Neuroinflammation plays a critical role in secondary damage that follows traumatic SCI. In particular, pro-inflammatory cytokines and chemokines have been shown to play a critical role in exacerbating the secondary injury that follows the primary injury to the central nervous system. This study examined the effects of CCL3, a pro-inflammatory chemokine with a role in neuroinflammatory conditions, on tissue damage and locomotor recovery after contusion SCI. We detected an increase in the expression of CCL3 and the receptors CCR1 and CCR5 in the spinal cord that persisted long-term after the injury. To analyze the impact of CCL3 after SCI, using the CCL3−/− mice, we demonstrated a mild improvement of locomotor recovery, reduced lesion sizes as well as a reduction of the inflammatory response, characterized by the neutrophil influx and expression of pro-inflammatory cytokines.
Several studies have addressed the expression of CCL3 in the injured spinal cord. For instance, in mice and rats, CCL3 is upregulated within the first day after SCI [29–31, 39] and stays elevated for prolonged periods, with observation times of up to 42 days after injury [33, 40, 41]. A similar pattern of CCL3 regulation has been observed after traumatic brain injury in mice . These results are in accordance with our findings regarding the expression levels of CCL3 mRNA or protein, which were equally upregulated at early time points after SCI and stayed elevated long-term.
Similarly, we are reporting elevated levels of the receptors CCR1 and CCR5 after SCI. CCR1 mRNA expression was elevated at day 1 after injury and remained upregulated until day 7, while protein expression levels were elevated at day 7 to day 28. CCR5, whose ligands include CCL3, CCL4, CCL3L1 and CCL5, showed mRNA upregulation at day 3 and 7 after injury, while protein expression was increased by day 7 and maintained on a high level until day 28 after injury. In a rat model of SCI, CCR1 expression is also increased depending on injury severity, but shows the highest levels at later time points after SCI . After mouse traumatic brain injury, in contrast, CCR1 mRNA is upregulated after injury but returns to normal levels after seven days in cortex, striatum and thalamus, while staying upregulated in the hippocampus for up to five weeks. However, CCR5 mRNA stays elevated for several weeks after traumatic brain injury, while protein levels for both CCR1 and CCR5 return to baseline after seven days . The observed similarities highlight a role for CCL3 and its receptors after traumatic injuries to the CNS. The differences in expression duration could be caused by species differences and variations in tissue expression.
The third receptor CCL3 can interact with is CCR4 . In our model of thoracic contusion SCI, no significant regulation of CCR4 was detected in the injured tissue at either time point, neither on the mRNA nor on the protein level. CCR4 is expressed on T regulatory (Tregs) or Type 2 T- helper (Th2) cells and has previously been detected on FoxP3/ GFP + Tregs after SCI, using flow cytometry . We cannot exclude that the Q-PCR on the whole injured spinal cord tissue is not sensitive enough to detect CCR4 expression levels on the individual cell level.
After demonstrating the regulation of CCL3 and its receptors CCR1 and CCR5 after SCI, we assessed the impact of CCL3 on locomotor recovery and secondary damage in CCL3−/−mice compared to wild-type controls. In a first experiment, CCL3−/−mice initially showed significant improvement of locomotor recovery. However, at later time points, the behavioral scores of CCL3−/−mice deteriorated and were no longer different from wild-type mice. We then performed two follow up experiments in female and male CCL3−/−mice, using a slightly milder contusion force (40 kdyn) to determine the effect of CCL3 on recovery under these conditions. While the BMS scores were elevated using this setup, we only detected a non-significant trend to improved outcome in CCL3−/−mice. A more detailed analysis of the locomotor behavior early after injury to showed that a significantly higher percentage of CCL3−/−mice demosntrated plantar placement at three, five and seven days after injury in both male and female mice.
The mild functional improvement was reflected by smaller lesion sizes in CCL3−/−mice, affecting mainly the gray area, while the myelin content was not significantly different between wild-type and CCL3−/−mice. The detrimental effect of CCL3 and its pathway has been demonstrated in other disease models. CCL3−/−mice show reduced lesion sizes and plaque formation in atherosclerosis  and blockade of CCR5, using neutralizing antibodies, promotes locomotor recovery after SCI . However, it is important to note that CCR5 is the receptor for various chemokines besides CCL3. In addition to functional recovery after SCI, CCL3 also plays a well described role in neuropathic pain. Higher levels of CCL3 were associated with the impaired locomotor recovery and neuropathic pain in a rat SCI model . After chronic constriction injury (CCI) of the sciatic nerve, CCL3, CCR1 and CCR5 mRNA and protein are upregulated in the spinal cord. Neuropathic pain could be inhibited by blockade with CCL3 neutralizing antibodies  or Maraviroc, a CCR5 antagonist . In our model, however, no differences in thermo- or mechano-sensitivity were detected between wild-type and CCL3−/−mice (data not shown). A possible explanation for this could be the differences in pain development between central and peripheral nervous tissue. Next, we investigated the mechanism by which CCL3 deficiency promotes functional recovery and reduced tissue damage. CCL3 is an inflammatory chemokine, which can induce the production of pro-inflammatory cytokines and the chemotactic mobilization of immune cells into inflammatory tissues .
