pretreatment with epo reduces the injury and dysfunction caused by ischemia/reperfusion in the mouse...
TRANSCRIPT
Kidney International, Vol. 66 (2004), pp. 983–989
HORMONES – CYTOKINES – SIGNALING
Pretreatment with EPO reduces the injury and dysfunctioncaused by ischemia/reperfusion in the mouse kidney in vivo
NIMESH S.A. PATEL, EDWARD J. SHARPLES, SALVATORE CUZZOCREA, PRABAL K. CHATTERJEE,DOMENICO BRITTI, MUHAMMAD M. YAQOOB, and CHRISTOPH THIEMERMANN
Centre for Experimental Medicine, Nephrology & Critical Care, William Harvey Research Institute, Queen Mary - University ofLondon, London, United Kingdom; Department of Clinical and Experimental Medicine and Pharmacology, University ofMessina,Messina, Italy; Department of Pharmacology, School of Pharmacy and Biomolecular Sciences, University of Brighton,Brighton, United Kingdom; and Department of Veterinary and Agricultural Science, University of Teramo, Teramo Italy
Pretreatment with EPO reduces the injury and dysfunctioncaused by ischemia/reperfusion in the mouse kidney in vivo.
Background. Erythropoietin (EPO) is a potent stimulatorof erythroid progenitor cells and is known to be up-regulatedduring states of hypoxia. Here we investigate the effects of renalischemia/reperfusion (I/R) on the degree of renal dysfunctionand injury with recombinant human EPO in mice when given aseither a 3-day pretreatment, or upon reperfusion of the kidney.
Methods. Mice were treated with EPO (1000 IU/kg/day sub-cutaneously) for 3 days, or treated with EPO (1000 IU/kg sub-cutaneously) upon reperfusion, and subsequently subjected tobilateral renal artery occlusion (30 minutes) and reperfusion(24 hours). At the end of experiments, the following indica-tors and markers of renal injury and dysfunction were mea-sured: plasma urea, creatinine, and aspartate aminotransferase(AST), tissue myeloperoxidase (MPO) activity [for polymor-phonuclear leukocyte (PMN) infiltration], and tissue malon-aldehyde (MDA) levels (for tissue lipid peroxidation). Kidneyswere used for histologic evaluation of renal injury.
Results. EPO was able to significantly attenuate the renaldysfunction and injury associated with I/R, as well as the tissueinjury. The increase in renal MPO activity and, hence, the degreeof PMN infiltration were also significantly reduced in EPO-treated mice. In addition, lipid peroxidation as a result of renalI/R injury was also attenuated in EPO-treated mice.
Conclusion. The protection afforded by the pretreatmentregime of EPO was greater than that of administering EPOas a single bolus upon reperfusion. We propose that differentmechanisms underlie the protective effects seen with EPO whengiven as either a daily pretreatment or as a single bolus, whichneed to be further investigated.
Erythropoietin (EPO) is a glycoprotein hormone pro-duced primarily by the adult kidney in the regulation
Key words: kidney, reperfusion injury, erythropoietin, mouse.
Received for publication December 5, 2003and in revised form February 23, 2004, and April 1, 2004Accepted for publication April 16, 2004
C© 2004 by the International Society of Nephrology
of red blood cell production, exerting its hematopoi-etic effects by stimulating the proliferation of committederythroid progenitor cells and their development into ma-ture erythrocytes [1]. There are several recent reportsdocumenting that EPO reduces the injury caused by is-chemia/reperfusion (I/R) of the brain [2, 3], eye [4], gut[5], heart [6, 7], and kidney [8]. It is, however, less clearwhether EPO has to be given as a pretreatment or can begiven upon reperfusion of a previously ischemic tissue.For instance, in the heart pretreatment of rats with EPO(5000 IU/kg) at 24 hours and 0.5 hour prior to ischemiareduces tissue injury and cardiac dysfunction, while theacute administration of EPO (5000 IU/kg) on reperfusionafter coronary artery occlusion also reduces infarct size inthe rat [6]. Hitherto, there have been no attempts to com-pare the efficacy of pretreatment of animals with EPO toan acute administration of EPO either before the onsetof the ischemic insult or before the onset of reperfusion.This is important, as the mechanisms underlying the pro-tective effects of EPO in these two therapeutic regimensmay well be different. For instance, pretreatment of micewith EPO (1000 IU/kg) for 3 days mobilizes endothelialprogenitor cells [9], and procedures which mobilize pro-genitors have been shown to protect the heart [10] andthe hind limb [9] against ischemic injury.
