the rip1-kinase inhibitor necrostatin-1 prevents osmotic...

BASIC RESEARCH www.jasn.org

The RIP1-Kinase Inhibitor Necrostatin-1 PreventsOsmotic Nephrosis and Contrast-Induced AKI in Mice

Andreas Linkermann,* Jan-Ole Heller,* Ágnes Prókai,† Joel M. Weinberg,‡ Federica DeZen,* Nina Himmerkus,§ Attila J. Szabó,† Jan H. Bräsen,|¶ Ulrich Kunzendorf,* andStefan Krautwald*

*Division of Nephrology and Hypertension, Christian-Albrechts-University, Kiel, Germany; †First Department ofPaediatrics, Semmelweis University, Budapest, Hungary; ‡Renal Division, University of Michigan Medical Center, AnnArbor, Michigan; Institutes for §Physiology and |Pathology, Christian-Albrechts-University, Kiel, Germany; and¶Pathology Hamburg-West, Institute for Diagnostic Histopathology and Cytopathology, Hamburg, Germany

ABSTRACTThe pathophysiology of contrast-induced AKI (CIAKI) is incompletely understood due to the lack of anappropriate in vivomodel that demonstrates reduced kidney function before administration of radiocon-trast media (RCM). Here, we examine the effects of CIAKI in vitro and introduce a murine ischemia/reper-fusion injury (IRI)–based approach that allows induction of CIAKI by a single intravenous application ofstandard RCM after injury for in vivo studies. Whereas murine renal tubular cells and freshly isolated renaltubules rapidly absorbed RCM, plasma membrane integrity and cell viability remained preserved in vitro

and ex vivo, indicating that RCM do not induce apoptosis or regulated necrosis of renal tubular cells. Invivo, the IRI-based CIAKI model exhibited typical features of clinical CIAKI, including RCM-induced os-motic nephrosis and increased serum levels of urea and creatinine that were not altered by inhibition ofapoptosis. Direct evaluation of renal morphology by intravital microscopy revealed dilation of renaltubules and peritubular capillaries within 20 minutes of RCM application in uninjured mice and similar,but less dramatic, responses after IRI pretreatment. Necrostatin-1 (Nec-1), a specific inhibitor of thereceptor-interacting protein 1 (RIP1) kinase domain, prevented osmotic nephrosis and CIAKI, whereasan inactiveNec-1 derivate (Nec-1i) or the pan-caspase inhibitor zVADdid not. In addition, Nec-1 preventedRCM-induced dilation of peritubular capillaries, suggesting a novel role unrelated to cell death for the RIP1kinase domain in the regulation of microvascular hemodynamics and pathophysiology of CIAKI.

J Am Soc Nephrol 24: 1545–1557, 2013. doi: 10.1681/ASN.2012121169

Contrast-induced AKI (CIAKI) is the consensusname for whatwas formally called contrast-inducednephropathy or radiocontrast-induced AKI.1–3

CIAKI is a common and potentially serious com-plication4 after the administration of contrast me-dia,5–7 especially in patients who are at risk for AKI,and is the most common cause of iatrogenic, inpa-tient, drug-induced AKI,3,8,9 with outstanding im-plications for patients with diabetes.1 CIAKI wasrecognized as the third commonest cause of hos-pital-acquired renal failure accounting for 11% ofthe cases10 even beforemagnetic resonance imagingcontrast media were found to be associated withnephrogenic systemic fibrosis. Preclinical researchthus far has failed to unravel the underlying path-ophysiology of CIAKI.

Programmed cell death (PCD) was used synon-ymously with apoptosis until regulated necrosis(RN) was discovered.11 Apoptosis has been pro-posed to contribute to CIAKI12–14 and asialoery-thropoietin was recently demonstrated in thiscontext to prevent CIAKI.15 Apoptosis is a process

Received December 11, 2012. Accepted April 22, 2013.

Published online ahead of print. Publication date available atwww.jasn.org.

Correspondence: Dr. Ulrich Kunzendorf, Division of Nephrol-ogy and Hypertension, Christian-Albrechts-University Kiel,Schittenhelmstr. 12, 24105 Kiel, Germany. Email: [email protected]

Copyright © 2013 by the American Society of Nephrology

J Am Soc Nephrol 24: 1545–1557, 2013 ISSN : 1046-6673/2410-1545 1545

BASICRESE

ARCH

http://www.jasn.org

mailto:[email protected]

mailto:[email protected]

Figure 1. Regular doses of RCM do not induce cell death in TKPTS cells, MMCs, and glENDs. (A) TKPTS cells are left untreated or aretreated with 50 ml/ml standard RCM (Imeron) for the indicated time points. Rapid nuclear RCM uptake is visualized by light microscopyfollowed over a period of 25 minutes. (B) Greyscale quantification of A and Supplemental Figure 1, A and B, after application of RCM.(C) TKPTS cells, MMCs, and glENDs are left untreated or treated with 50 ml/ml RCM for the indicated time periods. Jurkat cells serve as

1546 Journal of the American Society of Nephrology J Am Soc Nephrol 24: 1545–1557, 2013


http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2012121169/-/DCSupplemental

positive controls for inductionof apoptosis after stimulationwith 100 ng/ml anti-FasmAb. The classic apoptosismarkers cleaved caspase-3(upper blot) and full-length PARP-1 versus cleaved PARP-1 (central blot) indicate the induction of apoptosis in Jurkat cells, but not in any ofthe remaining cell lines tested. GAPDH serves as a loading control. (D and E) Cells are stimulated as in C for the indicated time periods.Positivity for the apoptosis-marker annexin V (D) and the necrosis marker 7-AAD (E) are depicted. Again, Jurkat cells serve as a positivecontrol that is known to undergo secondary necrosis 8 hours after induction of apoptosis. GAPDH, glyceraldehyde 3-phosphate de-hydrogenase.

