyping for microsatellite markers.
Selective genotyping of the BC progeny was used to expedite the detection of the chromosomal location of the modifier genes that vary the severity of the kidney phenotype. An end-mapping strategy was initially utilized to determine which chromosomes might contain a modifier. This strategy generates a series of chromosomal haplotypes for each selected progeny by typing only markers at the ends of each chromosome and provides an efficient method for excluding chromosomes that are unlikely to carry a modifier (5, 14). The 20 BC progeny with the highest kidney score (total percent cystic kidneys) were typed using this method to generate a cohort of candidate chromosomes for more detailed analysis. Selective genotyping of the 20 progeny with the highest kidney scores and the 20 progeny with the lowest kidney scores was then performed with markers spaced 1015 cM along these chromosomes to map the modifier loci grossly.
Microsatellite markers with FVB/N and C3H alleles that differ in size by at least 6 bp and that map to within 10 cM of the distal and proximal ends of each chromosome were chosen for the end-mapping strategy. Then, markers that mapped to 10- to 15-cM intervals spaced along the candidate chromosomes were chosen from the list of microsatellite markers that were determined to be variant between the two strains for the low-resolution interval mapping study (15). PCR primer pairs for these markers were purchased from Research Genetics (Huntsville, AL). Genomic DNA was prepared from tail biopsies or liver tissue according to standard protocols, and PCR amplification was performed as follows: an aliquot of 20 ng of genomic DNA was amplified in a 20-µl PCR reaction containing 22 mM Tris·HCl (pH 8.4), 55 mM KCl, 0.66 µM of each primer, 1.65 mM MgCl2, 880 µM dNTPs, and 0.44 U of Taq polymerase (GIBCO-BRL). PCR reactions were amplified in a thermal cycler (MJ Research) as follows: 1 cycle 95°C for 5 min; 30 cycles of 95°C for 45 s, 4555°C for 45 s, and 72°C for 2 min and 30 s with a final elongation step at 72°C for 5 min. PCR products were resolved in 3% agarose gels.
QTL analysis.
Genotyping data obtained from the end-mapping strategy were analyzed with a software program developed by Neuhaus and Beier (14) for interval haplotype analysis. For further analysis of the candidate chromosomes, the 20 progeny with the lowest kidney scores were genotyped for the end markers, and all 40 extreme animals in the two trait sets were genotyped for 23 additional markers within each interval, depending on the size of the interval. These genotyping data and quantitative trait scores were analyzed using Map Manager QT version b28 (11). Permutation tests were done in 1-cM steps for 500 permutations using Map Manager QT. The threshold values of the permutation test are taken from the guidelines of Lander and Kruglyak (8a) and correspond to the 37th, 95th, and 99.9th percentiles, which are suggestive, significant, and highly significant, respectively.
 |
RESULTS
|
---|
Kidney histology.
The kidney lesions in the orpk-C mutant mice are morphologically similar to the orpk-F mutant mice; however, the orpk-C mutants develop renal cysts at a slower rate and have fewer cysts at all stages of cyst development (Fig. 1). Proximal tubular cysts were not detected until 1520 days of age. The progressive shift in cyst localization from proximal tubules to collecting ducts occurs, and the renal cysts are all localized to collecting tubules by 70 days of age, in contrast to 1014 days of age in orpk-F mutants. Morphometric analysis of kidney sections from 20- to 30-day-old orpk-F and orpk-C mutant animals showed a marked difference in the percentage of the total kidney area that was cystic (kidney score). The mean values were 18.03% and 0.58% for the orpk-F and orpk-C mutants, respectively. The mean values for the kidney scores of the 27- to 35-day-old BC and F2 mutant offspring were 6.92% and 3.92%, respectively (Fig. 2).

