NFATc1 Identifies a Population of Proximal Tubule Cell Progenitors
Melissa Langworthy*,,
Bin Zhou,
Mark de Caestecker*,,
Gilbert Moeckel and
H. Scott Baldwin*,
* Department of Cell and Developmental Biology, Division of Pediatric Cardiology, Department of Pediatrics, Nephrology Division, Department of Medicine, and Renal Pathology Division, Department of Pathology, Vanderbilt University Medical Center, Nashville, Tennessee
Correspondence: Dr. H. Scott Baldwin, Division of Pediatric Cardiology, 2213 Garland Avenue, 9435-A Medical Research Building IV, Nashville, TN 37232-2495. Phone: 615-322-2703; Fax: 615-322-6541; E-mail: scott.baldwin{at}vanderbilt.edu
Received for publication January 23, 2008.
Accepted for publication September 22, 2008.
Recovery from acute kidney injury requires regeneration of tubulecells. Because calcineurin induces nuclear transport of NFATcproteins, whose expression pattern correlates with the nephronsegments injured by calcineurin inhibitors, we hypothesizedthat NFATc1 plays a role in modifying epithelial regenerationafter injury. To test this, we induced proximal tubular cell(PTC) injury in Balb/c mice and Nfatc1+/– mice with mercuricchloride; the PTCs of Nfatc1+/– mice demonstrated increasedapoptosis, sustained injury, and delayed regeneration. To attenuateNFATc1 activity further, we injected cyclosporin A daily. CyclosporinA–treated Nfatc1+/– mice demonstrated rapid andsevere injury after administration of mercuric chloride, withincreased serum creatinine, increased apoptosis, decreased PTCproliferation, and increased mortality compared with similarlytreated wild-type mice. Using a novel NFATc1 transgenic linethat reports activation of an NFATc1 enhancer domain criticalfor NFATc1 autoamplification, we demonstrated accentuated NFATc1expression in a PTC subpopulation after mercuric chloride–inducedinjury. In addition, NFATc1-labeled, apoptosis-resistant PTCsproliferated to repair the damaged proximal tubule segment.These data provide evidence for a resident progenitor PTC populationand suggest a role for NFATc1 in the regeneration of injuredproximal tubules.
Acute kidney injury (AKI) results from functional and structuralchanges after exposure to an environmental and/or an occupationaltoxin or results from treatment with chemotherapeutic and/orimmunosuppressive drugs.1 Despite the prevalence of AKI, themechanisms of injury and repair are still poorly understood.
NFATc proteins were originally described as transcription factorsin activated T cells. Phosphorylated NFATc proteins are cytoplasmicin resting T cells. After a sustained increase in intracellularCa2+, the Ca2+-dependent phosphatase calcineurin dephosphorylatesNFATc proteins, inducing their nuclear transport. Active calcineurinisoforms A and Aβ have been identified in the proximaltubule,2 and NFATc (NFATc1 through 4) proteins are expressedpredominantly in cortical tubules,3,4 correlating with the nephrotoxicityassociated with cyclosporin A (CsA). Thus, NFATc may play arole in proximal tubular cell (PTC) injury and repair, whichhas not been previously appreciated.
We examined the requirement of NFATc1 for regeneration of injuredPTCs and showed that NFATc1 is important for maintaining a PTCsubpopulation that participates in proximal tubule repair afterAKI. Mice generated by two independent single-exon deletionsof NFATc1 showed that Nfatc1–/– mice die in uteroof cardiac valve and septal defects5,6 and are therefore notaccessible for postnatal studies; however, after we inducedAKI by mercuric chloride (HgCl2) administration, Nfatc1+/–mice demonstrated increased apoptosis and delayed regeneration.To attenuate NFATc1 activity further, we administered low-dosageCsA, which exaggerated the defects observed in Nfatc1+/–mice and resulted in renal failure and death after HgCl2 injury.Using transgenic mouse lines that express LacZ (NFATc1-P2-LacZ7)or Cre-recombinase (NFATc1-P2-Cre) under the control of an NFATc1autoregulatory enhancer, we demonstrated that NFATc1 is expressedin a subset of PTCs after HgCl2 injury. These cells are resistantto apoptosis and act as a progenitor cell population that proliferatesto repopulate the proximal tubule. This is the first phenotypeidentified in the Nfatc1+/– mouse and suggests a rolefor NFATc1 in the regeneration of injured PTCs by an apoptosis-resistant subpopulation of PTCs.
Nfatc1+/– Mice Have Sustained AKI
To determine whether genetic attenuation of NFATc1 affectedthe severity of renal injury and rate of PTC repair, we inducedAKI in wild-type (WT) and heterozygous null Nfatc1+/–mice by administration of a single dose of HgCl2 (8.14 mg/kg)that causes a marked injury in the S2 and S3 proximal tubulesegments in Balb/c mice.9 Compared with WT mice, we did notobserve morphologic or histologic differences in Nfatc1+/–mice before HgCl2 administration (Figure 1, A and F). One dayafter HgCl2 injury, we observed AKI marked by tubular dilation,cellular necrosis, nuclear dropout, and the loss of definedbrush border membranes in PTCs in both WT and Nfatc1+/–mice (Figure 1, B and G). At day 3, PTCs in WT mice containedlarge basophilic nuclei characteristic of proliferation, whereasNfatc1+/– mice showed fewer basophilic PTC nuclei, suggestingthat proximal tubules remained in an injured state (Figure 1,C and H). At day 5, PTCs of WT mice showed reorganization, whereasthe PTCs of Nfatc1+/– mice were disorganized or dilated(Figure 1, D and I). At day 10, the injured PTCs of WT miceregenerated and recovered from injury as indicated by definedcellular membranes and brush borders (Figure 1E); however, theinjury from HgCl2 was sustained at 10 d in Nfatc1+/– micebecause the nephron segment was not remodeled into a propertubule and had disorganized interstitial cells (Figure 1J).
