Brain Advance Access originally published online on February 2, 2005
Brain 2005 128(4):854-866; doi:10.1093/brain/awh407
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Involvement of monocyte chemoattractant protein-1, macrophage inflammatory protein-1
and interleukin-1ß in Wallerian degeneration
Centre for Research in Neuroscience, McGill University Health Center, Montreal, Quebec, Canada
Correspondence to: Dr Samuel David, Centre for Research in Neuroscience, Montreal General Hospital Research Institute, Livingston Hall, Room L7-210, 1650 Cedar Ave, Montreal, Quebec, Canada, H3G 1A4 E-mail: sam.david{at}mcgill.ca
.
Received September 13, 2004. Revised December 19, 2004. Accepted December 22, 2004.
 |
Summary |
|---|
 Top Summary IntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Wallerian degeneration in the CNS and PNS consists of degradation and phagocytosis of axons and their myelin sheath distal to the site of injury. This process of degeneration, which requires an effective macrophage response, occurs rapidly in peripheral nerves but is slow in the CNS. Rapid Wallerian degeneration in peripheral nerves may contribute to subsequent axonal regeneration. We show that there is a marked increase in mRNA expression of three pro-inflammatory molecules, the chemokines monocyte chemoattractant protein-1 (MCP-1) and macrophage inflammatory protein-1
(MIP-1), and the cytokine interleukin-1ß (IL-1ß), in the mouse sciatic nerve but not in the spinal cord undergoing Wallerian degeneration. Neutralizing MCP-1, MIP-1 and IL-1ß in the lesioned sciatic nerve with function-blocking antibodies suppressed macrophage responses and myelin clearance. Injecting recombinant MCP-1, MIP-1 or IL-1ß into the normal, uninjured spinal cord triggered the expression of a number of chemokines and cytokines. Furthermore, injecting recombinant MCP-1/MIP-1 or IL-1ß into the dorsal column of the spinal cord undergoing Wallerian degeneration triggered rapid macrophage/microglial activation and myelin clearance. These findings provide direct evidence that MCP-1, MIP-1 and IL-1ß are important regulators of macrophage responses that lead to rapid myelin breakdown and clearance in Wallerian degeneration.
Key Words: Wallerian degeneration; spinal cord injury; phagocytosis; macrophage; axonal injury
Abbreviations:
GM-CSF = granulocyte–macrophage colony-stimulating factor; IFN-
= interferon-; IL-1ß = interleukin-1ß; IL-10 = interleukin-10; MCP-1 = monocyte chemoattractant protein-1; MIP-1 = macrophage inflammatory protein-1; RANTES = regulated upon activation, normal T cell expressed and secreted; RPA = RNase protection assay; TGF-ß1 = transforming growth factor-ß1; TNF- = tumour necrosis factor-
|
Introduction |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Wallerian degeneration, which consists of the breakdown and phagocytosis of damaged axons and their myelin sheaths distal to the site of injury, was first described >150 years ago (Waller, 1850
). This type of degeneration is remarkably slow in the mammalian CNS, and takes several months to complete (Perry et al., 1987; Stoll et al., 1989b; George and Griffin, 1994; Lawson et al., 1994; Buss and Schwab, 2003). The initial axonal breakdown occurs rapidly, undergoing granular degeneration of cytoskeletal structures via the action of proteases (Waller et al., 1991; George and Griffin, 1994; Coleman and Perry, 2002). However, the removal of the axonal and myelin debris is very slow. The clearance of myelin after injury may be important for axon regeneration in the CNS because of the presence of axon growth inhibitors in myelin (Bandtlow and Schwab, 2000; David and Lacroix, 2003) one of which, Nogo-A, can be detected 2 months after lesion in spinal cord white matter tracts undergoing Wallerian degeneration (Buss and Schwab, 2003). The slow rate of Wallerian degeneration in the CNS is reflected by an equally slow macrophage response, which occurs gradually over a period of several months (Perry et al., 1987; Stoll et al., 1989b; George and Griffin, 1994; Buss and Schwab, 2003). In marked contrast to the CNS, Wallerian degeneration occurs rapidly in the PNS. The macrophage response required to clear myelin from the distal portion of damaged peripheral nerves is well underway in the first week (Perry et al., 1987; Bruck, 1997; Bendszus and Stoll, 2003; Mueller et al., 2003), and myelin and axonal debris are cleared within 7–14 days (Stoll et al., 1989a; Bruck, 1997). Macrophages in degenerating peripheral nerves are mainly of haematogenous origin (Bendszus and Stoll, 2003; Mueller et al., 2003), while phagocytes in the degenerating CNS white matter appear to be largely of microglial origin (Buss and Schwab, 2003). Understanding the factors that regulate the rapid macrophage responses in the PNS during Wallerian degeneration, and comparing the differences in the expression of these molecules in the lesioned PNS and CNS, may provide insights into the reasons for slow Wallerian degeneration in the CNS.
Chemokines and cytokines are important regulators of the immune response (Allan and Rothwell, 2001; Ransohoff, 2002). Recent studies have reported increased expression of various chemokines and cytokines in the injured PNS and CNS (Stoll and Jander, 1999; Rotshenker, 2001). Most of the studies on the spinal cord, however, have focused on changes in chemokine and cytokine expression at the site of lesion, not in distal areas undergoing Wallerian degeneration. Furthermore, direct functional assessment of the role of these molecules in vivo in Wallerian degeneration in the PNS and CNS is lacking. Therefore, involvement of these molecules and how they mediate these changes remains to be fully understood.
