Experimental Neurology 158, 351–365 (1999) Article ID exnr.1999.7118, available online at http://www.idealibrary.com on
Depletion of Hematogenous Macrophages Promotes Partial Hindlimb Recovery and Neuroanatomical Repair after Experimental Spinal Cord Injury Phillip G. Popovich,* Zhen Guan,† Ping Wei,† Inge Huitinga,‡ Nico van Rooijen,§ and Bradford T. Stokes† *Department of Medical Microbiology & Immunology and †Department of Physiology, The Ohio State University College of Medicine and Public Health, 333 W. 10th Avenue, Columbus, Ohio 43210; and ‡The Netherlands Institute for Brain Research and §Department of Cell Biology and Immunology, Vrije Universiteit, Van der Boechorststraat 7, 1081 BT Amsterdam, The Netherlands Received January 15, 1999; accepted April 9, 1999
Traumatic injury to the spinal cord initiates a series of destructive cellular processes which accentuate tissue damage at and beyond the original site of trauma. The cellular inflammatory response has been implicated as one mechanism of secondary degeneration. Of the various leukocytes present in the spinal cord after injury, macrophages predominate. Through the release of chemicals and enzymes involved in host defense, macrophages can damage neurons and glia. However, macrophages are also essential for the reconstruction of injured tissues. This apparent dichotomy in macrophage function is further complicated by the overlapping influences of resident microglial-derived macrophages and those phagocytes that are derived from peripheral sources. To clarify the role macrophages play in posttraumatic secondary degeneration, we selectively depleted peripheral macrophages in spinal-injured rats during a time when inflammation has been shown to be maximal. Standardized behavioral and neuropathological analyses (open-field locomotor function, morphometric analysis of the injured spinal cord) were used to evaluate the efficacy of this treatment. Beginning 24 h after injury and then again at days 3 and 6 postinjury, spinal cord-injured rats received intravenous injections of liposome-encapsulated clodronate to deplete peripheral macrophages. Within the spinal cords of rats treated in this fashion, macrophage infiltration was significantly reduced at the site of impact. These animals showed marked improvement in hindlimb usage during overground locomotion. Behavioral recovery was paralleled by a significant preservation of myelinated axons, decreased cavitation in the rostrocaudal axis of the spinal cord, and enhanced sprouting and/or regeneration of axons at the site of injury. These data implicate hematogenous (blood-derived) macrophages as effectors of acute secondary injury. Furthermore, given the selective nature of the depletion regimen and its proven efficacy when administered after injury, cell-specific
immunomodulation may prove useful as an adjunct therapy after spinal cord injury. r 1999 Academic Press Key Words: neuroinflammation; macrophages; microglia; regeneration; spinal cord injury; liposomes; immunosuppression.
Leukocyte infiltration is consistently observed in human and experimental spinal cord injury (SCI). Spinal cord inflammation has been implicated in promoting secondary injury at the injury site and in lesion extension into nearby rostral/caudal spinal levels (15, 32, 63). Of the various inflammatory cells present, macrophages are present in greatest number and for the longest duration after injury. In peripheral tissues, macrophage activation is essential for restructuring and repairing injured tissues. However, in the process of wound repair, macrophages release nitrogen/oxygen metabolites, cytokines, and proteases that are capable of damaging healthy cells. While this type of bystander damage is easily overcome by regenerative peripheral tissues, CNS neurons are less likely to regenerate. Recent studies also have implicated infiltrating macrophages as the principal source of chondroitin sulfate proteoglycans (36). These molecules are present throughout the site of a CNS injury and are putative inhibitors of axon growth (29, 36). Therefore, since SCI elicits a robust macrophage response, it is conceivable that these cells mediate posttraumatic secondary injury and/or contribute to the failure of CNS axonal regeneration. Indeed, attenuation of the macrophage response has been shown to be effective in improving neurologic function and in decreasing axon/myelin damage after experimental SCI (16, 17, 42). Liposome-encapsulated dichloromethylene bisphosphonate (Cl2MBP or clodronate), when injected intravenously, induces selective apoptotic cell death in monocytes and phagocytic macrophages (49, 75, 76). Depletion
0014-4886/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.
POPOVICH ET AL.
occurs rapidly (within 24 h) and without directly affecting other peripheral leukocytes (65). CNS microglia are also unaffected as the liposomes are unable to cross the blood–brain barrier (10, 58). Subsequently, this technique has been used to define whether hematogenous macrophages play a role as pathologic effector cells in models of CNS autoimmunity (10, 50), stroke (68), and Wallerian degeneration (24). In the present study, we hypothesized that by minimizing peripheral macrophage accumulation at the site of trauma using the clodronate liposome technique, we would attenuate tissue damage and promote neurologic recovery in rats subjected to a model of spinal contusion injury. Our present results demonstrate the efficacy of this targeted immunotherapy in improving indices of neuroanatomical repair and functional recovery. MATERIALS AND METHODS
Spinal cord injury. A total of 42 female Lewis rats were used for these experiments (12 weeks old at time of injury, 190–220 g). Under ketamine/xylazine (80 mg kg⫺1/50 mg kg⫺1 ) anesthesia, rats received a partial laminectomy of the eighth thoracic vertebrae after which the spinal cord was displaced a distance of 0.9 mm (moderate injury level) over 23 ms using the Ohio State spinal contusion injury device (22, 71). Postoperative care consisted of hydration with lactated Ringer’s solution (5 cc, s.q.), manual voiding of bladders two to three times/day, and daily injection of antibiotics (s.q., gentamycin, 1 mg/kg). After injury, individual rats were randomly assigned into a treatment group (injury control, n ⫽ 12; macrophage depleted, n ⫽ 22; PBS only, n ⫽ 8). On days of behavioral evaluation, injections were given after testing was completed. Injured rats survived either 7 (n ⫽ 8) or 28 days (n ⫽ 30) postinjury and were assigned into their survival groups by a lab member not involved in behavioral analysis. Postinjury evaluation of biomechanical parameters (e.g., injury force and displacement) revealed significant interanimal variability in four animals. No behavioral or morphometric data is reported for these animals. Liposome preparations. Multilamellar liposomes were prepared as described previously (75). Briefly, 86 mg phosphatidylcholine and 8 mg cholesterol (Sigma, St. Louis, MO) at a molar ratio of 6:1 (140 µmol of total lipid) were dissolved in 10 ml chloroform in a 500-ml round-bottom flask then dried in vacuo on a rotary evaporator to form a film. The film was dispersed into liposomes after the addition of 2.5 g dichloromethylene bisphosphonate (Cl2MBP or clodronate; a gift from Boehringer Mannheim GmbH, Mannheim, Germany) to 10 ml PBS. The preparations were kept at room temperature for 2 h, sonicated for 3 min at 20°C (50 Hz), and then kept at room temperature for an additional 2 h. The liposomes were centrifuged at 100,000 ⫻ g
