SHOCK, Vol. 00, No. 00, pp.

00Y00, 2008

RECOMBINANT HUMAN ERYTHROPOIETIN PREVENTS LIPOPOLYSACCHARIDE-INDUCED VASCULAR HYPOREACTIVITY IN THE RAT

Roberta dEmmanuele di Villa Bianca,* Rosalinda Sorrentino,* Emma Mitidieri,* Stefania Marzocco, Giuseppina Autore, Christoph Thiemermann, Aldo Pinto, and Raffaella Sorrentino*

*Dipartimento di Farmacologia Sperimentale; Universita degli studi di Napoli, Federico II, Napoli; ` Dipartimento di Scienze Farmaceutiche, Universita degli Studi di Salerno, Salerno, Italy; and William ` Harvey Research Institute, Barts and the London School of Medicine and Dentistry, London, UK
Received DD Month YYYY; first review completed DD Month YYYY; accepted in final form DD Month YYYY

ABSTRACTErythropoietin (EPO) is a hypoxia-inducible hormone that is essential for normal erythropoiesis in the bone marrow. Administration of recombinant humanYEPO is currently being used for the therapy of anemia associated with chronic renal failure and cancer. Moreover, EPO reduces organ injury in experimental hemorrhagic as well as in splanchnic artery occlusion shock and preserves cardiac function after experimental cardiac I/R. Erythropoietin receptors are widely distributed in the cardiovascular system, including endothelial, smooth muscle, cardiac, and other cell types, and nonhematopoietic effects of EPO are increasingly recognized. Thus, the vasculature may be a biological target of EPO. Therefore, the aim of our study was to investigate whether EPO exerts a protective effect in septic shock by modulating vascular dysfunction and hyporeactivity. Rats received EPO (300 U/kg, i.v.) or vehicle 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.). In vivo and ex vivo (aortic rings) experiments were performed to evaluate the vascular response to contracting and vasodilating agents. The expression of iNOS, intercellular adhesion molecule 1, poly(ADP)ribose polymerase, Bcl-xl, and Bcl-2 was evaluated by Western blot analysis in the rat aorta. We demonstrate that EPO significantly prevents LPS-induced vascular hyporeactivity and endothelial dysfunction. Interestingly, EPO inhibited the increase in iNOS, poly(ADP)ribose polymerase, and intercellular adhesion molecule 1 expression in the aorta of endotoxemic rats and attenuated the decline in the expression of both Bcl-xl and Bcl-2 caused by LPS. In conclusion, our data support the view that EPO has important nonerythropoietic effects protecting organ and tissue against injury and indicate that EPO may be useful in the therapy of patients with septic shock. KEYWORDSErythropoietin, septic shock, endothelial dysfunction, vasculature injury, rat

INTRODUCTION Erythropoietin (EPO) is a hypoxia-induced hormone that is essential for normal erythropoiesis, produced primarily by the adult kidney. Erythropoietin targets erythroid progenitor cells in the bone marrow to increase the number of mature red blood cells (RBCs) (1). The production of recombinant human (rh)YEPO has revolutionized the treatment of anemia associated with chronic renal failure and chemotherapy, and it has been used as prophylaxis to prevent anemia after surgery. The EPO receptor is widely distributed in the cardiovascular system, that is, located on endothelial cells (2), smooth muscle cells (3), and cardiomyocytes (4). Although, EPO receptors are localized on endothelial and smooth muscle cells, EPO has no direct vasoconstrictor effect in either rabbit aorta or human renal artery, but it enhances the contractions caused by norepinephrine by increasing the synthesis of constrictor prostanoids and endothelin 1 (5). The probable source of these vasoconstrictor autacoids is the endothelium, as endothelial removal attenuates the increase in contraction mediated by EPO. In endothelial cells from umbilical vein, EPO causes an increase in prostaglandin F2! and thromboxane B2 and a
Address reprint requests to Raffaella Sorrentino, Dipartimento di Farmacologia ` Sperimentale, Universita di Napoli Federico II, Via D. Montesano, 49, 80131 Napoli, Italy. E-mail: rafsorre@unina.it. DOI: 10.1097/SHK.0b013e31818909c0 Copyright 2008 by the Shock Society

