Qin, Y.
Citation
Qin, Y. (2011, October 18). Cell adhesion signalling in acute renal failure. Retrieved from https://hdl.handle.net/1887/17953
Version: Corrected Publisher’s Version
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden
Downloaded from: https://hdl.handle.net/1887/17953
Note: To cite this publication please use the final published version (if applicable).
Epac-Rap signalling reduces cellular stress and ischemia-induced kidney failure
Geurt Stokman1, Yu Qin1, Hans-Gottfried Genieser2, Frank Schwede2, Emile de Heer3, Johannes L. Bos4, Ingeborg M. Bajema3, Bob van de Water1, Leo S. Price1
1Division of Toxicology, Leiden/Amsterdam Center for Drug Research, Leiden University, Leiden, The Netherlands; 2Biolog Life Science Institute, Bremen, Germany; 3Department of Pathology, Leiden
University Medical Center, Leiden, The Netherlands; 4Department of Physiological Chemistry, University Medical Center, Utrecht University, Utrecht, The Netherlands
Abstract
Renal ischemia/reperfusion injury is associated with the loss of tubular epithelial cell-cell and cell-matrix interactions which contribute to renal failure. The Epac-Rap signalling pathway is a potent regulator of cell-cell and cell-matrix adhesion. The cyclic AMP analogue 8-pCPT- 2’-O-Me-cAMP has been shown to selectively activate Epac, whereas the addition of an acetoxymethyl (AM) ester to 8-pCPT-2’-O-Me-cAMP enhanced in vitro cellular uptake.
Here we demonstrate that pharmacological activation of Epac-Rap signalling using acetoxymethyl-8-pCPT-2’-O-Me-cAMP preserves cell adhesions during hypoxia in vitro, maintaining the barrier function of the epithelial monolayer. Intrarenal administration in vivo of 8-pCPT-2’-O-Me-cAMP also reduced renal failure in a mouse model for ischemia/reperfusion injury. This was accompanied by decreased expression of the tubular cell stress marker clusterin-α, and lateral expression of β-catenin after ischemia indicative of sustained tubular barrier function. Our study emphasizes the undervalued importance of maintaining tubular epithelial cell adhesion in renal ischemia and demonstrates the potential of pharmacological modulation of cell adhesion as a new therapeutic strategy to reduce the extent of injury in kidney disease and transplantation.
J Am Soc Nephrol 2011; 22: 859-872
Introduction
Renal ischemia/reperfusion (I/R) injury is a frequently occurring form of acute renal failure in patients 1 and an important contributing factor to chronic allograft dysfunction following renal transplantation 2. Ischemia impairs the capacity of the tubular epithelium to maintain the integrity of its cytoskeleton and affects adherens and tight junction stability 3-5 and cell- matrix interactions 6. Loss of cell adhesion is one of the earlier responses of the tubular epithelium observed during I/R injury 7, preceding tubular epithelial cell (TEC) death and inflammatory cell influx. Loss of cell-cell or cell-matrix adhesion has been found to correlate with loss of cell function, pro-apoptotic signalling and cell death 8, 9.
The second messenger cyclic adenosine monophosphate (cAMP) controls a variety of intracellular signalling pathways including protein kinase A (PKA), ion channel activity and as more recent studies demonstrate, also Epac (Exchange Proteins directly Activated by cAMP) 10. Upon activation by cAMP, Epac functions as an exchange factor for the small GTPase Rap1, mediating replacement of GDP for GTP consequently leading to activation of Rap1. Among the processes known to be influenced by Rap1 are integrin-mediated adhesion to the extracellular matrix 11, 12 and the preservation of cell-cell contacts 13, 14. The cAMP analog 8-pCPT-2’-O-Me-cAMP, colloquially named ‘007’, selectively activates Epac with low affinity for PKA when used in the μM concentration range, and can therefore distinguish between Epac-mediated signalling events and PKA-mediated events 15. A recent modification to 8-pCPT-2’-O-Me-cAMP being the introduction of an acetoxymethyl-ester (8-pCPT-2’-O- Me-cAMP-AM or 007-AM) was found to greatly enhance the in vitro biological activity of the compound presumably by facilitating cellular uptake without affecting its intracellular efficacy 16.
Epac is expressed by endothelial and epithelial cells with the highest expression level reported to be in the kidney10. Both isoforms of Epac (Epac1 and Epac2) are expressed by all three segments of the proximal tubules and are localized in the brush border 17. In mice, Epac1 has been implicated in the regulation of the Na+/H+ exchanger 3 activity 18. The Epac- Rap pathway increases endothelial barrier function in vitro by promoting maturation of VE cadherin-mediated cell-cell adhesions 19 whereas inhibition of Rap1 activity prevents cadherin-based cell adhesion 13.
The development of Epac-selective cAMP analogs offers a novel strategy for preserving cell adhesion during ischemic injury. Here we demonstrate that 8-pCPT-2’-O-Me-cAMP- induced activation of the Epac-Rap signalling pathway protects against I/R injury in a mouse model. In an in vitro hypoxia model we show that 8-pCPT-2’-O-Me-cAMP-AM strengthens cell-cell and cell-matrix contacts and preserves the integrity and barrier function of the epithelial monolayer. These findings suggest that therapeutic strategies aimed at the preservation of TEC cell-cell or cell-matrix adhesion in general and specifically via activation of the Epac-Rap pathway during I/R injury have an important clinical potential.
Results
Epac-Rap signalling is functional in conditionally immortalized proximal tubular epithelial cells (IM-PTEC)
We studied Epac-Rap signalling in vitro using conditionally immortalized IM-PTEC. At restrictive culture conditions, SV40 expression decreases (Figure 1A), and cells loose their immortalized status, reflected by reduced proliferation and acquisition of a characteristic
‘cobblestone’ epithelial phenotype (data not shown). IM-PTEC cells form confluent monolayers and express the tight junction protein ZO-1 (Figure 1E). Transcripts for the proximal tubular epithelial markers megalin (gp330) and sodium-glucose exchange transporters SGLT1 and SGLT2 20 were detected (Figure 1B). These results show that IM- PTEC cultured under restrictive conditions are of proximal tubular origin and display the related characteristics and morphology in culture.
Transcripts for Epac1 and -2 were also detected in IM-PTEC (Figure 1B) and Epac1 expression was confirmed (Figure 1C and F) and strongly upregulated under restrictive conditions. Consistent with expression of Epac1 protein, exposure to forskolin, 8-pCPT-2’- O-Me-cAMP and 8-pCPT-2’-O-Me-cAMP-AM induced Rap1 activation (Figure 1G).
Activation of Rap1 by 8-pCPT-2’-O-Me-cAMP-AM was more pronounced compared to that
the presence of tight junctions and epithelial epithelial phenotype of the IM-PTEC. (F) Conditionally immortalized IM-PTEC were cultured under permissive (per) or restrictive (res) conditions and analyzed for expression of Epac1 by Western blotting. Tubulin (tub) was used as a loading control.
(G) IM-PTEC were exposed to vehicle (saline) as control, 50 µM 8-pCPT-2’-O-Me-cAMP (007), 10 µM Forskolin or 2.5 µM 8-pCPT-2’-O-Me-cAMP-AM (007-AM). Lysates were used for detection of activated GTP-bound Rap1 levels by pull down analysis followed by immunoblotting. Total Rap1 expression was confirmed by Western blotting.
Figure 1. (A) IM-PTEC were cultured under restrictive conditions and lysates were analyzed for expression of SV40 over a period of 8 days. Long term culture under restrictive conditions almost completely blocked SV40 expression.