In our SCI model, the absence of CCL3 strongly reduced the early influx of neutrophils into the injured tissue, which might contribute to the functional improvement and reduced tissue damage. In the spinal cord, like in many other injured tissues , neutrophils are the first cell type to infiltrate into the tissue, peaking one day after injury followed by a quick reduction in numbers . Several studies have identified neutrophils as a promising target after SCI [7, 51, 52]. However, in other studies, neutrophil depletion has led to impaired recovery and increased tissue damage . CCL3 and its receptors CCR1 and CCR5 are important mediators of intravascular adherence and transmigration of neutrophils [54, 55]. Absence or inhibition of CCL3 also leads to reduced neutrophil invasion and tissue protective effects in atherosclerosis , lung ischemia-reperfusion injury  and focal cerebral ischemia . Similarly, CCR5 blockade results in reduced neutrophil recruitment  after SCI. The influence of CCL3 on immune cell infiltration is not limited to neutrophils. Lysophosphatidylcholine (LPC) injection into the uninjured or hemisected spinal cord results in rapid and transient upregulation of CCL3 and recruitment of T-cells, neutrophils and monocytes [28, 58]. An overall reduction of infiltrating immune cells in the absence of CCL3 could contribute to the tissue immune response in the injured spinal cord of CCL3−/−mice, which was characterized by decreased expression levels of the pro-inflammatory cytokines il-1β, il-6 and tnf at early time points after injury. Importantly, tnf is not only a potent pro-inflammatory cytokine in the CNS, which is known to be regulated following acute and chronic inflammatory insults , but also it has been described as a potent inducer of apoptosis, contributing to the pathophysiology of many neurological disorders . Consistent with this, bax, which is involved in the apoptotic pathway, was upregulated in wild-type mice after SCI, but levels were reduced in the absence of CCL3. Thus, our results show that a reduction of tnf gene expression in the absence of CCL3 may decrease apoptosis after SCI. Further investigation is needed to determine the role of CCL3 in the apoptosis pathway during secondary damage after SCI.
Similarly, iNOS, as a direct marker of a pro-inflammatory reaction, was significantly decreased in the absence of CCL3. In contrast, arginase-1, which is often described as an indicator for anti-inflammatory phenotypes, was upregulated in CCL3−/−mice. This reduction of pro- and increase of anti-inflammatory responses is in accordance with findings in CCR5 blockade after SCI, where anti-inflammatory markers were increased as well .
These findings are further supported by results form a targeted gene array for cytokines and chemokines. We compared expression levels in the spinal cord of wild-type and CCL3−/−mice at seven and 21 days after injury. At seven days after injury, CCL3−/−mice downregulated 19 genes and did not upregulate any genes out of 84 genes screened. The majority of downregulated genes were pro-inflammatory cytokines and chemokines. This included chemotactic proteins, like fraktalkine  and il16, which is elevated in the plasma of SCI patients  and in rat SCI tissue, where it likely plays a role in immune activation and cell recruitment . Pro-inflammatory cytokines with reduced expression levels in CCL3−/− mice included tnf, which negatively impacts functional recovery after SCI , and mif, which is also elevated in the plasma of SCI patients , has neurotoxic properties  and is detrimental after SCI in mice . Interestingly, the anti-inflammatory factors il1ra and bmp7 were also upregulated. At day 21 after injury, only five genes were significantly regulated in CCL3−/− mice, again all of them reduced. Only one factor was regulated at both time points, which is thrombopoietin. While it was unexpected to detect the regulation of a protein involved in megakaryocyte maturation, thrombopoietin has previously been shown to be elevated in systemic lupus erythematosus and where it was strongly correlated with and likely induced by CCL3 .
Overall, at a time point when CCL3−/− mice and wild-type were behaviorally distinct, we detected a strong reduction in various cytokines and chemokines, the majority of which were pro-inflammatory and, in some cases, known to have negative impact after SCI. At a later time point, with reduced behavioral differences, most of the differentially regulated inflammatory factors also disappeared. The altered inflammatory environment in the injured spinal cord likely contributes to reduced secondary tissue damage after SCI. While an inflammatory reaction is important for tissue homeostasis and recovery, the prolonged, unresolved and exacerbated inflammation in spinal cord contusion injuries is related to pro-inflammatory responses, extensive secondary damage and a negative impact on functional recovery [64, 68].
To our knowledge, this is the first study investigating the impact of CCL3 on inflammation, secondary tissue damage and functional recovery after SCI. Taken together, the CCL3 pathway is an interesting target to reduce inflammation and decrease secondary damage after SCI.