- Email: [email protected]
Erythropoietin’s Beta Common Receptor Mediates Neuroprotection in Spinal Cord Neurons
CARDIOTHORACIC ANESTHESIOLOGY: The Annals of Thoracic Surgery CME Program is located online at http://www.annalsthoracicsurgery. org/cme/home. To take the CME activity related to this article, you must have either an STS member or an individual non-member subscription to the journal.
Lisa S. Foley, MD, David A. Fullerton, MD, Joshua Mares, BA, Mitchell Sungelo, BA, Michael J. Weyant, MD, Joseph C. Cleveland, Jr, MD, and T. Brett Reece, MD Department of Surgery, Division of Cardiothoracic Surgery, University of Colorado Denver, Aurora, Colorado
Background. Paraplegia from spinal cord ischemiareperfusion (SCIR) remains an elusive and devastating complication of complex aortic operations. Erythropoietin (EPO) attenuates this injury in models of SCIR. Upregulation of the EPO beta common receptor (bcR) is associated with reduced damage in models of neural injury. The purpose of this study was to examine whether EPOmediated neuroprotection was dependent on bcR expression. We hypothesized that spinal cord neurons subjected to oxygen-glucose deprivation would mimic SCIR injury in aortic surgery and EPO treatment attenuates this injury in a bcR-dependent fashion. Methods. Lentiviral vectors with bcR knockdown sequences were tested on neuron cell cultures. The virus with greatest bcR knockdown was selected. Spinal cord neurons from perinatal wild-type mice were harvested and cultured to maturity. They were treated with knockdown or nonsense virus and transduced cells were selected. Three groups (bcR knockdown virus, nonsense
control virus, no virus control; n [ 8 each) were subjected to 1 hour of oxygen-glucose deprivation. Viability was assessed. bcR expression was quantified by immunoblot. Results. EPO preserved neuronal viability after oxygen-glucose deprivation (0.82 ± 0.04 versus 0.61 ± 0.01; p < 0.01). Additionally, EPO-mediated neuron preservation was similar in the nonsense virus and control mice (0.82 ± 0.04 versus 0.80 ± 0.05; p [ 0.77). EPO neuron preservation was lost in bcR knockdown mice compared with nonsense control mice (0.46 ± 0.03 versus 0.80 ± 0.05; p < 0.01). Conclusions. EPO attenuates neuronal loss after oxygen-glucose deprivation in a bcR-dependent fashion. This receptor holds immense clinical promise as a target for pharmacotherapies treating spinal cord ischemic injury.
P
injured, though has potential for salvage. It is in this region where aberrant blood flow and immunologic processes can advance the lesion to an infarct, ultimately leading to second peak of neurologic decline after the initial insult [7]. Literature and clinical experience has shown that spinal cord injury after aortic surgery mirrors the bimodal injury pattern that characterizes cerebral ischemic injury [8–10]. The first peak in neurologic damage is due to the intraoperative ischemic insult. This injury primarily results in cellular necrosis, and is often subclinical. In response to cytokines released from necrotic cell death, a secondary inflammatory phase develops. Reperfusion delivers immune mediators that combine with proinflammatory cytokines surrounding the ischemic lesion to activate cell death pathways and amplify the damage [7]. Discovering ways to protect this region of vulnerable tissue from secondary insults and progression of injury is of great interest in neurologic research. Preventing this pathophysiologic response has proven promising in translational brain and spinal cord injury studies. Erythropoietin (EPO) and its receptor are upregulated in the brain and spinal cord after hypoxia and
araplegia after thoracoabdominal aneurysm repairs remains a devastating and elusive problem [1, 2]. Although advances in operative protective techniques have evolved over the past decade, there has yet to be a widespread improvement in the incidence of this complication [3–5]. Furthermore, there are no pharmacologic adjuncts proven to prevent paraplegia from spinal cord ischemia [5, 6]. Ischemia-reperfusion injury, as it occurs in neurologic tissue, has been studied more extensively in the brain than the spinal cord, using ischemic stroke models. These stroke studies have demonstrated that after an ischemic insult there are zones of injury radiating out from the primary lesion. At the center is a core of necrotic tissue, which is irreversibly damaged. This central necrotic core is surrounded by an area of stunned tissue that is severely
Accepted for publication July 20, 2017. Presented at the Sixty-third Annual Meeting of the Southern Thoracic Surgical Association, Naples, FL, Nov 9–12, 2016. Address correspondence to Dr Foley, 12631 E 17th Ave, C302, Aurora, CO 80045; email: [email protected].
Ó 2017 by The Society of Thoracic Surgeons Published by Elsevier Inc.
(Ann Thorac Surg 2017;104:1909–14) Ó 2017 by The Society of Thoracic Surgeons
0003-4975/$36.00 https://doi.org/10.1016/j.athoracsur.2017.07.052
ADULT CARDIAC
Erythropoietin’s Beta Common Receptor Mediates Neuroprotection in Spinal Cord Neurons
ADULT CARDIAC
1910
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
inflammatory signaling. It has emerged as a neuroprotective cytokine, with proven efficacy in murine models of spinal cord ischemia-reperfusion injury [11–14]. Two characteristics of EPO highlight its potential role in neuroprotection as well as its potential for therapeutic applications: (1) it is locally produced in the brain in response to ischemic injury and (2) it crosses the blood-brain barrier, likely through a transport mechanism due to its size [15]. These findings have implicated EPO’s innate function in neurologic signaling and potential for therapeutic uses. EPO activates 2 distinct receptors. The classically known receptor is a homodimer of 2 EPO receptor subunits (EPO-R þ EPO-R), which mediates hematopoiesis. Activating this receptor on progenitor erythrocytes halts apoptosis, allowing progenitor cells to mature into circulating erythrocytes. A newly identified EPO receptor consists of a heterodimer of an EPO-R subunit paired with the interleukin beta common receptor (bcR) subunit (known as bcR or CD131) that is expressed in various solid organ tissues including the brain, heart, and kidney. This receptor initiates antiapoptotic signaling and mediates tissue protection after injury in these various tissues [16]. bcR subunit expression is low at baseline and induced by hypoxia or metabolic stress. This receptor is primarily expressed in injured or metabolically stressed tissue and is believed to mediate cell preservation pathways. Expression of this receptor on injured cells precedes a local increase in tissue production of EPO temporally. Thus, the expression of this tissue-protective receptor occurs more rapidly than production and diffusion of its ligand, EPO, creating a therapeutic window where exogenous EPO administration could hasten the innate neuroprotective mechanisms. Our lab has previously demonstrated EPO treatment reduces paraplegia in a murine model of spinal cord ischemia-reperfusion injury [11, 17]. Additionally, we have shown improved outcomes with bcR upregulation and survival after ischemia-reperfusion injury [18]. These mechanisms were both observed but it remained unclear if the bcR was specifically mediating EPO’s neuroprotective effects. There is no bcR inhibitor available, limiting the ability to perform blocking studies. The purpose of this study was to determine if the bcR is essential for EPO-mediated neuroprotection. A murine spinal cord neuron in vitro model that mimics spinal cord ischemia-reperfusion injury was employed to better explore these cellular mechanisms. We hypothesized that bcR induction by ischemic injury could be blunted using a viral knockdown, and that lack of bcR expression would abrogate EPO’s neuroprotective effects.
