{"paper_id":"9c9a78dd-f0bc-4683-a346-1b37cffd1a6b","body_text":"INTRODUCTION\nErythropoietin (EPO) is a well-known\nstimulator of erythrocyte production and\nwidely used in the treatment of anemia\ncaused by kidney disease, cancer, or\nchronic inflammation (reviewed by Jelk-\nmann [1]). Over the past decade, it has\nbecome evident that EPO also possesses\nmany other biological activities that can\ngenerally be summarized as counteract-\ning the actions of proinflammatory cy-\ntokines and their deleterious effects in\ntissue injury (reviewed by Brines and Ce-\nrami [2]). In these nonhematopoietic ac-\ntivities, EPO is locally produced in the\nimmediate vicinity of the injury.\nCross-talk between the circulating\nhematopoietic and the local tissue protec-\ntive pools of EPO is avoided by the pres-\nence of EPO receptor isoforms that differ\ngreatly in their affinity for EPO. Specifi-\ncally, the hematopoietic receptor is a ho-\nmodimer composed of identical EPO re-\nceptor (EPOR) subunits that maintains\nerythropoiesis in response to circulating\nlevels of EPO in the 1–10 pMolar range\n(3). In contrast, the tissue protective re-\nceptor is postulated to be a heteromer\nformed by EPOR in assembly with\nCD131, the βcommon receptor which\nalso is used by GM-CSF, IL-3, IL-5, and\nother type I cytokines (4). The tissue pro-\ntective receptor exhibits a lower affinity\nfor EPO (2–20 nM) and therefore does not\nrespond to EPO at concentrations present\nwithin the circulation, but rather only to\nhigh levels of locally produced EPO. An\nadditional important functional separa-\ntion between the hematopoietic and tis-\nsue protective systems arises from the\ndifference in basal expression of the cor-\nresponding receptors. In hematopoiesis,\nthe homodimer is expressed continuously\nby a population of red cell precursors\nthat require constant circulating EPO to\nenable survival. In contrast, the tissue\nprotective receptor typically is not ex-\npressed until after the occurrence of in-\njury or significant metabolic stress and\nonly requires a brief exposure to EPO to\ntrigger sustained biological activity (2).\nWhen employed at the high doses re-\nquired for adequate tissue protection,\nEPO unfortunately possesses use-limiting\nside effects, particularly in converting the\nvasculature into a prothrombotic state,\nwhich leads to an increased risk of life-\nthreatening thromboses, as has been ob-\nserved particularly for injured patients\n(5) or those with cancer (6). The identifi-\ncation of distinct hematopoietic and tis-\nMOL MED 15(7-8)235-241, JULY-AUGUST 2009 | ERBAYRAKTAR ET AL. | 235\nNonerythropoietic Tissue Protective Compounds Are Highly\nEffective Facilitators of Wound Healing\nZübeyde Erbayraktar,1 Serhat Erbayraktar,1 Osman Yilmaz,1 Anthony Cerami,2 Thomas Coleman,3 and\nMichael Brines2\n1Dokuz Eylül University, Izmir, Turkey; 2Warren Pharmaceuticals, Ossining, New York, United States of America; 3The Feinstein\nInstitute for Medical Research, Manhasset, New York, United States of America\nErythropoietin (EPO) is a type I cytokine that utilizes different receptor isoforms either to maintain hematopoiesis or protect\nagainst injuries that arise from widely diverse etiologies. EPO also facilitates healing by reducing inflammation and mobilizing en-\ndothelial progenitor cells to participate in restorative neoangiogenesis, but it is unclear which EPO receptor isoform is responsible\nfor healing and whether this receptor use varies according to the type of wound. In the present studies carried out in the rat, w e\nhave utilized receptor-selective derivatives of EPO to determine which receptor type operates in (i) a nonischemic wound (skin\npunch biopsy), (ii) a permanently ischemic wound (raised musculocutaneous flap), (iii) an intermittent ischemic reperfusion\nwound (pressure or decubitus ulcer), or (iv) wounds complicated by infection (cecal ligation and perforation). Using these mod-\nels, we demonstrate that nonerythropoietic tissue protective compounds administered immediately following injury limit wound\nsize and accelerate eschar closure independent of wound type. Moreover, in a model of peritonitis-induced adhesions, daily ad-\nministration of the nonerythropoietic derivative carbamyl-EPO (10 μg/kg-bw) was associated with significantly lower serum TNFα\nconcentration, illness scores, increased survival, as well as decreased adhesion formation. These results confirm that wound heal-\ning is mediated by the tissue protective receptor isoform and argue that nonerythropoietic tissue protective molecules constitute\npromising new therapeutics for treatment of a wide variety of surgical wounds.