1. Effects of morphine on locomotor and sensory recovery.
1.1 Effects of intravenous morphine on locomotor recovery
- Insert Figure 4 here –
Figure 4. Effects of intravenous morphine on locomotor recovery. BBB scores were balanced across groups on day 1 post-injury. Morphine significantly undermined locomotor recovery in contused rats after 7 days of IV administration. Results shown as Mean ± S.E.M. * p < 0.05, n=10.
To assess the effects of morphine on recovery, locomotor behavior was scored daily using the BBB scale [35]. Figure 4 shows mean converted BBB scores (± S.E.M) for the first seven days after surgery. Locomotor scores on day one prior to drug treatment did not differ significantly between contused saline (1.70 ± 0.35) and contused morphine (1.65 ± 0.30) rats.
A two-way repeated measures ANOVA showed a main effect of day post-surgery (F (6, 108) = 19.72, p < 0.001), but not of treatment (F (1, 18) = 3.622, p = 0.07), on BBB scores. However, replicating our previous studies, our analysis revealed a significant interaction between day post-surgery and treatment (F (6, 108) = 7.694, p < 0.001). Saline-treated rats had significantly higher BBB scores than morphine-treated rats on days 6 and 7 post-surgery (t (126) = 2.739, p < 0.05; t (126) = 3.822, p < 0.05, respectively). Vehicle control rats recovered to an average converted BBB score of 5.30 ± 0.89, while morphine-treated rats recovered to an average converted BBB score of 2.30 ± 0.49. All sham rats, irrespective of treatment, had a converted BBB score of 12 (unconverted BBB score of 21) throughout the 7 day post-surgery assessment period.
1.2 Effects of morphine on sensory and mechanical reactivity
Figure 5 illustrates the effects of morphine on sensory recovery for the first seven days following sham or contusion injury. Sensory recovery was evaluated by assessing thermal (tail-flick test) and mechanical reactivity (von Frey filaments) on the first and last day of treatment, immediately before and 30-min after the intravenous administration of morphine.
- Insert Figure 5 here –
Figure 5. Effects of morphine on thermal and tactile recovery. Morphine produced robust analgesia on thermal (A) and tactile (B, C) reactivity. There was no reduction in the efficacy of morphine across days on either sensory test. Conversely, the development of hyperalgesia was observed by day 7 on the tactile reactivity (E, F) test but not for thermal reactivity (D). Contused rats treated with morphine displayed significantly lower thresholds for vocal (F), and a trend for reduction of motor (E), responses on the tactile reactivity test. Results shown as Mean ± S.E.M. *p < 0.05, n=10.
1.2.1 Assessment of the analgesic efficacy of morphine.
Morphine produced robust analgesia on the thermal reactivity test that persisted across the 7 days of administration (Figure 5A). Both sham and contused animals treated with morphine reached the maximum 8-sec tail-flick latency. A 3-way ANOVA revealed a significant main effect of drug treatment(F (1, 36) = 1092, p < 0.001). There were no effects of surgery (F (1, 36) = 0.05, p = 0.81) or day post injury (F (1, 36) = 2.97, p = 0.09), and no significant Surgery x Drug treatment interactions.
Similarly, morphine produced robust analgesia on the von Frey test of mechanical reactivity at all timepoints. A 3-way ANOVA revealed a main effect of surgery (F (1, 36) = 8.95, p < 0.05), a main effect of drug treatment (F (1, 36) = 44.62, p < 0.001), and a Surgery x Drug treatment interaction (F (1, 36) = 8.95, p < 0.05) on the motor response to mechanical stimulation (Figure 5B). Sham rats showed higher sensitivity to tactile stimulation than contused rats across all days. Morphine administration also resulted in less vocal responses to mechanical stimulation across all testing days (Figure 5C). There was a main effect of drug treatment (F (1, 36) = 10.79, p < 0.001), and no effect of surgery or significant interaction between drug treatment and surgery. We did not see evidence of tolerance to morphine developing across days. Analgesia on the thermal reactivity and mechanical stimulation tests persisted across the seven days of administration (Figure 5, A-C).