We now present work in which we compared the mRNA expression of a number of chemokines and cytokines in the degenerating PNS (sciatic nerve) and CNS (dorsal columns of spinal cord) using RNase protection assays (RPAs). Striking differences were observed in the expression of several of these molecules in the injured PNS and CNS, with the CNS showing very low-level expression. Infusion of function-blocking antibodies against chemokines and cytokines into the lesioned sciatic nerve, and injection of recombinant chemokines and cytokines into the spinal cord helped reveal the important role of MCP-1, MIP-1 and IL-1ß in Wallerian degeneration.
|
Material and methods |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
PNS and CNS lesions
Adult, female BALB/c mice were deeply anaesthetized with a mixture of ketamine–xylazine (50/10 mg/kg) and the sciatic nerve exposed and cut in the region of the upper thigh. In other mice, the spinal cord at the level of the 11th thoracic (T11) region was exposed and hemisected with a pair of micro-scissors to transect the ascending dorsal column sensory fibres (i.e. a dorsal hemisection). In order to ensure the complete axotomy of the dorsal column fibres, a micro-dissection knife (Fine Science Tools) was passed through the lesion site to the same depth (0.5 mm) as the hemisection injury as described previously (Huang et al., 1999
). Both groups of mice were sacrificed at various survival times ranging from 1, 3, 7, 14, 21 to 28 days. All surgical procedures were approved by the McGill University Animal Care Committee and followed the guidelines of the Canadian Council on Animal Care.
RNase protection assays
For each of the six post-lesion survival times, four mice were used for spinal cord hemisection and 12 mice were used for sciatic nerve transection. Corresponding tissue from animals that did not undergo surgery were taken as control. Experiments were repeated so that each set of RPAs was done up to four times. The RPA technique used is similar to that previously described by others (Babcock et al., 2003). Two 5 mm segments of the sciatic nerve and two 5 mm segments of the spinal cord distal to the injury were taken. One segment (‘near’ segment) extended 0–5 mm and the other (‘far’ segment) extended 10–15 mm distal to the site of injury. Because the spinal cord lesion transects the ascending sensory dorsal column fibres that have their cell bodies in the dorsal root ganglia, the ‘far’ segment in the spinal cord is 10–15 mm rostral to the injury site. Total RNA was extracted from the isolated tissues with TRIzol according to the manufacturer's instructions. Chemokines and cytokines were detected using a custom-made RiboQuant Multi-probe RNase Protection Assay system (BD Biosciences, San Diego, CA). Chemokines are grouped under two main groups that differ in the presence of an amino acid between the two cysteine residues near the N-terminus, i.e. CC and CXC chemokines, these being referred to as CC ligands (CCLs) and CXC ligands (CXCLs) (Ransohoff, 2002). The template set used for RPA included the following: MCP-1/CCL2, MIP-1/CCL3, IL-1ß, IL-10, granulocyte–macrophage colony-stimulating factor (GM-CSF), tumour necrosis factor- (TNF-), regulated upon activation, normal T cell expressed and secreted (RANTES/CCL5), transforming growth factor-ß1 (TGF-ß1) and interferon- (IFN-).
Riboprobes were labelled with [32P]dUTP and hybridized overnight with 8–15 µg of the RNA samples of spinal cord tissue or sciatic nerve tissue. The positive control consisted of spleen RNA from BALB/c mice infected with Plasmodium falciparum, the malaria-causing parasite (provided by Dr Mary Stevenson, McGill University). The hybridized RNA was treated with RNase and purified according to the Riboquant protocol. Resolution of the protected probes was performed in 5% polyacrylamide–Tris–borate–EDTA–urea gels, which were then dried and imaged with a PhosphorImager (Molecular Dynamics, Sunnyvale, CA) and quantified using the ImageQuant image analysis software. The length of the respective fragments identified the cytokine transcripts. For quantification, cytokine values were normalized to those of the L32, a control gene provided with the template, or glyceraldehyde-3-phosphate dehydrogenase.
Infusion of antibodies into the sciatic nerve
Adult female BALB/c mice were deeply anaesthetized and the right sciatic nerve exposed and cut at the level of the mid-thigh. A 7 day mini-osmotic pump (Alzet; 100 µl volume) filled with function-blocking antibodies against MCP-1/MIP-1, IL-1ß (all from R&D Systems, Minneapolis, MN) or control immunoglobulin at 100 µg/ml was placed subcutaneously in the region of the back. A polyethylene tube extended from the mini-pump to the cut sciatic nerve. The cut end of the distal sciatic nerve segment was inserted into the tubing and secured in place with 10-0 Ethicon sutures, and sealed with a fibrin glue (Tisseel VH kit; Baxter Healthcare Corp., Glendale, CA). After a 7 day survival, mice were perfused with 2.5% glutaraldehyde and 0.5% paraformaldehyde in 0.1 M phosphate buffer and the distal segment of the sciatic nerve processed for embedding in Epon. Sections (1 µm thick) were stained with toluidine blue for light microscopy, and ultra-thin sections picked-up on 200 mesh copper grids were stained with lead citrate for electron microscopy. The numbers of macrophages and intact myelin sheaths in the entire cross-section of the tibial branch of the sciatic nerve were quantified under the electron microscope. Counts were made from three nerves in each of the experimental and control groups. Data are presented as means ± SEM, and the Student's t test was used to establish statistical significance.
Microinjection of recombinant proteins into the spinal cord
In the first series of experiments (n = 3), 8- to 12-week-old female BALB/c mice were deeply anaesthetized as above and the T11 spinal segment exposed. A 1 µl injection of recombinant MCP-1, MIP-1, a combination of these two chemokines (MCP-1/MIP-1) or IL-1ß (all from R & D Systems, Minneapolis, MN) at a concentration of 10 ng/µl was injected in the dorsal column of the spinal cord using a glass micropipette with a tip diameter of 50 µm. Control mice were injected with vehicle (phosphate-bufferred saline; PBS). Animals were sacrificed at 6 and 24 h after injection. A 5 mm long segment surrounding the injection site from four mice was pooled and RNA extracted as described above.