View larger version (141K):
[in this window]
[in a new window]
|
Fig. 1. Histological sections of the kidneys (hematoxylin and eosin, x10) from a 20-day-old FVB/N orpk mutant mouse (orpk-F; A) and a 28-day-old C3HeB/FeJLe orpk mutant mouse (orpk-C; B). In contrast to the renal histology in the orpk-C mutants, large renal cysts are detectable in the orpk-F mutants at this age.
|
|

View larger version (14K):
[in this window]
[in a new window]
|
Fig. 2. Mean kidney scores (total percent cystic kidneys) of the mutant mice in the parental backgrounds (orpk-C and orpk-F), the backcross progeny (BC), and the intercross progeny (F2). Mean values are presented at top of each column, and bars indicate the standard deviation of the mean.
|
|
Liver histology.
The liver lesions in the orpk-C mutant animals are characterized by multifocal biliary hyperplasia, dysplasia, and portal fibrosis isolated to the portal triad regions (Fig. 3). The most striking feature of the orpk-C phenotype is the formation of excess biliary ductules, secondary to biliary hyperplasia, that expand the portal triad area. The lesions are seen in the first postnatal week and begin as mild biliary hyperplasia and dysplastic duct formation. As the mice mature, the biliary ducts continue to proliferate, forming multifocal areas of biliary hyperplasia and dysplasia centered on the periportal area. There is an absence of the proliferating nonparenchymal cells (oval cells) seen in the orpk-F mutant animals. However, the orpk-C mutants show a greater degree of fibrosis surrounding the dysplastic biliary ductules. The BC and F2 mutant offspring showed a greater degree of variation in their liver lesions, and all had the characteristics of both parental strains in varying degrees.

View larger version (155K):
[in this window]
[in a new window]
|
Fig. 3. Histological sections of the liver (hematoxylin and eosin, x10) from a 24-day-old orpk-F mutant (A) and a 20-day-old orpk-C mutant (B). The liver lesion is more extensive in the orpk-F mutants, with hyperplasia of oval cells (arrows in A) expanding the portal triad area into the periportal space. The liver lesions in the orpk-C mutant animals are not as diffuse and are characterized by multifocal biliary hyperplasia, dysplasia, and portal fibrosis isolated to the portal triad regions (arrow in B).
|
|
Pancreas histology.
The pancreatic lesions in the orpk-F mutant animals begin as pancreatic ductular hyperplasia and dysplasia that disrupt the normal acinar cell architecture. The proliferating ducts become the prominent features of the pancreas as the acinar cells disappear. This disappearance of the acinar cells may be secondary to apoptosis, since there is no evidence of necrosis or inflammation within the lesions. The islets of Langerhans are not affected by the ductular hyperplasia, although there appear to be fewer islets present in the sections examined. As the orpk-F mutant matures, the normal exocrine pancreatic tissue disappears, leaving scattered islets of Langerhans within the mesenteric adipose tissue. This pancreatic lesion is highly variable among orpk-F mutant animals, and the degree of severity seems to correlate with the survival of the mutants.
The pancreatic lesions in the orpk-C mutant animals also begin as pancreatic ductular hyperplasia and dysplasia that expand into the normal acinar cell tissue, but the lesions develop at a much slower rate than in the orpk-F mutants. The proliferating ductules progress to form cyst structures that enlarge to form space-occupying cysts within the abdomen (Fig. 4). The developing cysts can be seen microscopically as early as 1520 days of age and occur prominently in the cranial arm of the pancreas that lies near the greater curvature of the stomach and the spleen. The cysts are lined by a cuboidal epithelium that resembles normal pancreatic ductular epithelium and are fluid filled. The dysplastic pancreatic ductules and cysts are surrounded by extensive fibrosis, and the normal acinar cell structures are reduced (Fig. 5). The islets of Langerhans are not affected by the proliferating ductules, but are crowded by the expanding cysts, as is the rest of the normal pancreatic architecture. The pancreatic lesions of the BC and F2 mutant offspring had the microscopic characteristics of both parental strains with variable degrees of severity. None of the BC mutant offspring and only one F2 mutant offspring had a grossly observable pancreatic cyst by the age of death.

View larger version (104K):
[in this window]
[in a new window]
|
Fig. 4. Gross appearance of a large pancreatic cyst (arrow) occupying the cranial abdominal area in a 90-day-old orpk-C mutant.
|
|

View larger version (109K):
[in this window]
[in a new window]
|
Fig. 5. Histological section of the pancreas (hematoxylin and eosin, x10) from a 24-day-old orpk-C mutant showing the dysplastic pancreatic ductules and cysts (large arrow) surrounded by extensive fibrosis (small arrow).
|
|
Serum and urine analysis.