Thus, this study was designed to compare the effectsof EPO on the renal dysfunction and injury caused byI/R of the kidney of the mouse in vivo when given ei-ther as a pretreatment (1000 IU/kg subcutaneously for3 days) and when given in the same dose 5 minutes priorto reperfusion of the previously ischemic kidney.
METHODS
Animals
Fifty-nine male C57BL/6J mice (Charles River, Milan,Italy) weighing 25 to 30 g were used to assess therole of recombinant human EPO (St. Bartholomew’s
983
984 Patel et al: EPO attenuates renal I/R injury in the mouse
Hospital Pharmacy) when administered acutely andchronically in the pathogenesis of renal I/R in the mouse.Mice were allowed access to food and water ad libitumand were cared for in compliance with Italian regulationson protection of animals used for experimental and otherscientific purposes (D.M. 116192), as well as with theEuropean Economic Community regulations (O.J. ofE.C. L358/1 12/18/1986).
Mice were divided into the following four groups inexperiments involving the acute administration of EPO:(1) I/R group in which control mice underwent renalischemia for 30 minutes followed by reperfusion for24 hours (N = 11); (2) I/R EPO prereperfusion groupincluded mice that were administered EPO (1000 IU/kgsubcutaneously bolus) 5 minutes prior to reperfusion(N = 10); (3) sham group were sham-operated mice,which were subjected to the surgical procedures de-scribed above, but were not subjected to renal I/R (N =4);and (4) sham EPO prereperfusion group were mice thatwere treated identical to sham mice except for the admin-istration of EPO (1000 IU/kg subcutaneously) 5 minutesprior to sham reperfusion (N = 4).
In the above experiments, the route of administrationand dose of EPO were based on that of a previously re-ported experiment in the mouse [9]. Mice, which did notreceive EPO, were administered 8 mL/kg saline (vehi-cle for EPO) at equivalent time points (5 minutes priorto reperfusion). In experiments involving the chronic ad-ministration of EPO, mice were divided into the follow-ing four groups: (1) I/R group included control mice,which underwent renal ischemia for 30 minutes followedby reperfusion for 24 hours (N = 6); (2) I/R EPO pre-treatment group mice were administered EPO (1000IU/kg/day subcutaneously) for 3 days prior to I/R (N =6); (3) sham group were sham-operated mice, which weresubjected to the surgical procedures described above, butwere not subjected to renal I/R (N = 4); and (4) shamEPO pretreatment-group were mice that were treatedidentical to sham mice except for the administration ofEPO (1000 IU/kg/day subcutaneously) for 3 days prior tosham I/R (N = 4).
Mice, which did not receive EPO, were administered8 mL/kg/day saline (vehicle for EPO) at equivalent timepoints (72, 48, and 24 hours before operation).
Renal I/R model
Mice were anesthetized using chloral hydrate (125 mg/kg, intraperitoneally) and core body temperature main-tained at 37◦C using a homoeothermic blanket. Mice weremaintained under anesthesia for the duration of ischemia(i.e., 30 minutes). After performing a midline laparotomy,mice from the I/R groups were subjected to bilateral re-nal ischemia for 30 minutes, during which the renal arter-ies and veins were occluded using microaneurysm clamps
[11]. The time of ischemia chosen was based on that foundto maximize reproducibility of renal functional impair-ment, while minimizing mortality in these animals [11].After the renal clamps were removed, the kidneys wereobserved for a further 5 minutes to ensure reflow afterwhich 1 mL saline at 37◦C was injected into the abdomenand the incision was sutured in two layers. Mice werethen returned to their cages where they were allowedto recover from anesthesia and observed for 24 hours.Sham-operated mice underwent identical surgical pro-cedures to I/R mice except that microaneurysm clampswere not applied.
