Background
The human kidney is target for a vast array of disorders and kidney disease is increasing at an almost epidemic rate, imparting an enormous burden on patients and health care budgets. If the reparative mechanisms of the kidney are overrun, manifest organ injury will ensue, but the cellular source for renal epithelial regeneration has yet to be established unequivocally. The classical view that randomly surviving cells dedifferentiate and repopulate the injured tubules has recently been challenged by data favouring progenitor or stem cells as primary regenerative actors [
1]. Regardless of this debate, we and others have described remarkable changes of tubular phenotype in response to injury and during regeneration, and also that the parietal epithelial cells (PEC) of Bowman’s capsule normally express the markers of these regenerating cells [
2]. These tubular phenotypic alterations play an ill-defined role in disease resolution or progression. A protein that has consistently been found to become induced in the injured proximal tubular epithelium is caveolin-1/CAV1 [
3‐
7]. Given that caveolin-1/CAV1 and caveolae have been proposed to influence signaling, transport and disease processes in the kidney [
8‐
10] it is important to understand the mechanisms that underlie induction of caveolin-1/CAV1 and potentially other caveolar proteins in kidney disease.
Caveolae are 50–100 nm-sized membrane invaginations that play multiple roles in cell signaling and cholesterol homeostasis. They are present at high density in mesenchymal tissues such as striated and smooth muscle cells, adipocytes and in endothelial cells. Biogenesis of caveolae is a complex process, involving at least seven different genes from two families [
11]. Most well-studied are the caveolins (caveolin-1/CAV1, caveolin-2/CAV2, caveolin-3/CAV3), which are integral membrane proteins with an unusual topology, where the N- and C-termini are intracellular. Recently discovered additional components of caveolae are the four cavin isoforms: Polymerase I and transcript release factor (PTRF/CAVIN1) Serum deprivation response (SDPR/CAVIN2), Protein kinase C delta binding protein (PRKCDBP/CAVIN3), and Muscle-related coiled-coil protein (MURC/CAVIN4). Cavins are cytosolic proteins that form homo- and heterotrimers that constitute building blocks for the striated coat of caveolae [
11]. Rare null mutations in the genes responsible for caveolae result in lipo- and muscular dystrophies [
12‐
14]. In addition, homozygous loss of PTRF/CAVIN1 has effects on smooth muscle motility and cardiac rhythmicity [
14,
15]. A number of transcriptional control mechanisms for caveolin-1/CAV1 and expression of caveolae have been described. Among those with a positive regulatory impact are the myocardin family coactivators [
16], hypoxia-inducible factor 1α (HIF1α) [
17], peroxisome proliferator-activated receptor gamma (PPARG) [
18], sterol regulatory element-binding proteins (SREBP) [
19,
20] and forkhead box O (FOXO) transcription factors [
21]. Any one of these transcription factors could tentatively be mechanistically responsible for de novo expression of caveolar proteins in diseased kidney tubules. Free radicals have moreover been proposed to induce caveolae, but the underlying mechanisms have not been precisely defined [
22].
The myocardin family of transcriptional coactivators consists of four proteins: myocardin (MYOCD), megakaryoblastic leukemia 1 (MKL1), megakaryoblastic leukemia 2 (MKL2) and MEF2 activating motif and SAP domain containing transcriptional regulator (MAMSTR). These proteins stimulate transcription and play essential roles in striated and smooth muscle [
23,
24]. The term “coactivator” reflects that myocardin family members bind to DNA via other transcription factors. The best established example is through the serum response factor (SRF) and via DNA motifs referred to as CArG-boxes [
23,
24], but the cardiac isoform of myocardin may also influence transcription via MEF2 [
25]. MKL1 (Megakaryoblastic leukemia 1) moreover interacts with SMAD3 and this complex binds to a GCCG-like motif in the human Slug promoter [
26]. The latter effect is of relevance in the context of epithelial to mesenchymal transition. All myocardin family members have a conserved SAP domain (from SAF-A and -B/Acinus/PIAS) and a number of transcriptional targets appear to be SAP domain-dependent [
27]. How the SAP domain binds to DNA however remains unclear. Studies have identified important roles of myocardin family members in cardiovascular [
28,
29] and fibrotic [
30,
31] diseases, and it is becoming increasingly apparent that the pro-motile and pro-contractile effects of these coactivators are important for the metastatic process [
32,
33]. In cancer cells, it has been shown that MKL1 activity is strongly enhanced by down-regulation of its repressor SCAI (suppressor of cancer cell invasion) [
32]. MKL1 signaling is also prevented by binding to the globular form of actin (G-actin), whereas polymerization into filamentous actin (F-actin) dislocates MKL1, thereby inducing nuclear translocation and activation of MKL1 [
23]. Among the known regulatory factors positively influencing MKL1 is filamin A (FLNA). This is a structural protein that cross-links F-actin, attaches actin to integrins and also serves to anchor a set of plasma membrane proteins. FLNA acts through direct complex formation with MKL1 in the nucleus thereby enhancing transcription [
34].