Figure 1. Genetic attenuation of NFATc1 causes sustained HgCl2-induced AKI. (A and F) There was no detectable difference between WT (A) and Nfatc1+/– (F) mice before administration of HgCl2. (B and G) One day after a single injection of HgCl2 (3 mM), AKI, marked by tubular dilation, nuclear dropout, and loss of brush borders, is observed in both WT (B) and Nfatc1+/– mice (G). (C) Three days after HgCl2 injury, PTCs in WT mice contained large basophilic nuclei in the regenerating proximal tubule segment. (H) Nfatc1+/– mice showed fewer basophilic nuclei, suggesting that the PTC remained in an injured state. (D) Five days after injury, the WT mice have basophilic nuclei and fewer injured tubules. (I) Nfatc1+/– PTCs have basophilic nuclei 5 d after HgCl2 injury. (E) Ten days after injury, the injured PTCs of WT mice have regenerated and have recovered from the injury, as indicated by defined cellular membranes and brush borders. (J) Nfatc1+/– mice have disorganized PTC. (A through J) Hematoxylin- and eosin-stained sections. Arrow, nuclear dropout; >, proliferating nuclei; +, tubular dilation; , disorganization of the PTC nephron segment. (K) Acute injury scores to quantify HgCl2-induced injury. Acute injury scores were significantly higher in the WT and Nfatc1+/– mice after HgCl2. ***P < 0.001 by two-way ANOVA with Bonferroni posttest versus day 0. AKI scoring: 0 = normal; 1 = <10%; 2 = 10 to 25%; 3 = 26 to 75%; 4 = >75% of PTC injured. Magnification. x400.
To assess the severity of renal injury, we used a quantitativeevaluation of PTC injury defined by the severity of tubulardilation, cellular necrosis, nuclear dropout, and the loss ofdefined brush border membranes in PTCs.10 As indicated by thehistology, Nfatc1+/– mice had significant increased AKIcompared with WT mice (Figure 1K) (P < 0.0001, two-way ANOVA).
Nfatc1+/– Mice Have Decreased Nfatc1 Transcription Nfatc1 expression is induced through autoregulation of the endogenouspromoter by Nfatc1 and NFATc2.11,12 Quantitative real-time PCR(qRT-PCR) showed NFATc1 expression was significantly increasedin WT mice throughout the time course compared with Nfatc1+/–mice (P < 0.0001, two-way ANOVA; Figure 2A). Nfatc1 expressionwas significantly increased in WT mice at day 3, correlatingwith the period of PTC regeneration, and in situ hybridizationconfirmed expression of Nfatc1 in cortical tubules (SupplementalFigure 1). In Nfatc1+/– mice, Nfatc1 expression was attenuatedand did not significantly increase throughout the time course.Expression of other NFATc transcription factors (Nfatc2, Nfatc3,and Nfatc4) was not altered throughout the HgCl2 time coursein WT or Nfatc1+/– mice (Supplemental Figure 2), suggestingthat other members of the NFATc family of transcription factorsdo not compensate for the decreased Nfatc1 expression as describedpreviously in other cell types.13,14
Figure 2.Nfatc1+/– mice have significant AKI and decreased expression of NFATc1 mRNA and protein compared with WT. (A) Relative NFATc1 expression in RNA isolated from the cortex shows blunted NFATc1 expression in Nfatc1+/– mice after HgCl2 injury. After treatment with HgCl2, the WT mice have a significant increase in NFATc1 expression at day 3. ***P < 0.001 by two-way ANOVA with Bonferroni posttest versus day 0. (B) Western blot for NFATc1 and β-actin protein from isolated proximal tubules. Each lane represents PTC lysate from different mice. Open and closed arrows indicate the dephosphorylated activated and phosphorylated cytosolic forms of NFATc1, respectively. (C) Comparison of dephosphorylated NFATc1 expression levels normalized to β-Actin in WT and Nfatc1+/– mice. ***P < 0.001 and *P < 0.05 by two-way ANOVA with Bonferroni posttest versus day 0.
NFATc1 Protein Is Upregulated in Proximal Tubules
Proximal tubule fractions were isolated from three individualWT and Nfatc1+/– mice at each time point. Protein extractedfrom the proximal tubule was evaluated by Western blot (Figure 2B).Dephosphorylated transcriptionally active NFATc1 protein wasquantified and normalized to β-actin. NFATc1 expressionincreased throughout the HgCl2 time course and was significantlyincreased in WT mice compared with Nfatc1+/– mice (P =0.0014, two-way ANOVA; Figure 2C).
NFATc1 Attenuation Results in Increased Interstitial Collagen
Interstitial fibrosis is one of the hallmarks of CsA-inducednephrotoxicity8; therefore, we questioned whether genetic attenuationof NFATc1 might produce a fibrotic response in the setting ofAKI. After HgCl2 treatment, we observed cortical interstitialchanges and observed more extensive collagen deposition in thekidneys of Nfatc1+/– mice compared with WT mice (Figure 3,A and B). We quantified interstitial collagen in the cortexusing a point-counting assay (Figure 3C).15Nfatc1+/–mice had significantly increased interstitial collagen depositionafter injury compared with WT mice (P < 0.01, two-way ANOVA).
Figure 3. After treatment with HgCl2, mice with attenuated NFATc1 expression have interstitial collagen deposits and disrupted proximal tubule segments. (A and B) Trichrome staining 10 d after HgCl2 injury in WT (A), Nfatc1+/– (B). (C) Quantitative point-counting of interstitial collagen deposits. Nfatc1+/– mice had significantly higher interstitial collagen 3, 5, and 10 d after HgCl2 injury compared with that seen at day 0 **P < 0.01 by two-way ANOVA with Bonferroni posttest versus day 0.
NFATc1 Attenuation Results in Increased PTC Apoptosis
CsA nephrotoxicity triggers the apoptotic pathway in mitochondria,resulting in CsA-induced toxicity, in vivo and in vitro.16 InT cells, NFATc1 upregulation attenuates CsA-induced apoptosisin several cell types11,17,18; therefore, we hypothesized thatNFATc1 protected PTCs from apoptosis associated with AKI. Wequantified terminal deoxynucleotidyl transferase–mediateddigoxigenin-deoxyuridine nick-end labeling (TUNEL)-positivePTCs and showed a significant increase in apoptotic PTC in Nfatc1+/–mice compared with WT mice (P = 0.0431, two-way ANOVA). Thisis consistent with our previous observation of increased injuryin Nfatc1+/– mice and suggests that moderate attenuationof NFATc1 results in increased susceptibility to apoptosis inPTCs.