Measurement of biochemical parameters
At the end of the reperfusion period, 1 mL blood sam-ples were collected from anesthetized mice via cardiacpuncture. The samples were centrifuged (6000 rpm for3 minutes) to separate plasma. All plasma samples wereanalyzed for biochemical parameters within 24 hours af-ter collection. Plasma urea and creatinine concentrationswere used as indicators of renal (glomerular) function[11]. The rise in the plasma levels of aspartate aminotrans-ferase (AST), an enzyme located in the proximal tubule,was used as an indicator of reperfusion injury [11].
Histologic evaluation
Kidneys were removed from mice at the end of the ex-perimental period after tying the renal pedicle, and cutin a sagittal section into two halves. These tissue sam-ples were fixed by immersion in 10% (wt/vol) formalde-hyde in phosphate-buffered saline (PBS) (0.01 mol/L,pH 7.4) at room temperature for 1 day. After dehydra-tion using graded ethanol, the tissue was embedded inParaplast (Sherwood Medical, Mahwah, NJ, USA) andcut in fine (8 lm) sections and mounted on glass slides.Sections were then deparaffinized with xylene, counter-stained with hematoxylin and eosin, and viewed under alight microscope (Dialux 22) (Leitz, Milan, Italy).
Myeloperoxidase activity
Myeloperoxidase (MPO) activity in kidney sampleswas determined as an index of polymorphonuclear(PMN) leukocyte accumulation, as previously described[11]. Kidneys were homogenized in a solution contain-ing 0.5% hexa-decyl-trimethyl-ammonium bromide and10 mmol/L 3-(N-morpholino)-propane-sulfonic acid dis-solved in 80 mmol/L sodium phosphate buffer (pH 7), andcentrifuged for 30 minutes at 20,000g at 4◦C. An aliquot ofthe supernatant was then allowed to react with a solutionof tetra-methyl-benzidine (16 mmol/L) and 1 mmol/L hy-drogen peroxide. The rate of change in absorbance wasmeasured by a spectrophotometer at 650 nm. MPO ac-tivity was defined as the quantity of enzyme degrading
Patel et al: EPO attenuates renal I/R injury in the mouse 985
1 lmol peroxide/min at 37◦C and was expressed in unitsper gram weight of wet tissue.
Malondialdehyde measurement
Malondialdehyde (MDA) levels in kidney sampleswere determined as an indicator of lipid peroxidation, aspreviously described [11]. Tissues were homogenized ina 1.15% KCl solution. An aliquot of the homogenate wasadded to a reaction mixture containing 200 lL of 8.1%sodium dodecyl sulfate (SDS), 1500 lL of 20% aceticacid (pH 3.5), 1500 lL of 0.8% thiobarbituric acid, and700 lL of distilled water. The mixture was then boiledfor 1 hour at 95◦C and centrifuged at 3000g for 10 min-utes. The absorbency of the supernatant was measuredby spectrophotometry at 550 nm.
Materials
Unless otherwise stated, all compounds used in thisstudy were purchased from Sigma-Aldrich Company Ltd.(Milan, Italy). All solutions used in vivo were preparedusing non-pyrogenic saline [0.9% (wt/vol) NaCl] (BaxterHealthcare Ltd., Thetford, Norfolk, UK).
Statistical analysis
All values described in the text and figures are ex-pressed as mean ± standard error of the mean (SEM) forN observations. Each data point represents biochemicalmeasurements obtained from four to 13 separate animals.For histologic scoring, each data point represents analy-sis of kidneys taken from four to 13 individual animals.One-way analysis of variance (ANOVA) with Dunnett’spost test was performed using GraphPad Prism version4.00 for Windows (GraphPad Software, San Diego, CA,USA, www.graphpad.com) and a P value of less than 0.05was considered to be significant.
RESULTS
Effect of EPO on renal dysfunction caused by I/R: Plasmaurea and creatinine
When compared to sham-operated mice, I/R caused asignificant increase in the plasma levels of urea and cre-atinine in control mice (Fig. 1), suggesting a significantdegree of renal dysfunction. Administration of EPO, ei-ther acutely (1000 IU/kg prior to reperfusion) (Fig. 1Aand B) or chronically (1000 IU/kg/day for 3 days priorto I/R) (Fig. 1C and D), attenuated the renal dysfunctioncaused by I/R in control mice. It should be noted thatthe protection afforded by the chronic administration ofEPO was more pronounced.
