Erythropoietin and diabetes mellitus

Abstract

Erythropoietin (EPO) is a 30.4 kDa growth factor and cytokine that governs cell proliferation, immune modulation, metabolic homeostasis, vascular function, and cytoprotection. EPO is under investigation for the treatment of variety of diseases, but appears especially suited for the treatment of disorders of metabolism that include diabetes mellitus (DM). DM and the complications of this disease impact a significant portion of the global population leading to disability and death with currently limited therapeutic options. In addition to its utility for the treatment of anemia, EPO can improve cardiac function, reduce fatigue, and improve cognition in patients with DM as well as regulate cellular energy metabolism, obesity, tissue repair and regeneration, apoptosis, and autophagy in experimental models of DM. Yet, EPO can have adverse effects that involve the vasculature system and unchecked cellular proliferation. Critical to the cytoprotective capacity and the potential for a positive clinical outcome with EPO are the control of signal transduction pathways that include protein kinase B, the mechanistic target of rapamycin, Wnt signaling, mammalian forkhead transcription factors of the O class, silent mating type information regulation 2 homolog 1 (Saccharomyces cerevisiae), and AMP activated protein kinase. Therapeutic strategies that can specifically target and control EPO and its signaling pathways hold great promise for the development of new and effective clinical treatments for DM and the complications of this disorder.

Key Words: Protein kinase B, AMP activated protein kinase, Apoptosis, Autophagy, Forkhead, Metabolism, Factors of the O class, Diabetes mellitus, Erythropoietin, Stem cells, Silent mating type information regulation 2 homolog 1, Oxidative stress, Wnt1 inducible signaling pathway protein 1, Wnt

Core tip: Erythropoietin and the downstream signaling pathways of this cytokine that include protein kinase B, mechanistic target of rapamycin, Wnt signaling, Factors of the O class proteins, silent mating type information regulation 2 homolog 1 (Saccharomyces cerevisiae), and AMP activated protein kinase offer new avenues for the development of novel treatments for diabetes mellitus and the complications of this disease.

ERYTHROPOIETIN: DISCOVERY AND BIOLOGY

The concept of circulating agents that travel throughout the body may have initially originated from Ernest Starling[1]. In 1905 at the Royal College of Surgeons, Sterling introduced the term “hormones”, a term with Greek origins meaning to “excite” or “arouse”, to depict the action of chemicals that are dispersed in the body and can target specific organs. Earlier work prior to the presentation by Sterling also described processes that could come under the description as being defined as “hormonal”. Claude Bernard described the chemical release of glucose that was processed from glycogen in the liver[2]. Arnold Adolphe Berthold, another pioneer, also described messenger signals that could communicate among the different bodily organs[3].

Interestingly, almost as a counterpart to the discussions provided by Starling, Carnot et al[4] in 1906 presented the agent “hemopoietine”. This agent was detected in the blood of rabbits after prompted by bleeding that led to the production of immature erythrocytes in untreated rabbits. Subsequent work by other investigators also showed that bled animals could result in prominent reticulocytosis in the plasma[5-7]. Later, the agent responsible for reticulocytosis was termed erythropoietin (EPO). EPO was linked to depressed oxygen levels and was shown to increase hemoglobin levels in parabiotic rat experiments when one of the two rats experienced hypoxia[8]. Subsequently, purification of the EPO protein in humans was achieved and cloning of the EPO gene fostered recombinant EPO (rhEPO) production for clinical treatments[9,10].

EPO is located on chromosome 7 and is a single copy in a 5.4 kb region of the genomic DNA[11]. The EPO gene encodes for a polypeptide chain that has initially 193 amino acids. A 27 amino acid hydrophobic secretory leader at the amino-terminal to result in a 166 amino acid peptide in the EPO protein is then cleaved[12]. Additional post-translational processing occurs with the removal of a carboxy-terminal arginine¹⁶⁶ in the mature human and rhEPO to lead to a protein of 30.4 kDa with 165 amino acids[13-16].

