Experimental Models of Acute Renal Failure and Erythropoietin: What Evidence of a Direct Effect?

Abstract

The kidney can achieve a structural and functional recovery after the damage induced by ischemia and reperfusion. This is due to the regeneration of epithelial tubular cells, the intervention of immature cells mainly localized in the medulla, and a small number of bone marrow-derived stem cells. In many instances, however, recovery is delayed or does not occur at all. The mechanisms allowing the renal cells to de-differentiate still need to be clarified in order to find a therapeutic approach that can amplify this ability and then stop the fibroid involution and the progression toward renal failure. Several authors have hypothesized a protective effect of EPO against ischemic and cytotoxic renal damage and observed that patients precociously treated with EPO showed a slower progression of renal failure. EPO has been demonstrated to have proliferative and anti-apoptotic effects in ischemia-reperfusion models in the brain and cell cultures. Moreover, EPO can mobilize stem cells and increase the plasmatic levels and the renal expression of VEGF. These effects seem to be dose-dependent and could be due to the activation of signal transduction systems, like Jak and STAT. In the presence of high doses of exogenous EPO or during the treatment with long-acting EPO-like molecules, non-specific receptors may be activated through a low-affinity link. Further investigations are needed to determine new therapeutic applications for EPO and other analogous hormones. Very long-acting molecules or molecules with cyto-protective but no erythropoietic effect may represent useful tools in the study of the molecular mechanisms underlying EPO's action and may have a rapid and safe therapeutic application.

Keywords:

• apoptosis
• erythropoietin
• JAK
• renal failure
• STAT

INTRODUCTION

Ischemic damage represents one of the main causes of acute renal failure, characterized by a strong clinical impact and a high mortality rate (from 50–70% in intensive care unit patients requiring dialytic treatment) that has not changed significantly over the last 40 years. The reperfusion following ischemia can cause a further injury, involving especially the proximal tubule. The ischemia/reperfusion (I/R) damage is caused by several factors, including the reduction of glomerular filtration rate, leukocyte accumulation, production of oxygen- and nitrogen-reactive species, and loss of anti-oxidative systems. The early phases after ischemia are dominated by the processes leading to cellular injury and death.[1] Because the post-ischemic kidney can reach a complete structural and functional recovery, the second phase that overlaps with the first one is characterized by the repairing of the damaged organ, mainly due to the regenerative ability of tubular cells.[2] In many instances, however, recovery is delayed or does not occur at all. A better understanding of the cellular and molecular mechanisms leading to renal failure following ischemia and reperfusion is needed to find a therapeutic approach that minimizes injury and accelerates recovery.[3–5] In organs with a weaker repairing ability, such as heart and brain, the ischemic injury rapidly leads to infarction and fibrosis.[6–11] On the other hand, the fibroid involution of the renal parenchyma is mainly observed as a result of a chronic ischemic-hypoxic or inflammatory injury.[12] The ability of the kidney to repair the ischemic damage is due to several factors that are still not completely known.[13] The tubular cells, which constitute the main part of renal parenchyma, are epithelial cells and characterized by a more marked plastic and regenerative capacity[14],[15] with respect to cerebral or cardiac cells.[16],[17] Some authors have described also in the heart and brain progenitor cells that can de-differentiate and regenerate the damaged tissue.[18–20] Nevertheless, these cells are not enough to contrast the ischemic injury.[21] Few differentiated cells with a high proliferation potential and migratory activity have been described in the kidney, mainly in the medulla;[22] these elements can contribute to the post-ischemic functional recovery because they reach the basal membrane of the damaged tubules within a few days after the insult.[23],[24] Some authors have investigated the role of bone marrow-derived stem cells in the repairing of ischemia-reperfusion damage by analyzing the expressions of fluorescent proteins in transgenic (GFP) mice. Using three different markers of bone marrow-derived cells in chimerical mice, they have reported that in the post-ischemic phase, when a large number of new tubular cells is produced, bone marrow-derived stem cells do not localize in the damaged tubules.[25] This evidence suggests that the renal functional recovery is due to resident cells expressing de-differentiation markers to more than bone marrow-derived cells. In the kidney, these resident cells also express replication markers, like Ki67, PAX 2, and Kim-1. Thus, the role of the bone marrow-derived cells is particularly weak, as renal tubules have an intrinsic ability to repair. The renal GFP models show few chimerical tubules with a mix of resident cells and graft-derived cells (characterized by the expression of the GFP protein).[26] The mechanisms allowing the renal cells to de-differentiate still need to be clarified in order to find a therapeutic approach that can amplify this ability and then stop the fibroid involution and the progression toward ESRD. It is known that the kidney can react to a condition of systemic or local hypoxia by producing erythropoietin.

