Cystogenesis In ARPKD Results From Increased Apoptosis In Collecting Duct Epithelial Cells Of Pkhd1 Mutant Kidneys

BackgroundAutosomal recessive polycystic kidney disease (ARPKD) is one of the most common hereditary renal cystic diseases in infants and children, with an estimated incidence of ¹ in 20,000 live births and prevalence ¹ in 70 for heterozygosity. ARPKD is characterized by cystic dilatation of collecting ducts of the kidney and hepatic abnormalities consisting of bile duct dysgenesis and periportal fibrosis. Approximately 50% of patients with ARPKD present with their disease as neonates and are born with two very large kidneys with 60 to 90% of the renal tubules being ectatic. These neonates suffer a 30% mortality rate as a result of respiratory and/or renal dysfunction. ARPKD is caused by mutations in Pkhd1, which encodes a 16-kb transcript, contains at least 86 exons, and spans 470 kb on chromosome 6p12. The longest ORF is predicted to have 66 exons and to yield a 4074-amino acid membrane-associated receptor-like protein, fibrocystin/polyductin (FPC). FPC has been localized to primary cilia and basal body.The cystogenesis of kidney probably results in the molecular dysfunction of growth and development of renal collecting ducts in human. Illustrating the molecular mechanism of the cystogenesis will greatly contribute in the molecular mechanism of the growth and development of renal collecting ducts. To determine the molecular mechanism of the cystogenesis in ARPKD, we recently generated a mouse model for ARPKD that carries a targeted exon 15 and 16 deletion in the mouse orthologue of human Pkhd1. The homozygous mutant mice display hepatorenal cysts whose phenotypes are similar to those of human ARPKD patients. Pkhd1?/? mice that escaped embryonic lethality and survived into adulthood exhibited mild to severe tubular dilation or cyst formation in the kidney and liver accompanied by fibrosis and necrosis. Cystic or dilated-duct phenotypes were also seen in the pancreas and brain of Pkhd1?/? mice. To illustrate the role and associated mechanism of Pkhd1 loss in the cystogeneiss and development in ARPKD, we observe the effect of FPC loss on the proliferation and apoptosis of renal collecting duct epithelial cells and explore the related mechanisam using our Pkhd1 knockout mouse model and renal collecting duct cell lines derived from this model. We provided new concept for the pathogenic mechanism of ARPKD cystogenesis which will lead to new therapeutic strategies for human ARPKD.Part 1FPC loss impairs tubulomorphogenesis of collecting duct epithelial cells in Pkhd1 mutant kidneysObjective:Using our Pkhd1 knockout mouse model and renal collecting duct cell lines derived from this model, we observed the biological effect of FPC loss on the renal collecting duct epithilia cells.Methods:1. We mated the Pkhd1+/- mice with Immorto mice (Im) (both with C57Bl/6 congenic background to produce Im:: Pkhd1^-/- mice and their Im::WT littermates. To establish cell lines with or without Pkhd1, kidneys from an 8-week-old Im:Pkhd1^-/- mouse and its wildtype littermate were removed and minced finely with a scalpel. A Dolichus biflorus agglutinin (DBA)-based isolation approach was used to develop immortalized renal collecting duct cell lines from the kidneys. The collecting duct cell lines with and without Pkhd1 were selected from the Im:Pkhd1^-/- and its wildtype littermate cell pool. After using E-cadherin and cytokeratin as epithelial markers and DBA as the collecting duct marker to identify their origin, we detect the Pkhd1 expression in null-Pkhd1and their wildtype control at DNA, RNA and protein levels using PCR, RT-PCR, real time PCR and Western blot methods.2. We characterized the cell lines by performing experiments in 3-D Matrigel culture to examine whether the loss of FPC induced abnormal tubulomorphogenesis in vitro.3. We also analyzed the cell-cell contacts in these cell lines. E-cadherin and ZO-1 which are putative cell-cell junctional markers were detected using immunofluorescence staining and Western blot.4. To confirm the altered cell-cell interactions by loss of FPC, we measured the cell transepithelial resistance (TER) to determine if the integrity of the cell-cell contacts was impaired.5. We performed rhodamine-phalloidin staining to label cellular cytoplasmic actin to observe the locoliazation and shape changes