In another series of mice, the spinal cord was hemisected as described above. Five days later, a 1 µl injection of either recombinant MCP-1/MIP-1, recombinant IL-1ß or vehicle (PBS) was injected as described above into the dorsal column, 10 mm rostral to the hemisection, i.e. within the region in which the ascending sensory fibres will be undergoing Wallerian degeneration. Nine days later, i.e. 2 weeks after the hemisection, the mice were perfused with 4% paraformaldehyde and longitudinal cryostat sections (14 µm) of the spinal cord obtained for immunohistochemistry with the Mac-1 monoclonal antibody (American Type Culture Collection) that recognizes macrophages. Immunohistochemistry was done using standard protocols (Ousman and David, 2000). Tissue sections were counterstained with methyl green to stain cell nuclei. Counts of the number of Mac-1+ round macrophages (phagocytic macrophages) and short process-bearing microglia (activated microglia) were made from the white matter 250 µm adjacent to the injection site at 25x magnification using an ocular grid as previously described (Ousman and David, 2000). Only cells containing a cell nucleus were counted. A total of three mice were used for each of the experimental and control groups. Counts for each mouse were obtained from three tissue sections that were 45 µm apart. Some of the tissue sections were stained with Luxol fast blue to visualize myelin.
|
Results |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Expression of chemokines and cytokines in the injured mouse PNS and CNS
As chemokines and cytokines are important regulators of the immune response, we examined the mRNA expression of nine of these molecules in the transected adult mouse sciatic nerve and the spinal cord after dorsal hemisection at the lower thoracic region. This type of spinal cord lesion results in Wallerian degeneration of the ascending dorsal column fibres. The nine mRNAs studied were MCP-1/CCL2, MIP-1
/CCL3, IL-1ß, IL-10, GM-CSF, TNF-, RANTES/CCL5, TGF-1 and IFN-. Expression of mRNA was assessed using a multi-probe RPA, which allowed all nine mRNAs to be examined together. Since Wallerian degeneration is well underway in the first week in the sciatic nerve and occurs gradually over a period of 1 month in the spinal cord, we examined the expression of these chemokines and cytokines from 1 to 28 days after sciatic and spinal cord injury. Change in mRNA expression was assessed in two 5 mm distal segments of the lesioned sciatic nerve and spinal cord undergoing Wallerian degeneration: (i) a ‘far’ segment, between 10 and 15 mm distal to the injury; and (ii) a ‘near’ segment, between 0 and 5 mm distal to the lesion. IFN- and GM-CSF mRNA expression was not detected in either the lesioned sciatic nerve or spinal cord but was detected in the positive control which consisted of mRNA from the spleen of BALB/c mice infected with P. falciparum, the malaria-producing parasite. The lack of detection of GM-CSF in the lesioned sciatic nerve may be due to the sensitivity of the technique used, as earlier studies have shown it is expressed in Wallerian degeneration (Saada et al., 1996; Be'eri et al., 1998). The remaining seven chemokines and cytokines were differentially expressed.
Expression in the injured sciatic nerve
‘Far’ sciatic nerve segment
Changes in gene expression in this segment will reflect changes associated with Wallerian degeneration occurring at a distance from the lesion site (10–15 mm). A rapid increase in the mRNA expression of chemokines occurs by 1 day after injury (Fig. 1A). A striking feature of this early peak in expression at day 1 is the high level of expression of two pro-inflammatory chemokines (MCP-1 and MIP-1) and two anti-inflammatory cytokines (TGF-ß1 and IL-10). MCP-1 and MIP-1 show 5- and 2.5-fold increases, respectively. On the other hand, the anti-inflammatory cytokines, TGF-ß1 and IL-10, increase 
2- and 9-fold, respectively. The differences in these fold changes should be interpreted with caution, as the amount of each mRNA and the ultimate expression of the protein are critical factors. Nevertheless, these results indicate that there are marked increases in mRNA expression of both pro- and anti-inflammatory genes. Others such as IL-1ß, TNF- and RANTES mRNA show much less of an increase. The mRNA levels of these chemokines and cytokines decrease at 3 and 7 days post-injury before rising to a second peak at day 14. The prominent feature of the second peak is the 10-fold increase in IL-ß mRNA. In contrast to their levels in the first peak, the increase in MCP-1 in the second peak is much lower, while MIP-1 remains unchanged. The levels of all these mRNAs return to normal values between days 21 and 28.
|
‘Near’ sciatic nerve segment
Changes in expression in this segment reflect changes associated with Wallerian degeneration occurring nearer the site of lesion (0–5 mm). The ‘near’ segment also shows a rapid increase in expression of chemokines and cytokines at day 1 post-lesion (Fig. 1B). The striking difference in this segment at day 1 compared with the ‘far’ segment is the 37-fold increase in IL-1ß mRNA. The increase in MCP-1 mRNA is similar to that in the ‘far’ segment, but the level of MIP-1
mRNA is greater. The mRNAs for the two anti-inflammatory cytokines TGF-ß1 and IL-10 increase 4- to 5-fold. In contrast, TNF- and RANTES mRNA expression changed very slightly at day 1. The mRNA levels of many of these chemokines and cytokines return towards control levels after day 3. However, MCP-1 and IL-1ß mRNA levels were still elevated by 2-fold at day 28. These data show that there is a marked increase in the mRNA expression of pro-inflammatory chemokines MCP-1 and MIP-1 and the pro-inflammatory cytokine IL-1ß in the first 3 days after sciatic nerve injury.
Expression in the lesioned spinal cord
A dorsal hemisection at the level of lower thoracic spinal cord (T11) was done to induce Wallerian degeneration in the ascending dorsal column sensory fibres, which extend up to the dorsal column sensory nuclei in the medulla. In contrast to the degenerating peripheral nerve, there was minimal expression of chemokines and cytokines in the degenerating spinal cord as detected with RPA.
‘Far’ spinal cord segment
This segment of the spinal cord containing dorsal column fibres undergoing Wallerian degeneration is located 10–15 mm rostral to the hemisection. MIP-1 mRNA reaches a peak only at day 7, while MCP-1 mRNA shows an even longer delay, with its highest level at day 28 (Fig. 1C). These increases in MCP-1 and MIP-1 mRNA are much lower than that in the ‘far’ segment of the sciatic nerve, being 20- and 5-fold lower, respectively. There is no increase in IL-1ß and TNF- mRNA at any of the time points. Neither is there any significant change in the mRNA for the two anti-inflammatory cytokines TGF-ß1 and IL-10. An important caveat in interpreting these data is that since only the dorsal half of the spinal cord was lesioned (dorsal hemisection), there could be a diluting effect because RNA was extracted from the entire cross-section of the 5 mm segments of the spinal cord. However, even if one were to assume a 2- to 3-fold dilution, the differences with the sciatic nerve are in the 10-fold range, suggesting that the mRNA levels in the lesioned spinal cord are likely to be much lower than in the lesioned sciatic nerve.