The orpk-F mutants have urine with low specific gravity (1.0101.020) and low urine:serum osmolality ratio, indicating that the renal concentrating ability is compromised (Fig. 6). The orpk-C mutants retain their renal concentrating ability as measured by urine specific gravity (1.0401.068) and urine:serum osmolality ratio until late in the disease progression (Fig. 6). Serum creatinine and urea nitrogen concentrations are within normal limits in both the orpk-F and orpk-C mutant animals (data not shown).

View larger version (11K):
[in this window]
[in a new window]
|
Fig. 6. Renal concentrating ability was analyzed by comparing the mean urine osmolality:serum osmolality ratio in the orpk-F mutants (n = 10) and the orpk-C mutants (n = 12) and demonstrates a reduced concentrating ability in the orpk-F mutants. Mean values are presented within each column, and bars indicate the standard deviation of the mean.
|
|
Serum concentration of ALP, a liver enzyme used as a marker of biliary abnormalities, and serum concentration of bile acids are markedly elevated in orpk-F mutants, supporting the biliary origin of the lesions. Serum concentrations of ALP and bile acids in the orpk-C mutant animals are only mildly elevated compared with orpk-F mutants (Fig. 7).

View larger version (22K):
[in this window]
[in a new window]
|
Fig. 7. Mean serum alkaline phosphatase (ALP) and bile acid concentrations in orpk-C mice and orpk-F mice demonstrate marked elevations in the orpk-F mutants. Mean values are presented at top of each column, and bars indicate the standard deviation of the mean.
|
|
Genome-wide scan for modifiers.
Interval haplotype analysis of the genotyping data from the end-mapping strategy for the 20 BC progeny with the highest kidney scores (11.966.2%) identified 7 chromosomal intervals that may carry a candidate QTL. The 20 BC progeny with the lowest kidney scores (0.070.7%) and all F2 extreme progeny were then genotyped for the end markers. All 40 extreme progeny from the BC and F2 populations were genotyped for 23 additional markers within the interval. Linkage analysis to individual chromosomes was performed with Map Manager QT, and the FVB locus on distal chromosome 4 was found to be significantly associated with the highest kidney scores in both the BC and F2 populations, with LOD scores ("logarithm of the odds ratio") of 2.34 and 2.1, respectively (Fig. 8). This locus contributes 18% of the variance in the kidney phenotype for the BC mutant offspring and 14% of the variance in the kidney phenotype for the F2 mutant offspring.

View larger version (27K):
[in this window]
[in a new window]
|
Fig. 8. Map Manager QT was used to analyze chromosome 4 genotype and kidney score (total percent cystic kidneys) data from selected BC and F2 extreme progeny. A permutation test was used to calculate the threshold levels shown (dashed lines) for suggestive and significant linkage. LOD, logarithm of the odds ratio.
|
|
 |
DISCUSSION
|
---|
The major phenotypic variations of the orpk mutation on the FVB/N and C3H genetic backgrounds have been evaluated. As in human PKD and other mouse models of PKD, the pathologic features of the orpk phenotype include a spectrum of disease with variable severity in which the primary lesions of the kidneys and liver are expressed to different degrees. In the FVB/N genetic background, the kidney, liver, and pancreatic lesions are very severe, and the majority of the animals die within several weeks after birth. The orpk-C mutant mice have a much less severe phenotype and survive to several months of age. This dramatic decrease in severity is evident after only one backcross generation to C3H, suggesting dominant inheritance of one or more major modifiers from the C3H background. The most striking variation in severity was in the kidney phenotype. The kidney lesions are morphologically similar in the two genetic backgrounds, but the orpk-C mutants develop renal cysts at a slower rate and have fewer cysts at all stages of cyst development. The progressive shift from proximal tubule dilations to collecting tubule cysts is evident in both genetic backgrounds; however, the disease progression is much more rapid in the orpk-F mutants. The kidneys of orpk-C mutants do not show tubular dilations or cysts until they are 23 wk of age; renal cysts are not localized to collecting tubules until they are 23 mo of age, and end-stage renal disease is not evident until much later. These animals also retain their renal concentrating ability as measured by urine specific gravity and urine:serum osmolality ratio until late in the disease progression.