for 30 min and resuspended in 4 ml PBS. For macrophage depletion, animals received injections (2 ml/ injection) of Cl2MBP-liposomes at days 1, 3, and 6 postinjury into the tail vein. There were no adverse side effects associated with any of our injection schedules. Behavioral evaluation. All rats were acclimated to an enclosed open-field area for 4 days prior to the onset of behavioral testing. One day before injury, each animal was scored by two observers blinded to the experimental treatment using the standardized Basso– Beattie–Bresnahan (BBB) locomotor rating scale (7). Injured rats were tested daily for 1 week at which time one-half of the animals in each group were randomly selected and euthanized via intracardiac perfusion with 4% paraformaldehyde. The remaining animals were tested on days 10, 14, 18, 21, 24, and 28 postinjury. All remaining animals were euthanized after testing on day 28. The sensitivity of the BBB scale is sufficient to distinguish gross locomotor differences between spinalinjured rats. However, in cases where partial recovery of locomotion occurs without reestablishing a one-toone correspondence between alternating fore- and hindlimb stepping (i.e., FL/HL coordination), the 21-point BBB scale may not reflect changes in the finer details of locomotion (e.g., paw positioning, toe clearance). Indeed, others have noted little or no recovery of coordinated walking in rats receiving a moderate contusion injury (i.e., BBB scores plateau at 11 or 12) (7, 8). Therefore, since paw position and toe clearance are routinely documented once animals are able to step consistently, we supplemented BBB analyses by subscoring the fine details of locomotion at the plateau of the recovery phase (21–28 days postinjury). A subscore, 0–5, was given to each hindlimb based on paw rotation and toe clearance of the hindlimb. Subscores were assigned as follows: paw position, 0 ⫽ rotation at initial contact and liftoff, 1 ⫽ rotation at initial contact or liftoff and parallel at initial contact/liftoff, 2 ⫽ parallel at initial contact and liftoff; and toe clearance, 0 ⫽ no clearance, 1 ⫽ occasional clearance (ⱕ50% of the time), 2 ⫽ frequent clearance (51–95%), 3 ⫽ consistent (⬎95%) clearance. The cumulative scores of each hindlimb were summed to yield a single score (maximum score of 10/rat). Tissue processing. Anesthetized animals were euthanized by intracardiac perfusion with 300 ml of 4% paraformaldehyde in 0.1 M PBS. The spinal cord, liver, and spleen were removed and placed in fixative. Using the impact site as the central point, a 14-mm segment of injured spinal cord was embedded in paraffin. The remainder of the spinal tissue rostral and caudal to the injury site was immersed in 30% sucrose then frozen on dry ice. Paraffin blocks were serially sectioned at 15 µm in the transverse plane with every fourth, fifth, and sixth sections kept for morphometric analysis. Each set
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
of sections was used to stain for either macrophages, myelin sparing/cavitation, or axons (see below). In addition to their effects on circulating monocytes/ macrophages, intravenously injected clodronate liposomes deplete macrophages in the spleen and liver. As such, these organs were embedded in paraffin and were stained with ED1 (see below) to confirm macrophage depletion in each animal (77). Effective depletion was noted in each animal (data not shown). Morphometric analysis of the contusion lesion. As with the behavioral analyses, slide preparations were coded to facilitate blinded quantitation of morphometric data. Myelin sparing of luxol fast blue (LFB)-stained tissue sections was assessed over an 8-mm rostral– caudal interval of the contusion lesion using computerassisted morphometric techniques (MCID, Imaging Research, Inc., Ontario, Canada). Cresyl violet counterstains were used to facilitate identification of cell nuclei and to discriminate between cellular debris and intact/preserved tissue. Spared white matter was quantified by measuring the area of tissue identified by LFB-positive tissue. Computer-defined regions of LFB staining were always confirmed and when necessary, manually edited using high power light microscopy (50⫻). This type of analysis has been shown to correlate with locomotor recovery and is a preferred method for assessing the efficacy of various pharmacotherapies after rodent SCI (7, 8, 13, 14). Areas devoid of tissue or tissue containing only cellular debris or macrophages were designated as areas of cavitation. Often times spared gray matter or a fibrous connective tissue matrix were interposed between areas of cavitation. These regions were quantitatively discriminated from cavities or white matter and are collectively referred to as gray matter/matrix. In instances where tissue imperfections or processing artifacts (tears, inhomogeneity of staining, etc.) were present, data were not collected. Macrophage and axon analysis. Paraffin sections adjacent to those used for morphometric analysis were used for immunohistochemical detection of activated microglia and hematogenous macrophages (ED1: specific for a cytoplasmic protein with homology to the CD68 molecule in phagocytic macrophages; Harlan Bioproducts for Science; 1:1,000) (28) and axons (RT-97; specific for the 200 kd phosphorylated neurofilaments in large caliber axons; Boehringer Mannheim, 1:1,500). Paraffin-embedded tissues were dewaxed through xylene then rehydrated through descending concentrations of alcohol (100–70%) into PBS. Rehydrated tissues were incubated with primary antibodies then visualized using a modified immunoperoxidase technique. Macrophage activation/infiltration was quantified using proportional area measurements as defined previously (63, 64). Briefly, an increased proportional area value (area fraction of tissue occupied by immunohistochemically stained macrophages relative to the
area of the region sampled) indicates an increase in macrophage size or cell density. To quantify macrophages in white matter, LFB-stained tissues were used to create a template after which the digital templates were superimposed onto adjacent ED1-stained tissue sections. Macrophages were also quantified within the central region of the spinal cord that was circumscribed by the white matter templates. Within these central regions, no attempts were made to distinguish between macrophages in cavities, regions of spared gray matter, or fibrous connective tissue. RT-97-positive axons were quantified in an identical fashion. For axon quantification, these different types of analyses estimated the degree of axon sparing, regeneration, and/or sprouting within the center of the lesion and in spinal white matter, respectively (55). Spinal cord sections were captured digitally, color balance/tonal adjustments were made as needed, and dust, scratches, etc., introduced during the staining procedure were eliminated. Statistics. Open-field locomotor scores were analyzed by repeated measures analysis of variance with Tukey’s multiple comparison test at each time point. Behavioral subscores were analyzed using the Mann– Whitney nonparametric test. Comparisons of morphometric data between macrophage depleted and control groups were made using unpaired t tests or, in the case of unequal variances (F test), the Mann–Whitney nonparametric test was used. Regression analyses were used to correlate morphometric measures with behavioral recovery. Behavioral and morphometric differences were not observed between control groups. Therefore, to simplify presentation, control groups were pooled and are represented throughout the text and figures as injury/vehicle controls. Statistical significance was set at a value of P ⬍ 0.05. RESULTS
Clodronate liposomes are well tolerated by SCI rats. Previous studies using this method of macrophage depletion have not reported any adverse side effects of the treatment. Still, because of the chronic nature of our studies, we were concerned about increased susceptibility to bladder infections in macrophage-depleted (MD), spinal cord-injured animals. However, no chronic urinary problems or infectious complications were detected. Clodronate liposomes reduce macrophage infiltration into the injured spinal cord. A single intravenous injection of clodronate liposomes significantly reduces circulating monocytes and macrophages in organs with an open vascular supply (e.g., spleen, liver) (49, 77). Depletion is evident within 24 h and persists for as long as 1 month in spleen and liver (77). In the present study, we confirmed macrophage depletion in the periph-
POPOVICH ET AL.