decrease in PGI2 (5). At the same time, the release of endothelin 1 is increased by nearly 90% in the presence of EPO (6). Incubation of bovine pulmonary arterial endothelial cells with rh-EPO induces a rise of intracellular calcium concentration accompanied by an increase in endothelin 1 release and in preproendothelin 1 mRNA expression (7). On vascular smooth muscle cells, EPO causes an increase in angiotensin receptor messenger RNA, resulting in a parallel increase in the expression of angiotensin II (Ang II) receptors, which affects vasomotor tone and remodeling of vascular wall by enhancing cell proliferation (8). Furthermore, EPO inhibits apoptosis and induces proliferation and differentiation (9). Taken together, these findings indicate that EPO is involved in the regulation of vascular tone through an indirect effect. In addition, EPO reduces the organ injury and dysfunction of rats subjected to hypovolemic hemorrhagic shock (10); exerts beneficial effects in splanchnic artery occlusion shock, possibly by inhibiting iNOS activity (11); and improves skeletal muscle microcirculation and tissue bioenergetics in a mouse sepsis model (12). Moreover, EPO prevents LPS-induced apoptosis in bovine pulmonary artery endothelial cells, resulting in an increase in viability (13). In addition, EPO protects the myocardial structure and preserves cardiac function during ischemia and reperfusion (14Y16). The discovery that EPO plays a significant biological role in tissues outside the hematopoietic system has fuelled significant interest in EPO as a novel cytoprotective cytokine (17). Therefore, the aim of our study was
1

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK VOL. 00, NO. 00

DI

VILLA BIANCA

ET AL.

to investigate whether EPO exerts a protective effect in septic shock by modulating vascular dysfunction and vascular hyporeactivity. MATERIALS AND METHODS
Male Wistar rats weighing 250 to 300 g were used (Harlan, Udine, Italy). Animals were housed in an environment with controlled temperature (21-CY24-C) and lighting (12:12-h light-dark cycle). Standard chow and drinking water were provided ad libitum. A period of 7 days was allowed for acclimatization of rats before any experimental manipulation was undertaken. Animal use was in accordance with the guidelines of Italian and European Council for animal care. The animals were divided into four experimental groups: vehicle plus LPS-treated rats (LPS), EPO plus LPS treated rats (EPO+LPS), control-vehicle (sham), and control-EPO (sham-EPO). Erythropoietin (300 U/kg; i.v.) or vehicle (saline) was administered 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.) or vehicle (saline) injection.

In vivo experiments
Hemodynamic changes were evaluated in anesthetized rats by urethane injection (1 g/kg; i.p.). The trachea was cannulated to facilitate respiration; the right carotid artery was cannulated and connected to a pressure transducer (Bentley 800 Trantec; Basile, Comerio, Italy) for the measurement of MAP as well as heart rate, which were recorded using the Power Lab/800 (AD/Instruments). The left jugular vein was cannulated for administration of drugs. After surgical procedure, cardiovascular parameters were allowed to stabilize for 20 min. Anesthetized rats received EPO (300 U/kg, i.v.) or vehicle (saline) 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.) injection. A dose-response effect to phenylephrine (PE, 3, 10, and 30 2g/kg, i.v.) was performed after 4 h from LPS or vehicle (saline) injection. The recover time between each injection was 15 min.

FIG. 2. Effect of EPO on LPS-induced hyporeactivity to PE (3, 10, and 30 2g/kg, i.v.) after 4 h from LPS or vehicle injection. Anesthetized rats received EPO (300 U/kg, i.v.) or vehicle (saline) 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.) injection. Results are expressed as delta of increase in MAP of mean T SEM of LPS (n = 10), LPS+EPO (n = 7), shamEPO (n = 6), and sham (n = 5) groups. Data are analyzed by one-way ANOVA followed by Bonferroni as posttest. P G 0.05 was taken as significant. *P G 0.005 and **P G 0.01 vs. sham; -P G 0.05 and --P G 0.01 vs. LPS.

Ex vivo experiments
In another set of experiments, rats were treated as described above. At 4 h after injection of either LPS or vehicle (saline), rats were killed by cervical dislocation after exposure to isoflurane, and the thoracic aortas were excised, carefully cleaned of connective tissue and cut in rings (2Y3 mm). Rings were hooked in 2.5 mL water-jacketed organ baths filled with thermostated (37-C) and gassed (95% O2 and 5% CO2) Krebs solution with the following composition (in millimolars): NaCl, 115.3; KCl, 4.9; CaCl2, 1.46; MgSO4, 1.2; KH2PO4, 1.2; NaHCO3, 25.0; and glucose, 11.1. Aortic rings were connected to isometric force transducers (model 7002; Basile), and changes in tension were recorded continuously using a polygraph linearcorder (WR3310; Graphtec, Japan). Tissues were preloaded with 0.5 g and allowed equilibrating for at least 90 min; during this time, Krebs solution was changed about each 15 min. After equilibration, the tissues were used to evaluate vascular re-

activity. The contractile response was evaluated by using PE (1 nM up to 3 2M), h-endothelin 1(hYEt-1, 30 nM), U46619, a stable analog of thromboxane A2 (0.3 2M) and Ang I (0.1 2M). To assess the endothelial function, we performed a concentration-response curve with acetylcholine (ACh, 10 nM to 3 2M) in aortic rings precontracted by PE (0.3 2M). In another set of experiments to evaluate the release of NO, we estimated the increase in L-NAME (100 2M)Yinduced contraction added on stable tone of PE (0.3 2M).

