In a recent systematic study we demonstrated that the SCI initiates by day 3, an Inflammatory Phase with severe, phagocytic, CD68+/CD163−, macrophage infiltration of the site of injury that becomes converted into a cavity of injury (COI) [1
]. Macrophage infiltration in the COI occurs with initially marked elevation of pro-inflammatory cytokines, chemokines and other factors. Although the numbers of macrophages and levels of pro-inflammatory factors decline after 4 weeks post-SCI, presumably influenced by a spinal cord anti-inflammatory reaction, specifically astrogliosis [4
], damage to the spinal cord around the COI and severe inflammation therein is likely considerable [5
] and an anti-inflammatory treatment would be neuroprotective.
Infiltration by phagocytic macrophages in the COI is an obvious and characteristic feature of post-SCI pathology and their count can be performed in a standardized fashion in luxol fast blue and hematoxylin and eosin (LFB + H&E) stained cross sections of the injured spinal cord. The macrophage count in the COI can serve as a quantitative test directly addressing the severity of destructive inflammation post-SCI [3
]. Therefore, a reduction in counts of macrophages in the COI of rats administered an anti-inflammatory treatment would be considered neuroprotective [4
Previous studies on anti-inflammatory drugs demonstrated that subdural infusion in the vicinity of the crush lesion, allows for lowering of numbers of macrophages in the COI presumably by simple diffusion of administered dexamethasone, M-T7 or Serp-1 into the fluid filled COI [1
]. We recently reported that administration of the Myxoma virus-derived immune modulating serpin, Serp-1 [8
], infused locally for 7 days starting immediately after balloon crush SCI, reduced inflammation and improved neurologic scores in rat models [3
]. In another experiment, Serp-1 delivered in the chitosan hydrogel in the acute lesion created by the dorsal spinal column crush also resulted in anti-inflammatory effect associated with reduction of the size of the lesion and improvement in clinical scores [9
]. Serp-1 binds and inhibits tissue- and urokinase-type plasminogen activators, plasmin in the thrombolytic cascade, and thrombin and Factor X in the coagulation/thrombotic cascade [10
]. This serpin protein biologic has been shown to reduce vasculitis in vascular injury and in organ transplant models [11
]. Serp-1 was also tested in a Phase 2 clinical trial in patients with unstable coronary syndromes and coronary stent implant where it significantly reduced markers of heart damage [12
]. Serp-1 is thus a promising biologic for use in a variety of inflammatory conditions, including SCI [3
A successful preliminary pre-clinical study mandated by the Food and Drug Administration [13
] of an anti-inflammatory/neuroprotective compound needs to demonstrate its effectiveness, the optimal route of administration, the optimal effective dose and the duration of treatment necessary to eliminate destructive inflammation initiated by neurotrauma [4
]. Previously, we have established a novel test, phagocytic macrophage count in the cavity of injury (COI) [3
] which is directly relevant to the severity of destructive inflammation initiated by the SCI and allowed us to determine anti-inflammatory effectiveness of Serp-1, and its optimal route of administration defined as subdural infusion [3
]. We have also determined that a successful neuroprotective treatment of destructive inflammation post-SCI, will need to be considerably longer than 1–2 weeks [1
] to eliminate inflammation. This is related to the slowing of phagocytosis of myelin-rich necrotic debris and red blood cells in treated rats [1
]. We postulate that elimination of the destructive inflammation from the COI will render this cavity amenable to implantation of a functional bridge for axonal regeneration across this fluid-filled cavity in similar fashion to that achieved in the Long Evans Shaker (LES) dysmyelinated rat implanted with the choroid plexus [14
In this study, using the macrophage count test, we determined the optimal dose of Serp-1 administered in subdural infusion and the duration of the infusion to accelerate elimination of macrophage infiltration in the COI thus resulting in neuroprotection.
2. Experimental Section
This study was performed in 2 phases to determine; (1) the dose effect of Serp-1 infused subdurally for 1 week to inhibit macrophage infiltration of the cavity of injury (COI) post-SCI, and (2) the duration of such treatment by the optimal dose of Serp-1 leading to elimination of macrophages from the COI.
2.1. Ethical Considerations
Experiments using male, 16 weeks old Long Evans rats, 370–410 g, were approved by the Animal Research Ethics Board at McMaster University according to the Guides and Regulations of the Canadian Council of Animal Care. Given the invasive nature of the SCI, detection of the following End Point; lethargy, marked dehydration, hypothermia, and/or ruptured urinary bladder was followed by humane euthanasia and the rat was excluded from the study (n = 8).
