defining the acute kidney injury and repair transcriptome · pdf filedefining the acute kidney...

Defining the Acute Kidney Injury and Repair Transcriptome

Sanjeev Kumar, MD, MRCP, PhD, Jing Liu, PhD, and Andrew P. McMahon, PhD, FRS

Summary: The mammalian kidney has an intrinsic ability to repair after significant injury. However, this

0270-9295/ -& 2014 Elsevhttp://dx.doi.o

Financial supfor DiabeteM.), and bKidney Rese

Conflict of in

Department oand Edythe(CIRM) CeResearch, TSouthern C

Address repriBiology andCenter forKeck Schoo1425 San Pmed.usc.edu

404

process is inefficient: patients are at high risk for the loss of kidney function in later life. No therapy exists totreat established acute kidney injury (AKI) per se: strategies to promote endogenous repair processes andretard associated fibrosis are a high priority. Whole-organ gene expression profiling has been used to identifyrepair responses initiated with AKI, and factors that may promote the transition from AKI to chronic kidneydisease. Transcriptional profiling has shown molecular markers and potential regulatory pathways of renalrepair. Activation of a few key developmental pathways has been reported during repair. Whether these arecomparable networks with similar target genes with those in earlier nephrogenesis remains unclear. AlteredmicroRNA profiles, persistent tubular injury responses, and distinct late inflammatory responses highlightcontinuing kidney pathology. Additional insights into injury and repair processes will be gained by study of therepair transcriptome and cell-specific translatome using high-resolution technologies such as RNA sequenc-ing and translational profiling tailored to specific cellular compartments within the kidney. An enhancedunderstanding holds promise for both the identification of novel therapeutic targets and biomarker-basedevaluation of the damage-repair process.Semin Nephrol 34:404-417 C 2014 Elsevier Inc. All rights reserved.Keywords: Acute kidney injury, repair, transcriptome, TRAP, miRNA, development, cancer

The clinical syndrome of acute kidney injury(AKI) is characterized by an abrupt (within48 h) decrease in kidney function, frequently

caused by ischemia reperfusion injury (IRI), sepsis, ornephrotoxic insult.1–3 Despite advances in medicalcare, patients with AKI continue to have high morbid-ity and mortality; in-hospital mortality rates in crit-ically ill patients with AKI approach 50% to 70%.3,4

Furthermore, survivors also have a strikingly higherrisk of developing chronic kidney disease (pooledadjusted hazard ratio, 8.8; 95% confidence interval,3.1-25.5), and end-stage renal disease (pooled adjustedhazard ratio, 3.1; 95% confidence interval, 1.9-5.0)compared with non-AKI patient groups.5

The histologic features of human ischemic AKIinclude loss of the brush border typical of the proximaltubular epithelium, sloughing of tubular epithelial cells

see front matterier Inc. All rights reserved.rg/10.1016/j.semnephrol.2014.06.007

port: Supported by grants from the National Institutes and Digestive and Kidney Diseases and CIRM (A.P.y the John McKay Fellowship from the Universityarch Organization (S.K.).

terest statement: none.

f Stem Cell Biology and Regenerative Medicine, EliBroad-California Institute of Regenerative Medicinenter for Regenerative Medicine and Stem Cellhe Keck School of Medicine of the University ofalifornia, Los Angeles, CA.

nt requests to A. McMahon, Department of Stem CellRegenerative Medicine, Eli and Edythe Broad-CIRM

Regenerative Medicine and Stem Cell Research, Thel of Medicine of the University of Southern California,ablo St, Los Angeles, CA 90089. E-mail: amcmahon@

into the lumen resulting in focal loss of tubular epithelialcells, infiltration of inflammatory cells, and the appear-ance of Tamm-Horsfall protein-rich casts in the urine.6

After AKI, a repair process restores renal tubularepithelium and kidney function. The cellular mechanismsof repair have been scrutinized intensively using mousegenetic approaches. Agreement is increasing that surviv-ing cells within the renal tubular epithelium repair tubulardamage in the mouse, and likely the human kidney (seearticle by Marcus Moeller in this issue). Whether repair isa general capacity shared by surviving cells, or a morespecific function ascribed to a small subset of identifiableepithelial cells, has engendered considerable debate (seearticle by Marcus Moeller in this issue). It is clear that thereparative process is not as efficient or effective asdesired: fibrosis is evident despite the reacquisition ofbiochemical parameters such as plasma creatinineremoval, and progression to chronic kidney disease is afrequent long-term outcome.5

Fibrosis is associated with injury-invoked appearanceof α-smooth muscle actin–positive myofibroblasts.In fibrosis, Yang et al7 suggested G2/M-arrested pro-ximal tubular cells activate c-jun NH2-terminal kinasesignaling, initiating production of profibrotic cyto-kines. In fibroblasts, hypermethylation of RAS proteinactivator like 1, an inhibitor of the Ras oncoprotein,leads to prolonged fibroblast activation and fibrogen-esis.8 Once triggered, myofibroblasts synthesize adistinct collagen I–rich extracellular matrix that maypromote further fibrosis.

Initial suggestions that most fibrotic cells arise froman epithelial-to-mesenchymal conversion of renaltubule cells have been challenged; a revised view ofan extratubular origin for myofibroblasts is supportedby several fate-mapping studies. One view holds that

Seminars in Nephrology, Vol 34, No 4, July 2014, pp 404–417

dx.doi.org/10.1016/j.semnephrol.2014.06.007



mailto:amcmahon@&QJ;med.usc.edu

mailto:amcmahon@&QJ;med.usc.edu

Acute Kidney Injury Transcriptome 405

perivascular fibroblasts (pericytes) are the chief cul-prit,9 whereas another associates fibrosis resident non-pericyte intertubular fibroblasts and bone marrow–derived fibroblasts.10 The origins of injury-associatedmyofibroblasts are discussed in article by BenjaminHumphreys.

Harnessing and enhancing the kidney’s intrinsicmechanisms of repair, and developing approaches tosuppress and reverse renal fibrosis, are major goals ofrenal regenerative medicine. These strategies arefounded, and dependent, on our detailed knowledgeof the molecular and cellular events at play. Newapproaches to interrogate underlying mechanisms haveenhanced resolution at the molecular level by enablingsystematic, relatively unbiased, quantitative measure-ment of transcriptional and translational events. Fur-ther, the move from whole-organ analysis to abreakdown of responses in specific cellular compart-ments is increasing cellular resolution. These advanceswill facilitate the identification of new targets augment-ing renal repair processes and suppress renal scarring.

Here, we provide a brief overview of the cellularresponses initiated by AKI, with a particular focus onthe repair processes after ischemic AKI, review studiesthat have performed whole-kidney or cell-specificgene/transcript expression analysis temporally in thesetting of murine and human AKI, and discuss the roleof next-generation RNA-sequencing (RNA-seq) andtranslating ribosome affinity purification (TRAP)profiling in transcriptional and translational analyses,respectively, of the renal repair process.

BRIEF OVERVIEW OF CELLULAR RESPONSES AFTERISCHEMIC AKI

Injury and Repair of Nephron

Renal Tubule Damage

The proximal tubule is divided into three molecularly,histologically, and topographically distinct segments:S1, S2, and S3.11 The S3 segment, although highlydeveloped in rodents, is not as pronounced in humanbeings. The epithelial cells in the straight S3 segmentof the rodent proximal tubules located in the outerstripe of the outer medulla are exquisitely sensitive toischemic insults. Histologically, the ischemic injury isreadily discernible in this stripe in animal models ofischemic AKI induced by clamping of the renalpedicle. The S1 and S2 segments of the proximaltubule also respond to injury but the S3 segment showsthe most marked cell loss after AKI in the mousekidney.12,13 Although the medullary thick ascendinglimb (TAL) of the loop of Henle also resides in theouter medullary region, the TAL is relatively resistantto IRI. However, an AKI-like phenotype can beinduced experimentally by targeting apoptosis

specifically within the TAL.14 IRI regimens that effec-tively target the S3 segment of the proximal tubule (PT)have little effect on cells of the TAL. The differentialsensitivities off adjacent tubular epithelial cell typesmay reflect a distinct ability of TAL cells to switchfrom oxidative to glycolytic metabolism,15 to mountanti-apoptotic response (activating extracellular signal-related kinase and BCl-2 proteins),16 and increasedexpression of insulin-like growth factor-1 (IGF-1) andhepatocyte growth factor (HGF).17

Both proximal and distal tubules undergo cell deathin human AKI although biopsy specimens of renalallografts show significantly greater apoptosis in distaltubules whereas proximal tubular epithelial cells showmore marked proliferation.18 Focal areas of tubularepithelial cell loss in the TAL and proximal tubular S3segment have been reported in patients with ischemicacute tubular necrosis.19

