ARTICLE IN PRESS
Translating mechanisms of neuroprotection, regeneration, and repair to treatment of spinal cord injury Ahad M. Siddiqui*, Mohamad Khazaei*, Michael G. Fehlings*,†,{,1 *Department of Genetics and Development, Toronto Western Research Institute, University Health Network, Toronto, Ontario, Canada † Department of Surgery, University of Toronto, Toronto, Ontario, Canada { Institute of Medical Sciences, University of Toronto, Toronto, Ontario, Canada 1 Corresponding author: Tel.: +1-416-603-5229; Fax: +1-416-603-5745, e-mail address: [email protected]
Abstract One of the big challenges in neuroscience that remains to be understood is why the central nervous system is not able to regenerate to the extent that the peripheral nervous system does. This is especially problematic after traumatic injuries, like spinal cord injury (SCI), since the lack of regeneration leads to lifelong deficits and paralysis. Treatment of SCI has improved during the last several decades due to standardized protocols for emergency medical response teams and improved medical, surgical, and rehabilitative treatments. However, SCI continues to result in profound impairments for the individual. There are many processes that lead to the pathophysiology of SCI, such as ischemia, vascular disruption, neuroinflammation, oxidative stress, excitotoxicity, demyelination, and cell death. Current treatments include surgical decompression, hemodynamic control, and methylprednisolone. However, these early treatments are associated with modest functional recovery. Some treatments currently being investigated for use in SCI target neuroprotective (riluzole, minocycline, G-CSF, FGF-2, and polyethylene glycol) or neuroregenerative (chondroitinase ABC, self-assembling peptides, and rho inhibition) strategies, while many cell therapies (embryonic stem cells, neural stem cells, induced pluripotent stem cells, mesenchymal stromal cells, Schwann cells, olfactory ensheathing cells, and macrophages) have also shown promise. However, since SCI has multiple factors that determine the progress of the injury, a combinatorial therapeutic approach will most likely be required for the most effective treatment of SCI.
Progress in Brain Research, ISSN 0079-6123, http://dx.doi.org/10.1016/bs.pbr.2014.12.007 © 2015 Elsevier B.V. All rights reserved.
15
ARTICLE IN PRESS 16
Mechanisms of neuroprotection, regeneration, and repair in SCI
Keywords cell therapy, clinical translation, neuroprotection, neuroregeneration, spinal cord injury
1 INTRODUCTION 1.1 EPIDEMIOLOGY OF SPINAL CORD INJURY Spinal cord injury (SCI) refers to traumatic injury to the spinal cord that is not the result of disease. In recent years, advances in health care and rehabilitation have resulted in greater survival after injury. However, it is still the case that between one and two thirds of patients die on the way to the hospital (Bydon et al., 2014; Sekhon and Fehlings, 2001; Tator et al., 1993). The reported incidence of SCI ranges from 9.2 to 246 cases per million of the population a year depending on the area surveyed (Furlan et al., 2013). The incidence of SCI is highest among people in their late teens to early 20s and the elderly (Carroll, 1997; DeVivo, 2012; Hagen et al., 2010). However, the mean age has increased from 28.3 years in the 1970s to 37.1 years in 2005–2008 and it is expected to rise further as the population ages (DeVivo and Chen, 2011). The global prevalence ranges from 236 to 1298 per million of the population with the rate increasing over the last 30 years (Furlan et al., 2013). SCI occurs three to four times more often among males than females, however, the proportion of females is rising as the population ages (Putzke et al., 2003; DeVivo, 2012; Sekhon and Fehlings, 2001). Approximately half of SCI cases in the United States occur due to motor vehicle crashes (Putzke et al., 2003; DeVivo and Chen, 2011; Price et al., 1994). Other causes include violence (12%), sports (10%), and trips/falls (DeVivo, 2010). Among the elderly, falls are the leading cause of SCI and the incidence of this has been increasing as the population ages (Acton et al., 1993; DeVivo, 2012). Over half of SCIs occur at the cervical level of the spinal cord (Burney et al., 1993; DeVivo, 2010; Sekhon and Fehlings, 2001). It important to understand the mechanisms involved after SCI to develop better treatments that will help to improve the survival rate and quality of life of patients after injury.
1.2 PATHOPHYSIOLOGY OF SPINAL CORD INJURY The pathophysiology of SCI is a biphasic process that consists of a primary phase that involves the initial mechanical injury followed by a delayed secondary phase that involves processes such as vascular disruption, inflammation, and excitotoxicity.