Material and Methods Materials EPO was purchased from Sigma-Aldrich (St. Louis, Missouri). Anti-bcR antibody was purchased from Santa Cruz Biotechnology (Dallas, Texas). Lentiviral vectors containing gene-silencing plasmids were obtained through the Functional Genomics Core at the University of Colorado.
Ann Thorac Surg 2017;104:1909–14
Animals The University of Colorado Denver Health Sciences Center Animal Care and Use Committee approved all experiments. Experiments adhered to the Guide for the Care and Use of Laboratory Animals. Cells for in vitro experiments were obtained from postnatal day 2 to 3 pups from wild-type (C57/BL6) breeding pairs (Charles Rivers Laboratories, Frederick, Maryland). All cells harvested from a litter of pups were considered n ¼ 1.
Cell Culture Primary spinal cord neuron cultures were obtained from 2- to 3-day-old wild-type (C57-BL6) mice. The vertebral column was isolated immediately after euthanasia with isoflurane and decapitation. The spinal cord was flushed from the spinal canal en bloc with cold phosphatebuffered saline (pH 7.4). Cord tissue was minced and digested in a solution of Hibernate-A (Invitrogen, Carlsbad, California) with papain (Worthington, Lakewood, New Jersey). Neurons were isolated from the digested cord tissue using an Optiprep (Sigma-Aldrich) density gradient adapted from Brewer [18]. Neurons were plated in 1mL culture media of Neurobasal-A, B27, GlutaMAX (all obtained from Invitrogen), and penicillin or streptomycin (Gibco, New York, New York) on poly-D-Lysine (Sigma-Aldrich)–coated plates at approximately 300,000 cells/well. Cell were cultured in a 37 C, 5% CO2 humidified atmosphere. Half of the media was replaced every 2 days with fresh media containing AraC (Sigma-Aldrich) to prevent astrocyte growth. The cultures were confirmed to have >90% neurons as seen by morphology on light microscope as well as confirmed with microtubule-associated protein 2– positive neuronal staining. Cells were used at in vitro day 5 for experimentation, which is considered mature for neuronal cultures.
Lentiviral Knockdown of EPO bcR Lentiviral vectors containing short hairpin RNA gene silencing clones for the bcR subunit of the EPO receptor were obtained through the University of Colorado Denver Functional Genomics Core (University of Colorado Cancer Center, Aurora, CO). Similarly, lentiviral control vectors containing the same plasmid with a nonsense target sequence were obtained. Both the experimental and control plasmids contained a puromycin resistance gene for selection of transduced cells. We obtained 4 variations of a bcR knockdown virus and examined cellular viability after transduction with viability assays (as described subsequently). Additionally, we examined receptor expression levels after treatment to quantify the knockdown success. After identification of a knockdown virus with a high rate of receptor knockdown and minimal effect on viability was identified, this virus was used for all knockdown experiments. Mature spinal cord neuronal cultures were treated with control or bcR knockdown lentiviral vectors at a total concentration of 104 relative infectious units per milliliter for 16 hours with polybrene (2 mL of 1000X). Additionally,
Ann Thorac Surg 2017;104:1909–14
Oxygen-Glucose Deprivation and EPO Treatment In previous studies, the authors have shown that EPO treatment and bcR induction provided neuronal protection and prevented cell loss after ischemia in vivo and oxygen-glucose deprivation (OGD) in vitro [11, 19]. For this study, we further investigated that finding by subjecting neurons to OGD with EPO treatment in the absence of the bcR, by inhibiting its expression, to determine if this receptor was responsible for EPOmediated neuroprotection. On the day of the experiment, the experimental medium of Neurobasal-A without glucose (Invitrogen) was placed in a Ruskinn Bug Box Plus (The Baker Company, Sanford, Maine) humidified airtight hypoxic chamber for 2 hours. The Ruskinn Bug Box Plus was used per manufacturer’s protocol to maintain an environment of 95% N2 and 5% CO2 at 37 C. A hypoxic environment was verified with Anaerobic Indicator Strips (Oxoid Ltd, Basingstoke, United Kingdom) before placement of media in the chamber, and hypoxia of the media after 2 hours was tested with an indicator strip. The plates were then introduced into the hypoxic environment where puromycin selection culture medium was removed, cells were washed with phosphate-buffered saline, and 1 mL of hypoxic glucose-deprived medium was added to all cell culture wells. Neurons transduced with the bcR knockdown plasmid or the nonsense plasmid were subjected to 2 periods of OGD. The first period was a 30-minute exposure, which has previously been shown to be too short to induce significant injury, but enough hypoxia stimuli to induce bcR expression [5]. This first exposure was followed by 24 hours of suspension in normal cellular media to allow for expression of the bcR in control neurons. The neurons were then subjected to OGD for 1 hour, a dose previously shown to induce about 40% cell death in untreated spinal cord neurons. The hypoxic media used for these experiments contained EPO for the EPO treatment groups. It was at this time in the experiments that EPO treatment is introduced and continued throughout. After completion of OGD exposure, neurons were placed in culture media ( EPO) in a humidified, oxygenated culture chamber for 24 hours of reperfusion.
1911
At this time, final viability studies as well as receptor expression studies were performed.
Viability Studies For viability studies, neuronal cells were cultured on 24well plates. Cell viability was determined using an MTT Cell Viability Kit (Roche, Indianapolis, Indiana), as previously described by quantifiable chemiluminescence [5]. Absorbance was measured on a BioTek Synergy H1 Hybrid microplate reader (BioTek Instruments Inc, Winooski, Vermont). Cell viability is presented as the percentage absorbance relative to the control.
Western Blot Analysis Spinal cords were homogenized. Protein concentration in each sample was quantified with NanoDrop (Thermo Scientific, Wilmington, Delaware). A 4% to 20% gradient gel was used to separate proteins that were then transferred to a nitrocellulose membrane. After blocking in 5% milk in bovine serum albumin, the membranes were incubated at 4 C in 1:500 primary antibody (rabbit antimouse bcR). The membranes were then washed sequentially again and incubated in secondary anti-rabbit antibody at room temperature for 1 hour. Excess secondary antibody was washed off. Chemiluminescence was used to identify the target band and band density was quantified with ImageJ software (National Institutes of Health, Bethesda, Maryland). Quantification of receptor protein was compared with overall glyceradehyde-3phosphate expressed in each sample.