\n© 2009 The Feinstein Institute for Medical Research, www.feinsteininstitute.org\nOnline address: http://www.molmed.org\ndoi: 10.2119/molmed.2009.00051\nAddress correspondence and reprint requests to Michael Brines, Warren Pharmaceuticals,\n712 Kitchawan Road, Ossining, NY 10562. Phone: 914-762-7586; Fax: 914-762-4000; E-mail:\nmbrines@warrenpharma.com.\nSubmitted April 9, 2009; Accepted for publication April 16, 2009; Epub (www.molmed.org)\nahead of print April 16, 2009.\n\nsue protective EPO receptor isoforms ini-\ntiated the quest for molecules that exhibit\na high specificity for the tissue protective\nreceptor subtype to avoid the adverse ef-\nfects associated with EPO therapy. As a\nresult, a number of nonerythropoietic tis-\nsue protective compounds (TPCs) have\nbeen developed that interact exclusively\nwith the tissue protective receptor, in-\ncluding carbamyl-EPO (7), as well as\npeptides that mimic the three dimen-\nsional structure of EPO (8), for example,\npyroglutamate-helix B surface peptide\n(ARA 290, Araim Pharmaceuticals,\nOssining, NY, USA). The utilization of\nthese reagents has led to an understand-\ning of the major differences in biological\neffects of the hematopoietic and tissue\nprotective receptors. For example, while\nthe hematopoietic EPO receptor activates\nthe endothelium into a prothrombotic\nstate, increases platelet number and reac-\ntivity, and increases systemic blood pres-\nsure via constriction of vascular smooth\nmuscle, the tissue protective EPO recep-\ntor activates none of these activities (9).\nAmong the pleiotropic effects of EPO\nreported, a number of investigators have\nreported significant activity of EPO in\npromoting the healing of ischemic raised\ncutaneous skin flaps (10–14), ischemia-\nreperfusion in random musculocutaneous\nflaps (15), incisional wounds (10,13), and\nin colonic anastomoses (16,17). Addition-\nally, skin thermal burns also have been\nreported to respond in a beneficial way to\nEPO (18). In other models, EPO has been\nshown to increase the formation and\nquality of granulation tissue (19). In con-\ntrast to the direct, nonhematopoietic ac-\ntions, it has been postulated that EPO can\nbenefit healing of chronic wounds by\naugmenting the delivery of oxygen via\nincreases in the hematocrit (20).\nIt is currently unclear which receptor\nisoform transduces EPO’s activity in the\nsetting of wounds and, therefore,\nwhether TPCs could be reasonable candi-\ndates to avoid EPO’s adverse effects in\nthis therapeutic area. In the present work\nwe examined the action of TPCs on the\ndevelopment and resolution of injury in a\nvariety of wound etiologies: i) a predomi-\nnantly ischemic wound; ii) an ischemia-\nreperfusion wound model; iii) an inci-\nsional wound in which ischemia plays a\nlimited role; and iv) infectious peritonitis\nthat produces adhesions. Because we\nwere interested in modeling potential\nclinical scenarios that often cannot be an-\nticipated, dosing of TPCs was initiated\nonly following wound injury. The results\nshow that TPCs effectively limit injury,\naccelerate healing, and reduce adhesion\nformation, confirming a role for the tissue\nprotective heteroreceptor in this activity.\nMATERIALS AND METHODS\nAll experiments were performed under\nprotocols approved by the local animal\nuse and care committee and complied\nwith the standards in the Guide for the Care\nand Use of Laboratory Animals, prepared by\nthe Institute of Laboratory Animal Re-\nsources and published by the National\nAcademy Press (21). Male Sprague Daw-\nley rats of 275–300 gm body weight were\nused for these experiments. Each animal\nwas acclimated for at least 1 wk prior to\nexperimentation, maintained under a re-\nverse light–dark cycle, and were able to\naccess food and water ad libitum except as\nstated in each individual protocol. Follow-\ning surgical wounding procedures, each\nanimal was individually housed for the\nduration of the experiment.\nCompounds\nEPO was obtained from Dragon Phar-\nmaceuticals (Vancouver, BC, Canada).\nCarbamyl-EPO was prepared as pub-\nlished previously (7). ARA 290 is an 11\namino acid peptide of molecular weight\n1,258 synthesized by standard F-moc\nsolid phase peptide synthesis and puri-\nfied by HPLC and ion-exchange chro-\nmatography.