EPO has four glycosylated chains that include three N-linked and one O-linked acidic oligosaccharide side chains[17]. The N-linked glycosylation sites are at aspartate²⁴, aspartate³⁸, and aspartate⁸³ and the O-linked glycosylation site is at serine¹²⁶. Both the production and secretion of the mature EPO protein is dependent upon N- and O-linked chain integrity[18]. Replacement of asparagine³⁸ and asparagine⁸³ by glutamate or the replacement of serine¹²⁶ by glycine can impair EPO production and secretion[19].

Several factors determine the biological activity of EPO[20]. The two disulfide bonds formed between cysteine⁷ and cysteine¹⁶⁰ as well as cysteine²⁹ and cysteine³³ control the function of EPO[21]. EPO biological activity is lost with reduction of these disulfide bonds and with alkylation of the sulfhydryl groups. Almost 85% of EPO biological activity is restored with re-oxidization of EPO after reduction by guanidine[22]. In addition, EPO biological activity is maintained by the by the glycosylated chains[23] and EPO stability is fostered by the carbohydrate chains[24]. Free radical degradation of EPO is limited by both the glycosylated chains[23] and the oligosaccharides[25].

Currently, erythropoiesis-stimulating agents including EPO are approved for the treatment of anemia that results from chronic kidney failure, chemotherapy, human immunodeficiency virus, and to limit the number of blood transfusions for surgery[21,26]. The principal source for the production and secretion of EPO are the kidney peritubular interstitial cells[27]. Other organs that include the brain, uterus, and liver are also responsible for EPO production and secretion[17,27-30]. Expression of EPO is controlled by changes in oxygen tension and not by the concentration of red blood cells[28,31,32]. Hypoxia-inducible factor 1 (HIF-1) can control EPO expression and the EPO receptor (EPOR) to increase the production of EPO[11,28,33,34]. EPO and EPOR gene transcription occurs following HIF-1 activation. This gene transcription is governed by the transcription enhancer region in the 3’-flanking region of the EPO gene that binds to HIF-1[11,14]. HIF-1 also can foster pathways that provide cellular protection against injury[35-37]. Of note, EPO also can be generated from stimuli that may not directly involve hypoxia. During maturation of the brain that may be exposed to various toxic elements, EPO blood levels may be elevated and associated with greater disability[38]. Elevated EPO serum concentrations have been reported following xenon anesthesia in cardiac surgery[39]. Agents that decrease inflammation in cerebral microglia have been recently shown to lead to the release of EPO[40] and infection with malaria can result in significant serum levels of EPO[41]. Under some conditions during chronic hyperglycemia in adults, EPO levels may be depressed[42]. Conversely, EPO in the amniotic fluid of diabetic patients can be elevated and be suggestive of perinatal complications[43]. Furthermore, trophic factors such as insulin can stimulate EPO production in specific cells such as astrocytes[44].

EPO, OXIDATIVE STRESS, AND CELL SURVIVAL

As a cytoprotective agent, EPO promotes cellular survival, at least in part, through the control of oxidative stress mediated cell injury[45,46]. Reactive oxygen species (ROS) are released during oxidative stress[47]. This in turn can cause mitochondrial injury, DNA damage, and protein misfolding[48-52].

Following the generation of ROS, cell death pathways of programmed cell death can ultimately determine cell survival[53-62]. Two particular pathways of programmed cell death involve autophagy[50,63-65] and apoptosis[15,55,57,66,67]. EPO prevents autophagic cell injury in glomerular mesangial cells during lipopolysaccharide exposure[68]. Administration of EPO also limits excessive autophagy that precedes apoptosis during experimental neonatal necrotizing enterocolitis[69]. During hyperoxia exposure and oxygen toxicity to the developing rodent brain, EPO has been shown to modify the activity of autophagy and limit neonatal brain damage[70].