Figure 2. Regular doses of RCM do not induce cell death in freshly isolated primary renal tubules. (A) Primary murine proximal renaltubules are left untreated or are treated with 50 ml/ml RCM for the indicated time points. RCM uptake into the tubular compartment isvisualized by light microscopy and followed over a period of 25 minutes. (B) Greyscale quantification of A and Supplemental Figure 3, Aand B, after application of RCM. (C) Primary isolated murine renal tubules are left untreated (cold) or treated with 100 ml/ml RCM forthe indicated time periods. Tubules incubated in hypoxia for 60 minutes followed by 60 minutes of reoxygenation serve as positivecontrols. LDH release is measured and is shown as the percentage of overall LDH present in the whole tubules. (D) Primary renaltubules treated as in C are stained for positivity of propidium iodide. No statistically significant difference is observed in untreatedversus RCM-treated tubules after 240 minutes of incubation. DAPI, 4’,6-diamidino-2-phenylindole; PI, propidium iodide.

J Am Soc Nephrol 24: 1545–1557, 2013 Nec-1 Prevents CIAKI 1547

www.jasn.org BASIC RESEARCH



Figure 3. Blockade of apoptosis does not protect from RCM-induced osmotic nephrosis or AKI (CIAKI). (A) Eight-week-old male C57Bl/6mice undergo sham surgery or bilateral renal pedicle clamping 24 hours before intraperitoneal injection of either PBS or the pan-caspaseinhibitor zVAD followed by intravenous injection of RCM. Renal sections stained with periodic acid–Schiff are shown at magnifications of200-fold and 400-fold. Ischemia-reperfusion damage is not significantly altered in any of the groups (B), whereas quantification of



that is characterized by the activity of caspases that cleavehundreds of intracellular proteins to ultimately cause mem-brane blebbing, nuclear fragmentation, and regulated cellularshrinkage as a consequence of their proteolytic activity.16,17

Within this process, caspases are capable of cleaving NFs likepoly(ADP-ribose)-polymerase (PARP)-family proteins.18

PARP-1 has also been demonstrated to elicit a necrotic pheno-type in kidney cells and therefore exhibits a subroutine of theRN.19,20 It was suggested that tubular cell death by caspase-3–mediated apoptosis substantially contributes to the overall path-ogenesis ofCIAKI,14,15 and one report investigated the activationof the cell death molecules PARP, Bad, and BIM.14 On the basisof these findings, the currently proposed model ascribesapoptosis a major pathophysiologic function in CIAKI.12,13

Apart from PARP-mediated RN, necroptosis, another RNpathway, is mediated by activation of the “necrosome” con-sisting of receptor-interacting protein kinases 1 and 3 (RIP1and RIP3).11,21–23 Necroptosis involves all necrotic cellularhallmarks such as early loss of membrane integrity as well asrupture of the plasma membrane after cellular swelling. Werecently described the functional relevance of both apoptosisand necroptosis in AKI.24,25

Here, we demonstrate that necrostatin-1 (Nec-1), a highlyspecific inhibitor of the RIP1 kinase domain, prevents CIAKIin a new and easy-to-use preclinical model for the in vivoanalysis of CIAKI. Our model reliably mimics “osmoticnephrosis,” a pathologic feature that is typical of CIAKI inhumans. In vitro and in vivo, we found that apoptosis is ofminor pathophysiologic importance.Mechanistically, the dataimplicate RIP1 in the functional renal failure in vivo and pro-vide evidence for the prevention of CIAKI by the RIP1 kinaseinhibitor Nec-1 that also prevented the functional changes inthe peritubular vasculature after RCM injection as demon-strated by intravital microscopy. Because of the outstandingspecificity of Nec-1 that has been subject to extensive investi-gation,26–29 we consider it justified to conclude that a novelnon-cell death role of RIP1 might account for the functionalkidney failure in CIAKI. In addition, we introduce Nec-1 as apotential inhibitor of CIAKI.

RESULTS

Rapid Nuclear RCM Uptake in Kidney Cell Lines DoesNot Induce Cell Death In VitroPrevious reports suggested that kidney cells undergo PCDafterapplication of diverse RCM.12–15,30We aimed to investigate thedirect influence of clinically used standard RCM (Imeron) onrenal cells and initially demonstrated the direct uptake of RCM

into murine proximal tubular cells (TKPTS), mesangial cells(MMCs), and glomerular endothelial cells (glENDs) upon in-cubation with 50 ml RCM/ml (Figure 1A and SupplementalFigure 1). Greyscale analysis of contrast phase identified apeak RCM concentration within 15 seconds and a regular de-cline within 25 minutes (Figure 1B). We further investigatedRCM-induced cell death by evaluation of annexin V positivityand persistence of membrane integrity (exclusion of 7-AAD)upon medium (10 ml/ml) and high (50 ml/ml) concentrationsof RCM. No significant induction of cleaved caspase-3 orPARP-1 was detected in comparison with anti-Fas-treatedJurkat cells that serve as an apoptotic positive control, suggest-ing that classic caspase-mediated apoptosis is not initiatedafter RCM application (Figure 1C and Supplemental Figure2). Minimal levels of PARP1-cleavage in glENDs upon longWestern blot exposure (Supplemental Figure 3A) did not cor-relate with any detectable cell death (Supplemental Figure 3, Band C). Consistent with this, annexin V positivity in FACSanalysis was minimal (Figure 1D). To assess other necrotic-type cell death modalities, we applied 7-AAD that was ex-cluded over time in all renal cell lines (Figure 1E) at both 10and 50 ml/ml RCM. However, in accordance with previouslypublished data13–15 and Supplemental Figure 3A, very highconcentrations of RCM (250 ml/ml) did result in some an-nexin V positivity after 24 hours (Supplemental Figure 4A).With 400 ml RCM/ml, we found an increase in PARP-1 cleav-age that was unaffected by the addition of the caspase-8 in-hibitor zIETD, the pan-caspase inhibitor zVAD, TAT-crmA(a previously published fusion protein that utilized the viralcaspase-8 inhibitor31,32), or the cyclophilin D inhibitor cyclo-sporin A (Supplemental Figure 4B). We conclude that onlydoses that are high enough to induce artifacts lead to signifi-cant amounts of cell death evenwhen applied in cell culture for24 hours or longer.

