Role of genetic modifiers in an orthologous rat model of ARPKD

Abstract

Human data and animal models of autosomal recessive polycystic kidney disease (ARPKD) suggest that genetic factors modulate the onset and severity of the disease. We report here for the first time that ARPKD susceptibility is attenuated by introgressing the mutated Pkhd1 disease allele from the polycystic kidney (PCK) rat onto the FHH (Fawn-Hooded Hypertensive) genetic background. Compared with PCK, the FHH.Pkhd1 strain had significantly decreased renal cyst formation that coincided with a threefold reduction in mean kidney weights. Further analysis revealed that the FHH. Pkhd1 is protected from increased blood pressure as well as elevated plasma creatinine and blood urea nitrogen levels. On the other hand, liver weight and biliary cystogenesis revealed no differences between PCK and FHH.Pkdh1, indicating that genes within the FHH genetic background prevent the development of renal, but not hepatic, manifestations of ARPKD. Microarray expression analysis of kidneys from 30-day-old PCK rats revealed increased expression of genes previously identified in PKD renal expression profiles, such as inflammatory response, extracellular matrix synthesis, and cell proliferation genes among others, whereas the FHH.Pkhd1 did not show activation of these common markers of disease. This newly developed strain can serve as a tool to map modifier genes for renal disease in ARPKD and provides further insight into disease variability and pathophysiology.

current convention generally restricts the use of the term “polycystic kidney disease” (PKD) to two genetically distinct conditions: autosomal recessive PKD (ARPKD) (OMIM 263200) and autosomal dominant PKD (ADPKD) (OMIM 173900; 173910). ARPKD occurs less frequently than ADPKD with an incidence of ∼1/20,000 live births (compared with 1/400 for ADPKD) (50), and the phenotypic manifestations are usually far more severe. Approximately 30% of ARPKD cases result in neonatal death, and for patients surviving the neonatal period, ∼50% of affected individuals progress to end-stage renal disease within the first decade of life (5, 42). For patients surviving the perinatal period, a wide range of associated morbidities can develop, including systemic hypertension, renal failure, portal hypertension, and renal and hepatic fibrosis (6, 8). Despite this phenotypic variability, genetic linkage studies indicate that mutations at a single locus are responsible for all phenotypes of ARPKD (15).

ARPKD in humans is caused by mutations in a single gene, PKHD1 (polycystic kidney and hepatic disease 1), which encodes a large transmembrane receptor-like protein, called fibrocystin or polyductin, mapping to human chromosome 6p12 (40). Hundreds of different PKHD1 mutations have been described with alleles ranging from amino acid substitutions to protein frameshifts and resultant truncations (2, 3). Genotype-phenotype correlations are limited but have been observed between the type of PKHD1 mutation (i.e., truncating) and the severity of the PKD phenotype (2). Genetic heterogeneity of the PKHD1 mutation can explain much of the phenotypic variability, but it is also likely that genetic modifiers influence the manifestation of the disease (18, 41). Identifying genetic modifiers of ARPKD in humans has remained difficult due to genetic heterogeneity and environmental variability; therefore alternative methods are necessary to investigate modifiers of this disease.

Rodent models serve as a powerful tool to study PKD because they allow for extensive experimental control of both genetic and environmental variables. A number of murine models of ARPKD have been described (16, 24, 34, 37, 38). These models have been used to successfully identify some genetic modifiers of PKD (4, 35, 48); however, these previously identified genetic modifiers account for only a small percentage of the total genetic variability of the phenotype (35). Thus, there is a need to pursue additional orthologous models of ARPKD that can be utilized to map genetic modifiers of this disease. The PCK (polycystic kidney) rat serves as a particularly useful rodent model of PKD because it shares the same causative gene with humans affected by ARPKD. The PCK strain was developed by generations of inbreeding SD (Sprague-Dawley) rats carrying a spontaneous mutation in the Pkhd1 gene. PCK rats develop both renal and hepatic (biliary) cysts that closely mimic the human disease phenotype. Cysts in both organs are apparent during the first week of life and increase in size and number with age (24). Renal cysts are restricted predominantly to collecting tubule segments, a pattern closely resembling the human ARPKD renal phenotype (24, 30). As the only orthologous rodent model of ARPKD that mimics the human phenotype, the PCK rat is a powerful tool for investigating ARPKD genetic modifiers and pathophysiology of the disease. These findings may be translated into future therapeutic interventions for humans.

To facilitate the search for genetic modifiers that modulate ARPKD disease progression and severity, we sought to generate a congenic rat model that carries the PCK Pkhd1 mutation but changes the onset and/or progression of ARPKD. In the present study we transferred the Pkhd1 mutation from the PCK rat onto the genetic background of the FHH (Fawn-Hooded Hypertensive) rat. This newly developed strain, called FHH.Pkhd1, showed significant amelioration of renal disease but little to no difference in onset or degree of biliary abnormalities. At 6 mo of age, the disease-resistant FHH.Pkhd1 rats had nearly normal kidney weights and dramatically reduced renal cyst formation compared with PCK rats. Renal function, as assessed by serum creatinine and blood urea nitrogen (BUN), was also improved in FHH.Pkhd1 compared with PCK. These observed phenotypic improvements demonstrate that certain genetic elements of the FHH genome protect FHH.Pkhd1 rats from the development of renal manifestations of ARPKD. To define genes and pathways that are associated with renal cystogenesis in our model, we investigated transcriptional changes in kidneys from PCK, SD, FHH, and FHH.Pkhd1 rats by microarray analysis. The transcriptional profile of PCK kidneys closely resembles the profile of previously described PKD kidneys, whereas FHH.Pkhd1 kidneys do not share common markers of the disease. Furthermore, genes that we found to be misregulated to a similar degree in both PCK and FHH.Pkhd1 may provide insight into molecules affected by the Pkhd1 mutation itself and independent of disease progression. Phenotypic analysis of the FHH.Pkhd1 strain revealed that genetic background clearly influences renal manifestations of ARPKD. This strain provides a novel model for investigating the genetic modifiers and pathophysiology of ARPKD.

MATERIALS AND METHODS

Development of the FHH.Pkhd1 congenic strain.

Rats were housed in the Biomedical Resource Center of the Medical College of Wisconsin, an American Association for the Accreditation of Laboratory Animal Care-approved facility. All protocols used in these studies were approved by the local Institution Animal Care and Use Committee. Rats were fed 5LOD diet (Purina, Jefferson, WI) for the duration of all experiments.

A FHH male was crossed with a PCK female, and the male progeny were backcrossed to FHH females for five generations by marker-assisted breeding. In each generation, males were genotyped by fluorescent genotyping as previously described (33) for three markers within the Pkhd1 gene and an additional 98 markers evenly spaced throughout the genome. Males and females from the N5 generation were intercrossed, and pups homozygous for the Phkd1 mutated allele were selected for establishing the FHH.Phkd1 [FHH.PCK-(D9Rat35-D9Rat70)/Mcwi, RGD ID 5147594] colony, which provided the animals for characterization.