Ischemia-induced renal tubular adenosine triphosphatedepletion is likely an initiating insult in rodent IRI-associated AKI. Critical alterations in tubular dynamics,metabolism, and structure ultimately lead to necrotic and/or apoptotic cell death. These include depletion of cellularenergy stores, loss of basolateral distribution ofNaþKþadenosine triphosphatase and β-integrins (loss ofpolarity), disruption of the actin cytoskeleton and adher-ent and tight-junctions (shedding of brush border andsloughing of cells), accumulation of intracellular calcium,accumulation of hypoxanthine, and generation of reactiveoxygen species.20

Renal Tubule Repair

Damaged renal tubular epithelium may be repaired bysurviving epithelial cells, other cell types resident withinthe kidney, or cells that move into the injured organ. Onlydirect experimental analysis can distinguish among thesepossibilities; consequently, the most robust conclusionsare founded on fate-mapping strategies using mousegenetics. By using approaches that label renal tubulecells exclusively, Humphreys et al21 argued that repair bysurviving cells within the proximal tubule epithelium is abroad mechanism. Further analysis of clone size anddifferentiation markers suggests that repair in S1/S2segments is not mediated by a rare stem cell but isgeneral property of differentiated proximal tubule epithe-lial cells activated on injury.22 A contrasting view arguesfor repair from a small subset of CD24þ, CD133þ cellsthat reside within human renal tubules.23,24

Non-Nephron Components of Injury and Repair

Macrophage, Leukocytes, and Neutrophils

One of the earliest cellular responses to renal damage,seen within the first few hours after the triggeringstimulus, is neutrophil and macrophage infiltration;

S. Kumar, J.Liu, and A.P. McMahon406

key aspects of the engagement of an innate immuneresponse.25,26 Early monocyte/macrophage trafficking isfacilitated by chemokine (C-C motif) ligand 2 (CCL2)(monocyte chemoattractant protein 1)/chemokine (C-Cmotif) receptor 2 (CCR2) and chemokine (C-X3-Cmotif) ligand 1 (CX3CL1)/chemokine (C-X3-C motif)receptor 1 (CX3CR1) chemokine signaling pathways.26

Macrophage infiltration peaks at 24 hours and persistsfor at least 7 days. This component of proinflammatorymacrophages (M1) infiltrating the inflamed kidney isdistinct from resident macrophages and dendritic cells.26

As a consequence of IRI-triggered changes in theenvironment, macrophages may transition from aninitial proinflammatory state to acquire anti-inflamma-tory, pro-reparative properties.27 By 24 hours afterinjury initiation in the mouse kidney, macrophagesexpressed high levels of inducible nitric oxide synthase(a marker of proinflammatory M1 macrophages), andlow levels of arginase-1 (a marker of M2 macrophagesor alternatively activated macrophages). Subsequently,over the ensuing 6 days, flow-sorted macrophagesshowed increasing levels of arginase-1 paralleled bydecreasing levels of inducible nitric oxide synthase,indicative of an M1 to M2 transition within thispopulation. Liposomal clodronate–mediated depletionof M2 macrophages negatively impacts repair assessedat day 5 and day 7 after injury initiation, suggestingthat the transition to M2 macrophages is beneficial tothe repair process.

How these macrophages augment repair is unclearalthough production and secretion of a pro-reparativeWnt ligand (Wnt7b) (Wingless-related integration site(Wnt) family),28 and colony stimulating factor-1,29 aretwo possible mechanisms. However, macrophages alsomay drive AKI-associated fibrosis and increase the riskof subsequent chronic kidney injury; macrophageshave been implicated in fibrosis in chronic kidneydisease (CKD) models.30 See article by BenjaminHumphreys for a discussion of this topic.

Vascular Cells

The circulatory system is not only the source of oxygenfor the tubular nephron but the conduit for ingressinginflammatory cells; entry is enabled by the up-regulationof adhesion molecules and selectins, on the surface ofendothelial cells.20 Ischemia-reperfusion injury indu-ced alterations in renal microcirculation are thought tocompromise endothelial function. The capillaries in theouter medulla are uniquely susceptible to ischemicinsults because of various factors including dispropor-tionately reduced blood flow in the outer medullacompared with total kidney perfusion. The capillaryplexus of the outer strip is relatively sparse, suppliedfrom the small side branches that rise exclusively fromthe efferent arterioles of the juxtamedullary glomeruli.

Capillary rarefaction, a reduction in the numberof arterioles and capillaries, is observed on IRI.31 Acompromised microvascular density may exacerbate theinitial hypoxia and potentially contribute to the progres-sive development of interstitial fibrosis as in the fibroticscarring observed in Alzheimer's disease.32 According tothe chronic hypoxia hypothesis, capillary rarefaction is animportant factor driving a final common pathway to end-stage renal disease.33 The development of salt-sensitivehypertension and impaired urinary concentrating abilityare other functional consequences of vessel drop-out.

IMMEDIATE AND EARLY MOLECULAR RESPONSESTO AKI

Immediate-Early Damage Responses (Up to 4 HoursAfter Injury)

Both proximal and distal tubular epithelial cells mountan acute transcriptional response to IRI. The earliestgenes to be induced after in vivo injury (within 4 h afterinjury) the immediate-early genes include Fos, Jun, andEgr1.34 Fos is induced predominantly in the TAL.35 Thelatter observation suggests that the distal tubule, inaddition to the proximal tubule, also senses the acuteinsult. Subsequent microarray-based gene expressionprofiling studies encountered a similar immediate-earlyresponse, including Fos and Egr1, after renal IRI.36,37

An oxidative stress–induced increase in the intracellularcalcium ion concentration is one possible explanationfor this hyperacute response. Interestingly, in vitro, ratproximal tubular epithelial cells up-regulate Fos and Junmessenger RNA (mRNA) within 15 minutes of oxida-tive stress, attaining peak responses within 30 minutes,and returning to basal levels within 3 hours.38 Theprecise biological role of this immediate response, andthe interplay with ensuing transcriptional responses, isnot well understood.

Early Damage Responses (4 to 24 Hours After Injury)

A number of molecular approaches, including a repre-sentational difference analysis of complementary DNA(cDNA), were used in the late 1990s and early 2000s toexamine AKI responses at the molecular level, identi-fying two prominent injury indicators: Havcr1 (alsoknown as the kidney injury molecule-1 [Kim-1])39 andLcn2 (also known as the neutrophil gelatinase-associated lipocalin2 [NGAL]).36,40 A variety of ratand mouse models of AKI have sought commonmolecular themes and additional earlier biomarkers ofinjury responses.36,37,41,42 Collectively, these studieshave contributed significantly to our understanding ofthe acute responses after kidney injury, and identifiedfrontline biomarker candidates, several of which areundergoing clinical scrutiny for diagnostic efficacy.43


These profiling studies show shared early robustmolecular responses after AKI, irrespective of theunderlying insult.36,37,41,42 These include hemoxy-genase-1 (Ho1), Lcn2 (NGAL), Havcr1 (Kim1),annexin A2 (Anxa2), clusterin, and interleukin 6 (Il6).Induction of HO-1 and NGAL reflect the organ’sresponse to mitigate toxic effects of intracellular hemeand free catalytic iron, respectively, resulting fromrenal insult including IRI. Heme oxygenase convertsthe pro-oxidant, proinflammatory, and pro-apoptoticheme to biliverdin, a reaction that produces cytopro-tective molecules and endogenous toxic heme.44

NGAL also is induced in renal tubules, providing areservoir for excess iron.45 Iron in its free catalyticform is a mediator of renal IRI, triggering the inductionof toxic reactive oxygen species generation.46 NGAL:siderophore:Fe protects against ischemic IRI via up-regulation of hemoxygenase-1.47

Increased KIM-1 after renal IRI facilitates theclearance of dead cells, conferring endocytic andphagocytic phenotypes on epithelial cells with resultantinternalization of lipoproteins and epithelial cells.48 IRIleads to increased intracellular calcium and inducesAnnexin a2.49 Annexins are known to bind phospho-lipids in a Ca2þ-dependent manner, and participates invarious membrane-related events such as exocytosis,endocytosis, apoptosis, and binding to cytoskeletalproteins.50 Although the precise biological role ofannexin a2 in AKI has not been determined, its actionsmay contribute to inflammation by increasing IL-6production, akin to its role in lupus nephritis.51 IL-6 isa key proinflammatory cytokine up-regulated in bothischemic and toxic models of AKI. IL-6 is likely acritical driver of renal and extrarenal inflammatoryresponses after injury.52 Proximal tubule injury acti-vates macrophage-mediated production of IL-6, partic-ularly within the outer medullary region.53

Apoptosis is clearly one cellular mechanism, identi-fied by a cDNA microarray-based gene expressionprofiling study, underlying early loss of renal tubulecells on ischemic IRI.36 Several pro-apoptotic genes areup-regulated within the first 24 hours of IRI: membersof the extrinsic death receptor pathway (Fadd and Daxx)and the intrinsic mitochondrial apoptotic pathway (Badand Bak), and the anti-apoptotic gene, Bcl2. Renal IRIwas attenuated significantly in Bcl-2 transgenic micewith pre-activation of Bcl2,54 and in BH3 interactingdomain death agonist (Bid)-deficient mice.55 In contrast,loss of Bax inhibitor-1 enhanced injury.56

The striking down-regulation of the majority of thegenes involved in the mitochondrial metabolismmachinery at 24 hours after IRI is a common themeamong the significantly down-regulated gene sets inthe aforementioned studies. The biological and func-tional consequences of such a response is poorlyunderstood. Proximal tubular cells are highly enriched

in mitochondria and one possible explanation is that itreflects the proximal tubular cells attempt to attain amore protective hypometabolic state.