\nCutaneous Punch Biopsy\n(Nonischemic Wounding)\nAnimals were fasted from the evening\nbefore the procedure. Under isoflurane\nanesthesia, a region of skin 5 × 5 cm was\nshaved on the dorsum in the subscapular\nregion, washed with povidone-iodine so-\nlution, and followed by a sterile water\nrinse. Four full-thickness skin punches\nwere placed at the corners of a 2-cm\nsquare using a 3.5-mm-diameter (area =\n9.6 mm\n2) disposable sterile biopsy\npunch. A drop of 1% lidocaine solution\nwas placed into the wounds followed by\napplication of 1% lidocaine-saturated\ngelfoam attached by adhesive tape. Ani-\nmals then received one of three treat-\nments in a blinded fashion. Group 1 (n =\n9) received a daily subcutaneous saline\ninjection in a central location, equidistant\nfrom each punch wound. In a similar\nfashion, Group 2 (n = 6) received EPO at\n500 IU/kg-bw (~130 picomoles [pmol]/\nkg-bw), and Group 3 (n = 9) received\n1 nanomole [nmol]/kg-bw of the tissue\nprotective peptide ARA 290. Wound as-\nsessment was accomplished by digital\nplanimetry of eschar area using dimen-\nsionally calibrated serial digital photo-\ngraphs on d 0, 1, 3, 7, 10, and 14. Wound\nsize was expressed as the ratio between\nthe wound area and a size standard. (22).\nThis experiment was performed twice\nwith similar results.\nDorsal Skin Flap (Permanent\nIschemia)\nUnder isoflurane anesthesia, the dorsal\nskin was shaved, washed with povidone-\niodine solution, and rinsed with sterile\nwater. A caudally based 3- × 9-cm–long\ncutaneous pedicle flap was constructed\nfollowing the general method of McFar-\nlane et al. (23). In this procedure, a deep\nincision is made down to the areolar tis-\nsue plane along three sides through the\ndermis and adherent panniculus\ncarnosus. Following mobilization and el-\nevation to make certain that nutrient\nblood supply was severed on three sides,\nthe flap was immediately replaced into\nposition and the three edges sutured to-\ngether at 5-mm intervals using 3–0\nEthicon Products (Sommerville, NJ).\nAnimals were separated into three\ngroups, and received intravenous injec-\ntions in a blinded fashion twice weekly.\nGroup 1 (n = 9) received saline, Group 2\n(n = 6) received carbamylated EPO\n(CEPO) (3 μg/kg-bw), and Group 3 (n =\n6) CEPO (0.3 μg/kg-bw). Dimensionally-\n236 | ERBAYRAKTAR ET AL. | MOL MED 15(7-8)235-241, JULY-AUGUST 2009\nTPCs IN WOUND HEALING\n\ncalibrated photographs of the flaps were\nobtained weekly for 5 wks following sur-\ngery and eschar area determined by digi-\ntal planimetry.\nDecubitus Ulcer (Recurrent\nIschemia–Reperfusion)\nIn this protocol, pressure-induced is-\nchemia was produced to mimic that en-\ncountered by a debilitated hospitalized\npatient. Over a period of two successive d,\nwe performed five recurring 2-h ischemic\nepisodes, each separated by 30 min of\nreperfusion (12 h total), were performed,\nfollowed by a period of 12 h of ischemia.\nTo reliably elicit ischemia–reperfusion of\nthe skin in freely moving animals, a com-\nbination of an implanted steel plate with\nan externally applied permanent magnet\nwas employed (24). Briefly, under isoflu-\nrane anesthesia the left flank was shaved\nand disinfected. A steel plate (3.3 ×6.3 cm)\nwas then placed subcutaneously and\nsewn into the underlying muscle. Follow-\ning a 24-h recovery period, blood flow\ninto the overlying skin was arrested, ac-\ncording to the protocol, by application of\na neodynium magnet (47 × 22 × 10 mm\nthick possessing a surface area of 10.3 cm\n2)\nto the central portion of the steel plate.\nEach magnet was oriented to parallel the\nlong axis of the steel plates, centered to\nassure a similar compression for each ani-\nmal, and its position marked on the sur-\nface of the skin to allow accurate replace-\nment. Treatment groups (n = 8) were\nadministered saline, ARA 290 (1 nmol/\nkg-bw subcutaneously) on d 1 and 2,\nARA 290 (1 nmol/kg-bw subcutaneously),\nEPO (300 IU/kg-bw; ~80 pmol/kg-bw)\ndaily or CEPO (50 μg/kg-bw; ~1 nmol/\nkg-bw) once a day for the first 2 d and\nthe animals followed for a total of 14 d.\nDigital photographs were obtained and\nnecrotic area of the wound determined in\na blinded fashion using digital planime-\ntry. For the first day following injury, a\nclear determination of skin viability\ncould not be made reliably using photo-\ngraphs, so tissue viability was deter-\nmined by the presence of skin blanching\nand confirmed by the presence of bleed-\ning to a slight needle prick.\nCecal Ligation and Perforation\n(Infected Wound)\nTo produce a model of infectious peri-\ntonitis, an abdominal incision was made\nunder isoflurane anesthesia, followed by\nligation of the cecum and then two punc-\nture wounds made using a sterile 18 gauge\nneedle. Great care was taken to minimize\nthe handling of the intestines and to pro-\nduce a uniform abdominal wound. No\nfeces were expressed nor were antibiotics\nadministered. Four treatment groups were\nevaluated: sham (n = 8, saline, abdominal\nincision with minimal manipulation of in-\ntestine but without perforation); control \n(n = 14, cecal ligation and perforation\n[CLP] followed by saline); EPO (n = 8, \nCLP followed by EPO 10 μg/kg-bw; \n~250 pmol/kg-bw); and CEPO (n = 14,\nCLP followed by CEPO 10μg/kg-bw).