Lysates from the mouse breast tumor cell line MCF10 and primary mouse TEC were used as non-immortalized controls. Beta actin was used as a loading control. (B) Total RNA was isolated from IM-PTEC after one week of restrictive culture conditions and specific implication of Epac1, Epac2 and megalin transcripts was performed. PPIB and TBP are used as control genes. Transcription of SGLT1 and -2 was present in IM-PTEC and mouse kidney tissue sections. IM-PTEC were cultured on glass coverslips and stained for (C) Epac1 and (D) nuclei. (E) ZO-1 expression was confirmed by immunofluorescence staining establishing
of 8-pCPT-2’-O-Me-cAMP, most likely reflecting the improved cellular uptake of the AM ester-conjugate 16. These results demonstrate that IM-PTEC posses a functional Epac-Rap
Figure 2. (A) Cells cultured under normal conditions (control, upper panel) or subjected to 60 min of hypoxia (lower panel) were stained for HIF-1α, F-actin and nuclei. Note translocation of HIF-1α to nucleus in response to hypoxia. Original magnification of microscopy images: 600x. (B) Cells stained positive for HIF-1α were counted and expressed as % of the total number of cells. Data are expressed as mean±SEM (*P<0.0001).
signalling pathway that can be activated either via increases in endogenous cAMP levels or directly by Epac-selective cAMP analogs.
Epac-Rap activation by 8-pCPT-2’-O-Me-cAMP-AM reduces monolayer disruption and protects the tubular barrier function during in vitro hypoxia
To mimic I/R injury in vitro, cells were submerged in paraffin oil. This hypoxia model has been shown to lead to cellular ATP depletion and induce pro-inflammatory cytokine expression 21. Sixty min of hypoxia led to cytoskeletal remodeling, loss of cell-cell contacts (Figure 2A) and stabilization of hypoxia inducible factor-1α (HIF-1α) in IM-PTEC (Figure 2B) demonstrating that this model induces bona fide hypoxia and relevant cytoskeletal and adhesive changes.
Epac-Rap signalling has been shown to promote both inter-cellular and cell-matrix adhesion 11-14. We tested whether activation of Epac prevents loss of cell-cell contacts induced by hypoxia. Cells were exposed to 50 μM 8-pCPT-2’-O-Me-cAMP, 2.5 μM 8-pCPT- 2’-O-Me-cAMP-AM, 10 μM forskolin or vehicle for 30 min prior to induction of hypoxia. At regular intervals during hypoxia, cells were fixed, stained for filamentous (F)-actin and imaged using automated fluorescence microscopy. In vehicle-treated control cells progressive disruption of the epithelial monolayer during hypoxia occurred, manifested as stress fiber formation and loss of cell-cell contacts resulting in gap formation (Figure 2C and D).
Monolayer disruption was determined by quantifying F-actin coverage and the coinciding gap formation in the monolayer. Monolayer integrity under normoxic conditions was used as a reference; disruption after 75 min of hypoxia was considered maximal (Figure 2C). Pre- treatment with 8-pCPT-2’-O-Me-cAMP-AM and forskolin significantly prevented epithelial disruption resulting in approximately 75% reduction in the total area of gaps in the monolayer. Pre-treatment with 8-pCPT-2’-O-Me-cAMP did not result in significant protection against monolayer disruption and may be due to lower uptake of 8-pCPT-2’-O- Me-cAMP by cells in vitro and consequent weaker activation of Rap (Figure 1B). These results suggest that activation of Epac-Rap signalling preserves the integrity of the epithelial monolayer in response to hypoxic injury in vitro.
To determine if hypoxia-induced monolayer disruption affected the epithelial barrier
Figure 2. (C) Cells were pretreated with vehicle (control), 50 μM 8-pCPT-2’-O-Me-cAMP (007), 2.5 μM 8-pCPT-2’-O-Me-cAMP-AM (007-AM) or 10 μM forskolin prior to start of hypoxia. Cells were fixed and stained with rhodamine-conjugated phalloidin at the indicated time points and imaged.
Monolayer disruption was quantified using image analysis. Data are expressed in arbitrary units (mean±SEM), whereby monolayer disruption of control cells was set to the values 0 and 1, representing 0 and 75 min after the start of hypoxia, respectively. *P=0.048, **P=0.0175. (D) Representative images are shown for cells subjected to normoxic conditions and 75 min of hypoxia.
Arrow heads indicate examples of sites of monolayer disruption. Box outlines indicate image area shown as detail below. Original magnification: 20x. (E) Barrier function was determined by measuring transepithelial electrical resistance (TER). Data were normalized for the TER at steady state. Monolayers composed of control cells (white bars) display a reduction in barrier function which can be significantly reduced by treatment of cells with pCPT-2’-O-Me-cAMP-AM (black bars). Data are expressed as mean±SEM (*P=0.0008), and are composed of the combined results of three independent experiments.
function, we performed transepithelial electrical resistance (TER) measurements and examined whether protection of monolayer integrity by 8-pCPT-2’-O-Me-cAMP-AM improves epithelial barrier function. Stimulation with 8-pCPT-2’-O-Me-cAMP-AM in non- hypoxic cells increased the TER compared to controls (Figure 2E), indicating a functional effect of Epac activation on barrier function in non-injured cells. Hypoxia for 60 min reduced
Figure 3. (A) IM-PTEC were treated with 2.5 µM 8-pCPT-2’-O-Me-cAMP-AM (AM) or vehicle for 30 min and subjected to 60 min of hypoxia or maintained under normal culture conditions. After 60 min, cells were fixed directly. Cells were stained for β-catenin, ZO-1 and counterstained with Hoechst to visualize nuclei. Original magnification: 600x. (B) Digital image analysis of β-catenin staining was used to quantify the total amount of fluorescence calculated as the product of the area per field occupied by β-catenin and the mean fluorescent intensity per pixel. Control treated cells (white bars) subjected to hypoxia have significantly less cell junction associated β-catenin expression compared to hypoxic cells treated with 8-pCPT-2’-O-Me-cAMP-AM (black bars) (*P=0.0005). (C) β-catenin expression at the cell junctions was quantified using digital image analysis and used to calculate the cumulative perimeter length as a measure for the total length in pixels of all cell junctions per field.
the TER of vehicle treated cells by approximately 20% compared to steady state, normoxic conditions. This reduction was significantly inhibited by pre-treatment with 8-pCPT-2’-O- Me-cAMP-AM, resulting in a TER level equivalent to that under steady state conditions.
From these experiments we conclude that pre-treatment with 8-pCPT-2’-O-Me- cAMP-AM preserves the epithelial barrier function during hypoxia.
Exposure to 8-pCPT-2’-O-Me-cAMP-AM prevents loss of epithelial adherens junctions during in vitro hypoxia
To determine whether the protective effect of 8-pCPT-2’-O-Me-cAMP-AM on monolayer disruption during hypoxia results from an effect on cell-cell junctions, cells were stained for the adherens junction protein β-catenin and the tight junction protein ZO-1 (Figure 3A).