RCM Do Not Induce Cell Death in Freshly IsolatedRenal TubulesTo transfer our in vitro data into an ex vivo setting, we treatedfreshly isolated proximal tubule segments with RCM. Com-parable to the results in TKPTS cells, epithelial cell nuclei inthe proximal tubule segments rapidly took up contrast media(Figure 2A). Similar results were obtained for thick ascendinglimb segments and segments from the distal convoluted tu-bules (Supplemental Figure 5). Greyscale analysis revealedsimilar RCM uptake kinetics in all tubular segments investi-gated (Figure 2B). Given the rapid direct RCM uptake, weinvestigated RCM-induced cell death as measured by lactatedehydrogenase (LDH) release (Figure 2C). Tubules treated for60 minutes with hypoxia followed by 60 minutes

osmotic nephrosis appears exclusively in the RCM-treated mice (C). Subcapsular tubules that are affected by osmotic nephrosis arequantified inD.CIAKI isevaluated48hoursafter reperfusion (24hoursafterapplicationofRCM)bydeterminationof serumcreatinine (E) andserum urea (F) concentrations. We will further refer to the difference between the PBS-treated and the RCM + PBS-treated group as ourmurinemodel of CIAKI. n=8per group. *P,0.05; **P,0.01; ***P,0.001 comparedwith the IRI-pretreated PBSgroup. n.s., not statisticallysignificant .














Figure 4. The kinase domain of RIP1 mediates osmotic nephrosis and CIAKI. (A) Eight-week-old male C57Bl/6 mice undergo shamsurgery or bilateral renal pedicle clamping 24 hours before intraperitoneal injection of either PBS, the highly specific RIP1 kinase inhibitorNec-1, or the inactive derivate of necrostatin-1 (Nec-1i) in the presence or absence of RCM. (B) Histologic quantification of renal IRI is notsignificantly changed in any of the groups. (C) Osmotic nephrosis in RCM-treated mice was reduced in Nec-1–treated mice, but not in



those that received Nec-1i. Osmotic nephrosis in subcapsular tubules is quantified in D. Concentrations of serum creatinine (E) and serumurea (F) are evaluated 48 hours after reperfusion (24 hours after application of RCM). n=8per group. **P,0.05; ***P,0.001 comparedwiththe IRI-pretreated PBS group. Sham versus IRI-treated mice: P,0.001 (not indicated in E and F). n.s., not significant.

Figure 5. Nec-1 prevents osmotic nephrosis and CIAKI. (A and B) Eight-week-old mice undergo sham surgery or CIAKI treatment asdescribed for Figures 3 and 4, and kidney sections are stained for immunohistochemistry of RIP1 at 400-fold magnification in thepresence or absence of PBS, Nec-1, or Nec-1i as indicated (A) and RIP1-positive tubules/tubular cells are counted and quantified foreach group (B) (n=6 per group). (C) LDH release of primary freshly isolated renal tubules from wild-type mice (open bars) compared withtubules from RIP3-deficient mice (gray bars) after cold preparation or 60 minutes and 240 minutes of incubation at 37°C in PBS in thepresence and absence of RCM and Nec-1 as indicated. Wild-type tubules that are incubated 60 minutes in hypoxia followed by 60minutes of reoxygenation serve as positive controls. n=8 per group. *P,0.05 compared with the IRI-pretreated PBS group. n.s., notstatistically significant.



Figure 6. Functional influence of RCM on peritubular capillaries is affected by Nec-1 in vivo. Eight-week-old male C57BL/6N micereceive 250 ml RCM via the tail vein 15 minutes after intraperitoneal injection of DMSO or Nec-1 with and without IRI pretreatment, asindicated. IVM is performed to detect changes in the diameters of peritubular capillaries and renal tubules. Representative images aredepicted in A, and in higher magnification in B. Nec-1 prevents the return to baseline transtubular diameters after 20–25 minutes (C)



of reoxygenation served as a positive control. As expected, LDHrelease increased over time in the isolated tubules, but no furtherincrease in LDH release was detected when RCM was added,despite the use of high concentration (100 ml/ml). Accordingly,and in line with the finding from TKPTS cells in Figure 1E,positivity for propidium iodide in the tubules increased overtime without further increases caused by RCM (Figure 2D).Similarly, no significant changes were detected by Western blot-ting regarding cleaved caspase-3 andPARP-1 (Supplemental Fig-ure 6). From these data, we conclude that the minimal amountof RCM-induced cell death does not provide a convincing path-ophysiologic concept to explain organ failure in CIAKI. To func-tionally address this question, we developed a newmodel for theanalysis of CIAKI in vivo.

An Ischemia/Reperfusion Injury–Based Model SystemAllows Investigation of CIAKIIn humans, several studies have reported osmotic nephrosisas a typicalmorphologic feature that accompanies renal failurein CIAKI.33–35 With regard to serum parameters and histolog-ically, however, doses as high as 250 ml RCM injected intrave-nously did not result in any detectable alterations 24 hoursafter application (Supplemental Figure 7). In previously pub-lished reports,15,30 CIAKI only develops if the GFR is reducedbefore RCM application. This previous work utilized waterdeprivation, unilateral nephrectomy, indomethacin, and N-nitro-L-arginine methyl ester.30,36 We therefore treated unilater-ally nephrectomized mice with high doses of indomethacin(up to 100mg/kg) in addition to high doses ofN-nitro-L-argininemethyl ester using concentrations up to 100 mg/kg, respec-tively, after 16 hours of water deprivation followed by in-travenous application of 250 ml RCM. In our hands, this didnot induce detectable AKI as measured by unchanged serumurea or creatinine concentrations 24 or 48 hours after injectionof RCM (Supplemental Figure 8). Our subsequent approachwas based on achieving the required loss of renal functionthrough an easy-to-reproduce ischemia/reperfusion injury(IRI)–based setting. After 24 hours of reperfusion, serumurea and creatinine levels and histologic analysis of the renaldamage were used to demonstrate low SDs in this system. At 24hours after reperfusion, we analyzed various concentrations ofRCM to induce CIAKI and found that 250 ml of RCM that wasrapidly injected via the tail vein confers an ideal setting (Sup-plemental Figure 9). We subsequently used this dose to char-acterize the time course of CIAKI in this model for within thefirst 96 hours after RCM injection (Supplemental Figure 10).According to these data, we performed the following experi-ments with 250 ml of RCM applied via the tail vein (RCMgroup) and read out serum markers and histology 24 hours