Statistics Sample size for studies was determined by power analysis with plans for a power of 0.90 to and 2-sided alpha set at 0.05. A sample size of 4 was determined adequate for the differences expected to result and these experiments were performed with a sample size of 8 to account for the variability in starting samples that the knockdown step introduced. Data are presented as mean SD. Analysis was performed using StatView Version 5.0 (SAS Institute Inc, Cary North Carolina). One-way analysis of variance or t test were performed, and a p value of <0.05 was considered significant for all statistical comparisons.
Results Spinal Cord Cell Cultures As previously described, using the spinal cord neuron isolation protocol, we were able to isolate and maintain primary spinal cord neuronal cultures [17, 18]. These cultures matured appropriately, as demonstrated by their network of projections (Fig 1). Neuron cultures were maintained at >90% neurons present, verified under light microscopy and corroborated by microtubule-associated protein 2–positive neuronal staining. These cells proved robust enough to withstand transduction (introduction of genetic material using a virus vector), puromycin selection, and hypoxic experiments.
ADULT CARDIAC
2 wells on each 24-well plate were treated with 100 mL of Dulbecco’s modified Eagle medium (Gibco) alone, which is the media in which the lentiviral vectors were suspended. These 2 wells were used as references to determine the time at which all nontransduced cells had been killed by the puromycin selection process. After 16 hours of lentiviral vector treatment, the supernatant containing the virus (or no virus for the 2 reference wells) was removed from all wells and replaced with fresh media of Neurobasal-A with B27, GlutaMAX, and puromycin (1 mg/mL). The plates were examined under light microscopy until all nontransduced cells (all of the 2 reference wells) were dead. This selection was reliably completed by 2 days, at which time the neurons were ready to undergo experimental treatments.
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
Ann Thorac Surg 2017;104:1909–14
ADULT CARDIAC
1912
Fig 1. Primary spinal cord neurons in culture. Primary spinal cord neurons are identified by a network of projections when cultured to maturity.
Spinal Cord Neuron Viability After Transduction The first part of our project focused on identifying a lentiviral vector (LV) with minimal effect on neuron viability while also having excellent knockdown capabilities. In these studies, the viability is presented as a percentage absorbance relative to the control nonsense lentiviral treatment. A sample size of 8 litters was used for all cell culture experiments, compared against the average of 2 control neurons. The data demonstrated that LV2 was the least harmful to the neurons with an overall survival rate of 74.6% 1.96% (p < 0.01). LV1 treatment resulted in a survival rate of 55.4% 5.1%. Both LV3 and LV4 treatments were poorly tolerated by the neurons, with low survival rates at 29.3% 14.7% and 26.8% 1.7%, respectively (Fig 2).
Lentiviral Knockdown Studies The bcR is characteristically expressed in very low quantities at baseline. It is induced by hypoxia and other
Fig 3. Beta common receptor expression after transduction with knockdown virus. LV1, LV2, and LV4 successfully suppressed beta common receptor expression. Based on these findings along with the previous viability findings, LV2 was chosen as the knockdown virus used for all further experiments. (LV ¼ lentiviral vector.)
cellular stress stimuli. For this reason, a brief OGD exposure was performed on all transduced cells to induce the receptor and identify if the knockdown virus was successful at blunting this induction. These data are presented as bcR expression relative to the control cells treated with the nonsense control lentivirus (Fig 3). Nearly all viruses successfully prevented expression of the bcR after hypoxia, with LV1, LV2, and LV4 causing less than 5% receptor expression compared with control viruses (p < 0.01). Significant amount of bcR expression was still present after transduction with LV3. The results of the neuron viability studies were combined with bcR knockdown studies and LV2 was used for all knockdown experiments for the remainder of the project.
Neuron Viability After OGD With EPO Treatment EPO significantly preserved neuronal viability after OGD treatment (mean 0.82 0.04 versus 0.61 0.01; p < 0.01) (Fig 4). Additionally, EPO-mediated neuronal preservation was similar in the nonsense virus and cells not treated with virus (mean 0.82 0.04 versus 0.80 0.05; p ¼ 0.77). EPO attenuation of neuronal injury was lost in bcR knockdown cells compared with nonsense control cells (mean 0.46 0.03 versus 0.80 0.05; p < 0.01).
Comment
Fig 2. Neuron viability after transduction. LV2 resulted in the least cell loss after transduction experiments. (LV ¼ lentiviral vector.)
Delayed paraplegia is a devastating and humbling complication of complex thoracoabdominal aortic operations, which still haunts even the most skilled aortic surgeons. Investigation into the mechanisms of innate neuroprotection from ischemia provide novel avenues of potential treatment for preventing this injury. Our lab has previously shown that EPO treatment prevents paraplegia in a murine model of spinal cord ischemia [11, 19]. Additionally, we have shown that neuroprotective pathways of EPO treatment occur more robustly after bcR induction [17]. The results of the present study show that
Ann Thorac Surg 2017;104:1909–14
the interleukin bcR is an essential mediator in the neuroprotective pathway activated by EPO. Identifying and characterizing this receptor’s role in EPO-mediated neuroprotection holds vast clinical implications. Thrombotic side effects of EPO, mediated by the homodimer receptor on platelets, have hindered its clinical use for treating spinal cord ischemia. The bcR activates pathways separate from those of the homodimer receptor. Specifically targeting the bcR provides a clinical avenue to maximize the antiapoptotic and tissue protective properties of EPO without the unwanted stroke and clot risk associated with its use [7]. Derivatives of EPO, such as carbamylated EPO, and other nonhematopoietic versions of this cytokine have been found to provide neuroprotection against stroke, spinal cord compression, and diabetic neuropathy without hematologic effects [14]. Use of a tissue protective, but not hematopoietic, derivative of EPO in aortic surgery is a promising direction for identifying pharmacologic treatments for spinal cord ischemia. This study provides groundwork for further investigation into the neuroprotective potential of EPO and its derivatives in spinal cord ischemia by establishing the bcR’s essential role in this pathway. There are notable limitations to this study to mention. This study was performed using murine spinal cord neurons, a nonhuman rudimentary model. The steps to translate these findings to clinical practice are many. However, this spinal cold cell culture model using OGD has repeatedly mimicked our findings in a mouse model of spinal cord ischemia. We elected to use the cell culture model to conduct EPO and bcR knockdown experiments
1913
and distill out the components of this mechanism. Future studies are planned to further examine these findings in animal models, using derivatives of EPO specific for this receptor. EPO’s emerging role as a neuroprotective therapy is being studied in a variety of different clinical trials. Its role in neonatal neuroprotection after perinatal ischemia is currently of active interest globally [20]. It was also recently studied in aortic surgery, though that clinical trial was interrupted due to concerns for stroke risk reported elsewhere in the literature, although not seen in that trial [21]. Preclinical studies are confirming promising findings in animal models of spinal cord injury [22, 23]. These cell culture experiments are important tools for breaking down the components of in vivo findings to identify molecular targets for therapy. In summary, this study has proven that bcR knockdown is possible in cell culture of murine spinal cord neurons. More importantly, we have demonstrated that knocking down the bcR abrogates EPO-mediated neuroprotection, highlighting its central role in this mechanism. In conclusion, spinal cord ischemia-reperfusion injury remains a devastating and potentially preventable complication. Further research into the innate mechanisms of spinal cord neuroprotection may elucidate new avenues for therapeutic advances. The authors wish to thank the University of Colorado Denver, Department of Surgery, for their financial support of this project.