Recent experimental findings have defined roles for MKL1 and MKL2 in kidney cells. Work with kidney epithelial cell lines showed that disruption of intercellular junctions caused nuclear translocation of SRF and MKL2 and resultant transcriptional activation [
35]. A recent study also showed that MKL1 regulates fibrosis in diabetic nephropathy [
36].
The aim of the present work was to investigate the molecular mechanisms responsible for induction of caveolins and cavins in injured proximal tubules in situ and in tubular epithelial cells in vitro. We investigated the influence of hypoxia, free radicals and the myocardin family coactivator MKL1. Our studies support the view that MKL1 is responsible for the caveolin distribution in healthy and sclerotic human kidneys and for the induction of caveolar proteins in kidney epithelial cells in vitro. We furthermore present correlative evidence suggesting that SCAI repression and FLNA induction may contribute to MKL1 induced transcription and expression of caveolar proteins in tubular epithelial cells.
Methods
Immunohistochemistry
Tissues were fixed in 4% neutral buffered paraformaldehyde, paraffin embedded and cut at 3 μm. Immunohistochemistry was performed as described previously [
2]. Staining was performed using the EnVision system for detection and DAKO Techmate 500 equipment, according to the instructions of the manufacturer (DAKO, Glostrup, Denmark). Chromogen was diaminobenzidine (DAB). For double stainings the Ventana Benchmark Ultra automated staining equipment was used according to routine protocols provided by the manufacturer. As chromogens, prefilled cartridges of OptiView DAB IHC Detection Kit and Alkaline Phosphatase Red (Ventana Medical Systems, Tucson, AZ) were used. The antibodies used were directed against caveolin-1/CAV1 (D46G3, 1:800, Cell Signaling, Danvers, MA), caveolin-2/CAV2 (610685), 1:500) and caveolin-3/CAV3 (610421, 1:500) both BD Transduction Laboratories, PTRF/CAVIN1 (ab78553, 1:500, Abcam, Cambridge, UK), SDPR/CAVIN2 (AF5759, 1:2000, R&D Systems, Abingdon, UK), PRKCDBP/CAVIN3 (16250–1-AP, 1:500, Proteintech, Manchester, UK) and TNC (MB1, 1:50, Novocastra, Leica Biosystems, Newcastle, UK). Hematoxylin was used as nuclear counterstain. As negative control, kidney tissue was stained according to the same protocol with exclusion of the primary antibody incubation step (Additional file
1: Figure S1).
Transmission electron microscopy
Human kidney tissue was fixed in 2% glutaraldehyde buffered with 0.1 M cacodylate/0.1 M sucrose buffer, at pH 7.2. After 60 min the tissue was osmium treated, dehydrated and embedded in Agar 100 resin. Curing was performed at 60 °C for 48 h. Ultrathin sections were cut and stained with 2% uranyl acetate for 25 min and lead citrate for 2 min. TEM was performed using a JEOL 1230 microscope (Jeol, Tokyo, Japan) and 60 K digital images were captured. Caveolar diameters were determined using ImageJ (NIH, Bethesda, MD). Dimensions of caveolae in PECs were compared with caveolae in an existing image archive from human detrusor smooth muscle [
37].
