Effect of human mesenchymal stem cell secretome administration on morphine self-administration and relapse in two animal models of opioid dependence

Abstract

The present study investigates the possible therapeutic effects of human mesenchymal stem cell-derived secretome on morphine dependence and relapse. This was studied in a new model of chronic voluntary morphine intake in Wistar rats which shows classic signs of morphine intoxication and a severe naloxone-induced withdrawal syndrome. A single intranasal-systemic administration of MSCs secretome fully inhibited (>95%; p < 0.001) voluntary morphine intake and reduced the post-deprivation relapse intake by 50% (p < 0.02). Since several studies suggest a significant genetic contribution to the chronic use of many addictive drugs, the effect of MSCs secretome on morphine self-administration was further studied in rats bred as high alcohol consumers (UChB rats). Sub-chronic intraperitoneal administration of morphine before access to increasing concentrations of morphine solutions and water were available to the animals, led UChB rats to prefer ingesting morphine solutions over water, attaining levels of oral morphine intake in the range of those in the Wistar model. Intranasally administered MSCs secretome to UChB rats dose-dependently inhibited morphine self-administration by 72% (p < 0.001); while a single intranasal dose of MSC-secretome administered during a morphine deprivation period imposed on chronic morphine consumer UChB rats inhibited re-access morphine relapse intake by 80 to 85% (p < 0.0001). Both in the Wistar and the UChB rat models, MSCs-secretome administration reversed the morphine-induced increases in brain oxidative stress and neuroinflammation, considered as key engines perpetuating drug relapse. Overall, present preclinical studies suggest that products secreted by human mesenchymal stem cells may be of value in the treatment of opioid addiction.

Fig. 1. Simultaneous intranasal and intravenous administration of secretome derived from human preconditioned MSCs to Wistar rats that had chronically consumed morphine inhibits morphine intake and reduces post-deprivation morphine re-access intake.

A Daily morphine intake and (B) morphine preference over water. Wistar rats that had voluntarily ingested morphine for 21 days were simultaneously intranasally and intravenously treated with secretome (25μg protein) derived from 1 × 10⁶ preconditioned MSCs or vehicle (indicated by the arrow). Morphine intake was inhibited by 95% (p < 0.0001). Two-way ANOVA (treatment x day) of morphine intake and morphine preference following secretome administration (blue circles) indicate significant effect of treatment (F_(1,435) = 158.9 ****p < 0.0001), day (F_(28,435) = 30.4, p < 0.0001) and significant interaction (F treatment x day _(28,435) = 25.2, p < 0.0001) compared with control animals receiving vehicle (red circles). C Morphine relapse intake and (D) morphine preference over water after 12 days of morphine deprivation followed by 24-hours of morphine re-access. Rats treated previously with three intranasal and intravenous secretome doses (on day 22 of chronic morphine access and on days 2 and 9 of the deprivation period) (blue bars) ingested significantly lower (−50%) amount of morphine compared with vehicle-treated animals (red bar) (**p < 0.01, two-tailed Student t test). Data are presented as mean ± SEM; n = 9 for each experimental condition.

Fig. 2. Simultaneous intranasal and intravenous administration of secretome derived from human preconditioned MSCs normalizes the increased astrocyte reactivity and brain oxidative stress induced by chronic morphine intake in Wistar rats.

A Representative confocal microphotographs of GFAP immunoreactivity (green top) in hippocampal astrocytes and Iba-1 immunoreactivity (red, shown by arrows, center) in hippocampal microglia. Nuclei were counterstained with DAPI (blue, nuclear marker), scale bar 25 μm. B Quantification of length and (C) thickness of primary astrocytic processes. (D) Quantification of microglial density. Chronic morphine-drinking rats treated with vehicle displayed a marked increase in the length and thickness of astrocytic processes (**** p < 0.0001, One-way ANOVA followed by Tukey’s post-hoc test) and microglial density (red bars in B–D) (**p < 0.01, One-way ANOVA followed by Tukey’s post-hoc test) compared with water drinking rats (green bars in B–D). The simultaneous intranasal and intravenous administration of three secretome doses (on day 22 of the chronic morphine access and on days 2 and 9 of the deprivation period) significantly reduced the length and thickness of primary astrocytic processes (blue bars in B and C) (****p < 0.0001, One-way ANOVA followed by Tukey’s post-hoc test), but not the microglial density (blue bar in D) compared with the morphine + vehicle treated animals (red bars in B–D). (E) GSSG/GSH ratio in the hippocampus of rats chronically drinking morphine, deprived for 12 days, and allowed morphine re-access for 24 hours. Chronic morphine drinking rats treated with vehicle showed a three-fold increase in the GSSG/GSH ratio (red bar) (*p < 0.05, One-way ANOVA followed by Tukey’s post-hoc test) compared with rats drinking only water (green bar). The simultaneous intranasal and intravenous administration of three secretome doses (on day 22 of the chronic morphine access and on days 2 and 9 of the deprivation period) resulted in the full normalization of the GSSG/GSH ratio (blue bar) (*p < 0.05, One-way ANOVA followed by Tukey’s post-hoc test). F MDA levels in the neostriatum. Chronic morphine drinking rats treated with vehicle showed a two-fold increase in MDA levels (red bar) (****p < 0.0001, One-way ANOVA followed by Tukey’s post-hoc test) compared with rats drinking only water (green bar). Simultaneous intranasal and intravenous administration of three secretome doses resulted in the near complete normalization of the MDA levels (blue bar) (**p < 0.01, One-way ANOVA followed by Tukey’s post-hoc test). Data are presented as mean ± SEM; n = 9 for each experimental condition.