To examine the level of contamination of the genome in FHH.Pkhd1, we genotyped both the FHH and FHH.Pkhd1 strains using the Affymetrix customized single nucleotide polymorphism (SNP) chip (803,848 SNPs) according to the manufacturer's recommendations (Affymetrix, Santa Clara, CA). In brief, genomic DNA was digested and ligated to universal adaptors before subsequent PCR amplification. PCR products were purified and fragmented, and the fragmented products were end-labeled with biotin and hybridized to the array. The data files obtained from the Affymetrix customized SNP chip were analyzed by Affymetrix Power Tools (APT) in apt-1.12.0 version (Affymetrix). The genotype calling algorithm used in this study, BRLMM-P, has been previously described by Affymetrix (1). Each array included 803,848 polymorphic SNPs randomly located throughout the genome excluding chromosome Y. The overall success rate of SNP genotypes was 99.1%, with a median inter-SNP distance of 1,008 bp and a mean inter-SNP distance of 3,378 bp. If a SNP genotype could not be determined, the SNP was recognized as “no call” and was excluded from the present study.

From a total of 800,478 SNPs analyzed genome wide, and excluding the congenic region, we found only 0.30% of the genome containing the non-FHH allele after five generations of marker assisted backcross, which is less than expected after eight generations of backcrossing with random recombination (0.39%).

Blood pressure measurement.

All phenotypic analysis was performed on male rat strains carrying the Pkhd1 mutation (PCK and FHH.Pkhd1) as well as the respective genetic control strains (SD for PCK and FHH for FHH.Pkhd1). At 9 wk of age rats were anesthetized with 2% isoflurane, and a blood pressure transmitter (DSI, T11PA-C40) was surgically implanted subcutaneously with the catheter tip secured in the abdominal aorta via the femoral artery. After a 6 day recovery period, blood pressure was measured by radio telemetry in conscious freely moving animals one day each week for 3 h per day. Mean arterial pressure was calculated and averaged over the recording segment.

In vivo proteinuria.

At 15, 24, and 27 wk of age, rats were placed in metabolic cages (Lab Products, Seaford, DE) and allowed to acclimate for 24 h followed by a 24 h urine collection. Urine protein concentration was measured using the Coomassie Plus-Better Bradford Assay kit (Pierce, Rockford, IL).

Plasma and organ analysis.

At the end of the study, 27 wk old rats were anesthetized with isoflurane, and 2 ml of blood was drawn from the abdominal aorta, stored in heparinized vials and spun at 5,000 RPM to separate red blood cells from plasma. The plasma was collected and analyzed for creatinine and BUN by the ACE Autoanalyzer Clinical Chemistry System (Alfa Wassermann, West Caldwell, NJ). Following blood collection, kidneys and livers were collected, weighed, fixed in 10% buffered formalin (Sigma-Aldrich), embedded in paraffin, and sectioned for histological and immunohistological (IHC) analysis.

Kidneys and livers from all four genotypes were also removed from animals at postnatal day 30, 60, 90, or 120 and fixed in 10% buffered formalin and embedded in paraffin. The site of renal cystic lesions was determined by IHC using a collecting tubule-specific lectin and/or antibody.(10) Biotinylated dolichous biflorus agglutinin (DBA) specifically binds to rat collecting tubules, and a monoclonal antibody to the cell surface of collecting tubule principal cells, F-13, was used to verify DBA specificity. Renal cystic lesions in both PCK and FHH.Pkhd1 were primarily of collecting tubule origin.

Collecting tubule cyst growth was assessed using a cyst index as described previously (44). Briefly, a collecting tubule cystic index (CT-CI) was developing by measuring the size of renal cystic lesions in male PCK rats at 30 day intervals starting at day 30. We stained five to ten 8 μM sections, spaced 160 μM apart, with hematoxylin, and renal cysts from each section were measured (the longest axis) to determine the number and size of renal cystic lesions characteristic of PCK animals of that particular age.

Microarray expression analysis.

Both kidneys were harvested from SD, FHH, PCK, FHH.Pkhd1 male rats at 30 days of age (6 animals for each strain). The kidneys were bisected and stored in RNAlater at 4°C for 24 h. RNA was extracted using TRIzol (Invitrogen, Carlsbad, CA) method according to manufacturer's instructions. RNA from each strain was pooled, reverse transcribed, and hybridized to three Affymetrix Rat 230 arrays (Affymetrix). Image data were quantified with Affymetrix Expression Console Software and normalized with Robust Multichip Analysis (http://www.bioconductor.org/) to determine signal log ratios. ANOVA was conducted, and P values were determined using Partek Genomics Suite version 6.2 for all possible pairwise comparisons. To capture the most reliable data, limit the length of gene lists, and facilitate focused pathway analyses, differentially expressed probe sets were defined as those possessing a false discovery rate (FDR) of <0.05 between the compared groups for all analyses. The FDR was set to <0.05 to ensure 95% of the genes in the list are expected to be true positive under the model assumption (only <5% of chance being false positive, P < 0.05). This threshold was set to obtain a reasonable length of gene lists for follow-up analysis. The microarray data has been submitted to the Gene Expression Omnibus (http://www.ncbi.nlm.nih.gov/geo/) under the accession number GSE33056.

To test the reliability of the microarray data, we performed quantitative PCR (qPCR) using primers designed over selected Affymetrix probe sequences. The same RNA used in the microarray experiments was reverse transcribed using the SuperScript III reverse transcription kit according to the manufacturer's instructions (Invitrogen). Unique primers for qPCR were designed directly over the following microarray probe sequences; 1371970_AT (Fam111a), 1368224_AT (Serpina3n), 1391864_AT (Enpp6), 1369663_AT (Ephx2), 1391262_AT (LOC690251), 1397688_AT (Pcdh9), 1379420_AT (RGD1565002), and 1383783_AT (Pcdh9). Primer sequences are listed in Table 1. qPCR reactions were performed using GoTaq qPCR mastermix (Promega, Madison, WI), and amplification cycling was done on the 7900 HT thermocycler (Applied Biosystems) as follows: 95°C for 10:00, followed by 40 cycles of 95°C for 15 s and 60°C for 1 min (acquisition). The target qPCR cycle threshold (Ct) values were compared with GAPDH Ct values as an internal reference. Delta Ct values were compared between SD, PCK, FHH, and FHH.Pkdh1 by a one-way ANOVA followed by a Bonferroni post hoc test using Prism Graph Pad software (La Jolla, CA). The results of the qPCR verified the microarray results (i.e., significantly increased fold change in the microarray was also significantly increased by qPCR, or no significant difference by microarray was also found to be not significantly different by qPCR) (Table 2) for a majority (12 out of 16) of the PCR reactions, demonstrating the reliability of the microarray technique. qPCR data are presented as fold changes between PCK and SD and between FHH and FHH.Pkhd1.

Table 1.

Primer sequences designed over nine different Affymetrix probes used for qPCR

Target	Forward (5′-> 3′)	Reverse (5′-> 3′)
GAPDH	CTGCCTTCTCTTGTGACAAAGTG	GCCGTGGGTAGAGTCATACTGG
1368224_AT	TCCCCTGTATCTGCCTCAAC	ACGGACCAGCAAGTGTTCTT
1371970_AT	GCGATTGCTTCAATTCTTCC	CAATGGGAATGCCATACAAA
1391262_AT	TGTTGGAGCAGATGGTTTGT	TATGGGAGGGACGTATGGAA
1369663_AT	GGGGACACATCGAAGACTGT	GATAAGGCCACGTCCAGAAA
1391864_AT	TCAGAGCTGCTCCAATCAGA	GTGCTTTGGAATGAGCCCTA
1397688_AT	CAGGAGGCACGATTGAGAGT	CCTCCCTTATGTTGCAAAGC
1379420_AT	GCCTTATGGCTAACCTGCAA	TAGTCACTGGCCTGCCTTCT
1383783_AT	CCACGGTTCACCATCAATTT	TCATCAAACACCCCTCACAA

GAPDH primers were used as an internal reference. Sequences are listed in 5′ to 3′ orientation for both forward and reverse primers.