As with the renal lesions, the hepatic abnormalities detected in both genetic backgrounds are similar morphologically, but the orpk-F mutants develop a much more severe lesion, which progresses rapidly. The common feature of the liver phenotype in both genetic backgrounds is the biliary ductular hyperplasia and formation of numerous bile ductule-like structures. The orpk-F mutants also have a proliferation of an immature epithelial cell characterized as an oval cell. This proliferation expands the lesion from the portal triad into the periportal area, invading the hepatic parenchyma. The proliferating oval cells eventually form bridging lesions between portal triads, causing a diffuse hepatic lesion. Conversely, the lesion in the orpk-C mutants remains confined to the portal triads with the formation of an excess of biliary ductular structures that expand the portal area as the lesions progress. There is a marked reduction to an absence of the proliferating oval cells in the orpk-C mutants, but there is a greater degree of fibrosis within the liver lesions. The liver lesions in the C3H background are characterized by multifocal biliary hyperplasia, dysplasia, and portal fibrosis isolated to the portal triad regions. These features more closely resemble the liver lesion in human ARPKD. This difference in the cell types present in the lesions suggests a variation in the differentiation signals or response to these signals between the two genetic backgrounds when the orpk mutation is present. Homogeneous cell lines have been derived previously from the livers of mutant orpk-F and orpk-C mice (17). These cell lines showed differences in expression for a number of genes that have been well characterized in various liver cell lineages (17). It is still unclear how these differences contribute to the phenotype variations between the two strains and warrants further investigation. We were unable to quantify the histological differences between the two parental strains to perform a QTL analysis of the liver phenotype. However, we are in the process of using the serum ALP concentration as a quantitative trait to look at possible modifier genes that may influence this aspect of the liver phenotype. The elevation in serum ALP and bile acid concentrations seen in the orpk-F mutants may simply be a reflection of the severity of the liver lesion in these animals compared with the orpk-C mutants. The elevated ALP may also be due to the proliferation of the oval cells that occurs in the orpk-F mutants.
The pancreatic lesions have not previously been described in the orpk-F mutants and are also more severe in these animals than in the orpk-C mutants. The loss of normal exocrine pancreatic acinar tissue in the orpk-F mutants begins shortly after birth and is a major factor in their runted appearance and poor growth, since the exocrine pancreas is a key player in the digestion of nutrients. The loss of acinar tissue in the orpk-C mutants is much slower, occurring over several weeks; however, they develop pancreatic ductular cysts that grow into large cysts within the cranial abdominal cavity. Pancreatic cysts are uncommon in human ARPKD patients but have been reported to occur in 10% of patients with autosomal dominant PKD (ADPKD) (2). Pancreatic cystic disease has recently been reported in a mouse model of PKD resulting from a targeted mutation that disrupts the mouse homolog of PKD1, the major gene associated with ADPKD (10). The similarity of the pancreatic lesions in the orpk and Pkd1 mouse models suggests a common pathogenesis that leads to abnormal pancreatic tubular epithelial development and warrants further investigation. Once again, we were unable to place a quantitative value on the phenotypic differences between the parental strains to perform a QTL analysis for modifiers of the pancreatic phenotype. Histologically, the mutant BC and F2 offspring showed marked variation in the pancreatic lesions, and only one animal had a grossly visible pancreatic cyst by the age of death (2735 days of age). This was not surprising since most of the pancreatic cysts are not grossly visible in the orpk-C mutants until they are 23 mo of age.
The significant differences in the lesions on the FVB/N and C3H genetic backgrounds suggest the presence of important modifier genes that vary between these strains. This may also be the case in human ARPKD and ADPKD in which the severity of the phenotype and the nature of the extrarenal lesions can be quite variable. Based on the analysis of these phenotypes, the modifier genes appear to be dominant in the C3H background since the decrease in severity is evident in the progeny of the first generation backcross of orpk-F to C3H. The orpk-C mutants also provide a mouse model of the extrarenal lesions associated with other forms of PKD that can be studied to better understand the development of the epithelial proliferation and dysgenesis in the liver and pancreas.