Moreover, recent in vitro and in vivo studies have demonstrated that EPO attenuates ischemic cell damage. The EPO-mediated protective effects seem to be related to the anti-apoptotic, anti-oxidative, and anti-inflammatory properties of this hormone. Thus, exogenous EPO (rHuEPO), such as the endogenous hormone, could play a significant role in the prevention and in the repair of renal ischemic-hypoxic damage.

EPO THERAPY AT PRESENT DAY

Currently, recombinant erythropoietin represents an indispensable substitutive therapy for patients affected by renal failure, end stage renal disease (ESRD), and patients on dialysis.[27] The goal of the recombinant molecule and its long-acting forms is to maintain a good hemoglobin level and good oxygenation of the tissues.[28] The anaemia correction leads to an improvement in cardiac performance and life quality and to a reduction of left ventricular hypertrophy[29],[30] without any important side effects. Moreover, EPO is certainly useful, not only for patients with renal failure but also for oncologic patients,[31],[32] in which anaemia is due to the effect of both the tumoral mass and the heavy anti-blastic treatment.[33] High doses of EPO, in fact, together with growth factors like GM-CSF, can oppose the toxic effect of chemotherapy on bone marrow and improve the recovery of hematopoietic function. Nevertheless, a massive use of the hormone could lead to more significant side effects.[34],[35] Although there are not enough data in literature, some authors have hypothesized that some tumors express the EPO-receptor or produce erythropoietin.[36],[37] A recent trial on breast carcinoma has been stopped because of an unexpected increased frequency of metastatic lesions in the patients treated with high doses of EPO;[38] such reports could lead to a reduction in the therapeutic use of erythropoietin in oncology.

EPO'S PROTECTIVE ROLE

The molecular mechanisms underlying the therapeutic effect of EPO need to be better explained. Some authors observed that ESRD patients precociously treated with EPO show a slower progression of renal damage.[39] Several authors have hypothesized a protective effect of EPO against ischemic[40–42] and cytotoxic[43] renal damage in animals. Although their studies were observational and conducted on heterogeneous populations, some data suggest that a chronic treatment with low doses of EPO could really play a protective role.[44] This is also suggested by the presence of the EPO receptor and its mRNA in extracts of several organs, including the kidneys of humans, rats, and mice.[45] In some experimental models of ischemic damage, the administration of EPO was able to reduce the ischemic area in different organs, such as the heart and brain.[46–49] In these studies, characterized by a short follow-up period, the attention was focused on EPO's proliferative and anti-apoptotic effects (increased PCNA, reduced apoptosis studied by TUNEL assay). These effects, probably induced by the activation of EPO receptor,[50] have also been demonstrated in ischemia-reperfusion models in brain and in cell cultures. EPO's direct effect on the functional recovery after ischemia could be due to the activation of signal transduction systems, like Jak and STAT; STAT3, for example, can inhibit the apoptosis cascade. The STAT family includes cytoplasmatic transcription factors that receive signals coming from cytokines receptors on the cell surface and send them to the nucleus. Currently, seven members of STAT family are known, codified by different genes (i.e., STAT-1, STAT-2, STAT-3, STAT-4, STAT-5a, STAT-5b and STAT-6). STATs are activated by phosphorylation after the link between the cytokines and their specific receptors. STAT1α, STAT3, and STAT5 can be activated from erythropoietin but also from GM-CSF and IL-3. The single cytokine specificity is probably due to the different times of activation of the STATs' target genes.[51] The better known mechanism of renal protection in ischemia-reperfusion experimental models is represented by the reduction of apoptosis and the increase of proximal tubular cells proliferation, as demonstrated by the expression of proliferation markers like PCNA. Apoptosis has been studied by different methods (TUNEL assay, ISEL, Caspase-3 activity) and on different models of renal injury (see Table 1). Some authors analyzed PCNA alone or together with an evaluation of apoptosis in order to define the mitogenic potential of EPO therapy. Nevertheless, while the anti-apoptotic effect of EPO is quite clear in these experimental models, there are not enough data to demonstrate EPO's proliferative effect on kidney in vivo. This effect, in fact, should be analyzed using a specific marker of mitosis, such as Ki67, and not only by PCNA. Thus, the mechanisms through which EPO would protect the kidney from ischemic-hypoxic damage still need to be verified. Moreover, the role played by STAT and Jak in EPO-induced renal protection should be investigated, as these systems are certainly involved in the signal transduction from EPO receptor in erythroid cell lines[52],[53] and consequently in the expression of ant-apoptotic geneses, such as Bax e Bcl-Xl (see Table 2).[54],[55]