Table 2.

qPCR results validate the microarray results in 13 out of 16 comparisons

		SD vs. PCK qPCR			SD vs. PCK microarray
Probe target	Gene Name	SD	PCK	Significance	Fold Change	significance
1368224_AT	Serpina3	1 ± 0.24	74.47 ± 18.61	P < 0.05	27.44	P < 0.05
1371970_AT	Fam111A	1 ± 0.22	40.62 ± 4.89	P < 0.05	32.08	P < 0.05
1391262_AT	LOC690251	1 ± 0.13	11.86 ± 3.48	P < 0.05	−7.70	P < 0.05
1369663_AT	Ephx2	1 ± 0.17	2.58 ± 0.66	P < 0.05	−6.05	P < 0.05
1391864_AT	Enpp6	1 ± 0.17	0.24 ± 0.06	P < 0.05	−6.65	P < 0.05
1397688_AT	Pcdh9	1 ± 0.47	0.28 ± 0.03	P < 0.05	−2.06	P < 0.05
1379420_AT	RGD1565002	1 ± 0.14	1.14 ± 0.15	NS	1.01	NS
1383783_AT	Pcdh9	1 ± 0.32	0.43 ± 0.09	P < 0.05	−5.96	P < 0.05
		FHH vs. FHH.Pkhd1 qPCR			FHH vs. FHH.Pkhd1 microarray
		FHH	FHH.Pkhd1	Significance	Fold Change	significance
1368224_AT	Serpina3	1 ± 0.14	0.98 ± 0.27	NS	−1.14	NS
1371970_AT	Fam111A	1 ± 0.15	1.84 ± 0.15	P < 0.05	1.12	P < 0.05
1391262_AT	LOC690251	1 ± 0.22	0.48 ± 0.11	NS	−1.14	NS
1369663_AT	Ephx2	1 ± 0.10	1.49 ± 0.22	NS	1.03	NS
1391864_AT	Enpp6	1 ± 0.09	0.95 ± 0.12	NS	1.01	NS
1397688_AT	Pcdh9	1 ± 0.24	0.33 ± 0.07	P < 0.05	−2.32	P < 0.05
1379420_AT	RGD1565002	1 ± 0.08	1.80 ± 0.40	P < 0.05	1.04	NS
1383783_AT	Pcdh9	1 ± 0.17	0.52 ± 0.13	P < 0.05	−3.39	P < 0.05

PCR primers were designed over 9 different Affymetrix probes (probe target) that correspond to 8 different genes (Gene Name). delta Ct values (probe of interest Ct - GAPDH Ct) between SD, PCK, FHH, and FHH.Pkhd1 were compared by 1-way ANOVA to determine significance status (either significant [P < 0.05] or not significant [NS]). Data are presented as PCK fold change vs. SD and FHH.Pkhd1 fold change vs. FHH. Microarray fold changes and significance for each respective Affymetrix probe is listed to the right of the qPCR result. Microarray fold changes in boldface indicate that the result (fold change direction and significance status) was reproduced by qPCR.

Statistical analysis.

Data is presented as mean ± SE. We analyzed data by Student's t-test, one-way ANOVA, or two-way repeated-measures ANOVA followed by the Holm-Sidak multiple comparison test using the Sigma Plot 11.0 software. Data failing normality or equal variance were analyzed by a nonparametric ANOVA on Ranks.

RESULTS

Morphometric and histological evaluation of FHH.Pkhd1 kidneys and livers.

The body weights of FHH.Pkhd1 animals were not significantly different from FHH controls, and PCK body weights were not significantly different from SD controls (data not shown). However, kidneys, from 27 wk old FHH.Pkhd1 rats were significantly smaller in weight compared with those from PCK rats (2.118 g ± 0.199 vs. 5.798 g ± 0.380, P < 0.001), and FHH.Pkhd1 kidneys were not significantly enlarged compared with noncystic SD or FHH kidneys (Fig. 1A). Macroscopically, PCK kidneys appeared enlarged and severely mottled whereas FHH.Pkhd1 kidneys resembled SD and FHH kidneys. The smaller kidney size of the FHH.Pkhd1 rat kidneys appears to be due to attenuated renal cyst formation. Histological analysis revealed that FHH.Pkhd1 rats had smaller and fewer renal collecting duct cysts compared with PCK rats (Fig. 1B). These differences were significant at 30, 60, 90, and 120 days of age as assessed by a morphometrically derived semiquantitative CT-CI (P < 0.001 at all four time points) (Fig. 1C). Differences in renal cyst size and abundance were most striking in younger animals, indicating that the FHH genetic background confers resistance to the development and progressive enlargement of renal cystic lesions of ARPKD.

Fig. 1.

FHH.Pkhd1 rats have smaller kidneys and attenuated cyst formation compared with PCK rats. A: at 27 wk of age, kidneys from FHH.Pkhd1 rats are significantly smaller in weight than kidneys from PCK rats and not significantly larger than SD and FHH control kidneys. Number of animals was 6, 7, 12, and 15 for SD, PCK, FHH, and FHH.Pkdh1, respectively. B: H&E microscopic images taken with ×4 objective of PCK (1 and 3) and FHH.Pkhd1 (2 and 4) kidneys at postnatal day 30 (PN30) and postnatal day 90 (PN90). Bar indicates 100 μM. C: renal cysts are smaller and less abundant in FHH.Pkhd1 kidneys compared with PCK kidneys at 30, 60, 90, and 120 days of age as quantified by collecting tubule cystic index (CTCI). Number of animals was 10 for both PCK and FHH.Pkhd1 at each time point. Data are presented as means ± SE. ***P < 0.001. FHH, Fawn-Hooded Hypertensive; Pkhd1, polycystic kidney and hepatic disease 1 gene; PCK, polycystic kidney; SD, Sprague-Dawley; H&E, hematoxylin and eosin.

Livers from FHH.Pkhd1 rats tended to be slightly smaller (36.22 g ± 2.68) than livers from PCK rats (53.83 g ± 8.08), but this difference was not significant at 27 wk of age (Fig. 2A). Bile duct dilatation together with selective fibrosis of portal areas defines the congenital hepatic fibrosis phenotype (35). Liver fibrosis in the PCK and FHH.Pkhd1 occurs around virtually every cystic bile duct, and neither the PCK nor the FHH.Pkhd1 showed selective fibrosis (Fig. 2B). Therefore, we quantified liver fibrosis as an overall estimation of liver damage. The degree of fibrosis was evaluated by morphometric determination of the percent of area fibrosis using Masson's Trichrome staining. At 120 days of age the percent of the left lobe that was fibrotic in the FHH.Pkhd1 livers was 21.30 ± 5.6%. This was not significantly different from the percent of the left lobe of PN120 PCK male livers that was fibrotic (23.50 ± 4.8%) (Fig. 2C). Both PCK and FHH.Pkhd1 livers were significantly larger and had increased incidence of cyst formation compared with both SD and FHH livers (n = 6, 12, 7, and 15 for SD, FHH, PCK, and FHH.Pkhd1 respectively). These data indicate that the protective genetic elements on the FHH background play a more pronounced role in renal cyst formation and growth than in biliary cyst formation.