‘Near’ spinal cord segment
There is a rapid increase in the expression of MCP-1 and MIP-1 by day 1 in this segment that includes the site of hemisection (Fig. 1D). However, these increases in MCP-1 and MIP-1 mRNA are still 4.5- and 13-fold less than in the ‘near’ segment of the sciatic nerve. A delayed 2-fold increase in IL-1ß mRNA occurs between days 7 and 14. However, the peak level of IL-1ß is still 20-fold less than that in the ‘near’ segment of the sciatic nerve. There are no significant changes in TNF-, TGF-ß1 and IL-10 mRNA.
These results show that in the CNS in which Wallerian degeneration is very slow, there is only a minimal increase in chemokine and cytokine mRNA expression in areas undergoing Wallerian degeneration. In contrast, in injured peripheral nerves in which Wallerian degeneration is very rapid, there is a marked increase in mRNA expression of pro-inflammatory molecules, particularly of MCP-1, MIP-1 and IL-1ß. The presence of mRNA does not necessarily mean that the protein is expressed; however, previous immunohistochemical studies have reported increased expression of MCP-1 and MIP-1 (Taskinen and Roytta, 2000) and IL-1ß (Shamash et al., 2002) protein in the injured sciatic nerve.
Role of MCP-1/MIP-1 and IL-1ß in Wallerian degeneration in the sciatic nerve
We next assessed the role of the three pro-inflammatory molecules which are rapidly expressed at high levels in the degenerating peripheral nerve, namely the two chemokines MCP-1 and MIP-1 which are highly expressed in both the proximal and distal nerve segment, and IL-1ß which is most prominently expressed in the proximal nerve segment. The role of these molecules in Wallerian degeneration was assessed by infusing function-blocking antibodies against MCP-1 and MIP-1 or IL-1ß into the distal segment of the cut sciatic nerve using an Alzet mini-osmotic pump. Continuous infusion of these antibodies was carried out over a 7 day period. The number of macrophages and intact myelin profiles in cross-sections of the nerve was quantified by electron microscopy.
The number of macrophages in the sciatic nerve is reduced in mice treated with antibodies against MCP-1/MIP-1 or IL-1ß. The reduction in the number of macrophages is greater in mice treated with anti-MCP-1/MIP-1 than in mice treated with anti-IL-1ß, being reduced by 78 and 42%, respectively (Fig. 2A). These results suggest that the chemokines MCP-1 and MIP-1 and to a lesser extent the cytokine IL-1ß play a role in recruiting macrophages into the lesioned nerve, which previous studies have reported are mainly derived from the circulation. The effect of IL-1ß on macrophage recruitment may be mediated indirectly via increasing chemokine expression (see below). In control lesioned nerves, 85% of the macrophages appear to be phagocytic, while only 50% are phagocytic in nerves treated with anti-MCP-1/MIP-1, and 45% are phagocytic in nerves treated with anti-IL-1ß. The total number of phagocytic macrophages per mm2 is therefore reduced by 87% in anti-MCP-1/MIP-1-treated nerves, and by 69% in anti-IL-1ß-treated nerves (Fig. 2A). The severe reduction in the number of phagocytic macrophages in the anti-MCP-1/MIP-1-treated mice is accompanied by a 14-fold increase in the number of intact myelin sheaths, and a 6-fold increase in these profiles in anti-IL-1ß-treated nerves (Fig. 2B). Granular degenerative changes in the axon are seen in both antibody-treated groups, which is consistent with previous reports that axonal breakdown is independent of the macrophage response (Fig. 2F and I). The electron micrographs show that most of the axons and their myelin sheaths in control lesioned nerves are degraded and phagocytosed by macrophages by day 7 (Fig. 2C–E). However, many intact myelin profiles are present in nerves treated with anti-IL-1ß (Fig. 2F–H) or anti-MCP-1/MIP-1 (Fig. 2I–K). In anti-IL-1ß- and anti-MCP-1/MIP-1-treated mice, macrophages devoid of phagocytosed material were frequently found adjacent to degenerating myelin profiles (Fig. 2F–H and I–K). These results provide direct evidence that the chemokines, MCP-1 and MIP-1, and the cytokine IL-1ß play a crucial role in vivo in macrophage recruitment and activation, as well as in myelin breakdown and phagocytosis in lesioned peripheral nerves.
|
Chemokine/cytokine networks in the CNS
The expression studies detailed above show a strong correlation between poor expression of pro-inflammatory chemokines and cytokines and slow Wallerian degeneration in the spinal cord. It is possible that induction of one or a limited number of chemokines or cytokines in the spinal cord may be sufficient to trigger the coordinated expression of other chemokines and cytokines. The lack of such an initial trigger may account for the blunted expression of these molecules in the spinal cord undergoing Wallerian degeneration. We therefore assessed the effects of the three pro-inflammatory molecules (MCP-1, MIP-1
and IL-1ß), which were found to be important in the lesioned sciatic nerve. As our focus in these experiments was on pro-inflammatory molecules, we did not assess the effects of TGF-ß1 and IL-10 because of their anti-inflammatory properties.
Recombinant MCP-1 and MIP-1 either separately or in combination (MCP-1/MIP-1), recombinant IL-1ß, or vehicle (PBS) as control were microinjected into the normal mouse spinal cord at the lower thoracic level. The spinal cord tissue at the injection site was then assessed 6 and 24 h later for expression of the nine chemokines and cytokines using RPAs as detailed above. Changes in expression at each time point in the recombinant protein-injected groups were compared with that in the PBS-injected controls. As in Wallerian degeneration, IFN- and GM-CSF mRNA expression was not detected.
Effects of MCP-1 and MIP-1
Injection of recombinant MCP-1 results in an 2-fold increase in TGF-ß1 and RANTES mRNA at 24 h compared with PBS-injected controls (Fig. 3). Recombinant MIP-1 on the other hand increases IL-1ß mRNA between 2- and 3-fold at 6 h (Fig. 3), and IL-10 mRNA by 5-fold at 6 h (Fig. 3). The combination of recombinant MCP-1 and MIP-1 was much more effective in inducing higher expression of several mRNAs at 6 h, including MCP-1, MIP-1, IL-1ß, RANTES and IL-10, compared with either chemokine alone (Fig. 3).
|
Effects of IL-1ß
Recombinant IL-1ß induced a marked increase in the expression of both chemokine mRNAs compared with PBS-injected controls. MCP-1 mRNA was elevated by almost 12-fold, while MIP-1
mRNA was elevated by 5-fold at 6 h post-injection (Fig. 3). IL-1ß also increased RANTES expression 5-fold at 6 h post-injection. The expression of the two anti-inflammatory cytokines, IL-10 and TGF-ß, showed a pronounced increase, with IL-10 mRNA levels being raised by 15-fold at 6 h, and TGF-ß by 12- and 3-fold at 6 and 24 h post-injection, respectively.