Preparation of kidney epithelial cells and isolated tubules
Ethical permission was granted by the ethical committee at Lund University (LU680–08 and LU289–07) and the procedures followed were in accordance with the Helsinki Declaration. For isolation of primary kidney epithelial cells, kidneys were obtained from nephrectomies carried out due to localized renal tumors with written informed consent. Cortical tissue distant from the tumor was isolated and placed in cold Dulbecco’s modified Eagle’s medium (GE Healthcare, Logan, UT) with 10% fetal calf serum and 1% penicillin/streptomycin (Thermo Scientific, Waltham, MA). The tissue was rinsed, minced and subjected to overnight treatment (37 °C) with collagenase I (300 U/ml, Thermo Fischer Scientific Waltham, MA) and deoxyribonuclease I type II (200 U/ml, Sigma-Aldrich, St. Louis, MO). After trituration with a 10-ml pipette, the suspension was serially passed through tissue strainers with mesh sizes of 100 and 70 μm, excluding glomeruli. The suspension was thereafter treated with trypsin–EDTA for 5 min and passed through a 20-μm strainer, resulting in single cells.
For isolation of proximal tubules, cortical tissue was placed in DMEM medium supplemented with 10% fetal bovine serum (Saveen-Werner, Limhamn, Sweden) and 1% penicillin/streptomycin solution. One mm thick, cortical sections were prepared and incubated in 250 U/ml collagenase type II (Invitrogen) solution with shaking to digest the tissue. Proximal and distal tubules were isolated manually using watchmakers forceps and a standard cell culture inverted microscope under ocular inspection. Tubules were isolated according to their varying thickness, refractive appearance and degree of convolution. The isolation protocol was validated by immunohistochemistry and lectin staining. Positive markers for proximal tubules were: CD10 (Ventana, Tuscon, AZ) and Fluorescein labeled
Lotus tetragonolobus Lectin (Vector Laboratories, Burlingame, CA). Positive markers for distal tubules were E-cadherin (Ventana, Tuscon, AZ) and
Dolichos biflorus Agglutinin (Vector Laboratories, Burlingame, CA). Results from the validation experiments for isolation of proximal and distal tubules are shown in Additional file
2: Figure S2. The tubular contamination was negligible. Cell cultures were initiated from either isolated tubules or digested cortical tissue.
Primary kidney epithelial cells and cells from isolated proximal tubules were cultured in DMEM high glucose supplemented with 10% fetal bovine serum and 1% penicillin/streptomycin solution, at 37 °C and 5% CO
2. Cell culture experiments were also conducted using serum free medium [
38] Briefly, for these experiments DMEM/F12 medium (GE Healthcare, South Logan, UT) supplemented with insulin transferrin selenium (Thermo Fischer Scientific, Waltham, MA), 10 ng/ml of epidermal growth factor (Thermo Fischer Scientific, Waltham, MA), 10 pg/ml of triiodothyronine (Sigma Aldrich, St Louis, MO), 36 ng/ml hydrocortisone (Sigma Aldrich, St Louis, MO) and 1% penicillin/streptomycin solution was used. Cells were cultured until they reached 50% confluency, 100% confluency, or until 3 or 5 days post confluency.
Treatment protocols
To induce oxidative stress, cells were treated with 700 μM H2O2 (Sigma Aldrich, St Louis, MO) for 2 h. Thereafter medium was exchanged and cells were left to recover in the incubator for 24 h, 48 h and 72 h.
Hypoxia was generated in a Whitley H35 Hypoxystation (Don Whitley Scientific, Shipely, UK). Primary renal cell cultures were placed in the hypoxia chamber at 1% pO2 and medium was changed to hypoxic pre-conditioned media. Cells were incubated for 24 h and 72 h. Control cells were incubated in 21% pO2.
To inhibit MKL1, cells were treated with CCG-1423 (Sigma Aldrich, St Louis, MO). Cells exposed to the inhibitor were harvested at 50% confluency, 100% confluency, and at 3 and 5 days post confluency. At each time point, control and treated cells were collected for Western blot analysis. Cells were scraped in ice cold PBS, spun down and snap frozen in liquid nitrogen. Pellets were stored in −80 °C until further analysis.
For viral transduction, kidney epithelial cells were seeded in 6-well plates. At 24 h after seeding, cells were transduced using 25, 50 and 100 MOI (multiplicity of infection) of Ad-CMV-MKL1-eGFP (Vector Biolabs Burlingame, CA). As a control 25, 50 and 100 MOI respectively of Ad-CMV-null was used. Cells were incubated with virus for 96 h and then harvested for qPCR analysis.
