Inflammatory pain resolution by mouse serum-derived small extracellular vesicles

Abstract

Current treatments for chronic pain have limited efficacy and significant side effects, warranting research on alternative strategies for pain management. One approach involves using small extracellular vesicles (sEVs), or exosomes, to transport beneficial biomolecular cargo to aid pain resolution. Exosomes are 30-150 nm sEVs that can be beneficial or harmful depending on their source and cargo composition. We report a comprehensive multi-modal analysis of different aspects of sEV characterization, miRNAs, and protein markers across sEV sources. To investigate the short- and long-term effects of mouse serum-derived sEVs in pain modulation, sEVs from naïve control or spared nerve injury (SNI) model male donor mice were injected intrathecally into naïve male recipient mice. These sEVs transiently increased basal mechanical thresholds, an effect mediated by opioid signaling as this outcome was blocked by naltrexone. Mass spectrometry of sEVs detected endogenous opioid peptide leu-enkephalin. sEVs from naïve female mice have higher levels of leu-enkephalin compared to male, matching the analgesic onset of leu-enkephalin in male recipient mice. In investigating the long-term effect of sEVs, we observed that a single prophylactic intrathecal injection of sEVs two weeks prior to induction of the pain model in recipient mice accelerated recovery from inflammatory pain after complete Freund's adjuvant (CFA) injection. Our exploratory studies examining immune cell populations in spinal cord and dorsal root ganglion using ChipCytometry suggested alterations in immune cell populations 14 days post-CFA. Flow cytometry confirmed increases in CD206⁺ macrophages in the spinal cord in sEV-treated mice. Collectively, these studies demonstrate multiple mechanisms by which sEVs can attenuate pain.

Keywords: ChipCytometry; Exosomes; Extracellular vesicles; Inflammatory pain; Neuropathic pain.

Fig. 1. Characterization of male mouse serum-derived sEVs.

a Confirmation of mechanical hypersensitivity in male mice after SNI. Baseline testing (BL) was followed by SNI or sham surgery and mice were tested for mechanical sensitivity for 14 days. Animals were sacrificed after two weeks and sEVs isolated from their serum were used in all studies. Data shown are mean ± standard error of the mean, n=6. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Tukey posttest ** p <0.01, *** p <0.001, **** p <0.0001. b Nanoparticle tracking analysis (NTA) showing the mean concentrations of sEVs in serum of naïve, sham control and SNI model mice respectively. c The sEV particle size distribution shows particle sizes of sEVs from SNI model mice were smaller compared to the control groups. The mean diameters were 121.1 ± 4.1 nm for naïve, 115.7 ± 5.2 nm for sham and 101.1 ± 2.4 nm for SNI sEVs. One-way ANOVA followed by Tukey posttest ** p <0.01. d Particle size distribution and diameter range for sEVs from naïve mice e TEM images showing sEV morphology and size (top) and immunogold labeling for CD81 in sEVs (bottom) from SNI model. f Western blot for sEV markers and sEV-depleted serum from different groups of mice confirming the presence of exosome marker protein CD81. Negative marker albumin was absent in sEVs but present in the serum. g Nanoparticle tracking analysis (NTA) size distributions NIST bead standards (yellow), serum-derived sEVs and macrophage derived sEVs (n=2–4 samples per condition). h Concatenated microflow cytometry plots of calibration beads (polystyrene-dark purple and silica-light purple), serum derived sEVs (naïve-blue and SNI-red), and RAW 264.7 macrophage derived sEVs from cells with (Exo+ pink) or cells without (Exo green) LPS stimulation (n=2–4 samples per condition). i Representative gating scheme of CFSE positive sEVs for sEVs labeled by CFDA-SE with or without Triton. j Plot demonstrating the linear correlation of all sample protein concentrations with either microflow cytometry (purple) or NTA (yellow) particle concentrations.