Reactivation of developmental genes during tubulerepair/regeneration after AKI is a more widely believedparadigm than the data to support this view. Compa-rative cDNA microarray profiling between early andlate stages of nephrogenesis and adult mouse kidneysat 3, 12, and 24 hours after IRI has identified somecorrelated changes in genes encoding transcriptionalcomponents (Nmyc1 and Wt1), and growth factors(Gdnf and Mdk).57 Two key developmental pathwaysregulating nephrogenesis, the Wnt/β-catenin and Notchsignaling pathways, are reported to be activated afterAKI.28,58 Re-expression of Pax2, a key transcriptionalregulator in nephrogenesis, in injured tubular epithelialcells has been suggested to reflect a dedifferentiation oftubular epithelial cell.59

However, it remains unclear whether re-expressionof these IRI-activated genes reflects a similar regula-tory action to their normal role in ontogeny of thekidney. Furthermore, they raise the interesting questionas to what extent tubular epithelial cells truly dediffer-entiate. Direct unbiased analysis of relevant cells usingnew approaches for genome-wide discovery likely willprovide some clarity. Recent reports that humanembryonic stem cells and induced pluripotent stemcells can be coaxed in vitro into nephrogenic programsopens the door to comparing human nephrogenesiswith adult repair programs.60–62

Although the analysis of organ-wide injuryresponses provides broad information, the kidney is acomplex organ; in all likelihood, there is much to belearned from a closer examination of individual cellpopulations. Moreover, acute responses mediated byrelatively rare, but biologically significant, cell pop-ulations will be lost among the responses of moreabundant cellular compartments.

Fluorescent-activated cell sorting (FACs) cells isone approach to increase cellular resolution, althoughthe necessary processes of cell isolation can triggerinjury responses, exaggerating data variability anddiminishing reproducibility.63 A new approach, TRAP,provides a useful alternative strategy.64,65 TRAP hasthe additional advantage of focusing on the activelytranslated mRNA population at any stage; importantly,general mRNA profiling does not always predictchanges in protein abundance, indicating mechanismsfavoring translation of specific mRNAs.37

Recently, a generalized TRAP approach was devel-oped and applied to the mouse kidney to obtain cell-specific molecular signatures in an IRI-invoked AKImodel.66 Here, TRAP relies on affinity purification oftranslating ribosomes through an enhanced greenfluorescent protein (eGFP)-tagged, L10a ribosomalprotein subunit (L10a::eGFP), and subsequent profiling

Figure 1. TRAP RNA-seq work flow. pA = SV40 poly A; CAGGS = hybrid promoter composed of Cytomega-lovirus early enhancer fused to hicken β actin promoter; DAVID = Database for Annotation, Visualization andIntegrated Discovery. (see Liu et al66 for full details regarding approach).


of mRNAs stripped from the ribosome by microarrayor RNA-seq. Cell type specificity is governed by therequirement for CRE-recombinase–mediated removalof a transcription-blocking cassette upstream of anL10a::eGFP cDNA cassette: cell type–specific CRElines activate L10a::eGFP in distinct cell populations inthe kidney (Fig. 1). In a recent study,66 distinct CRElines enabled TRAP mRNA signatures to be generatedfor four critical cellular compartments in the kidney

after IRI injury: the nephron, vascular, macrophage/monocyte, and interstitial mesenchyme (Fig. 2).

Intersection of TRAP data, with gene lists fromvarious whole-organ gene expression profiling studiesperformed within 24 hours of AKI,36,37,41,42,67 showgeneral responses, mounted by all four major cellularcompartments, and cell-restricted responses were de-fined as responses noted in three, or fewer than three,cellular compartments (Tables 1 and 2, respectively).

Figure 2. Cell type–specific responses observed through celltype–specific TRAP analysis 24 hours after IRI. The heat mapdisplays the relative microarray probe intensity for a given geneacross samples. The representation of data was generated usingGenePattern (Broad Institute, Boston, MA). Six2CRE, labels thenephron; Foxd1CRE, labels the interstitial cells including mesan-gial and podocytes; Cdh5CRE, labels the vascular endothelium;Lyz2-CRE, labels the cells of myeloid lineage, notably, monocytes,macrophages, neutrophils, and dendritic cells. N, no surgery; S,sham surgery; I, IRI surgery.

Table 1. Comparison of Responses in AKI Models Analyzed by TraIdentified by TRAP Profiling Within the First 24 Hours of IRI or Cisp

StudiesGeneSymbol

Yoshida et al42, Yuen et al37 Akap12 A kinasancho

Yoshida et al42, Yuen et al37 Anxa2 AnnexinYoshida et al42 Anxa3 annexinHuang et al41, Yuen et al37 Cd44 CD44 aHuang et al41, Yoshida et al42, Yuenet al37, Supavekin et al36, Riss et al67

Cdkn1a Cyclin-d

Huang et al41, Yuen et al42, Riss et al67 Clu clusterinYuen et al37, Supavekin et al36 Egr1 Early grYuen et al37 Eif1a EukaryoYuen et al37 Fosl1 Fos-likeYuen et al37, Supavekin et al36, Risset al67

Hmox1 Heme o

Yuen et al37, Supavekin et al36 Lcn2 LipocaliYuen et al37, Supavekin et al36 Sphk1 SphingoYuen et al37, Supavekin et al36 Cldn7 Claudin


A number of common molecular responses providediagnostic evidence of AKI themes independent of theAKI trigger. Approximately 20% of differentiallyexpressed genes are shared in different models ofAKI: Table 1 summarizes the shared responses at 24hours. These include Lcn2, Ho1, Sphk1, p21, Cd44,Anxa2, Anxa3, Fosl1, and Clu.

Specific models also highlighted several molecularresponses of interest that were not common to all AKIstudies (Table 2). Such responses can be groupedfurther according to their site of activation, such asAdm, Myc, Tpm4, and Tnfrsf12a (nephron and inter-stitium/pericyte and endothelium); Anxa1, Cldn7, andHavcr1 (nephron and interstitium/pericyte and myeloid-lineage cells); Fos (endothelium and myeloid lineagecells); Hspa1a and Tubb5 (nephron and interstitium/pericytes); Fgb and Gdf15 (only nephron); Cxcl1, Il6,and Tagln (only interstitial/pericyte); Dnajb9 and inter-cellular adhesion molecule-1 (only endothelium); andVcam1 (only myeloid-lineage cells). TRAP unraveledlarge sets of unique responses in the vascular endothe-lium and interstitial/pericyte populations, in addition tothe expected responses of genes associated with theirrespective compartments (Table 2).

A striking IRI feature of the TRAP Cdh5-L10avascular endothelium compartment is evidence of path-way regulation for CXCR4, endothelin-1, Toll-likereceptor, IL-8, thrombopoietin, and Janus kinase-STAT(JAK-STAT) signaling. These data highlight vasculature-associated molecular pathways not readily discernible in

nscriptional and Translational Profiling: Shared Response Geneslatin- (Nephrotoxic) Induced Kidney Damage

Gene Name AKI Models

e (regulatory subunit of protein kinase A)r protein (gravin) 12

IRI

A2 IRIA3 IRIntigen Cisplatin, IRIependent kinase inhibitor 1A (P21) Cisplatin, IRI

Cisplatin, IRIowth response 1 IRItic translation initiation factor 1A IRIantigen 1 IRIxygenase (decycling) 1 IRI

n 2 IRIsine kinase 1 IRI7 IRI

Table 2. Comparison of Responses in AKI Models Analyzed by Transcriptional and Translational Profiling: Cell Compartment–SpecificRegulation of IRI and Cisplatin Invoked Responses Predicted Through Comparison With TRAP Responses Observed Within DistinctCell Populations in the Mouse Kidney

Studies Gene Symbol Gene Name AKI Models

TRAP IRI

Six2 Foxd1 Cdh5 Lyz2

Yoshida et al42,Yuen et al37

Fgb Fibrinogen β chain IRI √


Gdf15 Growth differentiation factor 15 IRI √


Havcr1 Hepatitis A virus cellular receptor 1 IRI √

Yoshida et al42 Cxcl1 Chemokine (C-X-C motif) ligand 1 IRI √Yuen et al37 Il6 Interleukin-6 IRI √Yuen et al37 Tagln Transgelin IRI √Yuen et al37 Dnajb9 Predicted gene 6568; DnaJ (Hsp40) homolog,

subfamily B, member 9IRI √

Yuen et al37 Icam1 Intercellular adhesion molecule-1 IRI √Yuen et al37 Hspa1a Heat shock protein 1B; heat shock protein 1A;

heat shock protein 1-likeIRI √ √

Yoshida et al42,Riss et al67

Tubb5 Tubulin, β 5 IRI √ √

Yoshida et al42 Vcam1 Vascular cell adhesion molecule 1 IRI √ √Supavekin et al36 Hbegf Hbegf heparin-binding epidermal growth

factor–like growth factorIRI √ √

Yuen et al37 Fos FBJ osteosarcoma oncogene IRI √ √ √Yuen et al37 Anxa1 Annexin A1 IRI √ √ √Huang et al41,Yuen et al37

