Crosstalk with renal proximal tubule cells drives acidosis-induced inflammatory response and dedifferentiation of fibroblasts via p38-singaling.

BACKGROUND: Tubulointerstitial kidney disease associated microenvironmental dysregulation, like acidification, inflammation and fibrosis, affects tubule cells and fibroblasts. Micromilieu homeostasis influences intracellular signaling and intercellular crosstalk. Cell-cell communication in turn modulates the interstitial microenvironment. We assessed the impact of acidosis on inflammatory and fibrotic responses in proximal tubule cells and fibroblasts as a function of cellular crosstalk. Furthermore, cellular signaling pathways involved were identified. METHODS: HK-2 (human proximal tubule) and CCD-1092Sk (human fibroblasts), in mono and coculture, were exposed to acidic or control media for 3 or 48 h. Protein expression of inflammation markers (TNF, TGF-ß and COX-2), dedifferentiation markers (N-cadherin, vinculin, ß-catenin and vimentin), fibrosis markers (collagen III and fibronectin) and phospho- as well as total MAPK levels were determined by western blot. Secreted collagen III and fibronectin were measured by ELISA. The impact of MAPK activation was assessed by pharmacological intervention. In addition, necrosis, apoptosis and epithelial permeability were determined. RESULTS: Independent of culture conditions, acidosis caused a decrease of COX-2, vimentin and fibronectin expression in proximal tubule cells. Only in monoculture, ß-Catenin expression decreased and collagen III expression increased in tubule cells during acidosis. By contrast, in coculture collagen III protein expression of tubule cells was reduced. In fibroblasts acidosis led to an increase of TNF, COX-2, vimentin, vinculin, N-cadherin protein expression and a decrease of TGF-ß expression exclusively in coculture. In monoculture, expression of COX-2 and fibronectin was reduced. Collagen III expression of fibroblasts was reduced by acidosis independent of culture conditions. In coculture, acidosis enhanced phosphorylation of ERK1/2, JNK1/2 and p38 transiently in proximal tubule cells. In fibroblasts, acidosis enhanced phosphorylation of p38 in a sustained and very strong manner. ERK1/2 and JNK1/2 were not affected in fibroblasts. Inhibition of JNK1/2 and p38 under coculture conditions reduced acidosis-induced changes in fibroblasts significantly. CONCLUSIONS: Our data show that the crosstalk between proximal tubule cells and fibroblasts is crucial for acidosis-induced dedifferentiation of fibroblasts into an inflammatory phenotype. This dedifferentiation is at least in part mediated by p38 and JNK1/2. Thus, cell-cell communication is essential for the pathophysiological impact of tubulointerstitial acidosis.

Hypothesis-generating analysis of the impact of non-damaging metabolic acidosis on the transcriptome of different cell types: Integrated stress response (ISR) modulation as general transcriptomic reaction to non-respiratory acidic stress?

Extracellular pH is an important parameter influencing cell function and fate. Microenvironmental acidosis accompanies different pathological situations, including inflammation, hypoxia and ischemia. Research focussed mainly on acidification of the tumour micromilieu and the possible consequences on proliferation, migration and drug resistance. Much less is known regarding the impact of microenvironmental acidosis on the transcriptome of non-tumour cells, which are exposed to local acidosis during inflammation, hypoxia, ischemia or metabolic derailment. In the present hypothesis-generating study, we investigated the transcriptional impact of extracellular acidosis on five non-tumour cell types of human and rat origin, combining RNA-Sequencing and extensive bioinformatics analyses. For this purpose, cell type-dependent acidosis resiliences and acidosis-induced transcriptional changes within these resilience ranges were determined, using 56 biological samples. The RNA-Sequencing results were used for dual differential-expression analysis (DESeq and edgeR) and, after appropriate homology mapping, Gene Ontology enrichment analysis (g:Profiler), Ingenuity Pathway Analysis (IPA®), as well as functional enrichment analysis for predicted upstream regulators, were performed. Extracellular acidosis led to substantial, yet different, quantitative transcriptional alterations in all five cell types. Our results identify the regulator of the transcriptional activity NCOA5 as the only general acidosis-responsive gene. Although we observed a species- and cell type-dominated response regarding gene expression regulation, Gene Ontology enrichment analysis and upstream regulator analysis predicted a general acidosis response pattern. Indeed, they suggested the regulation of four general acidosis-responsive cellular networks, which comprised the integrated stress response (ISR), TGF-ß signalling, NFE2L2 and TP53. Future studies will have to extend the results of our bioinformatics analyses to cell biological and cell physiological validation experiments, in order to test the refined working hypothesis here.

Acidosis Activates the Nrf2 Pathway in Renal Proximal Tubule-Derived Cells through a Crosstalk with Renal Fibroblasts.