Effect of EPO on renal dysfunction caused by I/R:Plasma AST
When compared to sham-operated mice, I/R caused asignificant increase in the plasma levels of AST in controlmice, suggesting significant reperfusion injury (Fig. 2).Administration of EPO, either acutely (1000 IU/kg priorto reperfusion) (Fig. 2A) or chronically (1000 IU/kg/dayprior to I/R) (Fig. 2B), attenuated the renal injury causedby I/R in control mice. It should be noted that the protec-tion afforded by the chronic administration of EPO wasmore pronounced.
Effect of EPO on renal injury caused by I/R:Histologic assessment
When compared to sham-operated mice (Fig. 3A), his-tological examination of kidneys obtained from controlmice subjected to I/R demonstrated a significant degreeof renal injury (Fig. 3B). Specifically, kidneys obtainedfrom these animals exhibited degeneration of tubularstructure, tubular dilatation, swelling and necrosis, lu-minal congestion, and eosinophilia. In contrast, renalsections obtained from mice treated with EPO acutely(1000 IU/kg prior to reperfusion) (Fig. 3C) andchronically (1000 IU/kg/day prior to I/R) (Fig. 3D)demonstrated a marked reduction in the severity of thesehistologic features of renal injury, when compared withkidneys obtained from control mice subjected to I/R only(Fig. 3B).
Effect of EPO on renal inflammation caused by I/R: MPOactivity and MDA levels
When compared to sham-operated mice, the kidneysobtained from control mice subjected to I/R demon-strated a significant increase in MPO activity (Fig. 4Aand C), suggesting increased PMN infiltration into renaltissues. The increase in the tissue level of MPO seen inmice administered EPO, either acutely (1000 IU/kg priorto reperfusion) (Fig. 4A) or chronically (1000 IU/kg/dayprior to I/R) (Fig. 4C) was significantly smaller than thoseseen in control mice subjected to I/R, which in the case ofchronic EPO treatment were similar to values obtainedfrom sham-operated mice. It should be noted that thechronic administration of EPO reduced the MPO levelsto levels which were not different from those observed insham-operated animals.
When compared to sham-operated mice, the kidneysobtained from control mice subjected to I/R demon-strated a significant increase in MDA levels (Fig. 4B andD), suggesting increased lipid peroxidation in renal tis-sues. The increase in the tissue level of MDA seen inmice administered EPO, either acutely (1000 IU/kg priorto reperfusion) (Fig. 4B) or chronically (1000 IU/kg/dayprior to I/R) (Fig. 4D) was significantly smaller than those
986 Patel et al: EPO attenuates renal I/R injury in the mouse
0
100
200
300
400
Pla
sma
urea
, mg/
dL
Sham
Sham
EPO
prer
eper
fusio
n I/R
I/R E
PO
prer
eper
fusio
n
A
0.0
0.5
1.0
1.5
2.0
Pla
sma
crea
tinin
e, m
g/dL
Sham
Sham
EPO
prer
eper
fusio
n I/R
I/R E
PO
prer
eper
fusio
n
B
0
100
200
300
400
Pla
sma
urea
, mg/
dL
Sham
Sham
EPO
pret
reat
men
t I/R
I/R E
PO
pret
reat
men
t
C
0.0
0.5
1.0
1.5
2.0
Pla
sma
crea
tinin
e, m
g/dL
Sham
Sham
EPO
pret
reat
men
t I/R
I/R E
PO
pret
reat
men
t
D Fig. 1. Effect of erythropoietin (EPO) onrenal function. Plasma urea (A) and creati-nine levels (B) were measured, from miceadministered EPO prior to reperfusion, asbiochemical markers of renal dysfunction sub-sequent to sham operation (Sham) (N =4); Sham EPO prereperfusion (1000 IU/kgprior to reperfusion) (N = 4); or renal is-chemia/reperfusion (I/R) (N = 11); I/R EPOprereperfusion (1000 IU/kg prior to reperfu-sion) (N = 10). Plasma urea (C) and crea-tinine levels (D) were measured, from miceadministered EPO as a 3-day pretreatment,as biochemical markers of renal dysfunctionsubsequent to sham operation (Sham) (N =4); Sham EPO pretreatment (1000 IU/kg/dayprior to I/R) (N = 4); or renal I/R (N = 6);I/R EPO pretreatment (1000 IU/kg/day priorto I/R) (N = 6). Data represent mean ± SEMfor number of observations, ∗P < 0.05 vs. renalI/R group.