We performed immunohistochemistry with proliferating cell nuclearantigen (PCNA) and quantified PTC stained nuclei. Both WT andNfatc1+/– mice demonstrated an increase in PTC proliferationat day 3; however, there was no difference in the number ofproliferating cells in Nfatc1+/– mice compared with WTcontrols (Figure 4B). HgCl2 nephrotoxicity was assessed by measuringcreatinine levels in serum collected throughout the time course.There was no difference in serum creatinine concentrations ofNfatc1+/– mice compared with WT mice (Figure 4C), suggestingthat there was no intrinsic renal pathology in Nfatc1+/–mice before HgCl2 administration. Thus, although tubular regenerationis delayed, PTCs that escape apoptosis and survive AKI are ultimatelyable to restore renal function.
Figure 4. Attenuation of NFATc1 causes increased apoptosis and decreased PTC proliferation after HgCl2 injury. (A) Apoptosis after HgCl2-induced AKI. One and 3 d after HgCl2 injury, there was a significant increase in the number of apoptotic PTCs in Nfatc1+/– mice. (B) Proliferation after HgCl2-induced AKI. (C) Serum creatinine concentrations are elevated in WT and Nfatc1+/– mice after HgCl2-induced AKI. ***P < 0.001, **P < 0.01, and *P < 0.05 by two-way ANOVA with Bonferroni posttest versus day 0.
Pharmacologic Attenuation with CsA Causes Severe Injury in Nfatc1+/– Mice
Because heterozygous Nfatc1 mice demonstrate only a moderatereduction in NFATc1 mRNA and protein, as discussed already,and because homozygous null mice die in utero and are unavailablefor postnatal studies, we sought to reduce NFATc1 expressionfurther. We treated WT mice daily with 10 mg/kg CsA before andafter HgCl2-induced AKI. Surprisingly, serum creatinine concentrationswere significantly increased in mice treated with 10 mg/kg CsAand HgCl2, and the severity of injury was associated with highmortality 5 d after HgCl2 treatment (Supplemental Figure 3);therefore, we reduced the dosage to 5 mg/kg CsA and repeatedthe HgCl2 time course in WT and Nfatc1+/– mice. WhereasCsA-treated WT mice survived the HgCl2 time course, CsA-treatedNfatc1+/– mice were removed from the study 3 d after HgCl2because the mice had severe dehydration, lost >20% body weight,developed impaired mobility, and had accelerated mortality (Figure 5A).
Figure 5. Genetic attenuation of NFATc1 combined with pharmacologic attenuation of NFATc proteins with CsA causes severe renal injury after HgCl2 injury. (A) Survival curve demonstrating high mortality in Nfatc1+/– mice treated with 5 mg/kg CsA. (B) Acute injury scores to quantify HgCl2-induced injury. Acute injury scores were significantly higher in CsA-treated Nfatc1+/– mice after HgCl2 (P = 0004 by two-way ANOVA). AKI scoring: 0 = normal; 1 = <10%; 2 = 10 to 25%; 3 = 26 to 75%; 4 = >75% of PTC injured. (C) After HgCl2 injury, there was a significant increase in the number of apoptotic PTCs in CsA-treated Nfatc1+/– mice (P = 0.0338 versus CsA-treated WT mice by two-way ANOVA). (D) Proliferation is significantly decreased in CsA-treated Nfatc1+/– mice after HgCl2 (P < 0.0001 versus CsA-treated WT mice by two-way ANOVA). (E). Serum creatinine concentrations. (B through E) *P < 0.05, **P < 0.01, and ***P < 0.001 by two-way ANOVA with Bonferroni posttest versus day 0.
We compared this unexpected heightened injury in CsA-treatedNfatc1+/– mice with CsA-treated WT mice. CsA-treated Nfatc1+/–mice had significantly increased AKI compared with CsA-treatedWT mice (P = 0.004, two-way ANOVA; Figure 5B). CsA-treated Nfatc1+/–mice had significantly increased numbers of apoptotic PTCs labeledby TUNEL staining compared with CsA-treated WT mice (P = 0.0338,two-way ANOVA; Figure 5C). CsA-treated Nfatc1+/– micealso demonstrated a significant decrease in proliferation comparedwith CsA-treated WT mice (P < 0.0001, two-way ANOVA; Figure 5D).Toxicity was clearly associated with HgCl2-induced AKI, becauseno significant changes in AKI, serum creatinine, apoptosis,or proliferation were observed in mice treated daily with eithervehicle or 5 mg/kg CsA alone. Thus, attenuation of NFATc1 usinga moderately low dosage of CsA resulted in increased and sustainedapoptosis and decreased proliferation of PTC, which resultedin impaired regeneration of the damaged PTC segment and ultimatelydeath. Interestingly, despite the high incidence of mortality,CsA-treated Nfatc1+/– mice did not have significantlyincreased serum creatinine concentrations compared with CsA-treatedWT mice (Figure 5E), suggesting that serum creatinine determinationsmay not provide a sensitive measurement of AKI severity.
Nfatc1-P2-LacZ Reporter Expression Documents Nfatc1 in a PTC Subpopulation
The transgenic line Nfatc1-P2-LacZ (Figure 6A) contains an intronicNfatc1 enhancer element that expresses β-galactosidase(β-Gal) in the endocardium of the developing mouse heart,recapitulating endogenous Nfatc1 expression.7 This P2 enhancerelement controls autoamplification of Nfatc1.7,12 We used Nfatc1-P2-LacZmice to delineate NFATc1 activation and observed expressionin smooth muscle cells of large renal arteries but no expressionin glomerular, interstitial, or tubular cell populations beforetoxin exposure (Figure 6B); however, LacZ expression was activated1 d after HgCl2 administration (Figure 6C). The number of cellsexpressing LacZ increased at day 3 (Figure 6D) and subsequentlydecreased at day 5 (Figure 6E). By day 10, the number of LacZ-positivePTCs was greatly reduced (data not shown), a trend analogousto the RT-PCR expression analysis of Nfatc1 described already(Figure 2B).