Crosstalk of renal epithelial cells with interstitial fibroblasts plays an important role in kidney pathophysiology. A previous study showed that crosstalk between renal epithelial cells and renal fibroblasts protects against acidosis-induced damage. In order to gain further mechanistic insight into this crosstalk, we investigated the effect of acidosis on the transcriptome of renal epithelial cells (NRK-52E) and renal fibroblasts (NRK-49F) in co-culture by RNASeq, bioinformatics analysis and experimental validation. Cells were exposed to acidic media or control media for 48 h. RNA and protein from whole cell lysate were isolated. In addition, cells were fractionated into cytosol, nucleus and chromatin. RNASeq data were analyzed for differential expression and pathway enrichment (ingenuity pathway analysis, IPA, QIAGEN). Total and phosphorylated protein expression was assessed by Western blot (WB). Transcription factor activity was assessed by luciferase reporter assay. Bioinformatic analysis using differentially expressed genes according to RNASeq (7834 for NRK-52E and 3197 for NRK-49F) predicted the antioxidant and cell-protective Nrf2 pathway as acidosis-induced in NRK-52E and NRK-49F cells. Activation of Nrf2 comprises enhanced Nrf2 phosphorylation, nuclear translocation, DNA binding and initiation of a cell protective transcriptional program. Our data show that acidosis enhances chromatin-associated Nrf2 expression and the abundance of phosphorylated Nrf2 in the chromatin fraction of NRK-52E cells in co-culture but not in monoculture. Furthermore, acidosis enhances the activity of a reporter for Nrf2 (ARE-luciferase). Despite the bioinformatics prediction, NRK-49F cells did not respond with Nrf2 activation. Transketolase (TKT) is an important regulator of antioxidant and homeostatic responses in the kidney and a canonical Nrf2 target gene. We show that protein and mRNA expression of TKT is increased in NRK-52E cells under co-culture but not under monoculture conditions. In conclusion, our data show that extracellular acidosis activates the cytoprotective transcription factor Nrf2 in renal epithelial cells co-cultivated with renal fibroblasts, thereby enhancing the expression of cytoprotective TKT. This protective response is not observed in monoculture. Activation of the Nrf2 pathway represents a co-operative cellular strategy of protection against acidosis.

Epithelial-Fibroblast Crosstalk Protects against Acidosis-Induced Inflammatory and Fibrotic Alterations.

Pathogenesis of chronic kidney disease (CKD) is accompanied by extracellular acidosis inflammation, fibrosis and epithelial-to-mesenchymal transition (EMT). The aim of this study was to assess the influence of acidosis on tubule epithelial cells (NRK-52E) and fibroblasts (NRK-49F) in dependence of cellular crosstalk. NRK-52E and NRK-49F were used in mono- and co-cultures, and were treated with acidic media (pH 6.0) for 48 h. The intracellular proteins were measured by Western blot. Secreted proteins were measured by ELISA. Distribution of E-cadherin was assessed by immunofluorescence and epithelial barrier function by FITC-dextran diffusion. Inflammation: Acidosis led to an increase in COX-2 in NRK-52E and TNF in NRK-49F in monoculture. In co-culture, this effect was reversed. EMT: Acidosis led to an increase in vimentin protein in both cell lines, whereas in co-culture, the effect was abolished. In NRK-52E, the E-cadherin expression was unchanged, but subcellular E-cadherin showed a disturbed distribution, and cellular barrier function was decreased. Fibrosis: Monoculture acidosis led to an increased secretion of collagen I and fibronectin in NRK-52E and collagen I in NRK-49F. In co-culture, the total collagen I secretion was unchanged, and fibronectin secretion was decreased. Intercellular crosstalk between epithelial cells and fibroblasts has a protective function regarding the development of acidosis-induced damage.

Epithelial-fibroblast cross talk aggravates the impact of the nephrotoxin ochratoxin A.