Fig. 1. Characterization of male mouse serum-derived sEVs.

a Confirmation of mechanical hypersensitivity in male mice after SNI. Baseline testing (BL) was followed by SNI or sham surgery and mice were tested for mechanical sensitivity for 14 days. Animals were sacrificed after two weeks and sEVs isolated from their serum were used in all studies. Data shown are mean ± standard error of the mean, n=6. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Tukey posttest ** p <0.01, *** p <0.001, **** p <0.0001. b Nanoparticle tracking analysis (NTA) showing the mean concentrations of sEVs in serum of naïve, sham control and SNI model mice respectively. c The sEV particle size distribution shows particle sizes of sEVs from SNI model mice were smaller compared to the control groups. The mean diameters were 121.1 ± 4.1 nm for naïve, 115.7 ± 5.2 nm for sham and 101.1 ± 2.4 nm for SNI sEVs. One-way ANOVA followed by Tukey posttest ** p <0.01. d Particle size distribution and diameter range for sEVs from naïve mice e TEM images showing sEV morphology and size (top) and immunogold labeling for CD81 in sEVs (bottom) from SNI model. f Western blot for sEV markers and sEV-depleted serum from different groups of mice confirming the presence of exosome marker protein CD81. Negative marker albumin was absent in sEVs but present in the serum. g Nanoparticle tracking analysis (NTA) size distributions NIST bead standards (yellow), serum-derived sEVs and macrophage derived sEVs (n=2–4 samples per condition). h Concatenated microflow cytometry plots of calibration beads (polystyrene-dark purple and silica-light purple), serum derived sEVs (naïve-blue and SNI-red), and RAW 264.7 macrophage derived sEVs from cells with (Exo+ pink) or cells without (Exo green) LPS stimulation (n=2–4 samples per condition). i Representative gating scheme of CFSE positive sEVs for sEVs labeled by CFDA-SE with or without Triton. j Plot demonstrating the linear correlation of all sample protein concentrations with either microflow cytometry (purple) or NTA (yellow) particle concentrations.

Fig. 2. miRNAs in serum-derived sEV samples of male mice.

a Principle component analysis for miRNAs in serum-derived sEVs from naïve, sham control, and two weeks post SNI model mice. One of the naïve sEV samples (ID: naive1) was considered as an outliner and was excluded from further analyses. Hierarchical clustering analysis of overall miRNA expression level with higher expression denoted by red and lower by blue shown and more related terms are closely grouped together. Pairwise distances between miRNAs shown for SNI model, sham and naïve control mice (n=3–4). b Venn diagram showing presence/absence analysis of miRNAs in sEVs from serum. HTG EdgeSeq/RNA-Seq detected unique and overlapping miRNAs in sEVs purified from the serum of SNI, sham, and naive control mice (n=4) two weeks after surgery. c A subset of differentially expressed miRNAs with the highest fold change in sEVs from SNI model compared to naïve control. d Gene Ontology (GO) terms and KEGG pathways enriched for the genes targeted by the miRNAs. Annotations are filtered to reduce redundancy, with no more than 80% overlap in the target genes. Only the annotations (FDR<=0.01) resulting from the miRNAs that are significantly differentially expressed between SNI vs. naive conditions are shown. Additional annotations from miRNAs expressed explicitly in SNI or naive conditions are available in the supplementary files.

Fig. 3. Intrathecal injection of 10 μg male serum-derived sEVs transiently increased basal mechanical but not the thermal threshold.

a Schematic of in vivo experiment. sEVs purified from serum of male C57BL/6 mice 2 weeks post SNI surgery or naïve mice were administrated intrathecally into another group of naïve male recipient mice followed by behavior tests. b Mechanical threshold was measured by von Frey filaments before and after the single intrathecal injection of sEVs. The recipient mice showed an increase in basal mechanical threshold (n=15). c sEVs did not impact basal thermal pain threshold in mice. Thermal threshold was measured by Hargreaves test before and after sEV injection. The recipient mice showed no significant difference (n=6). Data shown are mean ± SEM. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Bonferroni posttest * p <0.05, ** p <0.01, *** p <0.001. d Treatment with sEVs from naïve mice changed basal weight distribution of hind paws in recipient mice. The ratio of body weight distribution on two rear paws (RL/RR) was measured by dynamic weight bearing system at different time points after sEVs injection. Statistical analysis determined by two-way repeated-measures ANOVA followed by Bonferroni posttest ** p <0.01 PBS vs. naïve sEVs, # p<0.05 naïve sEVs vs SNI sEVs (n=5).