0
250
500
750
1000
1250
1500
1750
Pla
sma
AS
T, IU
/L
Sham
Sham
EPO
prer
eper
fusio
n I/R
I/R E
PO
prer
eper
fusio
n
A
0
250
500
750
1000
1250
1500
1750
Pla
sma
AS
T, IU
/L
Sham
Sham
EPO
pret
reat
men
t I/R
I/R E
PO
pret
reat
men
t
BFig. 2. Effect of erythropoietin (EPO) onrenal reperfusion-injury. Plasma aspartateaminotransferase (AST) (A) levels were mea-sured, from mice administered EPO prior toreperfusion, as a biochemical marker of renalinjury subsequent to sham operation (Sham)(N = 4); Sham EPO prereperfusion (1000IU/kg prior to reperfusion) (N = 4); or renalischemia/reperfusion (I/R) (N = 11); I/R EPOprereperfusion (1000 IU/kg prior to reperfu-sion) (N = 10). Plasma AST (B) levels weremeasured, from mice administered EPO as a3-day pretreatment, as a biochemical markerof renal injury subsequent to sham-operation(Sham) (N = 4); Sham EPO pretreatment(1000 IU/kg/day prior to I/R) (N = 4); orrenal I/R (N = 6); I/R EPO pretreatment(1000 IU/kg/day prior to I/R) (N = 6). Datarepresent mean ± SEM for number of obser-vations, ∗P < 0.05 vs. renal I/R group.
Patel et al: EPO attenuates renal I/R injury in the mouse 987
A B
C D
Fig. 3. Effect of erythropoietin (EPO) on re-nal injury: Histologic examination. A renalsection taken from a sham-operated mouse(A) is compared with that taken from a mousesubjected to renal ischemia/reperfusion (I/R)(B). A renal section taken from a mouse sub-jected to renal I/R administered 1000 IU/kgEPO prior to reperfusion (C) is represented.Addtionally, a renal section taken from amouse subjected to renal I/R administered1000 IU/kg/day EPO prior to I/R to is also rep-resented (D) (hematoxylin and eosin, originalmagnification ×150). Figures are representa-tive of at least three experiments performedon different days (N = 4 to 11 for all groups).
Sham
Sham
EPO
prer
eper
fusio
n I/R
I/R E
PO
prer
eper
fusio
n
0
10
20
30
40
50
60
70
80
Mye
lope
roxi
dase
act
ivity
,µ/
g w
et ti
ssue
A
Sham
Sham
EPO
prer
eper
fusio
n I/R
I/R E
PO
prer
eper
fusio
n
05
1015202530354045
Mal
onal
dehy
de le
vels
,µ m
ol/m
g w
et ti
ssue
B
Sham
Sham
EPO
pret
reat
men
t I/R
I/R E
PO
pret
reat
men
t
0
10
20
30
40
50
60
70
80
Mye
lope
roxi
dase
act
ivity
,µ/
g w
et ti
ssue
C
Sham
Sham
EPO
pret
reat
men
t I/R
I/R E
PO
pret
reat
men
t
05
1015202530354045
Mal
onal
dehy
de le
vels
,µ m
ol/m
g w
et ti
ssue
D Fig. 4. Effect of erythropoietin (EPO) on re-nal inflammation. Myeloperoxidase (MPO)activity (A) and malonaldehyde (MDA) levels(B) were measured from mice administeredEPO prior to reperfusion, subsequent to shamoperation (Sham) (N = 4); Sham EPO priorto reperfusion (1000 IU/kg prior to reperfu-sion) (N = 4); or renal ischemia/reperfusion(I/R) (N = 11); I/R EPO prior to reperfu-sion (1000 IU/kg prior to reperfusion) (N =10). MPO activity (C) and MDA levels(D) were measured from mice administeredEPO as a 3-day pretreatment, subsequent tosham operation (Sham) (N = 4); Sham EPOpretreatment (1000 IU/kg/day prior to I/R)(N = 4) or renal I/R (N = 6); I/R EPOpretreatment (1000 IU/kg/day prior to I/R)(N = 6). Data represent mean ± SEM fornumber of observations, ∗P < 0.05 vs. renalI/R wild-type group.