later (48 hours after reperfusion). These were compared withthe IRI-treated mice that received 250 ml of PBS instead ofcontrast media (PBS group) (Figure 3, A–G). We will furtherrefer to the difference between the RCM group and the PBSgroup as our model for CIAKI. In addition, we assessed theeffect of volume andN-acetylcysteine application in this modeland found no protective effects for these regimes (Supplemen-tal Figure 11). As quantified in Figure 3C and to the best of ourknowledge, this is the first in vivo model that reliably andclosely mimics the phenotype of tubular cell osmotic nephrosisin quantifiable resolution (Supplemental Figure 12). As expec-ted from the in vitro data, blockade of apoptosis did not in-fluence this CIAKI model in all parameters tested (Figure 3),but also did not worsen the outcome, as would be anticipatedin a purely necroptotic cell death.32

CIAKI Is Attenuated by Necrostatin-1, an Inhibitor ofthe Kinase Domain of Receptor-Interacting ProteinKinase-1Because other forms of PCD have recently emerged,20 andrenal cells preferentially appear to undergo programmed ne-crosis rather than apoptosis,19,25,37,38 we investigated the in-fluence of the RIP1 inhibitor necrostatin-1 (Nec-1) in the invivoCIAKImodel. Importantly for the design, and in line withpreviously published data,25 application of Nec-1 24 hoursafter reperfusion without RCM application did not alter theamount of IRI damage (Figure 4, A and B) and had no in-fluence on serum concentrations of urea and creatinine mea-sured 48 hours after reperfusion; thus, effects of Nec-1 at thetime 24 hours after ischemia/reperfusion (the time of RCM-administration) can be attributed to its effects on CIAKI. Wefound an almost complete prevention of RCM-induced os-motic nephrosis in the Nec-1–treated mice, but not in micetreated with an inactive derivate of Nec-132 called Nec-1i (Fig-ure 4C). CIAKI-induced affection of subcapsular tubules wasattenuated but not completely prevented by Nec-1 (Figure 4,D and E). Furthermore, CIAKI was markedly attenuated byNec-1 as demonstrated by the prevention of the RCM-inducedincrease in serum concentrations of creatinine and urea (Fig-ure 4, E and F). We conclude that the RIP1 kinase blocker,Nec-1, protects from CIAKI in our model.

CIAKI-Induced Loss of RIP1 Positivity Is Attenuatedby Nec-1To further investigate RIP1 biology in CIAKI, we performedimmunohistochemistry in sections from sham-treated andCIAKI-treated mice. RIP1 expression did not significantlychange upon renal IRI. In CIAKI, however, positivity for RIP1disappeared, and this effect was prevented byNec-1, but not by

and affects the dilation phase (injection to 10 minutes) in peritubular capillaries (D). Upon IRI pretreatment, application of Nec-1 leads tosignificantly increased transtubular diameters (E) andprevents the increase inperitubular capillary diameters that is seen in the controlmice(F). Note the artificial yellowappearanceof theheat-exposedarea in the intravitalmicroscope. This area is not included in thequantificationof tubular diameters (n.200per value) andperitubular capillaries (n.200per value). t tests reveal statistical significance asdepictedby thefollowing: *P,0.05; **P,0.01; ***P,0.001.













Nec-1i (Figure 5, A and B). Because RIP1 has been well de-scribed to activate RIP3 to form the so-called necroptosomeand mediate necroptosis11,20,39–41 but our in vitro results sug-gested that RCM do not induce cell death, we hypothesizedthat in CIAKI, the kinase domain of RIP1 might exert novelnon-cell death function. To exclude the involvement of RIP3in this context, we investigated RIP3-deficient freshly isolatedtubules and found no significant alterations in LDH release(Figure 5C), pointing to an effect that is RIP3 independent. Togain further information about the functional pathophysiol-ogy of CIAKI in vivo, we utilized intravital microscopy (IVM).

Nec-1 Induces Tubular Dilation and Affects the Kineticsof the Dilation of Peritubular Capillaries after RCMApplicationThe applicability of IVM to investigate AKI has recently beenestablished by others.42,43 We utilized IVM to detect the initialevents in tubules and peritubular capillaries upon RCM ap-plication in mice with an in vivo detection of renal tubules to adepth of 150 mm below the kidney capsule. Transtubular di-lation after intravenous application of 250 ml RCM injectionappeared throughout the first 20 minutes before diametersreturned to baseline levels. Upon a single intraperitoneal ap-plication of a single dose of Nec-1 15minutes before RCM, thereturn to baseline levels was prevented within the observationperiod. In line with this finding, we detected significantlywider tubules in the Nec-1 group after IRI pretreatment (Fig-ure 6, A, B, C, and E). In addition, and in line with a previousreport,44 RCM resulted in a significant increase in the diam-eter of the peritubular capillaries in the first 10minutes (6.9mmversus 7.9 mm, vasodilative phase) followed by a graduatedecline between 10 and 25 minutes (7.9 mm versus 5.8 mm,vasoconstrictive phase). As expected upon IRI pretreatment,peritubular capillaries remained in a dilated state, but diame-ters remained stable and significantly different to the wild-typemice in the Nec-1 group (Figure 6, A, B, D, and E). The lattereffect was also recently demonstrated for the vasa recta.45 Ac-cording to the laminar flow equation of Hagen-Poiseuille, theincreased blood flow might reflect the functional decrease inkidney function in CIAKI, without causing cell death, an effectthat is not observed upon Nec-1 treatment. In addition, theperitubular dilation was completely abolished by intraperito-neal application of Nec-1 15 minutes before application ofRCM, suggesting a functional role for RIP1 in this setting,even after IRI pretreatment. In combination with our in vitroresults, these data suggested that RCM might not kill tubularcells, but rather lead to a RIP1-associated functional renal fail-ure that involves vasodilation of the peritubular vessels, themechanism of which remains unclear.