1.2.2. Assessment of opioid-induced hyperalgesia
Thermal and mechanical reactivity were also assessed on Days 1 and 7 post injury, prior to drug administration, to examine the development of opioid-induced hyperalgesia. There were no significant differences in tail-flick reactivity between conditions on Day 1 post-injury. Across the 7 days post-injury, however, there was a main effect of day of testing (F (1, 36) = 7.21, p < 0.05) but no effect of surgery, drug treatment, or significant interactions (Figure 5D). Post hoc comparisons, revealed that the latency to tail-flick was increased in morphine-treated sham rats from Day 1 to Day 7 after injury (t (36) = 3.05, p < 0.05). Tail-flick latency did not change for the rest of the treatment groups across days.
Conversely, evidence of opioid-induced hyperalgesia was observed on the mechanical reactivity test (Figure 5, E-F). As with the assessment of tolerance, contused animals showed less motor reactivity to mechanical stimulation compared to sham rats (F (1, 36) = 11.43, p < 0.001). Additionally, statistical analyses of the vocal responses to mechanical stimulation revealed a main effect of day of testing (F (1, 36) = 6.810, p < 0.05), where Day 7 vocalization scores were significantly lower relative to Day 1 scores. There were also significant interactions between day of testing and surgery (F (1, 36) = 3.452, p < 0.05), and between day of testing and drug treatment (F (1, 36) = 12.64, p < 0.001). Post hoc analyses show that these effects were mainly driven by changes in the threshold for vocalizations in contused animals treated with morphine. As shown in Figure 5F, vocalization reactivity thresholds were significantly decreased from Day 1 to day 7 (t (36) = 4.69, p < 0.001) in the morphine-treated SCI rats, indicating the development of opioid-induced hyperalgesia.
2. Morphine increases the total number of microglia and macrophages
- Insert Figure 6 here –
Figure 6. Quantification of microglia and macrophages in CD86 set using flow cytometry. The contusion injury significantly increases the number of CD11b total positive cells (A), percentage of CD11b positive cells (B), total number of macrophages (C), and total number of microglia (D) at the site of injury relative to a sham surgery. After 3 days of morphine administration, contused animals also had a significantly higher total number of CD11b positive cells (A), percentage of CD11b positive cells (B), total number of macrophages (C), and total number of microglia (D) compared with vehicle SCI controls. There was no significant effect of treatment with 1 or 7 days of morphine administration. Results shown as Mean ± S.E.M. *p < 0.05, n=5-6.
Based on our previous studies [15], we hypothesized that morphine enhances inflammation and exacerbates the secondary injury cascade by increasing the number of immune cells at the injury site. To test this, we used flow cytometry to compare the expression of microglia and macrophages at the SCI lesion site after repeated i.v. administration of morphine versus vehicle. First, we quantified the total number of CD11b+ cells as an indirect measure of inflammation. To be considered positive (+), cell populations were selected using a gate that contained < 1% CD11b+ unstained cells.
Focusing on the M1 antibody panel, the contusion injury significantly increased the number and percentage of CD11b+ cells at 7 days post injury, compared to sham surgery (F (1, 20) = 33.54, p < 0.001 and F (1, 20) = 27.82, p < 0.001, respectively), regardless of drug treatment. Importantly, an effect of morphine on the number of CD11b+ cells was also observed after just 3 days of administration. At this timepoint there was a significant increase in the number of CD11b+ cells (t (8) = 6.67, p < 0.001, Figure 6 A and B) in morphine-treated SCI rats compared to their saline counterparts. The same effects of surgery and day post injury (F (1, 20) = 7.080, p < 0.05 and F (1, 20) = 7.392, p < 0.05) were seen in analyses of the M2 antibody panel. Again, there was significant increase in the number of CD11b+ cells (t (8) = 4.768, p < 0.001), macrophages (t (8) = 4.691, p < 0.001), and microglia (t (8) = 3.492, p < 0.001) on Day 3 post injury in the morphine-treated rats.