These data indicate that injections of MCP-1, MIP-1 or IL-1ß can induce the expression of other cytokines and chemokines that could trigger complex cellular responses. In particular, we show that the combination of MCP-1/MIP-1 induces rapid mRNA expression of several pro-inflammatory cytokines and chemokines, including IL-1ß, MCP-1 and MIP-1, while injection of IL-1ß induces a marked increase in the mRNA expression of MCP-1 and to a lower extent MIP-1. IL-1ß and the combination of MCP-1/MIP-1 also increase mRNA expression of the anti-inflammatory cytokines IL-10 and TGF-ß1. Early expression of sufficient levels of these two chemokines (MCP-1 and MIP-1) or IL-1ß in the spinal cord during Wallerian degeneration may therefore be sufficient to trigger the expression of various pro-inflammatory and anti-inflammatory chemokines and cytokines required for rapid recruitment and activation of macrophages, and rapid myelin clearance.
MCP-1/MIP-1 and IL-1ß induce rapid recruitment and activation of macrophages and rapid myelin phagocytosis in spinal cord white matter undergoing Wallerian degeneration
We next assessed whether intra-spinal injections of MCP-1/MIP-1 or IL-1ß into the dorsal column white matter undergoing Wallerian degeneration will stimulate macrophage recruitment and activation, and promote rapid myelin clearance. A single microinjection of recombinant MCP-1/MIP-1 or IL-1ß was made 10 mm rostral (distal) to the site of dorsal hemisection at the lower thoracic level 5 days after hemisection. Changes in the number of macrophages and activated microglia were assessed 9 days later using Mac-1 immunohistochemistry, i.e. 2 weeks after hemisection.
The number of large, round, Mac-1+ macrophages, which previously have been shown to be phagocytic macrophages (Ousman and David, 2000), almost doubled in the degenerating spinal cord near the site of injection of MCP-1/MIP-1 and IL-1ß compared with vehicle-injected controls (Fig. 4A). The differences between the MCP-1/MIP-1-treated and IL-1ß-treated groups are not statistically significant (Fig. 4A). Haematogenous macrophages as well as microglia may contribute to these large, round Mac-1+ macrophages. Unlike phagocytic macrophages, which are round in shape, activated microglial cells that have not yet become fully phagocytic have a ramified morphology with short cytoplasmic processes as characterized by Perry et al. (1993), and have strong Mac-1 immunoreactivity (Ousman and David, 2000). In contrast to these microglial cells, quiescent microglia have weak Mac-1 immunoreactivity and long branching processes (Perry et al., 1993; Ousman and David, 2000). Injection of recombinant IL-1ß doubled the number of activated microglia compared with mice receiving vehicle injections (Fig. 4B). However, recombinant MCP-1/MIP-1 failed to produce a significant increase in the number of activated microglia compared with vehicle-injected controls (Fig. 4B). These results suggest that IL-1ß is required for microglial activation.
|
Strong Mac-1-immunoreactive cells are found in the white matter adjacent to the site of injections of IL-1ß (Fig. 5B) and MCP-1/MIP-1
(Fig. 5C), but not in controls injected with PBS (Fig. 5A). The Mac-1 immunoreactivity in IL-1ß and MCP-1/MIP-1 spinal cords appeared to be weaker in the central core region at the injection site (Fig. 5B and C). At higher magnifications, these regions in the centre contain very large Mac-1+ cells with clear cytoplasm (Fig. 5D and E), which are characteristic of macrophages that have phagocytosed myelin (Ousman and David, 2000).
|
To assess further if the marked increase in macrophage recruitment and activation induced by IL-1ß or MCP-1/MIP-1
results in myelin clearance, adjacent tissue sections were stained for myelin using Luxol fast blue histochemistry. Myelin is cleared from the dorsal column white matter in mice injected with IL-1ß (Fig. 5G) or MCP-1/MIP-1 (Fig. 5H). These myelin-free regions correspond to areas containing large round Mac-1+ macrophages. In contrast, similarly lesioned spinal cords undergoing Wallerian degeneration injected with vehicle showed no detectable change in Luxol fast blue staining (Fig. 5F), indicating lack of myelin clearance. Axon regeneration could not be assessed in this model as the area of MCP-1/MIP-1- and IL-1ß-induced myelin clearance in the dorsal column is 10 mm away from the site of hemisection at which the cut end of the axons are located. Myelin is present between these two sites and would therefore continue to inhibit axon growth. |
Discussion |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
In this study, we assessed the changes in the mRNA expression of nine chemokines and cytokines in the adult mouse sciatic nerve and spinal cord undergoing Wallerian degeneration. The expression of the two chemokines, MCP-1 and MIP-1
, and the cytokine IL-1ß was markedly increased in both segments of the sciatic nerve distal to the injury. In addition to these pro-inflammatory molecules, two anti-inflammatory molecules, IL-10 and TGF-ß, also showed very high levels of expression in the injured sciatic nerve. In contrast to the sciatic nerve, the expression of all these molecules was markedly less in the spinal cord during Wallerian degeneration. The functional importance of these molecules was established by two sets of experiments: (i) infusion of function-blocking antibodies against MCP-1/MIP-1 and IL-1ß into the cut sciatic nerve blocked Wallerian degeneration, as judged by the reduction in the recruitment and activation of macrophages, and delayed myelin phagocytosis; and (ii) injection of recombinant MCP-1, MIP-1 or IL-1ß into the spinal cord white matter undergoing Wallerian degeneration led to rapid recruitment and activation of macrophages and enhanced myelin clearance.