Fig. 2.

FHH.Pkhd1 livers are not significantly protected from hepatic manifestations of the Pkhd1 mutation. A: at 27 wk of age FHH.Pkhd1 (n = 15) livers are not significantly smaller than PCK livers (n = 7). B: H&E stained liver sections (×10 objective) from 120-day-old FHH.Pkhd1 and PCK rats demonstrate colocalization of fibrosis and biliary cysts. C: FHH.Pkhd1 (n = 8) and PCK (n = 8) showed no significant differences in percent area liver fibrosis as assessed by morphometric quantification. NS, not significant.

Assessment of blood pressure and renal function in the FHH.Pkhd1 strain.

Blood pressure increased with time in the PCK rat reaching 122 ± 8 mmHg at 27 wk of age compared with the SD control, which remained relatively stable at 93 ± 5 mmHg (P = 0.007) (Fig. 3A). Introgression of the Pkhd1 locus from the PCK onto the FHH background did not induce changes in blood pressure compared with the control FHH strain (124 ± 4 mmHg in FHH.Pkhd1 and 123 ± 8 in FHH rats at 27 wk) (Fig. 3B); thus, blood pressure differences do not necessarily correlate with the presence of the Pkhd1 mutation.

Fig. 3.

Blood pressure and renal impairment are elevated in PCK, but not FHH.Pkhd1, compared with respective control strains. Blood pressure was measured in awake rats by radiotelemetry starting at 11 wk through 27 wk of age. MAP, mean arterial pressure. A: PCK rats have significantly elevated blood pressure compared with SD control strain beginning at week 20, and blood pressure continues to increase through 27 wk of age. B: the FHH.Pkdh1 strain demonstrated no changes in blood pressure compared with the FHH control strain. Data is presented as means ± SE. Number of animals is 6, 7, 6, and 8 for SD, FHH, PCK, and FHH.Pkdh1. C: 24 h proteinuria (UpV) measured at 15, 24, and 27 wk of age is elevated in PCK but not FHH.Pkhd1 compared with control. UpV is significantly higher in PCK compared with SD, whereas UpV is not significantly elevated in FHH.Pkhd1 compared with FHH. Both FHH and FHH.Pkhd1 excrete higher UpV compared with SD because FHH is a genetic model of progressive renal disease, and therefore this strain is proteinuric regardless of the Pkdh1 allele. Number of animals is 6, 6, 12, and 15 for SD, PCK, FHH, and FHH.Pkhd1, respectively. Data are presented as means ± SE. D: at 27 wk of age FHH.Pkhd1 rats have attenuated azotemia compared with PCK rats. PCK rats have higher blood urea nitrogen (BUN) compared with FHH.Pkhd1, SD, and FHH, whereas FHH.Pkhd1 demonstrates no elevation in BUN. E: plasma creatinine is also elevated in the PCK compared with the 3 other strains. FHH.Pkhd1 plasma creatinine is not significantly higher than SD or FHH. Data are presented as means ± SE. Number of animals is 6, 7, 6, and 9 for SD, FHH, PCK and FHH.Pkdh1, respectively. *P < 0.05, **P < 0.01, ***P < 0.001.

PCK demonstrated significantly higher urinary protein excretion (UpV) (up to ∼4-fold increase) than SD at 15, 24, and 27 wk of age (P < 0.001 at all time points), whereas FHH.Pkhd1 did not show a significant difference in UpV vs. FHH at any time point, although it had a tendency to be higher (Fig. 3C). UpV of both the FHH.Pkhd1 and FHH was elevated compared with SD. The FHH rat is a spontaneous model of hypertension that ultimately results in progressive renal disease, and it is well documented that FHH rats excrete high levels of UpV. This did not significantly increase when the Pkhd1 mutation was introduced.

Plasma BUN (36.33 ± 3.07) was elevated in PCK relative to FHH (20.50 ± 3.22 mg/ml P < 0.001), SD (22.33 ± 1.53 mg/ml, P = 0.005), and FHH.Pkhd1 (25.89 ± 3.59 mg/ml, P = 0.019) (Fig. 3D). PCK animals also demonstrated significantly higher plasma creatinine levels (0.80 ± 0.10 mg/dl) compared with FHH (0.48 ± 0.02 mg/dl, P < 0.001), SD (0.47 mg/dl ± 0.03 P < 0.001) and FHH.Pkhd1 (0.542 ± 0.05 mg/dl, P = 0.002) (Fig. 3E). FHH.Pkhd1 did not show a significant increase in BUN or plasma creatinine compared with either SD or FHH. Despite similar levels of proteinuria in both the PCK and FHH.Pkhd1 strains, renal function as assessed by BUN and plasma creatinine was clearly better in the FHH.Pkhd1 compared with PCK.

Detection of differentially expressed genes in PCK and FHH.Pkhd1 kidneys.

RNA from kidneys of 30 day old SD, PCK, FHH and FHH.Pkhd1 rats was used for gene expression profiling. This time point was selected because the differences between PCK and FHH.Pkhd1 renal cyst formation was greatest at this time point. To account for expression differences that may depend solely on genomic background, gene expression of each cystic strain was compared with gene expression of the appropriate control strain. Thus, gene expression of PCK was compared with SD to determine differential expression of PCK genes, and gene expression of FHH.Pkhd1 was compared with FHH to determine differential expression of FHH.Pkhd1 genes.

Differentially expressed genes were divided into two categories: 1) genes up- or downregulated in PCK and not expressed in the same pattern in FHH.Pkhd1 were considered “uniquely misregulated” in PCK, 2) genes up- or downregulated in FHH.Pkhd1 and not expressed in the same pattern in PCK were considered “uniquely misregulated” in FHH.Pkhd1. We found 411 (231 up and 180 down) genes to be uniquely misregulated in PCK and 237 (132 up and 105 down) genes uniquely misregulated in FHH.Pkhd1. These gene lists were each interrogated by The Database for Annotation, Visualization and Integrated Discovery (20, 21) to determine biological processes that were uniquely modulated in each group (Table 3).

Table 3.

Top networks queried by DAVID gene ontology annotation of genes uniquely misregulated in PCK or in FHH.Pkhd1 kidneys

