Erythropoietin (EPO), produced by the kidney in response to low serum oxygen, is a major stimulant for red blood cell production. Though this glycoprotein has only been medically approved for the treatment of anemia, there are indications that it can do more than just beef up erythrocyte numbers. Studies suggest that the hematopoietic factor can also protect neurons (see Siren et al., 2001), and that it may even improve cognition, properties that are of particular interest to those studying Alzheimer or other neurodegenerative diseases. These findings have spurred clinical trials to evaluate the benefit of EPO in several neurodegenerative paradigms, including stroke and schizophrenia. On November 16, at the 35th Annual Meeting of the Society for Neuroscience in Washington, DC, scientists gathered to address the role of EPO in the nervous system and to discuss its potential therapeutic benefit to both the peripheral and central nervous systems (CNS).

The first red-blooded evidence that EPO may be important in the CNS came in 1995, almost 20 years after the growth factor was purified. Max Gassman and colleagues at the University of Zurich-Irchel, Switzerland, detected EPO, and its receptor, in mouse brain (see Digicaylioglu et al., 1995). This revelation raised many questions, most notably about what a hematopoietic factor is doing in the CNS, given that red blood cells are not produced there. Ten years later, some of those questions have been answered, but pieces to the puzzle are continually being fitted into place.

It is still not fully clear how EPO expression is regulated in the nervous system, for example. At the Washington mini-symposium, Juan Chavez, from the Burke/Cornell Medical Research Institute, White Plains, New York, brought the issue up-to-date. He revealed that in the CNS, control of EPO expression is similar to that in the kidney, which produces the EPO that stimulates erythropoiesis. In the renal gland, expression of the protein is markedly stimulated by another protein called hypoxia inducible factor (Hif), which only survives when serum oxygen is low. When oxygen levels are normal, Hif is rapidly degraded by the kidney’s ubiquitin proteasome system. Chavez and colleagues found that a similar signaling process occurs in the brain, primarily in astrocytes where Hif levels increase under conditions of oxygen/glucose deprivation. However, he also reported that there is a subtle difference between renal and CNS regulation. In the brain, EPO appears to be controlled by just one member of the Hif family, Hif2. This became apparent after Chavez made conditional knockout mice that lose the Hif1 gene. These animals can still make CNS EPO, suggesting that Hif2, found only in endothelial cells and astrocytes, is the major hypoxia inducible factor in the brain (the role of the third Hif member, Hif3, is unclear). In support of this, Chavez reported that Hif2 (but not Hif1) binds to the astrocyte EPO promoter.

These findings are important to consider in the context of neuronal protection. Neurons die within minutes if they are deprived of oxygen, whereas astrocytes can survive for up to two hours in those conditions. But in co-cultures, neurons survive much longer without oxygen, reported Chavez. So how, exactly, does EPO protect neurons and can the mechanism be exploited for therapeutic purposes?

Murat Digicaylioglu and Stuart Lipton, at the Burnham Institute, La Jolla, California, answered the first part of that question several years ago when they showed that pretreatment of neurons with EPO protects them against ischemia-induced apoptosis, or programmed cell death (see ARF related news story). Last year they showed that the need for pretreatment could be dispensed with if EPO was added to cerebrocortical neurons together with insulin-like growth factor 1 (IGF-1), which has anti-apoptotic properties of its own. They also reported that this synergy was mediated by activation of Akt (protein kinase B), which has been implicated in the pathology of Alzheimer disease (see Digicaylioglu et al., 2004 and ARF related news story). At the Washington mini-symposium, Digicaylioglu revealed that this combination therapy also works in vivo. He reported that administering the combination intranasally to animals suffering from experimental stroke lowered infarct volumes and improved cognition. Though EPO is currently in clinical trials for ischemic stroke, these findings suggest that a combination therapy might prove an even better choice.

EPO as a therapeutic may not be restricted to ischemic events, either. Hannelore Ehrenreich, from the Max Planck Institute for Experimental Medicine, Gottingen, Germany, reported results of a clinical trial to measure the potential benefit of EPO in patients with schizophrenia. The rationale for this study was based on work in animal models and in studies of human patients. In mouse models of schizophrenia, for example, recombinant human EPO improved cognition, while immunohistochemical analysis of human autopsy samples revealed that, compared to normal subjects, people with schizophrenia have much higher levels of EPO receptor in both the cerebral cortex and the hippocampus. Together, these findings hinted that EPO supplementation may benefit schizophrenic patients, a suggestion that was supported by imaging studies. The latter demonstrated that EPO is taken up more readily in the brains of those with schizophrenia (see Ehrenreich et al., 2004).

