Interleukin 1β inhibition contributes to the antinociceptive effects of voluntary exercise on ischemia/reperfusion-induced hypersensitivity

Abstract

Issues of peripheral circulation have been increasingly suggested as an underlying cause of musculoskeletal pain in many conditions, including sickle cell anemia and peripheral vascular disease. We have previously shown in our model of transient ischemia and reperfusion (I/R) injury of the forelimb that individual group III and IV muscle afferents display altered chemosensitivity and mechanical thresholds 1 day after injury. Functional alterations corresponded to increased evoked and spontaneous pain-related behaviors and decreased muscle strength and voluntary activity-all actions that echo clinical symptoms of ischemic myalgia. These behavioral and physiological changes appeared to originate in part from the action of increased interleukin 1β (IL1β) in the injured muscles at its upregulated IL1 receptor 1 within the dorsal root ganglion. Here, we describe that two days of voluntary wheel running prior to I/R blocks both injury-induced IL1β enhancement and the subsequent development of ischemic myalgia-like behaviors. Furthermore, the protective effects of 2 days prior exercise on the I/R-evoked increases in pain-related behaviors were also paralleled with systemic injection of the IL1 receptor antagonist during I/R. Interleukin 1 receptor antagonist treatment additionally prevented the I/R-induced changes in mechanical and chemical sensitivity of individual primary muscle afferents. Altogether, these data strengthen the evidence that transient I/R injury sensitizes group III and IV muscle afferents via increased IL1β in the muscles to stimulate ischemic myalgia development. Targeting IL1β may, therefore, be an effective treatment strategy for this insidious type of muscle pain.

Figure 1. Two days of prior exposure to voluntary wheel running blocks ischemia and reperfusion injury (I/R)-induced changes in muscle IL1β and pain-related behaviors

(A) Singly-housed mice were either exposed to an in-cage running wheel or kept in traditional housing for two days prior to occlusion and reperfusion or sham surgeries. Behavioral assays were performed immediately prior to injury surgeries (baseline, BL), at one day following occlusion surgeries (D1), and at three days following surgery (D3). (B) Western blot of protein isolated from the affected forepaw muscles of similarly treated cohorts at D1 (n=3–4) shows that voluntary exercise-exposed animals are protected from I/R-related increases in IL1β. (C) Spontaneous/ongoing pain-like behavior induced by I/R is ablated in voluntary exercise exposed animals (significant interaction condition x time p<0.001). Decreases in paw withdrawal threshold to mechanical stimulation (D; significant main effect of condition p<0.001) and in grip strength (E; significant interaction condition x time p=0.43) are also prevented with 2d prior voluntary activity. Data are represented as mean ± SEM and were analyzed via two-way ANOVA or two-way RM ANOVA with Holm-Sidak post hoc tests. **p≤0.001 vs BL and other groups, *p<0.05 vs. BL and other groups.

Figure 2. Forced running immediately after reperfusion does not prevent increases in injury-associated pain-related behaviors or muscle IL1β

(A) Cohorts of age-matched Naïve and I/R mice underwent a moderate forced treadmill running protocol immediately following reperfusion surgery. (B) As measured by Western blot, the upregulation of muscle IL1β protein expression in forced run I/R vs Naïve animals was comparable to that seen in non-exercised animals (see Fig. 1B). (C) Paw guarding behaviors were significantly increased at D1 following forced running in animals with I/R (interaction effect of condition x time p<0.001). Similarly, forced running did not affect I/R-induced reductions in mechanical threshold (D; interaction effect of condition x time p=0.003) or grip strength (E; interaction effect of condition x time p=0.011), but also did not affect these behaviors in uninjured animals. Data from the non-exercised groups from Fig. 1 have also been included for comparison to forced run conditions. All data are represented as mean ± SEM. Behavioral data were analyzed via three-way ANOVA (condition x time x exercise) with Holm-Sidak, and Western blot data were analyzed using one-way ANOVA with Holm-Sidak. **p<0.001 vs BL and uninjured groups, *p<0.05 vs. BL and uninjured groups.

Figure 3. Treatment with the interleukin-1 receptor antagonist (IL1RA) attenuates increases in spontaneous pain-like behaviors and evoked mechanical hypersensitivity, but does not rescue grip strength impairment, associated with I/R

(A) Immediately before occlusion and following reperfusion surgery, mice were intraperitoneally injected with 3 mg/kg IL1RA or vehicle (H₂O). Pain-related behavioral assays were performed at BL and D1. (B) IL1RA ablates I/R-evoked increases in ongoing paw guarding (significant interaction condition x time p<0.001), and improves decreases in paw withdrawal threshold to mechanical stimulation (C; significant interaction condition x time p=0.010), but does not fully restore thresholds to BL levels (IL1RA+I/R BL vs D1 p<0.05). (D) Grip strength was similarly decreased on D1 from BL in both groups. Data are represented as mean ± SEM and were analyzed via two-way RM ANOVA with Holm-Sidak post hoc tests. **p<0.001 vs BL and other groups, *p<0.05 vs. BL.