ID	Term	Count	%	P Value
A: PCK misregulated biological processes
GO:0009611	response to wounding	26	7.28	4.27E-07
GO:0003013	circulatory system process	14	3.92	7.26E-06
GO:0008015	blood circulation	14	3.92	7.26E-06
GO:0016042	lipid catabolic process	13	3.64	8.40E-06
GO:0055114	oxidation reduction	29	8.12	1.22E-05
GO:0006954	inflammatory response	16	4.48	1.53E-05
GO:0016064	immunoglobulin mediated immune response	8	2.24	2.95E-05
GO:0019724	B cell mediated immunity	8	2.24	3.87E-05
GO:0002526	acute inflammatory response	10	2.80	4.83E-05
GO:0006775	fat-soluble vitamin metabolic process	7	1.96	5.10E-05
GO:0006952	defense response	21	5.88	6.76E-05
GO:0008202	steroid metabolic process	13	3.64	8.02E-05
GO:0019884	antigen processing and presentation of exogenous antigen	6	1.68	8.90E-05
GO:0002495	antigen processing and presentation of peptide antigen via MHC class II	5	1.40	1.06E-04
GO:0019886	antigen processing and presentation of exogenous peptide antigen via MHC class II	5	1.40	1.06E-04
GO:0006955	immune response	21	5.88	1.11E-04
GO:0002449	lymphocyte mediated immunity	8	2.24	1.42E-04
GO:0002504	antigen processing and presentation of peptide or polysaccharide antigen via MHC class II	5	1.40	1.87E-04
GO:0006873	cellular ion homeostasis	19	5.32	2.26E-04
GO:0002460	adaptive immune response based on somatic recombination of immune receptors built from immunoglobulin superfamily domains	8	2.24	2.34E-04
GO:0002250	adaptive immune response	8	2.24	2.34E-04
GO:0055082	cellular chemical homeostasis	19	5.32	2.67E-04
GO:0030168	platelet activation	5	1.40	4.70E-04
GO:0002478	antigen processing and presentation of exogenous peptide antigen	5	1.40	4.70E-04
GO:0002443	leukocyte mediated immunity	8	2.24	5.16E-04
GO:0006766	vitamin metabolic process	8	2.24	6.06E-04
GO:0050801	ion homeostasis	19	5.32	6.10E-04
GO:0048002	antigen processing and presentation of peptide antigen	6	1.68	7.79E-04
GO:0006631	fatty acid metabolic process	12	3.36	8.73E-04
GO:0019725	cellular homeostasis	20	5.60	9.49E-04
GO:0019882	antigen processing and presentation	8	2.24	9.51E-04
GO:0048878	chemical homeostasis	21	5.88	0.001
GO:0050778	positive regulation of immune response	10	2.80	0.001
GO:0050817	coagulation	7	1.96	0.001
GO:0007596	blood coagulation	7	1.96	0.001
GO:0002684	positive regulation of immune system process	13	3.64	0.002
GO:0007599	hemostasis	7	1.96	0.002
GO:0030005	cellular di-, tri-valent inorganic cation homeostasis	12	3.36	0.002
GO:0002252	immune effector process	9	2.52	0.002
GO:0042592	homeostatic process	26	7.28	0.002
GO:0044242	cellular lipid catabolic process	7	1.96	0.002
GO:0007610	behavior	18	5.04	0.003
GO:0055066	di-, tri-valent inorganic cation homeostasis	12	3.36	0.003
GO:0050878	regulation of body fluid levels	8	2.24	0.003
GO:0006953	acute-phase response	5	1.40	0.003
GO:0006721	terpenoid metabolic process	5	1.40	0.003
GO:0016125	sterol metabolic process	7	1.96	0.003
GO:0008217	regulation of blood pressure	8	2.24	0.004
GO:0055080	cation homeostasis	13	3.64	0.004
GO:0009062	fatty acid catabolic process	5	1.40	0.005
GO:0030003	cellular cation homeostasis	12	3.36	0.005
GO:0010033	response to organic substance	30	8.40	0.005
GO:0042060	wound healing	10	2.80	0.006
GO:0007626	locomotory behavior	11	3.08	0.006
GO:0001775	cell activation	12	3.36	0.006
GO:0019748	secondary metabolic process	7	1.96	0.007
GO:0042493	response to drug	14	3.92	0.009
GO:0045059	positive thymic T cell selection	3	0.84	0.009
GO:0043067	regulation of programmed cell death	23	6.44	0.010
GO:0010941	regulation of cell death	23	6.44	0.011
GO:0002520	immune system development	12	3.36	0.011
GO:0032101	regulation of response to external stimulus	9	2.52	0.011
GO:0008203	cholesterol metabolic process	6	1.68	0.011
GO:0043368	positive T cell selection	3	0.84	0.012
GO:0006874	cellular calcium ion homeostasis	9	2.52	0.013
GO:0048584	positive regulation of response to stimulus	11	3.08	0.013
GO:0043933	macromolecular complex subunit organization	20	5.60	0.013
GO:0010035	response to inorganic substance	12	3.36	0.013
GO:0008544	epidermis development	7	1.96	0.013
GO:0050880	regulation of blood vessel size	5	1.40	0.013
GO:0035150	regulation of tube size	5	1.40	0.013
GO:0030193	regulation of blood coagulation	4	1.12	0.014
GO:0006635	fatty acid beta-oxidation	4	1.12	0.014
GO:0006776	vitamin A metabolic process	4	1.12	0.014
GO:0055074	calcium ion homeostasis	9	2.52	0.014
GO:0001558	regulation of cell growth	9	2.52	0.015
GO:0045861	negative regulation of proteolysis	4	1.12	0.015
GO:0003018	vascular process in circulatory system	5	1.40	0.015
GO:0042311	vasodilation	4	1.12	0.017
GO:0030855	epithelial cell differentiation	7	1.96	0.017
GO:0042981	regulation of apoptosis	22	6.16	0.017
GO:0060429	epithelium development	11	3.08	0.017
GO:0030162	regulation of proteolysis	5	1.40	0.018
GO:0001523	retinoid metabolic process	4	1.12	0.018
GO:0016101	diterpenoid metabolic process	4	1.12	0.018
GO:0031214	biomineral formation	4	1.12	0.018
GO:0010038	response to metal ion	9	2.52	0.019
GO:0007398	ectoderm development	7	1.96	0.019
GO:0006875	cellular metal ion homeostasis	9	2.52	0.020
GO:0032844	regulation of homeostatic process	7	1.96	0.020
GO:0006935	chemotaxis	6	1.68	0.020
GO:0042330	taxis	6	1.68	0.020
GO:0048534	hemopoietic or lymphoid organ development	11	3.08	0.020
GO:0015671	oxygen transport	3	0.84	0.021
GO:0050818	regulation of coagulation	4	1.12	0.022
GO:0010959	regulation of metal ion transport	6	1.68	0.022
GO:0048545	response to steroid hormone stimulus	12	3.36	0.022
GO:0043068	positive regulation of programmed cell death	13	3.64	0.024
GO:0006720	isoprenoid metabolic process	5	1.40	0.024
GO:0055065	metal ion homeostasis	9	2.52	0.024
GO:0010942	positive regulation of cell death	13	3.64	0.025
GO:0045582	positive regulation of T cell differentiation	4	1.12	0.026
GO:0040008	regulation of growth	12	3.36	0.026
GO:0002253	activation of immune response	6	1.68	0.028
GO:0045061	thymic T cell selection	3	0.84	0.028
GO:0009991	response to extracellular stimulus	12	3.36	0.028
GO:0008629	induction of apoptosis by intracellular signals	4	1.12	0.030
GO:0045621	positive regulation of lymphocyte differentiation	4	1.12	0.030
GO:0051241	negative regulation of multicellular organismal process	8	2.24	0.030
GO:0051085	chaperone mediated protein folding requiring cofactor	3	0.84	0.032
GO:0009725	response to hormone stimulus	17	4.76	0.032
GO:0032846	positive regulation of homeostatic process	4	1.12	0.032
GO:0042573	retinoic acid metabolic process	3	0.84	0.036
GO:0030195	negative regulation of blood coagulation	3	0.84	0.036
GO:0009636	response to toxin	5	1.40	0.036
GO:0051924	regulation of calcium ion transport	5	1.40	0.036
GO:0043269	regulation of ion transport	6	1.68	0.036
GO:0048145	regulation of fibroblast proliferation	4	1.12	0.037
GO:0001869	negative regulation of complement activation, lectin pathway	2	0.56	0.037
GO:0001868	regulation of complement activation, lectin pathway	2	0.56	0.037
GO:0010890	positive regulation of sequestering of triglyceride	2	0.56	0.037
GO:0016485	protein processing	6	1.68	0.039
GO:0022411	cellular component disassembly	4	1.12	0.039
GO:0006692	prostanoid metabolic process	3	0.84	0.040
GO:0006693	prostaglandin metabolic process	3	0.84	0.040
GO:0022405	hair cycle process	4	1.12	0.042
GO:0022404	molting cycle process	4	1.12	0.042
GO:0034440	lipid oxidation	4	1.12	0.042
GO:0019395	fatty acid oxidation	4	1.12	0.042
GO:0001942	hair follicle development	4	1.12	0.042
GO:0042303	molting cycle	4	1.12	0.044
GO:0042633	hair cycle	4	1.12	0.044
GO:0006458	‘de novo’ protein folding	3	0.84	0.044
GO:0045058	T cell selection	3	0.84	0.044
GO:0009268	response to pH	3	0.84	0.044
GO:0015669	gas transport	3	0.84	0.044
GO:0051084	‘de novo’ posttranslational protein folding	3	0.84	0.044
GO:0051604	protein maturation	6	1.68	0.045
GO:0051240	positive regulation of multicellular organismal process	10	2.80	0.047
GO:0051605	protein maturation by peptide bond cleavage	5	1.40	0.047
GO:0043065	positive regulation of apoptosis	12	3.36	0.049
GO:0050819	negative regulation of coagulation	3	0.84	0.049
GO:0045597	positive regulation of cell differentiation	10	2.80	0.050
Table 3. B
B: FHH.Pkhd1 misregulated biological processes
GO:0006700	C21-steroid hormone biosynthetic process	3	1.71	0.001
GO:0048864	stem cell development	4	2.29	0.002
GO:0045449	regulation of transcription	27	15.43	0.002
GO:0034754	cellular hormone metabolic process	5	2.86	0.003
GO:0006397	mRNA processing	8	4.57	0.003
GO:0042445	hormone metabolic process	6	3.43	0.003
GO:0048863	stem cell differentiation	4	2.29	0.004
GO:0001568	blood vessel development	8	4.57	0.005
GO:0008207	C21-steroid hormone metabolic process	3	1.71	0.005
GO:0001944	vasculature development	8	4.57	0.006
GO:0016071	mRNA metabolic process	8	4.57	0.007
GO:0007617	mating behavior	3	1.71	0.007
GO:0016055	Wnt receptor signaling pathway	5	2.86	0.007
GO:0010817	regulation of hormone levels	6	3.43	0.014
GO:0019098	reproductive behavior	3	1.71	0.015
GO:0006396	RNA processing	9	5.14	0.017
GO:0007618	mating	3	1.71	0.021
GO:0008380	RNA splicing	6	3.43	0.024
GO:0019827	stem cell maintenance	3	1.71	0.025
GO:0000377	RNA splicing, via transesterification reactions with bulged adenosine as nucleophile	5	2.86	0.025
GO:0000375	RNA splicing, via transesterification reactions	5	2.86	0.025
GO:0000398	nuclear mRNA splicing, via spliceosome	5	2.86	0.025
GO:0048514	blood vessel morphogenesis	6	3.43	0.027
GO:0042446	hormone biosynthetic process	3	1.71	0.028
GO:0006694	steroid biosynthetic process	4	2.29	0.030
GO:0051705	behavioral interaction between organisms	3	1.71	0.034
GO:0032528	microvillus organization	2	1.14	0.036
GO:0030033	microvillus assembly	2	1.14	0.036
GO:0045777	positive regulation of blood pressure	3	1.71	0.036
GO:0043062	extracellular structure organization	5	2.86	0.037
GO:0043627	response to estrogen stimulus	5	2.86	0.046
GO:0030030	cell projection organization	8	4.57	0.046
GO:0048545	response to steroid hormone stimulus	7	4.00	0.050