References 1. Coselli JS, Lemaire SA, Conklin LD, K€ oksoy C, Schmittling ZC. Morbidity and mortality after extent II thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2002;73:1107–15. 2. Coselli J, Lemaire SA, Preventza O, et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg 2016;151:1323–38. 3. Conrad MF, Ergul EA, Patel VI, et al. Evolution of operative strategies in open thoracoabdominal aneurysm repair. J Vasc Surg 2011;53:1195–201. 4. Conrad MF, Ye JY, Chung TK, Davison JK, Cambria RP. Spinal cord complications after thoracic aortic surgery: Longterm survival and functional status varies with deficit severity. J Vasc Surg 2008;48:47–53. 5. Foley LS, Reece TB. Advances in spinal cord protection for complex aortic repairs. J Thorac Cardiovasc Surg 2016;151: 614–5. 6. Onose G, Anghelescu A, Muresanu DF, et al. A review of published reports on neuroprotection in spinal cord injury. Spinal Cord 2009;47:716–26. 7. Brines M, Cerami A. Erythropoietin-mediated tissue protection: reducing collateral damage from the primary injury response. J Intern Med 2008;264:405–32. 8. Stirling DP, Yong VW. Dynamics of the inflammatory response after murine spinal cord injury revealed by flow cytometry. J Neurosci Res 2008;86:1944–58. 9. Smith PD, Puskas F, Meng X, et al. The evolution of chemokine release supports a bimodal mechanism of spinal cord ischemia and reperfusion injury. Circulation 2012;126(Suppl 1):110–7.
ADULT CARDIAC
Fig 4. Spinal cord neuron viability after oxygen-glucose deprivation (OGD). Erythropoietin (EPO) significantly preserved neuronal viability after OGD treatment. EPO-mediated attenuation of neuronal injury was lost in beta common receptor (bcR) knockdown cells compared with nonsense control cells. (*Clinical significance, p < 0.05.)
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
ADULT CARDIAC
1914
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
10. Smith PD, Puskas F, Meng X, et al. Ischemic dose-response in the spinal cord: both immediate and delayed paraplegia. J Surg Res 2012;174:238–44. 11. Smith PD, Puskas F, Fullerton DA, et al. Attenuation of spinal cord ischemia and reperfusion injury by erythropoietin. J Thorac Cardiovasc Surg 2011;141:256–60. 12. Hirano K, Wagner K, Mark P, et al. Erythropoietin attenuates the sequels of ischaemic spinal cord injury with enhanced recruitment of CD34þ cells in mice. J Cell Mol Med 2012;16:1792–802. 13. Beattie MS. Inflammation and apoptosis: linked therapeutic targets in spinal cord injury. Trends Mol Med 2004;10:580–3. 14. Leist M, Ghezzi P, Grasso G, et al. Derivative of erythropoietin that are tissue protective but not erythropoietin. Science 2004;305:239–42. 15. Brines ML, Ghezzi P, Keenan S, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 2000;97:10526–31. 16. Brines M, Cerami A. The receptor that tames the innate immune response. Mol Med 2012;18:486–96. 17. Foley LS, Fullerton DA, Bennett DT, et al. Spinal cord ischemia-reperfusion injury induces erythropoietin receptor expression. Ann Thorac Surg 2015;100:41–6.
Ann Thorac Surg 2017;104:1909–14
18. Brewer G, Torricelli J. Isolation and culture of adult neurons and neurospheres. Nat Protoc 2007;2:1490–8. 19. Mares JM, Foley LS, Bell M, et al. Erythropoietin activates the phosporylated cAMP [adenosine 30 50 cyclic monophosphate] response element-binding protein pathway and attenuates delayed paraplegia after ischemia-reperfusion injury. J Thorac Cardiovasc Surg 2015;149:920–4. 20. Rangarajan V, Juul SE. Erythropoietin: emerging role of erythropoietin in neonatal neuroprotection. Pediatr Neurol 2014;51:481–8. 21. Messe SR, McGarvey ML, Bavaria JE, et al. A pilot study of darbepoetin alfa for prophylactic neuroprotection in aortic surgery. Neurocrit Care 2013;18:75–80. 22. Yang L, Yan X, Xu Z, Tan W, Chen Z, Wu B. Delayed administration of recombinant human erythropoietin reduces apoptosis and inflammation and promotes myelin repair and functional recovery following spinal cord compressive injury in rats. Restor Neurol Neurosci 2015;34: 647–63. 23. Yilmaz ER, Kermen H, Dolgun H, et al. Effects of darbepoetin-a in spinal cord ischemia-reperfusion injury in the rabbit. Acta Neurochir 2012;154:1037–44.
DISCUSSION DR SCOTT A. LEMAIRE (Houston, TX): That was very nicely presented. I would like to start by congratulating Dr Foley, Dr Reece, and their colleagues on a very elegant study to identify the critical role of interleukin beta common receptor (bcR) in erythropoietin’s protective role in ischemic neurons. The first question I have is about your endpoint. Your primary endpoint in this experiment was neuron viability. Did you assess any other endpoints, such as caspase 3, which is a marker for neuron apoptosis, or any downstream signaling pathways related to heterodimer receptor activation? DR FOLEY: We did in previous studies and we plan to in future studies. In previous studies using neuron viability we looked at markers of neuron cell death, we specifically looked at caspase, actually, but we did not in this study yet and we plan to. These are all n ¼ 8 and this is an exceptional amount of mice that went into this. DR LEMAIRE: My next question is about the other cells within the spinal cord. So astrocytes and microglia are also involved in erythropoietin’s modulation of neuroinflammation. Do you believe that the effects of erythropoietin on these cell types are also dependent on the heterodimer receptor?
DR FOLEY: In the last study that we presented actually at the STSA a couple of years ago, we demonstrated the bcR is upregulated in astrocytes as well, and we want to go further with it. This is our first thing using the lentiviral knockdowns, but, yes, I do think that it acts at least on the astrocytes, we know that for sure, but we have not done microglia yet. DR LEMAIRE: My last question is related to, you have got the in vitro model showing that this receptor is important and now about validation in the in vivo models. Other investigators have used bcR knockout mice to study neuropathic pain and other nerve diseases. Are you planning to expand your investigation by determining whether bcR-null mice remain susceptible to spinal cord ischemia despite treatment with erythropoietin or perhaps whether neurons from these animals would exhibit similar levels of apoptosis when subjected to oxygen and glucose depletion with and without erythropoietin? DR FOLEY: That is a very good question. We have actually been looking into using an animal knockdown, and we are using a neighboring lab to try to get that accomplished.