of F-actin by loss of FPC.6. We investigated the effect of null-Pkhd1 on the integrin-dependent adhesion to CI. We also tested the transwell migration capability upon CI between the cells with and without Pkhd1.7. To determine whether the lack of FPC also disrupts ciliogenesis in Pkhd1-deficient mice. We used IF with an anti-acetylatedα-tubulin antibody to examine the number and morphology of renal primary cilia in Pkhd1^-/- cells M10H2 and M10C7 and their wildtype littermate cells W10B6 and W10B2.8. Performing a tritiated thymidine proliferation assay, to determine whether the loss of FPC caused a decrease in renal epithelial proliferation.9. Pkhd1^-/- cells M10H2 and M10C7 and their wildtype littermate cells W10B6 and W10B2 were subjected to Phospho-Histone H3 staining to evaluate cell proliferation.10. At last, we used DeadEndTMflurometric TUNEL system or Caspase-3 Active Apoptosis Kit to examine whether apoptosis also occurred in vitro and in vivo.Results:1. To establish null-Pkhd1 cell lines, the kidneys from an 8-week-old Im::Pkhd1^-/- mouse and its Im::WT littermate were removed, and a Dolichus biflorus agglutinin (DBA)-based isolation approach was used to develop immortalized renal collecting duct cell lines of both genotypes. After a limiting dilution, 38 immortalized renal collecting duct cell colonies were isolated from each of the Im::Pkhd1^-/- and Im::WT cell pools. To identify the origin of the cell lines, we used E-cadherin and cytokeratin as epithelial markers and DBA as a collecting duct marker. By these biomarkers, 26 collecting duct cell lines with the Im::Pkhd1^-/- genotype were selected from the Im::Pkhd1^-/- cell pool, and 21 Im::WT collecting duct cell lines were selected from the Im::WT cell pool. Of these, two randomly selected lines from each pool (W10B6 and W10B2 for wildtype and M10H2 and M10C7 for Pkhd1^-/-) were used as the genotype-representative cell lines for further analysis. PCR genotyping was performed to identify cell lines with Pkhd1^-/- cells (M10H2 and M10C7) and wildtype cells (W10B6 and W10B2). A 380bp PCR band was observed for the Pkhd1^-/- allele whereas a 270bp PCRband was observed for the wild type. Quantitative PCR verified that the wildtype cell lines expressed Pkhd1 and the Pkhd1^-/- cell lines did not. To further confirm the genotypes of these cell lines, an anti-FPC monoclonal antibody hAR-C2m4E12, which is a subclone from hAR-C2m3C10, was used to detect the FPC expression levels in the cell lines by western blot. Consistent with the quantitative PCR results, the Pkhd1^-/- cell lines did not express any detectable FPC, while the wildtype cell lines expressed it.2. Most of the wildtype cells formed normal tubular structures in the 3-D culture, and only 5-10% of them failed to exhibit tubulogenesis. In sharp contrast, ⁹5% of the 3-D cultured Pkhd1^-/- cells failed to undergo tubulogenesis (Suppl. Fig. 1A-D). In addition, less than 5% of the colonies in the Pkhd1^-/- cell cultures had 3 or more branches, whereas ⁶0% of the wildtype tubular structures did.3. Although ZO-1 staining was observed on the cell-cell junctions for wildtype W10B6 and W10B2 cells, a more diffuse and discontinuous pattern of junctional staining was seen in Pkhd1^-/- cells (M10H2 and M10C7). In the wildtype cell lines, E-cadherin was predominantly seen at the cell-cell junctions; while in the Pkhd1^-/- cells junctional staining of E-cadherin was nearly indistinguishable from cytosolic. Given the IF staining showed different distribution patterns for E-cadherin and ZO-1 in wildtype and Pkhd1^-/- cells, we performed western blot analysis to determine whether there were variations in E-cadherin or ZO-1 expression levels. We found there was no detectable immunoreactive change in either protein among the cell lines. 