In regards to apoptotic cell death, EPO prevents apoptotic injury during oxidative stress in endothelial progenitor cells[71] and attenuates neuroinflammation that can result in apoptosis[72]. EPO can assist with erythroid differentiation and prevent cellular apoptosis[73] as well as promote ventricular-subventricular zone neurogenesis and oligodendrogenesis[74]. Derivatives of EPO, such as glutaraldehyde-EPO, can protect renal cells from apoptosis during ischemia/re-perfusion injury and oxidative stress[75]. Administration of EPO also can block apoptotic cell death during neuronal kainate-induced oxidative stress[76], wound injury[77], vascular oxygen-glucose deprivation[78-80], loss of protective zinc finger transcription factors[81], anoxia[82-84], astroglial glutamate toxicity[85], beta-amyloid (Aβ) toxicity[86-90], renal adriamycin-induced nephropathy[91], ischemic brain injury[92], and multi-organ dysfunction induced by thermal injury[93]. In addition, EPO is protective against retinal disease[94], sepsis[95,96], advanced glycation endproducts (AGEs) exposure in Schwann cells[97], elevated glucose[78,98-102], free radicals[103-108], and toxins that lead to microglial injury[30,40,90,94,109].

SIGNAL TRANSDUCTION PATHWAYS FOR EPO

EPO cytoprotection is tied to a number of cell pathways[3]. In particular, phosphoinositide 3-kinase (PI 3-K) and protein kinase B (Akt) can lead to increased cellular survival with EPO (Figure 1). PI 3-K phosphorylates membrane lipids and controls Akt transition from the cytosol to the plasma membrane. Phosphorylation of Akt occurs at serine⁴⁷³ and threonine³⁰⁸ by phosphoinositide dependent kinase (PDK) PDK1 and PDK2[110-112]. EPO leads to Akt phosphorylation on serine⁴⁷³ to activate this kinase. EPO uses the Akt pathway to protect against autophagy and apoptosis injury in gastrointestinal disease[69], maintain vascular integrity and reduce inflammation[113], limit Aβ toxicity in microglia and neurons[90,114-116], reduce injury from sepsis[95,117], increase survival in cardiomyocytes during cardiac hypoxic/re-oxygenation injury[118], and block oxidative stress injury[78,82,104,105,119-122]. Akt in conjunction with EPO also improves the function of cells. For example, EPO activates Akt to increase the adhesive properties of endothelial cells and improve the vasculogenic potential of peripheral blood mononuclear cells[123].

Figure 1 Erythropoietin signal transduction pathways that can lead to clinical benefit during diabetes mellitus. EPO governs a number of signal transduction pathways that involve protein kinase B (Akt), the mechanistic target of rapamycin (mTOR), Wnt and WISP1 signaling, mammalian forkhead transcription factors of the O class (FoxO), silent mating type information regulation 2 homolog 1 (Saccharomyces cerevisiae) (SIRT1), and AMP activated protein kinase (AMPK). EPO: Erythropoietin; Akt: Protein kinase B; mTOR: Mechanistic target of rapamycin; FoxO: Factors of the O class; SIRT1: Silent mating type information regulation 2 homolog 1 (Saccharomyces cerevisiae); AMPK: AMP activated protein kinase.

The mechanistic target of rapamycin (mTOR) is closely linked to PI 3-K and Akt[124] (Figure 1). mTOR is a 289-kDa serine/threonine protein kinase that is encoded by a single gene FRAP1[124,125]. mTOR is important for the function of mTOR Complex 1 (mTORC1) and mTOR Complex 2 (mTORC2)[126-129]. Neurons are protected against sepsis during exposure to EPO and activation of mTOR[95]. EPO prevents microglial cell injury through mTOR activation during oxidative stress[109] and Aβ toxicity[90]. During oxygen-glucose exposure in neurons, EPO affects multiple pathways of mTOR signaling[130] to include Akt and proline rich Akt substrate 40 kDa (PRAS40) to increase neuronal survival[79]. EPO and mTOR are required for the differentiation of neural precursor cells[131] and to control bone homeostasis with osteoblastogenesis and osteoclastogenesis[132]. EPO through mTOR can mediate resistance to hypoxia and oxidative stress in retinal progenitor cells[133] and also protect against increased activity of autophagy in epithelial cells[69]. Activation of mTOR prevents the induction of autophagy by phosphorylating autophagic related genes (Atg) and proteins that include Atg13 and ULKs to inhibit the UNC like kinase complex ULK-Atg13-FIP200[128]. Under some conditions, the concentration of EPO and activity of mTOR may be important for the degree of cellular protection that can be achieved. Elevated concentrations of EPO have been reported to lead to decreased phosphorylation and activity of mTOR with increased apoptotic cell death[134]. Increased mTOR activity also is tied to tumor cell growth[135-138].

