The critical role of Krüppel-like factors in kidney disease

Abstract

Krüppel-like factors (KLFs) are a family of zinc-finger transcription factors critical to mammalian embryonic development, regeneration, and human disease. There is emerging evidence that KLFs play a vital role in key physiological processes in the kidney, ranging from maintenance of glomerular filtration barrier to tubulointerstitial inflammation to progression of kidney fibrosis. Seventeen members of the KLF family have been identified, and several have been well characterized in the kidney. Although they may share some overlap in their downstream targets, their structure and function remain distinct. This review highlights our current knowledge of KLFs in the kidney, which includes their pattern of expression and their function in regulating key biological processes. We will also critically examine the currently available literature on KLFs in the kidney and offer some key areas in need of further investigation.

the word Krüppel refers to “cripple” in German. Consequently, when H. U. Gloor identified a critical mutation that led to severe body malformation in Drosophila, he labeled it Krüppel (14). Further work by Christiane Nusslein-Volhard and Eric Weischaus demonstrated that Krüppel mutations in Drosophila contribute to embryonic lethality due to significant defects in body segmentation (51). Krüppel-like factors (KLFs) are a subclass of zinc-finger family of DNA-binding transcriptional regulators that are involved in a broad range of cellular processes (i.e., cell differentiation, apoptosis, cell proliferation) (30). The COOH-terminal region of KLFs is highly conserved with three C2H2 zinc finger domains that mediate transcriptional activity by interacting with GC-rich DNA sequences. In contrast, the specificity of protein-protein and protein-DNA interactions of KLFs are determined by the NH₂-terminal region (30). KLFs are typically categorized into the following groups due to similarities in structure and transcriptional activity (1): KLF3, -8, and -12 (2); KLF1, -2, -4, -5, -6, and -7 (3); KLF9, -10, -11, -13, -14, and -16. However, KLF15 and KLF17 have not been classified since their interaction motifs have yet to be determined (30). In the past few years, there has been a dramatic increase in the number of publications on the expression and function of KLFs in the kidney. Furthermore, interrogation of recent gene expression arrays from deep sequencing of microdissected nephron segments of rat renal cortex demonstrate the diverse expression pattern of KLFs in the kidney (25) (Fig. 1). Here, we highlight the critical role of KLFs in kidney disease by reviewing recently reported data. Furthermore, we provide potential areas for further investigations.

Fig. 1.

Expression pattern of Krüppel-like factors in the kidney. Deep sequencing of microdissected nephron segments was performed in rat renal cortex by Lee et al. (25). From these reported findings (25), we extrapolated the Krüppel-like factors’ (KLFs) mRNA expression [reads per kilobase per million reads (RPKM)] from the expression arrays and highlighted its pattern of expression in each nephron segment. Nephron segments are as follows: G (glomeruli), S1 (1st segment of the proximal tubule), S2 (2nd segment of the proximal tubule), S3 (3rd segment of the proximal tubule), SDL (short descending limb of the loop of Henle), LDLOM (long descending limb of the loop of Henle in the outer medulla), LDLIM (long descending limb of the loop of Henle in the inner medulla), tAL (thin ascending limb of the loop of Henle), mTAL (medullary thick ascending limb of the loop of Henle), cTAL (cortical thick ascending limb of the loop of Henle), DCT (distal convoluted tubule), CNT (connecting tubule), CCD (cortical collecting duct), OMCD (outer medullary collecting duct), and IMCD (inner medullary collecting duct). Detailed methods for microdissection and RNA sequencing were previously provided (25).

KLFs in Glomerular Disease

Chronic kidney disease (CKD) is a leading risk factor for cardiovascular disease and stroke (15, 50). Podocytes (visceral glomerular epithelial cells) in normal mature kidneys are regarded as highly differentiated and quiescent cells. In many glomerular diseases such as Focal Segmental Glomerulosclerosis (FSGS) and HIV-associated nephropathy (HIVAN), podocytes are injured (31). In the setting of injury, podocytes undergo a major change in phenotype, resulting in a loss of podocyte cytoskeleton, actin stress fiber formation, and their terminal differentiation markers (3). A recent comparative promoter analysis of podocyte slit diaphragm molecules revealed that many podocyte-specific genes share KLF binding sites in their promoter region (7).