4. The TER was significantly lower in the null-Pkhd1 cells than in the wildtype ones after 3 days of transwell culture (*P<0.05), and this difference persisted to day 7.5. Under 100×microscope, the Pkhd1^-/- and wildtype cells were stained with a rhodamine-phalloidin (F-actin) antibody. We found that wildtype cells exhibit nominal cortical actin distribution and epithelial shape with fine and even stress fibers in sub-confluent cultures. Pkhd1^-/- cells cells lose their normal cortical actin distribution and exhibit an irregular shape with thick and enriched stress fibers. The cell-cell junction gaps and irregular cell-cell borders were seen in Pkhd1^-/- cells. Confocal microscope images showed that F-actin in cultured wildtype cells (W10B6 and W10B2) localized around the nucleas and cells. Fine, stress fiber distributed at cell-cell junctions. The cell-cell junction gaps and irregular cell-cell borders were seen in Pkhd1^-/- cells. Confocal microscope images of the Pkhd1^-/- cells (M10H2 and M10C7) and their wildtype littermate cells (W10B6 and W10B2) cultured for 7 days in 3-D CI gels using rhodamine-phalloidin. Extensive tubulomorphogenesis was seen in the wildtype cells, but it was rare in the Pkhd1^-/- cells.6. The Pkhd1^-/- cells adhered less well than that of wildtype cells at concentrations of CI from 0.125 to 2μg/ml. At 2μg/ml CI, the null-Pkhd1 cells showed only 40% cell adhesion compared to over 90% for the control cell lines. Transwell migration assay results indicated the absolute number of cells that migrated to the underside of the transwell. There was a significant decrease in the number of migrated Pkhd1^-/- cells compared to wildtype cells (W10B6 and W10B2).7. Compared with the cultured WT cells, a shortened ciliary structure and decreased ciliary staining were seen in the Pkhd1^-/- cells. The cilia stained in approximately 80% of WT cells and in fewer than 30% of Pkhd1^-/- cells. The mean length of primary cilia was 2.5μm in cultured WT cells and was <1.5μm in Pkhd1?/? littermate cells.8. Pkhd1^-/- cells M10H2 and M10C7 and their wildtype littermate cells W10B6 and W10B2 were incubated with 3H-thymidine, then the rate of 3H-thymidine incorporation was determined. The 3H-thymidine values were significantly decreased in the Pkhd1^-/- cells than that in the wildtype cells.9. Pkhd1^-/- cells M10H2 and M10C7 and their wildtype littermate cells W10B6 and W10B2 were subjected to Phospho-Histone H3 staining to evaluate cell proliferation. The percentage of cells positive for Phospho-Histone H3 was significantly greater in the cells with Pkhd1 than in those without. Western analyses with Phospho-Histone H3 and PCNA antibodies showed that both these proliferation markers were significantly downregulated in the null-Pkhd1 cells (M10H2 and M10C7), compared with their wildtype littermate cells W10B6 and W10B2. Furthermore, we tested immunohistochemical (IHC) and immunofluorescent (IF) staining of Phospho-Histone H3 and PCNA in the kidney of 6-week-old, 3-month-old and 6-months old Pkhd1^-/- mice and its wildtype littermate. Many positive-stained cells were seen in the wildtype kidneys, but only a few appeared in the corresponding region of Pkhd1^-/- kidneys.10. We examined apoptosis rates for the Pkhd1^-/- and wildtype cell lines using TUNEL (Terminal deoxynucleotidyl Transferase Biotin-dUTP Nick End Labeling) assays. Under routine culture conditions, approximately 5% of the wildtype cells and around 14% of the null-Pkhd1 cells were apoptotic. To confirm this result, we assayed active caspase-3 as an indictor of apoptosis in the same cell lines. In agreement with the TUNEL assay results, the percentage of apoptotic Pkhd1^-/- cells was significantly higher than that of the wildtype cells. To examine apoptosis, the same kidneys were subjected to caspase-3 staining. More positive staining was seen in the Pkhd1^-/- than in the wildtype kidneys 3-month-old, 6-months old and 12-months old. Since older Pkhd1^-/- mice exhibit severer cystic kidney, the kidneys of 12-month-old Pkhd1^-/- and wildtype mice were also examined by TUNEL staining. Increased positive staining was seen in the Pkhd1^-/- than that in wildtype kidneys. Notably, much apoptotic debris was observed in the lumen of the Pkhd1^-/- renal tubules.Conclusion:1. Establishment of renal collecting duct cell lines bearing null-Pkhd1 alleles from Pkhd1^-/- mutant kidneys.2. FPC expression is required for normal tubulogenesis in 3-D culture of renal collecting duct cells.3. Loss of FPC impairs normal renal collecting duct cell-cell contacts and decreases integrin-dependent cell adhesion.4. Loss of FPC exhibits aberrant ciliogenesis and disturbs normal actin cytoskeleton distribution in collecting duct epithelial cells5. Loss of FPC reduces cell proliferation and promotes apoptosis in vitro and in vivo.Part 2The role of apoptosis singnaling Pathway in collecting duct epithelial cells of Pkhd1 mutant kidneysObjective:To explore the molecular mechanism of Pkhd1 