Closely associated to the protective pathways of Akt and mTOR are the wingless pathways of Wnt proteins[139] (Figure 1). Crosstalk occurs among Wnt signaling pathways, Akt, and mTOR[140] to foster cellular survival during Aβ toxicity[141,142], reduce cerebral ischemia[143,144], promote progenitor cell activation during intestinal inflammation[145], prevent neuronal cell loss[146], limit 6-hydroxydopamine toxicity[147], enhance microglial and macrophage survival and function[148,149], and increase tissue fibrosis[150]. EPO employs the Wnt pathway to lead to cellular protection. During renal ischemia and reperfusion, EPO limits tubular cell apoptosis by increasing the expression of Wnt7b and β-catenin as well as by down-regulating specific micro-RNAs (miRNA)[151,152]. Through Wnt1, EPO protects against elevated glucose exposure in cerebral endothelial cells and maintains the expression of Wnt1[100]. In addition, EPO uses Wnt signaling to prevent immune cell loss during oxidative stress[109], prevent Aβ toxicity in microglia[90], limit the activity of forkhead transcription factors that result in apoptosis[99,153], and maintain the survival of mesenchymal stem cells[154]. Of note, both EPO and the pathways of Wnt signaling are proliferative in nature and have the potential to lead to tumorigenesis. For example, prolonged exposure of growth factors such as EPO that rely upon Wnt signaling can result in inflammation, blood-brain barrier injury[155], and tumor growth[156-158].

Cellular protection with EPO that relies upon Wnt signaling also can be associated with the modulation of mammalian forkhead transcription factors[159]. Mammalian FOXO proteins are assigned to the O class of the forkhead box class transcription factors[160,161] (Figure 1). These transcription factors consist of FOXO1, FOXO3, FOXO4, and FOXO6 and exist throughout the body[162]. FoxO proteins can impact cellular survival[163] and are homologous to DAuer Formation-16 (DAF-16), a transcription factor in Caenorhabditis elegans, that leads to lifespan extension and affects insulin signaling[164,165]. Under many circumstances, the activation of FoxO proteins results in apoptotic cell death[153]. FoxO3a expression increases in the hippocampus during cerebral ischemia[166] and FoxO3a may lead to cell cycle induction that can promote neuronal apoptotic cell death[167]. Loss of FoxO3a expression and prevention of nuclear shuttling of FoxO3a in microglial cells and neurons results in increased survival during oxidative stress[146,148]. Inhibitory phosphorylation of FoxO3a and the nuclear export of FoxO3a during periods of elevated glucose also protects vascular cells[80,99,168,169] and neuronal cells[170].

In endothelial cells, EPO uses Wnt1 to block FoxO3a activity and maintain cerebral endothelial survival during elevated glucose[99]. Without Wnt signaling, EPO also has been shown to phosphorylate FoxO3a and lead to its inactivation to block apoptosis in neuronal cells[73]. EPO can prevent endothelial cell injury during oxygen-glucose deprivation by preventing FoxO3a nuclear subcellular trafficking that would lead to “pro-apoptotic” protein transcription and translation[20,80]. EPO can oversee stem cell proliferation through FoxO protein regulation. Through the control of FoxO3a activity, EPO promotes the development of erythroid progenitor cells[57,73,171,172].