Western blot analysis for iNOS, intercellular adhesion molecule 1, poly(ADP)ribose polymerase, Bcl-xl, and Bcl-2 expression
Briefly, frozen arteries were homogenized in a lysis buffer (50 mM "-glycerophosphate, 100 2M Na3VO4, 2 mM MgCl2, 1 mM EGTA, 0.5% triton, and 1 mM DTT) containing protease inhibitors (20 2M pepstatin, 20 2M leupeptin, 1,000 U/mL aprotinin and 1 mM PMSF). Protein concentration was estimated by Bio-Rad protein assay using bovine serum albumin as standard. Equal amounts of protein (50 2g) were dissolved in Laemmli sample buffer, boiled and run on a sodium dodecyl sulfateYpolyacrylamide gel electrophoresis minigel, and then transferred to a Hybond polyvinylidene difluoride membrane. Membranes were then blocked for 40 min in PBS containing 5% (wt/vol) nonfat milk and subsequently probed overnight at 4-C with mouse monoclonal anti-iNOS or antiYintercellular adhesion molecule 1 (ICAM1) or antiYpoly(ADP)ribose polymerase (PARP) or antiYBcl-xl or antiYBcl-2 (1:2,500, 1:600, 1:400, 1:750, and 1:750 dilution, respectively, in PBS containing 5% wt/vol nonfat milk and 0.1% Tween-20). Blots were then incubated, after 4 washes in PBS containing 5% wt/vol nonfat milk and 1% Tween-20, with horseradish peroxidaseYconjugated goat antiYmouse IgG (1:5,000 for iNOS, Bcl-xl, and Bcl-2 detection) or with horseradish peroxidaseYconjugated goatYanti-rabbit IgG (1:5,000 for ICAM-1 and PARP detection) for 1 h at room temperature. Immunoreactive bands were visualized using ECL detection system according to the manufacturers instructions and exposed to Kodak X-Omat film. Protein bands on x-ray film were quantified by scanning densitometry (Imaging Densitometer GS-700; Bio-Rad) and results normalized with tubulin tissue expression, as reference protein. Densitometric analyses were performed using the NIH Image program.

Evaluation of RBCs, Hb, and hematocrit

FIG. 1. Effect of EPO on LPS-induced hypotension in anesthetized rats. Blood pressure values are reported as MAP and expressed in millimeters of mercury. The MAP was monitored for 4 h after LPS challenge (time 0). Insert, Early phase (0 Y 30 min). Anesthetized rats received EPO (300 U/kg, i.v.) or vehicle (saline) 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.) injection. Results are expressed as mean T SEM for LPS (n = 10), LPS+EPO (n = 7), sham-EPO (n = 6), and sham (n = 5) groups. Data are analyzed by two-way ANOVA followed by Bonferroni as posttest. P G 0.05 was taken as significant. ###P G 0.0001 vs. LPS; **P G 0.005 LPS or LPS+EPO vs. sham.

Blood samples were withdrawn before and after each treatment as described above to evaluate changes in hematocrit (HCT), RBCs, and hemoglobin (Hb) using a cell counter (Coulter AcT Diff 2; Instrumentation Laboratory, Italy).

Statistical analysis
All values in the figures, table, and text are expressed as mean T SEM. The statistical analysis was performed using GraphPad Prism (GraphPad Software, San Diego, Calif) by one or two-way ANOVA followed by Bonferroni post hoc test. A P G 0.05 was taken as significant. The increase in MAP induced by PE was expressed as delta compared with the basal value. The contraction afforded by a given drug ex vivo was expressed as dyne per