Under normal conditions, IM-PTEC exposed to vehicle or 8-pCPT-2’-O-Me-cAMP-AM display pronounced ZO-1 and β-catenin staining at the cell membrane. After 60 min of hypoxia, both ZO-1 and β-catenin localization were disrupted. Treatment with 8-pCPT-2’-O- Me-cAMP-AM before hypoxia reduced loss of β-catenin from the plasma membrane but did not significantly prevent loss of ZO-1. We used automated segmentation and analysis of the β-catenin images to measure two different parameters of β-catenin staining; total junctional staining, determined by the product of signal intensity and total staining area (Figure 3B) and cumulative junction length of the stain (Figure 3C). Both parameters were decreased during hypoxia. Pre-treatment with 8-pCPT-2’-O-Me-cAMP-AM significantly reduced the loss of β- catenin expression at the cell junctions compared to vehicle treated hypoxic cells and suggests that 8-pCPT-2’-O-Me-cAMP-AM-treatment protects adherens junctions from disassembly during hypoxia.
Figure 3. During hypoxia the cumulative length of cell junctions containing β-catenin in control treated cells (white bars) was significantly lower than that found in 8-pCPT-2’-O-Me-cAMP-AM treated cells (black bars) (*P=0.0001). (D) Immunostaining of β-catenin was performed in Epac1 deficient cells. Mock, siEpac1 and siGFP treated cells were subjected to hypoxia and stained for β- catenin. Mock and siGFP treated cells showed reduced cell junction disassembly following incubation with 8-pCPT-2’-O-Me-cAMP-AM. Cells treated with siEpac1 showed low responsiveness to 8-pCPT-2’-O-Me-cAMP-AM as is evident by impaired junction stability during hypoxia.
Interestingly, under normoxic culture conditions Epac1 deficiency did not appear to affect establishment of adherens junctions in monolayer formation when compared to controls (original magnification: 10x). (E) Four separate lysate samples of mock, siEpac1 and siGFP treated cells were used for determination of Epac1 expression by Western blotting. In contrast to control mock and siGFP treated cells, exposure to specific siRNA reduced Epac1 expression. Tubulin (tub) expression was used as a loading control. (F) Representative quantification of junction-associated β-catenin expression by digital image analysis expressed as the product of the area and the mean intensity of β-
catenin specific signal per whole field normalized to the respective vehicle treated normoxic control.
Mock and siGFP treated cells displayed low β-catenin expression during exposure to hypoxia (hyp) and vehicle (white bars). Exposure to 8-pCPT-2’-O-Me-cAMP-AM (black bars) prior to hypoxia retained β-catenin expression to levels similar as those found in both normoxic (nor) controls. In contrast, cells treated with siEpac1 showed a poor effect of 8-pCPT-2’-O-Me-cAMP-AM exposure on hypoxia-induced loss of junction-associated β-catenin (*P=0.048, **P=0.025, ns=non-significant).
Data are expressed as mean±SEM. Results are representative of three independent experiments.
Adherens junction stabilization by 8-pCPT-2’-O-Me-cAMP-AM is mediated by Epac1 and independent of PKA
To confirm that the effects of 8-pCPT-2’-O-Me-cAMP-AM on adherens junction stability are mediated by Epac1, we examined the effects of Epac1 protein knockdown by RNA interference. Mock transfected controls and siRNA to green fluorescent protein (GFP) were used to exclude adverse effects by the transfection procedure or non-specific oligonucleotide effects, respectively.
Expression of Epac1 was significantly reduced following exposure to mouse Epac1- specific siRNA compared to mock-treated cells and following exposure to GFP siRNA
Figure 4. (A) Cells were exposed to DMSO, PO4-AM, 8-pCPT-2’-O-Me-cAMP-AM (007-AM), N6- Bnz-cAMP-AM and the PKA inhibitor Rp-8-Br-cAMPS alone or in combination with 8-pCPT-2’-O- Me-cAMP-AM (007-AM). Rap1 activation was examined by pull-down analysis followed by immunoblotting. Inhibition of PKA by Rp-8-Br-cAMPS did not block activation of Rap1 by 8-pCPT- 2’-O-Me-cAMP-AM. Activation of PKA by N6-Bnz-cAMP-AM did not induce significant activation of Rap1. (B) Cells were treated with N6-Bnz-cAMP-AM, 8-pCPT-2’-O-Me-cAMP-AM (007-AM) or controls (DMSO and PO4-AM) or (C) DMSO and 8-pCPT-2’-O-Me-cAMP-AM (007-AM) alone or in combination with Rp-8-Br-cAMPS and subjected to hypoxia. B-catenin was stained and (D) analyzed by digital image analysis. Values were normalized versus normoxic (white bars) DMSO controls. 8-pCPT-2’-O-Me-cAMP-AM only prevented junction disassembly during hypoxia (grey bars) and was not inhibited by Rp-8-Br-cAMPS co-exposure (*P<0.0001). Data are expressed as mean±SEM. Results are representative of three independent experiments.
(Figure 3E). Disruption of Epac-Rap signalling blocked 8-pCPT-2’-O-Me-cAMP-AM- mediated protection against adherens junction disassembly during hypoxia but was unaffected in mock and siRNA controls (Figure 3D). Quantification of β-catenin staining supported these observations (Figure 3F). Knockdown of Epac1 by expression of shRNA specific for mouse Epac1 gave similar results on 8-pCPT-2’-O-Me-cAMP-AM-induced adherens junction stability during hypoxia (data not shown).
We also examined the contribution of PKA signalling to cell-cell junction stabilization.
Treatment of cells with the PKA selective activator N6-Bnz-cAMP-AM did not stabilize cell- cell junctions during hypoxia (Figure 4B and D). A control for the AM-ester, PO4-AM, was also without effect. Inhibition of PKA with Rp-8-Br-cAMPS did not inhibit the junction stabilization induced by treatment with 8-pCPT-2’-O-Me-cAMP-AM (Figure 4C and D) but Rp-8-Br-cAMPS prevented phosphorylation of serine 133 (pSer133) on CREB 22 (data not shown). Consistent with these observations, N6-Bnz-cAMP-AM did not induce Rap activation and Rp-8-Br-cAMPS did not inhibit Rap activation induced by 8-pCPT-2’-O-Me- cAMP-AM (Figure 4A). These results indicate that activation of PKA is not sufficient to prevent hypoxia-induced disruption of cell-cell junctions and is not required for the protective effects of 8-pCPT-2’-O-Me-cAMP-AM and together with the results of the Epac1 protein knockdown experiments show the crucial involvement of Epac1 in mediating the activity of 8-pCPT-2’-O-Me-cAMP-AM independent of PKA.
Exposure to 8-pCPT-2’-O-Me-cAMP-AM reduces loss of epithelial focal adhesions during in vitro hypoxia
We generated IM-PTEC with stable expression of paxillin labeled with GFP as a marker for
Figure 5. (A) IM-PTEC with stable expression of GFP-labeled paxillin were treated with 2.5µM 8- pCPT-2’-O-Me-cAMP-AM (007-AM) or vehicle for 30 min and subjected to 60 min of hypoxia or maintained under normal culture conditions. After 60 min cells were fixed, stained and analyzed by confocal microscopy. Cells cultured under normoxic conditions either with or without prior 8- pCPT-2’-O-Me-cAMP-AM exposure display paxillin containing focal adhesion complexes throughout the cell. Cells subjected to hypoxia lose the vast majority of their focal adhesion complexes and fibrillar adhesions. Cells treated with 8-pCPT- 2’-O-Me-cAMP-AM and subjected to 60 min of hypoxia maintain focal adhesions. Original magnification 600x. (B) Focal adhesion disassembly was quantified by determining the percentage per total area displaying GFP signal.
Data are expressed as mean±SEM (*P=0.04).