1.2.1 Primary Phase The primary phase of injury is mainly due to the spinal column exerting force on the spinal cord resulting in disruption of axons (Rowland et al., 2008). This is most commonly the result of a compressive/contusive injury that causes shearing, laceration, or acute stretching (Baptiste and Fehlings, 2006; Dasari et al., 2014; Sekhon and Fehlings, 2001). Injuries that fully transect the spinal cord are rare and usually some
ARTICLE IN PRESS 1 Introduction
connections are spared (Rowland et al., 2008). These spared but demyelinated axons are most commonly found at the subpial rim (McDonald and Belegu, 2006; Nashmi and Fehlings, 2001; Radojicic et al., 2005). This is therapeutically important since animal studies have shown significant neurological recovery with as little as 10% of the original axons being preserved (Fehlings and Tator, 1995; Kakulas, 2004). There is much interest in therapies that optimize recovery using existing connections.
1.2.2 Secondary Phase The secondary phase of injury is characterized by ischemia, excitotoxicity, vascular dysfunction, oxidative stress, and inflammation that leads to cell death (Braughler et al., 1985; Rowland et al., 2008; Wagner and Stewart, 1981). The processes in the secondary injury are often harmful to surviving bystander neurons and the injury of these neurons can lead to poor functional recovery (McDonald and Sadowsky, 2002; Vawda and Fehlings, 2013). In addition, it is during the secondary phase that an inhibitory environment is created that impairs endogenous regeneration and remyelination (Dasari et al., 2014). The secondary phase is made up of subphases that are divided temporally into the immediate, acute, subacute, intermediate, and chronic stages of SCI (Fig. 1).
FIGURE 1 Timeline summarizing the phases after spinal cord injury and the therapeutic aims best suited for that phase. The events that occur after spinal cord injury are divided into the immediate (first 2 h), acute (2–48 h), subacute (48 h–14 days), intermediate (14 days–6 months), and chronic (6 months and beyond) phases. These phases are characterized by changes in inflammation, hemorrhage, apoptosis, the blood-spinal cord barrier (BSCB), and the extracellular matrix. Some therapeutic aims are shown to be beneficial in certain phases of SCI since they target the events that occur in that phase.
17
ARTICLE IN PRESS 18
Mechanisms of neuroprotection, regeneration, and repair in SCI
1.2.2.1 Immediate Phase The immediate phase lasts for approximately the first 2 h of injury and constitutes the immediate aftermath of the injury (Norenberg et al., 2004). The rapid death of neurons and glia accompanies spinal shock that results in the immediate loss of function at and below level of the injury (Boland et al., 2011; Ditunno et al., 2004). The first sign of the immediate phase is the necrotic cell death of neurons due to ischemia, hemorrhaging, edema, and mechanical disruption of the cell membrane (Kakulas, 2004; Tator et al., 1993). Early in the immediate phase there is upregulation of TNF-a and IL-b (Davalos et al., 2005; David and Kroner, 2011; Donnelly and Popovich, 2008; Pineau and Lacroix, 2007). Another early event that occurs within minutes of SCI is the rise of extracellular glutamate to excitotoxic levels (Park et al., 2004; Wrathall et al., 1996).