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK MONTH 2008

EPO PREVENTS LPS-INDUCED VASCULAR INJURY

FIG. 3. Effect of EPO on vascular reactivity to different pressor agents in ex vivo experiments in rat aorta rings. The contractile response was evaluated after 4 h from treatments EPO (300 U/kg, i.v.) or vehicle (saline) 30 min before and 1 and 3 h after LPS (8 106 U/kg, i.v.) injection. A, PE (1 nmol up to 3 2M); **P G 0.001 vs. LPS and sham-EPO, ---P G 0.0001 vs. sham-EPO and LPS. B, Ang II (Ang, 0.1 2M); *P G 0.05 vs. LPS, **P G 0.01 vs. LPS, ***P G 0.001 vs. LPS, -P G 0.05 vs. EPO+LPS, and --P G 0.01 vs. EPO+LPS. C, U46619, a stable analog of thromboxane A2 (0.3 2M); *P G 0.05 vs. EPO+LPS and sham-EPO. D, hYEt-1(h-ET-1, 30 nM); *P G 0.05 vs. EPO+LPS, **P G 0.005 vs. LPS. A cumulative concentration curve was performed with PE, whereas a single concentration response was used for Ang, U46691, and hET-1. Results are expressed in dyne/mg of tissue as mean T SEM of LPS (n = 8), EPO+LPS (n = 8), sham-EPO (n = 5), and sham (n = 5) groups. Data are analyzed by two-way (A) and one-way (B, C, and D) ANOVA , followed by Bonferroni as posttest. P G 0.05 was taken as significant.

milligram of tissue. Any relaxation caused by a drug in the same ex vivo experiments was expressed as percentage of relaxation.

RESULTS
In vivo study

The baseline values for MAP were not significantly different between any of the groups studied. The MAP values were

100 T 3 (n = 10), 109 T 7 (n = 7), 105 T 3 (n = 6), and 102 T 4 mmHg (n = 5) for LPS, LPS+EPO, sham-EPO, and sham, respectively. When compared with sham, LPS administration caused a significant hypotension (P G 0.001; Fig. 1). Interestingly, EPO attenuated the hypotension observed in LPS-treated rats (P G 0.001; Fig. 1). The MAP values of sham rats treated with EPO (sham-EPO) were not different

FIG. 4. Effect of EPO on endothelial dysfunction caused by LPS (8 106 U/kg, i.v.). A, ACh (10 nmol to 3 2M) concentration-response curve in aorta rings precontracted by PE (0.3 2M); ***P G 0.001 vs. EPO+LPS and sham, **P G 0.01 vs. LPS. Results are expressed as % of relaxation to PE tone. B, L-NAME (100 2M)Yinduced contraction added on stable tone of PE (0.3 2M); *P G 0.05 vs. LPS; **P G 0.01 vs. LPS. Results are expressed as % of delta increase to PE tone. Data are expressed of mean T SEM of LPS (n = 12), EPO+LPS (n = 12), sham-EPO (n = 5), and sham (n = 5) groups and analyzed by two-way (A) and one-way (B) ANOVA, followed by Bonferroni as posttest. P G 0.05 was taken as significant.

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK VOL. 00, NO. 00

DI

VILLA BIANCA

ET AL.

FIG. 5. Effect of EPO on the expression of different protein in rat aorta after 4 h from treatments by Western blot analysis. A, iNOS expression (##P G 0.01 vs. sham, **P G 0.01 vs. LPS). B, ICAM-1 expression (###P G 0.001 vs. sham, *P G 0.0 vs. LPS). C, Cleaved PARP fragment expression (##P G 0.01 vs. sham, **P G 0.01 vs. LPS). Protein bands were quantified by scanning densitometry, and results are reported as integrated values (area density of the band). All values are expressed as means T SEM of three different experiments with three replicates in each. P G 0.05 was taken as significant.

from those measured in sham rat. The vascular reactivity in vivo to PE (3, 10, and 30 2g/kg i.v.) was evaluated 4 h after LPS or vehicle injection. Compared with the sham group, a significant hyporeactivity to PE was observed in the LPS group (3 and 10 2g/kg, P G 0.01; 30 2g/kg, P G 0.05; Fig. 2). Erythropoietin treatment completely reverted the vascular hyporeactivity to PE induced by LPS (3 2g/kg, P G 0.05; 10 and 30 2g/kg, P G 0.005; Fig. 2); indeed, LPS+EPO rats showed a reactivity to PE that was statistically indistinguishable to sham group. In the sham-EPO group, the PE-induced variation in blood pressure was similar to those produced in the sham and LPS+EPO groups (Fig. 2).
Ex vivo study

Vasoconstrictor responsesAt 4 h after administration of LPS, rats were killed, and aortic rings were suspended in organ baths and challenged with different contracting agents. The contraction of aortic rings caused by PE (3nM to 3 2M) was significantly reduced in ex vivo LPS-treated rats compared with the sham group (P G 0.0001; Fig. 3A). Erythropoietin treatment (LPS+EPO) attenuated the LPS-induced hyporeactivity to PE (P G 0.0001; Fig. 3A). In aortic rings collected from sham-EPO group, the concentration-response curve to