Quantitative PCR
Nucleic acid was extracted using the QIAshredder and RNeasy Mini Plus kit (Qiagen Inc., Hilden, Germany). The quality and concentration of RNA was determined using a Nanodrop spectrophotometer (Thermo Fischer Scientific, Waltham, MA). cDNA was synthesized with random primers and reverse transcriptase enzyme (MultiScribe, Applied Biosystems, Foster City, CA). The GeneAmp 7300 sequence detector was used for qPCR with the SYBR Green Master Mix (Applied Biosystems, Foster City, CA). Reactions were performed in triplicate, and three housekeeping genes (UBC, YWHAZ, and SDHA) were used for normalization (Table
1).
Table 1
Primer sequences used in the analysis
UBC | 5′-ATTTGGGTCGCGGTTCTTG | 5′-TGCCTTGACATTCTCGATGGT |
YWHAZ | 5′-ACTTTTGGTACATTGTGGCTTCAA | 5′-CCGCCAGGAGGACAAACCAGTAT |
SDHA | 5′-TGGGAACAAGAGGGCCATCTG | 5′-CCACCACTGCATCAAATTCATG |
CAV1 | 5′-GAAAGAAGATGGGGGAGGAG | 5′- AAAGTCCCCAAAGGCAGAAT |
CAV2 | 5′-ACGACTCCTACAGCCACCAC | 5′-CGTCCTACGCTCGTACACAA |
CAVIN1 | 5′-GCTCCTTCCGAACTTCCTCT | 5′-ACTTGGACAACCAGGACAGG |
CAVIN2 | 5′-CTTGTGCCTTGTCCCAAAAT | 5′-CGCGTAGCTACCCTCATAGC |
CAVIN3 | 5′- CTTGTGCCTTGTCCCAAAAT | 5′- TTATTGATGGTGAGCGCAAG |
For the qPCR results in Fig.
6, cells were lysed in Qiazol and total RNA was isolated with Qiagen miRNeasy mini kit in a Qiacube according to the manufacturer’s instructions (Qiagen, Hilden, Germany). qPCR reactions were performed on a real time thermal cycler (StepOnePlus™, Applied Biosystems, Foster City, CA) using Quantifast SYBR Green RT-PCR kit and Quantitect primer assays for: 18S, caveolin-1/CAV1, caveolin-2/CAV2, PTRF/CAVIN1 and SDPR/CAVIN2. The comparative Ct method (2–[delta][delta]Ct method) was used to quantify relative RNA levels for all qPCR experiments.
Western blotting
Frozen cell pellets were lysed with RIPA buffer containing cOmplete™ protease inhibitor cocktail (Sigma Aldrich, St Louis MO) for 20 min on ice. Every five minutes, lysates were vortexed. On completion of lysis, lysates were spun for 20 min at 4 °C to pellet remaining debris. Protein lysates from human kidney cortex were prepared by disrupting the tissue using a TissueLyser LT (Qiagen, Hilden, Germany) for two cycles of 30 s at 50 Hz in RIPA buffer. Lysates were further disrupted by sonication and spun for 20 min at 4°C to remove debris. Protein concentrations were determined using the Bradford assay. Western blotting was performed by standard methods using precast gels and the Trans-blot Turbo transfer system from BIO-RAD, (Hercules, CA). The following primary antibodies were used: caveolin-1/CAV1 (D46G3) from Cell Signaling (Danvers, MA), caveolin-2/CAV2 (610685) and HSP90 (610418) from BD Transduction Laboratories (San Jose, CA), PRKCDBP/CAVIN3 (16250–1-AP) from Proteintech (Rosemont, IL), β-actin (A5441) from Sigma-Aldrich (St Louis, MO) and PTRF/CAVIN1 (ab48824), SDPR/CAVIN2 (ab113876), PRX-SO3 (ab16830), Filamin A (ab76289) and SCAI (ab124688) from Abcam (Cambridge, UK). HRP-conjugated anti-mouse and anti-rabbit secondary antibodies from Cell Signaling (Danvers, MA) were used and bands were visualized using enhanced chemiluminescence (Pierce West Femto, Thermo Fischer Scientific, Waltham, MA) in an Odyssey Fc Imager (LI-COR Biosciences, Lincoln, NE).
Statistical analyses
Kidney mRNA expression data (
n = 32) were retrieved from the GTEx project (
https://gtexportal.org/home/) [
39] and normalized using the method of Robinson and Oshlack [
40]. Correlation coefficients were calculated using the Spearman method in Graph Pad Prism (GraphPad Software, La Jolla, CA).