FoxO protein activity is controlled by post-translation protein modifications that involve phosphorylation, ubiquitylation, and acetylation[162,173]. In regards to acetylation, FoxO proteins are deacetylated by histone deacetylases that includes the silent mating type information regulation 2 homolog 1 (Saccharomyces cerevisiae) (SIRT1)[54] (Figure 1). SIRT1 deacetylation of FoxO proteins can influence autophagic pathways such that glucose deprivation leads to increases in autophagic flux that maintain left ventricular function during periods of starvation[174]. SIRT1 may be required to promote cortical bone formation with osteoblast progenitors by deacetylation of FoxOs and preventing FoxO protein binding to β-catenin to inhibit Wnt signaling[175]. However, the degree of SIRT1 expression in relation to FoxO protein activity may be a significant determinant for cellular survival[160,161]. For example, during exercise a controlled up-regulation of FoxO3a and SIRT1 expression in cardiac tissue may be important to improve cell survival[176]. During oxidative stress, cell injury may be reduced with catalase expression regulated by FoxO1a expression and SIRT1 levels less than 7.5-fold. However, decreased cardiac function and apoptotic cell death in cardiomyocytes can ensue with elevated SIRT1 levels of 12.5-fold[177]. FoxO proteins, such as FoxO1, also can control SIRT1 transcription and increase SIRT1 expression[178]. Under some circumstances, SIRT1 and FoxO proteins may function synergistically to promote cell survival. Loss of the forkhead transcription factors FoxO1 and FoxO3 in combination with decreased SIRT1 activity during oxidative stress leads to a reduction in autophagy with chondrocyte cell death, demonstrating that SIRT1 with FoxO proteins may be required for cellular protection[179]. SIRT1 also has been shown to increase lifespan in higher organisms and offer protection against oxidative stress[180]. EPO relies upon SIRT1 activity to prevent cell injury during oxidative stress and elevated glucose[181]. EPO can raise cellular activity of SIRT1 and promote the subcellular trafficking of SIRT1 to the nucleus to protect endothelial cells during oxidative stress[80]. EPO is able to maintain adipose cell energy homeostasis and protect against metabolic disorders through SIRT1[101]. Pathways that involve Wnt signaling with the CCN family member Wnt1 inducible signaling pathway protein 1 (WISP1)[139] also require up-regulation of SIRT1 activity to block apoptotic pathways controlled by FoxO proteins[182] (Figure 1). WISP1 can increase neuronal survival by limiting FoxO3a activity and FoxO3a deacetylation, blocking caspase 1 and 3 activation, and promoting SIRT1 activity and trafficking to the cell nucleus[146].

NOVEL AVENUES FOR EPO AND METABOLIC DISEASE

Growth factors such as EPO offer potentially new treatment approaches for numerous disorders, but given the signal transduction pathways that are regulated by EPO, this agent provides exciting prospects for the treatment of diabetes mellitus (DM)[16,45]. DM affects at least 350 million individuals worldwide[182] and is increasing in incidence[183]. Of potentially greater concern are the numbers of undiagnosed individuals that just in the United States alone may exceed 8 million individuals who are believed to suffer from metabolic disorders[32,184,185]. DM can affect the entire body and involve the immune system[63,77,181,186-190], liver[55,191-196], musculoskeletal function[197-201], kidney[202-206], and cardiovascular system[163,188,207-213] to result in endothelial cell dysfunction[15,16,99,100,168,214,215] and atherosclerosis[45,67,199,216]. These disorders can easily affect other regions of the body such as the nervous system to lead to cognitive loss[14,217-219], visual deterioration[32,119,220,221], peripheral nerve disease[55], and ischemic disease of the brain[23,49,67,222-224].

EPO as well as its downstream pathways have been shown to have a high potential to treat multiple complications of DM[32] (Figure 2). In earlier work that examined diabetics and non-diabetics with severe congestive heart failure, EPO increased left ventricular ejection fraction, reduced fatigue, and lessened duration of hospital stay[225]. In patients with Type 1 DM and cognitive impairment related to hypoglycemia, administration of EPO leads to improvement in complex reaction time task assessing associated with attention and working memory[226]. EPO also could provide a small improvement to treat fatigue in patients with Type 2 DM and chronic kidney disease[227].

Figure 2 Targeting erythropoietin involves a balance that fosters clinical improvement over clinical disability. EPO can play a significant role in reducing disability and fostering clinical benefit during diabetes mellitus. Through its signal transduction pathways, EPO may improve organ and tissue function, reduce fatigue, improve vascular perfusion, maintain glucose homeostasis, assist with wound and tissue repair, and promote cellular proliferation, differentiation, and survival. However, the detrimental effects of EPO that can include tumor cell growth, hypertension, increased blood viscosity, and unchecked angiogenesis must be considered and eliminated for successful therapeutic treatments against diabetes mellitus. EPO: Erythropoietin.