Recently, a new targeted allele of the mouse Tg737 gene, the gene associated with the orpk mutation, has been described (13). The differences in the phenotype between the two mutations suggest that the orpk mutation is a hypomorphic allele of the Tg737 gene with the occurrence of low abundance and normal alternatively spliced forms of mRNA not affected by the transgene (12, 13). These residual transcripts may produce a functional gene product that rescues the orpk mutants through development; however, further studies are needed to determine the gene products arising from these transcripts and their role in the phenotype variations described here.
Stage 1 of our QTL analysis involved a genome-wide scan to detect potential modifier loci of the cystic kidney phenotype in orpk mice. Analysis of separate populations of severely affected and mildly affected mice in the BC progeny identified a possible modifying locus on chromosome 4. A significant association of a recessively inherited FVB/N-related locus on chromosome 4 with cystic kidney disease severity was found, which supports our hypothesis that the modifier genes are dominant in the C3H background. Stage 2, the low-resolution interval mapping stage, was performed with the BC and F2 populations using the Map Manager QTb program. The interval that explains most of the variance is at the distal end of chromosome 4 and has been found to span an interval of 10 cM between D4Mit221 and D4Mit233. The LOD scores for association with increased cystic kidney score for the BC and F2 populations exceeded the significance threshold established by permutation testing on our data sets. These LOD scores do not meet the criteria for significance set for backcross or intercross at the P > 0.05 level (LOD score > 3.3) (9). However, the fact that the markers for this interval segregated with the severe kidney scores with both populations makes the interval a strong candidate to contain a modifier gene. This initial genome-wide scan did not detect one strong locus and suggests that there are multiple loci contributing to the phenotypic variations. Our next step is to genotype the remaining animals in the BC and F2 populations for the chromosomal intervals that were found from the haplotype analysis to carry a candidate QTL. The data can then be analyzed with Map Manager QT to hopefully strengthen the significance of the locus on distal chromosome 4, and find weaker QTL by controlling for the affect of this locus on other weaker loci.
An initial search of databases available for the mouse genome identified several potential candidate genes within the candidate interval on distal chromosome 4 that could play a role in the pathogenesis of the cyst formation. Further narrowing of the interval and research on the genes within the interval is needed before a candidate gene can be identified. Identification of modifier genes will provide insight into the pathogenesis of PKD, and understanding the molecular and cellular nature of their products could lead to targets for diagnostics or therapeutics for PKD.
 |
ACKNOWLEDGMENTS
|
---|
We gratefully acknowledge Drs. Gene Rinchik and Ed Michaud for critical review of the manuscript. We also thank Tommy Jordan and Phil Snow of the University of Tennessee Photography Laboratory for assistance in image generation.
The research described in this report was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant 5-K08-DK-02466 (to C. Sommardahl).
 |
FOOTNOTES
|
---|
Article published online before print. See web site for date of publication (http://physiolgenomics.physiology.org).
Address for reprint requests and other correspondence: C. Sommardahl, Dept. of LACS, UTCVM, 2407 River Rd., Knoxville, TN 37996 (E-mail: csommard{at}utk.edu).
 |
REFERENCES
|
---|
-
Fry JL Jr, Koch WE, Jennette JC, McFarland E, Fried FA, and Mandell J. A genetically determined murine model of infantile polycystic kidney disease. J Urol 134: 828833, 1985.[ISI][Medline]
-
Gabow PA. Autosomal dominant polycystic kidney disease. Am J Kidney Dis 22: 511512, 1993.[ISI][Medline]
-
Guay-Woodford LM, Bryda EC, Christine B, Lindsey JR, Collier WR, Avner ED, DEustchio P, and Flaherty L. Evidence that two phenotypically distinct mouse PKD mutations, bpk and jcpk, are allelic. Kidney Int 50: 11581165, 1996.[ISI][Medline]
-
Hogan BCF and Lacy E. Manipulating the Mouse Embryo: A Laboratory Manual. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press, 1986.