DISCUSSION

Wepreviously reported that application ofNec-1 protects fromrenal IRI when it is applied before reperfusion.25

Mechanistically, the effect of Nec-1 on CIAKI presented in thisarticle is completely different, and must carefully be separatedfrom interference with necroptosis. Most importantly, additionofNec-1 24 hours after reperfusion does not influence the courseof IRI, which is a prerequisite for the present investigation inwhich we address three considerations that might be of interestin the context of CIAKI. First, we question the functional rele-vance of apoptosis, and PCD in general, in the pathogenesis ofCIAKI. Second, we introduce a new and easy-to-reproduceCIAKI mouse model. Third, we provide evidence for a non-cell death function for the kinase domain of RIP1 in regulationof vascular tone and CIAKI and its inhibitor Nec-1.

Apoptosis in tubular cells has been reported after variousiodide RCM in vitro such as 200 mg/ml of Iodixanol,14

320 mg/ml of Iodixanol,13 and 320 mg/ml of Ioversol.15

These agents have several limitations. First, these concentra-tions to our understanding are artificially high and do notmimic the clinical situation. Second, unlike RCM used in thisstudy, those RCM are not in everyday clinical use. Third, thedetection of apoptosis was performed using nonspecific ter-minal deoxynucleotidyl transferase–mediated digoxigenin-deoxyuridine nick-end labeling assays.39,46,47 Some reportsalso investigated caspase-3 activity, a very good marker forapoptosis, but the activity increased only by 1.4-fold.15 In typicalapoptotic settings, caspase-3 activity increases by$10-fold.31,48,49

Here we demonstrate that apoptosis only occurs in vitro usingartificially high RCM concentrations (Supplemental Figure 4),which is generally in line with some of the previously men-tioned reports.13,15 In vivo, we cannot find any protection fromCIAKI that is based on blockade of caspases (Figure 3) and donot find any significant amount of cell death that mightprovide a concept to explain AKI.

A preclinical model for CIAKI should meet three majorcriteria. First, itmust not be a lethalmodel to be comparable toclinically relevant CIAKI. Second, intravenous application ofcommonly used contrast media should lead to a measurableincrease in serum urea and creatinine concentrations. Third,typical histologic changes that are regularly seen in renalbiopsies, like osmotic nephrosis, should be easily detected bystandard histology. We undertook several approaches that hadbeen suggested in the literature, all of which did not meet theabove-mentioned three criteria in our hands using mice(Supplemental Figure 8). From several ischemia/reperfusionstudies, it is well known that kidney function was comparablein medium-sized groups 24 and 48 hours after reperfusionwith low SD in mice.25,50–52 Therefore, we chose the timepoint at 24 hours after reperfusion for the induction of CIAKI.In this regard, our model is restricted to the investigation ofthose compounds and knockout models that do not influenceIRI when applied 24 hours after reperfusion. For any agentthat is investigated in thismodel, it needs to be assured that IRIby 48 hours is not significantly different from the PBS controlat 48 hours. For instance, blockade of Fas ligand by the mAbMFL3 did exert protection from IRI when applied 24 hoursafter reperfusion (data not shown) with the consequence that





our model does not allow evaluation of the effectiveness of theblockade of this death receptor in CIAKI and RIP3-deficientmice could not be investigated due to the protection from IRIcompared with wild-type mice (data not shown). Nec-1, how-ever, did not influence IRI when applied later than 30 minutesafter the onset of reperfusion and certainly not when applied24 hours from the onset of reperfusion.25

Mechanistically, our study suggests a functional role of theRIP1 kinase domain that is not associated with cell death inCIAKI based on three lines of evidence. First, Nec-1 blocksCIAKI in our model. Second, the inactive Nec-1 derivateNec-1i25,28 does not affect this model, ruling out off-targeteffects of this compound. Third, the RIP1 expression pattern,which is unaffected by IRI (as reported earlier25), substantiallychanges after applications of RCM (Figure 5, A and B). RIP1was described to be involved in both apoptosis and necroptosis,but also in activating the NF-kB pathway after TNFR1 ligationby TNFa.53 We are only beginning to understand non-celldeath functions of RIP1 that can be prevented by Nec-1.54 Itis of interest in this regard that the histologic pattern of osmoticnephrosis is not restricted to CIAKI, but has also been de-scribed in severe sepsis.55–57 RIP1 is also critically involved insepsis as recently demonstrated in a model of TNFa-inducedshock,58 although in this model, we could not detect osmoticnephrosis due to the early death of the mice within 24 hoursafter TNFa application.32 Currently, we cannot exclude a com-mon underlying role for RIP1 in the pathogenesis of osmoticnephrosis in both CIAKI and sepsis because RIP1-deficientmice die perinatally.59 Organ-specific floxed RIP1-deficientmice will be required to further investigate the role of RIP-1in bothmodels. Unlike in the sepsis model, Nec-1 prevented allsigns of CIAKI almost completely. Besides hydration and ap-plication of N-acetylcysteine, both of which did not influenceour model (Supplemental Figure 11), and the recently sugges-ted application of atorvastatin,60 blocking the RIP1 kinase ac-tivity may be considered a new strategy to prevent CIAKI.