We recently reported that KLF15 or KKLF (kidney-enriched KLF) is required for restoration of podocyte differentiation markers under cell stress (Fig. 2) (29). KLF15 is an early inducible gene and putative binding sites for KLF15 are present in the promoter region of podocyte-specific genes such as Nephrin and Podocin (29). In addition, the overexpression of KLF15 increased the expression of Nephrin in wild-type and HIV-1 infected human podocytes (a model of podocyte dedifferentiation and injury) (29). Furthermore, knockdown of Klf15 exacerbated albuminuria and podocyte effacement with a reduction in podocyte differentiation markers in proteinuric murine models (29). Interestingly, KLF15 is a key regulator of retinoic acid (RA)-induced restoration of podocyte differentiation markers (Nephrin, Synaptopodin, and Podocin) and amelioration of podocyte injury in cultured human podocytes. Furthermore, the loss of Klf15 in mice attenuated RA-mediated restoration of podocyte differentiation markers after lipopolysaccharide treatment (29). KLF15 expression is also reduced in human primary glomerulopathies such as FSGS and HIVAN compared with healthy control subjects (29).

Fig. 2.

Role of Krüppel-like factor 15 in podocyte differentiation. We highlight the pathway by which Krüppel-like factor 15 (KLF15) mediates retinoic acid (RA) and glucocorticoids (GCs)-induced restoration of podocyte differentiation markers [actin cytoskeleton, increased expression of podocyte-specific cytoskeleton and slit diaphragm proteins (Nephrin, Podocin, and Synaptopodin)] under cell stress. GR, glucocorticoid receptor; cAMP, cyclic adenosine monophosphate; PKA, protein kinase A; CREB, cAMP response element-binding protein; GRE, GC response element; CRE, cAMP response element.

Glucocorticoids (GCs) are the initial treatment option for many glomerular diseases, such as Minimal Change Disease (MCD) and FSGS (46). Alternate immunosuppressive therapy is typically not considered until patients have failed GC therapy (35). Other than their immunomodulatory effects, recent studies demonstrate that GCs may exert their therapeutic benefits by direct action on podocytes (16, 35, 48, 49, 54). Similar to RA, we recently reported that the salutary benefits of GCs [stabilization of actin cytoskeleton and expression of podocyte-specific cytoskeleton proteins (Nephrin, Synaptopodin)] in the podocyte are directly mediated by KLF15 in cultured human podocytes and proteinuric murine models (27). In addition, the level of KLF15 expression in the glomeruli directly correlated with responsiveness to GCs in patients with primary FSGS and MCD (27).

Although Klf15^−/− mice exhibited increased susceptibility to glomerular injury, no overt injury was observed at baseline (29). Interestingly, screening of other Klfs in glomerular fractions in the Klf15^−/− mice revealed a significant increase in Klf4 and Klf6 expression, suggesting a potential compensatory mechanism (data not shown). Based on these findings, we initially investigated the role of KLF6 in glomerular disease. Similar to KLF15, KLF6 is expressed in the tubular as well as the glomerular compartment (10, 28). We also observed that KLF6 is an early injury response gene and podocyte-specific loss of Klf6 increased the susceptibility to FSGS in the resistant C57BL/6 mouse strain (28). Interestingly, KLF6 regulates the expression of cytochrome c oxidase (COX) assembly gene, critical to mitochondrial function in the podocyte. In addition, KLF6 binding sites occupy the promoter region of COX assembly gene. Consequently, the loss of KLF6 in cultured human podocytes exacerbated mitochondrial injury under cell stress, leading to activation of intrinsic apoptotic pathway and eventual cell death (28).