1.2.2.2 Acute Phase The immediate phase is not generally considered as a target for treatment as it is too early, clinically speaking, for treatment to be realistically administered. Due to this, it is thought that the acute phase is a better target for neuroprotective interventions since this may be the earliest that a patient arrives to hospital. The acute phase can be divided into the early acute and subacute stages. The early acute phase occurs between 2 and 48 h after injury. A hallmark of the secondary injury in the acute phase is the vascular disruption and hemorrhage that result in ischemia (Tator and Fehlings, 1991; Tator and Koyanagi, 1997). Although the vascular mechanism that leads to the prolonged ischemia is not fully understood, it is thought that disruption of the microvascular, hypotension, and increased interstitial pressure leads to hypoperfusion of the cord after injury (Kwon et al., 2004; Mautes et al., 2000; Ng et al., 2011; Tator and Fehlings, 1991). The process of hemorrhage and ischemia is closely related to the permeability of the blood–brain-barrier (BBB)/blood-spinal cord barrier (BSCB). SCI results in the permeability of the BBB/BSCB due to the direct mechanical disruption of the vasculature and the effect of inflammatory mediators on endothelial cells (Rowland et al., 2008; Zhang et al., 2012). BSCB permeability in rats reaches its peak 24 h after contusive/clip compression SCI and returns to control levels around 2 weeks after injury (Figley et al., 2014; Noble and Wrathall, 1989). BSCB permeability may be affected by inflammatory cytokines that are commonly upregulated by SCI (Pardridge, 2010; Pineau and Lacroix, 2007; Schnell et al., 1999). Although permeability of the BBB/BSCB after SCI is seen as a deleterious event, the permeability may provide an opportunity to introduce cell treatments and drugs that normally may not be able to cross the BBB/BSCB. The leakiness of the BBB/BSCB permits the infiltration of immune cells, such as T cells, neutrophils, and monocytes, into the CNS. The resident microglia continue to proliferate and become activated into the subacute stage. Microglia attract peripheral leukocytes and other immune cells through production of cytokines that upregulate production of chemokines (Donnelly and Popovich, 2008; Mueller et al., 2006; Tzekou and Fehlings, 2014). Within 24 h of the injury, neutrophils reach the lesion where they produce cytokines, MMPs, superoxide dismutase, and myeloperoxidase
ARTICLE IN PRESS 1 Introduction
(Donnelly and Popovich, 2008; Fleming et al., 2006; Guth et al., 1999; Noble et al., 2002; Taoka et al., 1997). Neutrophils act to aid in leukocyte chemotaxis and extravasation, as well as the activation of glia and mediation of the respiratory burst which may be harmful to neurons (Carlson et al., 1998; Tzekou and Fehlings, 2014). The monocytes begin to be recruited as the levels of neutrophils stop increasing at 48 h (Guth et al., 1999; Taoka et al., 1997). At about 72 h after injury, the monocytes begin to differentiate into macrophages and produce glutamate, TNF-a, IL-1 and IL-6, and prostanoids which may exacerbate secondary injury (Leskovar et al., 2000; Schwab et al., 2000). The number of macrophages begins to decrease 7 days after injury, but the activation of microglia can persist for weeks after injury (Donnelly and Popovich, 2008). The neuroinflammatory response after SCI has been shown to be a double-edged sword where activation of certain immune cells and inflammatory cytokines has been shown to have both beneficial and detrimental roles. Microglia and macrophages have also shown to have beneficial and detrimental effects after CNS injury, partly due to the fact that they may have proinflammatory (M1) and neuroprotective (M2) activation states (David and Kroner, 2011; Kigerl et al., 2009; Kobayashi et al., 2013). Ischemia and immune infiltration can lead to the production of oxidative stress and free radical production. Reactive oxygen (ROS) and nitrogen (NO) species can be produced by macrophages/microglia after SCI or even as a result of ischemia and reperfusion (Chatzipanteli et al., 2002; Sakamoto et al., 1991). The levels of ROS peak 12 h after injury and remain elevated for 1 week (Donnelly and Popovich, 2008). Inhibition of NO production has been shown to have neuroprotective effects after CNS injury (Chatzipanteli et al., 2002; Koeberle and Ball, 1999; Lo´pez-Vales et al., 2004; Pearse et al., 2003). One of the key facilitators of ROS-induced injury is the formation of peroxynitrite radical generated from the reaction between nitric oxide and superoxide which is involved in the initiation of neuronal apoptosis after experimental SCI (Bao and Liu, 2003; Xiong et al., 2007). Ionic dysregulation and excitotoxicity immediately follow SCI and contribute to the cellular damage and loss. The