Results are representative of three independent experiments.
focal adhesions 23 (Figure 5A). Hypoxia significantly reduced paxillin localization indicative of decreased matrix adhesion of cells (Figure 5B). Treatment with 8-pCPT-2’-O-Me-cAMP- AM partially preserved focal adhesions during hypoxia with a minor decrease in number of paxillin puncta compared to normoxic controls. Treatment with 8-pCPT-2’-O-Me-cAMP- AM also appeared to change the pattern of phosphorylated paxillin distribution, preserving staining preferentially in the proximity of the cell-cell junctions (data not shown).
Intrarenal administration of 8-pCPT-2’-O-Me-cAMP activates renal Rap1
Next we explored the therapeutic significance of Epac-Rap pathway activation and the potential of 8-pCPT-2’-O-Me-cAMP as a prototype drug in a mouse model for I/R injury.
Expression of Epac in mouse kidney tissue and the efficacy of intrarenal 8-pCPT-2’-O-Me- cAMP treatment on Epac activation were established. In accordance with previous studies using human and rat tissue 17, Epac is expressed by the tubular epithelium including the proximal segment of the nephron (Figure 6A). We also observed expression of Epac in parietal and visceral epithelium in the glomerulus (Figure 6B). We did not detect Epac expression in capillary endothelium.
8-pCPT-2’-O-Me-cAMP was administered by intrarenal injection. To demonstrate that this approach effectively results in activation of Rap1 we clamped the renal pedicles of mice and administered either saline or 8-pCPT-2’-O-Me-cAMP. Kidneys were collected after 30 min and analyzed for Rap1 activation by performing a Rap1 pull down assay on lysates of frozen kidney sections. Administration of 8-pCPT-2’-O-Me-cAMP resulted in a significant activation of Rap1 compared to saline-treated control kidneys (Figure 6D and E) demonstrating that our approach is effective in activating renal Epac. As the vast majority of Epac expressing cells in the kidney are of tubular epithelial origin, activation of Rap1 after treatment with 8-pCPT-2’-O-Me-cAMP is most likely caused by activation of Epac in the tubular epithelium.
Treatment with 8-pCPT-2’-O-Me-cAMP during ischemia preserves renal function and reduces tubular epithelial cell stress
We next examined whether activation of Epac-Rap signalling can affect the pathogenesis associated with I/R injury. Both renal pedicles of mice were clamped for 25 min and directly 8-pCPT-2’-O-Me-cAMP or saline was administered intrarenally. Animals subjected to I/R injury and treated with saline showed a significant increase in plasma urea and creatinine levels at day 1 after ischemia compared to sham-operated animals (Figure 7A), demonstrating that 25 min of ischemia is sufficient to induce renal failure. When animals were treated with 8-pCPT-2’-O-Me-cAMP, a significant reduction of plasma urea was observed, suggesting that activation of Epac reduced renal failure. Renal function returned to that found in sham-operated animals at day 2 and 3. We therefore examined tissue obtained at day 1 after ischemia to determine the effect of Epac activation.
Histological parameters of corticomedullary tubular damage were scored using PAS/D stained tissue sections. Sections of ischemic kidneys showed clear signs of tissue injury compared to sham-operated kidneys but no difference in the degree of tubular injury between saline and 8-pCPT-2’-O-Me-cAMP-treated ischemic kidneys was detected (data not shown).
Immunostainings for active caspase 3 were performed to study apoptosis. Although ischemia
increased tubular epithelial apoptosis, we detected no difference in the number of positive stained cells between control and treated ischemic kidneys (data not shown).
We measured expression of keratinocyte-derived chemokine (KC) as a biomarker for post-ischemic inflammation in mouse I/R injury 24. Expression of KC was measured in homogenates of ischemic kidney tissue. No significant difference in the level of KC found in homogenates of saline and 8-pCPT-2’-O-Me-cAMP-treated ischemic kidneys (data not shown).
To determine whether the protective effect of 8-pCPT-2’-O-Me-cAMP on cell-cell contacts is responsible for the reduction in loss of renal function, we stained tissue sections for β-catenin. Tissues from both groups of sham-operated animals showed a distinct lateral membrane staining pattern (Figure 6B). Sections from saline treated ischemic kidneys showed an irregular and more cytoplasmic staining pattern. In contrast to this, localization of β-catenin in 8-pCPT-2’-O-Me-cAMP-treated ischemic kidneys resembled the pattern found in sham-operated controls.
Clusterin is an urinary marker of tubular epithelial damage 25. Its expression is increased by reactive oxygen species and involved in the antioxidant response 26. Clusterin-α expression in ischemic tissue was significantly decreased following 8-pCPT-2’-O-Me-cAMP treatment compared to controls (Figure 7B and E). Expression of heme oxygenase-1 (HO-1) reflects the cellular response to oxidative stress 27. We found that the number of HO-1 expressing cells was significantly lower following 8-pCPT-2’-O-Me-cAMP treatment compared to ischemic controls (Figure 7C and E). These results suggest that treatment with 8-pCPT-2’-O-Me-cAMP reduces tubular epithelial cell stress during I/R injury. The occurrence of intraluminal obstruction in the renal papilla was scored by determining the number of tubules located in the inner stripe of the outer medulla containing cellular debris.
The incidence of intraluminal obstruction was significantly decreased in kidneys treated with 8-pCPT-2'-O-Me-cAMP compared to controls (Figure 7F and G)
Figure 6. Kidney sections were stained for (A and B) Epac1 expression or (C) labeled with secondary antibodies only. (D) GTP- Rap1 pull down analyses demonstrated that treatment with 8-pCPT-2’-O-Me-cAMP (007) induced Rap activation. Rap activation was quantified by densitometric analysis and expressed as the amount of active Rap1 over total Rap1. In a repeat experiment similar results were obtained.
Data are expressed as mean±SEM (*P=0.03). (E) Representative Western blot image with samples from saline (-) and 8- pCPT-2’-O-Me-cAMP (+) treated kidneys.
Discussion
In the present study we demonstrate that activation of the Epac-Rap signalling pathway in a mouse model for I/R injury reduces renal failure and tubular cell stress. Using an in vitro hypoxia model we established that 8-pCPT-2’-O-Me-cAMP-AM-induced Epac activation prevented disruption of the epithelial monolayer and maintained tubular barrier function by
Figure 7. (A) Plasma samples were collected 24 h after surgery from sham-operated and 24, 48 and 72 h for I/R animals. Plasma urea (upper graph) and creatinine (lower graph) were measured.
Treatment with 8-pCPT-2’-O-Me-cAMP (007) significantly reduced plasma urea at day 1 following ischemia compared to saline treated controls. Data are expressed as mean±SEM (*P=0.026).
preventing disassembly of adherens junctions and focal adhesion complexes. The second messenger cAMP is involved in vasopressin-induced water reabsorption in the collecting duct and renin production by juxtaglomerular cells (reviewed by Szaszák et al. 28). Activation of PKA by cAMP controls processes ranging from cellular metabolism, cycling and growth.
The discovery of Epac as an effector protein activated by cAMP has extended this range of cellular processes to include Rap-mediated events such as cell adhesion.
The phosphodiesterase inhibitors olprinone 29, 30 and rolipram 31 improve renal function and reduce inflammation following I/R injury. Anas et al. 30 attribute this effect to a cAMP- PKA or p38 MAPK-mediated decrease of NF-kB activity and interleukin-8 (IL-8) expression.
We did not observe any effect on renal expression of the mouse IL-8 ortholog KC following I/R injury, indicating that Epac activation during ischemia is likely not involved in inflammation.
Administration of 8-pCPT-2’-O-Me-cAMP reduced clusterin-α and HO-1 expression and prevented abnormal β-catenin distribution after ischemia. Although no signs of tubular dedifferentiation were yet observed, cytoplasmic β-catenin localization has been implicated in epithelial to mesenchymal transition 32.