To assess the effects on the peripheral versus resident immune response, a CD45 marker was used. CD11b+/CD45+ cells were selected and separated into discrete populations of high and low CD45 expression; we identified these cell populations as infiltrating macrophages (CD11b+/CD45High) and resident microglia (CD11b+/CD45Low) respectively, as previously done by others [41-44], but recognize that other cells may have been included in these populations. Similar to total CD11b+ quantification, after SCI and 3 days of morphine administration our analysis of the M1 antibody panel shows a significant increase in the number of both macrophages (t (8) = 7.03, p < 0.001), and microglia (t (8) = 2.79, p < 0.05, Figure 6 C and D, respectively), compared to vehicle-treated rats. Replicating this finding, for the M2 panel, morphine-treated SCI rats had an increased number of macrophages (t (8) = 4.691, p < 0.001) and microglia (t (8) = 3.492, p < 0.001) at the injury site, compared to vehicle-treated rats. There was no effect of treatment with 1 or 7 days of morphine on the number or percentage of CD11b+ macrophages (CD11b+/CD45High), or microglia (CD11b+/CD45Low). Together, these results show that morphine administration produces a transient, but dramatic surge in the number of immune cells, including macrophages and microglia, in a rodent model of SCI.
2.1: Effect of morphine on CD86+ microglia/macrophages
- Insert Figure 7 here –
Figure 7. Quantification of microglia/macrophages expressing CD86 (M1 marker) and KOR. Morphine significantly increases the number of CD86+ macrophages (C) in contused rats after 3 days of administration but had no effect on CD86+ microglia at any timepoint(A). However, 3 days of morphine administration significantly increased the number of CD86+ microglia and macrophages expressing KORs (B and D). Results shown as Mean ± S.E.M. *p < 0.05, n=5-6.
Immune cells like microglia and macrophages are highly heterogenous in phenotype and function. Thus, after quantifying the overall expression of cells, we further examined the phenotype of the cells present at the site of injury, quantifying the number of cells within each population expressing the CD86 marker (M1 panel). Our statistical analyses show there were no significant changes in the overall number of CD86+ microglia (t (8) = 0.85, p = 0.4226, Figure 7A) at any timepoint. However, morphine had divergent effects on CD86+ macrophages at different timepoints. Contused rats treated with morphine had significantly less CD86+ (Figure 7C) macrophages after 1 day of administration (t (10) = 2.33, p < 0.05). However, after 3 days of morphine administration the overall number of CD86+ macrophages significantly increased (t (8) = 2.658, p < 0.05, Figure 7C) relative to vehicle-treated contused rats. Not surprisingly, there was a main effect of surgery on CD86 expressing macrophages, with contused rats exhibiting significantly higher numbers of CD86+ macrophages than shams on Day 7 post injury (F (1, 20) = 16.07, p < 0.001, Figure 7C).
Additionally, we previously found that activation of KORs is both sufficient and necessary to mediate the negative effects of morphine on motor recovery in the rat model of SCI [12, 13]. To explore whether morphine administration altered the expression of opioid receptors on specific subpopulations of immune cells we quantified KOR expression within the CD86+ subpopulations of microglia and macrophages. Our results show that morphine significantly increased the number of CD86+ microglia expressing KORs after 3 days of drug administration (t (8) = 1.92, p < 0.05) in comparison to saline-treated animals (Figure 7B). There was an additional main effect of surgery on the expression of KORs after 7 days of treatment, with contused rats exhibiting significantly higher numbers of KOR+ microglia (F (1, 20) = 5.26, p < 0.05) compared to sham rats, irrespective of treatment (Figure 7B). There were no other significant effects or interactions on Days 1 and 7. For macrophages, after 1 day of morphine administration, there is a significant decrease in the overall number (Figure 7D) of CD86+ cells expressing KORs (t (10) = 2.574, p < 0.05). Again, this effect is reversed after 3 days of administration when there is a significant increase in the overall number of CD86+ macrophages expressing KORs relative to vehicle treated controls (t (8) = 5.645, p < 0.0001). There is also a main effect of surgery on the expression of KOR on CD86+ macrophages observed after day 7, with contused rats exhibiting a higher number of CD86+ macrophages expressing KORs (F (1, 20) = 6.96, p < 0.05) compared to sham rats (Figure 7D). These data show that both SCI and morphine administration increase the expression of CD86+ macrophages around the center of the lesion, as well as increasing the number of CD86+ macrophages expressing KORs.