ery using histological/morphological criteria as previously described (77). Specifically, we noted an absence of ED1 staining throughout the spleen and liver at 7 days postinjury. By 28 days postinjury, some ED1 immunoreactivity was noted (indicating some repopulation of the depleted macrophage pools) but was qualitatively less than that seen in control tissues (not shown). Using a multiple injection schedule (1, 3, and 6 days postinjury) we significantly reduced macrophage accumulation at the injury site during the first 7 days postinjury—a time when the majority of blood monocyte/ macrophage influx occurs (63). Significant depletion was evident within regions previously occupied by gray matter and propriospinal pathways (central spinal cord) (Figs. 1 and 2). ED1⫹ macrophages, presumably activated microglia, were still present in superficial white matter tracts. Although there was a tendency for fewer ED1⫹ cells to be present in these spinal regions by 28 days postinjury, the magnitude of the macrophage response began to approach that of control animals (Fig. 3). Remnants of gray matter and the accumulation of a cellular matrix were present in the chronically injured spinal cord of MD rats (Figs. 3D and 5). Effects of macrophage depletion on functional recovery after SCI. Similar to previous studies using this injury device (7, 13, 14, 22, 71) we observed pronounced hindlimb impairment immediately after SCI with partial recovery over the next 4 weeks (Fig. 4A). Typical recovery patterns in control animals progressed from flaccid hindlimb paralysis on days 1 and 3 postinjury to frequent or consistent plantar stepping by 4 weeks postinjury. Motor deficits were still apparent at this chronic postinjury time point and included paw rotation, inability to clear the toe during the swing phase of stepping, and hindlimbs that were widely abducted. The severity of these deficits was much greater in control animals compared to MD rats. Macrophage-depleted animals were able to position their paw parallel to the body and/or clear their toes during the step cycle (Fig. 4B). External paw rotation/ hindlimb abduction without toe clearance was observed in only 6 MD animals (38%) compared to 14 control animals (61%). Of the control animals that had a subscore ⬎0, one-third of these animals never cleared their toes during the step cycle. Only 1 macrophagedepleted animal showed this deficit. The characteristics of hindlimb use exhibited by MD animals are typical of normal uninjured rats. Further improvement in locomotion was evident in 31% of MD animals with 3 achieving frequent (51–94%) and 1 animal recovering consistent (95–100%) coordination. Only 1 of 14 injury/ vehicle control rats showed frequent coordinated forelimb–hindlimb stepping by 4 weeks postinjury. Indices of behavioral recovery correlated with increased white matter sparing (P ⬍ 0.01) and axon growth into the
lesion epicenter (P ⬍ 0.05; see below for morphometric details). Increased myelin sparing and decreased cavitation after macrophage depletion. Morphometric analyses of the lesion center did not reveal differences in section area (P ⫽ 0.23) between experimental and control groups. Thus, macrophage depletion did not affect parameters of edema or swelling mediated by posttraumatic inflammation. In contrast, we found significant preservation of spinal white matter at the impact site (Fig. 5A). Marked differences were also noted between MD and injury/vehicle controls when evaluating regions of cavitation. In MD rats, cavities were significantly reduced in size at the lesion site and in caudal spinal segments (Fig. 5B). Preservation of axons and regeneration at the injury site after macrophage depletion. Microscopic analysis of injured spinal cord sections also revealed a welldeveloped fibrous matrix and remnants of gray matter within the lesion epicenter of MD animals (Fig. 5C). Figure 6 demonstrates the relationship between spinal level, white matter sparing, cavitation, gray matter sparing, and the accumulation of a cellular matrix. The latter is presumably composed of glia, fibroblasts, ependyma, and endothelial cells (11, 45). Because of a lack of clearly demarcated boundaries, we did not discriminate between spared gray matter and the nonneuronal cellular matrix in our quantitative analysis (Fig. 5C). Neurofilament immunohistochemistry revealed few axons within the lesion center or white matter of control animals (Figs. 7A–7C). In contrast, robust axon growth was colocalized to regions of cellular matrix accumulation in each MD spinal cord (Fig. 7D). These axons appeared to grow without specific orientation and often surrounded blood vessels (Figs. 7E and 7G). We also observed significant sparing of large caliber axons within the spared white matter of MD animals (Fig. 7F). Quantitative image analysis of neurofilamentpositive axons within spared white matter and the lesion core (region of central canal to the edges of the spared white matter) revealed significantly more axon sparing and growth, respectively, after macrophage depletion (Fig. 8). DISCUSSION
The present data implicate blood-derived monocytes/ macrophages in the progression of secondary injury after acute spinal trauma. In rats depleted of hematogenous macrophages we observed improved overground locomotion that was accompanied by decreased intraspinal macrophage activation, increased sparing of myelinated axons, and decreased spreading of necrotic cavitation in the injured spinal cord. Furthermore, given that an injury of this severity destroys axons within the
FIG. 1. Clodronate liposomes decrease macrophage (ED1⫹ cells) influx into the spinal cord after contusion injury (7 days postinjury). Low power photomicrographs of injury control (A) and macrophage-depleted (MD) spinal cord sections (B). High-power fields of boxed regions (A and B) demonstrate differences in macrophage density between control (C) and MD tissues (D). Bars, 500 µm (A and B) and 100 µm (C and D).
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
POPOVICH ET AL.
FIG. 2. Macrophage/microglial proportional area measurements (⫾SEM) within spared white matter (A), central region of spinal cord (B), or with respect to the total spinal cord cross-sectional area (C). These quantitative data demonstrate the effective reduction of macrophage influx into the injured spinal cord (7 days postinjury) after clodronate liposome treatment. White matter ED1 proportional area is decreased but is not statistically different from injury/vehicle controls (A) suggesting that the total macrophage reduction observed in (C) is mostly a result of decreased influx into the central region of the spinal cord (B) [macrophage depleted (n ⫽ 4) and injury/vehicle controls (n ⫽ 5); **P ⬍ 0.01 in B and C].