KLF4 was initially described to exhibit differential expression in growth-arrested intestinal epithelial cells (37). Furthermore, induction of KLF4 results in G₁/S phase cell cycle arrest in colon cancer cell lines (5). Subsequent studies have clearly demonstrated its role as a negative regulator of cellular proliferation in several types of epithelial cells by inducing cell cycle arrest in the G₁/S phase (5, 40). Furthermore, KLF4 is an essential transcription factor in the induction of pluripotency from somatic cells (42). Hayashi et al. initially reported that KLF4 is expressed solely in podocytes based on colocalization studies by immunofluorescence, but subsequent studies from the same laboratory showed that KLF4 is also expressed in endothelial cells (18, 56). Consequently, further studies are required to validate these discrepant findings. Nonetheless, the authors demonstrate that KLF4 expression is reduced in proteinuric murine models and kidney biopsies of nephrotic syndrome, and this correlated with increased methylation of the Nephrin promoter and decreased Nephrin expression (18). Restoration of glomerular Klf4 by either gene transfer of Klf4-containing plasmids, using a podocyte-specific tetracycline-based inducible system, or by treatment with an angiotensin receptor blocker (ARB), attenuated albuminuria and reduced methylation of the Nephrin promoter in proteinuric murine models, suggesting an epigenetic-mediated regulation of podocyte-specific genes (18, 19).

KLF6 and KLF15 have been reported to play a role in mesangial cell function (Table 1). KLF6 expression was induced in cultured glomerular mesangial cells in the setting of sublytic complement-mediated apoptosis (55). In addition, KLF15 expression was reduced in mesangial cells during the proliferative phase of Thy-1 rat model of mesangial glomerulonephritis (21). Interestingly, modulating the expression of KLF15 in cultured mesangial cells inversely regulated the levels of E2F1, cyclin D1, and CDK2 expression. These studies confirm the findings of KLF15 as a key regulator of cell differentiation.

Table 1.

Reported expression and function of KLFs in the kidney

Name	References	Expression Pattern in the Kidney at Baseline	Experimental Models	Reported Function and Disease Association
KLF2	(1, 39, 43, 56, 57)	Endothelial cells	STZ (mice); Uni-Nx (mice); cultured HUVEC; glomerular endothelial cells; IRI (rats); human kidney biopsies	Diabetic nephropathy; compensatory renal hypertrophy; ischemic AKI
KLF4	(4, 18, 19, 36, 38, 52, 55)	Endothelial cells and podocytes	IRI (mice); Adriamycin (mice); PAN (mice); db/db (mice); UUO (mice); cultured human podocytes; proximal tubular cells; HUVEC; human kidney biopsies	Epigenetic regulation of podocyte gene expression; mediates statin-induced renoprotective effects in ischemic AKI; kidney fibrosis; endothelial injury in antibody-mediated rejection (post-kidney transplantation)
KLF5	(4, 11)	Collecting duct cells	UUO (mice); cultured IMCD cells; proximal tubular cells	Regulates inflammatory response in kidney fibrosis
KLF6	(10, 20, 28, 54)	Podocytes; tubular cells	Adriamycin (mice); STZ (rats); cultured human podocytes; primary mouse podocytes; HK2 cells; human kidney biopsies	Podocyte apoptosis (regulates mitochondrial function); kidney fibrosis (TGF-β pathway)
KLF12	(41)	Inner medullary collecting duct cells	Kidney cortex (mice); cultured MDCK cells	Regulates expression of urea transporter (UT-A1)
KLF15	(12, 13, 21, 27, 29, 44)	Podocytes; mesangial cells; tubular cells and fibroblasts	HIV-1 transgenic mice; Adriamycin (mice); LPS (mice); anti-glomerular antibody (mice); 5/6 Nx + high-protein diet (rats); Thy-1 mesangial GN (rats); cultured human podocytes; primary mouse podocytes; rat renal fibroblasts; human kidney biopsies	Regulates podocyte differentiation markers; attenuates mesangial cell proliferation; kidney fibrosis (inhibit TGF-β-induced CTGF signaling)

STZ, streptozotocin; Nx, nephrectomy, HUVEC, human umbilical vein endothelial cells; IRI, ischemia-reperfusion injury; PAN, puromycin nephropathy; AKI, acute kidney injury; UUO, unilateral ureteric obstruction; IMCD, inner medullary collecting duct; HK, human kidney; MDCK, Madin-Darby canine kidney; LPS, lipopolysaccharide.