A: genes up- or downregulated in PCK (>1.45 or < −1.45-fold) and not differentially expressed (<1.2 or > −1.2), or expressed in the opposite direction in FHH.Pkhd1 were considered “uniquely misregulated” in PCK and uploaded into to identify enriched biological processes (BP) to generate part A. B: genes up- or downregulated in FHH.Pkhd1(>1.45 of <−1.45-fold) and not differentially expressed (<1.2 or > −1.2), or expressed in the opposite direction in PCK were considered “uniquely misregulated” in FHH.Pkhd1 and uploaded into DAVID to identify enriched biological processes for part B. Gene ontology (GO) identifiers are listed in the lefthand column. Count indicates the number of molecules from the gene list in each BP, % is the percent of molecules out of the whole gene list that fell into each BP.

Biological processes that were uniquely differentially expressed in PCK involved immunological processes, inflammation, and epithelial cell differentiation among others (Table 3A). These networks fell into categories that correlate with disease pathways that have been previously described in human and rodent PKD (47). As expected, PCK kidneys uniquely overexpressed some previously described expression markers of PKD such as Aqp3, Myc, Clu, Ren, laminins, and collagens (11–14, 36). FHH.Pkhd1 kidneys did not show upregulation of some biological processes common to PKD such as cell proliferation or inflammatory response (Table 3B). Since the FHH.Pkhd1 animals were resistant to progression of renal cystic disease, genes uniquely expressed in this strain may play a role in protection from renal cystogenesis.

The most highly misregulated genes in PCK and FHH.Pkhd1 kidneys are listed in Table 4. A small number of the genes found to be most differentially expressed in the PCK rat have been previously associated with renal cystogenesis (i.e., Alox15) (28), or renal damage (i.e., Kim1) (25). Despite substantial expression differences in the PCK rat, many of the misregulated genes have not been linked to the pathophysiology of renal cyst formation or progressive enlargement before now.

Table 4.

Top 15 most up- or downregulated genes in PCK and FHH.Pkhd1 compared with their respective controls