Chemokine and cytokine involvement in Wallerian degeneration in the PNS
An important early feature of Wallerian degeneration in peripheral nerves is the rapid recruitment of macrophages into the nerve as early as 1–3 days after injury, which reach maximum numbers by day 14 (Bendszus and Stoll, 2003; Mueller et al., 2003). We therefore assessed the expression of two chemokines, MCP-1 and MIP-1, that have potent monocyte/macrophage chemotactic and recruitment properties (Adams and Lloyd, 1997; Ousman and David, 2001; Babcock et al., 2003). The expression of these chemokines reached a peak at day 1, which is in keeping with their role in recruiting monocytes. Expression of MCP-1 and MIP-1 was markedly increased in both the ‘near’ and ‘far’ nerve segments, with the increase in MIP-1 mRNA being greater in the ‘near’ segment that includes the lesion site. The higher level of MIP-1 mRNA expression in the ‘near’ sciatic nerve segment correlates with the rapid and greater recruitment of macrophages to the site of injury compared with more distal regions of the nerve (Leskovar et al., 2000).
Although previous studies have reported the expression of MCP-1 and MIP-1 in injured peripheral nerve (Toews et al., 1998; Taskinen and Roytta, 2000; Subang and Richardson, 2001), direct evidence for their role in lesioned nerves in vivo was lacking. To obtain direct evidence for the involvement of these chemokines in macrophage responses in the injured sciatic nerve, we carried out function-blocking experiments in vivo. The 78% reduction in the number of macrophages in nerves treated with anti-MIP-1 and MCP-1 antibodies indicates that these chemokines stimulate substantial macrophage recruitment into the degenerating sciatic nerve. This effect is not due to a reduction in cell proliferation because these macrophages are largely of haematogenous origin and therefore post-mitotic (Bendszus and Stoll, 2003; Mueller et al., 2003). In addition, MCP-1 and MIP-1 are also capable of either directly or indirectly stimulating macrophage activation as judged by their phagocytic activity. Blocking MCP-1 and MIP-1 with antibodies resulted in only 50% of the macrophages being phagocytic compared with 85% of phagocytic macrophages in control lesioned nerves. The combined effect of blocking these chemokines on macrophage recruitment and activation resulted in an 87% reduction in the number of phagocytic macrophages in the sciatic nerve.
Previous studies have shown an early increase in IL-1ß mRNA by RT–PCR (revese transcription–polymerase chain reaction) and IL-1ß protein by immunohistochemistry in crushed sciatic nerves (Gillen et al., 1998; Shamash et al., 2002). Our RPA results now show that in the ‘near’ nerve segment, the IL-1ß mRNA level at 1 day post-injury was the highest of all the chemokines and cytokines assessed. The high level of expression of this potent macrophage activator in the proximal segment correlates with the greater number of phagocytic macrophages near the site of lesion (Leskovar et al., 2000). In addition, the ‘far’ nerve segment undergoing Wallerian degeneration showed two peaks of IL-1ß mRNA expression, consisting of a lower first peak at day 1 and a more prominent second peak at day 14. The higher level of IL-1ß mRNA in the second peak at day 14 correlates with the time when phagocytic macrophages are at a maximum (Bendszus and Stoll, 2003; Mueller et al., 2003). Since Schwann cells and macrophages express IL-1ß (Shamash et al., 2002), the two peaks of IL-1ß probably reflect expression first by Schwann cells and later by Schwann cells and macrophages in the nerve. Our antibody-blocking studies show that anti-IL-1ß antibody causes a reduction in macrophage number in the degenerating nerve; however, this is less than in anti-MCP-1/MIP-1-treated nerves, suggesting that the two chemokines are more effective than IL-1ß in recruiting haematogenous macrophages. IL-1ß stimulates macrophage activation as much as the two chemokines as the antibody-blocking experiments show a similar decline in the percentage (50%) of phagocytic macrophages in the two groups.
TNF- and RANTES also showed smaller increases as early as day 1 in the ‘near’ nerve segment. RANTES was not shown to play a critical role in macrophage influx after entorhinodentate lesions (Babcock et al., 2003). Increases in TNF- mRNA and protein after peripheral nerve injury have been reported (Stoll et al., 1993; Wagner and Myers, 1996; Taskinen et al., 2000; Shamash et al., 2002), and studies on TNF- null mice suggest that it plays a role in macrophage recruitment but not activation (Liefner et al., 2000). The small change in TNF- mRNA compared with changes in MCP-1 and MIP-1 in the ‘near’ sciatic nerve segment in the present study suggests that the two chemokines are likely to play a more significant role in macrophage recruitment than TNF-.
Increased expression of the anti-inflammatory cytokine IL-10, determined by RT–PCR and in situ hybridization, has been reported after sciatic nerve injury (Jander et al., 1996; Be'eri et al., 1998; Gillen et al., 1998; Taskinen et al., 2000). Our results indicate that IL-10 and TGF-ß, another cytokine with anti-inflammatory properties, show very high levels of mRNA expression compared with the pro-inflammatory chemokines and cytokines in both sciatic nerve segments after injury. If these changes in mRNA expression lead to changes in protein expression, then these anti-inflammatory cytokines could play a role in restraining the pro-inflammatory changes in the injured nerve and help in sculpting a safe immune cell response.
Chemokine and cytokine involvement in Wallerian degeneration in the CNS
The ‘near’ spinal cord segment, which contains the site of hemisection, showed a rapid increase in expression of MCP-1 and MIP-1 by day 1. Although it may not be possible to compare expression levels directly at the site of injury in the spinal cord and sciatic nerve, the level of expression appears to be much lower in the hemisected spinal cord. More robust levels of expression of chemokines and cytokines have been reported in the first 24–48 h at the site of spinal cord contusion injury (McTigue et al., 1998; Streit et al., 1998; Lee et al., 2000; Pan et al., 2002). This difference may reflect the fact that the hemisections are small lesions that produce less damage compared with spinal cord contusion injuries. Likewise, large CNS injuries such as those to the cerebral cortex and entorhinal cortex cause a substantial increase in MCP-1 (Glabinski et al., 1996; Babcock et al., 2003), IL-1ß and TNF- (Rostworowski et al., 1997) at the site of lesion. The increased expression of MIP-1, MCP-1 and IL-1ß mRNA after spinal cord hemisection that we and others (Bartholdi and Schwab, 1997) have observed correlates with the rapid influx of immune cells and activation of macrophages at the site of lesion (Klusman and Schwab, 1997; Schnell et al., 1999a).