In experimental models of DM, EPO can reduce blood glucose levels in animal models of DM and obesity[228], protect against the detrimental effects of obesity in animal models[16], treat diabetic peripheral neuropathy[229], and block apoptosis in Schwann cells mediated by AGEs[97]. EPO has been shown to limit high glucose-induced oxidative stress in renal tubular cells[230], control cellular mitochondrial function[76,80,103,109,118], and maintain energy metabolism[15]. Through anti-inflammatory mechanisms and the blockade of apoptosis, EPO can protect pancreatic islet cells in models of type 1 DM and Type 2 DM[98]. Intravitreal administration of EPO in rodent models of DM can normalize gene expression that can lead to apoptotic and inflammatory cell death[231]. EPO is cardioprotective in DM models with the inhibition of glycogen synthase kinase -3β (GSK-3β)[232] that can limit Wnt signaling pathways[233]. Through increased angiogenesis and decreased apoptotic cell death, EPO can improve wound healing and wound closure in diabetic mice[77,234]. In vascular disease, EPO has been reported to protect the neuroglialvascular unit in a model of retinal neurodegeneration and secondary vasoregression[119]. EPO can directly protect against endothelial cell apoptosis during elevated glucose through activation of Wnt1[100] and the inhibition of GSK-3β and FoxO3a[99]. Improvement in vascular perfusion by EPO[123] also may afford indirect protection to assist with cognitive repair[235] and decrease peripheral nerve injury during DM[102].

Not all studies demonstrate a beneficial effect with EPO during DM, suggesting that focus upon the downstream signaling pathways of EPO with mTOR, Wnt signaling, FoxO proteins, and SIRT1 may yield greater utility for some clinical populations with complications of DM. In patients with DM and renal disease, EPO administration results in a two-fold increase in stroke that is not attributed to any baseline characteristic or to blood pressure, hemoglobin, platelet count, or treatment dose of EPO[236]. In mice that overexpress EPO, blood viscosity has been reported to be increased with a reduction in cerebral blood flow[237]. As a result, EPO may increase the risk for stroke through increased blood viscosity. Although systemic administration of EPO may block retinopathy in animal models[94], elevated EPO concentrations in patients with DM also may lead to proliferative diabetic retinopathy[238] that could be associated with excessive vascular growth. EPO can increase vascular responsiveness[239] and may lead to hypertension[26,57,240]. Sustained erythrocytosis with agents such as EPO may result in the activation of inflammatory pathways and blood-brain barrier dysfunction[155]. As a proliferative agent, EPO also can lead to new tumor growth as well as foster the progression of existing tumors[156-158,241].

The potential adverse effects of EPO may be avoided by targeting more specific pathways controlled by EPO such as mTOR and AMP activated protein kinase (AMPK)[40,208] (Figure 2). AMPK oversees the activity of the hamartin (tuberous sclerosis 1)/tuberin (tuberous sclerosis 2) (TSC1/TSC2) complex that is an inhibitor of mTORC1[135]. Metformin, an agent that controls hyperglycemia in DM, can reduce cardiomyopathy in experimental models of DM through AMPK activation[242]. EPO as well may dependent upon AMPK to promote antioxidant gene expression[243]. Furthermore, other EPO signaling pathways play a role in controlling AMPK. AMPK can increase nicotinamide phosphoribosyltransferase levels during glucose limitation resulting in elevated nicotinamide adenine dinucleotide[244] and lower levels of the SIRT1 inhibitor nicotinamide[245]. SIRT1 and AMPK activation promotes autophagy that offers endothelial cell protection during exposure to oxidized low density lipoproteins that can lead to atherosclerosis[246]. WISP1, a component of Wnt signaling, also controls the post-translational phosphorylation of AMPK that is involved in glucose homeostasis[124,247-249]. WISP1 regulates AMPK activation by decreasing phosphorylation of TSC2 at serine¹³⁸⁷, a target of AMPK, and increasing phosphorylation of TSC2 at threonine¹⁴⁶², a target of Akt[142]. The ability of WISP1 to modulate AMPK activity is vital for the regulation of cellular metabolism during DM[249]. AMPK activity is able to reduce insulin resistance and lessen oxidative stress through activation of autophagy[200]. AMPK can prevent myocardial ischemia in experimental models of DM[250], assist with proper metabolic function of cells[251], and limit adipocyte differentiation, lipid accumulation, and obesity[252]. Yet, similar to SIRT1, the degree of AMPK activity is a significant consideration in DM. AMPK activation can lead to apoptosis in pancreatic islet cells in some experimental models of Type 2 DM[253].