2.2. Serp-1 Protein Expression and Purification
Recombinant Serp-1 (m008.1L; NCBI Gene ID# 932146) was expressed and secreted by a Chinese hamster ovary (CHO) cell line (Viron Therapeutics Inc., London, ON, Canada). GMP-compliant purification was performed with greater than 95% purity as determined by Coomassie-stained SDS PAGE and reverse-phase HPLC. Serp-1 was endotoxin-free by LAL assay and stocked at −80 °C in 100 mM citrate buffer (pH 6.5) at a concentration of 1.8 mg/mL [12
2.3. Balloon Crush SCI in the Rat
The surgical procedure involved the balloon crush SCI and subdural infusion was performed in 58 rats (Table 1
) and has been previously described [1
Briefly, 58 rats were induced for SCI surgery in 5% isoflurane in 95% oxygen and maintained under 4% isoflurane in 96% oxygen. The laminectomy was created at T10 vertebrae and a 3Fogarty catheter (Balton, Warsaw, Poland) inserted towards the head over the dura to position the balloon over the mid-thoracic spinal cord. The balloon was inflated with 20 µL saline for 5 s, deflated and removed. For subdural infusion, a small cut was created in the dura over the dorsal spinal cord in the laminectomy and a rat intrathecal catheter (Alzet®, Durect Corporation, Cupertino, CA, USA) inserted over the spinal cord cranially to approximate the catheter tip with the site of the crush injury. The other end of the catheter was connected to the osmotic pump (Alzet) placed under the skin of the flank. For determination of the optimal dose of Serp-1, 0.008 mg, 0.04 mg or 0.2 mg of the protein in saline were loaded into 2ML1 osmotic pumps providing a constant flow of 10 µL/h over 7 days (Alzet). For determination of the duration of infusion required to eliminate macrophage infiltration from the cavity of injury (COI), 2ML4 pumps were loaded with 0.8 mg Serp-1 or saline providing a constant flow of 2.5 µL/h over 28 days. At 28 days post-SCI, the spent pumps were replaced by newly loaded ones via a skin incision and under isoflurane anesthesia. During the total 56 days treatment, 1.6 mg of Serp-1 was infused. Prior to awakening from anesthesia, all rats received injection of 0.4 mL ketoprofen analgesic (10 mg/mL, Anafen, Merial Canada, Inc., Baie d’Urfe, QC, Canada) and 5 mL saline subcutaneously.
2.4. Post-Operative Care and Clinical Testing
Post-surgical rats were attended 1–2 times per day. Hydration status, body temperature and presence of lethargy were assessed. Rats with dilated urinary bladder were gently voided and rats with hemorrhagic urine treated by intramuscular injection of 50 µL enrofloxacin (50 mg/mL, Baytril®, Mississauga, ON, Canada) for 3–5 days until blood cleared from urine. The function of the urinary bladder typically returned during the second week post-SCI.
To detect therapeutic effect of Serp-1 in SCI rats a simplified hind end locomotor test with 7 scoring levels was developed previously [7
] and used every day starting with day 1 post-SCI to assess the locomotor function of the hind legs as described in the Table 2
In addition, a pinch withdrawal reflex test (Table 3
) previously developed [7
] was used every day in post-SCI rats.
At 7 days post-SCI in the Serp-1 dosing study and at 14, 28, and 56 days post-SCI in the effective duration of administration study, rats were overdosed with 80 mg/kg body weight sodium pentobarbital, intraperitoneal, and the whole body perfusion with saline followed by formalin [4
]. The spines were dissected, post-fixed in formalin for 1–2 days and decalcified in formalin supplemented with 8% ethylenediaminetetraacetic acid (EDTA, Bioland Scientific, Paramount, CA, USA) [3
]. When soft, the spines were cut perpendicularly at 3 mm thick, consecutive segments including the laminectomy and the site of injury. Eight consecutive segments were processed, embedded in paraffin wax, 5 µm thick sections mounted on glass slides and stained with luxol fast blue and counterstained with hematoxylin and eosin (LFB + H&E). Additional sections were labelled with a primary anti-CD68 antibody and brown color developed as described previously [3
Macrophage Counts in the Cavity of Injury
The LFB + H&E stained sections of the spinal cord were analyzed under a light 50i Eclipse Nikon microscope by an experienced experimental neuropathologist and one area with the cavity of injury (COI) photographed per section with 40× objective; 3–5 sections per rat as described previously [3
]. The images measured 225 × 300 µm and included the periphery of the COI with adjacent spinal cord tissue taking 20% of the area of the image. Macrophages; large cells with a large oval nucleus and abundant vacuolated cytoplasm with blue granules of myelin and/or with red blood cells, were counted and the average for each rat in a treatment group (Table 1
) was averaged and standard deviation calculated [3
2.6. Statistical Analysis
The statistical analysis was performed using STATISTICA software, version 13.0 (StatSoft, USA) and GraphPad Prizm v8.4.3 with the significance level of 0.05. Normal distribution was assessed using Shapiro-Wilk and Kolmogorov-Smirnov tests. Variables were presented as the mean and stdev due to normal distribution and differences were tested using Student parametric t-test and one-way and two-way ANOVA with Holm-Sidak and Sidak post-hoc tests, respectively, where indicated.