Figure 6. The NFATc1-P2-LacZ reporter7 is activated and expressed in a subset of PTCs that do not apoptose after HgCl2-induced AKI. The HgCl2 time course was repeated in NFATc1-P2-LacZ mice. (A) Schematic of the NFATc1-P2-LacZ transgenic construct. (B) Before treatment with HgCl2, the NFATc1-P2-LacZ reporter is not active in any tubule epithelial, glomerular, or interstitial cell populations. (C) After HgCl2 injury, the NFATc1-P2-LacZ reporter is expressed at day 1, as seen by the nuclear localized X-Gal stain. (D and E) The NFATc1-P2-LacZ expression expands on day 3 (D), and expression is decreased by day 5 (E). (F) Immunohistochemistry performed using anti–β-Gal antibody and proximal tubule–specific marker LTL (brown) reveals that the NFATc1-P2-LacZ reporter is expressed specifically in the proximal tubule segment of the nephron in a subset population of PTCs. Representative image from day 3. (G) Immunofluorescence performed with β-Gal (red) and cleaved caspase 328 antibodies show 3 d after injury that the NFATc1-P2-LacZ PTCs are not apoptotic. (H) Immunofluorescence performed with β-Gal and cleaved caspase 3 antibodies before HgCl2 injury. (B through E) X-Gal staining with nuclear fast red counterstain. (F) Hematoxylin stained nuclei. (G and H) DAPI-stained nuclei (blue). Magnifications: x100 in B through E; x400 in F; x200 in G and H.
Activation of Nfatc1-P2-LacZ was specific to a subset of PTCsas seen by immunohistochemistry with Lotus tetragonolobus lectin(LTL; Figure 6F). To confirm that the Nfatc1-P2 enhancer, whichcontains the minimal HSP promoter element, is responsible foractivation of the LacZ reporter in PTCs, we transfected LLC-PK1cells with a Nfatc1-P2-HSP-LacZ plasmid and demonstrated a three-foldincrease in X-Gal staining after treatment with HgCl2 comparedwith LLC-PK1 cells transfected with HSP-LacZ plasmid (data notshown). To characterize this PTC subpopulation further, we performedimmunofluorescence on adjacent sections with β-Gal andcleaved caspase 3 antibodies. Nfatc1-P2-LacZ expression didnot co-localize with apoptotic PTCs (Figure 6G). These datasuggest that the P2 locus, which reports Nfatc1 autoamplification,is activated in and identifies a population of apoptosis-resistantPTCs.
NFATc1-P2-Cre Reporter Identifies a Progenitor Subpopulation of PTCs
We questioned whether apoptosis-resistant PTCs marked by theNfatc1-P2 enhancer domain contributed to PTC regeneration. Nfatc1-P2-Cremice, a transgenic line that utilizes the Nfatc1-P2 enhancerelement to drive Cre recombinase expression (Figure 7A), werecrossed to R26R reporter mice.19 This lineage analysis approachallowed us to identify genetically all PTCs that activated Nfatc1-P2and their subsequent progeny. We repeated the HgCl2 time coursewith Nfatc1-P2-Cre mice. Before treatment with HgCl2, LacZ wasnot expressed in tubular, glomerular, or interstitial cells,indicating that the P2 promoter element was not activated duringrenal development (Figure 7B); however, the Cre reporter wasactivated and cytoplasmic β-Gal was observed by day 5 (Figure 7C).Contrasting previous experiments with Nfatc1-P2-LacZ in whichLacZ-positive cells decreased after day 5 indicating decreasedNfatc1 expression, X-Gal–stained PTCs in Nfatc1-P2-Cremice increased by day 10 (Figure 7D). The increased number ofNfatc1-P2-Cre–labeled PTCs thus represents the clonalexpansion of the initial Nfatc1-expressing PTCs. Staining ofserial sections with X-Gal and LTL confirmed Nfatc1-P2-Cre expressionin PTCs (Figure 7, E and F).
Figure 7. Lineage analysis using the NFATc1-P2-Cre reporter marks a progenitor subpopulation of PTCs that proliferate after AKI. (A) Schematic of the NFATc1-P2-Cre transgenic construct. (B) Before HgCl2 treatment, NFATc1-P2-Cre is not expressed in any tubule epithelial, glomerular, or interstitial cell population, confirming that the reporter construct was not activated previously during any stage of development. (C and D) After HgCl2 injury, the NFATc1-P2-Cre reporter is activated and expressed in PTCs cells at day 5 (C) and the number of X-Gal positive cells increases by day 10 (D). (E through F) Staining of serial sections with X-Gal (E) and LTL (F). Red arrows reference proximal tubule segments on adjacent sections. (G) Co-staining 5 d after HgCl2 injury shows co-localization of the NFATc1-P2-Cre (blue, X-Gal staining of R26R reporter) and BrdU-labeled (brown, arrows) PTC identifying a subset population of PTCs that proliferate after injury. (H) Quantification of PTCs cells after BrdU pulse/chase that are positive for X-Gal or BrdU or both reveal that at day 3, there were an equal number of X-Gal+/BrdU+ and X-Gal–/BrdU+ populations each composing approximately 5% of the PTCs counted. At day 10, 27% of the PTCs are NFATc1-P2-Cre derivatives. (I) After treatment with HgCl2 and daily injections of BrdU, there is a three-fold increase in the number of X-Gal+/BrdU+ PTCs compared with X-Gal–/BrdU+ PTCs. (B through E) Nuclear fast red counterstain. Magnifications: x100 in B through D; x400 in E.
To quantify regeneration, we stained sections with X-Gal andperformed immunohistochemistry with the bromodeoxyuridine (BrdU)antibody and quantified X-Gal–BrdU–, X-Gal+BrdU–,X-Gal–BrdU+, and X-Gal+BrdU+ PTCs (Figure 7, G and H).Five days after HgCl2, the number of X-Gal+BrdU+ PTCs was 2.5-foldgreater than X-Gal–BrdU+ PTCs. Ten days after HgCl2, >25%of PTCs were derived from Nfatc1-P2-Cre PTCs. The number ofX-Gal–BrdU+ PTCs was not altered throughout the time course.To determine the relative contribution of Nfatc1 progeny tothe total regenerative potential of PTCs, we performed dailyBrdU injections throughout the period of HgCl2–inducedinjury and repair. After 10 consecutive days of BrdU injections(Figure 7I), 30.6% of PTCs labeled X-Gal+BrdU+ compared with9.6% of PTCs labeled X-Gal–BrdU+. These data indicatethat Nfatc1-P2 enhancer–labeled PTCs serve as a progenitorpopulation to repair the damaged proximal tubule segment becauseat least 75% of the PTCs that are generated by proliferationafter injury are progeny of the initial Nfatc1-expressing PTCs.Furthermore, the close approximation of Nfatc1-P2-Cre–positivecells and their X-Gal+BrdU+ progeny suggests that the damagedPTC segment is regenerated by clonal expansion of an apoptosis-resistantsubpopulation.