PE was significantly increased compared with the sham group (P G 0.0001; Fig. 3A). In the same way, aortic rings of LPS rats showed a significant (P G 0.05) reduced response to Ang (0.1 2M), U-46619 (0.3 2M), or hYEt-1 (30 nM) compared with the sham group (Fig. 3, BYD). Similarly to response to PE, aortic rings of LPS+EPO rats exhibited a contraction to Ang (0.1 2M), U-46619 (0.3 2M), or hYEt-1 (30 nM) completely comparable to sham group (Fig. 3, BYD). Erythropoietin treatment did not affect the response to Ang, U-46619, or hYEt-1 in sham animals. Acetylcholine-induced relaxation and NO releaseTo assess the integrity of the endothelium, an ACh-induced concentration-response curve was performed. In endotoxemic rats, ACh-induced relaxation was significantly reduced (P G 0.0001) compared with the sham group, indicating the development of an endothelial dysfunction, which was prevented by EPO treatment (LPS+EPO group; Fig. 4A). Indeed, AChinduced relaxation in LPS+EPO aortas was similar to that observed in the sham group (Fig. 4A). Erythropoietin treatment did not modify the relaxation induced by ACh in sham animals (sham-EPO; Fig. 4A). The increase in contracting force of aorta rings, displaced by adding L-NAME (100 2M), an inhibitor of NOS, on the stable tone of PE-induced contraction

FIG. 6. Effect of EPO on the expression of different proteins in rat aorta after 4 h from treatments by Western blot analysis. A, Bcl-2 expression (###PG 0.001 vs. sham, **P G 0.01 vs. LPS. B, Bcl-xl expression ((###P G 0.001 vs. sham, *P G 0.05 vs. LPS). Protein bands were quantified by scanning densitometry, and results are reported as integrated values (area density of the band). All values are expressed as means T SEM of three different experiments with three replicates in each. P G 0.05 was taken as significant.

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK MONTH 2008

TABLE 1. Basal Treatment Sham LPS Epo+LPS Sham-EPO Hb, g/dL 14.10 T 0.2 14.38 T 0.4 14.02 T 0.7 14.68 T 0.6 RBCs, 106/2L 7.85 T 1.4 9.10 T 0.3 8.54 T 0.4 9.68 T 0.4 HCT, % 48.30 T 2.1 51.12 T 1.7 47.08 T 2.0 50.38 T 0.8

EPO PREVENTS LPS-INDUCED VASCULAR INJURY

4h Hb, g/dL 14.33 T 2.3 14.23 T 0.4 14.18 T 1.1 13.90 T 1.0 RBCs, 106/2L 10.77 T 2.7 9.50 T 0.3 9.46 T 0.9 9.91 T 0.8 HCT, % 52.00 T 1.2 52.88 T 1.5 53.48 T 5.0 52.58 T 3.0

Red blood cell count and HCT were measured in rats at different points: basal (at the beginning of the experiment) and 4 h after LPS challenge. Hb, RBC count, and HCT were not changed in LPS rats. In addition, EPO treatment did not modify Hb, RBC count, and HCT either in sham or LPS+EPO group. The values are reported as mean T SEM of sham (n = 5), LPS (n = 10), LPS+EPO (n = 7), and sham-EPO (n = 6) groups.