proper regulation of calcium is an important process in preventing cell death, and its dysregulation leads to cell death through mitochondrial dysfunction, production of free radicals, and activation of calpains (Schanne et al., 1979; Vosler et al., 2009). Furthermore, extracellular levels of glutamate rise after injury as a direct consequence of disruption of membrane transporters which maintain homeostasis of ions and glutamate (Llado´ et al., 2004). This results in overactivation of the glutamate receptor leading to an increased influx of sodium and calcium ions through the NMDA and AMPA receptors, dysregulation of metabolic and mitochondrial activity, and loss of osmotic balance that ultimately results in excitotoxic cell death (Agrawal and Fehlings, 1997; Gerardo-Nava et al., 2013; Park et al., 2004; Wang et al., 2012). Due to the role of excitotoxicity in SCI, there has been great interest in using drugs to control it through antagonism of NMDA and other receptors. The ultimate consequences of the processes described earlier during the acute phase are cell death and demyelination. The majority of neuronal cell death after
19
ARTICLE IN PRESS 20
Mechanisms of neuroprotection, regeneration, and repair in SCI
SCI occurs through necrosis, although apoptosis also plays an important role (Beattie et al., 2002; Keane et al., 2001; Yu et al., 2009). However, oligodendrocyte cell death occurs through apoptosis, partially dependent on activation of the Fas receptor and through p75 receptor signaling (Ackery et al., 2006; Casha et al., 2001, 2005; Chu et al., 2007; Crowe et al., 1997; Liu et al., 2009). Most of the Fas receptors in the spinal cord are found on oligodendrocytes and expressed by activated microglia and lymphocytes (Austin and Fehlings, 2008; Casha et al., 2001, 2005). The interaction between the two leads to apoptosis through the activation of caspases (Austin and Fehlings, 2008). There have been studies showing that blocking Fas-mediated cell death can lead to functional recovery and be used in the treatment of SCI (Ackery et al., 2006; Robins-Steele et al., 2012; Yu et al., 2009). The subacute phase lasts from approximately 2 days to 2 weeks after SCI. Astrocytes initially go through necrotic cell death but, in the subacute phase, become hypertrophic and proliferative (Rowland et al., 2008). The large cytoplasmic processes come together to become the gliotic scar that forms a physical and chemical barrier to regeneration (Fawcett and Asher, 1999; Hagg and Oudega, 2006; KarimiAbdolrezaee et al., 2010, 2012; Reier and Houle, 1988; Young, 2014). The cells in scar tissue release inhibitory molecules, such as chondroitin sulfate proteoglycans (CSPGs) (Fawcett and Asher, 1999; Fitch and Silver, 2008). To combat the inhibition to regeneration from the glial scar, there is interest in developing therapeutics that can remove the glial scar to promote regeneration.
1.2.2.3 Intermediate Phase The immediate phase begins approximately 2–3 weeks after injury and continues to 6 months after injury. During this phase, reactive gliosis continues as the scar begins to mature. There is also axonal sprouting of the corticospinal tract and the reticulospinal fibers (Hill et al., 2001) during this stage. Although this endogenous attempt at sprouting axons does not translate to significant functional recovery, it presents an attractive target for therapeutic intervention.
1.2.2.4 Chronic Phase The last phase in SCI is the chronic phase which begins at around 6 months postinjury and lasts for the lifetime of the patient. During the chronic phase, the lesion begins to stabilize with scar formation and cyst/syrinx development (Li and Lepski, 2013; Rowland et al., 2008). Cysts arise due to the clearance of debris by microglia and macrophages due to progressive loss of neural tissue (Basso et al., 1996; Fleming et al., 2006; Norenberg et al., 2004). Usually some axons are spared at the rim of the cysts but the cysts present a physical barrier to neuronal regeneration (Kramer et al., 2013). In addition, Wallerian degeneration of the axons continues and years for the cell bodies and axons to be removed (Beattie et al., 2002; Ehlers, 2004). Many of the therapeutic strategies used during the chronic stage aim to promote regeneration, promote plasticity, or to improve function of spared axons.
ARTICLE IN PRESS 2 Clinical intervention
2 CLINICAL INTERVENTION 2.1 CURRENT PRACTICE Historically, patients with SCI had a poor prognosis and were left unmonitored in hospital wards until their vertebrae healed. However, modern medical advances have led to improvements in patient care. The use of spineboards and the practice of immobilizing patients at the site of injury have acted to reduce injury and mortality (Anon, 2002a). In addition, surgical and pharmacological interventions are becoming more common.