In summary,we conclude that living renal cells are put into anonfunctioning state by RCM and that this functional in-hibition, rather than cell death, is prevented by Nec-1,suggesting a role for the kinase domain of RIP1 in this process.With respect to the fact that probably no other kinase inhibitorwas examined in this much detail for its selectivity,26–28 weconsider it justified to conclude that the RIP1 kinase domainessentially mediates the mentioned effects without the needfor its default partner RIP3. These results should focus ourefforts on the inhibition of the RIP1 kinase domain in thisscenario rather than on the prevention of cell death. Finally,the possibility that Nec-1 and presumably other RIP1 kinaseinhibitors can strongly alter peritubular perfusion might be ofrelevance beyond CIAKI.

CONCISE METHODS

Complete methods are available in the Supplemental Material.

ACKNOWLEDGMENTS

We thank Vishva Dixit for the RIP3-deficient mice, Elsa Bello-Reuss

for the TKPTS cells, and Harald Schöcklmann for the MMC and

glEND cells. We also thank Silvia Iversen, Mareike Newsky, Alina

Pape, and Katja Bruch (Kiel) and Nancy F. Roeser (Ann Arbor) for

excellent technical support.

This work was funded by grants from the German Society for

Nephrology (to A.L.); Novartis PharmaGmbH,Germany (toA.L. and

S.K.); Fresenius Medical Care, Germany (to S.K. and U.K.); Else

Kröner-Fresenius Stiftung (to U.K.); INTERREG 4A, South Denmark-

Schlewig- K.E.R.N. program no.: 62-1.2-10, J-no.: 10/13123 (to U.K.);

and the National Institutes of Health (NIH-DK34275 to J.M.W.).

A.J.S. was funded by a grant from TAMOP (4.2.2/B-10/1-2010-0013).

This work is part of the thesis of J.O.H.

DISCLOSURESNone.

REFERENCES

1. Calvin AD, Misra S, Pflueger A: Contrast-induced acute kidney injuryand diabetic nephropathy. Nat Rev Nephrol 6: 679–688, 2010

2. McCullough PA: Radiocontrast-induced acute kidney injury. Nephron,

Physiol 109: 61–72, 20083. Weisbord SD, Mor MK, Resnick AL, Hartwig KC, Sonel AF, Fine MJ,

Palevsky PM: Prevention, incidence, and outcomes of contrast-inducedacute kidney injury. Arch Intern Med 168: 1325–1332, 2008

4. Maioli M, Toso A, Leoncini M, Gallopin M, Musilli N, Bellandi F: Per-sistent renal damage after contrast-induced acute kidney injury: In-cidence, evolution, risk factors, and prognosis. Circulation 125: 3099–3107, 2012

5. BudanoC, LevisM,D’AmicoM,Usmiani T, FavaA, Sbarra P, BurdeseM,Segoloni GP, Colombo A, Marra S: Impact of contrast-induced acutekidney injury definition on clinical outcomes.AmHeart J 161: 963–971,2011

6. Solomon R: Contrast-induced acute kidney injury (CIAKI). Radiol ClinNorth Am 47: 783–788, v, 2009

7. Weisbord SD, Palevsky PM: Contrast-induced acute kidney injury:Short- and long-term implications. Semin Nephrol 31: 300–309, 2011

8. Rudnick M, Feldman H: Contrast-induced nephropathy: What arethe true clinical consequences? Clin J Am Soc Nephrol 3: 263–272,2008

9. RudnickMR, Berns JS, Cohen RM,Goldfarb S: Contrastmedia-associatednephrotoxicity. Curr Opin Nephrol Hypertens 5: 127–133, 1996

10. Nash K, Hafeez A, Hou S: Hospital-acquired renal insufficiency. Am J

Kidney Dis 39: 930–936, 200211. Vandenabeele P, Galluzzi L, Vanden Berghe T, Kroemer G: Molecular

mechanisms of necroptosis: An ordered cellular explosion.Nat RevMol

Cell Biol 11: 700–714, 201012. Zhao Y, Tao Z, Xu Z, Tao Z, Chen B, Wang L, Li C, Chen L, Jia Q, Jia E,

Zhu T, Yang Z: Toxic effects of a high dose of non-ionic iodinatedcontrast media on renal glomerular and aortic endothelial cells in agedrats in vivo. Toxicol Lett 202: 253–260, 2011

13. Quintavalle C, Brenca M, DeMicco F, Fiore D, Romano S, RomanoMF,Apone F, Bianco A, Zabatta MA, Troncone G, Briguori C, Condorelli G:In vivo and in vitro assessment of pathways involved in contrast media-induced renal cells apoptosis. Cell Death Dis 2: e155, 2011





14. Romano G, Briguori C, Quintavalle C, Zanca C, Rivera NV, Colombo A,Condorelli G: Contrast agents and renal cell apoptosis. Eur Heart J 29:2569–2576, 2008

15. Yokomaku Y, Sugimoto T, Kume S, Araki S, Isshiki K, Chin-Kanasaki M,Sakaguchi M, Nitta N, Haneda M, Koya D, Uzu T, Kashiwagi A: Asia-loerythropoietin prevents contrast-induced nephropathy. J Am Soc

Nephrol 19: 321–328, 200816. Boatright KM, Salvesen GS: Mechanisms of caspase activation. Curr

Opin Cell Biol 15: 725–731, 200317. Pop C, Salvesen GS: Human caspases: Activation, specificity, and

regulation. J Biol Chem 284: 21777–21781, 200918. Kroemer G, Galluzzi L, Brenner C: Mitochondrial membrane per-

meabilization in cell death. Physiol Rev 87: 99–163, 200719. Devalaraja-Narashimha K, Padanilam BJ: PARP-1 inhibits glycolysis in

ischemic kidneys. J Am Soc Nephrol 20: 95–103, 200920. Linkermann A, De Zen F, Weinberg J, Kunzendorf U, Krautwald S:

Programmed necrosis in acute kidney injury. Nephrol Dial Transplant

27: 3412–3419, 201221. Challa S, Chan FK: Going up in flames: Necrotic cell injury and in-

flammatory diseases. Cell Mol Life Sci 67: 3241–3253, 201022. Christofferson DE, Yuan J: Necroptosis as an alternative form of pro-

grammed cell death. Curr Opin Cell Biol 22: 263–268, 201023. Declercq W, Vanden Berghe T, Vandenabeele P: RIP kinases at the

crossroads of cell death and survival. Cell 138: 229–232, 200924. Linkermann A, Himmerkus N, Rölver L, Keyser KA, Steen P, Bräsen JH,