\nCompounds or saline were administered\nintravenously by tail vein once immedi-\nately following manipulation of the intes-\ntine, and then continued daily by in-\ntraperitoneal doses. Animals were scored\ndaily for signs of illness by noting the\npresence (+1) or absence (0) of the follow-\ning twelve criteria: piloerection, immobil-\nity, inability to hold onto an inclined cork-\nboard, inability to use claws, assuming\nhunchbacked posture, abnormal gait, lack\nof exploration of surroundings, unable to\ngrasp a string within 30 sec, diminished\nreflexes, lack of appetite, weight loss, or\nmoribund. Animals were either eutha-\nnized when and if appearing moribund or\nfollowed for 8 d after the procedure and\nthen killed. Immediately following death,\nthe abdomen was opened and adhesions\nscored using the scale of Bothin et al. (25).\nStatistics\nWound sizes in different treatment\ngroups were analyzed using nonpara-\nmetric methods (Mann–Whitney U or\nWilcoxon test), by log-rank (Mantel–Cox\ntest), or by repeated measures analysis as\nappropriate using JMP software (SAS\nInc, Cary, South Carolina, USA).\nRESULTS\nIn the skin biopsy wound model, we\ndetected no differences among the loca-\ntion of the punch wounds throughout the\nexperiment and little change in eschar\nsize was observed over the first 3 d fol-\nlowing injury, irrespective of treatment\ntype. Thereafter, wound size decreased\nrapidly, with complete closure occurring\nby d 14 in the saline treatment group.\nMaximum difference in wound size be-\ntween the treatment groups was largest\non d 7 (Figure 1A). At this time, wounds\non the animals that received the tissue\nRESEARCH ARTICLE\nMOL MED 15(7-8)235-241, JULY-AUGUST 2009 | ERBAYRAKTAR ET AL. | 237\nFigure 1.Tissue protective compounds improve wound healing in a dermal punch biopsy\n(nonischemic) model. (A) Punch wound size from an animal representative of each treat-\nment group at d 7 following injury. The ARA 290 group (1 nmol/kg-bw) was characterized by a\nmore complete closure of the wound compared with animals that received either PBS or\nEPO (500 IU/kg-bw). (Standard represents the initial lesion size.) (B) At 7 d following injury, the\nARA 290 group was significantly more advanced in healing of the clean, uncomplicated der-\nmal punch biopsy wound compared with saline or EPO (\nP < 0.001).\n\nprotective peptide ARA 290 were 84%\nhealed and differed significantly from the\nsaline group, which was 63% healed and\nthe EPO group, which was 67% healed\n(Figure 1B). Thus, daily subcutaneous in-\njections of ARA 290 (1 nmol/kg-bw;\nequivalent to ~ 3,000 IU EPO on a molar\nbasis), a compound with a very short\nserum half-life of ~ 2 min, effectively\nshortened healing time in a model of\nwounding that is not accompanied by\nsignificant hypoxia.\nIn the random musculocutaneous flap\nmodel characterized by permanent ische-\nmia due to the interruption of nutrient\narteries, CEPO treatment at the higher\ndose (30 μg/kg intravenously; Figure 2A)\nbut not the lower dose (3 μg/kg; data\nnot shown) was associated with an in-\ncreased rate of healing. In addition to the\ndegree of eschar resolution, the pattern\nof closure also differed between treat-\nment groups. Animals that received\nsaline exhibited a lateral contraction of\nthe wound, leading to a long, thin scar\n(Figure 2B). In contrast, the CEPO group\nhealed without constriction, producing a\nmore rectangular scar.\nIn the decubitus ulcer model, animals\nreceived the first dose of a TPC 36 h after\nthe onset of recurrent ischemic episodes.\nThe results (Figure 3) show that animals\ntreated with tissue protective com-\npounds exhibited a much smaller wound\nsize after 3 d than did animals that re-\nceived saline alone (3–3.9 cm\n2 versus\n6.4 cm2). After attaining a peak wound\nsize, the saline group remained about the\nsame for the duration of the experiment.\nIn contrast, treatment with ARA 290 ad-\nministered as two daily doses (1 nmol/\nkg-bw) resulted in a smaller peak wound\nsize (49% of maximum), which also re-\nmained the same size for the duration of\nthe experiment. Two doses of the tissue\nprotective compound CEPO, which has a\nplasma half-life of ~5–6 h, was similar to\ndaily dosing of EPO: both the EPO and\nCEPO groups that received daily dosing\nshowed a larger reduction in peak\nwound area (38% and 30% respectively),\nbut thereafter declined. In the terminal\nphase of the experiment, animals that\n238 | ERBAYRAKTAR ET AL. | MOL MED 15(7-8)235-241, JULY-AUGUST 2009\nTPCs IN WOUND HEALING\nFigure 2. Tissue protective compounds improve wound healing in a raised skin flap (per-\nmanent ischemia) model. Both treated and untreated rats with a permanently ischemic,\ncaudally based cutaneous flap exhibited a large necrotic wound area that took more\nthan 5 wks to fully heal. (A) At 35 d following injury, wounds in the saline group were signifi-\ncantly larger (\nP < 0.05) than the group that received CEPO (30 μg/kg twice weekly in-\ntraperitoneally). (B) Representative examples of wound size from both treatment groups.