P < 0.05 was considered significant. For the remainder of our experiments, pair-wise comparisons were made using an unpaired Student’s t-test. Multiple comparisons were done using one-way ANOVA followed by the Bonferroni post-hoc test.
Discussion
We find that caveolins (caveolin-1/CAV1, caveolin-2/CAV2, but not caveolin-3/CAV3) and cavins (PTRF/CAVIN1, SDPR/CAVIN2, and PRKCDBP/CAVIN3, but not MURC/CAVIN4) are highly expressed in blood vessels and in the PECs of Bowman’s capsule in healthy human kidney tissue, but absent from proximal tubules. In sclerotic kidneys on the other hand, caveolar proteins (caveolin-1/CAV1, caveolin-2/CAV2, SDPR/CAVIN2 and PRKCDBP/CAVIN3) are also expressed in atrophying proximal tubules. In vitro studies showed that caveolins and cavins are induced in epithelial cells in culture and that this can be inhibited by an MKL1 inhibitor and mimicked by viral overexpression of MKL1. We also show that CAV1/CAV2 mRNA expression levels correlate with MKL1 and with the archetypal MKL1 target tenascin C (TNC) in human kidney tissue. TNC and caveolin-1/CAV1 stainings coincide in both normal and diseased kidneys. Together these findings argue that MKL1 activity is an important determinant of caveolin-1/CAV1 distribution in healthy and diseased kidneys. It is interesting that PECs are endowed with such a broad repertoire of caveolar proteins, normally only encountered in mesenchymal cell types. To our knowledge this is clearly a unique feature of this epithelium. The stratum basale of stratified squamous epithelia, although structurally dissimilar, is the only other epithelial tissue that at least partially recapitulates this structural aspect.
Our findings are well in line with the recent finding that myocardin and MKL1 regulate caveolin and cavin expression in human coronary artery smooth muscle cells [
16]. Induction of many of the target genes of these transcriptional coactivators requires binding to the serum response factor [
23,
24]. Caveolins and cavins, with exception for PTRF/CAVIN1, were in that study however unaffected by SRF knock down [
16], arguing that they are controlled by the so called SAP domain. This domain is conserved in all myocardin family members [
23] suggesting that all of them may harbor the potential to induce caveolae. An archetypal SAP domain-dependent target of MKL1 is tenascin C [
43]. In contrast to caveolin-1/CAV1, which is an integral membrane protein, this protein resides in the extracellular matrix where it functions as an integrin ligand. Distinct localizations were indeed revealed in our co-staining experiments for TNC and caveolin-1/CAV1 which supported overall co-localization. It could be argued that coincident detection of nuclear MKL1 in the same cells would further strengthen the hypothesis that MKL1 was responsible for caveolin-1/CAV1 induction. An assessment of MKL1 distribution at a single time point is, however, a poor reflection of time-integrated MKL1 activity. In fact, distribution of a long-lived target protein, such as TNC, might be a better record of historical MKL1 activity.
Our studies on primary kidney epithelial cells demonstrated an MKL1-dependent induction of caveolins and cavins in culture. This finding is in good agreement with the work by Zhuang et al. [
7], showing that while proximal tubular cells in situ are negative for these proteins, they are readily detectable when proximal tubules, or cells derived from them, are cultured in vitro. It has been previously demonstrated that cell dissociation, cell shape changes, and spreading cause MKL1 activation [
35,
41]. Similar processes could be responsible for MKL1 activation and the following caveolin-1/CAV1 induction in the cultured kidney epithelial cells studied herein. MKL1 activity has also been shown to be modified by a number of coactivators and repressors. In cancer cells, MKL1 signaling has been shown to be inhibited by nuclear complex formation with SCAI, and in the original study a qPCR based tissue screen indicated high SCAI levels in the kidney [
32]. We therefore hypothesized that SCAI may exert an inhibitory effect on MKL1 activity in proximal tubular cells. Furthermore, MKL1 signaling is intimately connected to the ratio of filamentous to globular actin. FLNA is a protein that functions to interconnect F-actin strands and also acts as a nuclear enhancer of MKL1 activity. Having noticed high levels of caveolar proteins in cultivated tubular cells, but absence of tubular expression in normal human kidney, we decided to analyze the presence of these modifying factors. We thus blotted for caveolin-1/CAV1, PTRF/CAVIN1, FLNA and SCAI in cortical tissue lysates from human kidney and lysates from primary tubular cells cultivated from the cortex of the same kidneys. Interestingly, FLNA was absent from cortical tissue lysates, but strongly induced in parallel with the increase in caveolin-1/CAV1 and PTRF/CAVIN1 expression in cultured proximal tubular cells. SCAI expression on the other hand virtually disappeared when the cortical cells were cultured. These results support the notion that MKL1 activity in tubular cells is enhanced by increased levels of FLNA and reduced expression of SCAI when cells are cultured in vitro.