Myc Myelocytomatosis oncogene Cisplatin, IRI √ √ √

Yuen et al37 Tnfrsf12a Tumor necrosis factor–receptor superfamily,member 12a

IRI √ √ √

Yuen et al37 Tpm4 Tropomyosin 4; predicted gene 7809 IRI √ √ √Yuen et al37 Adm Adrenomedullin IRI √ √ √


a microarray-based analysis of total kidney RNA sam-ples. Similarly, specific responses from infiltrating cellsof the myeloid lineage (Lyz2-L10a) readily are shown byTRAP. These included IL-6, peroxisome proliferator-activated receptor, glucocorticoid, IL-17A, and extracel-lular signal-related kinase 5 signaling. Furthermore, thedata identify chemokine receptors (Ccr1 and Cxcr2) andgrowth factors (Csf1) linked to IRI. The glutamate-leucine-arginine (ELR) motif containing CXC chemo-kines (ELR þ CXC) attract polymorphonuclear leuko-cytes to the sites of acute inflammation. Gene ontologyalso shows a significant enrichment of IRI-induced genesassociated with hepatic fibrosis/hepatic stellate cell acti-vation, suggesting parallels between kidney and liver infibrotic programs, and furthermore, that fibrosis-inducingactivities are present within Lyz2-derived myeloid line-ages and Foxd1-derived, interstitial/pericytes lineages24 hours after IRI, before histologically apparentfibrosis (late fibrosis).

For several of these early response genes additionaldata are available from genetic studies on their actions inAKI-induced renal dysfunction. In general, a significantacute injury response involves myeloid-lineage cells

aggravating ischemic AKI. Genetic knockout of Cd44,Edn1, intercellular adhesion molecule-1, and Il6 protectagainst ischemic AKI.53,68,69 In contrast, AKI is wors-ened upon removal of Cdkn1a, Hmox1, and Atf3, arguingfor a primary protective role of this group of factors.70–72

INTERMEDIATE AND LATE MOLECULARRESPONSES TO AKI: 48 HOURS OR LONGER AFTERINJURY

Chronic Inflammation and Extracellular MatrixRemodeling After Ischemic AKI

To date, relatively few published studies have examinedwhole-organ or cell-specific molecular signatures at laterstages of the injury responses defined here as later than48 hours after the insult, when evident repair is underway(Table 3). Despite differences in species, insults, molec-ular profiling platforms, and genes under investigation,a unifying theme is the inability of the postischemickidney to return to a pre-injury histologic or basal mo-lecular state. Persistence of the proinflammatory milieuremains a major feature in the later postischemic kidney,

Figure 3. Analysis of gene expression changes observed in twostudies of unilateral IRI-invoked kidney damage 48 hours, or later,after injury initiation. Genes showing a more than 1.5-fold expres-sion change were compared in studies by Ko et al73 (profiling at 3,10, and 28 days after IRI) and Riss et al67 (profiling at 2, 5, 7, and14 days after IRI). Genes shared between studies, or unique toeach study, are indicated.

Table 3. Kidney mRNA Profiling After AKI

Studies Species AKI Model Time Points Profiling Platforms

Basile et al86 Rat IRI D 35 Customized cDNA microarrayKrishnamoorthyet al98

Rat IRI D 5 mRNA profiling

Ko et al73 Mouse uIRI D 3, d 10, d 28 mRNA profilingTran et al96 Mouse LPS (septic AKI) 42 h mRNA profilingStroo et al77 Mouse uIRI 24 h, 7 d Chemokine pathway–specific microarrayKo et al78 Mouse uIRI 6 h, d 3, d 10,

d 28Th1-Th2-Th3 RT2 profiler polymerasechain reaction array

Riss et al67 Mouse uIRI D 1, d 2, d 5,d 7, d 14

Mouse cDNA microarrays

Abbreviation: uIRI, unilateral ischemia reperfusion injury.


exaggerated by distinct late proinflammatory molecularresponses after injury. Of note, published studies haveused unilateral clamping of the renal pedicle in studyinglater responses to ischemic AKI. This makes parallelassessment of kidney function problematic in unilateralversus bilateral IRI models. Consequently, drawing clear-cut conclusions on whether an observed invoked molec-ular response is equivalent to survival is only possible inbilateral injury models.

Figure 3 summarizes shared and unique molecularfeatures between comparable gene expression profilingstudies examining molecular responses in a repairingkidney after unilateral IRI injury.67,73 A gene listcontaining genes that remained continuously increased(41.5-fold change) throughout the intermediate (48 h today 7) and late phases (day 7 onward) was collated forboth studies. Five genes were shared between the twostudies and the remaining 26 and 25 genes, respectively,were specific to each study (Fig. 3). Interestingly, theshared genes suggest persistent inflammation (comple-ment C3, and suppressor of cytokine signaling 3) andcontinuous extracellular matrix remodeling (matrix gla-protein and cathepsin S) after injury. Extracellularmatrix turnover is mediated by a number of elastolyticproteinases, including Zn2þ/Ca2þ-dependent matrixmetalloproteinases and cathepsin cysteine proteases,such as cathepsin S.74 C3 is a key component of thecomplement cascade, a fundamental innate defensesystem. C3 is situated at the crossroads of three majorcomplement activation pathways, yielding severaleffector molecules with powerful inflammatoryeffects.75 Indeed, complement C3 influences long-term kidney transplant outcomes.76 Defense, immune,and inflammatory responses persist after renal IRI.73

Among gene ontology analysis of data, complementactivation, chemotaxis, cell adhesion, and antigenpresentation emerge at day 10 after injury.73

The temporal profiling of chemokines using chemo-kine pathway–specific microarrays identified 14 newgenes up-regulated on day 7 after unilateral IRI in themouse, reflecting a changing component of infiltrating

cell types or a changing activation profile of existingcells within the damaged kidney.77 Among this groupare the CC chemokines (Ccl2, Ccl6, Ccl12, and Ccl17),whose actions are linked to the attraction of mono-nuclear cells to sites of chronic inflammation, andCX3CL1 (also known as fractalkine, a member of theCX3C family). Ccl2, also known as monocyte chemo-attractant protein 1, is a potent agonist for monocytes,dendritic cells, memory T cells, and basophils.

FACs sorted T cells infiltrating the postischemickidney also display transcriptome changes as suggestedby the T helper (Th)1-Th2-Th3 polymerase chain reac-tion array analysis.78 Four weeks after injury and/orinsult, genes associated with co-stimulatory pathway for


antigen presentation, and cellular and humoral responses,are up-regulated in T cells, however, the biologicalsignificance is unclear. Interestingly, T-cell co-stimula-tory pathways increasingly are being considered astherapeutic targets to suppress T-cell–mediated immuno-logic injury and prolong renal allograft survival rates.79

MicroRNAs (miRNAs) play important roles in variedcell-biological processes including development, apopto-sis, proliferation, and differentiation. When miRNAs aredown-regulated globally in the cortices of the proximaltubule through the proximal tubule–specific knock-out ofdicer, an enzyme critical for miRNA genesis, the kidneyis reported to be less susceptible to renal dysfunctionsecondary to ischemic AKI.80 Emerging evidence sug-gests that a few miRNAs remain significantly increasedlate in the course of postischemic kidney (Table 4).81–83

Table 5 illustrates the shared and unique microRNAresponses among the unilateral IRI and unilateral ureteralobstruction (UUO) models. The shared responses includemiR-21, miR-20a, miR-119a-3p, and miR-146a. Whethertheir increase plays a part in the persistence of chronicinflammatory processes has not been examined directly.However, miR-146a is known to regulate innate immuneand inflammatory responses via post-translational inhib-ition of key target genes.84 A 12-month-old, age-matchedcomparative microarray analysis for 511 miRNAs inkidneys of B6.MRLc1 mice (a model of inflammation-driven spontaneous CKD) versus C57BL6 mice showedthe highest expression level (2.2-fold) of miR-146a in thekidneys of CKD mice.85 These observations raise theintriguing possibility that miR-146a may contribute tosustained inflammation after renal IRI. Interestingly, asimilar differential expression profile in the immunodefi-cient RAG-2/common ϒ-chain double-knockout micesuggests a negligible contribution of the infiltratinglymphocytes to the overall miRNA signature of thepostischemic kidney.81

Studies in the rat also show a persistently altered geneprofile after AKI injury. Basile et al86 sought to identifyalterations in renal gene expression in the recovered rats

Table 4. miRNA Expression Changes Linked to IRI-Induced AKI

Study Species Injury Time Point Profiling

Godwin et al81 Mouse uIRI 24 h, d 3,d 7, d 14,d 28

miRNA profiling(μParaflomicrofluidic ar

Chau et al83 Mouse uIRI D 10 miRNA profiling(Agilent micro

Saikumaret al82

Rat IRI D 5 miRNA profiling


35 days after bilateral renal ischemia reperfusion injury:serum creatinine levels had returned to baseline a weekafter IRI. By using a customized cDNA microarray toexamine 2,000 rat genes, 16 genes persistently werealtered at 35 days. Among the 12 up-regulated genes,osteopontin (Opn), complement C4 (proinflammatory),and S100A4 remained up-regulated after serum crea-tinine levels normalized. The role of Opn was exploredfurther in Opn-deficient mice: a reduction of naturalkiller cell infiltration was observed correlating withdecreased tissue damage 5 days after IRI consistentwith a negative impact of endogenous Opn in the naturalrepair process.87 However, a second study found nodifferences in functional or morphologic consequencesof ischemic AKI up to 7 days after IRI in a unilateralinjury model.88 Although both studies agree on reducedlevels of infiltrating immune cells, the former highlightsnatural killer cells and the latter highlights macrophages.Differences in surgical models or mouse strains mayunderlie differences between these independent findings.