\nNotably, wounds in the saline group healed by transverse constriction which was not ap-\nparent in CEPO group.\nFigure 3. Tissue protective compounds improve wound healing in a decubitus ulcer\n(ischemia-reperfusion) model. (A) Recurrent ischemia-reperfusion (I/R) episodes caused by\ncompression of the skin lead to a large necrotic wound that reaches a peak in size by d 5.\nAnimals were subjected to I/R episodes for the first 36 h and then received a first treatment\nvia tail vein. Wound size first measured on d 3 was smallest in animals that received nonery-\nthropoietic tissue protective molecules. The saline group was characterized by a large\nwound that did not shrink appreciably over the observational period. Although two doses\nof the tissue protective peptide ARA 290 (with an ~2-min plasma half-life) reduced the ini-\ntial lesion size at d 3, but no further reduction in size was noted, as compared with the\nother treatment groups that received daily doses (EPO and ARA 290) or two treatments\nwith a nonerythropoietic tissue protective compound with a longer plasma half-life (CEPO;\nhalf-life ~5–6 h). (B) Representative examples of difference at d 8 between wound of a\nsaline-treated animal compared with one having received ARA 290.\n\n\nhad received intensive dosing of ARA\n290 exhibited the most complete and\nrapid resolution of the wound over the\n13 d of the study (to ~16% of the maxi-\nmum wound size; Figure 3B). Thus, like\nthe skin punch biopsy model, daily treat-\nment with ARA 290 resulted in a better\noutcome than that for EPO in this model\nsystem.\nRats undergoing cecal ligation and\nperforation displayed signs of illness\nwithin 24 h following the procedure (Fig-\nure 4A). The group that received CEPO\nexhibited a milder course of symptoms\nthat reached a peak by d 3, and then di-\nminished slightly over the observation\ntime of 8 d. In contrast, the saline group\nreached a maximum illness score by d 4,\nwhich was twice as severe as the CEPO\ngroup. Additionally, infected animals\nfrom both groups lost about 10% of basal\nbody weight over the observation period\n(data not shown). Although the saline\ngroup reached a nadir of weight loss ear-\nlier than the CEPO group, from d 3, both\ngroups exhibited the same degree of\nweight loss, even though the illness\nscores differed greatly. Serial serum\nTNFαlevels obtained following CLP rose\nrapidly in all treatment groups except\nsham, peaking by 3 h, and diminishing\nslowly over the ensuing 24 h. The mean\nvalue of the CEPO treatment group was\ncomparable to the sham group, each\nbeing only about half that of the saline\ntreatment group (Figure 4B).\nAs shown in Figure 4C, animals that re-\nceived CEPO exhibited an approximate\nthree-fold improved survival rate com-\npared with either the EPO or saline treat-\nment groups. Animals that survived were\neuthanized at d 8 following the proce-\ndure. Interestingly, the severity of adhe-\nsions was significantly much worse in the\nsaline group compared with the CEPO\ngroup (means of ~ 10 versus 7; Figure\n4D). Although the mean adhesion score of\nthe EPO group was less than that of the\nsaline group, the effect did not reach the\nlevel of significance. Figure 4E is a repre-\nsentative photograph showing adhesions\npresent in a surviving saline animal com-\npared with one that received CEPO.\nDISCUSSION\nUsing four wound models, each char-\nacterized by a different pathophysiology,\nwe have shown that nonerythropoietic\ntissue protective molecules limit maxi-\nmum wound size as well as accelerate\nclosure. Additionally, in a model of infec-\ntious peritonitis, carbamyl-EPO, a non-\nerythropoietic tissue protective molecule,\nsuppresses production of the proinflam-\nRESEARCH ARTICLE\nMOL MED 15(7-8)235-241, JULY-AUGUST 2009 | ERBAYRAKTAR ET AL. | 239\nFigure 4. Tissue protective compounds improve wound healing in a cecal ligation and\nperforation infectious peritonitis model. (A) Following wounding, signs of illness (see Mate-\nrials and Methods) intensified, reaching a peak by 2–3 d. Notably, illness in the CEPO-\ntreated animals was delayed and less severe than in the saline treatment group. (B)\nSerum TNFαlevels from tail vein samples at 3 h following the procedure. The CEPO group\nexhibited a lower peak level (\nP < 0.05) and was similar to the sham group, consistent with\nthe less severe clinical scores observed. Horizontal lines are the group means. (C) The\nmortality in the saline group was significantly more than for CEPO (\nP < 0.01), whereas\nEPO treatment was not associated with a significant increase in survival. (D) Scoring the\nseverity of adhesions showed that CEPO treatment was associated with a marked re-\nduction (\nP < 0.01 compared with saline) as well as a nonsignificant trend in reduction as-\nsociated with EPO treatment. (E) Representative photos demonstrating the difference\nbetween a saline animal (left) and CEPO (right). Note the white, fibrinous membrane al-\nmost completely enveloping the opened abdominal surface. This is consistent with highly\ninflammatory processes, which are not present in CEPO-treated animals.