Gene Symbol	Gene Name	Fold Change
		PCK vs. SD (up-)
Fam111A	family with sequence similarity 111, member A	32.08
Serpina3	serpin peptidase inhibitor, clade A , member 3	27.44
Havcr1(Kim1)	hepatitis A virus cellular receptor 1	8.90
Akr1b15	aldo-keto reductase family 1, member B15	5.54
Dusp15	dual specificity phosphatase 15	4.96
Cps1	carbamoyl-phosphate synthase 1, mitochondrial	4.83
Col17A1	collagen, type XVII, alpha 1	4.15
Slc7a12	solute carrier family 7, member 12	3.64
Parm1	peptidase domain containing associated with muscle regeneration 1	3.63
Snap91	synaptosomal-associated protein, 91 kDa homolog	3.61
Lypd2	LY6/PLAUR domain containing 2	3.32
Lamc2	laminin, gamma 2	3.22
Pard6 g	par-6 partitioning defective 6 homolog gamma	3.14
Hrg	histidine-rich glycoprotein	3.06
Mal2	mal, T-cell differentiation protein 2	3.03
		PCK vs. SD (down-)
LOC690251	Sumo1/sentrin/SMT3 specific peptidase 5	−7.70
Enpp6	ectonucleotide pyrophosphatase/phosphodiesterase 6	−6.65
Ephx2	epoxide hydrolase 2, cytoplasmic	−6.05
Hla-dqa1	major histocompatibility complex, class II, DQ alpha 1	−6.00
Mx1	myxovirus resistance 1, interferon-inducible protein p78	−5.57
Resp18	regulated endocrine-specific protein 18 homolog	−5.01
Rt1-t24-3	RT1 class I, locus T24, gene 3	−4.15
Eml1	echinoderm microtubule associated protein like 1	−3.86
Ahsp	alpha hemoglobin stabilizing protein	−3.76
Ugt2b15	UDP glucuronosyltransferase 2 family, polypeptide B15	−3.64
Plekhh1	pleckstrin homology domain containing, family H member 1	−3.57
Pcdh9	protocadherin 9	−3.50
Gc	group-specific component	−3.27
Slco1a1	solute carrier organic anion transporter family, member 1a1	−3.14
Alox15	arachidonate 15-lipoxygenase	−3.05
		FHH.Pkhd1 vs. FHH (up-)
Ddx6	DEAD (Asp-Glu-Ala-Asp) box polypeptide 6	7.11
Cd2ap	CD2-associated protein	4.84
Cyp11B2	cytochrome P450, family 11, subfamily B, polypeptide 2	4.57
Akr1b7	aldo-keto reductase family 1, member B7	3.51
Cdkn1B	cyclin-dependent kinase inhibitor 1B	3.35
Scd	stearoyl-CoA desaturase	2.89
Bptf	bromodomain PHD finger transcription factor	2.69
Rab1b	RAB1B, member RAS oncogene family	2.44
S100a9	S100 calcium binding protein A9	2.36
Sfrp1	secreted frizzled-related protein 1	2.33
Znf609	zinc finger protein 609	2.29
LOC100363366	amyloid beta (A4) precursor-like protein 2-like	2.24
Lsm14B	LSM14B, SCD6 homolog B	2.21
Cbx1	chromobox homolog 1	2.15
Dnajc7	DnaJ (Hsp40) homolog, subfamily C, member 7	2.14
		FHH.Pkhd1 vs. FHH (down-)
Slco1a1	solute carrier organic anion transporter family, member 1a1	−18.26
Dhrs7	similar to dehydrogenase/reductase member 7	−10.75
Ugt2b15	UDP glucuronosyltransferase 2 family, polypeptide B15	−10.12
Cyp4a22	cytochrome P450, family 4, subfamily A, polypeptide 22	−9.70
Snca	synuclein, alpha	−9.33
Akr1c12	aldo-keto reductase family 1, member C12	−7.64
Akr1c4	aldo-keto reductase family 1, member C4	−6.22
Tff3	trefoil factor 3	−5.05
Slc7a12	solute carrier family 7, member 12	−3.77
RGD1564865	similar to 20-alpha-hydroxysteroid dehydrogenase	−3.73
Pzp	pregnancy-zone protein	−3.60
Cyp2d9	cytochrome P450, family 2, subfamily d, polypeptide 9	−3.29
Apcs	amyloid P component, serum	−3.01
Trpv1	transient receptor potential cation channel, subfamily V, member 1	−2.91
Pcdh9	protocadherin 9	−2.80

We identified genes that were up- or downregulated in the same direction in both the FHH.Pkhd1 and PCK (“commonly misregulated”) (Table 5). Similar gene regulation in both PCK and FHH.Pkhd1 kidneys suggests that the expression of these genes may be in response to the Pkhd1 mutation itself and independent of to the severity of cystic disease. These genes that are misexpressed to a similar degree in PCK and FHH.Pkhd1 may provide insight into a number of unknown biological functions of fibrocystin. Several of the genes identified to be commonly misregulated in both ARPKD strains have been directly associated with renal function, or renal cysts before, [Cps1,(26) Areg (9, 39), and P2RY1 (19)], but many of these genes have not been previously identified.

Table 5.

Genes found to be up- or downregulated in the same direction in PCK and FHH.Pkhd1 kidneys at 30 days of age

		Fold Change
Gene Symbol	Gene Name	FHH.Pkhd1 vs. FHH	PCK vs. SD
Common upregulated genes
Cps1	carbamoyl-phosphate synthase 1, mitochondrial	1.45	4.83
Mttp	microsomal triglyceride transfer protein	2.08	2.63
Kng1	kininogen 1	1.47	2.53
Snx10	sorting nexin 10	1.44	2.28
Fgb	fibrinogen beta chain	1.61	1.98
Klf5	Kruppel-like factor 5	1.47	1.75
Slc22A13	solute carrier family 22, member 13	1.41	1.63
Gabrp	gamma-aminobutyric acid A receptor, pi	1.46	1.49
Rab1b	RAB1B, member RAS oncogene family	2.44	1.45
S100A9	S100 calcium binding protein A9	2.36	1.45
Igfbp5	insulin-like growth factor binding protein 5	1.44	1.44
LOC100302372	hypothetical protein LOC100302372	1.82	1.43
Scnn1A	sodium channel, nonvoltage-gated 1 alpha	1.81	1.41
Common downregulated genes
Resp18	regulated endocrine-specific protein 18 homolog	−1.50	−5.01
Ugt2B15	UDP glucuronosyltransferase 2 family, polypeptide B15	−10.12	−3.64
Pcdh9	protocadherin 9	−2.80	−3.50
Slco1a1	solute carrier organic anion transporter family, member 1a1	−18.26	−3.14
Areg	amphiregulin	−1.48	−2.42
Cyp2C9	cytochrome P450, family 2, subfamily C, polypeptide 9	−1.46	−2.22
P2RY1	purinergic receptor P2Y, G-protein coupled, 1	−1.70	−2.12
Akr1C4	aldo-keto reductase family 1, member C4	−6.22	−2.07
Mmp9	matrix metallopeptidase 9	−1.42	−1.81
Akr1c12	aldo-keto reductase family 1, member C12	−7.64	−1.77
Dhrs7	similar to dehydrogenase/reductase member 7	−10.75	−1.76
TrpV1	transient receptor potential cation channel, subfamily V, member 1	−2.91	−1.74
Col9A1	collagen, type IX, alpha 1	−1.96	−1.64
Bhlhe41	basic helix-loop-helix family, member e41	−1.79	−1.63
Ttc21B	tetratricopeptide repeat domain 21B	−1.58	−1.61
Ganc	glucosidase, alpha; neutral C	−1.50	−1.58
Tnrrsf11B	tumor necrosis factor receptor superfamily, member 11b	−1.51	−1.58
Igh-2	immunoglobulin heavy chain 2	−1.79	−1.56
Tff3	trefoil factor 3	−5.05	−1.54
LOC100361122	SRY-box containing gene 9	−1.44	−1.51
Snca	synuclein, alpha	−9.33	−1.51
RGD1562717	similar to ABI gene family, member 3 binding protein	−1.43	−1.48
Fgf13	fibroblast growth factor 13	−1.46	−1.45
C1QL3	complement component 1, q subcomponent-like 3	−1.41	−1.43
Igtp	interferon gamma induced GTPase	−1.57	−1.42
Ccrn4L	CCR4 carbon catabolite repression 4-like	−1.58	−1.40

This list includes genes that are differentially expressed by >1.4, or <−1.4-fold in both FHH.Pkhd1 and PCK compared to their respective controls.