Fig. 4. Detection of leu-enkephalin in sEVs and reversal of short-term mechanical antinociception in male mice induced by naïve male and female serum-derived sEVs using naltrexone.

a Liquid chromatography with tandem mass spectrometry of serum sEVs samples. Representative MS/MS spectrum showing the peaks of mass-to-charge ratios (m/z) at 397 and 425 derived from the collision-induced dissociation of leu-enkephalin in sEVs. b Confirmation of leu-enkephalin in sEVs by competitive enzyme immunoassay. Levels of leu-enkephalin in serum-derived sEVs and serum from naïve male and female mice. Both sEV and serum samples were pooled from n=5 mice of each sex. The standard curve for leu-enkephalin is also shown. c Quantification of leu-enkephalin from the competitive enzyme immune assay. d Schematic and timeline of of in vivo experiments in mice administrated with either leu-enkephalin alone or naltrexone (i.p., 10mg/kg) and/or naïve sEVs (i.t., 10 μg/mice) at the same time, with behavior testing 15 min, 3h and 24h post-administration. e Dose response of i.t. leu-enkephalin injections into male recipient mice and mechanical withdrawal thresholds as measured by von Frey. f Male recipients that had received male sEVs had higher mechanical threshold than the group receiving both naltrexone and male sEVs at 3 and 24 hours post-injection. (n=5–11). g Male recipients who had received female sEVs had higher mechanical threshold than the group receiving both naltrexone and male sEVs at 15 minutes post-injection (n=5). Data shown are mean ± SEM. Statistical analysis for e & g was determined by two-way repeated-measures ANOVA followed by Bonferroni test. Analysis for f was determined by a mixed-effects model with Tukey post-hoc group comparisons. * p <0.05, ** p <0.01, *** p <0.001.

Fig. 5. The pretreatment with 10 μg of male serum-derived sEVs caused a transient delay in the development of mechanical allodynia in SNI model of male mice.

a Schematic of experiment showing the serum-derived sEVs from naïve and SNI model donor mice two weeks after injury injected intrathecally into another group of naïve recipient mice. After two weeks, the SNI surgery was performed on these recipient mice followed by behavior tests to assess the effect of sEVs on SNI-induced pain. b Mechanical hypersensitivity measured by von Frey filaments. The recipient mice that received 10 μg naïve or SNI sEVs then underwent SNI surgery showed increased mechanical threshold at 1d and 3d compared to PBS control (n=9). These results indicate sEVs from both naïve and SNI model can induce a transient delay in SNI-induced mechanical hypersensitivity in the initial stages. c There was no significant differences in thermal hypersensitivity as measured by Hargraves after SNI surgery. Data shown are mean ± SWM. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Bonferroni test, **** p < 0.0001.

Fig. 6. Prophylactic intrathecal injections of 10 μg of male serum-derived sEVs promoted resolution of CFA induced mechanical hypersensitivity in male mice.

a Schematic representation of experimental design. The sEVs from male C57BL/6 naïve and SNI model donor mice two weeks after injury were injected intrathecally into nine weeks old male C57BL/6 recipient mice. CFA model was established in these recipient mice 2 weeks post sEVs injection. b Mechanical allodynia in CFA model mice. Mice were prophylactically injected with 10 μg sEVs or PBS two weeks prior to CFA. Paw withdrawal thresholds were measured at different time points after CFA injection (n=6). Ten μg sEV accelerated recovery from CFA-induced mechanical hypersensitivity compared with PBS control group. Multiple comparisons between different time points in CFA model are shown on the right. c There were no significant differences in thermal hypersensitivity as measured by Hargreaves test. Data shown are mean ± SEM. Multiple comparation between different time points in CFA model is shown in adjacent table. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Bonferroni test * p <0.05, *** p < 0.001, ### p < 0.001, **** p < 0.0001. d Pretreatment with naive sEVs reverse CFA-induced weight distribution changes on rear paws. sEVs from male C57BL/6 naïve mice were injected intrathecally into nine weeks old male C57BL/6 recipient mice. CFA was injected into rear right (RR) paw of these recipient mice 2 weeks post sEV injection. The ratio of body weight distribution on two rear paws (RL/RR) were measured by dynamic weight bearing system at different time points after CFA injection (n=5). Ten μg naïve sEV significantly reversed CFA-induced decreases of body weight distribution (RL/RR) on the injured paw after three days compared to PBS control group. Though sEVs from SNI model had a similar trend, it was not significant. The significance was observed on day 3 between SNI sEVs and PBS control. Data shown are mean ± SEM. Statistical analysis was determined by two-way repeated-measures ANOVA followed by Bonferroni posttest * p <0.05.