Fibrosis After Ischemic AKI

The persistence of AKI biomarker up-regulation sev-eral weeks after normalization of serum creatininelevels after IRI injury could reflect early indicationsof a progression from AKI to CKD.73

NGAL/Lipocalin2-deficient mice are relatively pro-tected against the development of renal lesions (tubularinjury, interstitial fibrosis) after 75% nephron reductionsurgery.89 Kim-1 expression correlates directly withinterstitial fibrosis in human allografts,90 and increasedurinary Kim-1 is an independent predictor of long-termrenal graft loss.91 In mice, chronic expression of Kim-1in renal epithelial cells in the absence of an insult led toprogressive interstitial kidney inflammation with fib-rosis.92 These mice developed a phenotype analogousto the clinical progression of human CKD includingproteinuria, anemia, hyperphosphatemia, hypertension,and cardiac hypertrophy. These studies raise the

Key miRNAs Biologics

ray)

miR-21, miR-20a, miR-146a,miR-199a-3p, miR-214,miR-192, miR-187, miR-805, and miR-194

miR-199a-3p:prosurvival; miR-21,miR-192, miR-146a:profibrotic

array)42 miRNAs up-regulated (24shared with UUO); miR-21,miR-214, miR-199a-3p,miR-146a

miR-21: profibrotic

33 miRNAs up; miR-21,miR-155, miR-18a

miRNA-21 and miRNA-155: potential urinarybiomarker

Table 5. Comparative Analysis of Unique and Shared miRNA Expression Changes Between uIRI and UUO Models of Kidney Injury

Study Injury miRNAs

Unique in Godwin et al81 uIRI miR-187, miR-192, miR-194, miR-805Shared in Godwin andChau et al81,83

uIRI/UUO miR-21, miR-214, miR-199a-3p, miR-146a, miR-20a

Unique in Chau et al83 uIRI/UUO miR-25, miR-93, miR-132, miR-183, miR-223, miR-350, miR-674,miR-199a-5p, miR-142-3p, miR-343-3p, miR-18a, miR-99b, miR-199b,miR-let-7i, miR-19a, miR-15b, miR-106b, miR-92a, miR-15a



possibility that Kim-1 does not simply play a role inengulfment of apoptotic cells in early AKI but maytrigger pathogenic proinflammatory and profibroticeffects. Removing Kim-1 activity at different periodsafter an AKI trigger will provide important insightsinto Kim-1 action, and the potential role of Kim-1 inprogressive renal pathology.

The distinct wave of chemokines observed a weekafter injury (discussed earlier) also could play a role infibrotic pathogenesis, a possibility supported by theclose temporal association between chemokine expres-sion and histologically apparent fibrosis at 2 weeksfurther strengthens this possibility. Mice deficient forthe CX3CL1 receptor, Cx3CR1, showed significantlyreduced infiltration of macrophages correlating withdecreased fibrosis, particularly in the outer medullaryregion.93 Treatment with a CX3CR1-neutralizing anti-body also reduced fibrosis. The contribution of chemo-kines to the reparative processes, including recruitmentand cross-talk of immune cells, is likely to continue asa major focus in identifying the initiating and prop-agating factors in injury-related fibrotic disease.

miRNAs also may contribute to the progression ofAKI to CKD. By using miRNA microarrays, Chauet al83 identified 14 miRNAs unique to UUO, 18unique to IRI studies, and a set of 24 miRNAs up-regulated in both UUO and unilateral IRI models ofkidney damage 10 days after initiation of injury. Thiscommon set includes miRNA-214, miRNA-199a-3p,miRNA-21, miRNA-20a, and miRNA-146a, as dis-cussed earlier for unilateral IRI-associated microarraychanges.81 Interestingly, miRNA-21 knock-out micehad significantly less injury-induced fibrosis, and anti–miR-21 oligonucleotide treatment of the wild-typemice reduced fibrosis in the setting of unilateral IRIand UUO.83 Of the IRI-specific set, miRNA-192 isdown-regulated rapidly and persistently after ischemicAKI.81 Loss of miR-192 correlates with tubulointer-stitial fibrosis and a reduction in renal function inpatients with established diabetic nephropathy, andtransforming growth factor-β decreases miR-192expression,94 suggesting that a similar transforminggrowth factor-β–driven repression of miRNA-192 afterischemic injury could promote kidney fibrosis. The

miRNA action on their mRNA targets may modulateproduction of up to 30% of total cellular proteins, smallchanges in large networks from miRNA imbalancecould have a major pathologic impact. However,identifying the most critical protein components medi-ating pathology in these networks will be a majorchallenge.95

Tubular Proliferation/Repair after Ischemic AKI

Recent gene expression profiling studies have providedinsights into potential strategies and therapeutic targetsto augment tubular proliferation responses. By using alipopolysaccharide (LPS)-induced septic AKI model,Tran et al96 compared differential gene expressionamong kidneys with persistent injury versus thoseshowing functional recovery 42 hours after LPS admin-istration. Gene ontology analysis identified oxidativephosphorylation and mitochondrial dysfunction amongthe top three enriched pathways. More specifically,expression of peroxisome proliferative activated recep-tor, gamma, coactivator 1 alpha (PGC-1α), a transcrip-tional regulator of mitochondria and oxidative metabolicprograms, was down-regulated significantly in kidneysthat failed to recover from LPS treatment. Proximaltubule–specific knockout of PGC-1α resulted in persis-tent renal dysfunction after LPS treatment, consistentwith a role for PGC-1α in functional recovery fromendotoxemia. Reduced expression of PGC-1α in tubularepithelium also is associated with mitochondrial dys-function in cisplatin-induced proximal tubule injury.97

Collectively, these studies argue for a more comprehen-sive understanding of the mitochondrial enzyme machi-nery in tubular repair processes.

Cell proliferation is a marked, and likely essential,response to effective repair of AKI. Krishnamoorthyet al98 performed whole-genome expression profilingof rat cortex and medulla and identified fibrinogen(Fg)α, Fgβ, and Fgγ to be increased persistently afterIRI throughout the study (end point, 5 days after IRI),with a tubular expression similar to Kim-1. Admin-istration of Fgβ-derived Bβ15-42 peptide promotedtubular cell proliferation and protected againstischemia-induced AKI in the mouse.


Hypertension After AKI

As discussed earlier, inflammatory and immune-associated cellular responses are predominant withinthe transcriptome after AKI injury. Less clear, butlikely of importance, is the pathophysiology thatresults from fluid and electrolyte disturbances, andhypertension associated with AKI.

Basile et al86 reported that the vasodilator kallikreinremains strongly down-regulated 35 days after renal IRIin the rat. The kallikrein system plays an important rolein blood pressure regulation, salt sensitivity, and electro-lyte excretion. A kallikrein-deficient rat strain manifestspolydipsia and hypertension in response to increasedsodium intake as well as progressive renal scarring.99

Mining the repair transcriptome with a particular focuson understanding hemodynamic and electrolytederangements may provide valuable new insights intothe pathophysiology of salt and volume overload,hyperkalemia, hypophosphatemia, urinary concentratingdefects, metabolic acidosis, and hypertension that fre-quently is encountered after human AKI.

CELL-SPECIFIC GENE EXPRESSION PROFILING OFTHE REPAIRING KIDNEY

The T-cell profiling study examining infiltrating cellsthrough FACs and polymerase chain reaction arrayanalysis takes an expansive (multistage), cell type–specific focus on kidney damage and repair.78 Theimmune response is complex, varied immune cellshave been shown to play a role in acute and chronicpathogenesis of AKI. Sifting anti-inflammatory, pro-reparative responses from deleterious proinflammatoryand profibrotic triggers is a challenge, especially if theformer reflects a rare cell population whose transcrip-tional signature is a minor component of a largerproinflammatory response in the whole organ.