lesion center by 1–2 days postinjury, increased axonal immunoreactivity noted in the lesion center of macrophage-depleted animals suggests a cellular environment more conducive to axonal sprouting and regeneration. From these data, the overlap between secondary injury and the onset of endogenous repair mechanisms seem to involve complex interactions between infiltrating macrophages and resident CNS cells. Future studies are necessary to clarify these relationships and define the neuroanatomical basis for the behavioral improvements documented in this study. The acute inflammatory response and recovery of function. To avoid ambiguity in data interpretation, previous studies designed to evaluate therapeutic efficacy after SCI have relied on complete or partial transection of defined spinal pathways (21, 26, 44). Using a model of spinal contusion injury, we and others have attempted to model the most prevalent of human injuries. Unfortunately, contusion injury disrupts numerous ascending and descending systems making it difficult to assess structure/function relationships without using a battery of behavioral tests (43, 51, 52). However, with the present model it is clear that general features of overground locomotion correlate with preservation of myelinated axons at the injury site (7, 8, 13, 14). Thus, we hypothesized that clodronate liposomes would reduce macrophage-mediated destruction of spinal tissue thereby improving general features of motor recovery. Therefore, standardized measures of motor recovery and morphometric analysis were used to screen the therapeutic efficacy of a well-defined macrophage depletion technique. Using the Basso–Beattie–Bresnahan locomotor rating scale (7), we were able to detect and quantify distinct improvements in the quality of hindlimb use (paw rotation and hindlimb flexion/toe clearance during stepping) between MD and injury/vehicle control animals, a partial restoration of function similar to that observed in full or partial transection models of spinal
injury (26, 44, 67). Improved hindlimb use may be explained by a greater preservation of myelinated axons at the impact site. Alternatively, local axonal sprouting/regeneration accompanied by significant reduction of necrotic cavitation into caudal spinal levels may facilitate recovery of spinal segmental reflexes without involvement of supraspinal projections (4, 9, 33). Given that the integrity of the neural substrate is significantly improved and that axon growth into the lesion center is enhanced, more comprehensive behavioral studies in combination with anatomical tract tracing/axon phenotyping techniques are necessary to define the mechanisms underlying functional recovery after macrophage depletion. Other studies will focus on how cells of the immune system coordinate these effects on neuronal function and behavioral recovery. Specifically, the quality of the macrophage response, either directly or through its effects on glial–glial/ glial–neuronal interactions, will determine whether secondary injury progresses or neural regeneration is achieved. Spinal cord injury and inflammation. Similar to spinal trauma caused by vertebral fracture–dislocation injuries in man, experimental spinal contusion injury produces a cyst-like region of central necrosis surrounded by a preserved rim of axons—many of which are dysfunctional or show varying degrees of dysmyelination. Over time, demyelination extends into rostral and caudal spinal levels (2, 23). The mechanisms underlying this secondary damage and lesion extension are not known but may be related to inflammatory cell-mediated oligodendrocyte apoptosis (27, 70). The marked loss of gray matter and axons within the lesion center occurs within the first few days postinjury (11, 45, 55) and is related temporally and spatially with microglial activation (25, 63), increased blood–spinal cord barrier permeability (60, 61), and marked elevation of inflammatory neurotoxins (18, 62), chemokines (56), and cytokines (69, 72). Similar data have been
FIG. 3. Phagocytic macrophages accumulate in the chronically injured epicenter of macrophage-depleted (MD) rats. By 1 month postinjury, we did not observe quantitatively different levels of ED1 immunoreactivity between injury/vehicle control (A,C) and MD tissues (B,D). However, qualitative differences were evident in select regions of gray matter. Note the presence of a substantial tissue matrix (box in B; arrows in D) surrounded by ED1⫹ macrophages/microglia in the MD tissue. Only large phagocytic macrophages are present in control tissues (C). Bars, 500 µm (A and B) and 100 µm (C and D).
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
POPOVICH ET AL.
FIG. 4. Macrophage depletion (MD) improves the fine details of hindlimb locomotor recovery (paw position and toe clearance). Although injured/vehicle control animals are able to step, their hindlimbs are abducted from the body and stepping occurs without control of toe clearance or paw position. However, the quality of the step cycle after MD is much like that of uninjured control animals. Not evident in the 21-point BBB scale (A) is the significant and continuous improvement (21–28 days postinjury) of these more refined parameters of stepping (B; *P ⬍ 0.05); i.e., there is a trend for MD animals to achieve frequent/consistent coordinated locomotion by 4 weeks postinjury (see Materials and Methods) and to place their paw parallel to the body during the step cycle without dragging their toes. Each point represents an individual animal with the bar representing the group mean [macrophage depletion (n ⫽ 13) and injury/vehicle controls (n ⫽ 14)].
obtained in models of spinal hemisection (6, 32), ischemic spinal injury (42), and lateral compression injury (84) and intimate a role for inflammation in the progressive enlargement of the primary injury site. The effectiveness of anti-inflammatory treatments in promoting functional recovery and anatomical repair support this notion (16, 42, 46, 47, 73). Interestingly, in the absence of any intervention, spinal-injured animals regain some motor function during the peak inflammatory response causing some to argue that leukocytes and resident microglia are essential for wound healing after SCI (53, 82, 83). Our current data suggest that hematogenous macrophage depletion can enhance recovery, perhaps by allowing resident nonneuronal cells, namely, microglia, to play a greater role the recovery process. It is also possible that more effective recovery could be achieved if neutrophil function were attenuated prior to macrophage depletion (within the
first 24 h—a time when neutrophils have maximally infiltrated the injury site) (74, 84). Neutrophils, like macrophages, have been implicated as mediators of secondary neuronal injury (57, 73, 84). A closer look at the variables influencing immune cell activation within the CNS may help clarify the dichotomous role played by the inflammatory response after SCI. CNS macrophage heterogeneity: microglia vs hematogenous macrophages. Activated microglia and extravasated blood monocytes constitute the majority of inflammatory cells present at the site of a SCI (15, 25, 32, 63). Whether these cells act in a coordinated fashion to resolve the injury site or whether they adopt distinct functional repertoires is unclear. However, the contribution and functional status of one macrophage subset relative to the other is likely to be influenced by several factors including: (i) the nature of the primary insult (e.g., transection, infection, blunt trauma, ischemia);
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
FIG. 5. Macrophage depletion (MD) inhibits or delays degeneration of myelinated axons at the impact site and in nearby caudal segments. There is a significant preservation of myelinated axons at the site of impact (A), decreased cavitation caudal to the epicenter (B) concomitant with marked accumulation of a fibrous tissue matrix, and preservation of portions of the gray matter (C). Spinal cord drawings adjacent to A–C are representative epicenter sections from control animals and demonstrate the regions quantified for each graph (arrows indicate general regions). A single section from each animal was quantified at each level of the spinal cord. Each data point represents the mean ⫾ SEM of n ⫽ 10 animals at 28 days postinjury (except at 3–4 mm rostral/caudal where each point represents n ⫽ 4–6). Uninjured white matter area at the levels quantified range between 3 and 4 mm2. *P ⬍ 0.05; **P ⬍ 0.01; ***P ⬍ 0.001.
POPOVICH ET AL.
FIG. 6. Line drawings demonstrating the effect of macrophage depletion (MD) on lesion cavitation (red shading), tissue matrix formation (black shading), and gray matter sparing at 28 days postinjury. Sections are representative of luxol fast blue/cresyl violet stained tissues selected randomly at 1-mm intervals beginning 3 mm rostral to the lesion epicenter (left to right, rostral-to-caudal). Note the large necrotic cavities (red) and the absence of gray matter sparing at and caudal to the lesion center in control animals (A) compared to MD animals (B). In MD tissues, when gray matter is not present it is replaced by a tissue matrix of unknown composition (black shading). Bar, 1 mm.