In this study, we report four major findings. First, NFATc1is upregulated during regeneration after AKI, and modest attenuationof NFATc1 expression through genetic deletion results in increasedapoptosis, delayed regeneration, and increased fibrosis in responseto injury. Second, moderate attenuation of NFATc expressionby CsA in Nfatc1+/– mice results in an increased susceptibilityto renal injury and death after AKI. Third, using Nfatc1-P2-LacZmice, we identified a resident apoptosis-resistant PTC subpopulationcharacterized by autoamplification of NFATc1 after HgCl2 injury.Fourth, using a novel transgenic reporter mouse line, Nfatc1-P2-Cre,we identified a resident progenitor PTC subpopulation that regeneratesthe damaged proximal tubule segment.
Our RT-PCR analysis of renal cortex mRNA shows that Nfatc1 mRNAis upregulated in WT mice during the period of regenerationafter HgCl2 treatment. In addition, protein isolated from theproximal tubule segment shows that transcriptionally activeNFATc1 protein was increased during regeneration. Interestingly,only a modest decrease in NFATc1 mRNA and protein produced byheterozygous gene deletion resulted in significantly increasedPTC apoptosis and impaired regeneration after HgCl2-mediatedAKI. To our knowledge, this is the first evidence of a roleof NFATc1 in homeostasis of the mature kidney. The criticaldownstream targets of NFATc1 activation in the PTC essentialto regeneration are the focus of current investigations.
Severe renal injury resulting from the combined genetic andpharmacologic attenuation of NFATc activity with 5 mg/kg perd CsA in Nfatc1+/– mice supports a role for NFATc1 activationin PTC regeneration. Calcineurin is involved in dephosphorylationof NFATc and other cytoplasmic proteins, and the effects ofCsA are therefore pleiotropic; however, given our observationthat NFATc1 is the only NFAT family member that seems to beaffected by HgCl2-induced AKI, it is reasonable to speculatethat the primary effect of CsA in these studies is on the modulationof NFATc1 activity. Because Nfatc1–/– mice die inutero, definitive postnatal studies will require developmentof tissue-specific deletion models currently in progress.
Perhaps most surprising was the dramatic renal injury that occurredafter HgCl2 administration in mice receiving 5 and 10 mg/kgCsA. Previously, administration of higher dosages (100 mg/kgCsA) did not induce functional or structural abnormalities20;however, CsA-treated Nfatc1+/– mice and 10-mg/kg CsA–treatedWT mice caused increased PTC apoptosis, decreased proliferation,renal failure, and death after HgCl2-induced AKI. CsA inhibitionof NFATc1 activity has been previously associated with increasedapoptosis and reduced cell proliferation in monocytes,17 retinoblastoma,18and T cells.11 Furthermore, ectopic expression of a constitutivelyactive isoform of NFATc1 protected cells from apoptosis andpromoted proliferation,21 and expression of the autoregulatedNFATc1/A isoform attenuated induced cell death in T cells.11Thus, inhibition of NFATc1, either genetically or pharmacologically,is likely to result in activation of proapoptotic genes afterexposure to a renal toxin, and our studies suggest that CsA,even at low dosages not associated with chronic changes, mayresult in an increased vulnerability to renal insult that hasnot been previously appreciated.
This study also provides additional insights into the mechanismsof PTC regeneration and repair after toxic insult. The injuredkidney must undergo regeneration and proliferation to restorerenal function. It has been postulated that these cells arisefrom adjacent less injured cells, a renal stem cell population,or an external stem cell population circulating in the bloodstream.22–24 Our use of Nfatc1-P2-LacZ and Nfatc1-P2-Cremice suggests a subpopulation of resident PTCs, identified usingthe Nfatc1-P2 enhancer, act as progenitor cells. Nfatc1-P2-LacZand Nfatc1-P2-Cre transgenic lines were not active during development.We speculate that NFATc1 functions as an active transcriptionfactor during recovery from injury as the P2 enhancer elementreports autoamplification of NFATc1. This population subsequentlyundergoes proliferation to reconstitute the proximal tubuleand provides the predominant source of proliferating cells requiredfor PTC repair.
In conclusion, we propose a model (Figure 8) for PTC regeneration.When exposed to an injury stimulus, it is assumed that PTCsreceive the same injury. Some of the PTCs undergo apoptosis;however, a resident progenitor population, genetically markedusing the Nfatc1-P2 autoamplification/enhancer domain, is acutelyresistant to apoptosis and subsequently participates in regenerationof the damaged proximal tubule. Continued investigation is warrantedto delineate further the molecular basis for PTC heterogeneityand to develop therapeutic interventions that will attenuatePTC damage or enhance the regenerative activity of the residentprogenitor population after injury.
Mice
Homozygous null Nfatc1 mice created by excising the Rel DNA-bindingdomain are maintained as a heterozygous colony on the Balb/cbackground and genotyped as described previously.5 The Nfatc1-P2-LacZtransgenic mice containing a minimal HSP promoter and the enhancerdomain of Nfatc1 were engineered and genotyped 5'-GCAGCAGGCAGGGTCACAGAGA-3'and 5'-ACCCCAGGCTGCAAGGAGGATT-3'.7Nfatc1-P2-Cre transgenicmice contain the minimal Nfatc1-P1 promoter characterized byZhou et al.,12 a Cre recombinase cassette, and the Nfatc1 enhancerdomain. Stable lines of Nfatc1-P2-Cre mice were genotyped usingCre-5'-CCTGGAAAATGCTTCTGTCCG-3' and 5'-CAGGGTGTTATAAGCAATCCC-3'.Nfatc1-P2-LacZ and Nfatc1-P2-Cre mice are maintained on theB6D2F1 background. The Institutional Animal Care and Use Committeeof Vanderbilt University approved all animal studies.