For example, Foxp3þCD4þ regulatory T cells(Tregs) play a critical role in immune homeostasisincluding self-tolerance and their ability to suppressinflammation. Several lines of evidence suggest thatTregs may promote repair after ischemic AKI. Adoptivetransfer of Tregs at 24 hours after injury,100 or IL-1/anti–IL-2 complex–mediated expansion of the intrinsic Tregpool, promote functional renal recovery exemplified byreduced serum blood urea nitrogen and creatinine levelsin a treatment cohort 5 days after injury.101 Further, Tregdepletion aggravates ischemic renal dysfunction.102

HUMAN KIDNEY INJURY TRANSCRIPTOME ANDREPAIR

Kidney transplantation can be considered a highlyscrutinized in vivo model of human AKI in whichthe inciting insults can be identified precisely in a

temporal manner, unlike other settings of human AKI.Immediately after kidney transplant, significant AKIcauses delayed graft function, an independent predictorof allograft rejection and graft loss. To define thetranscripts induced by human AKI and assess theirimpact on renal allograft outcomes, Famulski et al103

performed microarray-based gene expression profilingof kidney tissue from 26 kidney transplant patientsidentified as a "pure AKI" cohort. The controls con-sisted of a set of 11 age-matched pristine protocolbiopsy specimens from a different transplant cohort.The transcript score (geometric mean of the fold-increase in the top 30 transcripts versus controlnephrectomies) correlated with reduced graft function,renal recovery, and requirement for renal replacementtherapy assessed at 6 months after AKI.

The intragraft molecular signature predominantlyreflected the renal parenchymal response to AKI, andshowed similarities to cancer, cell adhesion, cell move-ment, and re-expression of developmental programs.Responses significantly overlapped with those observedin IRI-induced AKI mouse models, indicating a broadconservation of molecular and cellular processes.

Interestingly, similar transcriptional signatures alsowere encountered in transplants with other causes ofallograft dysfunction including chronic antibody-mediated graft injury and recurrent primary renaldisease.104 The considerable similarity between theintergraft transcriptomes, irrespective of the underlyingpathology, suggests that the human transplant kidneymounts a shared robust response to varied insults.Induction of AKI-associated transcripts (predominantlyparenchymal) was reported to be a better predictor offuture graft loss than fibrosis, inflammation, or expres-sion of collagen genes.104,105 Although these findingssuggest that fibrosis may not be as significant afactor to the overall progression of AKI, it remainshighly likely that fibrosis contributes to longer-terminjury, notably postrecovery onset of CKD. Althoughthere are procedural and analytic hurdles to overcome,transcriptomic profiling of renal allograft biopsy speci-mens could become a useful complement to currentapproaches for risk stratification of transplant patients.

GENE EXPRESSION PROFILING: REPAIR/REGENERATION VERSUS CANCER

Transcriptional profiling studies comparing unilateral IRIin the mouse at various time points with expressionprofiling of human renal cell carcinoma has drawnparallels between these distinct kidney insults.67 Of 361differentially expressed genes identified in both condi-tions, the majority (77%) showed concordant expressionchanges, either up-regulation or down-regulation, sugges-tive of related biological processes: cell proliferation, cell


migration, cell adhesion, and cell death gene ontologyterms are shared between IRI and cancer samples.

CONCLUSIONS AND FUTURE DIRECTIONS

The move from biased (array-based) to unbiased (next-generation sequencing) assessment of the transcrip-tome will provide important new insights into bothcoding and noncoding components of injury/repairresponses in AKI.106,107 Deep sequencing of RNA,RNA-seq, enables a complete survey of mRNAs andnoncoding RNAs (eg, miRNA and long noncodingRNAs) that serves not only for discovery-identifyingnovel genes and variant transcripts, but as a morerigorous measure of observed responses.

In the developing kidney, the Genito UrinaryDevelopment Molecular Anatomy Project (GUDMAR)initiative (www.GUDMAP.org) has produced a wealthof high-quality annotated data from direct visualizationand indirect microarray-based analysis of gene expres-sion. These data now are being complemented by high-quality RNA-seq data sets.108 These data already haveproduced numerous novel insights into already well-studied developmental processes. This bodes well forinsights that will be obtained in less well-scrutinizedevents induced on AKI in the adult kidney.

Understanding how the various cell types in thekidney communicate to regulate the intrinsic repairmechanisms, and their contribution to postinjury fib-rosis, remains a major challenge. Here, a continuedfocus on cell-type–specific signatures, and a broad-ening of analysis beyond early injury, is likely toprovide important new insights. Clearly, the majorclinical goal is to develop new analytic tools todiagnose both short- and long-term outcomes, and, atthe same time, to develop new therapeutic strategies toimprove existing repair processes and reduce the long-term risk of CKD after AKI.

REFERENCES1. Mehta RL, Kellum JA, Shah SV, et al. Acute Kidney Injury

Network: report of an initiative to improve outcomes in acutekidney injury. Crit Care. 2007;11:R31.

2. Waikar SS, Liu KD, Chertow GM. Diagnosis, epidemiologyand outcomes of acute kidney injury. Clin J Am Soc Nephrol.2008;3:844-61.

3. Chertow GM, Burdick E, Honour M, Bonventre JV, BatesDW. Acute kidney injury, mortality, length of stay, and costsin hospitalized patients. J Am Soc Nephrol. 2005;16:3365-70.

4. VA/NIH Acute Renal Failure Trial Network, Palevsky PM,Zhang JH, et al. Intensity of renal support in critically ill patientswith acute kidney injury. N Engl J Med. 2008;359:7-20.

5. Coca SG, Singanamala S, Parikh CR. Chronic kidney diseaseafter acute kidney injury: a systematic review and meta-analysis. Kidney Int. 2012;81:442-8.

6. Bonventre JV, Yang L. Cellular pathophysiology of ischemicacute kidney injury. J Clin Invest. 2011;121:4210-21.

7. Yang L, Besschetnova TY, Brooks CR, Shah JV, BonventreJV. Epithelial cell cycle arrest in G2/M mediates kidney fibrosisafter injury. Nat Med. 2010;16:535-43,1p following 143.

8. Bechtel W, McGoohan S, Zeisberg EM, et al. Methylationdetermines fibroblast activation and fibrogenesis in the kidney.Nat Med. 2010;16:544-50.

9. Kramann R, DiRocco DP, Humphreys BD. Understanding theorigin, activation and regulation of matrix-producing myofi-broblasts for treatment of fibrotic disease. J Pathol. 2013;231:273-89.

10. LeBleu VS, Taduri G, O'Connell J, et al. Origin and function ofmyofibroblasts in kidney fibrosis. Nat Med. 2013;19:1047-53.

11. Helbert MJ, Dauwe SE, Van der Biest I, Nouwen EJ, De BroeME. Immunodissection of the human proximal nephron: flowsorting of S1S2S3, S1S2 and S3 proximal tubular cells.Kidney Int. 1997;52:414-28.

12. Heyman SN, Evans RG, Rosen S, Rosenberger C. Cellularadaptive changes in AKI: mitigating renal hypoxic injury.Nephrol Dial Transplant. 2012;27:1721-8.

13. Shanley PF, Rosen MD, Brezis M, Silva P, Epstein FH, Rosen S.Topography of focal proximal tubular necrosis after ischemia withreflow in the rat kidney. Am J Pathol. 1986;122:462-8.

14. Srichai MB, Hao C, Davis L, et al. Apoptosis of the thickascending limb results in acute kidney injury. J Am SocNephrol. 2008;19:1538-46.

15. Bagnasco S, Good D, Balaban R, Burg M. Lactate productionin isolated segments of the rat nephron. Am J Physiol.1985;248:F522-6.

16. Gobe G, Zhang XJ, Willgoss DA, Schoch E, Hogg NA,Endre ZH. Relationship between expression of Bcl-2 genesand growth factors in ischemic acute renal failure in the rat.J Am Soc Nephrol. 2000;11:454-67.

17. Gobe GC, Johnson DW. Distal tubular epithelial cells of thekidney: potential support for proximal tubular cell survivalafter renal injury. Int J Biochem Cell Biol. 2007;39:1551-61.

18. Oberbauer R, Rohrmoser M, Regele H, Muhlbacher F, MayerG. Apoptosis of tubular epithelial cells in donor kidneybiopsies predicts early renal allograft function. J Am SocNephrol. 1999;10:2006-13.

19. Olsen TS, Hansen HE. Ultrastructure of medullary tubules inischemic acute tubular necrosis and acute interstitial nephritisin man. APMIS. 1990;98:1139-48.

20. Devarajan P. Update on mechanisms of ischemic acute kidneyinjury. J Am Soc Nephrol. 2006;17:1503-20.

21. Humphreys BD, Valerius MT, Kobayashi A, et al. Intrinsicepithelial cells repair the kidney after injury. Cell Stem Cell.2008;2:284-91.

22. Kusaba T, Humphreys BD. Controversies on the origin ofproliferating epithelial cells after kidney injury. Pediatr Neph-rol. 2014;29:673-9.

23. Romagnani P, Lasagni L, Remuzzi G. Renal progenitors: anevolutionary conserved strategy for kidney regeneration. NatRev Nephrol. 2013;9:137-46.

