The cat as a naturally occurring model of renal interstitial fibrosis: Characterisation of primary feline proximal tubular epithelial cells and comparative pro-fibrotic effects of TGF-β1

Abstract

Chronic kidney disease (CKD) is common in both geriatric cats and aging humans, and is pathologically characterised by chronic tubulointerstitial inflammation and fibrosis in both species. Cats with CKD may represent a spontaneously occurring, non-rodent animal model of human disease, however little is known of feline renal cell biology. In other species, TGF-β1 signalling in the proximal tubular epithelium is thought to play a key role in the initiation and progression of renal fibrosis. In this study, we first aimed to isolate and characterise feline proximal tubular epithelial cells (FPTEC), comparing them to human primary renal epithelial cells (HREC) and the human proximal tubular cell line HK-2. Secondly, we aimed to examine and compare the effect of human recombinant TGF-β1 on cell proliferation, pro-apoptotic signalling and genes associated with epithelial-to-mesenchymal transition (EMT) in feline and human renal epithelial cells. FPTEC were successfully isolated from cadaverous feline renal tissue, and demonstrated a marker protein expression profile identical to that of HREC and HK-2. Exposure to TGF-β1 (0-10 ng/ml) induced a concentration-dependent loss of epithelial morphology and alterations in gene expression consistent with the occurrence of partial EMT in all cell types. This was associated with transcription of downstream pro-fibrotic mediators, growth arrest in FPTEC and HREC (but not HK-2), and increased apoptotic signalling at high concentrations of TGF- β1. These effects were inhibited by the ALK5 (TGF-β1RI) antagonist SB431542 (5 μM), suggesting they are mediated via the ALK5/TGF-β1RII receptor complex. Taken together, these results suggest that TGF-β1 may be involved in epithelial cell dedifferentiation, growth arrest and apoptosis in feline CKD as in human disease, and that cats may be a useful, naturally occurring model of human CKD.

Fig 1. Morphology of FPTEC, HREC and HK-2.

Representative photomicrographs of FPTEC, HREC and HK-2 cells. (A) FPTEC 24hrs post isolation. Small tissue clumps and individual cells are adhered to the tissue culture plastic. (B) FPTEC 96 hours post isolation. Proliferating colonies of epithelial cells form from the isolated tissue. (C) FPTEC 7 days post isolation. Colonies mature into densely packed epithelial monolayers. (D) FPTEC, passage 1. Cells maintain their epithelial morphology and demonstrate formation of “domes” (thick arrow) at confluency. (E) FPTEC, passage 1: High power image illustrating cuboidal morphology and close cell-cell adhesion. (F) FPTEC, passage 3. Larger, irregular cells begin to appear, indicating cultures are developing replicative senescence. (G) FPTEC, passage 4. The few remaining adherent cells are greatly enlarged and irregular. (H) HK-2 cells. The HK-2 cell line formed monolayers of cuboidal, epithelial cells and did not form domes. (I) HREC, passage 2. These cultures were initially heterogeneous, grew into a cuboidal monolayer. Images are representative of three experiments.

Fig 2. Immunofluorescence studies of FPTEC, HREC and HK-2.

Immunofluorescence staining of FPTEC, HREC and HK-2. Cell nuclei were stained with DAPI (blue). FPTEC were (A) >85% positive for cytokeratin AE1/AE3 expression (green), (B) 100% positive for vimentin expression (green), (C) α-klotho expression (red), and negative for (D) desmin and (E) vWF. HREC were 100% positive for (F) cytokeratin (green), (G) vimentin (green), (H) α-klotho (red) and negative for (I) desmin. Similarly, HK-2 were 100% positive for (J) cytokeratin (green), (K) vimentin (green), (L) α-klotho (red) and negative for (M) desmin and (N) vWF. (O) Isotype controls were negative. Images are representative of three experiments.

Fig 3. Brush border enzyme expression in FPTEC and HK-2.

Photomicrographs of cells stained for the demonstration of ALP (blue/mauve) and GGT (red/orange) activities. FPTEC stained positive for both ALP and GGT activity. Expression was heterogenous in both cases and ranged from intense to barely present. HK-2 demonstrated lower ALP activity, with staining generally less intense than the FPTEC, with some heterogenicity, but stained intensely and homogeneously positive for GGT expression. Images are representative of three experiments.

Fig 4. Effect of TGF-β1 on phosphorylation of Smad2 in FPTEC, HREC and HK-2.