Based on these observations, Ehrenreich and colleagues initiated a proof-of-principle trial. They administered EPO as a neuroprotective “add-on” treatment to patients taking neuroleptics. Out of 44 patients that started the 12-week trial, 39 finished. Ehrenreich reported that the major outcome was dramatically improved cognition—though EPO had no effect on psychic pathology, as judged by scores in the Positive And Negative Syndrome Scale (PANSS), a commonly used test to evaluate schizophrenia patients. Further details of the trial can be found at Ehrenreich’s website.

The results of this small trial suggest that EPO can have a potent neuroprotective effect in humans, slowing or preventing the recently documented neurodegenerative component of schizophrenia (see Thompson et al., 2001). This raises the question of whether the glycoprotein may also benefit patients with other neurodegenerative diseases, such as Alzheimer disease (AD). Only a clinical trial can answer this question, but Ehrenreich and colleagues have just published data showing that in mice, administration of EPO for as little as two weeks prevents the slow neurodegeneration that continues for months after an acute trauma (see Siren et al., 2005). These findings, coupled with the clinical trial data, suggest that the hematopoietic factor can have a profound impact on the neurodegenerative process and might be beneficial in AD or other degenerative diseases. In fact, Wendy Campana, from the University of California at San Diego, and colleagues reported in a poster that EPO receptor mRNA is downregulated in AD brain compared to normal controls. The loss was significant and greatest in those with later-stage disease, who had about 30 percent lower expression of the receptor. The authors also reported that EPO levels are lower in AD patients, too, though not to the same extent. Finally, Campana and colleagues reported that transgenic mice producing human mutant amyloid-β precursor protein have 50-60 percent less EpoR mRNA in the frontal cortex by the time they are six months old. All these findings suggest that EPO might be beneficial for AD patients.

In the peripheral nervous system, too, EPO may turn out to give neurons a boost and, again, this may be dependent on the action of neuronal helper cells. Campana showed several years ago that EPO attenuates apoptosis when it is initiated by damage to spinal axons in rats (see Sekiguchi et al., 2003) and that it can protect dorsal root ganglia neurons against injury in vivo (see Campana and Myers, 2003). In a very recent paper, she reported that injured sciatic nerve fibers upregulate production of endogenous EPO, and that one of the targets for this EPO is its cognate receptor on Schwann cells. In fact, Campana and colleagues reported that these cells, which produce the myelin sheath that surrounds axons, proliferate rapidly if they are treated with EPO (see Li et al., 2005). At the Washington symposium, she revealed that EPO receptors are massively upregulated on the surface of Schwann cells within a minute after addition of the growth factor. These findings are relevant to axonal degeneration diseases such as multiple sclerosis (MS), and, in fact, studies have shown that EPO can lead to improvements in animal models of MS (see, for example, Agnello et al., 2002 and Savino et al., 2005).

The importance of the Schwann cell connection is supported by work from the labs of Sanjay Keswani and Ahmet Hoke (who chaired the mini-symposium) at Johns Hopkins University, Baltimore, Maryland. Their work brings us back full circle to the question of EPO regulation. Last year they showed that Schwann cells produce EPO in response to nitric oxide (NO) (see Keswani et al., 2004). At the symposium, Keswani reported that NO exerts its effects by increasing levels of Hif1 in Schwann cells. This finding appears to mark a major difference between regulation of EPO in the PNS and CNS given Chavez’s findings that Hif2 mediates production of the glycoprotein in CNS astrocytes (see above). But no matter which of the Hif family is involved, it could turn out that EPO acts on the same downstream targets in both central and peripheral nervous systems. Keswani reported that EPO protects peripheral axons in a phosphatidyl-inositol-3-kinase (PI3K)-dependent manner. This is curious, given that Digicaylioglu and colleagues found that Akt, which mediates PI3K signaling, plays a role in EPO/IGF-1-mediated protection of cortical neurons (see above). Even more curious for AD researchers, perhaps, is Keswani’s finding that phosphorylation of glycogen synthase kinase 3β (GSK3β) is essential for EPO’s effects. GSK3β, of course, is well-known as a tau kinase that is silenced by Akt (see ARF related SfN story). The findings give a tantalizing hint that EPO might protect against tau phosphorylation, a major pathological hallmark of AD.—Tom Fagan.