deletion in cystogenesis in ARPKD through reducing proliferation and increasing apoptosis.Methods:1. Western blot was use to detect the expression of phosphorylated Akt (pS473) and test its effects on the apoptosis cell singnaling Pathway Bax-Caspase9-Caspase 3 in collecting duct epithelial cells of Pkhd1 mutant kidneys.2. We used western blot analyses to examine the Akt-downstream singnaling Ras-Raf-MEK-ERK.3. Four major FAK phosphorylation sites which regulate multiple cellular processes from cell migration and polarity to proliferation and apoptosis were examined by Western blot.4. Lysates from our null-Pkhd1 and wildtype cell lines were examined for changes in the phosphorylation of PI3K and PDK1.Results:1. Western blot analysis showed that phosphorylated Akt (pS473) was significantly reduced in the null-Pkhd1 cells compared to the wildtype-littermate cells after collagen I (CI) induction. Because Bax and Bcl2 are putative downstream factors of Akt, and their dysfunction can also induce abnormal apoptosis, we next compared their factors between the null-Pkhd1 and wildtype cells. We found that no significant difference in Bcl2 expression between the cells with and without Pkhd1, at all time points examined after CI induction (data not shown). In contrast, Bax expression was significantly higher in the null-Pkhd1 cells compared to the wildtype cells. The basal and collagen I (CI)-inducible levels of Bax were both much lower in the null-Pkhd1 cells than that in the wildtype cells. Since both caspase-9 and caspase-3 are putative downstream factors of Bax, we next examined their expression levels in the same cell lines. Western analyses showed that the caspase-9 and -3 were significantly elevated in the null-Pkhd1 cells compared to the wildtype cells. 2. Western blot results indicated that there was no significant difference in the B-Raf phosphorylation in the M10H2 and M10C7 (Pkhd1^-/-) cell lines versus the wildtype-littermate W10B6 and W10B2 cell lines. We therefore shifted our focus to c-Raf, which is another important regulator of cell proliferation. Compared to the wildtype cell lines, western results showed that the null-Pkhd1 cells exhibited a significant downregulation of phosphorylated c-Raf, in the M10H2 and M10C7 (Pkhd1^-/-) cell lines versus the wildtype-littermate W10B6 and W10B2 cell lines.we examined the Ras/c-Raf downstream factors MEK and ERK, both of which are key markers for cell proliferation, using lysates from the Pkhd1^-/- and wildtype cell lines. The results showed that the phosphorylated MEK and ERK were significantly decreased in the null-Pkhd1 cells at all time points after CI induction. Given that Raf is the first effector that is positioned downstream of Ras, we next investigated if the aberrant c-Raf phosphorylation was caused by Ras dysregulation. Western analyses of lysates from the Pkhd1^-/- and wildtype cell lines showed that the level of Ras was significantly downregulated in the basal condition and after 30 minutes of CI induction in the null-Pkhd1 cells.3. Besides FAKpY861, which we reported previously, the phosphorylations of FAKpY397, 576, and 925 were also significantly decreased in the M10H2 and M10C7 (Pkhd1^-/-) cells compared to the wildtype-littermate W10B6 and W10B2 lines after CI induction.4. We analyzed the PI3K phosphorylation using anti-PI3K class III and PI3Kp110αantibodies. We observed no difference in PI3K class III phosphorylation between the cells with and without Pkhd1. In contrast, the basal and induced levels of PI3Kp110αwere much lower in the null-Pkhd1 M10H2 and M10C7 cells than that in wildtype-littermate W10B6 and W10B2 cells. Notably, in the null-Pkhd1 cells, a low level of PI3Kp110αwas observed in the basal and CI-induced condition. Since PDK1 is a downstream factor of PI3K that is activated by PI3K's phosphorylation on p110, we analyzed the phosphorylation of PDK1 between the null-Pkhd1 and wildtype cells. Similar to PI3Kp110, both the basal and induced pPDK1 levels were significantly lower in the null-Pkhd1 cells than that in wildtype cells.Conclusion:1. FAK phosphorylation promotes Bax and provoke apoptosis in null-Pkhd1 cells via the PI3Kp110α/PDK1 inactivation.2. Downregulation of the Ras/c-Raf/MEK/ERK cascade may be the molecular mechanism underlying the decreased renal epithelial proliferation in collecting duct epithelial cells of Pkhd1 mutant kidneys.