988 Patel et al: EPO attenuates renal I/R injury in the mouse
seen in control mice subjected to I/R. The reduction inthe degree of lipid peroxidation afforded by EPO wassimilar when EPO was given as a pretreatment or givenupon reperfusion.
There were no differences in any of the above parame-ters measured between sham-operated EPO (1000 IU/kg)-treated mice, sham-operated EPO (1000 IU/kg/day)-treated mice or sham-operated mice (see Figs. 1, 2,and 4).
DISCUSSION
EPO, the critical hormone promoting the survivaland differentiation of erythroid progenitor cells, is cur-rently being used in the therapy of patients with chronicrenal failure suffering from anemia [12]. Here we demon-strate for the first time that EPO also reduces the re-nal dysfunction and injury caused by bilateral occlusion(30 minutes) and reperfusion (24 hours) in anesthetizedmice. Specifically, in mice subjected to renal I/R, EPO(1000 IU/kg given 5 minutes prior to reperfusion) signif-icantly attenuated the (1) renal dysfunction (increases inplasma creatinine and urea); (2) reperfusion injury (in-crease in plasma AST); (3) characteristic histologic signsof marked tubular injury; (4) PMN accumulation (MPOassay); and (5) lipid peroxidation (MDA assay). All thisdata confirmed a well-known pattern of renal dysfunctionand injury caused by I/R of the kidney [13–15]. Moreover,our findings are in agreement with the notion that renalI/R causes both renal and tubular dysfunction [16], as wellas secondary inflammation [17].
We also report that the pretreatment of mice with EPO(1000 IU/kg/day for 3 days) also reduced the renal in-jury, dysfunction, and inflammation caused by bilateralrenal artery occlusion in the mouse. We have previouslydemonstrated that EPO reduced both necrotic and apop-totic cell death in vitro and reduced renal injury in a ratmodel of I/R. Importantly, this study shows that EPO, incontrast to other candidate therapies for acute renal fail-ure, notably insulin-like growth factor-1 (IGF-1), is effec-tive in different species [18]. Most notably, we discoveredthat the degree of protection against injury, dysfunction,and inflammation afforded by the pretreatment regimenwith EPO was greater than the one afforded by the acuteadministration of EPO upon reperfusion. There are sev-eral potential explanations for this finding.
First, one could argue that the pretreatment proto-col has resulted in favorable systemic or renal hemody-namic effects. However, although acute administration ofEPO (in the rat) had no effect on blood pressure, corticaland pyramidal perfusion, the repetitive administration ofEPO for up to 1 week caused a small reduction in corticalblood flow [19]. This would argue against hemodynamicchanges affecting preinsult glomerular filtration rate, but
might contribute to delayed pre-conditioning (discussedbelow).
Second, as the half-life of EPO is ∼10 hours in therodent, it is possible and likely that the repetitive ad-ministration of 1000 IU/kg subcutaneously per day for3 days (pretreatment protocol) resulted in a highersteady-state plasma concentration, when compared tothe peak concentration achieved from a single dose ad-ministration prior to reperfusion (1000 IU/kg prior toreperfusion).