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
FIG. 7. Increased axon sparing and regeneration in the injured spinal cord of macrophage-depleted (MD) rats (D–G) relative to injury/vehicle control animals (A–C). Neurofilament immunohistochemistry reveals greater preservation of axons within the lateral and ventral funiculi of the injury epicenter of MD rats (D,F) compared to control rats (A,B). Also, only in MD tissues was sprouting and/or regeneration of axons increased within the central regions (compare A and D) of the spinal cord. These latter regions contained a nonneuronal tissue matrix rather than large necrotic cavities (see Figs. 5 and 6). Axon sprouts or newly regenerating axons in control tissues were typically found near or growing along blood vessels (double arrow in C). Occasionally, immunoreactive axons were found in these central spinal regions that were oriented perpendicular to the plane of section (single arrows in C). After macrophage depletion, tortuous axon bundles (small arrows in E) occupy the central spinal cord of MD tissues (D,E) where they exit/enter the region previously occupied by gray matter. Like control tissues, axons were often seen wrapped around blood vessels in MD tissues (G). Bars, 500 µm (A,D); 50 µm (B,C,E–G).
POPOVICH ET AL.
FIG. 8. Quantitative assessment of axon sparing and sprouting/ regeneration at the lesion epicenter (see Fig. 7). Data are expressed as the area occupied by immunoreactive axons relative to the area of spared white matter (A) or to the central spinal cord (B). See Materials and Methods for details of analysis. Each graph represents the mean value of n ⫽ 10 rats. ***P ⬍ 0.001.
(ii) interactions with resident and infiltrating cells; (iii) distance from the site of the injurious stimulus, related to blood–brain barrier damage and chemotactic gradients; and (iv) exposure to humoral factors or substrates of non-CNS origin (e.g., complement protein, peripheral nervous system components, or drugs). These variables will affect microglial/macrophage activation and whether secondary damage or repair occurs. It is clear that activated microglia and blood monocytes can be triggered to produce different levels and types of cytokines and neurotoxic molecules (e.g., quinolinate, superoxide) (1, 3, 38–40, 59). Coupled to the release of these compounds are differences in phagocytic potential, both to particulate antigens and to myelin (30, 59). At the ultrastructural level, microglia and hematogenous macrophages harvested from sites of CNS trauma or infection are morphologically distinct and respond differently to modulatory signals from astrocytes (31, 41). Previously, we described marked heterogeneity in the morphology and expression of
surface antigens on microglia/macrophages after SCI (63). Heterogeneous neurotrophin expression in vivo and differential responses to neurotrophins by microglia in vitro suggest that these cells may elicit unique functional properties in the pathological CNS (35). Future studies will be needed to clarify the potential for macrophage/microglial functional heterogeneity and whether the interaction of these cells affects other nonneuronal cells involved in repair of the injured CNS. Macrophage depletion vs macrophage transplantation to promote neural regeneration. In the periphery, macrophages are essential for wound repair and regeneration of injured tissues. After SCI, intrinsic repair programs are initiated that result in the formation of a matrix of endothelial sprouts, ependymal chords emanating from the central canal, reactive glia (astrocytes and microglia), and infiltrating Schwann cells (11, 45). The onset of repair precedes hematogenous macrophage infiltration but not the activation of resident microglia. Therefore, interventions that alter the kinetics or nature of the inflammatory response to trauma might affect later stages of repair, including axon regeneration. Since the matrix that forms after injury is not sufficient to maintain axonal growth, it is possible that infiltrating macrophages antagonize the efforts of resident cells to repair the injury site. Our data and previous work by Fitch and Silver (36) support the concept that the inefficient progression of endogenous repair and the formation of an axon restrictive growth environment is mediated by blood–brain barrier damage, acute infiltration of blood monocytes, and the accumulation of inhibitory extracellular matrix molecules. Acute macrophage depletion may limit chondroitin sulfate proteoglycan deposition and reduce the phagocytosis-coupled release of antibacterial agents (e.g., superoxide, hydrogen peroxide, hypochlorous acid), quinolinic acid, or proteolytic enzymes. These latter compounds, although innocuous in the regenerative tissues of the periphery, could cause inefficient repair, progressive necrosis/apoptosis, and destruction of healthy tissues and neural/glial progenitors within the CNS. Recent studies have shown that supplementing the injury site with microglia or macrophages can promote neuroanatomical repair and functional recovery (37, 54, 66, 67). Conceptually, this approach would seem to contradict our data and that of others showing the beneficial effect of immunosuppression after acute SCI (16, 42, 46, 47, 72). However, the transplant studies demonstrate that CNS regenerative failure may be partially overcome if the quality of the macrophage response is altered. Indeed, the microenvironment in which microglia and/or macrophages are activated can influence their neurotrophic or neurotoxic effector potential (81). A recent report by Rapalino, Lazarov-Spiegler,
MACROPHAGE DEPLETION AND SPINAL CORD INJURY
and colleagues (67) provides an excellent example of how ‘‘environment’’ dictates macrophage effector function. In their studies, blood monocytes were incubated with degenerating peripheral nerve segments then transplanted into a transected spinal cord. Axonal regeneration was supported and partial recovery of locomotor function was achieved. Furthermore, the regenerative potential of the macrophages was dependent on prior activation by peripheral nerve but not optic nerve segments (80), presumably because peripheral nerves lack the undefined inhibitory factors which suppress macrophage function in the CNS (48, 53). These studies, like ours, demonstrate that a more detailed understanding of the signals controlling macrophage activation is needed. The clinical relevance of manipulating macrophages. Given the magnitude of the pathology caused by SCI, promoting functional regeneration seems a difficult goal. However, sparing of as little as 1–5% of fibers in the contused or transected spinal cord is sufficient to support some locomotor function below the level of injury (8, 34, 78). Increased sparing beyond this threshold dramatically improves motor recovery (12, 13, 79). In the present study we have shown that delayed depletion of a distinct inflammatory cell population, i.e., hematogenous macrophages, promotes partial functional recovery and anatomical repair within the injured spinal cord. The white matter sparing observed in this study is of the same magnitude (increased ⬃5% relative to injury control tissues) as previously reported for spinal-injured rats given methylprednisolone (MP) (13), the clinical standard for treating acute SCI (19, 20). However, despite MP’s anti-inflammatory properties, a recent study was unable to demonstrate a relationship between decreased macrophage influx after SCI and attenuation of secondary injury (5). This may be a result of the nonspecific suppression of both beneficial and deleterious macrophage and microglial cell functions. The selective uptake of clodronate liposomes and subsequent depletion of hematogenous macrophages may help us to learn more about the role of infiltrating inflammatory cells, their interactions with other cells of the CNS, and the signals that control immunological activation. Furthermore, the neuroanatomical sparing achieved using this method of macrophage depletion may prove useful in combination with other experimental therapies (e.g., intraspinal transplantation, exogenous neurotrophin administration) designed to promote axonal growth, regeneration, and remyelination. Ultimately, such studies could help define methods for selective immunomodulation which may help preserve or promote recovery of function after SCI. ACKNOWLEDGMENTS We thank Pat Walters, Yi Fei Chen, and Annemarie Sanders for their technical expertise. We thank Drs. Dana McTigue, Lyn Jake-
man, and Michele Basso for their thoughtful critique of the manuscript. This work was supported in part by NIH Grants NS-33696 and NS-10165, Sandoz Pharmaceutical Corporation & Sandoz Research Institute, and the Medical Directors Support Fund of the Ohio State University.