BleichM, Kunzendorf U, Krautwald S: Renal tubular Fas ligandmediatesfratricide in cisplatin-induced acute kidney failure. Kidney Int 79: 169–178, 2011

25. Linkermann A, Bräsen JH, Himmerkus N, Liu S, Huber TB, KunzendorfU, Krautwald S: Rip1 (receptor-interacting protein kinase 1) mediatesnecroptosis and contributes to renal ischemia/reperfusion injury. Kid-ney Int 81: 751–761, 2012

26. Biton S, Ashkenazi A: NEMO and RIP1 control cell fate in responseto extensive DNA damage via TNF-a feedforward signaling. Cell 145:92–103, 2011

27. Degterev A, Huang Z, BoyceM, Li Y, Jagtap P, Mizushima N, Cuny GD,Mitchison TJ, Moskowitz MA, Yuan J: Chemical inhibitor of non-apoptotic cell deathwith therapeutic potential for ischemic brain injury.Nat Chem Biol 1: 112–119, 2005

28. Degterev A, Hitomi J, Germscheid M, Ch’en IL, Korkina O, Teng X,Abbott D, Cuny GD, Yuan C, Wagner G, Hedrick SM, Gerber SA,Lugovskoy A, Yuan J: Identification of RIP1 kinase as a specific cellulartarget of necrostatins. Nat Chem Biol 4: 313–321, 2008

29. KaiserWJ, Upton JW, LongAB, Livingston-Rosanoff D, Daley-Bauer LP,Hakem R, Caspary T, Mocarski ES: RIP3 mediates the embryonic le-thality of caspase-8-deficient mice. Nature 471: 368–372, 2011

30. Billings FT 4th, Chen SW, KimM, Park SW, Song JH,Wang S, Herman J,D’Agati V, Lee HT: alpha2-Adrenergic agonists protect against radio-contrast-induced nephropathy in mice.Am J Physiol Renal Physiol 295:F741–F748, 2008

31. Krautwald S, Ziegler E, Rölver L, Linkermann A, Keyser KA, Steen P,Wollert KC, Korf-Klingebiel M, Kunzendorf U: Effective blockage ofboth the extrinsic and intrinsic pathways of apoptosis in mice by TAT-crmA. J Biol Chem 285: 19997–20005, 2010

32. Linkermann A, Bräsen JH, De Zen F, Weinlich R, Schwendener RA,Green DR, Kunzendorf U, Krautwald S: Dichotomy between RIP1- andRIP3-mediated necroptosis in tumor necrosis factor-a-induced shock.Mol Med 18: 577–586, 2012

33. Heinrich MC: Osmotic nephrosis and contrast media. Am J Kidney Dis

52: 629–630, author reply 629–630, 200834. DickenmannM,Oettl T,MihatschMJ:Osmotic nephrosis: Acute kidney

injury with accumulation of proximal tubular lysosomes due to admin-istration of exogenous solutes. Am J Kidney Dis 51: 491–503, 2008

35. Moreau JF, Droz D, Sabto J, Jungers P, Kleinknecht D, Hinglais N,Michel JR: Osmotic nephrosis induced by water-soluble triiodinated

contrast media in man. A retrospective study of 47 cases. Radiology115: 329–336, 1975

36. Lee HT, Jan M, Bae SC, Joo JD, Goubaeva FR, Yang J, Kim M: A1adenosine receptor knockout mice are protected against acute radio-contrast nephropathy in vivo. Am J Physiol Renal Physiol 290: F1367–F1375, 2006

37. Devalaraja-Narashimha K, Diener AM, Padanilam BJ: Cyclophilin Dgene ablation protects mice from ischemic renal injury. Am J Physiol

Renal Physiol 297: F749–F759, 200938. Sanz AB, Sanchez-NiñoMD,Ortiz A: TWEAK, amultifunctional cytokine

in kidney injury. Kidney Int 80: 708–718, 201139. Galluzzi L, Vitale I, Abrams JM, Alnemri ES, Baehrecke EH,

Blagosklonny MV, Dawson TM, Dawson VL, El-Deiry WS, Fulda S,Gottlieb E, Green DR, Hengartner MO, Kepp O, Knight RA, Kumar S,Lipton SA, Lu X, Madeo F, Malorni W, Mehlen P, Nuñez G, Peter ME,Piacentini M, Rubinsztein DC, Shi Y, Simon HU, Vandenabeele P,White E, Yuan J, Zhivotovsky B, Melino G, Kroemer G: Moleculardefinitions of cell death subroutines: Recommendations of the No-menclature Committee on Cell Death 2012.Cell Death Differ 19: 107–120, 2012

40. Moquin D, Chan FK: Themolecular regulation of programmed necroticcell injury. Trends Biochem Sci 35: 434–441, 2010

41. Oberst A, Green DR: It cuts both ways: Reconciling the dual roles ofcaspase 8 in cell death and survival.Nat RevMol Cell Biol 12: 757–763,2011

42. Dunn KW, Sandoval RM, Kelly KJ, Dagher PC, Tanner GA, Atkinson SJ,Bacallao RL, Molitoris BA: Functional studies of the kidney of livinganimals using multicolor two-photon microscopy. Am J Physiol Cell

Physiol 283: C905–C916, 200243. Peti-Peterdi J, Burford JL, Hackl MJ: The first decade of using multi-

photonmicroscopy for high-power kidney imaging.Am J Physiol Renal

Physiol 302: F227–F233, 201244. Porter GA: Experimental contrast-associated nephropathy and its

clinical implications. Am J Cardiol 66: 18F–22F, 199045. Sendeski MM, Persson AB, Liu ZZ, Busch JF, Weikert S, Persson PB,