Table 1 Markers of direct action of EPO therapy to inhibit apoptosis and increase proliferation index

Damage models	Apoptosis assay	PCNA	Other antigen
Bagnis et al.[43]	None	Yes	None
Vesey et al.[41]	In situ end labelling (ISEL) nuclear positivity was used for quantification of apoptosis in vitro and in vivo	Yes	None
Tanaka H, et al. Expression and function of Ets-1 during experimental acute renal failure in rats. J Am Soc Nephrol. 2004;15:3083–3092.	None	Yes	None
Spandou E, et al. Erythropoietin attenuates renal injury in experimental acute renal failure ischemic/reperfusion model. NDT. 2006;21:330–336.	TUNEL assay	No	Bax NFkB
Sharples et al.[42]	TUNEL assay	No	Caspase-3 activity
Johnson DW, et al. Delayed administration of darbepoetin or erythropoietin protects against ischemic acute renal injury and failure. Kidney Int. 2006 May; 69(10):1806–1813.	Quantification of apoptosis	No	Bcl-2,Bcl-XL, Bax

Table 2 Animal models of EPO administration in renal ischemia/reperfusion

Histology and Western blotting	EPO dose	Model	Jak 1	Jak 2	Stat1	Stat3	Stat5	Others
Ates E, et al. Protective effect of erythropoietin on renal ischemia and reperfusion injury. J Surg. 2005 Dec; 75(12):1100–1105	1000 U/kg 2 h before I/R	Rat I/R	No	No	No	No	No	TNFα IL-2
Spandou E, et al. Erythropoietin attenuates renal injury in experimental acute renal failure ischemic/reperfusion model. NDT. 2006;21:330–336.	500 U/kg i.p. 20 min. before I/R	Rat I/R	No	No	No	No	No	Bax and NF-κB
Patel NS, et al. Pretreatment with EPO reduces the injury and dysfunction caused by ischemia/reperfusion in the mouse kidney in vivo. Kidney Int. 2004 Sep; 66(3):983–989	1000 U/kg/day pretreated and before reperfusion	Mouse I/R	No	No	No	No	No	None
Sharples et al.[42]	300 U/kg either pre-ischemia or just before reperfusion	Rat I/R	No	No	No	No	No	None
Gong H, et al. EPO and alpha-MSH prevent ischemia/reperfusion-induced down-regulation of AQPs and sodium transporters in rat kidney. Kidney Int. 2004 Aug; 66(2):683–695	500 U/Kg at the reperfusion or four hours after	Rat I/R	No	No	No	No	No	None
Vesey et al.[63]	In vitro 6.25–400 U/mL in vivo 5000 U/kg	Culture with different [O^2] and rat I/R	No	No	No	No	No	None
Nemoto et al.[62]	500 or 3000 U/kg starting at time of ischemia	Rat I/R	No	No	No	No	No	None
Johnson DW, et al. Delayed administration of darbepoetin or erythropoietin protects against ischemic acute renal injury and failure. Kidney Int. 2006 May; 69(10):1806–1813.	5000 U/kg at the time of ischemia or six hours before reperfusion	Rat I/R	No	No	No	No	No	None

Note. None of the authors described intra-cellular signal activation. Nuclear factor NF-kB was studied in one work only.