Figure 4. Effects of IL1RA treatment on I/R-induced alterations on Group III and IV primary muscle afferents as assessed by ex vivo electrophysiology

(A). Compared to previous pooled data from uninjured mice, IL1RA+I/R mice were no different in total number of mechanosensitive or chemosensitive cells, but Vehicle+I/R mice had significantly more mechanically sensitive cells (B). (C) The decrease in mechanical threshold following I/R seen in vehicle-treated animals (vs naïve comparison p=0.037) is absent in IL1RA+I/R animals (vs naïve comparison p=0.056, vs Vehicle+I/R p<0.001; representative traces during mechanical stimulation of the muscles, D). (E) Although the total number of metabolite responders was unaffected by I/R or IL1RA, proportions of cells sensitive to Low, High, or Both metabolite solutions were significantly altered from naïve and IL1RA+I/R proportions in vehicle-treated animals (p=0.045 and 0.007, respectively), whereas naïve and IL1RA+I/R metabolite proportions did not differ (p>0.05). I/R also causes an increase in IF to low metabolites that may be partially rescued with IL1RA treatment (F; Vehicle+I/R vs naïve p=0.033, IL1RA p>0.05 vs other groups). Differences in phenotype distribution were analyzed with χ², and response properties were analyzed with Kruskal-Wallis one-way ANOVA on ranks with Dunn’s test. Phenotype data are presented as percentage of cells responsive to a certain stimulus out of the total tested for that stimulus within that condition, and response properties are represented as mean ± SEM, wherein these reference values from pooled naïve data are reflected as solid and dashed lines, respectively. * indicates p<0.05 compared to naïve; # indicates p>0.05 compared to naïve and p<0.05 vs Vehicle+I/R, ^ indicates p<0.05 vs naïve but not IL1RA+I/R.

Figure 5. DRG ASIC3 upregulation after I/R is prevented with IL1RA treatment

(A) Representative immunocytochemical staining of ASIC3 (blue) and P2X3 (red) protein in whole DRGs from naïve, Vehicle+I/R and IL1RA+I/R conditions. (B) The Vehicle+I/R condition contained significantly more cells immunopositive for ASIC3 than IL1RA+I/R animals (p=0.018), but P2X3 expression did not differ (C; p>0.05). (D) Co-expression of ASIC3 and P2X3 was significantly increased in Vehicle+I/R mice compared to IL1RA+I/R mice (p<0.039). Scale bar for all images, 60 μm. Experimental data are represented as mean ± SEM and were analyzed with Kruskal-Wallis one-way ANOVA on ranks with Dunn’s test. Solid and dashed lines represent the mean and upper or lower extremes of the SEM, respectively, of pooled naïve comparison data. *p < 0.05 compared to other group(s).

Figure 6. Acute exercise with IL1RA treatment decreases pain-related behaviors and increases voluntary activity following I/R similar to 2d prior exercise exposure

Mice were briefly habituated to a running wheel, then intraperitoneally injected with 3 mg/kg IL1RA or vehicle (H₂O) before occlusion surgery, and placed back into running wheel housing for remainder of study. Following reperfusion surgery, they were again injected with IL1RA or vehicle (A). (B) IL1RA treatment with acute exercise exposure attenuates increased guarding score evoked by I/R, and ensures D3 recovery in treated mice (significant interaction condition x time p<0.001). (C) Although IL1RA+I/R animals showed a mild decrease from BL in paw withdrawal threshold on D1, they did not differ from uninjured mice at any time point (significant interaction condition x time p=0.042). (D) Grip strength in IL1RA+I/R animals also did not differ at any time point (significant interaction condition x time p=0.006). IL1RA treatment protects from I/R-associated decrease in voluntary activity. Vehicle+I/R animals ran shorter distances on D1 compared to both IL1RA+I/R and uninjured mice (E; significant main effects of condition p=0.035 and time p<0.001), but average velocity only differed from IL1RA+I/R animals on D1 (F; significant main effects of condition p=0.047 and time p=0.002). (G) Following normalization to average naïve 2d distance, total 2d post-I/R distance traveled from animals in Fig. 1 was compared to that from Vehicle+I/R and IL1RA+I/R groups, showing that 2d prior exercise exposure and IL1RA treatment similarly recover voluntary activity following I/R (Vehicle+I/R p≤0.014 vs all other conditions). Data are represented as mean ± SEM and were analyzed via two-way RM ANOVA (B–F) or one-way ANOVA (G) with Holm-Sidak post hoc tests. ^p<0.05 vs other time points, *p<0.05 vs other groups, # p<0.05 vs uninjured controls only, ^>p=0.034 vs IL1RA+I/R only.

Figure 7. Reward-seeking behavior is unaffected by I/R and/or IL1RA

Saccharin preference was assayed in an additional cohort of mice with injury and/or injection treatment (n=5–6). During the week leading up to I/R surgery, mice of the various conditions were habituated to saccharin exposure during two overnight periods. Then, following reperfusion surgery mice were again provided with both water and saccharin solution. (A) Total liquid intake in these mice over the night following I/R surgery was no different between groups (p>0.05). (B) There was no clear effect of IL1RA or I/R on saccharin preference (p>0.05). Data are represented as mean ± SEM and were analyzed via two-way ANOVA (drug x injury).