2.2 SURGICAL DECOMPRESSION Surgical decompression helps to restore spinal stability and maintains cord perfusion (Mothe and Tator, 2013; Wilson and Fehlings, 2011). Spinal decompression surgery includes various procedures intended to relieve symptoms caused by pressure, or compression, on the spinal cord. There is controversy about the role and timing of surgical decompression after an acute SCI (Fehlings et al., 2001). Numerous experimental studies of decompression after SCI have been performed in various animal models including primates, dogs, cats, and rodents. These studies have consistently shown that neurological recovery is enhanced by early decompression (Fehlings and Perrin, 2005). To further address the questions regarding the optimal timing for performing decompression, our center was part of the “Surgical Timing in Acute Spinal Cord Injury Study” (STASCIS) on a multicenter, international, prospective cohort of patients. The results demonstrate a more favorable neurologic recovery among those treated with early (defined as
Translating mechanisms of neuroprotection, regeneration, and repair to treatment of spinal cord injury Ahad M. Siddiqui*, Mohamad Khazaei*, Michael G. Fehlings*,†,{,1 *Department of Genetics and Development, Toronto Western Research Institute, University Health Network, Toronto, Ontario, Canada † Department of Surgery, University of Toronto, Toronto, Ontario, Canada { Institute of Medical Sciences, University of Toronto, Toronto, Ontario, Canada 1 Corresponding author: Tel.: +1-416-603-5229; Fax: +1-416-603-5745, e-mail address: [email protected]
Abstract One of the big challenges in neuroscience that remains to be understood is why the central nervous system is not able to regenerate to the extent that the peripheral nervous system does. This is especially problematic after traumatic injuries, like spinal cord injury (SCI), since the lack of regeneration leads to lifelong deficits and paralysis. Treatment of SCI has improved during the last several decades due to standardized protocols for emergency medical response teams and improved medical, surgical, and rehabilitative treatments. However, SCI continues to result in profound impairments for the individual. There are many processes that lead to the pathophysiology of SCI, such as ischemia, vascular disruption, neuroinflammation, oxidative stress, excitotoxicity, demyelination, and cell death. Current treatments include surgical decompression, hemodynamic control, and methylprednisolone. However, these early treatments are associated with modest functional recovery. Some treatments currently being investigated for use in SCI target neuroprotective (riluzole, minocycline, G-CSF, FGF-2, and polyethylene glycol) or neuroregenerative (chondroitinase ABC, self-assembling peptides, and rho inhibition) strategies, while many cell therapies (embryonic stem cells, neural stem cells, induced pluripotent stem cells, mesenchymal stromal cells, Schwann cells, olfactory ensheathing cells, and macrophages) have also shown promise. However, since SCI has multiple factors that determine the progress of the injury, a combinatorial therapeutic approach will most likely be required for the most effective treatment of SCI.
Progress in Brain Research, ISSN 0079-6123, http://dx.doi.org/10.1016/bs.pbr.2014.12.007 © 2015 Elsevier B.V. All rights reserved.
15
ARTICLE IN PRESS 16
Mechanisms of neuroprotection, regeneration, and repair in SCI
Keywords cell therapy, clinical translation, neuroprotection, neuroregeneration, spinal cord injury
1 INTRODUCTION 1.1 EPIDEMIOLOGY OF SPINAL CORD INJURY Spinal cord injury (SCI) refers to traumatic injury to the spinal cord that is not the result of disease. In recent years, advances in health care and rehabilitation have resulted in greater survival after injury. However, it is still the case that between one and two thirds of patients die on the way to the hospital (Bydon et al., 2014; Sekhon and Fehlings, 2001; Tator et al., 1993). The reported incidence of SCI ranges from 9.2 to 246 cases per million of the population a year depending on the area surveyed (Furlan et al., 2013). The incidence of SCI is highest among people in their late teens to early 20s and the elderly (Carroll, 1997; DeVivo, 2012; Hagen et al., 2010). However, the mean age has increased from 28.3 years in the 1970s to 37.1 years in 2005–2008 and it is expected to rise further as the population ages (DeVivo and Chen, 2011). The global prevalence ranges from 236 to 1298 per million of the population with the rate increasing over the last 30 years (Furlan et al., 2013). SCI occurs three to four times more often among males than females, however, the proportion of females is rising as the population ages (Putzke et al., 2003; DeVivo, 2012; Sekhon and Fehlings, 2001). Approximately half of SCI cases in the United States occur due to motor vehicle crashes (Putzke et al., 2003; DeVivo and Chen, 2011; Price et al., 1994). Other causes include violence (12%), sports (10%), and trips/falls (DeVivo, 2010). Among the elderly, falls are the leading cause of SCI and the incidence of this has been increasing as the population ages (Acton et al., 1993; DeVivo, 2012). Over half of SCIs occur at the cervical level of the spinal cord (Burney et al., 1993; DeVivo, 2010; Sekhon and Fehlings, 2001). It important to understand the mechanisms involved after SCI to develop better treatments that will help to improve the survival rate and quality of life of patients after injury.
1.2 PATHOPHYSIOLOGY OF SPINAL CORD INJURY The pathophysiology of SCI is a biphasic process that consists of a primary phase that involves the initial mechanical injury followed by a delayed secondary phase that involves processes such as vascular disruption, inflammation, and excitotoxicity.