Besides promoting cadherin-mediated cell junction adhesion 33 activation of Rap1 by guanine nucleotide exchange factors (Rap1GEFs) such as C3G, PDZ-GEF and Epac promotes clustering of integrins thereby enhancing cell-matrix adhesion properties 34.
Figure 7. (B) Clusterin-α expression in kidney sections was analyzed using immunostainings and image analysis software. Representative images of each group are shown. Clusterin-α expression in saline treated animals was increased at day 1 after ischemia, compared to sham operated animals.
Image details on the bottom row are magnifications of the boxed areas. Original magnification: 200x.
(C) Representative tissue immunostainings for HO-1 on kidney sections from sham-operated animals and animals sacrificed at day 1 after ischemia. White arrows indicate examples of positive cells.
Original magnification: 200x. (D) Paraffin embedded tissue sections from animals sacrificed at day 1 after ischemia were stained for β-catenin (white) and counterstained with Hoechst 33258 (blue). B- catenin is localized at the cell membrane in tissue from sham-operated animals. During I/R injury, β- catenin shows an irregular, more pronounced staining pattern in saline treated control animals, whereas its expression in 8-pCPT-2’-O-Me-cAMP (007) treated animals shows an expression pattern that is more equal to that found in shams. Original magnification: 400x. (E) Analysis of HO-1 and clusterin-α tissue stainings. (left) HO-1 positive cells located in the corticomedullary area were counted. Significantly more HO-1 positive cells were found in ischemic control-treated kidneys (white bars) compared to kidneys treated with 8-pCPT-2’-O-Me-cAMP (007, black bars). (right) Digital image analysis was used for signal quantification of clusterin-α staining. Treatment with 8- pCPT-2’-O-Me-cAMP (007, black bars) significantly reduced clusterin-α expression during I/R injury compared to saline treated controls (white bars). Data are expressed as mean±SEM (*P=0.005,
**P=0.04). (F) Incidence of intraluminal obstruction with cellular debris of tubules located in the inner stripe of the outer medulla was examined in PAS/D stainings on kidney sections from control and 8-pCPT-2’-O-Me-cAMP (007) treated animals (left panel). Right panel images are magnifications of boxed areas are magnified Black arrows indicate presence of nuclei. Original magnification: 400x. (G) Quantification of tubule obstruction show a significant decrease in the number of obstructed tubules in the inner stripe of the outer medulla at day 1 after ischemia after 8- pCPT-2’-O-Me-cAMP (007) treatment (black bars) compared to saline controls (white bars). Data are expressed as mean±SEM (*P=0.05).
Similarly we found that Epac activation preserved the focal adhesion component paxillin during hypoxia (Figure 5A).
In accordance with previous findings describing an important role for Epac in mediating endothelial barrier function 19, we found that activation of Epac reduced loss of barrier function during hypoxia (Figure 2E). The inability of 8-pCPT-2’-O-Me-cAMP-AM to prevent tight junction disassembly during in vitro hypoxia is consistent with previous findings that Rap activation in epithelial cells stabilizes adherens junctions but not tight junctions 14. Previous studies have demonstrated that adherens junctions stabilization was sufficient to strengthen epithelial barrier function 35.
Our findings using ECIS suggest that loss of TER is reduced, but we can not rule out that this effect on TER may be influenced by Epac-mediated cell spreading affecting the intercellular space in an adhesion independent fashion. Here we demonstrated that hypoxia- induced loss of focal adhesions is reduced by activation of Epac using paxillin as a functional marker (Figure 5). In addition, we found that Epac activation induced focal adhesion dynamics by promoting phosphorylation of Tyr118 on paxillin23 during hypoxia (data not shown).
Loss of cell-cell adhesion has been linked to renal failure in patients undergoing allograft transplantation 5. Furthermore, loss of matrix adhesion is thought to lead to exfoliation of viable tubular epithelial cells in the urine and has been described in patients with acute renal failure and animal models 36. In the present study, we have not examined if this process occurs in our in vivo experiments but did find decreased intraluminal obstruction in 8-pCPT- 2’-O-Me-cAMP-treated kidneys following ischemia (Figure 7F and G). Our in vitro data suggest that Epac-mediated cell junction and focal adhesion stabilization underlie these findings however direct evidence of this proposed mechanism will require additional experiments.
Addition of the AM-ester to 8-pCPT-2’-O-Me-cAMP increases its biological activity by facilitating cellular uptake 16. The stability of 8-pCPT-2’-O-Me-cAMP-AM is however dependent on the absence of extracellular esterases which would remove the AM-ester from the molecule. Due to high levels of esterases in serum, we chose to use 8-pCPT-2’-O-Me- cAMP injected intrarenally in our in vivo experiments. It is likely that Epac-specific cAMP analogs that have improved renal uptake can widen the therapeutic window of this prototype drug in follow-up studies.
In conclusion, we propose that enhancement of tubular epithelial cell adhesion in general and specifically activation of the Epac-Rap signalling pathway represents a novel therapeutic strategy for reducing renal failure during early I/R injury.
Concise methods Antibodies and reagents
Epac antibodies were generated in the laboratory of J.L. Bos, Utrecht, The Netherlands. For Epac1 detection the mouse monoclonal 5D3 was used for Western blot analysis 13 and the rabbit polyclonal 2293 for immunostainings 17. Rabbit anti-Rap1 (121), goat anti-clusterin (M-18) and mouse anti-beta actin-IgG were purchased from Santa Cruz Biotech (San Cruz, CA). Rabbit anti-ZO-1-IgG was from Zymed (Burlington, NC), the anti-pY118-paxillin and anti-cleaved caspase 3 were from Cell Signaling (Danvers, MA), the anti-β-catenin from BD
Biosciences (San Jose, CA). The mouse anti-tubulin antibody was purchased from Sigma (St.Louis, MO). The mouse anti-HIF1a and anti-SV40 antibodies were from Abcam (Cambridge, UK). Secondary antibodies conjugated to HRP were from Jackson Immunoresearch (Newmarket, UK); antibodies conjugated to Alexa-488 and Cy3 and rhodamine conjugated phalloidin were from Invitrogen (Breda, The Netherlands). Forskolin was purchased from Calbiochem (Nottingham, UK); PO4-AM3, 8-pCPT-2’-O-Me-cAMP, 8- pCPT-2’-O-Me-cAMP-AM (13), N6-Bnz-cAMP and Rp-8-Br-cAMPS were from BIOLOG (Bremen, Germany). The rabbit anti-HO-1 (SPA-895) antibody was purchased from Enzo Lifesciences (Zandhoven, Belgium)
Animals and experimental I/R model
Eight week old wild type male C57BL/6 mice were purchased from Charles River (Maastricht, The Netherlands). Mice were anesthetized using Dormicum (Roche, Woerden, The Netherlands) and Hypnorm (Vetapharma, Leeds, UK). Both renal pedicles were clamped for 25 min using B-2 vascular clamps (S&T AG, Neuhausen, Switzerland). Intrarenal treatment with 8-pCPT-2’-O-Me-cAMP was performed directly after placement of clamps, by a double 20 µl administration containing 1.45 mM 8-pCPT-2’-O-Me-cAMP or vehicle (saline) in both renal poles of each kidney during ischemia (n=10 per group). All animals received one post-operative dose of buprenorfine (subcutaneous, 0.15 mg/kg, Schering- Plough, Brussels, Belgium). Shams (n=6 per group) received identical treatment without clamping of the renal arteries. Animals were sacrificed at 24, 48 or 72 h after ischemia, sham-operated animals after 24 h only. Blood samples were collected by heart puncture and transferred to heparin-coated containers containing separation gels (BD, Alphen a/d Rijn, The Netherlands). Both kidneys were removed and fixed in 4% formaldehyde or snap-frozen in liquid nitrogen. All experimental procedures were approved by the Animal Care and Use Committee of Leiden University, the Netherlands.