The effect of a 30 minute treatment with 10ng/ml TGF-β1 on the expression of phosphorylated Smad2 (p-Smad2) in FPTEC, HREC and HK-2 was assessed by western blotting. Densitometric analysis of p-Smad2 was undertaken in ImageJ, normalised to β-actin and is expressed as fold change in relation to control. Incubation with TGF-β1 resulted in a significant increase in p-Smad2 expression in FPTEC (P = 0.038), HREC (P = 0.0393) and HK-2 (P < 0.0001). Data were analysed using the Student’s t-test. The columns represent the mean normalised density measurement and error bars represent the standard deviation. Images are representative of three experiments. *P<0.05, ****P<0.0001.

Fig 5. Effect of TGF-β1 on FPTEC, HREC and HK-2 cell morphology.

Representative photomicrographs of FPTEC, HREC and HK-2 after incubation with vehicle and 0.1–10 ng/ml TGF-β1 for 72 h. There was a concentration-dependent loss of the morphology evident in vehicle-treated cells across all three cell types, with cells adopting a hypertrophic, elongated fusiform appearance. These alterations were inhibited by the TGF-β1R1/ALK5 antagonist SB431542 (5 μM). Images are representative of three experiments.

Fig 6. TGF-β1 mediated expression of genes related to EMT in FPTEC.

Expression levels of commonly accepted EMT marker genes were assessed in FPTEC by RT-qPCR after incubation with TGF-β1 (0.1 to 10 ng/ml). Target gene mRNA copy number was normalised to GAPDH/RPS7 and is expressed as fold change in relation to control. Data represent four experimental repeats using cells from different cats and were analysed using the one-way ANOVA with Dunnett’s post-hoc analysis. The columns represent the mean normalised mRNA copy number and error bars represent the standard deviation.*P<0.05 **P<0.01 ***P<0.001 ****P<0.0001.

Fig 7. TGF-β1-mediated expression of downstream mediator mRNA in FPTEC.

Expression of paracrine and autocrine mediator mRNA after incubation with TGF-β1 was assessed in FPTEC by RT-qPCR. Target gene mRNA copy number was normalised to GAPDH/RPS7 and is expressed as fold change in relation to control. Data represent four experimental repeats using cells from different cats and were analysed using the one-way ANOVA with Dunnett’s post-hoc analysis. The columns represent the mean normalised mRNA copy number and error bars represent the standard deviation. *P<0.05 **P<0.01 ***P<0.001 ****P<0.0001.

Fig 8. Comparative effect of TGF-β1 and the TGF-β1RI/ALK5 inhibitor SB431542 on target gene expression in FPTEC, HREC and HK-2.

Effects of TGF-β1 (+/- 5 μM SB431542) on target gene expression was assessed in FPTEC, HREC and HK-2 by RT-qPCR. Target gene mRNA copy number was normalised to GAPDH/RPS7 and is expressed as fold change in relation to control. Data represent three experimental repeats using cells from different cats/batches and were analysed using the one-way ANOVA with Dunnett’s post-hoc analysis. The columns represent the mean normalised mRNA copy number and error bars represent the standard deviation.*P<0.05 **P<0.01 ***P<0.001.

Fig 9. Effect of TGF-β1 on proliferation of FPTEC, HREC and HK-2.

FPTEC, HREC and HK-2 were seeded onto 96-well plates at sub-confluent density and proliferation assessed by cell counting in DAPI-stained cultures after incubation with the indicated concentrations of TGF-β1 for 72 h. Data from FPTEC represent four experimental repeats using cells from different cats; data from HREC and HK-2 cells represent three experimental repeats. Data were analysed using the one way ANOVA with Dunnett’s post-hoc analysis. The columns represent the mean relative cell number and error bars represent the standard deviation.*P<0.05 **P<0.01 ***P<0.001.

Fig 10. TGF-β1-induced changes in caspase 3/7 activity in FPTEC and HK-2.

Pro-apoptotic signalling was assessed in HK-2 and FPTEC via measurement of caspase 3/7 activity. Data from FPTEC represent three experimental repeats using cells from different cats; data from HK-2 cells represent three experimental repeats. Data were analysed using the one way ANOVA with Dunnett’s post-hoc analysis. The columns represent the mean normalised mRNA copy number and error bars represent the standard deviation.*P<0.05 **P<0.01 ***P<0.001.