\n\n\nmatory cytokine TNFα, decreases mor-\ntality, and is highly effective in reducing\nscarring. Because TPCs cannot interact\nwith the hematopoietic homodimeric re-\nceptor (7), the beneficial effects of TPCs\nare mediated through the heteromeric\ntissue protective receptor. An additional\nconclusion is that effects via the erythro-\npoietic receptor (for example, endothelial\ncell activation) could account for the in-\nferior activity of EPO on wound healing\nwhen compared with TPCs in some of\nthe models tested.\nTissue damage is a certainty of life.\nNormally, injury is contained and then a\nrepair and regeneration phase returns tis-\nsues to a normal state. The biology of\nwounding involves an acute injury phase\nthat activates hemostatic, inflammatory,\nand proliferative cellular responses that\nare triggered by molecular signals (26).\nThese “alarmins” subsequently initiate\nthe recruitment of inflammatory cells\nthat both amplify and modulate tissue\ndamage in a potentially positive feed-\nback manner (27,28). One prominent\npathological consequence of this process\nis tissue apoptosis and necrosis. Prior\nwork has shown that a considerable por-\ntion of necrotic injury arises from an in-\ntensification of ischemia due to nonfunc-\ntional capillaries caused by endothelial\ncell apoptosis, leukocyte plugging, en-\ndothelial cell swelling, and vascular leak-\nage, leading to tissue compression by in-\nterstitial edema (29,30). If wound volume\nis successfully contained, endothelial\nprogenitor cells migrate into the injured\nregion supporting the growth of new\nblood vessels within granulation tissue.\nTissues respond to a wide variety of\ninjuries by triggering a stereotyped re-\nsponse involving a balance between de-\nstructive proinflammatory cytokines (for\nexample, TNFα) and antiinflammatory\nresponses (for example, locally-produced\nEPO) (2). However, these are mutually\nsuppressive such that within the vicinity\nof a wound, the proinflammatory milieu\nleads to suppressed EPO synthesis. A\nmajor role of nonerythropoietic tissue\nprotective compounds is to antagonize\nthe activities of proinflammatory cy-\ntokines by reducing their production and\nattenuating their biological effects (2).\nIn contrast to EPO, the expression of\nthe tissue protective receptor is upregu-\nlated within the wound, setting the stage\nfor a potential role of exogenously ad-\nministered TPCs to rescue cellular targets\notherwise programmed to undergo apo-\nptosis. A critical potential therapeutic tar-\nget is the capillary endothelium, in that\ncapillary dropout frequently occurs be-\ncause of apoptosis of the endothelial\ncells, amplifying injury. Administration\nof TPCs could rescue these capillaries,\ndiminishing wound volume. This phe-\nnomenon has been recently reported to\noccur in the rodent ischemic random\nmusculocutaneous flap model (14) in\nwhich EPO administration increased the\nproportion of functional capillaries\nwithin the wound. Another important ef-\nfect at the capillary level: EPO has been\nobserved to reduce capillary plugging\ndramatically by leukocytes primed for\nadherence by proinflammatory mole-\ncules (11,12,14,15). However, EPO itself\ncan strongly activate the endothelium\nproducing a prothrombotic state which\ncould neutralize a portion of its thera-\npeutic benefit. This disadvantage is\navoided by use of TPCs.\nIn addition to preserving existing cap-\nillaries, EPO has also been shown to ac-\ncelerate healing in the ischemic flap\nmodel by amplifying reparative angio-\ngenesis (10–12). It is known that in organ\ninjury, such as a rodent myocardial in-\nfarction model (31), EPO mobilizes en-\ndothelial progenitor cells (CD34\n+/flk-1)\nthat subsequently travel via the circula-\ntion and migrate into injury sites to drive\nneoangiogenesis, ultimately reducing tis-\nsue hypoxia. Further, tissue-specific stem\ncells are recruited by EPO to take up resi-\ndence within the region of injury and dif-\nferentiate into mature, functional cells\n(32). Finally, EPO also improves collagen\ndeposition and modulates the number of\nfibroblasts within a wound site (13).