Third, pretreatment of mice with EPO for 3 daysmay well have resulted in the up-regulation of protec-tive genes, including endothelial nitric oxide synthase(eNOS), magnesium superoxide dismutase (Mn-SOD)and heat shock protein 70 (HSP70), which in turn couldhave contributed to the observed protective effect, mim-icking the effects of delayed preconditioning. Delayedischemic preconditioning is protective in animal mod-els of ischemic myocardial and brain injury, and ismediated by the up-regulation of multiple factors, includ-ing EPO, vascular endothelial growth factor and hemeoxygenase-1, which improve tissue performance in re-sponse to metabolic stress under the control of the oxy-gen sensor hypoxia-inducible factor-1a (HIF-1a). Theeffects of delayed preconditioning were lost in HIF-1a+/− mice [20]. The crucial role of the down stream ef-fects of EPO production in delayed preconditioning ofbrain injury has been demonstrated by the attenuation ofprotection observed with the use of a soluble EPO recep-tor [21], and in vitro, medium transfer from hypoxic andglucose-starved astrocytes to untreated neurons inducedprotection against hypoxia in neuronal cells, which wasattenuated strongly by the application of a soluble EPOreceptor and anti-EPO receptor antibodies [22]. In therat kidney, preconditioning with EPO 24 hours beforeischemia was associated with reduced injury, mediatedin part by the up-regulation of HSP70 and the antiapop-totic protein Bcl-2 [23]. The main difference is that unliketreatment at reperfusion, preconditioning with the pre-treatment protocol enables increased protein synthesisand not simply activation of existing proteins.
Fourth, the recent findings that bone marrow–derivedstem cells contribute to the regeneration of proximaltubule epithelial cells following I/R injury [24] also cre-ates an attractive hypothesis that the protection observedwith EPO pretreatment is partially due to an increase inthe mobilization and circulation of endothelial progen-itor cells and possibly bone marrow–derived stem cells.It has been shown in mice that EPO treatment causeda significant increase in the percentage of proliferatinglin−/sca-1+ stem cells in the bone marrow and mobiliza-tion of endothelial progenitor cells [9, 25].
There is evidence that the pretreatment protocol ofEPO used here results in enhanced levels of circulatingendothelial progenitor cells. Most notably, it has been
Patel et al: EPO attenuates renal I/R injury in the mouse 989
suggested [9] that the increase in circulating endothelialprogenitor cells by EPO contributes to the beneficial ef-fects of EPO in animal models of I/R. It is, however, notknown whether enhanced circulating levels of endothe-lial cell progenitors protect the kidney against I/R injuryor improve recovery of renal function after an acute is-chemic event.
CONCLUSION
Our data has demonstrated, for the first time, that re-combinant EPO significantly reduces the renal injury anddysfunction associated with I/R in the mouse. The de-gree of protection was dependent on the timing of ad-ministration of EPO, with pretreatment for 3 days prov-ing to be more effective than a single treatment at thetime of reperfusion. Understanding the mechanisms ofsignaling pathways that are involved in the protection ofthe kidney from ischemic injury may lead to advances inthe clinical management of patients with syndromes ofacute ischemic injury, and our data support the use ofEPO in clinical trials in the treatment of ischemic renalfailure.
ACKNOWLEDGMENT
NSAP and this work are supported by a PhD-Studentship of theWilliam Harvey Research Foundation. EJS is funded through a TrainingFellowship from the National Kidney Research Fund (NKRF).
Reprint requests to Professor Christoph Thiemermann, Centre forExperimental Medicine, Nephrology & Critical Care, William HarveyResearch Institute, Queen Mary - University of London, CharterhouseSquare, London, EC1M 6BQ, UK.E-mail: [email protected]
REFERENCES
1. JELKMANN W: Erythropoietin: Structure, control of production, andfunction. Physiol Rev 72:449–489, 1992