Lisa S. Foley, MD, David A. Fullerton, MD, Joshua Mares, BA, Mitchell Sungelo, BA, Michael J. Weyant, MD, Joseph C. Cleveland, Jr, MD, and T. Brett Reece, MD Department of Surgery, Division of Cardiothoracic Surgery, University of Colorado Denver, Aurora, Colorado
Background. Paraplegia from spinal cord ischemiareperfusion (SCIR) remains an elusive and devastating complication of complex aortic operations. Erythropoietin (EPO) attenuates this injury in models of SCIR. Upregulation of the EPO beta common receptor (bcR) is associated with reduced damage in models of neural injury. The purpose of this study was to examine whether EPOmediated neuroprotection was dependent on bcR expression. We hypothesized that spinal cord neurons subjected to oxygen-glucose deprivation would mimic SCIR injury in aortic surgery and EPO treatment attenuates this injury in a bcR-dependent fashion. Methods. Lentiviral vectors with bcR knockdown sequences were tested on neuron cell cultures. The virus with greatest bcR knockdown was selected. Spinal cord neurons from perinatal wild-type mice were harvested and cultured to maturity. They were treated with knockdown or nonsense virus and transduced cells were selected. Three groups (bcR knockdown virus, nonsense
control virus, no virus control; n [ 8 each) were subjected to 1 hour of oxygen-glucose deprivation. Viability was assessed. bcR expression was quantified by immunoblot. Results. EPO preserved neuronal viability after oxygen-glucose deprivation (0.82 ± 0.04 versus 0.61 ± 0.01; p < 0.01). Additionally, EPO-mediated neuron preservation was similar in the nonsense virus and control mice (0.82 ± 0.04 versus 0.80 ± 0.05; p [ 0.77). EPO neuron preservation was lost in bcR knockdown mice compared with nonsense control mice (0.46 ± 0.03 versus 0.80 ± 0.05; p < 0.01). Conclusions. EPO attenuates neuronal loss after oxygen-glucose deprivation in a bcR-dependent fashion. This receptor holds immense clinical promise as a target for pharmacotherapies treating spinal cord ischemic injury.
P
injured, though has potential for salvage. It is in this region where aberrant blood flow and immunologic processes can advance the lesion to an infarct, ultimately leading to second peak of neurologic decline after the initial insult [7]. Literature and clinical experience has shown that spinal cord injury after aortic surgery mirrors the bimodal injury pattern that characterizes cerebral ischemic injury [8–10]. The first peak in neurologic damage is due to the intraoperative ischemic insult. This injury primarily results in cellular necrosis, and is often subclinical. In response to cytokines released from necrotic cell death, a secondary inflammatory phase develops. Reperfusion delivers immune mediators that combine with proinflammatory cytokines surrounding the ischemic lesion to activate cell death pathways and amplify the damage [7]. Discovering ways to protect this region of vulnerable tissue from secondary insults and progression of injury is of great interest in neurologic research. Preventing this pathophysiologic response has proven promising in translational brain and spinal cord injury studies. Erythropoietin (EPO) and its receptor are upregulated in the brain and spinal cord after hypoxia and
araplegia after thoracoabdominal aneurysm repairs remains a devastating and elusive problem [1, 2]. Although advances in operative protective techniques have evolved over the past decade, there has yet to be a widespread improvement in the incidence of this complication [3–5]. Furthermore, there are no pharmacologic adjuncts proven to prevent paraplegia from spinal cord ischemia [5, 6]. Ischemia-reperfusion injury, as it occurs in neurologic tissue, has been studied more extensively in the brain than the spinal cord, using ischemic stroke models. These stroke studies have demonstrated that after an ischemic insult there are zones of injury radiating out from the primary lesion. At the center is a core of necrotic tissue, which is irreversibly damaged. This central necrotic core is surrounded by an area of stunned tissue that is severely
Accepted for publication July 20, 2017. Presented at the Sixty-third Annual Meeting of the Southern Thoracic Surgical Association, Naples, FL, Nov 9–12, 2016. Address correspondence to Dr Foley, 12631 E 17th Ave, C302, Aurora, CO 80045; email: [email protected].
Ó 2017 by The Society of Thoracic Surgeons Published by Elsevier Inc.
(Ann Thorac Surg 2017;104:1909–14) Ó 2017 by The Society of Thoracic Surgeons
0003-4975/$36.00 https://doi.org/10.1016/j.athoracsur.2017.07.052
ADULT CARDIAC
Erythropoietin’s Beta Common Receptor Mediates Neuroprotection in Spinal Cord Neurons
ADULT CARDIAC
1910
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
inflammatory signaling. It has emerged as a neuroprotective cytokine, with proven efficacy in murine models of spinal cord ischemia-reperfusion injury [11–14]. Two characteristics of EPO highlight its potential role in neuroprotection as well as its potential for therapeutic applications: (1) it is locally produced in the brain in response to ischemic injury and (2) it crosses the blood-brain barrier, likely through a transport mechanism due to its size [15]. These findings have implicated EPO’s innate function in neurologic signaling and potential for therapeutic uses. EPO activates 2 distinct receptors. The classically known receptor is a homodimer of 2 EPO receptor subunits (EPO-R þ EPO-R), which mediates hematopoiesis. Activating this receptor on progenitor erythrocytes halts apoptosis, allowing progenitor cells to mature into circulating erythrocytes. A newly identified EPO receptor consists of a heterodimer of an EPO-R subunit paired with the interleukin beta common receptor (bcR) subunit (known as bcR or CD131) that is expressed in various solid organ tissues including the brain, heart, and kidney. This receptor initiates antiapoptotic signaling and mediates tissue protection after injury in these various tissues [16]. bcR subunit expression is low at baseline and induced by hypoxia or metabolic stress. This receptor is primarily expressed in injured or metabolically stressed tissue and is believed to mediate cell preservation pathways. Expression of this receptor on injured cells precedes a local increase in tissue production of EPO temporally. Thus, the expression of this tissue-protective receptor occurs more rapidly than production and diffusion of its ligand, EPO, creating a therapeutic window where exogenous EPO administration could hasten the innate neuroprotective mechanisms. Our lab has previously demonstrated EPO treatment reduces paraplegia in a murine model of spinal cord ischemia-reperfusion injury [11, 17]. Additionally, we have shown improved outcomes with bcR upregulation and survival after ischemia-reperfusion injury [18]. These mechanisms were both observed but it remained unclear if the bcR was specifically mediating EPO’s neuroprotective effects. There is no bcR inhibitor available, limiting the ability to perform blocking studies. The purpose of this study was to determine if the bcR is essential for EPO-mediated neuroprotection. A murine spinal cord neuron in vitro model that mimics spinal cord ischemia-reperfusion injury was employed to better explore these cellular mechanisms. We hypothesized that bcR induction by ischemic injury could be blunted using a viral knockdown, and that lack of bcR expression would abrogate EPO’s neuroprotective effects.