BACKGROUND: Chronic nephropathies result from different pathogenic agents, including nutritional factors triggering vicious pathophysiological cycles. Ochratoxin A (OTA) is a globally occurring nephrotoxic mycotoxin detectable in a variety of foodstuff and suspected to cause tubulointerstitial damage. The underlying mechanisms are not sufficiently understood, compromising risk assessment. Because crosstalk of proximal tubule cells with fibroblasts is crucial for tubulointerstitial damage, we investigated the effects of OTA in co-culture of these two cell types. METHODS: Rat renal proximal tubule cells (NRK-52E) and renal fibroblasts (NRK-49F) were exposed to nanomolar OTA concentrations under mono- and/or co-culture conditions for up to 48 h. We determined the impact on inflammation-, EMT- and fibrosis-associated proteins as well as microRNAs by western blot or qPCR, respectively. Alterations in cell morphology were quantitatively assessed. The roles of miRs, COX-2 and ERK1/2 in OTA-induced effects were investigated by specific inhibition. FINDINGS: Only under co-culture condition, OTA caused an increase of vimentin, fibronectin and miR-21 and a decrease of collagen III, E-cadherin, COX-2 and WISP1 mRNA abundance in NRK-52E cells. In NRK-49F cells, OTA induced an increase of N-cadherin, COX-2, WISP1 in co-culture only. The OTA-induced increase of fibronectin in NRK-52E cells was prevented by simultaneous inhibition of miR-21 and -200a, COX-2 or ERK1/2. The OTA-induced increase of COX-2 in NRK-49F cells was prevented by inhibition of miR-21 and -200a or ERK1/2. INTERPRETATION: Our results show that the complete nephropathic potential of nanomolar OTA, leading to EMT, is unveiled when cellular crosstalk is possible. In monoculture, the nephropathic potential is underestimated. RESEARCH IN CONTEXT: Chronic nephropathies are a severe health burden and the result of different pathogenic mechanisms, including nutritional factors that trigger vicious pathophysiological cycles. Ochratoxin A (OTA) is a ubiquitous, globally occurring nephrotoxic mycotoxin detectable in a variety of foodstuff and suspected to cause tubulointerstitial damage. Because underlying pathomechanisms are unclear, risk assessment is problematic. Crosstalk of proximal tubule cells (the main target of OTA) with fibroblasts is crucial for the development of tubulointerstitial damage. We show that during co-culture of proximal tubule cells and fibroblasts, OTA-induced effects (e.g. epithelial-mesenchymal transition (EMT)) change significantly as compared to monoculture. Our results show that the complete nephropathic potential of OTA is unveiled when cellular crosstalk is possible. In monoculture, the nephropathic potential of OTA is underestimated.

Synergistic action of the nephrotoxic mycotoxins ochratoxin A and citrinin at nanomolar concentrations in human proximal tubule-derived cells.

Increased ochratoxin A (OTA) or citrinin (CIT) concentrations in food correlate with increased prevalence of tubule-interstitial nephropathy. We tested the hypothesis that co-exposure of human proximal tubule-derived epithelial cells (HK-2) to OTA and CIT promotes synergistic events indicative for inflammation, epithelial-to-mesenchymal-transition (EMT) or fibrosis. We measured markers of inflammation, EMT and fibrosis and investigated the role of MAP-kinases. Only concurrent but not individual exposure to OTA and CIT at nanomolar concentrations led to (i) an increase of TNF protein and mRNA, (ii) a decrease of COX-2 protein and mRNA, (iii) a decrease of E-cadherin protein and (iv) an increase of vimentin and α-SMA protein. Cell shape shifted from a cobblestone- to a spindle-like phenotype indicating EMT. Extra- and intracellular collagen III protein content was increased. Concomitant mRNA expression changes were observed for TNF, COX-2, E-cadherin and α-SMA indicating transcriptional regulation. This was not the case for vimentin and collagen III mRNA indicating posttranscriptional regulation. Inhibition of ERK 1/2 and JNK 1/2 reduced the effect on TNF but not on α-SMA mRNA indicating an involvement of these kinases. Phosphorylation of ERK1/2 was increased by CIT, OTA alone and the mycotoxin combination. In contrast, the phosphorylation of JNK1/2 was unchanged. In conclusion, nanomolar OTA and CIT act synergistically favouring nephropathic processes.

Structure-activity relationship of ochratoxin A and synthesized derivatives: importance of amino acid and halogen moiety for cytotoxicity.

The enigma why the mycotoxin ochratoxin A (OTA) impairs cell and organ function is still not solved. However, an interaction with target molecules is a prerequisite for any observed adverse effect. This interaction depends on characteristics of the target molecule as well as on the OTA molecule itself. OTA has different structural moieties which may be relevant for these interrelations including a halogen (chlorine) and an amino acid group (phenylalanine). To test their importance for the impact of OTA, detailed structure-activity studies with various OTA derivatives were performed. For this, 23 OTA derivatives were available, which were modified by either an exchange of the halogen moiety against another halogen (fluorine, iodine or bromine) or by the amino acid moiety against another one (tyrosine or alanine) or a combination of both. Additionally, the configuration of the 3R carbon atom was changed to 3S. These derivatives were tested in human renal cells for their ability to induce cell death (cytotoxicity, apoptosis, necrosis), their impact on collagen protein secretion and for their influence on gene expression. It turned out that the substitution of the amino acid moiety against tyrosine or alanine almost completely prevented the adverse effects of OTA. The exchange of the halogen moiety had minor effects and the inversion of the stereochemistry at C3 did not prevent the effects of OTA. Therefore, we conclude that the amino acid moiety of OTA is indispensable for the interaction of OTA with its target molecules.