Histology, renal function and immunohistochemistry
Renal tissue was fixed in 4% formaldehyde for 24 h and embedded in paraffin in a routine fashion. Four µm thick sections were cut and used for all stainings. To determine tubular damage, sections were pretreated with α-amylase (Sigma), stained with periodic acid-Schiff reagent (PAS/D) and counterstained using hematoxylin. Plasma urea and creatinine concentrations were measured in a routine fashion using the autoanalyzer facility at the clinical diagnostics department of the Academic Medical Center, Amsterdam, The Netherlands. Tubular damage was scored semi-quantitatively on 10 non-overlapping fields in the corticomedullary area as described previously 21 using the criteria tubular dilatation, brush border shedding, cast deposition and epithelial necrosis on a scale from 0 – 5 based on the percentage of tubules involved: 0 = no tubular damage; 1 = ≤10%; 2 = 11 – 25%; 3 = 26 – 50%; 4 = 51 – 75%; 5 = 76 – 100% of tubules. Intraluminal obstruction was scored by counting the number of tubules located in the inner stripe of the medulla that displayed intraluminal cellular debris. For all immunostainings, tissue sections were dewaxed, treated with 0.37% H2O2 in methanol for 15 min. For the HO-1 staining, sections were boiled (10’) in a 10 mM citrate buffer (pH6.0) prior to blocking with normal goat serum (Jackson).
Primary antibodies were labeled with HRP-conjugated secondary antibody. Visualization was
performed using 3,3’-diaminobenzidine and sections were counterstained with hematoxylin.
Sections were imaged using a Leica DM6000B light microscope (Rijswijk, The Netherlands).
Reverse transcription polymerase chain reaction (RT-PCR)
Total RNA was isolated from cells using an RNeasy® Mini Kit (Qiagen, Venlo, The Netherlands) according to the manufacturer’s protocol. CDNA was synthesized using SuperScript® III Reverse Transcriptase (invitrogen). Oligonucleotide primers are compiled in Table 1. For amplification REDTaq® DNA polymerase (Sigma) was used. For analysis of Epac1, Epac2 and megalin transcription an annealing temperature of 58°C and for SGLT1 and SGLT2 62°C was used. PCR products were analyzed on a standard 3% agarose gel and visualized with a BioCapt gel imager (Vilber Lourmat, Torcy, France).
Table 1. Primers used for RT-PCR (mouse genes)
Gene Forward Reverse
Epac1 5’-GCTGCTGCTCTTACCAGCTA-3’ 5’-CTCTGAAATCCAGGGAGTCCT-3’
Epac2 5’-CCTGGAAAAAGGAATCACACTG-3’ 5’-CAGACACTTTAACATCCAAAGACC-3’
Megalin 5’-ACAGGTGTGACCATGTCAGTG-3’ 5’-GCTCCGTTGGCACAAGTAAG-3’
TBP 5’-CAGGAGCCAAGAGTGAAGAAC-3’ 5’-GGAAATAATTCTGGCTCATAGCT-3’
PPIB 5’-AGACTTCACCAGGGGAGATG-3’ 5’-GGTGTCTTTGCCTGCATTG-3’
SGLT1 5’-GAATGGAACGCCTTGGTTT-3’ 5’- AGATACTCCGGCATCGTCAC-3’
SGLT2 5’-GGCACAGTTGGTGGCTACTT-3’ 5’-AGAGCGCATTCCACTCAAAT-3’
TBP: Tata box binding protein; PPIB: Peptidylpropyl isomerase B; SGLT1: Sodium/glucose cotransporter 1; SGLT2: Sodium/glucose cotransporter 2
Rap1-GTP pull down assay on cell and tissue lysates
To determine in vitro Rap1 activation, cells were lysed for 15 min in a lysis buffer containing 10% glycerol, 1% Nonidet P40, 50 mM Tris-HCl, pH 7.4, 200 mM NaCl, 2.5 mM MgCl2
supplemented with 1 μM aprotonin and 2 μM leupeptide. To determine in vivo renal Rap1 activation, ten 10 µm thick cryosections per sample were used for analysis. Sections were incubated with lysis buffer for 30 min at 4°C. Lysates were centrifuged and the supernatants were incubated with Gluthation Sepharose 4B (Roche) beads coated with RalGDS-RBD fusion protein as described previously 37. Samples were then used for Rap1 immunoblotting as described below.
Immunoblotting
Cells were lysed in the above mentioned lysisbuffer supplemented with protease inhibitor cocktail II (Sigma-Aldrich), sodium fluoride and vanadate. After centrifugation, supernatants were boiled for 5 min in Laemmli samplebuffer containing ß-mercaptoethanol, then subjected to protein separation and blotted on Immobilon-P (Millipore, Amsterdam, The Netherlands).
Immunoblots were blocked in Tris-buffered saline with 5% (w/v) bovine serum albumin and incubated overnight with primary antibodies. For detection, immunoblots were incubated with peroxidase-conjugated secondary antibodies and the presence of proteins was visualized
using ECL+ (Amersham, Little Chalfont, UK) on a Typhoon imager (GE Healthcare, Diegem, Belgium).
Tissue homogenates and ELISA
Frozen kidney samples were thawed in PBS containing 1% (v/v) Triton X-100 (Sigma- Aldrich) and 1mM EDTA supplemented with protease inhibitors. Renal homogenates were made using a Potter tissue grinder. Homogenates were centrifuged and the supernatant was stored at -80°C. A Duoset ELISA kit specific for detection of mouse KC (R&D Systems, Abingdon, UK) was performed according to the supplied protocol using 3,3′,5,5′- Tetramethylbenzidine (TMB, Sigma-Aldrich) as a substrate. Renal KC levels were corrected for total protein contents, measured using the Bradford technique (BioRad, Veenendaal, The Netherlands).
Cells and in vitro hypoxia
IM-PTEC were isolated from Immorto mice as described previously 21 labeled with antibodies to neprilsyin/CD10 and aquaporin 4 combined, as markers for proximal tubular epithelium 38, 39 and sorted by flow cytometry on a FacsAria cell sorter (BD Biosciences).
Cells were grown in HK-2 medium (DMEM/F12 medium (Invitrogen) with 10% fetal bovine serum (Hyclone, Etten-Leur, The Netherlands), 5 µg/ml insulin and transferrin, 5 ng/ml sodium selenite (Roche), 20 ng/ml tri-iodo-thyrionine (Sigma Aldrich), 50 ng/ml hydrocortisone (Sigma Aldrich) and 5 ng/ml prostaglandin E1 (Sigma Aldrich) with L- glutamine and antibiotics (both from Invitrogen) and mouse interferon-γ (IFN-γ, 1 ng/ml, R&D)) at 33°C in 5% CO2 and 95% air. From this cell population, monoclonal cell lines were generated by limiting dilution and examined for down-regulation of SV40 activity during restrictive conditions (culture temperature at 37°C in the absence of IFN-γ) by Western blotting (Figure 1F) with recurrence of the cobble stone-like morphology (data not shown) and megalin transcription (Figure 1G). One clone was used for all experiments described below and named IM-PTEC hereafter. IM-PTEC were transfected retrovirally with a LZRS vector encoding GFP-tagged paxillin 40 and a single cell clone (IM-PTEC/GFP-pax) was used for analysis of paxillin localization.