Experimental Protocol
All experiments were performed on 6- to 8-wk-old female mice.A single subcutaneous dose of 8.14 mg/kg HgCl2 (Fisher Scientific,Pittsburgh, PA) in normal saline causes significant AKI in miceon the Balb/c background.25 HgCl2 accumulates in PTCs, preferentiallyin the S2 and S3 segments.9 Balb/c, Nfatc1+/–, and Nfatc1-P2-LacZmice were administered HgCl2 and killed 24 h (1 d), 72 h (3d), 5 d, or 10 d later and compared with mice not administeredHgCl2, 0 d. Balb/c mice were analyzed at 0, 1, 3, 5, and 10d (n = 8, 8, 10, 8, and 5, respectively). Nfatc1+/– micewere analyzed at 0, 1, 3, 5, and 10 d (n = 8, 7, 8, 11, and5, respectively). To block transcriptional activity of all NFATcproteins in Balb/c and Nfatc1+/– mice, we administeredCsA daily, beginning 7 d before treatment with HgCl2, untileach end point. CsA prepared in polyoxyethylated castor oiland absolute alcohol (Bedford Labs, Bedford, OH) was dilutedin normal saline and administered in the peritoneal cavity ata daily dose of 5 mg/kg body wt (WT: n = 5 mice per dose pertime point; Nfatc1+/–: n = 3 mice per dose per time point).A vehicle group was treated with polyoxyethylated castor oiland absolute alcohol daily prepared in normal saline (n = 3per end point; NFATc1-P2-LacZ: n = 3 per end point). BrdU (CellProliferation Labeling Reagent; GE Healthcare, Piscataway, NJ)was administered to Nfatc1-P2-Cre//R26R mice followed by a 2-h(day 3 and day 5) or 5-d (day 10) chase (n = 3 for each timepoint, or for 10 consecutive days after HgCl2, n = 5). We didnot observe any gross abnormalities in the intestines, stomach,heart, lungs, spleen, pancreas, or liver in any group studied.
Blood Serum Creatinine Concentration
Blood drawn from the orbital vein at each end point was analyzedto determine the serum creatinine concentration as describedby Dunn et al.26
Histology
Dissected kidneys were bisected and fixed in PBS containing4% paraformaldehyde overnight at 4°C, dehydrated in an ethanoland xylene gradient, and embedded in paraffin. Five-micrometersections were cut on a microtome. The AKI score for tubularinjury was assessed in hematoxylin- and eosin-stained sectionsusing a semiquantitative scale in which 10 high-power fields(x200) were scored for the percentage of cortical tubules showingepithelial necrosis and assigned a score as described previously10:0 = normal; 1 = <10%; 2 = 10 to 25%; 3 = 26 to 75%; 4 = >75%.Weigert's Iron Hematoxylin Solution (Sigma, St. Louis, MO) andthe Accustain Trichrome Stain (Sigma) were used to stain andscore interstitial collagen deposits.
β-Gal Detection
Kidneys were bisected; fixed in PBS containing 0.2% glutaraldehyde,5 mM EGTA, and 100 mM MgCl2 for 4 h at room temperature witha solution change after 2 h; and cryoprotected in PBS containing15 and 30% sucrose before embedding in OCT.27 Ten-micrometersections were cut; dried at room temperature for 30 min; washedin a PBS detergent containing 2 mM MgCl2, 0.01% Na-deoxycholate,and 0.02% NP40; stained in X-Gal solution (pH 7.50); and postfixedin 4% paraformaldehyde, counterstained with eosin or nuclearfast red, and mounted in Permount (Fisher).
RNA Extraction
A fragment of the outer cortex was snap-frozen and stored at–80°C. RNA was isolated from 40 mg of tissue usingthe Versagene Total RNA Purification Kit, and DNA contaminationwas removed using the Versagene DNase Treatment Kit. RNA qualityand concentration were determined using the RNA 6000 Nano LabChipKit (Agilent, Santa Clara, CA). Samples within each experimentalgroup were pooled, and single-stranded cDNA was prepared from10 µg of total RNA using oligo (dT)16 primer and TranscriptorReverse Transcriptase (Roche, Indianapolis, IN).
qRT-PCR
qRT-PCR was performed in a Light Cycler (Roche). All experimentswere done using the Light Cycler DNA Master SYBR Green I Kit,0.5µM of each primer, 3 to 5 mM MgCl2, and 200 ng of cDNA.DNA primer sets were designed using different exons, when possible,to ensure that the product was from mRNA and not genomic DNA.Primers were as follows: For glyceraldehyde-3-phosphate dehydrogenase,5'-CACTGGCATGGCCTTCCGTG-3' and 5'-AGGAAATGAGCTTGACAAAG-3"; forNfatc1, 5'-GGTGGCCTCGAACCCTATC-3' and 5'-TCA GTCTTTGCTTCCATCTCCC-3'.The specificity of the amplified product was evaluated usingthe melting curve analysis and a no template control reactionincluded in each run. The housekeeping gene glyceraldehyde-3-phosphatedehydrogenase was used to normalize the data. The relative changein gene expression was determined using the critical threshold(Ct) and the equation Fold Induction = 2^(–Ct) using theWT 0-d sample as the calibrator sample, and a two-fold increaseor decrease in expression was considered significant.28
Isolation of Proximal Tubule
PTCs were isolated using modification of the methods of Vinayet al.29 and as modified by Xu et al.30 Briefly, cortices fromHgCl2-treated WT or Nfatc1+/– mice were collected andminced in Krebs-Hensleit saline (KHS) buffer containing 115mM NaCl, 24 mM NaHCO3, 10 mM HEPES, 5 mM glucose, 5 mM KCl,2 mM NaH2PO4, 1.5 mM MgSO4, and 1 mM alanine. This solutionwas then enriched with 0.15% (wt/vol) collagenase type I, 0.5%(wt/vol) BSA, and 0.01% soybean trypsin inhibitor and incubatedat 37°C for 1 h to digest the cortices. The suspension wasstrained through a 100-µm sieve, washed, and centrifugedat 600 rpm three times. The pellets were combined with 47% Percoll(Amersham, Piscataway, NJ) solution mixed with 2x KHS and centrifugedat 16,300 rpm for 30 min at 4°C. The lowest band enrichedwith proximal tubule segments was washed with KHS buffer threetimes and used for protein isolation. The purity of the isolatedproximal tubule fractions was >95% as determined by microscopicanalysis.