DISCUSSION

The genes responsible for both recessive and dominant forms of PKD have been identified (22, 32). A handful of key molecular pathways have been associated with ARPKD cyst formation and progressive enlargement (47); however, the precise molecular mechanisms that govern disease pathogenesis remain unknown. Previous publications have documented that the genetic background influences the severity of monogenic diseases such as PKD in humans (41, 49), but identifying genetic modifiers in humans remains very difficult. Therefore, many groups have turned to genetic mapping of PKD modifiers in rodent models. To further facilitate the search for genes that modulate ARPKD disease progression and severity we created a novel congenic rat model by transferring the Pkhd1 allele from the PCK rat onto the FHH genetic background. Phenotypic analysis demonstrated that the newly developed FHH.Pkhd1 strain is resistant to renal cyst development and has improved renal function compared with the PCK rat.

FHH, FHH.Pkdh1, and PCK animals were all proteinuric, but the mechanism of UpV is not the same in these strains. Increased UpV in the PCK rat is due to the renal lesions resulting from the Pkhd1 mutation, whereas elevated UpV in the FHH and FHH.Pkhd1 results from the FHH genetic background. The FHH strain is a genetic model of renal failure, and these rats develop UpV and focal segmental glomerular sclerosis at a young age and eventually die of end-stage renal failure (7, 27, 29). Thus, the cause of high UpV in the FHH and FHH.Pkhd1 is independent of the Pkdh1 mutation. Although the overall level of UpV is similar between PCK and FHH.Pkhd1, renal function is clearly improved in the FHH.Pkhd1 as indicated by BUN and plasma creatinine, and these measurements closely correlate with renal cyst formation.

While renal cyst formation was greatly diminished in the FHH.Pkhd1 animals, the FHH genetic background did not confer the same degree of protection against biliary manifestations and hepatic fibrosis. FHH.Pkdh1 livers tended to be smaller than PCK livers; however, this difference was not significant at 27 wk of age. The difference in protective effect of renal compared with biliary cystogenesis demonstrates that genetic elements on the FHH background specifically modulate kidney specific pathways. Therefore, the genes responsible for attenuated renal cyst formation in the FHH are likely kidney specific and not expressed in the liver.

Hypertension is a common feature of ARPKD in humans (13, 17), but in the case of the FHH.Pkhd1, there was no increase in blood pressure over time. This could be because elevated blood pressure is secondary to kidney disease, and FHH.Pkhd1 had milder renal impairment than PCK. Alternatively, the absence of blood pressure elevations in the FHH.Pkhd1 could be attributed to a lack of renin angiotensin system (RAS) activation, which is thought to mediate the hypertension observed in PKD in humans (13, 31) as well as in the PCK rat (14, 23). ACE and renin were both upregulated in PCK kidneys [1.47 (P = 0.003)- and 1.61 (P < 0.001)-fold, respectively] compared with SD, but there was no major increase in ACE and a modest increase in renin gene expression in FHH.Pkhd1 compared with FHH [−1.03 (not significant)- and 1.15 (P < 0.001)-fold, respectively]. These expression results support the hypothesis that activation of the RAS may be responsible for, or at least contributing to, hypertension in the PCK rat.

Identifying molecules and pathways that modulate the progression of PKD can provide important targets for therapeutic intervention and may provide clues to disease severity. A number of studies have reported comprehensive transcriptional changes in PKD patients as well as murine models of PKD, which have led to the identification of numerous genes misregulated in cystic vs. healthy kidneys (11, 28, 36, 44, 46). Although these studies provide insight into pathways that are modulated in PCK, it is unclear whether differentially expressed genes are directly involved in cyst formation as a result of the Pkhd1 mutation or if expression differences are secondary to disease progression. By investigating mRNA expression in the PCK vs. SD and FHH.Pkhd1 vs. FHH, we were able to segregate genes and pathways that may be differentially expressed due to progression of cyst formation (PCK vs. SD) from those genes up- or downregulated in response to the Pkhd1 mutation itself (genes common in PCK vs. SD and FHH.Pkhd1 vs. FHH), independent of the state of disease progression.

Not surprisingly, we found a number of genes previously described to be differentially expressed in murine and human cystic kidneys to also be differentially expressed in PCK kidneys but not FHH.Pkhd1 kidneys. The altered expression of these genes correlates with the severity of renal cystogenesis, suggesting that regulation of many genes associated with PKD is due to secondary phenotypic differences (i.e., state of disease progression) and are not directly modulated by the Pkhd1 mutation. Inflammatory and immune response as well as epithelial growth and cell proliferation were major biological processes activated in PCK kidneys, which have been previously described in cystic kidneys (11, 28, 44, 46).

The transcriptional analysis of both PCK and FHH.Pkhd1 revealed genes that may be directly regulated by the Pkhd1 mutation, as they were differentially expressed to a similar degree in both FHH.Pkhd1 and PCK. Pcdh9 (Protocadherin 9) was similarly downregulated in both FHH.Pkdh1 (−2.8-fold) and PCK (−3.5-fold) compared with control strains, suggesting that the expression of this gene is directly associated with the Pkhd1 mutation itself and may be involved in pathways downstream of fibrocystin. Little is known about Pcdh9, but it has been demonstrated that knock-down of protocadherin family members Fat4 and Fat1 result in renal cyst formation in mice and zebrafish, respectively (43, 45). These studies demonstrate that certain protocadherins are necessary for normal renal epithelial arrangement. The similar downregulation of Pcdh9 in PCK and FHH.Pkhd1 kidneys along with evidence that protocadherin knockdown causes renal cyst formation suggests that Pcdh9 expression may be directly influenced by the fibrocystin mutation. Molecules up- or downregulated in the same direction in both PCK and FHH.Pkhd1, but to a very different degree (i.e., Cps1, Ugt2B15, and Slco1a1) could be misregulated due to secondary phenotypic effects and not directly influenced by the Pkhd1 mutation, since the degree of regulation appears to be independent of the presence of the Pkhd1 mutation. However, further studies are required to address this hypothesis. The microarray analysis carried out in the present study is a powerful comparison because we identified molecules and pathways that are associated with the cystic phenotype of ARPKD as well as molecules that may be directly modulated by the Pkhd1 mutation.

In summary, the newly developed FHH.Pkhd1 rat strain has attenuated susceptibility to renal cystogenesis and insufficiency compared with PCK rats even though both strains carry the same Pkhd1 mutant allele. In future studies, the FHH.Pkhd1 and PCK strains can be used to identify genetic modifiers of ARPKD aggression by consomic mapping and F2 linkage analysis. Molecules and pathways found to influence ARPKD susceptibility can provide therapeutic targets for treating ARPKD, and can aid in treating and predicting outcomes of this devastating disease.

GRANTS

This study was performed with financial support from National Institute of Diabetes and Digestive and Kidney Diseases Grant NIDDK-1-P50 DK-079306 to E. D. Avner.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: C.C.O., W.E.S., H.J.J., E.D.A., and C.M. conception and design of research; C.C.O., M.H., W.E.S., S.-W.T., B.X., and C.M. analyzed data; C.C.O., E.D.A., and C.M. interpreted results of experiments; C.C.O. prepared figures; C.C.O. drafted manuscript; C.C.O., E.D.A., and C.M. edited and revised manuscript; C.C.O. and C.M. approved final version of manuscript; M.H. performed experiments.