In contrast to the rapid recruitment of large numbers of macrophages at the site of CNS lesions (Popovich et al., 1999; Schnell et al., 1999b; Mabon et al., 2000; Ma et al., 2002), macrophages fail to be recruited rapidly into CNS white matter undergoing Wallerian degeneration (Perry et al., 1987; George and Griffin, 1994; Bendszus and Stoll, 2003; Buss and Schwab, 2003). Our results suggest that the minimal expression of MCP-1, MIP-1 and IL-1ß in the distal segment of the spinal cord may account for the lack of recruitment and activation of macrophages into degenerating white matter. We show that injection of recombinant MCP-1 and MIP-1 into spinal cord white matter undergoing Wallerian degeneration can indeed induce a rapid increase in the number of activated macrophages and rapid myelin phagocytosis. The origin of these activated macrophages is not known, but they are likely to be of monocytic and microglial origin. It has been reported by others that monocyte recruitment and myelin clearance at the site of contusion injury in the spinal cord is reduced in CCR2 chemokine receptor null mice, a major receptor for MCP-1 (Ma et al., 2002). We have also shown previously that MCP-1 and MIP-1 contribute importantly to the recruitment of monocytes and activation of macrophages in lysophosphatidylcholine-induced demyelination in mouse spinal cord (Ousman and David, 2001). These and other studies indicate that under certain conditions, peripheral macrophages can be recruited into the CNS (Graca and Blakemore, 1986; Tran et al., 1998; Hinks and Franklin, 1999; Popovich et al., 1999). In addition to their potential beneficial effects on axon regeneration by clearing myelin, macrophages have also been shown to secrete growth factors that promote remyelination (Hinks and Franklin, 2000; Kotter et al., 2001).
Injection of recombinant IL-1ß into the spinal cord undergoing Wallerian degeneration had a similar effect on macrophage recruitment as the chemokine injections. However, unlike the chemokines, IL-1ß also increased the number of activated microglia i.e. strong Mac-1-immunoreactive cells with short, cytoplasmic processes. An important finding in the present study was that microinjection of the chemokines MCP-1 and MIP-1 into the normal, unlesioned spinal cord induced expression of IL-1ß and other cytokines, and injection of IL-1ß induced a marked increase in MCP-1 and MIP-1 as well as of other cytokines. Furthermore, MCP-1, MIP-1 and IL-1ß also induced expression of the anti-inflammatory cytokines IL-10 and TGF-ß. The latter responses may underlie the protective effects noted after injection of pro-inflammatory cytokines, including IL-1ß, into spinal cord lesions (Klusman and Schwab, 1997). Cytokine networks involving pro- and anti-inflammatory cytokines have been proposed to function at the site of CNS injury (Klusman and Schwab, 1997), and in injured peripheral nerves undergoing Wallerian degeneration (Shamash et al., 2002). Our present data suggest that there is also likely to be a dynamic interaction between various chemokines and cytokines in which the expression of one regulates the expression of others. The expression profile of members of this chemokine/cytokine network will be crucial for the effective recruitment and activation of macrophages in Wallerian degeneration in the CNS and PNS, and is also likely to be important in shutting down this immune response to prevent abnormal inflammation and tissue damage.
A number of chemokines and cytokines have been suggested to play a role in Wallerian degeneration in the PNS. We now provide direct evidence from in vivo functional studies of the important role that MCP-1, MIP-1 and IL-1ß play in Wallerian degeneration in both the PNS and CNS. Future studies will address whether the rapid removal of myelin from injured CNS white matter tracts by treatment with such molecules will promote better axon regeneration.
|
Notes |
|---|
* These authors contributed equally to this work
|
Acknowledgements |
|---|
This work was supported by a grant from the Canadian Institutes of Health Research (CIHR) to S.D. F.P. received a post-doctoral fellowship from the McGill University Health Centre; and S.L. was supported by a post-doctoral fellowship from the Fonds de la Recherche en Santé du Québec (FRSQ).
|
References |
|---|
|
TopSummaryIntroductionMaterial and methodsResultsDiscussionReferences |
|---|
Adams DH, Lloyd AR. Chemokines: leucocyte recruitment and activation cytokines. Lancet 1997; 349: 490–5.[CrossRef][ISI][Medline]
Allan SM, Rothwell NJ. Cytokines and acute neurodegeneration. Nat Rev Neurosci 2001; 2: 734–44.[CrossRef][ISI][Medline]
Babcock AA, Kuziel WA, Rivest S, Owens T. Chemokine expression by glial cells directs leukocytes to sites of axonal injury in the CNS. J Neurosci 2003; 23: 7922–30.
Bandtlow CE, Schwab ME. NI-35/250/nogo-a: a neurite growth inhibitor restricting structural plasticity and regeneration of nerve fibers in the adult vertebrate CNS. Glia 2000; 29: 175–81.[CrossRef][ISI][Medline]
Bartholdi D, Schwab ME. Expression of pro-inflammatory cytokine and chemokine mRNA upon experimental spinal cord injury in mouse: an in situ hybridization study. Eur J Neurosci 1997; 9: 1422–38.[CrossRef][ISI][Medline]
Be'eri H, Reichert F, Saada A, Rotshenker S. The cytokine network of wallerian degeneration: IL-10 and GM-CSF. Eur J Neurosci 1998; 10: 2707–13.[Medline]
Bendszus M, Stoll G. Caught in the act: in vivo mapping of macrophage infiltration in nerve injury by magnetic resonance imaging. J Neurosci 2003; 23: 10892–6.