  1. Grosche from the Nun Study demonstrated with autopsy that brain cell death: i.e. brain atrophy, is what produces the dementia of AD rather than the density of plaque and tangle deposits. Heme deficiency has been identified as one cause of brain cell death. Aluminum overload and folic acid deficiency are key but often overlooked factors in brain atrophy. Both interfere with response to erythropoietin: e.g. oral aluminum dosing of rats with normal renal function has been shown to directly impair erythropoiesis.


    . Hippocampal volume as an index of Alzheimer neuropathology: findings from the Nun Study. Neurology. 2002 May 28;58(10):1476-82. PubMed.

    . Heme deficiency may be a factor in the mitochondrial and neuronal decay of aging. Proc Natl Acad Sci U S A. 2002 Nov 12;99(23):14807-12. Epub 2002 Nov 4 PubMed.

    . Hyporesponsiveness to recombinant human erythropoietin. Nephrol Dial Transplant. 2001;16 Suppl 7:25-8. PubMed.

    . Oral aluminum administration to rats wih normal renal function. 1. Impairment of erythropoiesis. Hum Exp Toxicol. 1998 Jun;17(6):312-7. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Erythropoietin a Neuroprotectant?
  2. Presenilins Work Overtime to Control Akt, Tau, and Aβ—n-3 Fatty Acids Aid and Abet
  3. SfN: Where, How Does Intraneuronal Aβ Pack Its Punch? Part 4

Paper Citations

  1. . Erythropoietin prevents neuronal apoptosis after cerebral ischemia and metabolic stress. Proc Natl Acad Sci U S A. 2001 Mar 27;98(7):4044-9. PubMed.
  2. . Localization of specific erythropoietin binding sites in defined areas of the mouse brain. Proc Natl Acad Sci U S A. 1995 Apr 25;92(9):3717-20. PubMed.
  3. . Acute neuroprotective synergy of erythropoietin and insulin-like growth factor I. Proc Natl Acad Sci U S A. 2004 Jun 29;101(26):9855-60. PubMed.
  4. . Erythropoietin: a candidate compound for neuroprotection in schizophrenia. Mol Psychiatry. 2004 Jan;9(1):42-54. PubMed.
  5. . Mapping adolescent brain change reveals dynamic wave of accelerated gray matter loss in very early-onset schizophrenia. Proc Natl Acad Sci U S A. 2001 Sep 25;98(20):11650-5. PubMed.
  6. . Global brain atrophy after unilateral parietal lesion and its prevention by erythropoietin. Brain. 2006 Feb;129(Pt 2):480-9. PubMed.
  7. . ISSLS prize winner: Erythropoietin inhibits spinal neuronal apoptosis and pain following nerve root crush. Spine (Phila Pa 1976). 2003 Dec 1;28(23):2577-84. PubMed.
  8. . Exogenous erythropoietin protects against dorsal root ganglion apoptosis and pain following peripheral nerve injury. Eur J Neurosci. 2003 Sep;18(6):1497-506. PubMed.
  9. . Schwann cells express erythropoietin receptor and represent a major target for Epo in peripheral nerve injury. Glia. 2005 Sep;51(4):254-65. PubMed.
  10. . Erythropoietin exerts an anti-inflammatory effect on the CNS in a model of experimental autoimmune encephalomyelitis. Brain Res. 2002 Oct 11;952(1):128-34. PubMed.
  11. . Delayed administration of erythropoietin and its non-erythropoietic derivatives ameliorates chronic murine autoimmune encephalomyelitis. J Neuroimmunol. 2006 Mar;172(1-2):27-37. PubMed.
  12. . A novel endogenous erythropoietin mediated pathway prevents axonal degeneration. Ann Neurol. 2004 Dec;56(6):815-26. PubMed.

External Citations

  1. website

Further Reading

No Available Further Reading