Protein Isolation and Western Blotting
Proximal tubule segments were pelleted and lysed in 25 mM HEPES(pH 7.4), 150 mM NaCl, 5 mM EDTA, 1% Triton X-100, 10% glycerol,50 mM NaF (tyrosine phosphatase inhibitor), 2 mM NaVO4 (phosphataseinhibitor), and 1% protease cocktail inhibitor (Sigma). Concentrationwas determined by Bradford assay, and 50 µg of proteinwas analyzed per lane. NFATc1 was detected by Western blot analysis(mouse monoclonal, 7A6; BD Pharmingen, San Jose, CA). Proteinloading was confirmed by Western blot analysis for β-actin(mouse monoclonal; Sigma-Aldrich). To quantify the concentrationof active NFATc1 protein, we quantified the dephosphorylatedprotein band by densitometry (EpiChem; UVP Bioimaging System,Upland, CA) and the normalized results as a ratio of NFATc1to β-actin for each preparation as compared with WT samplesat day 0.
Antibody Staining
PCNA and TUNEL stainings were performed on paraffin sectionsusing the PCNA Staining Kit (Zymed, South San Francisco, CA)and ApopTag Apoptosis Detection Kit (Serologicals, Billerica,MA), respectively. Ten high-power fields from each mouse werescored blindly for the number of PTCs that were TUNEL or PCNApositive. Immunofluorescence with the anti–β-Galantibody (Dr. Lim, University of Michigan, Ann Arbor, MI) andanti–cleaved caspase 3 (Cell Signaling, Danvers, MA) weredetected on adjacent 10-µm cryosections with TRITC- andFITC-labeled goat anti-rabbit antibodies (Jackson Immunoresearch,West Grove, PA), respectively, and nuclei were stained usingHoechst Dye 33342 (Molecular Probes, Eugene, OR) and mountedin Vectashield Mounting Media (Vector, Laboratories, Burlingame,CA). Immunohistochemistry with X-Gal and biotinylated LTL (Vector)were detected on adjacent 10-µm cryosections. BrdU-labelednuclei were identified using the anti-BrdU antibody (Abcam,Cambridge, MA) and biotinylated LTL on X-Gal–stained slidesto score the NFATc1-P2-Cre expression pattern.
Statistical Analysis
All data are presented as means ± SEM. All scoring wasperformed blinded. Prism 4 software was used for all statisticalmeasurements. Two-way ANOVA was used to assess the relationshipbetween genotype, CsA treatment and acute kidney score, serumcreatinine, apoptosis, proliferation, collagen, and NFATc1 mRNAlevel at each time point. Differences between groups were assessedwith two-way ANOVA and posttest using Bonferroni correctionto compare days 1, 3, 5, and 10 with day 0 for each genotypeand to reduce type I error. A result was considered to be significantat P < 0.05.
This work was supported by National Institutes of Health PediatricNephrology Center grant DK44757 and a Center in Molecular ToxicologyPilot Grant ES000267.
We are thankful to Stephen Dunn (Thomas Jefferson University,Philadelphia, PA), Satish Rao, and Zachary Sendar (Universityof California, San Diego, CA) for performing the HPLC serumcreatinine analysis. The anti–β-Gal antibody wasgraciously provided by Dr. Kim Lim's laboratory (Universityof Michigan). We are also thankful to Dr. David Frank for criticalreview of the manuscript.
Footnotes
Published online ahead of print. Publication date availableat www.jasn.org.
Supplemental information for this article is available onlineat http://www.jasn.org/.
Basile DP: Novel approaches in the investigation of acute kidney injury.
J Am Soc Nephrol 18
: 7
–9, 2007[Free Full Text]
Tumlin JA, Someren JT, Swanson CE, Lea JP: Expression of calcineurin activity and alpha-subunit isoforms in specific segments of the rat nephron.
Am J Physiol 269
: F558
–F563, 1995[Medline]
Gooch JL, Barnes JL, Garcia S, Abboud HE: Calcineurin is activated in diabetes and is required for glomerular hypertrophy and ECM accumulation.
Am J Physiol Renal Physiol 284
: F144
–F154, 2003[Abstract/Free Full Text]
Puri S, Magenheimer BS, Maser RL, Ryan EM, Zien CA, Walker DD, Wallace DP, Hempson SJ, Calvet JP: Polycystin-1 activates the calcineurin/NFAT signaling pathway.
J Biol Chem 279
: 55455
–55464, 2004[Abstract/Free Full Text]
Ranger AM, Grusby MJ, Hodge MR, Gravallese EM, de la Brousse FC, Hoey T, Mickanin C, Baldwin HS, Glimcher LH: The transcription factor NF-ATc is essential for cardiac valve formation.
Nature 392
: 186
–190, 1998[CrossRef][Medline]
de la Pompa JL, Timmerman LA, Takimoto H, Yoshida H, Elia AJ, Samper E, Potter J, Wakeham A, Marengere L, Langille BL, Crabtree GR, Mak TW: Role of NF-ATc transcription factor in morphogenesis of cardiac valves and septum.
Nature 392
: 182
–186, 1998[CrossRef][Medline]
Zhou B, Wu B, Tompkins KL, Boyer KL, Grindley JC, Baldwin HS: Characterization of Nfatc1 regulation identifies an enhancer required for gene expression that is specific to pro-valve endocardial cells in the developing heart.
Development 132
: 1137
–1146, 2005[Abstract/Free Full Text]
Myers BD, Sibley R, Newton L, Tomlanovich SJ, Boshkos C, Stinson E, Luetscher JA, Whitney DJ, Krasny D, Coplon NS, et al.: The long-term course of cyclosporine-associated chronic nephropathy.
Kidney Int 33
: 590
–600, 1988[Medline]
Hultman P, Enestrom S: Localization of mercury in the kidney during experimental acute tubular necrosis studied by the cytochemical Silver Amplification method.
Br J Exp Pathol 67
: 493
–503, 1986[Medline]
Ramesh G, Reeves WB: TNF-alpha mediates chemokine and cytokine expression and renal injury in cisplatin nephrotoxicity.