\nAs discussed previously, the affinity of\nthe tissue protector subtype is lower than\nthat of the hematopoietic system (re-\nviewed in Brines and Cerami [2]). This re-\nquires substantially higher doses of EPO\nfor treatment of tissue injury than for ery-\nthropoeisis and, as a result, invariably in-\nteracts with both receptor isoforms. This\nis exemplified in the skin flap model in\nwhich CEPO was effective at 3 μg/kg-bw\n(equivalent to ~ 300 IU/kg-bw of EPO),\nbut not at 0.3 μg/kg-bw which is equiva-\nlent to ~ 30 IU/kg-bw, a purely hemato-\npoietic dose of EPO. In addition to\nhematopoiesis, the homodimer also me-\ndiates a number of other activities. These\ninclude constriction of vascular smooth\nmuscle, activation of vascular endothelial\ncells into a proinflammatory state (ex-\npression of selectins, increased platelet\nproduction, and so on) which are gener-\nally undesirable when one contemplates\nthe use of EPO as a therapeutic agent for\ntissue protection. Nonerythropoetic mol-\necules completely avoid this issue (9).\nIt is especially notable that in the decu-\nbitus ulcer model, ARA 290 administered\ndaily produced an initially smaller lesion\nthan animals receiving EPO. Several\ngroups using the musculocutaneous skin\nflap model have reported that high dose\nEPO is associated with significantly infe-\nrior effects on healing than lower doses\nof EPO (14,33). In these cases, nonbenefi-\ncial activities ascribed to EPO, for exam-\nple, reduction of blood supply by pro-\nmoting thrombosis or from adverse\nrheological effects arising from increases\nin the hematocrit are explained by inter-\nactions with the homodimeric receptor.\nIn these experiments, treatment was\ninitiated only following the wounding\ninjury, mirroring many typical clinical\nsituations in which therapy in advance\nof wounding is not practical. The bio-\nlogical effects can be explained by a\ntriggering of the tissue protective recep-\ntor, as ARA 290, which is highly effec-\ntive, is a compound that has a half-life\nof only about 2 minutes when adminis-\ntered intravenously in the rat (8). Such\nlong lasting effects are explained by the\ntriggering of complex gene expression\nprograms that provide for sustained ac-\ntivities. Thus, in promoting recovery\nfrom wounding, nonerythropoietic tis-\nsue protective ligands do not need to be\n240 | ERBAYRAKTAR ET AL. | MOL MED 15(7-8)235-241, JULY-AUGUST 2009\nTPCs IN WOUND HEALING\n\npresent continuously to promote benefi-\ncial effects.\nThe potential clinical utility of TPCs in\nwound healing is clear for a reduction of\nhealing time in a variety of wound types,\nincluding those surgically-induced or\nthose caused by a debilitated patient\nstate, for example, decubitus ulcers. Po-\ntential additional benefits in modulation\nof scarring following peritoneal infec-\ntions, or other inflammatory processes\nlike endometriosis, bear further investi-\ngation. Finally, these compounds could\nhave other beneficial effects not exam-\nined in the current experiments, for ex-\nample, a neurotrophic activity (2) that\nlikely enhances the regrowth of severed\ncutaneous nerves.\nACKNOWLEDGMENTS\nWe thank Deborah Diaz, Daniel\nGomez, Annie Zhu, and Xiao-wen Xie\nfor their invaluable technical assistance.\nDISCLOSURE\nAraim Pharmaceuticals is a spin-off of\nWarren Pharmaceuticals. A Cerami and\nM Brines are employees of Warren Phar-\nmaceuticals, which is developing tissue\nprotective compounds for clinical use.\nREFERENCES\n1. Jelkmann W. (2007) Erythropoietin after a century\nof research: younger than ever. Eur. J. Haematol.\n78:183–205.\n2. Brines M, Cerami A. (2008) Erythropoietin-\nmediated tissue protection: reducing collateral\ndamage from the primary injury response. J. In-\ntern. Med. 264:405–32.\n3. Jelkmann W, Bohlius J, Hallek M, Sytkowski AJ.\n(2008) The erythropoietin receptor in normal and\ncancer tissues. Crit. Rev. Oncol. Hematol. 67:39–61.\n4. Brines M, et al. (2004) Erythropoietin mediates\ntissue protection through an erythropoietin and\ncommon beta-subunit heteroreceptor. Proc. Natl.\nAcad. Sci. U. S. A. 101:14907–12.\n5. Corwin HL, et al. (2007) Efficacy and safety of\nepoetin alfa in critically ill patients. N. Engl. J.\nMed. 357:965–76.\n6. Bennett CL, et al. (2008) Venous thromboembolism\nand mortality associated with recombinant ery-\nthropoietin and darbepoetin administration for\nthe treatment of cancer-associated anemia. JAMA\n299:914–24.\n7. Leist M, et al. (2004) Derivatives of erythropoietin\nthat are tissue protective but not erythropoietic.\nScience 305:239–42.\n8. Brines M, et al. (2008) Nonerythropoietic, tissue-\nprotective peptides derived from the tertiary\nstructure of erythropoietin. Proc. Natl. Acad. Sci.