KLFs in Tubular Injury

KLF15 was initially described in the kidney and was subsequently coined as the “Kidney-enriched KLF” (KKLF) (44). Specifically, the authors demonstrated by immunofluorescence that KLF15 was expressed in cells where kidney-specific CLC chloride channels were absent. Interestingly, KLF15 suppresses the expression of CLC-K1 and CLC-K2 channels by transcriptionally regulating myc-associated zinc-finger protein in rat kidney cells (44). In contrast, KLF12 is expressed in the collecting ducts and colocalized to the urea transporter UT-A1 in mouse kidney (41). The authors showed that KLF12 is expressed 15 days postconception and regulates the expression of UT-A1 by increasing its promoter activity in mice (41). However, it is unclear whether KLF12 or KLF15 has a functional role in chloride or urea transport, respectively. Furthermore, gene expression arrays from Lee et al. (25) highlight a substantial level of expression in KLF9 and KLF10 compared with other KLFs in several segments of the nephron. However, little is known about the significance of KLF9 and KLF10 in the kidney. KLF9 was identified as a potential downstream target of mineralocorticoid receptor in mouse distal convoluted tubular epithelial cells in culture (45), but further studies are required to assess whether the expression of these zinc finger proteins in the tubule contributes to channelopathies in the kidney.

KLFs in Interstitial Inflammation

As in endothelial cells, KLF2 and KLF4 are expressed in T cells. Generation of regulatory T cells is critical to preventing autoimmunity in several renal and nonrenal disorders. Specifically, KLF2 was recently shown to regulate the production of T regulatory cells in mice (33), which might serve as a critical target in the treatment of cellular rejection post-kidney transplantation. In addition, KLF4 was reported as a key regulator of macrophage polarization in mice (26). Liao et al. (26) demonstrated that KLF4 expression was markedly upregulated in M2 subtype macrophages compared with M1 macrophages in mice, suggesting an increase in anti-inflammatory and pro-fibrotic phenotype. Furthermore, direct interaction between KLF4 and STAT6 in macrophages inhibited M1 programming by blocking the cofactors for nuclear factor-kappa B (NF-κβ) signaling, while activating signaling for M2 macrophages in mice (26). Since tubulointerstitial inflammation is a critical process in several renal disorders, further investigations in the role of KLFs in mediating inflammation specifically in the kidney are required.

KLFs in Fibrosis

Although the anti-inflammatory effects of KLF4 are clear, the precise role of KLF4 in renal fibrosis remains debatable. KLF4-mediated regulation of the canonical TGF-β pathway appears to be cell-specific. For instance, overexpression of KLF4 in human proximal tubular cells suppressed macrophage migration inhibitory factor (MIF) and monocyte chemotactic protein-1 (MCP-1) levels, key mediators of TGF-β pathway (32). In contrast, other laboratories have reported that KLF4 transcriptionally upregulates the expression of TGF-β in cultured cardiac fibroblasts (23). In turn, TGF-β signaling also regulates KLF4 expression. Specifically, TGF-β1 induces KLF4 phosphorylation via canonical and noncanonical pathways, which interacts with Smad2 to cooperatively activate TGF-β1 receptor in HEK 293T cells (22). In addition, TGF-β signaling through the Cdh1/APC pathway leads to ubiquitination and proteosomal degradation of KLF4 (22). Recent studies demonstrate that KLF4 expression is reduced in mice that underwent unilateral ureteric obstruction (4, 53). However, it remains unclear if this is merely an association or a consequence of fibrosis. Consequently, further studies are required to investigate the role of KLF4 in a cell-specific manner in kidney fibrosis.