2. SIREN AL, FRATELLI M, BRINES M, et al: Erythropoietin preventsneuronal apoptosis after cerebral ischemia and metabolic stress.Proc Natl Acad Sci USA 98:4044–4049, 2001
3. AGNELLO D, BIGINI P, VILLA P, et al: Erythropoietin exerts an anti-inflammatory effect on the CNS in a model of experimental autoim-mune encephalomyelitis. Brain Res 952:128–134, 2002
4. JUNK AK, MAMMIS A, SAVITZ SI, et al: Erythropoietin administra-tion protects retinal neurons from acute ischemia-reperfusion in-jury. Proc Natl Acad Sci USA 99:10659–10664, 2002
5. SQUADRITO F, ALTAVILLA D, SQUADRITO G, et al: Recombinanthuman erythropoietin inhibits iNOS activity and reverts vasculardysfunction in splanchnic artery occlusion shock. Br J Pharmacol127:482–488, 1999
6. CALVILLO L, LATINI R, KAJSTURA J, et al: Recombinant human ery-thropoietin protects the myocardium from ischemia-reperfusion in-jury and promotes beneficial remodeling. Proc Natl Acad Sci USA100:4802–4806, 2003
7. PARSA CJ, MATSUMOTO A, KIM J, et al: A novel protective effectof erythropoietin in the infarcted heart. J Clin Invest 112:999–1007,2003
8. SHARPLES EJ, PATEL N, BROWN P, et al: Erythropoietin protectsthe kidney against the injury and dysfunction caused by ischemia-reperfusion injury by the inhibition of the caspase cascade and apop-tosis. J Am Soc Nephrol 14:F-PO957, 2003
9. HEESCHEN C, AICHER A, LEHMANN R, et al: Erythropoietin is a potentphysiologic stimulus for endothelial progenitor cell mobilization.Blood 102:1340–1346, 2003
10. ASSMUS B, SCHACHINGER V, TEUPE C, et al: Transplantation of pro-genitor cells and regeneration enhancement in acute myocardialinfarction (TOPCARE-AMI). Circulation 106:3009–3017, 2002
11. CHATTERJEE PK, PATEL NSA, SIVARAJAH A, et al: GW274150, apotent and highly selective inhibitor of iNOS, reduces experi-mental renal ischemia/reperfusion injury. Kidney Int 63:853–865,2003
12. CODY J, DALY C, CAMPBELL M, et al: Recombinant human erythro-poietin for chronic renal failure anaemia in pre-dialysis patients.Cochrane Database Syst Rev CD003266, 2001
13. PALLER MS: The cell biology of reperfusion injury in the kidney. JInvest Med 42:632–639, 1994
14. SHERIDAN AM, BONVENTRE JV: Pathophysiology of ischemic acuterenal failure. Contrib Nephrol 132:7–21, 2001
15. KRIBBEN A, EDELSTEIN CL, SCHRIER RW: Pathophysiology of acuterenal failure. J Nephrol 12 (Suppl 2):S142–S151, 1999
16. PALLER MS: Pathophysiologic mechanisms of acute renal failure,in Mechanisms of Injury in Renal Disease and Toxicity, edited byGoldstein RS, Ann Arbor, CRC Press, 1994, pp 3–13
17. DRAGUN D, HOFF U, PARK JK, et al: Ischemia-reperfusion injuryin renal transplantation is independent of the immunologic back-ground. Kidney Int 58:2166–2177, 2000
18. FERNANDEZ M, MEDINA A, SANTOS F, et al: Exacerbated inflamma-tory response induced by insulin-like growth factor I treatment inrats with ischemic acute renal failure. J Am Soc Nephrol 12:1900–1907, 2001
19. HUANG C, DAVIS G, JOHNS EJ: Study of the actions of human recom-binant erythropoietin on rat renal haemodynamics. Clin Sci (Lond)83:453–459, 1992
20. CAI Z, MANALO DJ, WEI G, et al: Hearts from rodents exposedto intermittent hypoxia or erythropoietin are protected againstischemia-reperfusion injury. Circulation 108:79–85, 2003
21. PRASS K, SCHARFF A, RUSCHER K, et al: Hypoxia-induced stroke tol-erance in the mouse is mediated by erythropoietin. Stroke 34:1981–1986, 2003
22. RUSCHER K, FREYER D, KARSCH M, et al: Erythropoietin is aparacrine mediator of ischemic tolerance in the brain: Evidencefrom an in vitro model. J Neurosci 22:10291–10301, 2002
23. YANG CW, LI C, JUNG JY, et al: Preconditioning with erythropoi-etin protects against subsequent ischemia-reperfusion injury in ratkidney. FASEB J 17:1754–1755, 2003
24. LIN F, CORDES K, LI L, et al: Hematopoietic stem cells contribute tothe regeneration of renal tubules after renal ischemia-reperfusioninjury in mice. J Am Soc Nephrol 14:1188–1199, 2003
25. BAHLMANN FH, DEGROOT K, DUCKERT T, et al: Endothelial progen-itor cell proliferation and differentiation is regulated by erythropoi-etin. Kidney Int 64:1648–1652, 2003