\nU. S. A. 105:10925–30.\n9. Coleman TR, et al. (2006) Cytoprotective doses of\nerythropoietin or carbamylated erythropoietin have\nmarkedly different procoagulant and vasoactive ac-\ntivities. Proc. Natl. Acad. Sci. U. S. A. 103:5965–70.\n10. Buemi M, \net al. (2004) Recombinant human ery-\nthropoietin stimulates angiogenesis and healing\nof ischemic skin wounds. Shock 22:169–73.\n11. Buemi M, et al. (2002) Recombinant human ery-\nthropoietin influences revascularization and\nhealing in a rat model of random ischaemic\nflaps. Acta. Derm. Venereol. 82:411–7.\n12. Rezaeian F, et al. (2008) Erythropoietin protects crit-\nically perfused flap tissue. Ann. Surg. 248:919–29.\n13. Sayan H, Ozacmak VH, Guven A, Aktas RG,\nOzacmak ID. (2006) Erythropoietin stimulates\nwound healing and angiogenesis in mice. J. In-\nvest. Surg. 19:163–73.\n14. Harder Y, et al. (2009) Erythropoietin reduces\nnecrosis in critically ischemic myocutaneous tis-\nsue by protecting nutritive perfusion in a dose-\ndependent manner. Surgery 145:372–83.\n15. Kim EK, Hong JP . (2007) The effect of recombi-\nnant human erythropoietin on ischemia-reperfu-\nsion injury: an experimental study in a rat TRAM\nflap model. Plast. Reconstr. Surg. 120:1774–81.\n16. Fatouros M, Dalekos GN, Mylonakis E, Vekinis\nG, Kappas AM. (1999) Alterations in body\nweight, breaking strength, and wound healing in\nWistar rats treated pre- and postoperatively with\nerythropoietin or granulocyte macrophage-\ncolony stimulating factor: evidence of a previ-\nously unknown anabolic effect of erythropoietin?\nJ. Lab. Clin. Med. 133:253–9.\n17. Fatouros MS, et al. (1999) Influence of growth fac-\ntors erythropoietin and granulocyte macrophage\ncolony stimulating factor on mechanical strength\nand healing of colonic anastomoses in rats. Eur. J.\nSurg. 165:986–92.\n18. Galeano M, et al. (2006) Recombinant human ery-\nthropoietin improves angiogenesis and wound\nhealing in experimental burn wounds. Crit. Care\nMed. 34:1139–46.\n19. Haroon ZA, Amin K, Jiang X, Arcasoy MO.\n(2003) A novel role for erythropoietin during fib-\nrin-induced wound-healing response. Am. J.\nPathol. 163:993–1000.\n20. Keast DH, Fraser C. (2004) Treatment of chronic\nskin ulcers in individuals with anemia of chronic\ndisease using recombinant human erythropoietin\n(EPO): a review of four cases. Ostomy Wound\nManage. 50:64–70.\n21. National Research Council, Commission on Life\nSciences, Institute of Laboratory Animal Re-\nsources. (1996) Guide for the care and use of lab-\noratory animals. Washington, DC: National\nAcademy Press. 140 pp.\n22. Kiecolt-Glaser JK, Marucha PT, Malarkey WB,\nMercado AM, Glaser R. (1995) Slowing of wound\nhealing by psychological stress. Lancet 346:1194–6.\n23. McFarlane RM, Deyoung G, Henry RA. (1965)\nThe design of a pedicle flap in the rat to study\nnecrosis and its prevention. Plast. Reconstr. Surg\n .\n35:177–82.\n24. Peirce SM, Skalak TC, Rodeheaver GT. (2000) Is-\nchemia-reperfusion injury in chronic pressure\nulcer formation: a skin model in the rat. Wound\nRepair Regen. 8:68–76.\n25. Bothin C, Okada M, Midtvedt T, Perbeck L.\n(2001) The intestinal flora influences adhesion\nformation around surgical anastomoses. Br. J.\nSurg. 88:143–5.\n26. Oppenheim JJ, Yang D. (2005) Alarmins: chemo-\ntactic activators of immune responses. Curr.\nOpin. Immunol. 17:359–65.\n27. Klune JR, Dhupar R, Cardinal J, Billiar TR, Tsung\nA. (2008) Hmgb1: endogenous danger signaling.\nMol. Med. 14:476–84.\n28. Yang H, Wang H, Czura CJ, Tracey KJ. (2005) The\ncytokine activity of HMGB1. J. Leukoc. Biol. 78:1–8.\n29. Menger MD, Pelikan S, Steiner D, Messmer K.\n(1992) Microvascular ischemia-reperfusion injury\nin striated muscle: significance of “reflow para-\ndox.” Am. J. Physiol. 263: H1901–6.\n30. Menger MD, Steiner D, Messmer K. (1992) Mi-\ncrovascular ischemia-reperfusion injury in stri-\nated muscle: significance of “no reflow.” Am. J.\nPhysiol. 263: H1892–900.\n31. Westenbrink BD, et al. (2007) Erythropoietin im-\nproves cardiac function through endothelial pro-\ngenitor cell and vascular endothelial growth fac-\ntor mediated neovascularization. Eur. Heart J.\n28:2018–27.\n32. Wang L, et al. (2006) Matrix metalloproteinase 2\n(MMP2) and MMP9 secreted by erythropoietin-\nactivated endothelial cells promote neural pro-\ngenitor cell migration. J. Neurosci. 26:5996–6003.\n33. Saray A, et al. (2003) Effect of chronic and short-\nterm erythropoietin treatment on random flap\nsurvival in rats: an experimental study. Laryngo-\nscope 113:85–9.\nRESEARCH ARTICLE\nMOL MED 15(7-8)235-241, JULY-AUGUST 2009 | ERBAYRAKTAR ET AL. | 241","source_license":"CC-BY-4.0","license_restricted":false}