(0.3 2M), was assumed as an index of NO release. Aortas of LPS-treated rats showed a significant (P G 0.01) increase in NO release than in aortas obtained from the sham group. In LPS+EPO rats, the NO release was significantly (P G 0.05) reduced compared with the LPS group (Fig. 4B) and not statistically different from the sham-EPO and sham groups (Fig. 4B). Western blot studyWestern blot analysis revealed a significant (P G 0.005) increase in iNOS protein expression in aorta of LPS group compared with sham, LPS+EPO, and sham+EPO groups (Fig. 5A). The expression of iNOS proteins resulted significantly attenuated in LPS+EPOYtreated rats; whereas no difference was observed in sham+EPO and LPS+EPO compared with the sham group. Endotoxemia (LPS group) also resulted in an increased of ICAM-1 and PARP cleaved form (89 kd) protein expression (P G 0.01; Fig. 5, B and C, respectively), and both of these effects were significantly (P G 0.05) attenuated by EPO treatment. Bcl-2 and Bcl-xl protein expressions, which inhibit both apoptosis and proliferation, were significantly reduced in the aorta of LPS-treated rats (P G 0.01; Fig. 6, A and B). Interestingly, EPO treatment significantly attenuated the decline in the expression of both Bcl-2 and Bcl-xl caused by LPS (P G 0.05; Fig. 6, A and B). Evaluation of RBCs, HCT, and HbRed blood cell count, Hb, and HCT (%) were measured in all rats before each treatment (basal) and 4 h after challenge with LPS or vehicle. Red blood cell, Hb, and HCT values were not changed in each group (Table 1). DISCUSSION The presence of EPO receptors on several cell types, that is, neurons, cardiomyocytes, endothelial, vascular smooth muscle, and kidney cells such as proximal tubule epithelial and mesangial cells and the glomerulus (2Y4, 18, 19), indicates that EPO, as a growth factor or cytokine, may have effects on many cells and tissues. We demonstrated here that EPO attenuates both hypotension and vascular hyporeactivity in an in vivo and ex vivo model of endotoxemic shock in rat. Similar beneficial effects have been reported for EPO in other animal models of shock, including hemorrhagic and splanchnic artery occlusion shock (10, 11). We found that EPO prevented early hypotension (30 min) and, moreover, the long-lasting hypotension induced by LPS administration. The response to the !1 receptor agonist PE was significantly reduced, indicating the development of a vascular hyporeactivity in endotoxemic rats. Most notably, EPO significantly

improved the PE response in LPS-treated rats in vivo. Our in vivo data correlated with our ex vivo results obtained in the isolated aorta. Specifically, EPO pretreatment prevented LPSinduced hyporeactivity to PE in isolated aortic rings. Surprisingly, EPO significantly augmented the response to PE in aortic rings not subjected to endotoxemia, while such an effect was not observed in vivo (above). The latter result may be due to effects of anesthesia and/or by blood pressure autoregulation. The increase observed in !1 response induced by EPO ex vivo is in accord with the studies of Bode-Boger and colleagues (5, 6). Indeed, they have demonstrated that the enhanced contractile response to noradrenalin involves a shift in the balance of constrictor and relaxing prostanoids (5, 6). This imbalance in prostanoid production seems to be closely related to !1 adrenergic receptor pathway. Thus, we propose that the increase in the response to PE afforded by EPO, which was not observed when we used thromboxane A2, endothelin 1, or Ang as vasoconstrictor, may be secondary to the release by EPO of endogenous vasoconstrictor autacoids (5Y8). However, EPO also attenuated the LPS-induced vascular hyporeactivity to the vasoconstrictor responses elicited by h-endothelin, thromboxane A2, and Ang II. Moreover, the improvement in vascular hyporeactivity induced by LPS may be related to an increase in intracellular calcium concentration induced by EPO (20). Interestingly, EPO attenuated the impairment in ACh vasodilator response caused by LPS, indicating that this hormone also prevented the endothelial dysfunction caused by LPS. In fact, it has been demonstrated that EPO attenuates the cardiac injury and dysfunction caused by regional myocardial ischemia in the rat and rabbit (15, 21, 22). This beneficial effect of EPO has been ascribed to preservation of endothelial function and vascular flow in coronary vessel secondary to antiapoptotic effects of EPO. Our molecular data are in this direction; in fact, we found that the impaired expression of the antiapoptotic proteins Bcl-2 and Bcl-xl in aortas of LPS group was significantly attenuated by EPO treatment. All of the above data support the hypothesis that EPO prevents the endothelial dysfunction and injury caused by LPS. Noteworthy are the results obtained for PARP, a nuclear enzyme activated by DNA damage. During DNA damage, PARP is activated as a DNA-repair enzyme by formation of poly(ADP-ribose) in a process that consumes ATP. The excessive DNA damage associated with a huge activation of PARP leads to a cascade of events, known as the Bsuicide pathway[ (23) that causes necrosis (24) and apoptosis (25). Poly(ADP)ribose polymerase inhibitors were

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.