Fig. 7. Immune markers present on male serum-derived sEVs of naïve and SNI mice, as well as naïve female mice.

a-e Comparison of tetraspanin composition, leukocyte and macrophage markers and characteristics of sEV derived from male naïve and SNI model mice. sEVs were captured using specific antibody-coated spots against CD81, CD9, F4/80, and CD45 and analyzed using the ExoView R100 platform. Counts are after subtraction of mouse IgG isotype control background. a Total fluorescent particle counts of sEVs on tetraspanin capture spots analyzed using fluorescent antibodies against CD81/CD9/F4/80 and CD45. b Size distribution of sEVs on tetraspanin capture spots analyzed using the SP-IRIS mode of the ExoView R100 platform. c sEVs captured per label. Plotted is the mean ± SEM of three independent biological replicates. d Radar plots of the distribution of markers detected on sEVs captured by the respective antibody marker. e Representative visualization of the individual vesicles captured on a single assay spot with respective fluorescent detector markers (CD9- blue, CD45- red, F4/80- green) with naïve sEVS on top and SNI sEVs on bottom. f-h Characterization of markers on serum-derived sEVs using a MACSPlex flow cytometry bead-based capture and fluorescent detection assay. Comparisons are normalized within data set and weighted to represent alterations in both percent and fluorescent intensity. f sEVs from male naïve control and SNI model mice. sEVs used were from a pool of serum of 5–6 mice g serum-derived sEVs from male and female naive control mice, isolated from 3 separate pools of n=10 mice per pool (n=3). h The geoMFI (geometric mean fluorescence intensity) of male and female sEVS of different capture antibody beads represented in g.

Fig. 7. Immune markers present on male serum-derived sEVs of naïve and SNI mice, as well as naïve female mice.

a-e Comparison of tetraspanin composition, leukocyte and macrophage markers and characteristics of sEV derived from male naïve and SNI model mice. sEVs were captured using specific antibody-coated spots against CD81, CD9, F4/80, and CD45 and analyzed using the ExoView R100 platform. Counts are after subtraction of mouse IgG isotype control background. a Total fluorescent particle counts of sEVs on tetraspanin capture spots analyzed using fluorescent antibodies against CD81/CD9/F4/80 and CD45. b Size distribution of sEVs on tetraspanin capture spots analyzed using the SP-IRIS mode of the ExoView R100 platform. c sEVs captured per label. Plotted is the mean ± SEM of three independent biological replicates. d Radar plots of the distribution of markers detected on sEVs captured by the respective antibody marker. e Representative visualization of the individual vesicles captured on a single assay spot with respective fluorescent detector markers (CD9- blue, CD45- red, F4/80- green) with naïve sEVS on top and SNI sEVs on bottom. f-h Characterization of markers on serum-derived sEVs using a MACSPlex flow cytometry bead-based capture and fluorescent detection assay. Comparisons are normalized within data set and weighted to represent alterations in both percent and fluorescent intensity. f sEVs from male naïve control and SNI model mice. sEVs used were from a pool of serum of 5–6 mice g serum-derived sEVs from male and female naive control mice, isolated from 3 separate pools of n=10 mice per pool (n=3). h The geoMFI (geometric mean fluorescence intensity) of male and female sEVS of different capture antibody beads represented in g.

Fig. 8. ChipCytometry profiling of immune cell populations in the spinal cord of day 14 CFA male mice after receiving male serum-derived sEVs from naïve and SNI mice.