Material and Methods Materials EPO was purchased from Sigma-Aldrich (St. Louis, Missouri). Anti-bcR antibody was purchased from Santa Cruz Biotechnology (Dallas, Texas). Lentiviral vectors containing gene-silencing plasmids were obtained through the Functional Genomics Core at the University of Colorado.
Ann Thorac Surg 2017;104:1909–14
Animals The University of Colorado Denver Health Sciences Center Animal Care and Use Committee approved all experiments. Experiments adhered to the Guide for the Care and Use of Laboratory Animals. Cells for in vitro experiments were obtained from postnatal day 2 to 3 pups from wild-type (C57/BL6) breeding pairs (Charles Rivers Laboratories, Frederick, Maryland). All cells harvested from a litter of pups were considered n ¼ 1.
Cell Culture Primary spinal cord neuron cultures were obtained from 2- to 3-day-old wild-type (C57-BL6) mice. The vertebral column was isolated immediately after euthanasia with isoflurane and decapitation. The spinal cord was flushed from the spinal canal en bloc with cold phosphatebuffered saline (pH 7.4). Cord tissue was minced and digested in a solution of Hibernate-A (Invitrogen, Carlsbad, California) with papain (Worthington, Lakewood, New Jersey). Neurons were isolated from the digested cord tissue using an Optiprep (Sigma-Aldrich) density gradient adapted from Brewer [18]. Neurons were plated in 1mL culture media of Neurobasal-A, B27, GlutaMAX (all obtained from Invitrogen), and penicillin or streptomycin (Gibco, New York, New York) on poly-D-Lysine (Sigma-Aldrich)–coated plates at approximately 300,000 cells/well. Cell were cultured in a 37 C, 5% CO2 humidified atmosphere. Half of the media was replaced every 2 days with fresh media containing AraC (Sigma-Aldrich) to prevent astrocyte growth. The cultures were confirmed to have >90% neurons as seen by morphology on light microscope as well as confirmed with microtubule-associated protein 2– positive neuronal staining. Cells were used at in vitro day 5 for experimentation, which is considered mature for neuronal cultures.
Lentiviral Knockdown of EPO bcR Lentiviral vectors containing short hairpin RNA gene silencing clones for the bcR subunit of the EPO receptor were obtained through the University of Colorado Denver Functional Genomics Core (University of Colorado Cancer Center, Aurora, CO). Similarly, lentiviral control vectors containing the same plasmid with a nonsense target sequence were obtained. Both the experimental and control plasmids contained a puromycin resistance gene for selection of transduced cells. We obtained 4 variations of a bcR knockdown virus and examined cellular viability after transduction with viability assays (as described subsequently). Additionally, we examined receptor expression levels after treatment to quantify the knockdown success. After identification of a knockdown virus with a high rate of receptor knockdown and minimal effect on viability was identified, this virus was used for all knockdown experiments. Mature spinal cord neuronal cultures were treated with control or bcR knockdown lentiviral vectors at a total concentration of 104 relative infectious units per milliliter for 16 hours with polybrene (2 mL of 1000X). Additionally,
Ann Thorac Surg 2017;104:1909–14
Oxygen-Glucose Deprivation and EPO Treatment In previous studies, the authors have shown that EPO treatment and bcR induction provided neuronal protection and prevented cell loss after ischemia in vivo and oxygen-glucose deprivation (OGD) in vitro [11, 19]. For this study, we further investigated that finding by subjecting neurons to OGD with EPO treatment in the absence of the bcR, by inhibiting its expression, to determine if this receptor was responsible for EPOmediated neuroprotection. On the day of the experiment, the experimental medium of Neurobasal-A without glucose (Invitrogen) was placed in a Ruskinn Bug Box Plus (The Baker Company, Sanford, Maine) humidified airtight hypoxic chamber for 2 hours. The Ruskinn Bug Box Plus was used per manufacturer’s protocol to maintain an environment of 95% N2 and 5% CO2 at 37 C. A hypoxic environment was verified with Anaerobic Indicator Strips (Oxoid Ltd, Basingstoke, United Kingdom) before placement of media in the chamber, and hypoxia of the media after 2 hours was tested with an indicator strip. The plates were then introduced into the hypoxic environment where puromycin selection culture medium was removed, cells were washed with phosphate-buffered saline, and 1 mL of hypoxic glucose-deprived medium was added to all cell culture wells. Neurons transduced with the bcR knockdown plasmid or the nonsense plasmid were subjected to 2 periods of OGD. The first period was a 30-minute exposure, which has previously been shown to be too short to induce significant injury, but enough hypoxia stimuli to induce bcR expression [5]. This first exposure was followed by 24 hours of suspension in normal cellular media to allow for expression of the bcR in control neurons. The neurons were then subjected to OGD for 1 hour, a dose previously shown to induce about 40% cell death in untreated spinal cord neurons. The hypoxic media used for these experiments contained EPO for the EPO treatment groups. It was at this time in the experiments that EPO treatment is introduced and continued throughout. After completion of OGD exposure, neurons were placed in culture media ( EPO) in a humidified, oxygenated culture chamber for 24 hours of reperfusion.
1911
At this time, final viability studies as well as receptor expression studies were performed.
Viability Studies For viability studies, neuronal cells were cultured on 24well plates. Cell viability was determined using an MTT Cell Viability Kit (Roche, Indianapolis, Indiana), as previously described by quantifiable chemiluminescence [5]. Absorbance was measured on a BioTek Synergy H1 Hybrid microplate reader (BioTek Instruments Inc, Winooski, Vermont). Cell viability is presented as the percentage absorbance relative to the control.
Western Blot Analysis Spinal cords were homogenized. Protein concentration in each sample was quantified with NanoDrop (Thermo Scientific, Wilmington, Delaware). A 4% to 20% gradient gel was used to separate proteins that were then transferred to a nitrocellulose membrane. After blocking in 5% milk in bovine serum albumin, the membranes were incubated at 4 C in 1:500 primary antibody (rabbit antimouse bcR). The membranes were then washed sequentially again and incubated in secondary anti-rabbit antibody at room temperature for 1 hour. Excess secondary antibody was washed off. Chemiluminescence was used to identify the target band and band density was quantified with ImageJ software (National Institutes of Health, Bethesda, Maryland). Quantification of receptor protein was compared with overall glyceradehyde-3phosphate expressed in each sample.
Statistics Sample size for studies was determined by power analysis with plans for a power of 0.90 to and 2-sided alpha set at 0.05. A sample size of 4 was determined adequate for the differences expected to result and these experiments were performed with a sample size of 8 to account for the variability in starting samples that the knockdown step introduced. Data are presented as mean SD. Analysis was performed using StatView Version 5.0 (SAS Institute Inc, Cary North Carolina). One-way analysis of variance or t test were performed, and a p value of <0.05 was considered significant for all statistical comparisons.