Bruck W. The role of macrophages in Wallerian degeneration. Brain Pathol 1997; 7: 741–52.[ISI][Medline]
Buss A, Schwab ME. Sequential loss of myelin proteins during Wallerian degeneration in the rat spinal cord. Glia 2003; 42: 424–32.[CrossRef][ISI][Medline]
David S, Lacroix S. Molecular approaches to spinal cord repair. Annu Rev Neurosci 2003; 26: 411–40.[CrossRef][ISI][Medline]
George R, Griffin JW. Delayed macrophage responses and myelin clearance during Wallerian degeneration in the central nervous system: the dorsal radiculotomy model. Exp Neurol 1994; 129: 225–36.[CrossRef][ISI][Medline]
Gillen C, Jander S, Stoll G. Sequential expression of mRNA for proinflammatory cytokines and interleukin-10 in the rat peripheral nervous system: comparison between immune-mediated demyelination and Wallerian degeneration. J Neurosci Res 1998; 51: 489–96.[Medline]
Glabinski AR, Balasingam V, Tani M, Kunkel SL, Strieter RM, Yong VW, et al. Chemokine monocyte chemoattractant protein-1 is expressed by astrocytes after mechanical injury to the brain. J Immunol 1996; 156: 4363–8.[Abstract]
Graca DL, Blakemore WF. Delayed remyelination in rat spinal cord following ethidium bromide injection. Neuropathol Appl Neurobiol 1986; 12: 593–605.[ISI][Medline]
Hinks GL, Franklin RJ. Distinctive patterns of PDGF-A, FGF-2, IGF-I, and TGF-beta1 gene expression during remyelination of experimentally-induced spinal cord demyelination. Mol Cell Neurosci 1999; 14: 153–68.[CrossRef][ISI][Medline]
Hinks GL, Franklin RJ. Delayed changes in growth factor gene expression during slow remyelination in the CNS of aged rats. Mol Cell Neurosci 2000; 16: 542–56.[CrossRef][ISI][Medline]
Huang DW, McKerracher L, Braun PE, David S. A therapeutic vaccine approach to stimulate axon regeneration in the adult mammalian spinal cord. Neuron 1999; 24: 639–47.[CrossRef][ISI][Medline]
Jander S, Pohl J, Gillen C, Stoll G. Differential expression of interleukin-10 mRNA in Wallerian degeneration and immune-mediated inflammation of the rat peripheral nervous system. J Neurosci Res 1996; 43: 254–9.[CrossRef][ISI][Medline]
Klusman I, Schwab ME. Effects of pro-inflammatory cytokines in experimental spinal cord injury. Brain Res 1997; 762: 173–84.[CrossRef][ISI][Medline]
Kotter MR, Setzu A, Sim FJ, Van Rooijen N, Franklin RJ. Macrophage depletion impairs oligodendrocyte remyelination following lysolecithin-induced demyelination. Glia 2001; 35: 204–12.[CrossRef][ISI][Medline]
Lawson LJ, Frost L, Risbridger J, Fearn S, Perry VH. Quantification of the mononuclear phagocyte response to Wallerian degeneration of the optic nerve. J Neurocytol 1994; 23: 729–44.[CrossRef][ISI][Medline]
Lee YL, Shih K, Bao P, Ghirnikar RS, Eng LF. Cytokine chemokine expression in contused rat spinal cord. Neurochem Int 2000; 36: 417–25.[CrossRef][ISI][Medline]
Leskovar A, Moriarty LJ, Turek JJ, Schoenlein IA, Borgens RB. The macrophage in acute neural injury: changes in cell numbers over time and levels of cytokine production in mammalian central and peripheral nervous systems. J Exp Biol 2000; 203: 1783–95.[Abstract]
Liefner M, Siebert H, Sachse T, Michel U, Kollias G, Bruck W. The role of TNF-alpha during Wallerian degeneration. J Neuroimmunol 2000; 108: 147–52.[Medline]
Ma M, Wei T, Boring L, Charo IF, Ransohoff RM, Jakeman LB. Monocyte recruitment and myelin removal are delayed following spinal cord injury in mice with CCR2 chemokine receptor deletion. J Neurosci Res 2002; 68: 691–702.[CrossRef][ISI][Medline]
Mabon PJ, Weaver LC, Dekaban GA. Inhibition of monocyte/macrophage migration to a spinal cord injury site by an antibody to the integrin alphaD: a potential new anti-inflammatory treatment. Exp Neurol 2000; 166: 52–64.[CrossRef][Medline]
McTigue DM, Tani M, Krivacic K, Chernosky A, Kelner GS, Maciejewski D, et al. Selective chemokine mRNA accumulation in the rat spinal cord after contusion injury. J Neurosci Res 1998; 53: 368–76.[CrossRef][ISI][Medline]
Mueller M, Leonhard C, Wacker K, Ringelstein EB, Okabe M, Hickey WF, et al. Macrophage response to peripheral nerve injury: the quantitative contribution of resident and hematogenous macrophages. Lab Invest 2003; 83: 175–85.[ISI][Medline]
Ousman SS, David S. Lysophosphatidylcholine induces rapid recruitment and activation of macrophages in the adult mouse spinal cord. Glia 2000; 30: 92–104.[CrossRef][ISI][Medline]
Ousman SS, David S. MIP-1alpha, MCP-1, GM-CSF, and TNF-alpha control the immune cell response that mediates rapid phagocytosis of myelin from the adult mouse spinal cord. J Neurosci 2001; 21: 4649–56.
Pan JZ, Ni L, Sodhi A, Aguanno A, Young W, Hart RP. Cytokine activity contributes to induction of inflammatory cytokine mRNAs in spinal cord following contusion. J Neurosci Res 2002; 68: 315–22.[Medline]
Perry VH, Brown MC, Gordon S. The macrophage response to central and peripheral nerve injury. A possible role for macrophages in regeneration. J Exp Med 1987; 165: 1218–23.
Perry VH, Andersson PB, Gordon S. Macrophages and inflammation in the central nervous system. Trends Neurosci 1993; 16: 268–73.[CrossRef][ISI][Medline]
Popovich PG, Guan Z, Wei P, Huitinga I, van Rooijen N, Stokes BT. Depletion of hematogenous macrophages promotes partial hindlimb recovery and neuroanatomical repair after experimental spinal cord injury. Exp Neurol 1999; 158: 351–65.[CrossRef][ISI][Medline]
Ransohoff RM. The chemokine system in neuroinflammation: an update. J Infect Dis 2002; 186 Suppl 2: S152–6.[Medline]
Rostworowski M, Balasingam V, Chabot S, Owens T, Yong VW. Astrogliosis in the neonatal and adult murine brain post-trauma: elevation of inflammatory cytokines and the lack of requirement for endogenous interferon-gamma. J Neurosci 1997; 17: 3664–74.