Hippenstiel S, Patzak A: Iodinated contrast media cause endothelialdamage leading to vasoconstriction of human and rat vasa recta. Am J

Physiol Renal Physiol 303: F1592–F1598, 201246. Galluzzi L, Aaronson SA, Abrams J, Alnemri ES, Andrews DW,

Baehrecke EH, Bazan NG, Blagosklonny MV, Blomgren K, Borner C,Bredesen DE, Brenner C, Castedo M, Cidlowski JA, Ciechanover A,Cohen GM, De Laurenzi V, De Maria R, Deshmukh M, Dynlacht BD, El-Deiry WS, Flavell RA, Fulda S, Garrido C, Golstein P, Gougeon ML,GreenDR,GronemeyerH, HajnóczkyG,Hardwick JM,HengartnerMO,Ichijo H, Jäättelä M, Kepp O, Kimchi A, Klionsky DJ, Knight RA,Kornbluth S, Kumar S, Levine B, Lipton SA, Lugli E,Madeo F,MalomiW,Marine JC, Martin SJ, Medema JP, Mehlen P, Melino G, Moll UM,Morselli E, Nagata S, Nicholson DW, Nicotera P, Nuñez G, Oren M,Penninger J, Pervaiz S, Peter ME, Piacentini M, Prehn JH, Puthalakath H,Rabinovich GA, Rizzuto R, Rodrigues CM, Rubinsztein DC, Rudel T,Scorrano L,SimonHU, StellerH,TschoppJ, TsujimotoY, Vandenabeele P,Vitale I, Vousden KH, Youle RJ, Yuan J, Zhivotovsky B, Kroemer G:Guidelines for the use and interpretation of assays for monitoringcell death in higher eukaryotes. Cell Death Differ 16: 1093–1107,2009

47. Kroemer G, Galluzzi L, Vandenabeele P, Abrams J, Alnemri ES,Baehrecke EH, Blagosklonny MV, El-Deiry WS, Golstein P, Green DR,Hengartner M, Knight RA, Kumar S, Lipton SA, Malorni W, Nuñez G,Peter ME, Tschopp J, Yuan J, Piacentini M, Zhivotovsky B, Melino G;Nomenclature Committee on Cell Death 2009: Classification of celldeath: Recommendations of the Nomenclature Committee on CellDeath 2009. Cell Death Differ 16: 3–11, 2009

48. Chekeni FB, Elliott MR, Sandilos JK, Walk SF, Kinchen JM, LazarowskiER, Armstrong AJ, Penuela S, Laird DW, Salvesen GS, Isakson BE,Bayliss DA, Ravichandran KS: Pannexin 1 channels mediate ‘find-me’



signal release and membrane permeability during apoptosis. Nature467: 863–867, 2010

49. McStay GP, Salvesen GS, Green DR: Overlapping cleavage motif se-lectivity of caspases: Implications for analysis of apoptotic pathways.Cell Death Differ 15: 322–331, 2008

50. Kinsey GR, Sharma R, Huang L, Li L, Vergis AL, Ye H, Ju ST, Okusa MD:Regulatory T cells suppress innate immunity in kidney ischemia-reperfusion injury. J Am Soc Nephrol 20: 1744–1753, 2009

51. Ko GJ, Jang HR, Huang Y, Womer KL, Liu M, Higbee E, Xiao Z, Yagita H,RacusenL,HamadAR,RabbH:BlockingFas ligandon leukocytesattenuateskidney ischemia-reperfusion injury. J Am Soc Nephrol 22: 732–742, 2011

52. Li L, Huang L, Ye H, Song SP, Bajwa A, Lee SJ, Moser EK, Jaworska K,Kinsey GR, Day YJ, Linden J, Lobo PI, Rosin DL, Okusa MD: Dendriticcells tolerized with adenosine A2AR agonist attenuate acute kidneyinjury. J Clin Invest 122: 3931–3942, 2012

53. Festjens N, Vanden Berghe T, Cornelis S, Vandenabeele P: RIP1, a ki-nase on the crossroads of a cell’s decision to live or die. Cell DeathDiffer 14: 400–410, 2007

54. Kang TB, Yang SH, Toth B, Kovalenko A, Wallach D: Caspase-8 blockskinase RIPK3-mediated activation of the NLRP3 inflammasome. Im-munity 38: 27–40, 2013

55. Doi K, Leelahavanichkul A, Yuen PS, Star RA: Animal models of sepsisand sepsis-induced kidney injury. J Clin Invest 119: 2868–2878, 2009

56. Ebcioglu Z, Cohen DJ, Crew RJ, Hardy MA, Ratner LE, D’Agati VD,MarkowitzGS:Osmotic nephrosis in a renal transplant recipient.KidneyInt 70: 1873–1876, 2006

57. Venkatachalam MA, Weinberg JM: The tubule pathology of septicacute kidney injury: A neglected area of research comes of age. KidneyInt 81: 338–340, 2012

58. Duprez L, Takahashi N, Van Hauwermeiren F, Vandendriessche B,Goossens V, Vanden Berghe T, Declercq W, Libert C, Cauwels A,Vandenabeele P: RIP kinase-dependent necrosis drives lethal systemicinflammatory response syndrome. Immunity 35: 908–918, 2011

59. Zhang H, Zhou X, McQuade T, Li J, Chan FK, Zhang J: Functionalcomplementation between FADD and RIP1 in embryos and lympho-cytes. Nature 471: 373–376, 2011

60. Quintavalle C, Fiore D, De Micco F, Visconti G, Focaccio A, Golia B,Ricciardelli B, Donnarumma E, Bianco A, Zabatta MA, Troncone G,Colombo A, Briguori C, Condorelli G: Impact of a high loading dose ofatorvastatin on contrast-induced acute kidney injury. Circulation 126:3008–3016, 2012

This article contains supplemental material online at http://jasn.asnjournals.org/lookup/suppl/doi:10.1681/ASN.2012121169/-/DCSupplemental.





the rip1-kinase inhibitor necrostatin-1 prevents osmotic...

Documents