Fig. 3. Simultaneous intranasal and intravenous administration of secretome derived from human preconditioned MSCs to Wistar did not modify the morphine-induced increases in the brain expression of μ-opioid receptor or glutamate transporters.

Quantification of the mRNA levels of the (A) μ-opioid receptor OPRM1; (B) xCT glutamate transporter and (C) GLT-1 glutamate transporter, determined by RT-qPCR in prefrontal cortex (top) and nucleus accumbens (bottom) of rats chronically drinking morphine, deprived for morphine for 12 days and allowed morphine re-access for 24 hours. Data were normalized against the mRNA level of the housekeeping gene GAPDH in the same sample. Morphine intake induced a significant increase in μ-opioid receptor mRNA levels in prefrontal cortex and nucleus accumbens (red bars in A) (*p < 0.05, One-way ANOVA followed by Tukey’s post-hoc test), in xCT mRNA levels in prefrontal cortex and nucleus accumbens (red bars in B) (*p < 0.05, **p < 0.01 One-way ANOVA followed by Tukey’s post-hoc test), and in GLT-1 mRNA levels in prefrontal cortex (red bars in C) (*p < 0.05, One-way ANOVA followed by Tukey’s post-hoc test), compared with water drinking animals (green bars). Simultaneous intranasal and intravenous administration of three secretome doses (on day 22 of the chronic morphine access and on days 2 and 9 of the deprivation period) had no impact on the mRNA levels of these molecules. Data are presented as mean ± SEM; n = 9 for each experimental condition.

Fig. 4. Repeated intranasal administration of secretome derived from human preconditioned MSCs to UChB rats that had chronically consumed morphine inhibits its intake and reduces post-deprivation relapse.

A-Left Daily morphine intake and (B-Left) morphine preference over water. UChB rats that had consumed morphine for 76 days were intranasally treated with three doses of secretome derived from preconditioned MSCs or vehicle (indicated by arrows). Two-way ANOVA (treatment × day) of morphine intake following three intranasal doses of MSC secretome indicates significant effect of treatment [F_(1,180) = 195.4, p < 0.0001], day [F_(29,180) = 8.76, p < 0.0001], and significant interaction [F_(29,180) = 10.15, p < 0.001] compared with control rats receiving vehicle. Tukey´s post-hoc analysis revealed that secretome treatment (blue circles) inhibited morphine intake during the 14 days recorded period, versus vehicle-treated control (red circles) (***p < 0.001). The inhibition of morphine intake induced by the third secretome dose (72%) was significantly higher (**p < 0.01) than the inhibition induced by the second dose (50%) and the inhibition induced by the latter was significantly higher (*p < 0.05) than the inhibition induced by the first dose (34%) (two-tailed Student t test). (A Right) Morphine intake and (B Right) morphine preference over water of UChB rats after five days of morphine deprivation followed by one day of free choice of 44 mg/l and 50 mg/l morphine re-access. Rats treated previously with four intranasal secretome doses (blue bar) ingested a 74% lower amount of morphine than vehicle treated animals (red bar) (****p < 0.0001; Two-tailed Student t test). (C Left) Daily water intake showing that intranasal administration of three secretome doses (blue circles) significantly increased water intake compared with vehicle treated rats (red circles). Two-way ANOVA (treatment × day) indicates significant effect of treatment [F_(1,216) = 213.9, p < 0.0001], day [F_(35,216) = 104.7, p < 0.0001], and significant interaction [F_{(35, 216)} = 11.92, p < 0.0001]. Tukey’s post-hoc analysis revealed that secretome treatment increases water intake during the 14 days recorded versus vehicle-treated control (***p < 0.001). The higher water intake induced by the third secretome dose (80%) was significantly higher than the increase induced by the first dose (57%) (*p < 0.05, two-tailed Student t test). (C Right) Water intake after five days of morphine deprivation followed by one day of 44 mg/l and 50 mg/l morphine solutions re-access. Rats treated previously with four intranasal secretome doses (blue bar) ingested a significantly higher amount of water than vehicle control animals (red bar) (****p < 0.0001, Two-tailed Student t test). (D) Total fluid intake and animals body weights of UChB rats intranasally treated with secretome derived from preconditioned MSCs (blue circles) or vehicle (red circles). Data showed that body weight or total fluid intake were not affected by secretome administration, indicating that effects induced by this treatment were specific for morphine intake. Data are presented as mean ± SEM; n = 4 for each experimental condition.