We can rule out PEC or other contaminating cells as an explanation for increased levels of caveolar proteins in cultured kidney cortex cells. This is because caveolar protein induction was seen in small micro-dissected proximal tubular segments that should contain only proximal epithelial cells and no PECs. If increased levels of caveolar proteins were due to some rare cell, only a minor fraction of outgrowing cells would stain positive for caveolin-1/CAV1. When proximal tubules were manually isolated and cells dissociated, close to 100% of the resulting colonies were positive for caveolin-1/CAV1.
The induction of caveolar proteins has been suggested to occur through different mechanisms. Previous studies on the effects of hypoxia on caveolae have yielded contradictory results. Initial studies by Wang et al. [
17] showed higher expression of caveolin-1/CAV1 in clear-cell renal cell carcinoma, where HIF1α and HIF2α are constitutively active. This was mimicked by hypoxia and was normalized by expression of the von Hippel Lindau (VHL) factor. A conserved hypoxia response element capable of binding HIF1α and HIF2α was moreover identified in the caveolin-1/CAV1 promoter, and VHL loss was associated with an increased density of caveolae. Together, these findings supported the view that caveolae are positively regulated by HIF. Regazzetti et al. [
44] on the other hand found that hypoxia reduced cavin-1 and cavin-2 levels in adipocytes whereas caveolin-1/CAV1 levels remained unchanged. Hypoxia moreover reduced the density of caveolae at the ultrastructural level and these changes were mimicked by the HIF-1α inhibitor echinomycin. The apparently opposing results of these studies could be due to the use of different cells. We found, using primary kidney epithelial cells, that hypoxia reduced the mRNA levels of caveolin-2/CAV2 and SDPR/CAVIN2 whereas the caveolin-1/CAV1 level remained unchanged, in good agreement with the findings of Regazzetti et al. [
44]. We conclude that hypoxia is unlikely to be a critical factor for induction of caveolar proteins in sclerotic kidneys.
A previous study has demonstrated that sublethal hydrogen peroxide concentrations up-regulate PTRF/CAVIN1 levels in human and mouse fibroblasts. H
2O
2 moreover increased association of caveolin-1/CAV1 with PTRF/CAVIN1 and increased the density of caveolae [
22]. We speculated that this mechanism could be responsible for ectopic caveolin expression in sclerotic kidneys. We were however unable to induce caveolins and cavins using H
2O
2 despite testing different treatment protocols and close monitoring of cell viability. We cannot explain why fibroblasts and kidney epithelial cells respond differently to free radicals, but the proteomes of these cells are likely to be vastly different. The repressive effect of H
2O
2, especially on caveolin-2/CAV2, PTRF/CAVIN1 and SDPR/CAVIN2 protein levels and the lack of effect on caveolin-1/CAV1 protein, argued against a major role of oxidative stress in induction of caveolar proteins in kidney disease.
This study is not the first to document increased staining for caveolin-1/CAV1 in diseased proximal tubular epithelial cells. Vallés et al. [
6] demonstrated increased proximal tubular staining for caveolin-1/CAV1 following longstanding ureteral obstruction. Caveolin-1/CAV1 induction has also been demonstrated in ischemic [
3,
4] and toxic [
5] kidney injury. Induction of caveolin-1/CAV1 thus appears to represent a conserved response of injured/regenerating proximal tubules. This may be a reflection of underlying MKL1 activation in regenerating epithelia. Recent work has established a role of MKL1 in kidney fibrosis [
36,
42], but associated induction of caveolin-1/CAV1 was not demonstrated. The upstream mechanism of MKL1 activation in regenerating kidneys is presently unclear. Co-activation by FLNA and repression of SCAI, an inhibitor of MKL1 activity [
32] that has recently been demonstrated to play a role in kidney fibrosis [
42], are attractive possibilities.