1.2.1 Primary Phase The primary phase of injury is mainly due to the spinal column exerting force on the spinal cord resulting in disruption of axons (Rowland et al., 2008). This is most commonly the result of a compressive/contusive injury that causes shearing, laceration, or acute stretching (Baptiste and Fehlings, 2006; Dasari et al., 2014; Sekhon and Fehlings, 2001). Injuries that fully transect the spinal cord are rare and usually some
ARTICLE IN PRESS 1 Introduction
connections are spared (Rowland et al., 2008). These spared but demyelinated axons are most commonly found at the subpial rim (McDonald and Belegu, 2006; Nashmi and Fehlings, 2001; Radojicic et al., 2005). This is therapeutically important since animal studies have shown significant neurological recovery with as little as 10% of the original axons being preserved (Fehlings and Tator, 1995; Kakulas, 2004). There is much interest in therapies that optimize recovery using existing connections.
1.2.2 Secondary Phase The secondary phase of injury is characterized by ischemia, excitotoxicity, vascular dysfunction, oxidative stress, and inflammation that leads to cell death (Braughler et al., 1985; Rowland et al., 2008; Wagner and Stewart, 1981). The processes in the secondary injury are often harmful to surviving bystander neurons and the injury of these neurons can lead to poor functional recovery (McDonald and Sadowsky, 2002; Vawda and Fehlings, 2013). In addition, it is during the secondary phase that an inhibitory environment is created that impairs endogenous regeneration and remyelination (Dasari et al., 2014). The secondary phase is made up of subphases that are divided temporally into the immediate, acute, subacute, intermediate, and chronic stages of SCI (Fig. 1).
FIGURE 1 Timeline summarizing the phases after spinal cord injury and the therapeutic aims best suited for that phase. The events that occur after spinal cord injury are divided into the immediate (first 2 h), acute (2–48 h), subacute (48 h–14 days), intermediate (14 days–6 months), and chronic (6 months and beyond) phases. These phases are characterized by changes in inflammation, hemorrhage, apoptosis, the blood-spinal cord barrier (BSCB), and the extracellular matrix. Some therapeutic aims are shown to be beneficial in certain phases of SCI since they target the events that occur in that phase.
17
ARTICLE IN PRESS 18
Mechanisms of neuroprotection, regeneration, and repair in SCI
1.2.2.1 Immediate Phase The immediate phase lasts for approximately the first 2 h of injury and constitutes the immediate aftermath of the injury (Norenberg et al., 2004). The rapid death of neurons and glia accompanies spinal shock that results in the immediate loss of function at and below level of the injury (Boland et al., 2011; Ditunno et al., 2004). The first sign of the immediate phase is the necrotic cell death of neurons due to ischemia, hemorrhaging, edema, and mechanical disruption of the cell membrane (Kakulas, 2004; Tator et al., 1993). Early in the immediate phase there is upregulation of TNF-a and IL-b (Davalos et al., 2005; David and Kroner, 2011; Donnelly and Popovich, 2008; Pineau and Lacroix, 2007). Another early event that occurs within minutes of SCI is the rise of extracellular glutamate to excitotoxic levels (Park et al., 2004; Wrathall et al., 1996).