Cells were grown in flasks at restrictive conditions for 7 days, passed to the appropriate assay plates at high density and cultured for an additional two days. Cells were briefly serum- starved in DMEM/F12 for 2 h. Before being subjected to hypoxia, cells were pre-treated with 50 µM 8-pCPT-2’-O-Me-cAMP, 2.5 µM 8-pCPT-2’-O-Me-cAMP-AM, 10 µM forskolin or vehicle for 30 min. Hypoxia was induced by submersion of the monolayer in paraffin oil (Bufa, Uitgeest, The Netherlands) for 60 min as described previously 21. Cells on coverslips were fixed directly by adding 4% formaldehyde solution under the oil layer. Cells were washed with PBS after 10 min and processed for staining.
Short interfering (si)RNA and short hairpin (sh)RNA mediated Epac1 knockdown
Mouse-specific siGENOME SMARTpool siRNA to Epac1 and siRNA to green fluorescent protein (GFP) were purchased from Dharmacon (Lafayette, CO). Knockdown of target proteins by siRNA was performed by reverse transfection after 5 days of culture on restrictive conditions. For transfections, 100 nM siRNA were diluted in INTERFERin
(Polyplus, New York, NY) reagent and added to cell suspensions. Mock treated cells were exposed to transfection reagent only. Cells were plated in 24-well glass bottom SensoPlates (Greiner Bio-One, Alphen a/d Rijn, The Netherlands) and medium was changed after 24 h.
Cells were cultured under normal conditions for an additional two days before the experiment.
Lentiviral shRNA constructs (TRC1 Library, Mission™shRNA, Sigma-Aldrich) were provided by M Rabelink, LUMC. Lentiviral particles were generated via transfection of HEK293T packaging cells with 5 different mouse Epac1 specific shRNA constructs.
Lentiviral transduced IM-PTEC were cultured under permissive conditions and puromycin selection. Epac1 protein expression in puromycin resistant cell pools was determined by Western blotting after restrictive culture conditions. Cell assays were performed in the absence of puromycin.
Cell and cryosection immunofluorescence staining
Formaldehyde fixed cells on glass coverslips were permeabilized and blocked in PBS containing 0.05% (v/v) Triton X-100 and 0.5% (w/v) bovine serum albumin (TBP). All antibodies were diluted in TBP. Ten μm cryosections were used for immunostainings.
Sections were air dried and fixed in 4% buffered formaldehyde. Sections were permeabilized in 0.2% Triton X-100 in PBS and blocked with 5% normal horse serum (Jackson) in PBS with 0.05% Triton X-100. All antibodies were diluted in PBS with 0.05% Triton X-100. Cells and sections were counterstained with Hoechst 33258 dye. For the immunofluorescence staining of β-catenin, paraffin-embedded tissue sections were obtained and treated as described above. Autofluorescence was quenched by incubating slides in 0.1 M glycine-PBS (pH7.4) for 20 min after epitope retrieval followed by a PBS wash. Blocking and antibody labeling were performed similarly as above. A secondary antibody labeled with Cy3 and 1 μg/ml Hoechst 33258 in PBS were used to label the anti-β-catenin and nuclei, respectively.
Coverslips and sections were mounted with Aqua-poly/mount (Polysciences, Eppelheim, Germany). Stainings were imaged using a Nikon E600 epifluorescence microscope (Nikon, Tokyo, Japan), except paxillin stainings on a Bio-Rad Radiance 2100 confocal laser scanning microscope (Bio-Rad, Hercules, CA) using a 60x Plan Apo NA1.4 objective lens (Nikon).
Monolayer analysis, image analysis and signal quantification
IM-PTEC were cultured in 24 well plates, exposed to 8-pCPT-2’-O-Me-cAMP, 8-pCPT-2’- O-Me-cAMP-AM and forskolin and subjected to hypoxia in duplo as described above. At 15 min intervals, cells were fixed and stained with rhodamine-phalloidin to visualize F-actin.
Plates were imaged using a BD Pathway 855 high-content bioimager (BD Biosciences) using a long-working distance objective lens (20x magnification). Six images per well were made.
Phalloidin staining was analyzed using Image-Pro Plus v6.2 analysis software (MediaCybernetics, Gleichen, Germany). Monolayer disruption was scored by quantifying the area of gaps in phalloidin staining caused by cell-cell detachment. A score of 1 represents maximal disruption, which was caused by hypoxia treatment of control cells (with no further treatments), whereas a fully intact monolayer (control cells no hypoxia) was ascribed a score of 0 for disruption. Values from stimulated cells were expressed accordingly.
β-catenin localization was analyzed using 20 separate 600x magnified fields per treatment using Image-Pro Plus v6.2 analysis software. Stainings were analyzed by measuring cell
junction area, junction length and mean pixel intensity of junctions. For knockdown experiments, cells were stained ‘in plate’ for β-catenin as described above and imaged using the BD Pathway high-content bioimager with a low magnification (10x) long-working distance lens. Junction β-catenin signal was selected and quantified using Image-Pro Plus v6.2 analysis software by determining the percentage of area containing specific staining per whole well. Focal adhesions were identified by expression of GFP and phospho-paxillin and analyzed by determining the total area of all focal adhesions per field using Image-Pro Plus v6.2. All experiments were repeated three times.
Clusterin expression was quantified by using five 20x magnifications per stained cryosection. Clusterin expression was expressed as the percentage of the area with positive signal per total field in the corticomedullary region of the kidney using Image-Pro Plus v6.2 analysis software. HO-1 expressing cells were counted in the corticomedullary area of the kidney and expressed as the number of cells per high-power field (HPF) using a 40x objective.
Epithelial barrier function measurement
Epithelial barrier function was determined using the electric cell-substrate impedance sensing (ECIS) method on an ECIS 1600R using 8W10E electrode array slides (Applied Biophysics, Troy, NY). All measurements were performed using 400 Hz frequency. Cells were subjected to pre-stimulation and hypoxia as described above. After 60 min, an equal volume of DMEM/F12 medium was added to the cells submerged in paraffin oil, enabling re- continuation of the ECIS measurement. Barrier function was determined before and during pre-treatment with 8-pCPT-2’-O-Me-cAMP-AM and directly after recovery from hypoxia.
Statistical analyses
Results are expressed as mean ± standard error of the mean (SEM). Data were tested for normality using the Kolmogorov-Smirnow test and analyzed using an unpaired t test. Tubular injury scores were analyzed using the non-parametric Mann-Whitney U Test. Values of P≤0.05 were considered statistically significant. All statistical analyses were performed using Graphpad Prism4 (GraphPad Software, San Diego, California, USA).
Acknowledgements
This study was supported by grants from the Dutch Kidney Foundation (GS) and The Netherlands Toxicogenomics Center (NTC)/The Netherlands Genomics Initiative (NGI) (LSP).
The authors wish to thank Hans de Bont for help with image analysis and Jantine van Dijk for assistance with the TER measurements.
H.G. Genieser owns BIOLOG Life Science Institute, which sells 8-pCPT-2'OMe-cAMP for research purposes.
Reference
1. Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med 2007; 357: 797-805.
2. Chapman JR, O'Connell PJ, Nankivell BJ.
Chronic renal allograft dysfunction. J Am Soc Nephrol 2005; 16: 3015-3026.