J Clin Invest 110
: 835
–842, 2002[CrossRef][Medline]
Chuvpilo S, Jankevics E, Tyrsin D, Akimzhanov A, Moroz D, Jha MK, Schulze-Luehrmann J, Santner-Nanan B, Feoktistova E, König T, Avots A, Schmitt E, Berberich-Siebelt F, Schimpl A, Serfling E: Autoregulation of NFATc1/A expression facilitates effector T cells to escape from rapid apoptosis.
Immunity 16
: 881
–895, 2002[CrossRef][Medline]
Zhou B, Cron RQ, Wu B, Genin A, Wang Z, Liu S, Robson P, Baldwin HS: Regulation of the murine NFATc1 gene by NFATc2.
J Biol Chem 277
: 10704
–10711, 2002[Abstract/Free Full Text]
Diehl S, Krahl T, Rinaldi L, Norton R, Irvin CG, Rincón M: Inhibition of NFAT specifically in T cells prevents allergic pulmonary inflammation.
J Immunol 172
: 3597
–3603, 2004[Abstract/Free Full Text]
Wilkins BJ, De Windt LJ, Bueno OF, Braz JC, Glascock BJ, Kimball TF, Molkentin JD: Targeted disruption of NFATc3, but not NFATc4, reveals an intrinsic defect in calcineurin-mediated cardiac hypertrophic growth.
Mol Cell Biol 22
: 7603
–7613, 2002[Abstract/Free Full Text]
Ma J, Nishimura H, Fogo A, Kon V, Inagami T, Ichikawa I: Accelerated fibrosis and collagen deposition develop in the renal interstitium of angiotensin type 2 receptor null mutant mice during ureteral obstruction.
Kidney Int 53
: 937
–944, 1998[Medline]
Justo P, Lorz C, Sanz A, Egido J, Ortiz A: Intracellular mechanisms of cyclosporin A-induced tubular cell apoptosis.
J Am Soc Nephrol 14
: 3072
–3080, 2003[Abstract/Free Full Text]
Roy MK, Takenaka M, Kobori M, Nakahara K, Isobe S, Tsushida T: Apoptosis, necrosis and cell proliferation-inhibition by cyclosporine A in U937 cells (a human monocytic cell line).
Pharmacol Res 53
: 293
–302, 2006[CrossRef][Medline]
Eckstein LA, Van Quill KR, Bui SK, Uusitalo MS, O'Brien JM: Cyclosporin a inhibits calcineurin/nuclear factor of activated T-cells signaling and induces apoptosis in retinoblastoma cells.
Invest Ophthalmol Vis Sci 46
: 782
–790, 2005[Abstract/Free Full Text]
Soriano P: Generalized lacZ expression with the ROSA26 Cre reporter strain.
Nat Genet 21
: 70
–71, 1999[CrossRef][Medline]
Andoh TF, Lam TT, Lindsley J, Alpers CE, Bennett WM: Enhancement of chronic cyclosporine nephrotoxicity by sodium depletion in an experimental mouse model.
Nephrology 3
: 471
–478, 1997[CrossRef]
Neal JW, Clipstone NA: A constitutively active NFATc1 mutant induces a transformed phenotype in 3T3–L1 fibroblasts.
J Biol Chem 278
: 17246
–17254, 2003[Abstract/Free Full Text]
Morigi M, Benigni A, Remuzzi G, Imberti B: The regenerative potential of stem cells in acute renal failure.
Cell Transplant 15
[Suppl 1]: S111
–S117, 2006[CrossRef][Medline]
Gupta S, Verfaillie C, Chmielewski D, Kren S, Eidman K, Connaire J, Heremans Y, Lund T, Blackstad M, Jiang Y, Luttun A, Rosenberg ME: Isolation and characterization of kidney-derived stem cells.
J Am Soc Nephrol 17
: 3028
–3040, 2006[Abstract/Free Full Text]
Sagrinati C, Netti GS, Mazzinghi B, Lazzeri E, Liotta F, Frosali F, Ronconi E, Meini C, Gacci M, Squecco R, Carini M, Gesualdo L, Francini F, Maggi E, Annunziato F, Lasagni L, Serio M, Romagnani S, Romagnani P: Isolation and characterization of multipotent progenitor cells from the Bowman's capsule of adult human kidneys.
J Am Soc Nephrol 17
: 2443
–2456, 2006[Abstract/Free Full Text]
Miyaji T, Hu X, Star RA: Alpha-melanocyte-simulating hormone and interleukin-10 do not protect the kidney against mercuric chloride-induced injury.
Am J Physiol Renal Physiol 282
: F795
–F801, 2002[Abstract/Free Full Text]
Dunn SR, Qi Z, Bottinger EP, Breyer MD, Sharma K: Utility of endogenous creatinine clearance as a measure of renal function in mice.
Kidney Int 65
: 1959
–1967, 2004[CrossRef][Medline]
Ma W, Rogers K, Zbar B, Schmidt L: Effects of different fixatives on beta-galactosidase activity.
J Histochem Cytochem 50
: 1421
–1444, 2002[Abstract/Free Full Text]
Livak KJ, Schmittgen TD: Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) method.
Methods 25
: 402
–408, 2001[CrossRef][Medline]
Vinay P, Gougoux A, Lemieux G: Isolation of a pure suspension of rat proximal tubules.
Am J Physiol 241
: F403
–F411, 1981[Medline]
Xu J, Yao B, Fan X, Langworthy MM, Zhang MZ, Harris RC: Characterization of a putative intrarenal serotonergic system.
Am J Physiol Renal Physiol 293
: F1468
–F1475, 2007[Abstract/Free Full Text]
Top
Abstract
Introduction
and Aβ have been identified in the proximal tubule,2 and NFATc (NFATc1 through 4) proteins are expressed predominantly in cortical tubules,3,4 correlating with the nephrotoxicity associated with cyclosporin A (CsA). Thus, NFATc may play a role in proximal tubular cell (PTC) injury and repair, which has not been previously appreciated. 
, disorganization of the PTC nephron segment. (K) Acute injury scores to quantify HgCl2-induced injury. Acute injury scores were significantly higher in the WT and Nfatc1+/– mice after HgCl2. ***P < 0.001 by two-way ANOVA with Bonferroni posttest versus day 0. AKI scoring: 0 = normal; 1 = <10%; 2 = 10 to 25%; 3 = 26 to 75%; 4 = >75% of PTC injured. Magnification. x400.