A recently completed detailed study on the pathogenesis of SCI revealed that when trauma resulting in local massive injury to the white matter initiates, by day 3 post-SCI, an Inflammatory Phase characterized by severe infiltration by phagocytic CD68+/CD163− macrophages and its destructive, extraordinarily prolonged the course lasting beyond 16 weeks [4
]. The destructive inflammation initiated by SCI presents an obvious target for anti-inflammatory treatments to achieve neuroprotection.
Anti-inflammatory treatments previously attempted have not benefited from sufficient understanding of the pathogenesis of SCI and high-dose, short-term intravenous infusion of methylprednisolone [18
] has been shown to cause severe side effects [3
] and has recently been discouraged [25
]. Other compounds including riluzole, glibenclamide and cethrin [26
], and also fumaric acid esters [27
], estrogen [28
], endaravone [29
], mitramycine A [30
], and N
] have recently been studied in SCI animal models and clinical trials but only in short term treatments in initial stages of SCI. The effects of these experimental treatments have not been measured in a fashion addressing pathogenesis of SCI leaving a possibility that some of these compounds may potentially be found neuroprotective when tested in properly designed pre-clinical studies.
Recently elucidated pathogenesis of SCI and our previous studies on short term treatment, 1–2 weeks [1
], indicate that reduction of numbers of macrophages in the COI results in slower phagocytosis of myelin-rich necrotic debris and red blood cells requiring extended treatment to enable fewer macrophages to phagocytize and remove this immunogenic material. We hypothesized that continuous infusion of Serp-1, an immunomodulating protein with anti-inflammatory action would shorten the macrophage-rich inflammation down from 16 weeks in the rat model [4
] resulting in neuroprotective effect.
Despite their brevity, experiments with infusion for 7 days however, have had their use in determination for Serp-1 of the; (1) robust anti-inflammatory effect, (2) optimal route of administration as subdural infusion, demonstrated in the previous study [3
] and, (3) dose effect with 0.2 mg/week showing the strongest macrophage inhibition as determined in this study. The neuroprotective effect of an anti-inflammatory treatment however, requires sufficiently long continuity of administration, at least 8 weeks, to eliminate inflammation. The subdural infusion of Serp-1 for 56 days almost completely extinguished the inflammation by lowering the numbers of macrophages in the COI to very few, similar to numbers counted at 16 weeks in untreated rats [4
]. Therefore, we consider that the treatment resulted in overall inhibition and shortening of the destructive inflammation initiated by SCI and can be considered neuroprotective.
The phagocytic, pro-inflammatory macrophage count is of direct consequence to the severity of destructive inflammation initiated by SCI [4
], therefore it can serve as a reliable test for anti-inflammatory and neuroprotective effect of candidate drugs in pre-clinical studies. Importantly, macrophage count-lowering effect of Serp-1 infusion was associated with persistence of myelin-rich necrotic debris among scattered macrophages at 2 and 4 at weeks post-SCI, not observed in saline treatments at these time points in this and in the previous study [4
]. However, since it is based on histologic analysis of the spinal cord, the macrophage count is not suitable for clinical studies and in vivo laboratory assays measuring biomarkers of the spinal cord damage and of the severity of inflammation in the body fluids such as blood plasma and the cerebrospinal fluid need to be considered and developed for systematic evaluation of neuroprotective efficacy of candidate drugs in individual SCI patients along the duration of this very destructive and protracted inflammatory disease.
The hind end locomotor test and hind limb pinch withdrawal test have been used previously in a 7 day study where Serp-1 was infused subdurally [3
] and also in a 28 day study where Serp-1 was delivered from a chitosan hydrogel injected into the dorsal column crush [9
]. In both studies, a beneficial effect of Serp-1 administration was observed and supported by lower counts of macrophages in the COI infused with Serp-1 [3
] and reduced size of the crush injury with Serp-1 hydrogel [9
]. In the present study however, both tests revealed no higher scores for Serp-1 treatments for 7 days (Figure 1
A) and for 56 days (Figure 2
A) despite the macrophage counts and histologic analysis consistently indicating anti-inflammatory effect of Serp-1 infusion. It needs to be highlighted that for both Serp-1 and saline treatments, the scores in both neurological tests used in this study stabilized during the week 4 of the treatment and did not change during the remaining 4 weeks (Figure 2
A) despite the histologic evidence of active inflammatory disease. Given the above considerations the interpretation of the results of two neurologic tests used remains difficult [7
Although the experimental subjects, the SCI rats, were not randomized in this study as recommended previously [32
] all rats were male LE of the same age raised in the same colony, therefore of the same quality [15
]. In rats coded to obscure the identity to the examiner, the results of two simplified neurological exams and also of the macrophage count in the COI test were reliable and consistently reproducible.