3. Saenz-Morales D, Escribese MM, Stamatakis K, et al. Requirements for proximal tubule epithelial cell detachment in response to ischemia: role of oxidative stress. Exp Cell Res 2006; 312: 3711-3727.
4. Bush KT, Tsukamoto T, Nigam SK.
Selective degradation of E-cadherin and dissolution of E-cadherin-catenin complexes in epithelial ischemia. Am J Physiol Renal Physiol 2000; 278: F847-852.
5. Kwon O, Nelson WJ, Sibley R, et al.
Backleak, tight junctions, and cell- cell adhesion in postischemic injury to the renal allograft. J Clin Invest 1998; 101: 2054-2064.
6. Zuk A, Bonventre JV, Brown D, et al.
Polarity, integrin, and extracellular matrix dynamics in the postischemic rat kidney. Am J Physiol 1998; 275: C711-731.
7. Alderliesten M, de Graauw M, Oldenampsen J, et al. Extracellular signal- regulated kinase activation during renal ischemia/reperfusion mediates focal adhesion dissolution and renal injury. Am J Pathol 2007;
171: 452-462.
8. Cordes N. Integrin-mediated cell-matrix interactions for prosurvival and antiapoptotic signaling after genotoxic injury. Cancer Lett 2006; 242: 11-19.
9. Meredith JE, Jr., Schwartz MA. Integrins, adhesion and apoptosis. Trends Cell Biol 1997;
7: 146-150.
10. de Rooij J, Zwartkruis FJ, Verheijen MH, et al. Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature 1998; 396: 474-477.
11. Reedquist KA, Ross E, Koop EA, et al. The small GTPase, Rap1, mediates CD31-induced integrin adhesion. J Cell Biol 2000; 148: 1151- 1158.
12. Knox AL, Brown NH. Rap1 GTPase regulation of adherens junction positioning and cell adhesion. Science 2002; 295: 1285-1288.
13. Price LS, Hajdo-Milasinovic A, Zhao J, et al. Rap1 regulates E-cadherin-mediated cell- cell adhesion. J Biol Chem 2004; 279: 35127- 35132.
14. Hogan C, Serpente N, Cogram P, et al.
Rap1 regulates the formation of E-cadherin- based cell-cell contacts. Mol Cell Biol 2004; 24:
6690-6700.
15. Enserink JM, Christensen AE, de Rooij J, et al. A novel Epac-specific cAMP analogue demonstrates independent regulation of Rap1 and ERK. Nat Cell Biol 2002; 4: 901-906.
16. Vliem MJ, Ponsioen B, Schwede F, et al. 8- pCPT-2'-O-Me-cAMP-AM: an improved Epac- selective cAMP analogue. Chembiochem 2008;
9: 2052-2054.
17. Li Y, Konings IB, Zhao J, et al. Renal expression of exchange protein directly activated by cAMP (Epac) 1 and 2. Am J Physiol Renal Physiol 2008; 295: F525-533.
18. Honegger KJ, Capuano P, Winter C, et al.
Regulation of sodium-proton exchanger isoform 3 (NHE3) by PKA and exchange protein directly activated by cAMP (EPAC).
Proc Natl Acad Sci U S A 2006; 103: 803-808.
19. Kooistra MR, Corada M, Dejana E, et al.
Epac1 regulates integrity of endothelial cell junctions through VE-cadherin. FEBS Lett 2005;
579: 4966-4972.
20. Tabatabai NM, Blumenthal SS, Lewand DL, et al. Mouse kidney expresses mRNA of four highly related sodium-glucose cotransporters:
regulation by cadmium. Kidney Int 2003; 64:
1320-1330.
21. Leemans JC, Stokman G, Claessen N, et al.
Renal-associated TLR2 mediates ischemia/reperfusion injury in the kidney. J Clin Invest 2005; 115: 2894-2903.
22. Jensen BO, Selheim F, Doskeland SO, et al.
Protein kinase A mediates inhibition of the
thrombin-induced platelet shape change by nitric oxide. Blood 2004; 104: 2775-2782.
23. Zaidel-Bar R, Milo R, Kam Z, et al. A paxillin tyrosine phosphorylation switch regulates the assembly and form of cell-matrix adhesions. J Cell Sci 2007; 120: 137-148.
24. Miura M, Fu X, Zhang QW, et al.
Neutralization of Gro alpha and macrophage inflammatory protein-2 attenuates renal ischemia/reperfusion injury. Am J Pathol 2001;
159: 2137-2145.
25. Hidaka S, Kranzlin B, Gretz N, et al.
Urinary clusterin levels in the rat correlate with the severity of tubular damage and may help to differentiate between glomerular and tubular injuries. Cell Tissue Res 2002; 310: 289-296.
26. Kim JH, Yu YS, Min BH, et al. The role of clusterin in retinal development and free radical damage. Br J Ophthalmol 2007; 91: 1541-1546.
27. Nath KA. Heme oxygenase-1: a provenance for cytoprotective pathways in the kidney and other tissues. Kidney Int 2006; 70: 432-443.
28. Szaszak M, Christian F, Rosenthal W, et al.
Compartmentalized cAMP signalling in regulated exocytic processes in non-neuronal cells. Cell Signal 2008; 20: 590-601.
29. Mizutani A, Murakami K, Okajima K, et al.
Olprinone reduces ischemia/reperfusion- induced acute renal injury in rats through enhancement of cAMP. Shock 2005; 24: 281- 287.
30. Anas C, Ozaki T, Maruyama S, et al.
Effects of olprinone, a phosphodiesterase III inhibitor, on ischemic acute renal failure. Int J Urol 2007; 14: 219-225.
31. Okusa MD, Linden J, Huang L, et al.
Enhanced protection from renal ischemia- reperfusion [correction of ischemia:reperfusion]
injury with A(2A)-adenosine receptor activation
and PDE 4 inhibition. Kidney Int 2001; 59:
2114-2125.
32. Hertig A, Verine J, Mougenot B, et al. Risk factors for early epithelial to mesenchymal transition in renal grafts. Am J Transplant 2006;
6: 2937-2946.
33. Kooistra MR, Dube N, Bos JL. Rap1: a key regulator in cell-cell junction formation. J Cell Sci 2007; 120: 17-22.
34. Boettner B, Van Aelst L. Control of cell adhesion dynamics by Rap1 signaling. Curr Opin Cell Biol 2009; 21: 684-693.
35. Guo X, Rao JN, Liu L, et al. Regulation of adherens junctions and epithelial paracellular permeability: a novel function for polyamines.
Am J Physiol Cell Physiol 2003; 285: C1174- 1187.
36. Racusen LC, Fivush BA, Li YL, et al.
Dissociation of tubular cell detachment and tubular cell death in clinical and experimental
"acute tubular necrosis". Lab Invest 1991; 64:
546-556.
37. Franke B, Akkerman JW, Bos JL. Rapid Ca2+-mediated activation of Rap1 in human platelets. EMBO J 1997; 16: 252-259.
38. Kim YH, Earm JH, Ma T, et al. Aquaporin- 4 expression in adult and developing mouse and rat kidney. J Am Soc Nephrol 2001; 12: 1795- 1804.
39. Ronco P, Pollard H, Galceran M, et al.
Distribution of enkephalinase (membrane metalloendopeptidase, E.C. 3.4.24.11) in rat organs. Detection using a monoclonal antibody.
Lab Invest 1988; 58: 210-217.
40. Danen EH, van Rheenen J, Franken W, et al. Integrins control motile strategy through a Rho-cofilin pathway. J Cell Biol 2005; 169:
515-526.