SHOCK VOL. 00, NO. 00

DI

VILLA BIANCA

ET AL.

found to be beneficial in many pathophysiological conditions associated with oxidative stress and, moreover, PARP knockout mice proved to be resistant to LPS-induced shock (26). Our data showed an increase in the formation of PARP in aortas obtained from rats subjected to endotoxemia that was completely reverted by EPO administration. This result, in addition to the observed effect on Bcl-2 and Bcl-xl, indicates that EPO may prevent LPS-induced apoptosis in aortas. A further enzyme involved in cell damage is iNOS, which expression in many cell types, including macrophages and vascular smooth muscle cells, is triggered by LPS (27). The overproduction of NO by iNOS could be responsible in part for the endothelial and DNA damage, because simultaneous production of NO and superoxide anions in LPS-treated animals leads to the production of peroxynitrite and subsequently nitrotyrosine formation in the blood vessels (28). Erythropoietin contributed to the protective effect of vascular wall by inhibition of iNOS expression. Furthermore, the reduction in iNOS expression was accompanied by significant diminution in NO formation, as demonstrated by L-NAMEYinduced contraction, which could be involved in the improvement of vascular hyporeactivity. Similarly, it has been reported that EPO protects against splanchnic artery occlusion shock by inhibiting the expression of iNOS protein (11). A further index of LPS-induced cellular damage is ICAM-1 expression. Erythropoietin also attenuated the expression of the adhesion molecule ICAM-1, resulting in a reduced adhesion of neutrophils to the endothelium. It should be noted that the observed beneficial effects of EPO are not related to any improvement in HCT or Hb, as the dose of EPO used had no effect on these parameters. Our data support the hypothesis that EPO has important nonerythropoietic effects, which may protect tissues and organs against injury. Taken together, our data support the view that a low dose of rh-EPO exerts beneficial effects in rodent models of endotoxemia. As sustaining blood pressure is critical for the survival of humans with severe septic shock (29), we speculate that our data indicate that EPO may be useful in the therapy of patients with septic shock, but clearly, further studies are warranted to support this working hypothesis. ACKNOWLEDGMENTS
The authors thank medical veterinarians Dr Lucia DEsposito, Giovanni Esposito, and Angelo Russo for animal care assistance.

REFERENCES
1. Gabrilove J: Overview: erythropoiesis, anemia, and the impact of erythropoietin. Semin Hematol 371:1Y3, 2000. 2. Anagnostou A, Liu Z, Steiner M, Chin K, Lee ES, Kessimian N, Noguchi CT: Erythropoietin receptor mRNA expression in human endothelial cells. Proc Natl Acad Sci 91(9):3974Y3978, 1994. 3. Ammarguellat F, Llovera M, Kelly PA, Goffin V: Low doses of EPO activate MAP kinases but not JAK2-STAT5 in rat vascular smooth muscle cells. Biochem Biophys Res Commun 284(4):1031Y1038, 2001. 4. Wu H, Lee SH, Gao J, Liu X, Iruela-Arispe ML: Inactivation of erythropoietin leads to defects in cardiac morphogenesis. Development 26(16):3597Y3605, 1999. 5. Bode-Boger SM, Boger RH, Kehn M, Radermacher J, Frolich JC: Recombinant human erythropoietin enhances vasoconstrictor tone via endothelin-1 and constrictor prostanoids. Kidney Int 50:1255Y1265, 1996.