REFERENCES 1. Alberati-Giani, D., P. Ricciardi-Castagnoli, C. Kohler, and A. M. Cesura. 1996. Regulation of the kynurenine metabolic pathway by interferon-␥ in murine cloned macrophages and microglial cells. J. Neurochem. 66: 996–1004. 2. Balentine, J. D. 1978. Pathology of experimental spinal cord trauma. II. Ultrastructure of axons and myelin. Lab. Invest. 39: 254–266. 3. Banati, R. B., G. Rothe, G. Valet, and G. W. Kreutzberg. 1991. Respiratory burst activity in brain macrophages: a flow cytometric study on cultured rat microglia. Neuropathol. Appl. Neurobiol. 17: 223–230. 4. Barbeau, H., and S. Rossignol. 1987. Recovery of locomotion after chronic spinalization in the adult cat. Brain Res. 412: 84–95. 5. Bartholdi, D., and M. E. Schwab. 1995. Methylprednisolone inhibits early inflammatory processes but not ischemic cell death after experimental spinal cord lesion in the rat. Brain Res. 672: 177–186. 6. Bartholdi, D., and M. E. Schwab. 1997. Expression of proinflammatory cytokine and chemokine mRNA upon experimental spinal cord injury in mouse: An in situ hybridization study. Eur. J. Neurosci. 9: 1422–1438. 7. Basso, D. M., M. S. Beattie, and J. C. Bresnahan. 1995. A sensitive and reliable locomotor rating scale for open field testing in rats. J. Neurotrauma 12: 1–21. 8. Basso, D. M., M. S. Beattie, and J. C. Bresnahan. 1996. Graded histological and locomotor outcomes after spinal cord contusion using the NYU weight-drop device versus transection. Exp. Neurol. 139: 244–256. 9. Basso, D. M., M. Murray, and M. E. Goldberger. 1994. Differential recovery of bipedal and overground locomotion following complete spinal cord hemisection in cats. Res. Neurol. Neurosci. 7: 95–110. 10. Bauer, J., I. Huitinga, W. Zhao, H. Lassmann, W. F. Hickey, and C. D. Dijkstra. 1995. The role of macrophages, perivascular cells, and microglial cells in the pathogenesis of experimental autoimmune encephalomyelitis. GLIA 15: 437–446. 11. Beattie, M. S., J. C. Bresnahan, J. Komon, C. A. Tovar, M. Van Meter, D. K. Anderson, A. I. Faden, C. Y. Hsu, L. J. Noble, S. K. Salzman, and W. Young. 1997. Endogenous repair after spinal cord contusion injuries in the rat. Exp. Neurol. 148: 453–463. 12. Behrmann, D. L., J. C. Bresnahan, and M. S. Beattie. 1993. A comparison of YM-14673, U-50488H, and Nalmefene after spinal cord injury in the rat. Exp. Neurol. 119: 258–267. 13. Behrmann, D. L., J. C. Bresnahan, and M. S. Beattie. 1994. Modeling of acute spinal cord injury in the rat: Neuroprotection and enhanced recovery with methylprednisolone, U-74006F, and YM-14673. Exp. Neurol. 126: 1–16. 14. Behrmann, D. L., J. C. Bresnahan, M. S. Beattie, and B. R. Shah. 1992. Spinal cord injury produced by consistent displacement of the cord in rats: behavioral and histologic analysis. J. Neurotrauma 9(3): 197–217. 15. Blight, A. R. 1985. Delayed demyelination and macrophage invasion: A candidate for secondary cell damage in spinal cord injury. CNS Trauma 2: 299–315. 16. Blight, A. R. 1994. Effects of silica on the outcome from
POPOVICH ET AL. experimental spinal cord injury: Implication of macrophages in secondary tissue damage. Neuroscience 60: 263–273. Blight, A. R., T. I. Cohen, K. Saito, and M. P. Heyes. 1995. Quinolinic acid accumulation and functional deficits following experimental spinal cord injury. Brain 118: 735–752. Blight, A. R., K. Saito, and M. P. Heyes. 1993. Increased levels of the excitotoxin quinolinic acid in spinal cord following contusion injury. Brain Res 632: 314–316. Bracken, M. B., M. J. Shepard, W. F. Collins, T. R. Holford, W. Young, D. S. Baskin, H. M. Eisenberg, E. Flamm, L. LeoSummers, J. Maroon, et al. 1990. A randomized, controlled trial of methylprednisolone or naloxone in the treatment of acute spinal-cord injury: Results of the Second National Acute Spinal Cord Injury Study. N. Engl. J. Med. 322: 1405–1411. Bracken, M. B., M. J. Shepard, T. R. Holford, L. Leo-Summers, E. F. Aldrich, M. Fazl, M. Fehlings, D. L. Herr, P. W. Hitchon, L. F. Marshall, et al. 1997. Administration of methylprednisolone for 24 or 48 hours or tirilazad mesylate for 48 hours in the treatment of acute spinal cord injury: Results of the Third National Acute Spinal Cord Injury Randomized Controlled Trial—National Acute Spinal Cord Injury Study. JAMA 277: 1597–1604. Bregman, B. S., E. Kunkel-Bagden, L. Schnell, H. N. Dai, D. Gao, and M. E. Schwab. 1995. Recovery from spinal cord injury mediated by antibodies to neurite growth inhibitors. Nature 378: 498–501. Bresnahan, J. C., M. S. Beattie, F. D. Todd III, and D. H. Noyes. 1987. A behavioral and anatomical analysis of spinal cord injury produced by a feedback-controlled impaction device. Exp. Neurol. 95: 548–570. Bresnahan, J. C., J. S. King, G. F. Martin, and D. Yashon. 1976. A neuroanatomical analysis of spinal cord injury in the rhesus monkey (Macaca mulatta). J. Neurol. Sci. 28: 521–542. Bruck, W., I. Huitinga, and C. D. Dijkstra. 1996. Liposomemediated monocyte depletion during wallerian degeneration defines the role of hematogenous phagocytes in myelin removal. J. Neurosci. Res. 46: 477–484. Carlson, S. L., M. E. Parrish, J. E. Springer, K. Doty, and L. Dossett. 