Results Spinal Cord Cell Cultures As previously described, using the spinal cord neuron isolation protocol, we were able to isolate and maintain primary spinal cord neuronal cultures [17, 18]. These cultures matured appropriately, as demonstrated by their network of projections (Fig 1). Neuron cultures were maintained at >90% neurons present, verified under light microscopy and corroborated by microtubule-associated protein 2–positive neuronal staining. These cells proved robust enough to withstand transduction (introduction of genetic material using a virus vector), puromycin selection, and hypoxic experiments.
ADULT CARDIAC
2 wells on each 24-well plate were treated with 100 mL of Dulbecco’s modified Eagle medium (Gibco) alone, which is the media in which the lentiviral vectors were suspended. These 2 wells were used as references to determine the time at which all nontransduced cells had been killed by the puromycin selection process. After 16 hours of lentiviral vector treatment, the supernatant containing the virus (or no virus for the 2 reference wells) was removed from all wells and replaced with fresh media of Neurobasal-A with B27, GlutaMAX, and puromycin (1 mg/mL). The plates were examined under light microscopy until all nontransduced cells (all of the 2 reference wells) were dead. This selection was reliably completed by 2 days, at which time the neurons were ready to undergo experimental treatments.
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
Ann Thorac Surg 2017;104:1909–14
ADULT CARDIAC
1912
Fig 1. Primary spinal cord neurons in culture. Primary spinal cord neurons are identified by a network of projections when cultured to maturity.
Spinal Cord Neuron Viability After Transduction The first part of our project focused on identifying a lentiviral vector (LV) with minimal effect on neuron viability while also having excellent knockdown capabilities. In these studies, the viability is presented as a percentage absorbance relative to the control nonsense lentiviral treatment. A sample size of 8 litters was used for all cell culture experiments, compared against the average of 2 control neurons. The data demonstrated that LV2 was the least harmful to the neurons with an overall survival rate of 74.6% 1.96% (p < 0.01). LV1 treatment resulted in a survival rate of 55.4% 5.1%. Both LV3 and LV4 treatments were poorly tolerated by the neurons, with low survival rates at 29.3% 14.7% and 26.8% 1.7%, respectively (Fig 2).
Lentiviral Knockdown Studies The bcR is characteristically expressed in very low quantities at baseline. It is induced by hypoxia and other
Fig 3. Beta common receptor expression after transduction with knockdown virus. LV1, LV2, and LV4 successfully suppressed beta common receptor expression. Based on these findings along with the previous viability findings, LV2 was chosen as the knockdown virus used for all further experiments. (LV ¼ lentiviral vector.)
cellular stress stimuli. For this reason, a brief OGD exposure was performed on all transduced cells to induce the receptor and identify if the knockdown virus was successful at blunting this induction. These data are presented as bcR expression relative to the control cells treated with the nonsense control lentivirus (Fig 3). Nearly all viruses successfully prevented expression of the bcR after hypoxia, with LV1, LV2, and LV4 causing less than 5% receptor expression compared with control viruses (p < 0.01). Significant amount of bcR expression was still present after transduction with LV3. The results of the neuron viability studies were combined with bcR knockdown studies and LV2 was used for all knockdown experiments for the remainder of the project.
Neuron Viability After OGD With EPO Treatment EPO significantly preserved neuronal viability after OGD treatment (mean 0.82 0.04 versus 0.61 0.01; p < 0.01) (Fig 4). Additionally, EPO-mediated neuronal preservation was similar in the nonsense virus and cells not treated with virus (mean 0.82 0.04 versus 0.80 0.05; p ¼ 0.77). EPO attenuation of neuronal injury was lost in bcR knockdown cells compared with nonsense control cells (mean 0.46 0.03 versus 0.80 0.05; p < 0.01).
Comment
Fig 2. Neuron viability after transduction. LV2 resulted in the least cell loss after transduction experiments. (LV ¼ lentiviral vector.)
Delayed paraplegia is a devastating and humbling complication of complex thoracoabdominal aortic operations, which still haunts even the most skilled aortic surgeons. Investigation into the mechanisms of innate neuroprotection from ischemia provide novel avenues of potential treatment for preventing this injury. Our lab has previously shown that EPO treatment prevents paraplegia in a murine model of spinal cord ischemia [11, 19]. Additionally, we have shown that neuroprotective pathways of EPO treatment occur more robustly after bcR induction [17]. The results of the present study show that
Ann Thorac Surg 2017;104:1909–14
the interleukin bcR is an essential mediator in the neuroprotective pathway activated by EPO. Identifying and characterizing this receptor’s role in EPO-mediated neuroprotection holds vast clinical implications. Thrombotic side effects of EPO, mediated by the homodimer receptor on platelets, have hindered its clinical use for treating spinal cord ischemia. The bcR activates pathways separate from those of the homodimer receptor. Specifically targeting the bcR provides a clinical avenue to maximize the antiapoptotic and tissue protective properties of EPO without the unwanted stroke and clot risk associated with its use [7]. Derivatives of EPO, such as carbamylated EPO, and other nonhematopoietic versions of this cytokine have been found to provide neuroprotection against stroke, spinal cord compression, and diabetic neuropathy without hematologic effects [14]. Use of a tissue protective, but not hematopoietic, derivative of EPO in aortic surgery is a promising direction for identifying pharmacologic treatments for spinal cord ischemia. This study provides groundwork for further investigation into the neuroprotective potential of EPO and its derivatives in spinal cord ischemia by establishing the bcR’s essential role in this pathway. There are notable limitations to this study to mention. This study was performed using murine spinal cord neurons, a nonhuman rudimentary model. The steps to translate these findings to clinical practice are many. However, this spinal cold cell culture model using OGD has repeatedly mimicked our findings in a mouse model of spinal cord ischemia. We elected to use the cell culture model to conduct EPO and bcR knockdown experiments
1913
and distill out the components of this mechanism. Future studies are planned to further examine these findings in animal models, using derivatives of EPO specific for this receptor. EPO’s emerging role as a neuroprotective therapy is being studied in a variety of different clinical trials. Its role in neonatal neuroprotection after perinatal ischemia is currently of active interest globally [20]. It was also recently studied in aortic surgery, though that clinical trial was interrupted due to concerns for stroke risk reported elsewhere in the literature, although not seen in that trial [21]. Preclinical studies are confirming promising findings in animal models of spinal cord injury [22, 23]. These cell culture experiments are important tools for breaking down the components of in vivo findings to identify molecular targets for therapy. In summary, this study has proven that bcR knockdown is possible in cell culture of murine spinal cord neurons. More importantly, we have demonstrated that knocking down the bcR abrogates EPO-mediated neuroprotection, highlighting its central role in this mechanism. In conclusion, spinal cord ischemia-reperfusion injury remains a devastating and potentially preventable complication. Further research into the innate mechanisms of spinal cord neuroprotection may elucidate new avenues for therapeutic advances. The authors wish to thank the University of Colorado Denver, Department of Surgery, for their financial support of this project.