1.2.2.2 Acute Phase The immediate phase is not generally considered as a target for treatment as it is too early, clinically speaking, for treatment to be realistically administered. Due to this, it is thought that the acute phase is a better target for neuroprotective interventions since this may be the earliest that a patient arrives to hospital. The acute phase can be divided into the early acute and subacute stages. The early acute phase occurs between 2 and 48 h after injury. A hallmark of the secondary injury in the acute phase is the vascular disruption and hemorrhage that result in ischemia (Tator and Fehlings, 1991; Tator and Koyanagi, 1997). Although the vascular mechanism that leads to the prolonged ischemia is not fully understood, it is thought that disruption of the microvascular, hypotension, and increased interstitial pressure leads to hypoperfusion of the cord after injury (Kwon et al., 2004; Mautes et al., 2000; Ng et al., 2011; Tator and Fehlings, 1991). The process of hemorrhage and ischemia is closely related to the permeability of the blood–brain-barrier (BBB)/blood-spinal cord barrier (BSCB). SCI results in the permeability of the BBB/BSCB due to the direct mechanical disruption of the vasculature and the effect of inflammatory mediators on endothelial cells (Rowland et al., 2008; Zhang et al., 2012). BSCB permeability in rats reaches its peak 24 h after contusive/clip compression SCI and returns to control levels around 2 weeks after injury (Figley et al., 2014; Noble and Wrathall, 1989). BSCB permeability may be affected by inflammatory cytokines that are commonly upregulated by SCI (Pardridge, 2010; Pineau and Lacroix, 2007; Schnell et al., 1999). Although permeability of the BBB/BSCB after SCI is seen as a deleterious event, the permeability may provide an opportunity to introduce cell treatments and drugs that normally may not be able to cross the BBB/BSCB. The leakiness of the BBB/BSCB permits the infiltration of immune cells, such as T cells, neutrophils, and monocytes, into the CNS. The resident microglia continue to proliferate and become activated into the subacute stage. Microglia attract peripheral leukocytes and other immune cells through production of cytokines that upregulate production of chemokines (Donnelly and Popovich, 2008; Mueller et al., 2006; Tzekou and Fehlings, 2014). Within 24 h of the injury, neutrophils reach the lesion where they produce cytokines, MMPs, superoxide dismutase, and myeloperoxidase
ARTICLE IN PRESS 1 Introduction
(Donnelly and Popovich, 2008; Fleming et al., 2006; Guth et al., 1999; Noble et al., 2002; Taoka et al., 1997). Neutrophils act to aid in leukocyte chemotaxis and extravasation, as well as the activation of glia and mediation of the respiratory burst which may be harmful to neurons (Carlson et al., 1998; Tzekou and Fehlings, 2014). The monocytes begin to be recruited as the levels of neutrophils stop increasing at 48 h (Guth et al., 1999; Taoka et al., 1997). At about 72 h after injury, the monocytes begin to differentiate into macrophages and produce glutamate, TNF-a, IL-1 and IL-6, and prostanoids which may exacerbate secondary injury (Leskovar et al., 2000; Schwab et al., 2000). The number of macrophages begins to decrease 7 days after injury, but the activation of microglia can persist for weeks after injury (Donnelly and Popovich, 2008). The neuroinflammatory response after SCI has been shown to be a double-edged sword where activation of certain immune cells and inflammatory cytokines has been shown to have both beneficial and detrimental roles. Microglia and macrophages have also shown to have beneficial and detrimental effects after CNS injury, partly due to the fact that they may have proinflammatory (M1) and neuroprotective (M2) activation states (David and Kroner, 2011; Kigerl et al., 2009; Kobayashi et al., 2013). Ischemia and immune infiltration can lead to the production of oxidative stress and free radical production. Reactive oxygen (ROS) and nitrogen (NO) species can be produced by macrophages/microglia after SCI or even as a result of ischemia and reperfusion (Chatzipanteli et al., 2002; Sakamoto et al., 1991). The levels of ROS peak 12 h after injury and remain elevated for 1 week (Donnelly and Popovich, 2008). Inhibition of NO production has been shown to have neuroprotective effects after CNS injury (Chatzipanteli et al., 2002; Koeberle and Ball, 1999; Lo´pez-Vales et al., 2004; Pearse et al., 2003). One of the key facilitators of ROS-induced injury is the formation of peroxynitrite radical generated from the reaction between nitric oxide and superoxide which is involved in the initiation of neuronal apoptosis after experimental SCI (Bao and Liu, 2003; Xiong et al., 2007). Ionic dysregulation and excitotoxicity immediately follow SCI and contribute to the cellular damage and loss. The proper regulation of calcium is an important process in preventing cell death, and its dysregulation leads to cell death through mitochondrial dysfunction, production of free radicals, and activation of calpains (Schanne et al., 1979; Vosler et al., 2009). Furthermore, extracellular levels of glutamate rise after injury as a direct consequence of disruption of membrane transporters which maintain homeostasis of ions and glutamate (Llado´ et al., 2004). This results in overactivation of the glutamate receptor leading to an increased influx of sodium and calcium ions through the NMDA and AMPA receptors, dysregulation of metabolic and mitochondrial activity, and loss of osmotic balance that ultimately results in excitotoxic cell death (Agrawal and Fehlings, 1997; Gerardo-Nava et al., 2013; Park et al., 2004; Wang et al., 2012). Due to the role of excitotoxicity in SCI, there has been great interest in using drugs to control it through antagonism of NMDA and other receptors. The ultimate consequences of the processes described earlier during the acute phase are cell death and demyelination. The majority of neuronal cell death after