1998. Acute inflammatory response in spinal cord following impact injury. Exp. Neurol. 151: 77–88. Cheng, H., Y. H. Cao, and L. Olson. 1996. Spinal cord repair in adult paraplegic rats: Partial restoration of hind limb function. Science 273: 510–513. Crowe, M. J., J. C. Bresnahan, S. L. Shuman, J. N. Masters, and M. S. Beattie. 1997. Apoptosis and delayed degeneration after spinal cord injury in rats and monkeys. Nature Med. 3: 73–76. Damoiseaux, J. G., E. A. Dopp, W. Calame, D. Chao, G. G. MacPherson, and C. D. Dijkstra. 1994. Rat macrophage lysosomal membrane antigen recognized by monoclonal antibody ED1. Immunology 83: 140–147. Davies, S. J., M. T. Fitch, S. P. Memberg, A. K. Hall, G. Raisman, and J. Silver. 1997. Regeneration of adult axons in white matter tracts of the central nervous system. Nature 390: 680–683. DeJong, B. A., and M. E. Smith. 1997. A role for complement in phagocytosis of myelin. Neurochem. Res. 22: 491–498. DeWitt, D. A., G. Perry, M. Cohen, C. Doller, and J. Silver. 1998. Astrocytes regulate microglial phagocytosis of senile plaque cores of Alzheimer’s disease. Exp. Neurol. 149: 329–340. Dusart, I., and M. E. Schwab. 1993. Secondary cell death and the inflammatory reaction after dorsal hemisection of the rat spinal cord. Eur. J. Neurosci. 6: 712–724. Edgerton, V. R., R. R. Roy, J. A. Hodgson, R. J. Prober, C. P. de Guzman, and R. de Leon. 1992. Potential of adult mammalian lumbosacral spinal cord to execute and acquire improved locomo-
tion in the absence of supraspinal input. J. Neurotrauma 9(Suppl. 1): S119–S128. Eidelberg, E., D. Straehley, R. Erspamer, and C. J. Watkins. 1977. Relationship between residual hindlimb-assisted locomotion and surviving axons after incomplete spinal cord injuries. Exp. Neurol. 56: 312–322. Elkabes, S., E. M. DiCicco-Bloom, and I. B. Black. 1996. Brain microglia/macrophages express neurotrophins that selectively regulate microglial proliferation and function. J. Neurosci. 16: 2508–2521. Fitch, M. T., and J. Silver. 1997. Activated macrophages and the blood–brain barrier: inflammation after CNS injury leads to increases in putative inhibitory molecules. Exp. Neurol. 148: 587–603. Franzen, R., J. Schoenen, P. Leprince, E. Joosten, G. Moonen, and D. Martin. 1998. Effects of macrophage transplantation in the injured adult rat spinal cord: A combined immunocytochemical and biochemical study. J. Neurosci. Res. 51: 316–327. Giulian, D. 1987. Ameboid microglia as effectors of inflammation in the central nervous system. J. Neurosci. Res. 18: 155–171, 132–133. Giulian, D., J. Chen, J. E. Ingeman, J. K. George, and M. Noponen. 1989. The role of mononuclear phagocytes in wound healing after traumatic injury to adult mammalian brain. J. Neurosci. 9: 4416–4429. Giulian, D., M. Corpuz, S. Chapman, M. Mansouri, and C. Robertson. 1993. Reactive mononuclear phagocytes release neurotoxins after ischemic and traumatic injury to the central nervous system. J. Neurosci. Res. 36: 681–693. Giulian, D., J. Li, S. Bartel, J. Broker, X. Li, and J. B. Kirkpatrick. 1995. Cell surface morphology identifies microglia as a distinct class of mononuclear phagocyte. J. Neurosci. 15: 7712–7726. Giulian, D., and C. Robertson. 1990. Inhibition of mononuclear phagocytes reduces ischemic injury in the spinal cord. Ann. Neurol. 27: 33–42. Goldberger, M. E., B. S. Bregman, C. J. Vierck, Jr., and M. Brown. 1990. Criteria for assessing recovery of function after spinal cord injury: behavioral methods. Exp. Neurol. 107: 113–117. Grill, R., K. Murai, A. Blesch, F. H. Gage, and M. H. Tuszynski. 1997. Cellular delivery of neurotrophin-3 promotes corticospinal axonal growth and partial functional recovery after spinal cord injury. J. Neuroscience 17: 5560–5572. Guth, L., C. P. Barrett, E. J. Donati, F. D. Anderson, M. V. Smith, and M. Lifson. 1985. Essentiality of a specific cellular terrain for growth of axons into a spinal cord lesion. Exp. Neurol. 88: 1–12. Guth, L., Z. Zhang, N. A. DiProspero, K. Joubin, and M. T. Fitch. 1994. Spinal cord injury in the rat: Treatment with bacterial lipopolysaccharide and indomethacin enhances cellular repair and locomotor function. Exp. Neurol. 126: 76–87. Guth, L., Z. Zhang, and E. Roberts. 1994. Key role for pregnenolone in combination therapy that promotes recovery after spinal cord injury. Proc. Natl. Acad. Sci. USA 91: 12308–12312. Hirschberg, D. L., and M. Schwartz. 1995. Macrophage recruitment to acutely injured central nervous system is inhibited by a resident factor: Basis for an immune–brain barrier. J. Neuroimmunol. 61: 89–96. Huitinga, I., J. G. Damoiseaux, N. van Rooijen, E. A. Do¨pp, and C. D. Dijkstra. 1992. Liposome mediated affection of monocytes. Immunobiology 185: 11–19. Huitinga, I., N. van Rooijen, C. J. de Groot, B. M. Uitdehaag, and C. D. Dijkstra. 1990. Suppression of experimental allergic encephalomyelitis in Lewis rats after elimination of macrophages. J. Exp. Med. 172: 1025–1033.
MACROPHAGE DEPLETION AND SPINAL CORD INJURY 51.