CONCLUSIONS AND FUTURE PERSPECTIVES

In the global population, DM is a significant cause of disability and death. Treatment options to limit the onset and progression of this disease are insufficient and warrant the development of novel treatments. EPO, as a cytoprotective agent that controls a broad array of signal transduction pathways offers exceptional promise for the treatment of DM and pathways of oxidative stress. EPO has been shown in diabetic patients to improve cardiac function, reduce fatigue, and improve cognition. In experimental models of DM, EPO can reduce blood glucose levels, limit peripheral neuropathy, maintain mitochondrial function and energy metabolism, and block programmed cell death in many cell types such as Schwann cells, endothelial cells, neurons, pancreatic islet cells, and cardiomyocytes.

However, several challenges exist to move EPO forward as an effective treatment for DM. EPO has been reported to increase the risk of stroke in patients with DM and renal disease and has been demonstrated to increase blood viscosity in animal studies. EPO may be contraindicated in hypertensive patients and may contribute to elevated mean arterial blood pressure. Elevated concentrations of EPO have been linked to proliferative diabetic retinopathy that may be associated with excessive microvascular angiogenesis. Finally, EPO, as a growth factor and proliferative agent, may lead to new tumor growth and also promote the growth of existing tumors, especially in the treatment of patients with cancer and anemia.

Further investigations that assess the protective capacity of EPO and limit any potential detrimental clinical outcomes are warranted. New work has been directed to improving the molecular stability, solubility, and immunogenicity of EPO for improved therapeutic strategies to treat the complications of DM. Glycoengineering, a method that introduces N-linked glycosylation consensus sequences into proteins to increase serum half-life and biological activity, has been examined for EPO[254]. Darbepoetin alpha is one such example of a hyperglycosylated EPO derivative. Darbepoetin alpha has an increased serum half-life when compared to recombinant EPO[255] and is considered more potent than recombinant EPO[256]. EPO mimetic proteins are other avenues being pursued that can be used to activate the EPOR, potentially increase treatment half-life and maintain potency when compared to EPO, and lessen immunogenicity[257,258]. For example, CNTO 530 has been shown to increase reticulocytes, red blood cells and total hemoglobin in β- thalassemic mice[259].

A promising investigative course also could target the downstream signaling pathways of EPO that include Akt, mTOR, Wnt signaling, FoxO proteins, SIRT1, and AMPK. EPO employs Akt and mTOR for stem cell maintenance and differentiation, resistance against oxidative stress, and the regulation of autophagy. In experimental models of DM, EPO relies upon Wnt signaling, β-catenin, and the inhibition of GSK-3β to block apoptotic cell death. EPO also governs FoxO proteins and SIRT1 to protect against DM apoptotic vascular injury, maintain adipose cell energy homeostasis, and modulate autophagic flux to improve cardiac function during metabolic disturbances. Pathways that involve EPO and AMPK also offer interesting targets to maximize clinical efficacy and minimize unwanted side effects. AMPK reduces insulin resistance and lessens oxidative stress through activation of autophagy, prevents myocardial ischemia in models of DM, and limits adipocyte lipid accumulation and obesity. WISP1 controls AMPK activity for the regulation of cellular metabolism during DM. In addition, SIRT1 and AMPK in conjunction with SIRT1 can increase autophagy activity to provide endothelial cell protection during exposure to oxidized low-density lipoproteins. However, it should be noted that consideration of these pathways may still require use of EPO or an EPO analogue since therapeutic success may be dependent on modulation of more than one of these down-stream pathways of EPO. In addition, one needs to emphasize that each of these pathways also can lead to undesirable biological outcomes under some circumstances such as tumorigenesis, pancreatic islet cell death, and cardiac dysfunction. Carefully targeting future investigations for EPO and its relevant signal transduction pathways for specific clinical disturbances of DM should offer the greatest promise for novel therapeutic strategies.