BEYOND THE DIRECT EFFECTS

The super-family of cytokines receptors includes several receptors that show similar characteristics maintained in the different species, with highly homologous regions and sub-units.[56],[57] Moreover, GM-CSF, G-CSF, and IL-2 show a high homology, with the α-chain of EPO receptor and a common β-chain that joins the link increasing the affinity.[58],[59] As well as EPO, IL3, IL5, IL6, and IFNγ are also involved in inflammatory, ischemic, and cytotoxic events. All of these cytokines induce phosphorylation and the subsequent activation of Jak2 and STAT3. Because the stimuli activating Jak and STAT are very heterogeneous, it is not easy to identify a univocal mechanism. For instance, some data suggest that Stat1-α may play a key role in EPO-induced cell proliferation,[60] while the role of Stat5 is still unclear. Some authors hypothesize that Stat5 activation may be necessary but not sufficient to stimulate the maximal EPO-induced cell growth.[61] This analysis is further complicated by the variability in experimental methods and models. Furthermore, because the cytokine receptors show a certain homology, it is possible that a ligand binds different receptors in a non-specific way: the protective effect of EPO, for example, seems to be dose-dependent both in vitro and in vivo.[62] In particular, its proliferative effects increase proportionally with the hormone concentration in the culture medium.[63] In the experimental models, EPO dose is from 10 to 100 times higher than in the therapy of patients with RF. Such a high concentration may saturate all of the available specific receptors and then activate in a non-specific manner different signal transduction systems in different tissues and organs. This datum has been seen in a heart I/R model.[64] Moreover, some works demonstrated STAT3 and STAT5 activation in in vivo and in vitro models in which high concentrations of EPO were used.[65–67] Although Jak2 and Stat5 are involved in the expansion of erythroid precursors,[68],[69] the activation of other proteins belonging to the same family is unusual: maybe an excess of EPO could saturate its specific receptors and then bind other receptors and activate alternative patterns through a low-affinity link. It has been demonstrated that EPO can mobilize stem cells through a mechanism that is not well known.[70] EPO administration increases the plasmatic levels and the renal expression of VEGF:[71–73] this effect may accelerate the post-ischemic repairing, mainly through a vascular stimulation and the recruitment of stem cells.[74],[75] The anti-inflammatory effect of EPO should also be considered: EPO can inhibit pro-inflammatory cytokines such as TNF, MCP-1, and IL-6 in homogenized cerebral tissue 24 hours after the medial cerebral artery occlusion.[76] Furthermore, EPO delays the increase of TNF levels without altering its maximal values and reduces Il-6 concentration.[77]

CONCLUSIONS

EPO's role in the complex network of cytokines still needs to be clarified. This hormone, the plasmatic concentration of which is about 4–26 mU/mL, has a well-defined primary target. It seems reductive to attribute to EPO a protective effect toward the ischemia-reperfusion renal damage involving only its anti-apoptotic properties. Although the ischemic kidney treated with EPO shows a reduction in the number of apoptotic cells, this cannot be enough to explain the functional recovery described by many authors. Nevertheless, the increased number of PCNA-positive cells cannot be considered a demonstration of a hypothetic EPO's proliferative effect.

To better understand this point, we should define the exact localization of EPO receptor in the different organs, tissues, and cell types in physiologic and pathologic conditions and verify a correspondence between the receptor position and the observed effects. It has been widely demonstrated that EPO and EPO receptor play a key role during prenatal growth,[78] not only stimulating red blood cells production but also enhancing vascular system development. These different effects are likely related to the activation of different signal transduction pathways. In this context, the role of EPO receptor seems to be particularly relevant, because it is expressed not only in the kidney but also in other organs. The behavior of this protein in the different cell types and the destiny of EPO after the saturation of all the available specific receptors still need to be defined. The activation of non-specific receptors through a low-affinity link may occur in particular experimental conditions: similar conditions could be created by high doses of exogenous EPO (for example, during the treatment of oncological patients) or by the administration of long-acting molecules. Further investigation is needed to determine new therapeutic applications for EPO and other analogous hormones. Very long-acting molecules[79] or molecules with cyto-protective but no erythropoietic effects[80],[81] may represent useful tools in the study of the molecular mechanisms underlying EPO's action, and they may have a rapid and safe therapeutic application.

Notes

23. Lin F, et al. Hematopoietic stem cells contribute to the regeneration of renal tubules after renal ischemia-reperfusion injury in mice. J Am Soc Nephrol. 2003;141188–1199. See also: Kale S, et al. Bone marrow stem cells contribute to repair of the ischemically injured renal tubule. J Clin Invest. 2003;112:42–49.

27. Fink JC, et al. Use of erythropoietin before the initiation of dialysis and its impact on mortalità. Am. J Kidney Dis.2001;37:348–355. Also see: Valderràbano F, et al. Pre-dialisys survey on anaemia management. Nephrol Dial Transplant2003;18:89–100.

29. Hayashi T, et al. Cardiovascular effect of normalizing the hematocrit level during erythropoietin therapy in predialysis patients with chronic renal failure. Am. J. Kidney Dis.2000;35:250–256. See also: Portolès J, et al. Cardiovascular effects of recombinant human erythropoietin in predialysis patients. Am. J. Kidney Dis.1997;29:541–548.