Kerasidis, H., J. R. Wrathall, and K. Gale. 1987. Behavioral assessment of functional deficit in rats with contusive spinal cord injury. J. Neurosci. Methods 20: 167–189. Kunkel-Bagden, E., H. N. Dai, and B. S. Bregman. 1993. Methods to assess the development and recovery of locomotor function after spinal cord injury in rats. Exp. Neurol. 119: 153–164. Lazarov-Spiegler, O., O. Rapalino, G. Agranov, and M. Schwartz. 1998. Restricted inflammatory response in the CNS: A key impediment to axonal regeneration? Mol. Med. Today 4: 337– 342. Lazarov-Spiegler, O., A. S. Solomon, A. B. Zeev-Brann, D. L. Hirschberg, V. Lavie, and M. Schwartz. 1996. Transplantation of activated macrophages overcomes central nervous system regrowth failure. FASEB J. 10: 1296–1302. McTigue, D. M., P. J. Horner, B. T. Stokes, and F. H. Gage. 1998. Neurotrophin-3 and brain-derived neurotrophic factor induce oligodendrocyte proliferation and myelination of regenerating axons in the contused adult rat spinal cord. J. Neurosci. 18: 5354–5365. McTigue, D. M., M. Tani, K. Krivacic, A. Chernosky, G. S. Kelner, D. Maciejewski, R. Maki, R. M. Ransohoff, and B. T. Stokes. 1998. Selective chemokine mRNA accumulation in the rat spinal cord after contusion injury. J. Neurosci. Res. 53: 368–376. Means, E. D., and D. K. Anderson. 1983. Neuronophagia by leukocytes in experimental spinal cord injury. J. Neuropathol. Exp. Neurol. 42: 707–719. Micklus, M. J., N. H. Greig, J. Tung, and S. I. Rapoport. 1992. Organ distribution of liposomal formulations following intracarotid infusion in rats. Biochim. Biophys. Acta 1124: 7–12. Mosley, K., and M. L. Cuzner. 1996. Receptor-mediated phagocytosis of myelin by macrophages and microglia: effect of opsonization and receptor blocking agents. Neurochem. Res. 21: 481– 487. Noble, L. J., and J. R. Wrathall. 1989. Distribution and time course of protein extravasation in the rat spinal cord after contusive injury. Brain Res. 482: 57–66. Popovich, P. G., P. J. Horner, B. B. Mullin, and B. T. Stokes. 1996. A quantitative spatial analysis of the blood–spinal cord barrier. I. Permeability changes after experimental spinal contusion injury. Exp. Neurol. 142: 258–275. Popovich, P. G., J. F. Reinhard, Jr., E. M. Flanagan, and B. T. Stokes. 1994. Elevation of the neurotoxin quinolinic acid occurs following spinal contusion injury. Brain Res. 633: 348–352. Popovich, P. G., P. Wei, and B. T. Stokes. 1997. The cellular inflammatory response after spinal cord injury in Sprague– Dawley and Lewis rats. J. Comp. Neurol. 377: 443–464. Popovich, P. G., J. Y. Yu, and C. C. Whitacre. 1997. Spinal cord neuropathology in rat experimental autoimmune encephalomyelitis: Modulation by oral administration of myelin basic protein. J. Neuropathol. Exp. Neurol. 56: 1323–1338. Qian, Q., M. A. Jutila, N. van Rooijen, and J. E. Cutler. 1994. Elimination of mouse splenic macrophages correlates with increased susceptibility to experimental disseminated candidiasis. J. Immunol. 152: 5000–5008. Rabchevsky, A. G., and W. J. Streit. 1997. Grafting of cultured microglial cells into the lesioned spinal cord of adult rats enhances neurite outgrowth. J. Neurosci. Res. 47: 34–48. Rapalino, O., O. Lazarov-Spiegler, E. Agranov, G. J. Velan, E. Yoles, M. Fraidakis, A. Solomon, R. Gepstein, A. Katz, A. Belkin, et al. 1998. Implantation of stimulated homologous macrophages results in partial recovery of paraplegic rats. Nature Med. 4: 814–821.
Schroeter, M., S. Jander, I. Huitinga, O. W. Witte, and G. Stoll. 1997. Phagocytic response in photochemically induced infarction of rat cerebral cortex: The role of resident microglia. Stroke 28: 382–386. Semple-Rowland, S. L., A. Mahatme, P. G. Popovich, D. A. Green, G. Hassler, B. T. Stokes, and W. J. Streit. 1995. Analysis of TGF-␤1 gene expression in contused rat spinal cord using quantitative RT-PCR. J. Neurotrauma 12: 1003–1014. Shuman, S. L., J. C. Bresnahan, and M. S. Beattie. 1997. Apoptosis of microglia and oligodendrocytes after spinal cord contusion in rats. J. Neurosci. Res. 50: 798–808. Stokes, B. T., D. H. Noyes, and D. L. Behrmann. 1992. An electromechanical spinal injury device with dynamic sensitivity. J. Neurotrauma 9: 187–195. Streit, W. J., S. L. Semple-Rowland, S. D. Hurley, R. C. Miller, P. G. Popovich, and B. T. Stokes. 1998. Cytokine mRNA profiles in contused spinal cord and axotomized facial nucleus suggest a beneficial role for inflammation and gliosis. Exp. Neurol. 152: 74–87. Taoka, Y., K. Okajima, M. Uchiba, K. Murakami, N. Harada, M. Johno, and M. Naruo. 1998. Activated protein C reduces the severity of compression-induced spinal cord injury in rats by inhibiting activation of leukocytes. J. Neurosci. 18: 1393–1398. Taoka, Y., K. Okajima, M. Uchiba, K. Murakami, S. Kushimoto, M. Johno, M. Naruo, H. Okabe, and K. Takatsuki. 1997. Role of neutrophils in spinal cord injury in the rat. Neuroscience 79: 1177–1182. van Rooijen, N., and A. Sanders. 1994. Liposome mediated depletion of macrophages: Mechanism of action, preparation of liposomes and applications. J. Immunol. Methods 174: 83–93. van Rooijen, N., A. Sanders, and T. K. van den Berg. 1996. Apoptosis of macrophages induced by liposome-mediated intracellular delivery of clodronate and propamidine. J. Immunol. Methods 193: 93–99. van Rooijen, N., and R. van Nieuwmegen. 1984. Elimination of phagocytic cells in the spleen after intravenous injection of liposome-encapsulated dichloromethylene diphosphonate: An enzyme-histochemical study. Cell Tissue Res 238: 355–358. Windle, W. F., J. Smart, and J. Beers. 1958. Residual function after subtotal spinal cord transection in adult cats. Neurology 8: 518–521. Wrathall, J. R., D. Choiniere, and Y. D. Teng. 1994. Dosedependent reduction of tissue loss and functional impairment after spinal cord trauma with the AMPA/kainate antagonist NBQX. J. Neurosci. 14: 6598–607. Zeev-Brann, A. B., O. Lazarov-Spiegler, T. Brenner, and M. Schwartz. 1998. Differential effects of central and peripheral nerves on macrophages and microglia. GLIA 23: 181–190. Zhang, S. C., and S. Fedoroff. 1996. Neuron–microglia interactions in vitro. Acta Neuropathol. (Berl.) 91: 385–395. Zhang, Z., and L. Guth. 1997. Experimental spinal cord injury: Wallerian degeneration in the dorsal column is followed by revascularization, glial proliferation, and nerve regeneration. Exp. Neurol. 147: 159–171. Zhang, Z., L. Guth, and O. Steward. 1998. Mechanisms of motor recovery after subtotal spinal cord injury: Insights from the study of mice carrying a mutation (WldS) that delays cellular responses to injury. Exp. Neurol. 149: 221–229. Zhang, Z., C. J. Krebs, and L. Guth. 1997. Experimental analysis of progressive necrosis after spinal cord trauma in the rat: etiological role of the inflammatory response. Exp. Neurol. 143: 141–152.