6. Bode-Boger SM, Boger RH, Kuhn M, Radermacher J, Frolich JC: Endothelin release and shift in prostaglandin balance are involved in the modulation of vascular tone by recombinant erythropoietin. J Cardiovasc Pharmacol 20(suppl 12):S25YS28, 1992. 7. Carlini RG, Gupta A, Liapis H, Rothstein M: Endothelin-1 release by erythropoietin involves calcium signaling in endothelial cells. J Cardiovasc Pharmacol 26(6):889Y892, 1995. 8. Barrett JD, Zhang Z, Zhu JH, Lee DB, Ward HJ, Jamgotchian N, Hu MS, Fredal A, Giordani M, Eggena P: Erythropoietin upregulates angiotensin receptors in cultured rat vascular smooth muscle cells. J Hypertens 16(Pt 1Y12): 1749Y1757, 1998. 9. Kouri MC, Boudurant MC: Maintenance by erythropoietin of viability and maturation of murine erythroid precursor cells. J Cell Physiol 137:65Y74, 1988. ` 10. Buemi M, Allegra A, Squadrito F, Buemi AL, Lagana A, Aloisi C, Frisina N: Effects of intravenous administration of recombinant human erythropoietin in rats subject to hemorrhagic shock. Nephron 65(3):440Y443, 1993. 11. Squadrito F, Altavilla D, Squadrito G, Campo GM, Arlotta M, Quartarone C, Saitta A, Caputi AP: Recombinant human erythropoietin inhibits iNOS activity and reverts vascular dysfunction in splanchnic artery occlusion shock. Br J Pharmacol 27(2):482Y488, 1999. 12. Kao R, Xenocostas A, Rui T, Yu P, Huang W, Rose J, Martin CM: Erythropoietin improves skeletal muscle microcirculation and tissue bioenergetics in a mouse sepsis model. Crit Care 11(3):R58YR64, 2007. 13. Carlini RG, Alonzo EJ, Dominguez J, Blanca I, Weisinger JR, Rothstein M, Bellorin-Font E: Effect of recombinant human erythropoietin on endothelial cell apoptosis. Kidney Int 55(2):546Y553, 1999. 14. Abdelrahman M, Sharples EJ, McDonald MC, Collin M, Patel NS, Yaqoob MM, Thiemermann C: Erythropoietin attenuates the tissue injury associated with hemorrhagic shock and myocardial ischemia. Shock 22(1):63Y69, 2004. 15. Parsa CJ, Matsumoto A, Kim J, Riel RU, Pascal LS, Walton GB, Thompson RB, Petrofski JA, Annex BH, Stamler JS, et al.: A novel protective effect of erythropoietin in the infarcted heart. J Clin Invest 112(7):999Y1007, 2003. 16. Lipsic E, van der Meer P, Henning RH, Suurmeijer AJ, Boddeus KM, van Veldhuisen DJ, van Gilst WH, Schoemaker RG: Timing of erythropoietin treatment for cardioprotection in ischemia/reperfusion. J Cardiovasc Pharmacol 44(4):473Y479, 2004. 17. Smith KJ, Bleyer AJ, Little WC, Sane DC: The cardiovascular effect of erythropoietin. Cardiovascular Res 59:538Y548, 2003. 18. Noguchi CT, Asavaritikrai P, Teng R, Jia Y: Role of erythropoietin in the brain. Crit Rev Oncol Hematol 64(2):159Y171, 2007. 19. Westenfelder C, Biddle DL, Baranowski RL: Human, rat, and mouse kidney cells express functional erythropoietin receptors. Kidney Int 56(3):1159Y1160, 1999. 20. Neusser M, Tepel M, Zidek W: Erythropoietin increases cytosolic free calcium concentration in vascular smooth muscle cells. Cardiovasc Res 27(7): 1233Y1236, 1993. 21. Parsa CJ, Kim J, Riel RU, Pascal LS, Thompson RB, Petrofski JA, Matsumoto A, Stamler JS, Koch WJ: Cardioprotective effects of erythropoietin in the reperfused ischemic heart: a potential role for cardiac fibroblasts. J Biol Chem 279(20):20655Y20662, 2004. 22. Lin X, Fujita M, Kanemitsu N, Kimura Y, Tambara K, Premaratne GU, Nagasawa A, Ikeda T, Tabata Y, Komeda M: Sustained-release erythropoietin ameliorates cardiac function in infracted rat-heart without inducing polycythemia. Circ J 71(1):132Y137, 2007. 23. Gaal JC, Smith KR, Pearson CK: Cellular euthanasia mediated by a nuclear enzyme: a central role for nuclear ADP-ribosylation in cellular mechanism. Trends Biochem 12:129Y130, 1987. 24. Ha HC, Snyder SH: Poly(ADP-ribose) polymerase is a mediator of necrotic cell death by ATP depletion. Proc Natl Acad Sci 96(24):13978Y13982, 1999. 25. Moroni F, Meli E, Peruginelli F, Chiarugi A, Cozzi A, Picca R, Romagnoli P, Pellicciari R, Pellegrini-Giampietro DE: Poly(ADP-ribose) polymerase inhibitors attenuate necrotic but non apoptotic neuronal death in experimental models of cerebral ischemia. Cell Death Diff 8:921Y932, 2001. 26. Veres B, Gallyas F, Varbiro G, Berente Z, Osz E, Szekeres G, Szabo C, Sumegi B: Decrease of the inflammatory response and induction of the Akt/ protein kinase B pathway by poly-(ADP-ribose) polymerase 1 inhibitor in endotoxin-induced septic shock. Biochem Pharmacol 65:1373Y1382, 2003. 27. Panaro MA, Brandonisio O, Acquafredda A, Sisto M, Mitolo V: Evidences for iNOS expression and nitric oxide production in the human macrophages. Curr Drug Targets Immune Endocr Metabol Disord 3:210Y221, 2003. 28. Zingarelli B, Day BJ, Crapo JD, Salzman AL, Szabo C: The potential role of peroxynitrite in the vascular contractile and cellular energetic failure in endotoxic shock. Br J Pharmacol 120:259Y267, 1997. 29. Annane D: Replacement therapy with hydrocortisone in catecholaminedependent septic shock. J Endotoxin Res 7:305Y309, 2001.

Copyright @ 2008 by the Shock Society. Unauthorized reproduction of this article is prohibited.