24. Sagrinati C, Netti GS, Mazzinghi B, et al. Isolation andcharacterization of multipotent progenitor cells from theBowman's capsule of adult human kidneys. J Am SocNephrol. 2006;17:2443-56.

25. Swaminathan S, Griffin MD. First responders: understandingmonocyte-lineage traffic in the acutely injured kidney. KidneyInt. 2008;74:1509-11.

26. Li L, Huang L, Sung SS, et al. The chemokine receptorsCCR2 and CX3CR1 mediate monocyte/macrophage traffick-ing in kidney ischemia-reperfusion injury. Kidney Int.2008;74:1526-37.

27. Lee S, Huen S, Nishio H, et al. Distinct macrophagephenotypes contribute to kidney injury and repair. J Am SocNephrol. 2011;22:317-26.

www.GUDMAP.org

http://refhub.elsevier.com/S0270-9295(14)00080-1/sbref1





















































































28. Lin SL, Li B, Rao S, et al. Macrophage Wnt7b is critical forkidney repair and regeneration. Proc Natl Acad Sci U S A.2010;107:4194-9.

29. Menke J, Iwata Y, Rabacal WA, et al. CSF-1 signals directlyto renal tubular epithelial cells to mediate repair in mice.J Clin Invest. 2009;119:2330-42.

30. Duffield JS. Macrophages in kidney repair and regeneration.J Am Soc Nephrol. 2011;22:199-201.

31. Basile DP, Donohoe D, Roethe K, Osborn JL. Renal ischemicinjury results in permanent damage to peritubular capillariesand influences long-term function. Am J Physiol RenalPhysiol. 2001;281:F887-99.

32. Zlokovic BV. Neurovascular mechanisms of Alzheimer'sneurodegeneration. Trends Neurosci. 2005;28:202-8.

33. Venkatachalam MA, Griffin KA, Lan R, Geng H, Saikumar P,Bidani AK. Acute kidney injury: a springboard for progres-sion in chronic kidney disease. Am J Physiol Renal Physiol.2010;298:F1078-94.

34. Ouellette AJ, Malt RA, Sukhatme VP, Bonventre JV. Expres-sion of two "immediate early" genes, Egr-1 and c-fos, inresponse to renal ischemia and during compensatory renalhypertrophy in mice. J Clin Invest. 1990;85:766-71.

35. Witzgall R, Brown D, Schwarz C, Bonventre JV. Localizationof proliferating cell nuclear antigen, vimentin, c-Fos, andclusterin in the postischemic kidney. Evidence for a hetero-genous genetic response among nephron segments, and a largepool of mitotically active and dedifferentiated cells. J ClinInvest. 1994;93:2175-88.

36. Supavekin S, Zhang W, Kucherlapati R, Kaskel FJ, MooreLC, Devarajan P. Differential gene expression following earlyrenal ischemia/reperfusion. Kidney Int. 2003;63:1714-24.

37. Yuen PS, Jo SK, Holly MK, Hu X, Star RA. Ischemic andnephrotoxic acute renal failure are distinguished by their broadtranscriptomic responses. Physiol Genomics. 2006;25:375-86.

38. Maki A, Berezesky IK, Fargnoli J, Holbrook NJ, Trump BF.Role of [Ca2þ]i in induction of c-fos, c-jun, and c-myc mRNAin rat PTE after oxidative stress. FASEB J. 1992;6:919-24.

39. Ichimura T, Bonventre JV, Bailly V, et al. Kidney injurymolecule-1 (KIM-1), a putative epithelial cell adhesion moleculecontaining a novel immunoglobulin domain, is up-regulated inrenal cells after injury. J Biol Chem. 1998;273:4135-42.

40. Mishra J, Ma Q, Prada A, et al. Identification of neutrophilgelatinase-associated lipocalin as a novel early urinary bio-marker for ischemic renal injury. J Am Soc Nephrol. 2003;14:2534-2543.

41. Huang Q, Dunn RT 2nd, Jayadev S, et al. Assessment ofcisplatin-induced nephrotoxicity by microarray technology.Toxicol Sci. 2001;63:196-207.

42. Yoshida T, Kurella M, Beato F, et al. Monitoring changes ingene expression in renal ischemia-reperfusion in the rat.Kidney Int. 2002;61:1646-54.

43. Koyner JL, Parikh CR. Clinical utility of biomarkers of AKIin cardiac surgery and critical illness. Clin J Am Soc Nephrol.2013;8:1034-42.

44. Nath KA. Heme oxygenase-1 and acute kidney injury. CurrOpin Nephrol Hypertens. 2014;23:17-24.

45. Schmidt-Ott KM, Mori K, Li JY, et al. Dual action ofneutrophil gelatinase-associated lipocalin. J Am Soc Nephrol.2007;18:407-13.

46. McCord JM. Oxygen-derived free radicals in postischemictissue injury. N Engl J Med. 1985;312:159-63.

47. Mori K, Lee HT, Rapoport D, et al. Endocytic delivery oflipocalin-siderophore-iron complex rescues the kidney fromischemia-reperfusion injury. J Clin Invest. 2005;115:610-21.

48. Ichimura T, Asseldonk EJ, Humphreys BD, Gunaratnam L,Duffield JS, Bonventre JV. Kidney injury molecule-1 is a

phosphatidylserine receptor that confers a phagocytic phe-notype on epithelial cells. J Clin Invest. 2008;118:1657-68.

49. Cheng CW, Rifai A, Ka SM, et al. Calcium-binding proteinsannexin A2 and S100A6 are sensors of tubular injury andrecovery in acute renal failure. Kidney Int. 2005;68:2694-703.

50. Camors E, Monceau V, Charlemagne D. Annexins and Ca2þhandling in the heart. Cardiovasc Res. 2005;65:793-802.

51. Yung S, Cheung KF, Zhang Q, Chan TM. Anti-dsDNAantibodies bind to mesangial annexin II in lupus nephritis.J Am Soc Nephrol. 2010;21:1912-27.

52. Klein CL, Hoke TS, Fang WF, Altmann CJ, Douglas IS,Faubel S. Interleukin-6 mediates lung injury following ische-mic acute kidney injury or bilateral nephrectomy. Kidney Int.2008;74:901-9.

53. Kielar ML, John R, Bennett M, et al. Maladaptive role of IL-6 inischemic acute renal failure. J Am Soc Nephrol. 2005;16:3315-25.

54. Suzuki C, Isaka Y, Shimizu S, et al. Bcl-2 protects tubularepithelial cells from ischemia reperfusion injury by inhibitingapoptosis. Cell Transplant. 2008;17:223-9.

55. Wei Q, Yin XM, Wang MH, Dong Z. Bid deficiencyameliorates ischemic renal failure and delays animal death inC57BL/6 mice. Am J Physiol Renal Physiol. 2006;290:F35-42.

56. Wei Q, Dong G, Chen JK, Ramesh G, Dong Z. Bax and Bakhave critical roles in ischemic acute kidney injury in globaland proximal tubule-specific knockout mouse models. KidneyInt. 2013;84:138-48.

57. Devarajan P, Mishra J, Supavekin S, Patterson LT, StevenPotter S. Gene expression in early ischemic renal injury: cluestowards pathogenesis, biomarker discovery, and novel ther-apeutics. Mol Genet Metab. 2003;80:365-76.

58. Kobayashi T, Terada Y, Kuwana H, et al. Expression andfunction of the Delta-1/Notch-2/Hes-1 pathway during exper-imental acute kidney injury. Kidney Int. 2008;73:1240-50.

59. Imgrund M, Grone E, Grone HJ, et al. Re-expression of thedevelopmental gene Pax-2 during experimental acute tubularnecrosis in mice 1. Kidney Int. 1999;56:1423-31.

60. Takasato M, Er PX, Becroft M, et al. Directing humanembryonic stem cell differentiation towards a renal lineagegenerates a self-organizing kidney. Nat Cell Biol. 2014;16:118-26.

61. Taguchi A, Kaku Y, Ohmori T, et al. Redefining the in vivoorigin of metanephric nephron progenitors enables generationof complex kidney structures from pluripotent stem cells. CellStem Cell. 2014;14:53-67.

62. Xia Y, Nivet E, Sancho-Martinez I, et al. Directed differ-entiation of human pluripotent cells to ureteric bud kidneyprogenitor-like cells. Nat Cell Biol. 2013;15:1507-15.

63. Emery B, Barres BA. Unlocking CNS cell type heterogeneity.Cell. 2008;135:596-8.

64. Doyle JP, Dougherty JD, Heiman M, et al. Application of atranslational profiling approach for the comparative analysis ofCNS cell types. Cell. 2008;135:749-62.

65. Heiman M, Schaefer A, Gong S, et al. A translational profilingapproach for the molecular characterization of CNS cell types.Cell. 2008;135:738-48.

66. Liu J, Krautzberger AM, Sui SH, et al. Cell-specific translationalprofiling in acute kidney injury. J Clin Invest. 2014;124:1242-54.