KLF5 is reportedly expressed in the collecting duct (Table 1). Fujiu et al. (11) recently demonstrated that mice with haploinsufficiency for Klf5 exhibited markedly less kidney injury after unilateral ureteric obstruction. The authors further showed that the loss of Klf5 contributed to less macrophage M2 subtype accumulation in the kidney. Conversely, cooperative interaction between KLF5 and C/EBPα increased chemotactic proteins that contribute to M1 type macrophage accumulation in mice. The authors also showed that ablation of Klf5 specifically in the collecting duct contributed to progression of kidney fibrosis in mice (11). Recent in vitro studies demonstrate that increasing the matrix stiffness in cultured mouse proximal tubular cells leads to significant upregulation of KLF5 (4). In contrast, increased matrix stiffness leads to suppression of KLF4 expression in these cells. The authors also showed that increasing matrix stiffness activated ERK signaling, which contributed to Yes-associated protein 1 (YAP1) mediated stabilization of KLF5. In addition, reducing matrix stiffness directly lowered ERK/MAPK signaling, YAP1, and KLF5 expression while preserving KLF4 expression in cultured proximal tubular cells (4).

In addition to its critical role in enhancing mitochondrial function in the podocyte, KLF6 is also expressed in the proximal tubule cells in the kidney (Table 1). Previous studies have demonstrated that KLF6 transcriptionally regulates TGF-β expression in liver fibrosis. Similarly, Holian et al. (20) showed that overexpression of KLF6 in cultured proximal tubule cells exposed to high glucose reduced the expression of epithelial markers while increasing mesenchymal markers. However, these findings have yet to be demonstrated in murine models of kidney fibrosis.

Recent studies have also highlighted that KLF15 might play a role in the progression of kidney fibrosis (Table 1). By immunostaining, KLF15 is expressed in the glomerular as well as the tubulointerstitial compartments of the kidney (27, 29). Furthermore, KLF15 expression was reduced in 5/6 nephrectomized rats on a high-protein diet (12). Interestingly, protein restriction increased KLF15 expression with subsequent attenuation in kidney injury in these 5/6 nephrectomized rats (12). Furthermore, global Klf15^−/− mice demonstrated increased susceptibility to glomerulosclerosis after uninephrectomy. Interestingly, non-Smad-dependent TGF-β signaling suppressed KLF15 expression in cultured rat renal fibroblasts. Conversely, overexpression of KLF15 directly inhibited canonical TGF-β-induced CTGF signaling in cultured rat renal fibroblasts (13). Collectively, these studies demonstrate that KLFs play an important role in kidney fibrosis. However, redundancy and cooperative interaction between KLFs and TGF-β signaling in kidney fibrosis have yet to be explored.

KLFs in Renovascular Injury

Several members of the KLF family contribute to vascular homeostasis (Table 1). Although some overlap exists between KLFs in their role in endothelial biology, they remain distinct in the mechanisms by which they regulate key downstream targets. Since structural and functional similarities exist between KLF2 and KLF4, it is not surprising that both have been reported to play an active role in maintaining endothelial homeostasis. Furthermore, the specific role of these KLFs in endothelial biology has been highlighted in detail in recent reviews (2, 9).

Although initially characterized in epithelial cells, KLF4 is highly expressed in vascular endothelial cells and modulates their antithrombotic and anti-inflammatory properties. Furthermore, induction of KLF4 in cultured endothelial cells increases the expression of endothelial nitric oxide synthase (eNOS) and thrombomodulin (THBD) (47), whereas knockdown of KLF4 leads to increased Vascular cell adhesion molecule (VCAM-1) expression in TNF-α-treated endothelial cells (17). In response to proinflammatory stimuli, KLF4 is induced early with subsequent activation of anti-inflammatory pathways in cultured endothelial cells (17). Specifically in the kidney, endothelial KLF4 was protective against acute kidney injury (AKI) in the murine model of ischemia-reperfusion injury, where the loss of endothelial Klf4 increased AKI, inflammation, and the expression of adhesion molecules (ICAM-1, VCAM-1) after IRI (56). However, these findings have yet to be validated by other laboratories since the conditional deletion of Klf4 in mice was performed using tyrosine kinase promoter, which is expressed in endothelial cells and hematopoietic cells. Endothelial expression of adhesion molecules is mediated by the NF-κβ pathway (8). Interestingly, KLF4 was demonstrated to regulate NF-κβ pathway by directly inhibiting p65, the subunit required for NF-κβ activation, thereby modulating the expression of endothelial adhesion molecules in cultured endothelial cells (56). The authors also report that the endothelial-specific loss of Klf4 attenuates the pleotropic effects of statins in AKI in mice (56).