Fig. 5. Intranasal administration of secretome derived from human preconditioned MSCs reduces morphine-induced astrocyte activation and brain oxidative stress in UChB rats.

A Representative confocal microphotographs of GFAP immunoreactivity (red top) for hippocampal astrocytes and Iba-1 immunoreactivity (green, shown by arrows, center) for hippocampal microglia. Nuclei were counterstained with DAPI (blue, nuclear marker), scale bar 25 μm. B Quantification of length and (C) thickness of primary astrocytic processes. D Quantification of microglia density. Chronic morphine-ingesting rats treated with vehicle displayed a marked increase in the length and thickness of astrocytic processes (****p < 0.0001, One-way ANOVA followed by Tukey’s post-hoc test) and microglial density (red bars in B–D) (***p < 0.001, One-way ANOVA followed by Tukey’s post-hoc test) compared with water drinking rats (green bars in B–D). The intranasal administration of secretome significantly reduced the length and thickness of primary astrocytic processes (blue bars in B and C) (****p < 0.0001, One-way ANOVA followed by Tukey’s post-hoc test) compared with the morphine vehicle treated animals (red bars in B and C). E Quantification of GSSG/GSH ratio in the hippocampus. Chronic morphine drinking rats treated with vehicle showed a 2.5-fold increase in GSSG/GSH ratio (red bar) (***p < 0.001, One-way ANOVA followed by Tukey’s post-hoc test) compared with rats drinking only water (green bar). Intranasal administration of secretome resulted in the fully normalization of the GSSG/GSH ratio (blue bar) (***p < 0.001, One-way ANOVA followed by Tukey’s post-hoc test). F Quantification of MDA levels in neostriatum. Chronic morphine drinking rats treated with vehicle showed a 60% increase in MDA levels (red bar) (**p < 0.01, One-way ANOVA followed by Tukey’s post-hoc test) compared with rats drinking only water (green bar). Intranasal secretome administration resulted in the full normalization of MDA levels (blue bar) (**p < 0.01, One-way ANOVA followed by Tukey’s post-hoc test). Data are presented as mean ± SEM; n = 4 for each experimental condition.

Fig. 6. A single intranasal administration of secretome derived from human preconditioned MSCs during the deprivation period to UChB rats that had chronically consumed morphine markedly reduces post-deprivation relapse.

A Morphine relapse intake and (B) Morphine preference over water of UChB rats that had voluntarily consumed morphine for 57 days were morphine deprived for six days and intranasally treated with a single dose of secretome derived from preconditioned MSCs or vehicle (indicated by the arrow) on day 61. Relapse morphine intake and morphine preference upon re-access to the 80 and 90 mg/l morphine solutions over water was significantly reduced in secretome treated rats (80% to 85% ****p < 0.0001, two-tailed Student t test) on the two re-access days (blue bars in A and B) compared with vehicle treated animals (red bars in A and B). C Daily water intake showing that secretome administration during the morphine deprivation period significantly increased water intake (days 64 and 65) (****p < 0.0001, Two-tailed Student t test) upon morphine re-access (blue bar) compared with vehicle-treated animals (red bar). D Total fluid intake and animals body weight of UChB rats intranasally treated with secretome derived from preconditioned MSCs (blue circles) or vehicle (red circles) during the deprivation period. Data shows that these parameters were not affected by secretome administration. Data are presented as mean ± SEM; n = 6 for each experimental condition.