-
Iakoubova OA, Dushkin H, and Beier DR. Localization of a murine recessive polycystic kidney disease mutation and modifying loci that affect disease severity. Genomics 26: 107114, 1995.[ISI][Medline]
-
Iakoubova OA, Dushkin H, and Beier DR. Genetic analysis of a quantitative trait in a mouse model of polycystic kidney disease. Am J Respir Crit Care Med 156: S72S77, 1997.[Abstract/Free Full Text]
-
Kaplan BS, Fay J, Shah V, Dillon MJ, and Barratt TM. Autosomal recessive polycystic kidney disease. Pediatr Nephrol 3: 4349, 1989.[ISI][Medline]
-
Kuida S and Beier DR. Genetic localization of interacting modifiers affecting severity in a murine model of polycystic kidney disease. Genome Res 10: 4954, 2000.[Abstract/Free Full Text]
-
Lander E and Kruglyak L. Genetic dissection of complex traits: guidelines for interpreting and reporting linkage results. Nat Genet11: 241247, 1995.[ISI][Medline]
-
Lander ES and Schork NJ. Genetic dissection of complex traits. Science 265: 20372048, 1994.[ISI][Medline]
-
Lu W, Peissel B, Babakhanlou H, Pavlova A, Fan X, Larson C, Brent G, and Zhou J. Perinatal lethality with kidney and pancreas defects in mice with targetted Pkd1 mutation. Nat Genet 17: 179181, 1997.[ISI][Medline]
-
Manly KF and Olsen JM. Overview of QTL mapping software and introduction to Map Manager QT. Mamm Genome 10: 327334, 1999.[ISI][Medline]
-
Moyer JH, Lee-Tischler MJ, Kwon HY, Schrick JJ, Avner ED, Sweeney WE, Godfrey VL, Cacheiro NLA, Wilkinson JE, and Woychik RP. Candidate gene associated with a mutation causing recessive polycystic kidney disease in mice. Science 264: 13291333, 1994.[ISI][Medline]
-
Murcia NS, Richards WG, Yoder BK, Mucenski ML, Dunlap JR, and Woychik RP. The Oak Ridge Polycystic Kidney (orpk) disease gene is required for left-right axis determination. Development 127: 23472355, 2000.[Abstract/Free Full Text]
-
Neuhaus IM and Beier DR. Efficient localization of mutations by haplotype analysis. Mamm Genome 9: 150154, 1998.[ISI][Medline]
-
Neuhaus IM, Sommardahl CS, Johnson DK, and Beier DR. Microsatellite DNA variants between the FVB/N and C3HeB/FeJLe and C57BL/6J mouse strains. Mamm Genome 8: 506509, 1997.[ISI][Medline]
-
Richards WG, Yoder BK, Isfort RJ, Detilleux PG, Foster C, Neilsen N, Woychik RP, and Wilkinson JE. Oval cell proliferation associated with the murine insertional mutation TgN737Rpw. Am J Pathol 149: 19191930, 1996.[Abstract]
-
Richards WG, Yoder BK, Isfort RJ, Detilleux PG, Foster C, Neilsen N, Woychik RP, and Wilkinson JE. Isolation and characterization of liver epithelial cell lines from wild-type and mutant TgN737Rpw mice. Am J Pathol 150: 11891197, 1997.[Abstract]
-
Takahashi H, Calvet JP, Dittemore-Hoover D, Yoshida K, Grantham JJ, Gattone VH II. A hereditary model of slowly progressive polycystic kidney disease in the mouse. J Am Soc Nephrol 1: 980989, 1991.[Abstract]
-
Woo DDL, Nguyen DK, Khatibi N, and Olsen P. Genetic identification of two major modifying loci of polycystic kidney disease in pcy mice. J Clin Invest 100: 19341940, 1997.[Abstract/Free Full Text]
-
Woo DDL, Tran T, and Kurtz I. Genetic identification of two major modifying loci that control the progression of polycystic kidney disease in the cpk mice (Abstract). J Am Soc Nephrol 7: 1609, 1996.