Similar to KLF4, KLF2 expression is regulated by laminar shear stress and proinflammatory stimuli and confers an anti-inflammatory and antithrombotic phenotype in the vascular endothelium (2, 17, 34). Klf2 is expressed early during mammalian development, at embryonic day 8.5 (E8.5) (24), and both global and endothelial-specific Klf2 knockout mice die as early as E14 due to a loss in blood vessel integrity, leading to hemorrhage and high-output heart failure (24, 52). In glomerular endothelial cells in culture, the expression of KLF2, and its downstream targets endothelial nitric oxide (eNOS), thrombomodulin (THBD), and endothelin-1, is upregulated upon exposure to chronic laminar flow by activation of the ERK5 signaling pathway (39). In human diabetic kidney biopsy samples and in a streptozotocin (STZ) rat model of diabetes, glomerular KLF2 was reduced compared with controls (57). Endothelial-specific knockdown of Klf2 exacerbated diabetic kidney injury and increased expression of angiogenesis markers (57). More recently, we reported that these mice are also susceptible to increased glomerular injury in the setting of glomerular hyperfiltration after unilateral uninephrectomy (58). The increased shear stress with glomerular hyperfiltration initially contributes to a renoprotective induction in Klf2 expression in cultured endothelial cells. Consequently mice with knockdown of Klf2 have lost this compensatory response, thereby exacerbating glomerular injury. Interestingly, endothelial-specific loss of Klf2 contributes to podocyte injury in these models, suggesting a potential cross-talk that needs to be further explored. Similar to KLF4, statins also induce the expression of KLF2 after ischemia reperfusion injury in mice (43).

After kidney transplantation, endothelial injury secondary to thrombotic microangiopathy (TMA) can occur as a result of chronic antibody-mediated rejection (chronic ABMR) or calcineurin inhibitor toxicity. Although these lesions are well characterized histologically, mechanisms mediating endothelial injury post kidney transplantation remain poorly understood. Interestingly, glomerular KLF2 and KLF4 expression was significantly downregulated in microdissected glomeruli from human kidney biopsies with TMA posttransplantation (1). In contrast, KLF4 expression was the highest increased transcript from gene expression arrays performed on renal cortex with chronic ABMR compared with T cell-mediated rejection (36, 38). Interestingly, the differentially expressed endothelial associated transcripts in chronic ABMR involved in inflammation, angiogenesis, adhesion, and thrombosis were regulated by induction of KLF4 expression in an independent study (47).

Since KLF2 and KLF4 share some redundancy in their function in endothelial cells, it is not surprising that they are both required for vasculogenesis in mouse embryos (6). For instance, single knockouts live longer during embryogenesis than the double knockouts of Klf2 and Klf4 due to severity in vascular injury (6). These data suggest that KLF2 and KLF4 might cooperatively interact to regulate the function of key endothelial transcripts in the setting of vascular injury in the kidney.

Conclusions

In this review, we highlight the diverse and intersecting roles of KLFs in maintaining homeostasis in the kidney. Although a majority of the literature has focused on the role of KLFs in development and regeneration, there is clear evidence from multiple laboratories that they also play a vital role in the progression of human disease. In recent years, there has been a dramatic rise in studies on KLFs in glomerular disease, inflammation, kidney fibrosis, and vascular biology. Nonetheless, significant gaps lie ahead in bridging the link between the expression and function of these zinc-finger transcription factors in renal physiology and disease.

GRANTS

This work was supported by funds from the National Institute of Diabetes and Digestive and Kidney Diseases (DK-102519), the American Heart Association (16-GRNT-31280004), and Dialysis Clinic, Inc., to S. K. Mallipattu.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

S.K.M. performed experiments; S.K.M. analyzed data; S.K.M. interpreted results of experiments; S.K.M. and C.C.E. prepared figures; S.K.M. and C.C.E. drafted manuscript; S.K.M., C.C.E., and J.C.H. edited and revised manuscript; S.K.M., C.C.E., and J.C.H. approved final version of manuscript.