References 1. Coselli JS, Lemaire SA, Conklin LD, K€ oksoy C, Schmittling ZC. Morbidity and mortality after extent II thoracoabdominal aortic aneurysm repair. Ann Thorac Surg 2002;73:1107–15. 2. Coselli J, Lemaire SA, Preventza O, et al. Outcomes of 3309 thoracoabdominal aortic aneurysm repairs. J Thorac Cardiovasc Surg 2016;151:1323–38. 3. Conrad MF, Ergul EA, Patel VI, et al. Evolution of operative strategies in open thoracoabdominal aneurysm repair. J Vasc Surg 2011;53:1195–201. 4. Conrad MF, Ye JY, Chung TK, Davison JK, Cambria RP. Spinal cord complications after thoracic aortic surgery: Longterm survival and functional status varies with deficit severity. J Vasc Surg 2008;48:47–53. 5. Foley LS, Reece TB. Advances in spinal cord protection for complex aortic repairs. J Thorac Cardiovasc Surg 2016;151: 614–5. 6. Onose G, Anghelescu A, Muresanu DF, et al. A review of published reports on neuroprotection in spinal cord injury. Spinal Cord 2009;47:716–26. 7. Brines M, Cerami A. Erythropoietin-mediated tissue protection: reducing collateral damage from the primary injury response. J Intern Med 2008;264:405–32. 8. Stirling DP, Yong VW. Dynamics of the inflammatory response after murine spinal cord injury revealed by flow cytometry. J Neurosci Res 2008;86:1944–58. 9. Smith PD, Puskas F, Meng X, et al. The evolution of chemokine release supports a bimodal mechanism of spinal cord ischemia and reperfusion injury. Circulation 2012;126(Suppl 1):110–7.
ADULT CARDIAC
Fig 4. Spinal cord neuron viability after oxygen-glucose deprivation (OGD). Erythropoietin (EPO) significantly preserved neuronal viability after OGD treatment. EPO-mediated attenuation of neuronal injury was lost in beta common receptor (bcR) knockdown cells compared with nonsense control cells. (*Clinical significance, p < 0.05.)
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
ADULT CARDIAC
1914
FOLEY ET AL NEUROPROTECTION IN SPINAL CORD NEURONS
10. Smith PD, Puskas F, Meng X, et al. Ischemic dose-response in the spinal cord: both immediate and delayed paraplegia. J Surg Res 2012;174:238–44. 11. Smith PD, Puskas F, Fullerton DA, et al. Attenuation of spinal cord ischemia and reperfusion injury by erythropoietin. J Thorac Cardiovasc Surg 2011;141:256–60. 12. Hirano K, Wagner K, Mark P, et al. Erythropoietin attenuates the sequels of ischaemic spinal cord injury with enhanced recruitment of CD34þ cells in mice. J Cell Mol Med 2012;16:1792–802. 13. Beattie MS. Inflammation and apoptosis: linked therapeutic targets in spinal cord injury. Trends Mol Med 2004;10:580–3. 14. Leist M, Ghezzi P, Grasso G, et al. Derivative of erythropoietin that are tissue protective but not erythropoietin. Science 2004;305:239–42. 15. Brines ML, Ghezzi P, Keenan S, et al. Erythropoietin crosses the blood-brain barrier to protect against experimental brain injury. Proc Natl Acad Sci U S A 2000;97:10526–31. 16. Brines M, Cerami A. The receptor that tames the innate immune response. Mol Med 2012;18:486–96. 17. Foley LS, Fullerton DA, Bennett DT, et al. Spinal cord ischemia-reperfusion injury induces erythropoietin receptor expression. Ann Thorac Surg 2015;100:41–6.
Ann Thorac Surg 2017;104:1909–14
18. Brewer G, Torricelli J. Isolation and culture of adult neurons and neurospheres. Nat Protoc 2007;2:1490–8. 19. Mares JM, Foley LS, Bell M, et al. Erythropoietin activates the phosporylated cAMP [adenosine 30 50 cyclic monophosphate] response element-binding protein pathway and attenuates delayed paraplegia after ischemia-reperfusion injury. J Thorac Cardiovasc Surg 2015;149:920–4. 20. Rangarajan V, Juul SE. Erythropoietin: emerging role of erythropoietin in neonatal neuroprotection. Pediatr Neurol 2014;51:481–8. 21. Messe SR, McGarvey ML, Bavaria JE, et al. A pilot study of darbepoetin alfa for prophylactic neuroprotection in aortic surgery. Neurocrit Care 2013;18:75–80. 22. Yang L, Yan X, Xu Z, Tan W, Chen Z, Wu B. Delayed administration of recombinant human erythropoietin reduces apoptosis and inflammation and promotes myelin repair and functional recovery following spinal cord compressive injury in rats. Restor Neurol Neurosci 2015;34: 647–63. 23. Yilmaz ER, Kermen H, Dolgun H, et al. Effects of darbepoetin-a in spinal cord ischemia-reperfusion injury in the rabbit. Acta Neurochir 2012;154:1037–44.
DISCUSSION DR SCOTT A. LEMAIRE (Houston, TX): That was very nicely presented. I would like to start by congratulating Dr Foley, Dr Reece, and their colleagues on a very elegant study to identify the critical role of interleukin beta common receptor (bcR) in erythropoietin’s protective role in ischemic neurons. The first question I have is about your endpoint. Your primary endpoint in this experiment was neuron viability. Did you assess any other endpoints, such as caspase 3, which is a marker for neuron apoptosis, or any downstream signaling pathways related to heterodimer receptor activation? DR FOLEY: We did in previous studies and we plan to in future studies. In previous studies using neuron viability we looked at markers of neuron cell death, we specifically looked at caspase, actually, but we did not in this study yet and we plan to. These are all n ¼ 8 and this is an exceptional amount of mice that went into this. DR LEMAIRE: My next question is about the other cells within the spinal cord. So astrocytes and microglia are also involved in erythropoietin’s modulation of neuroinflammation. Do you believe that the effects of erythropoietin on these cell types are also dependent on the heterodimer receptor?
DR FOLEY: In the last study that we presented actually at the STSA a couple of years ago, we demonstrated the bcR is upregulated in astrocytes as well, and we want to go further with it. This is our first thing using the lentiviral knockdowns, but, yes, I do think that it acts at least on the astrocytes, we know that for sure, but we have not done microglia yet. DR LEMAIRE: My last question is related to, you have got the in vitro model showing that this receptor is important and now about validation in the in vivo models. Other investigators have used bcR knockout mice to study neuropathic pain and other nerve diseases. Are you planning to expand your investigation by determining whether bcR-null mice remain susceptible to spinal cord ischemia despite treatment with erythropoietin or perhaps whether neurons from these animals would exhibit similar levels of apoptosis when subjected to oxygen and glucose depletion with and without erythropoietin? DR FOLEY: That is a very good question. We have actually been looking into using an animal knockdown, and we are using a neighboring lab to try to get that accomplished.