19
ARTICLE IN PRESS 20
Mechanisms of neuroprotection, regeneration, and repair in SCI
SCI occurs through necrosis, although apoptosis also plays an important role (Beattie et al., 2002; Keane et al., 2001; Yu et al., 2009). However, oligodendrocyte cell death occurs through apoptosis, partially dependent on activation of the Fas receptor and through p75 receptor signaling (Ackery et al., 2006; Casha et al., 2001, 2005; Chu et al., 2007; Crowe et al., 1997; Liu et al., 2009). Most of the Fas receptors in the spinal cord are found on oligodendrocytes and expressed by activated microglia and lymphocytes (Austin and Fehlings, 2008; Casha et al., 2001, 2005). The interaction between the two leads to apoptosis through the activation of caspases (Austin and Fehlings, 2008). There have been studies showing that blocking Fas-mediated cell death can lead to functional recovery and be used in the treatment of SCI (Ackery et al., 2006; Robins-Steele et al., 2012; Yu et al., 2009). The subacute phase lasts from approximately 2 days to 2 weeks after SCI. Astrocytes initially go through necrotic cell death but, in the subacute phase, become hypertrophic and proliferative (Rowland et al., 2008). The large cytoplasmic processes come together to become the gliotic scar that forms a physical and chemical barrier to regeneration (Fawcett and Asher, 1999; Hagg and Oudega, 2006; KarimiAbdolrezaee et al., 2010, 2012; Reier and Houle, 1988; Young, 2014). The cells in scar tissue release inhibitory molecules, such as chondroitin sulfate proteoglycans (CSPGs) (Fawcett and Asher, 1999; Fitch and Silver, 2008). To combat the inhibition to regeneration from the glial scar, there is interest in developing therapeutics that can remove the glial scar to promote regeneration.
1.2.2.3 Intermediate Phase The immediate phase begins approximately 2–3 weeks after injury and continues to 6 months after injury. During this phase, reactive gliosis continues as the scar begins to mature. There is also axonal sprouting of the corticospinal tract and the reticulospinal fibers (Hill et al., 2001) during this stage. Although this endogenous attempt at sprouting axons does not translate to significant functional recovery, it presents an attractive target for therapeutic intervention.
1.2.2.4 Chronic Phase The last phase in SCI is the chronic phase which begins at around 6 months postinjury and lasts for the lifetime of the patient. During the chronic phase, the lesion begins to stabilize with scar formation and cyst/syrinx development (Li and Lepski, 2013; Rowland et al., 2008). Cysts arise due to the clearance of debris by microglia and macrophages due to progressive loss of neural tissue (Basso et al., 1996; Fleming et al., 2006; Norenberg et al., 2004). Usually some axons are spared at the rim of the cysts but the cysts present a physical barrier to neuronal regeneration (Kramer et al., 2013). In addition, Wallerian degeneration of the axons continues and years for the cell bodies and axons to be removed (Beattie et al., 2002; Ehlers, 2004). Many of the therapeutic strategies used during the chronic stage aim to promote regeneration, promote plasticity, or to improve function of spared axons.
ARTICLE IN PRESS 2 Clinical intervention
2 CLINICAL INTERVENTION 2.1 CURRENT PRACTICE Historically, patients with SCI had a poor prognosis and were left unmonitored in hospital wards until their vertebrae healed. However, modern medical advances have led to improvements in patient care. The use of spineboards and the practice of immobilizing patients at the site of injury have acted to reduce injury and mortality (Anon, 2002a). In addition, surgical and pharmacological interventions are becoming more common.
2.2 SURGICAL DECOMPRESSION Surgical decompression helps to restore spinal stability and maintains cord perfusion (Mothe and Tator, 2013; Wilson and Fehlings, 2011). Spinal decompression surgery includes various procedures intended to relieve symptoms caused by pressure, or compression, on the spinal cord. There is controversy about the role and timing of surgical decompression after an acute SCI (Fehlings et al., 2001). Numerous experimental studies of decompression after SCI have been performed in various animal models including primates, dogs, cats, and rodents. These studies have consistently shown that neurological recovery is enhanced by early decompression (Fehlings and Perrin, 2005). To further address the questions regarding the optimal timing for performing decompression, our center was part of the “Surgical Timing in Acute Spinal Cord Injury Study” (STASCIS) on a multicenter, international, prospective cohort of patients. The results demonstrate a more favorable neurologic recovery among those treated with early (defined as