2.2: Effect of morphine on CD206+ microglia/macrophages
- Insert Figure 8 here –
Figure 8. Quantification of microglia/macrophages expressing CD206 (M2 marker) and KOR. Morphine significantly increases the number of CD206+ microglia (A) and CD206+ macrophages (C) after 3 days of administration. Morphine administration also increases the number of CD206+ microglia and macrophages expressing KORs (B and D). Results shown as Mean ± S.E.M. *p < 0.05, n=6.
Another subpopulation of glia cells that are relevant within the context of SCI are CD206 expressing cells. Transplantation of CD206+ microglia and macrophages into the site of lesion after an SCI has been reported to promote recovery of function in rodents [45-47]. Thus, as we did before, we used flow cytometry to quantify the number of glia expressing the CD206 and KOR markers (M2 panel, Figure 8). After 1 day of administration, saline-treated rats had a modest but significantly higher number of CD206+ microglia (t (8) = 2.698, p < 0.05, Figure 8A) compared to morphine-treated rats. Conversely, after 3 days of morphine administration there was a significant increase in the overall number of CD206+ microglia (t (8) = 3.456, p < 0.001, Figure 8A) and CD206+ macrophages (t (8) = 4.423, p < 0.001, Figure 8C) in morphine-treated rats versus saline controls.
Similarly, after 1 day of treatment, saline-treated rats showed significantly higher numbers of CD206+ microglia expressing KORs (t (8) = 2.659, p < 0.05, Figure 8B) compared to morphine-treated rats. Again, this effect reversed after 3 days of morphine, compared to vehicle, administration when there was an increase in the total number of CD206+ microglia expressing KORs (t (8) = 3.905, p < 0.001), and an increase in the overall number of CD206+/KOR+ macrophages (t (8) = 4.926, p < 0.001, Figure 8D). There was an additional main effect of surgery, where contused rats showed significantly higher numbers of CD206+ macrophages (F (1, 20) = 11.04, p < 0.001, Figure 8B), and CD206+ KOR+ macrophages (F (1, 20) = 15.70, p < 0.001, Figure 8D) compared to sham animals. There were no other significant main effects or interactions for days 1 and 7. Therefore, similar to the CD86+ population, our results show that SCI alone increases the number of CD206+ macrophages/microglia. Morphine administration further increases the expression of CD206+ cells after SCI, as well as increasing the number of CD206+/KOR+ macrophages and microglia. Interestingly, the morphine effects seem to be larger in CD206+ cells compared to CD86+ cell populations.
3. Effect of morphine on protein expression in CD11b+ cells
In addition to increasing the number of microglia and macrophages, we hypothesized that morphine administration may change the function of these cells after SCI. To address this, we used western blot analyses to examine whether morphine-induced activation of microglia and macrophages after SCI engages signaling pathways associated with the production of pro-inflammatory cytokines and neurotoxicity.
3.1 Effect of morphine on protein expression 30-minutes post-administration
- Insert Figure 9 here –
Figure 9. Quantification of b-arrestin pathway protein, Dynorphin, and Pro-Dynorphin in CD11b+ cells from center of lesion 30-minutes after 3 days of drug administration. Morphine administration significantly increases the expression of b-arrestin (A) and ERK 1 (C) in contused rats compared to their saline-treated counterparts. Morphine administration significantly increases the expression of pro-dynorphin (E) and dynorphin (F) in contused rats compared to their saline-treated counterparts. Loading control: protein extracted from CD11b+ cells from spinal cord tissue from sham rats treated with saline. Results shown as Mean ± S.E.M. *p < 0.05, n=6.