Alterations in immune cell population in 5–7 μM sections spinal cord 14 days after CFA of sEV treated mice determined using ChipCytometry. Groups include naïve, PBS+CFA, male naïve sEVs+CFA, and male SNI sEVs+CFA (n=2). a An example of image overlay in spinal cord. b Immune cells in spinal cord 14 days post CFA as count c, d Percentage of CD45⁺ cells in spinal cord represented as pie charts (c) and bar graph (d). Bar graphs showing e NK cells f NKT cells g neutrophils and h macrophages and CD206⁺ macrophages.

Fig. 8. ChipCytometry profiling of immune cell populations in the spinal cord of day 14 CFA male mice after receiving male serum-derived sEVs from naïve and SNI mice.

Alterations in immune cell population in 5–7 μM sections spinal cord 14 days after CFA of sEV treated mice determined using ChipCytometry. Groups include naïve, PBS+CFA, male naïve sEVs+CFA, and male SNI sEVs+CFA (n=2). a An example of image overlay in spinal cord. b Immune cells in spinal cord 14 days post CFA as count c, d Percentage of CD45⁺ cells in spinal cord represented as pie charts (c) and bar graph (d). Bar graphs showing e NK cells f NKT cells g neutrophils and h macrophages and CD206⁺ macrophages.

Fig. 9. ChipCytometry profiling of immune cell populations in the dorsal root ganglia of day 14 CFA male mice after receiving male serum-derived sEVs from naïve and SNI mice.

Alterations in immune cell population in 5–7 μm sections of DRG 14 days after CFA of sEV treatment determined using ChipCytometry. Groups include naïve, PBS+CFA, male naïve sEVs+CFA, and male SNI sEVs+CFA (n=2). a An example sample image overlay in DRG. b CD45⁺ immune cells in DRG 14 days post CFA as count. c, d Percentage of CD45⁺ cells in DRG as pie chart (c) and as bar graph (d). Bar graphs showing e NK cells f NKT cells g neutrophils and h macrophages and M2 macrophages (CD206⁺).

Fig. 9. ChipCytometry profiling of immune cell populations in the dorsal root ganglia of day 14 CFA male mice after receiving male serum-derived sEVs from naïve and SNI mice.

Alterations in immune cell population in 5–7 μm sections of DRG 14 days after CFA of sEV treatment determined using ChipCytometry. Groups include naïve, PBS+CFA, male naïve sEVs+CFA, and male SNI sEVs+CFA (n=2). a An example sample image overlay in DRG. b CD45⁺ immune cells in DRG 14 days post CFA as count. c, d Percentage of CD45⁺ cells in DRG as pie chart (c) and as bar graph (d). Bar graphs showing e NK cells f NKT cells g neutrophils and h macrophages and M2 macrophages (CD206⁺).

Fig. 10. Flow cytometry profiling of immune cell populations in the spinal cord and dorsal root ganglia of day 14 CFA male mice after receiving male or female serum-derived sEVs from naïve mice.

Alterations in immune cell population from whole lumbar spinal cord sections and ipsilateral DRG of male mice 14 days after CFA. Mice received either male or female serum-derived naïve sEVs 14 days prior to CFA. Groups are represented as naïve, vehicle+CFA, male sEVs+CFA, or female sEVs+CFA (n=5). a Distribution of immune cells in the spinal cord represented as % of all CD45⁺ cells or without microglia and macrophages. b Bar graphs of NK cell and microglia populations in the spinal cord. c Bar graphs showing percentage of total macrophages (%CD45⁺) or CD206⁺ and CD86⁺ macrophages (%macrophages) in the spinal cord. d Distribution of immune cells in the DRG represented as % of all CD45⁺ cells. e Bar graphs depicting percentage of NK cells (%CD45⁺) or mature NKp46⁺ NK Cells (%NK cells) in the DRG. f Bar graphs showing percentage of total macrophages (%CD45⁺) or CD206⁺ and CD86⁺ macrophages (%macrophages) in the DRG. g Bar graphs showing the percentage of neutrophils (%CD45⁺) or activated CD11b⁺ neutrophils (%NK cells) in the DRG. Data shown are mean ± SEM. Statistical analysis was determined by one-way ANOVA followed by Bonferroni posttest * p <0.05, ** p <0.01, *** p <0.001.