67. Riss J, Khanna C, Koo S, et al. Cancers as wounds that do notheal: differences and similarities between renal regeneration/repair and renal cell carcinoma. Cancer Res. 2006;66:7216-24.

68. Kelly KJ, Williams WW Jr, Colvin RB, et al. Intercellularadhesion molecule-1-deficient mice are protected againstischemic renal injury. J Clin Invest. 1996;97:1056-63.

69. Arfian N, Emoto N, Vignon-Zellweger N, Nakayama K, YagiK, Hirata K. ET-1 deletion from endothelial cells protects thekidney during the extension phase of ischemia/reperfusioninjury. Biochem Biophys Res Commun. 2012;425:443-9.









































































































































70. Megyesi J, Andrade L, Vieira JM Jr, Safirstein RL, Price PM.Positive effect of the induction of p21WAF1/CIP1 on the courseof ischemic acute renal failure. Kidney Int. 2001;60:2164-72.

71. Nath KA, Haggard JJ, Croatt AJ, Grande JP, Poss KD, AlamJ. The indispensability of heme oxygenase-1 in protectingagainst acute heme protein-induced toxicity in vivo. Am JPathol. 2000;156:1527-35.

72. Li HF, Cheng CF, Liao WJ, Lin H, Yang RB. ATF3-mediatedepigenetic regulation protects against acute kidney injury.J Am Soc Nephrol. 2010;21:1003-13.

73. Ko GJ, Grigoryev DN, Linfert D, et al. Transcriptionalanalysis of kidneys during repair from AKI reveals possibleroles for NGAL and KIM-1 as biomarkers of AKI-to-CKDtransition. Am J Physiol Renal Physiol. 2010;298:F1472-83.

74. Aikawa E, Aikawa M, Libby P, et al. Arterial and aortic valvecalcification abolished by elastolytic cathepsin S deficiency inchronic renal disease. Circulation. 2009;119:1785-94.

75. Walport MJ. Complement. First of two parts. N Engl J Med.2001;344:1058-66.

76. Brown KM, Kondeatis E, Vaughan RW, et al. Influence ofdonor C3 allotype on late renal-transplantation outcome.N Engl J Med. 2006;354:2014-23.

77. Stroo I, Stokman G, Teske GJ, et al. Chemokine expression inrenal ischemia/reperfusion injury is most profound during thereparative phase. Int Immunol. 2010;22:433-42.

78. Ko GJ, Linfert D, Jang HR, et al. Transcriptional analysis ofinfiltrating T cells in kidney ischemia-reperfusion injuryreveals a pathophysiological role for CCR5. Am J PhysiolRenal Physiol. 2012;302:F762-73.

79. Riella LV, Sayegh MH. T-cell co-stimulatory blockade intransplantation: two steps forward one step back!. Expert OpinBiol Ther. 2013;13:1557-68.

80. Wei Q, Bhatt K, He HZ, Mi QS, Haase VH, Dong Z. Targeteddeletion of Dicer from proximal tubules protects against renalischemia-reperfusion injury. J Am Soc Nephrol. 2010;21:756-61.

81. Godwin JG, Ge X, Stephan K, Jurisch A, Tullius SG, IacominiJ. Identification of a microRNA signature of renal ischemiareperfusion injury. Proc Natl Acad Sci U S A. 2010;107:14339-44.

82. Saikumar J, Hoffmann D, Kim TM, et al. Expression, circu-lation, and excretion profile of microRNA-21, -155, and -18afollowing acute kidney injury. Toxicol Sci. 2012;129:256-67.

83. Chau BN, Xin C, Hartner J, et al. MicroRNA-21 promotesfibrosis of the kidney by silencing metabolic pathways. SciTransl Med. 2012;4:(121ra18)

84. Li L, Chen XP, Li YJ. MicroRNA-146a and human disease.Scand J Immunol. 2010;71:227-31.

85. Ichii O, Otsuka S, Sasaki N, Namiki Y, Hashimoto Y, Kon Y.Altered expression of microRNA miR-146a correlates with thedevelopment of chronic renal inflammation. Kidney Int.2012;81:280-92.

86. Basile DP, Fredrich K, Alausa M, et al. Identification ofpersistently altered gene expression in the kidney after func-tional recovery from ischemic acute renal failure. Am JPhysiol Renal Physiol. 2005;288:F953-63.

87. Zhang ZX, Shek K, Wang S, et al. Osteopontin expressed intubular epithelial cells regulates NK cell-mediated kidneyischemia reperfusion injury. J Immunol. 2010;185:967-73.

88. Persy VP, Verhulst A, Ysebaert DK, De Greef KE, De BroeME. Reduced postischemic macrophage infiltration and inter-stitial fibrosis in osteopontin knockout mice. Kidney Int.2003;63:543-53.

89. Viau A, El Karoui K, Laouari D, et al. Lipocalin 2 is essentialfor chronic kidney disease progression in mice and humans.J Clin Invest. 2010;120:4065-76.

90. Schroppel B, Kruger B, Walsh L, et al. Tubular expression ofKIM-1 does not predict delayed function after transplantation.J Am Soc Nephrol. 2010;21:536-42.

91. van Timmeren MM, Vaidya VS, van Ree RM, et al. Highurinary excretion of kidney injury molecule-1 is an independ-ent predictor of graft loss in renal transplant recipients.Transplantation. 2007;84:1625-30.

92. Humphreys BD, Xu F, Sabbisetti V, et al. Chronic epithelialkidney injury molecule-1 expression causes murine kidneyfibrosis. J Clin Invest. 2013;123:4023-35.

93. Furuichi K, Gao JL, Horuk R, Wada T, Kaneko S, MurphyPM. Chemokine receptor CCR1 regulates inflammatory cellinfiltration after renal ischemia-reperfusion injury. J Immunol.2008;181:8670-6.

94. Krupa A, Jenkins R, Luo DD, Lewis A, Phillips A, Fraser D.Loss of MicroRNA-192 promotes fibrogenesis in diabeticnephropathy. J Am Soc Nephrol. 2010;21:438-47.

95. Lewis BP, Burge CB, Bartel DP. Conserved seed pairing,often flanked by adenosines, indicates that thousandsof human genes are microRNA targets. Cell. 2005;120:15-20.

96. Tran M, Tam D, Bardia A, et al. PGC-1alpha promotesrecovery after acute kidney injury during systemic inflamma-tion in mice. J Clin Invest. 2011;121:4003-14.

97. Portilla D, Dai G, McClure T, et al. Alterations of PPARalphaand its coactivator PGC-1 in cisplatin-induced acute renalfailure. Kidney Int. 2002;62:1208-18.

98. Krishnamoorthy A, Ajay AK, Hoffmann D, et al. Fibrinogenβ-derived Bbeta(15-42) peptide protects against kidney ische-mia/reperfusion injury. Blood. 2011;118:1934-42.

99. Madeddu P, Vio CP, Straino S, Salis MB, Milia AF, Emanueli C.Renal phenotype of low kallikrein rats. Kidney Int. 2001;59:2233-42.

100. Kinsey GR, Sharma R, Huang L, et al. Regulatory T cellssuppress innate immunity in kidney ischemia-reperfusioninjury. J Am Soc Nephrol. 2009;20:1744-53.

101. Kim MG, Koo TY, Yan JJ, et al. IL-2/anti-IL-2 complexattenuates renal ischemia-reperfusion injury through expan-sion of regulatory T cells. J Am Soc Nephrol. 2013;24:1529-36.

102. Gandolfo MT, Jang HR, Bagnasco SM, et al. Foxp3þregulatory T cells participate in repair of ischemic acutekidney injury. Kidney Int. 2009;76:717-29.

103. Famulski KS, de Freitas DG, Kreepala C, et al. Molecularphenotypes of acute kidney injury in kidney transplants. J AmSoc Nephrol. 2012;23:948-58.

104. Famulski KS, Reeve J, de Freitas DG, Kreepala C, Chang J,Halloran PF. Kidney transplants with progressing chronicdiseases express high levels of acute kidney injury transcripts.Am J Transplant. 2013;13:634-44.

105. Halloran PF, Reeve JP, Pereira AB, Hidalgo LG, FamulskiKS. Antibody-mediated rejection, T cell-mediated rejection,and the injury-repair response: new insights from the GenomeCanada studies of kidney transplant biopsies. Kidney Int.2014;85:258-64.

106. Wang Z, Gerstein M, Snyder M. RNA-Seq: a rev-olutionary tool for transcriptomics. Nat Rev Genet. 2009;10:57-63.

107. Wilhelm BT, Marguerat S, Watt S, et al. Dynamic repertoireof a eukaryotic transcriptome surveyed at single-nucleotideresolution. Nature. 2008;453:1239-43.

108. Brunskill EW, Lai HL, Jamison DC, Potter SS, Patterson LT.Microarrays and RNA-Seq identify molecular mechanismsdriving the end of nephron production. BMC. Dev Biol.2011;11:15.



































































































































defining the acute kidney injury and repair transcriptome · pdf filedefining the acute kidney...

Documents