Activation of classic opioid receptors (ORs) on immune cells has been shown to induce the inflammatory b-arrestin pathway [18, 22, 23]. Thus, we used western blot analysis to estimate changes in the concentration of proteins downstream of b-arrestin in CD11b+ cells (microglia and macrophages), 30 minutes post-morphine administration. Quantification revealed a significant increase in b-arrestin expression in morphine-treated rats (t (10) = 2.49, p < 0.05) and ERK 1 (t (10) = 2.10, p < 0.05) relative to saline-treated contused controls (Figure 9 A and B). There was also a trend toward an increase in p38 MAPK and ERK 2 expression in morphine-treated rats, but these effects were not significant (t (10) = 1.17, p < 0.13 and t (10) = 1.38, p < 0.09, respectively). OR activation also induces ERK 1/2 which is a kinase responsible for CREB upregulation of dynorphin gene expression [48]. Supra-physiological levels of dynorphin within the spinal cord can induce neurotoxicity and cause motor dysfunction [49], so we used western blot analysis to look at components of this pathway. Morphine administration significantly increased the expression of pro-dynorphin (t (10) = 2.014, p < 0.05) and dynorphin (t (10) = 2.715, p < 0.05), compared to saline-treated rats (Figure 9 E and F).
3.2 Effect of morphine on protein expression 24-hours post-administration
- Insert Figure 10 here –
Figure 10. Quantification of b-arrestin pathway proteins in CD11b+ cells from center of lesion 24-hours after morphine administration. Total Protein Stain as internal loading control for normalization of target signals (A). Protein collected from rats treated for 3 days with i.v. drug administration. Loading control: protein extracted from whole spinal cord tissue from sham rats treated with saline (lanes 2-3). Western blot image (B) protein was collected from rats that received 3 days of consecutive i.v. drug treatment and samples taken 24-hours after the last drug dose and processed with LI-COR system. Figure (B) western blot images were cropped to remove irrelevant sections of the image and display only the proteins of interest. The contusion injury significantly increased the expression of b-arrestin (C) and p38 MAPK (D) on day 3. Additionally, morphine significantly decreases the expression of p38 MAPK by day 7 compared to saline-treated rats (D). Results shown as Mean ± S.E.M. *p < 0.05, n=6.
Twenty four hours post morphine administration (3 days of administration), there was also a main effect of surgery with SCI significantly increasing the expression of the b-arrestin protein compared to sham surgery (F (1, 20) = 4.85, p < 0.05), irrespective of drug treatment (Figure 10 C). Similarly, the contusion injury significantly increased the expression of p38 MAPK in CD11b+ cells compared to the sham surgery (F (1, 20) = 4.57, p < 0.05) irrespective of treatment (Figure 10D). Surprisingly, after 7 days of drug administration there was a significant decrease in the expression of p38 MAPK in morphine-treated rats (t (10) = 2.465, p < 0.05) as compared to saline-treated animals.
While there was a trend toward an increase in protein expression of ERK 1 and ERK 2 in the CD11b+ cells extracted from the rats treated with morphine for 3 days (t (10) = 1.32, p = 0.10 and t (10) = 1.06, p = 0.16, respectively), there were no main effects of surgery or drug-treatment at any time point (Figure 10 E and F). There were also no significant effects of surgery on the expression of pro-dynorphin or dynorphin. However, similar to p38 MAPK levels, there was a significant decrease in the expression of pro-dynorphin after day 7 in morphine-treated rats (t (10) = 2.502, p < 0.05) compared to saline-treated rats (Figure 11A). There were no additional effects of treatment for Days 1 or 3 of treatment. Our statistical analyses did not show main effects of treatment for dynorphin on any of the test days (Figure 11B).
- Insert Figure 11 here –
Figure 11. Quantification of Dynorphin and its precursor Pro-Dynorphin in CD11b+ cells taken from center of lesion 24-hours after administration. Morphine administration significantly decreases the expression of pro-dynorphin compared to saline-treated subjects by day 7 (A). Western blot image (C) protein was collected from subjects that received 3 days of consecutive i.v. drug treatment. Samples taken 24-hours after the last drug dose and processed with LI-COR system. Figure (C) western blot images were cropped to remove irrelevant sections of the image and display only the proteins of interest. Loading control: protein extracted from whole spinal cord tissue from sham subjects treated with saline (lanes 2-3). REVERT Total protein stain used as internal control. Tissue used for western blot analysis was collected 24-hours after the last drug-administration. Results shown as Mean ± S.E.M. *p < 0.05, n=6.