Doxycycline attenuates l-DOPA-induced dyskinesia through an anti-inflammatory effect in a hemiparkinsonian mouse model

Abstract

The pharmacological manipulation of neuroinflammation appears to be a promising strategy to alleviate l-DOPA-induced dyskinesia (LID) in Parkinson's disease (PD). Doxycycline (Doxy), a semisynthetic brain-penetrant tetracycline antibiotic having interesting anti-inflammatory properties, we addressed the possibility that this compound could resolve LID in l-DOPA-treated C57BL/6 mice presenting either moderate or intermediate lesions of the mesostriatal dopaminergic pathway generated by intrastriatal injections of 6-OHDA. Doxy, when given subcutaneously before l-DOPA at doses of 20 mg kg^-1 and 40 mg kg^-1, led to significant LID reduction in mice with moderate and intermediate dopaminergic lesions, respectively. Importantly, Doxy did not reduce locomotor activity improved by l-DOPA. To address the molecular mechanism of Doxy, we sacrificed mice with mild lesions 1) to perform the immunodetection of tyrosine hydroxylase (TH) and Fos-B and 2) to evaluate a panel of inflammation markers in the striatum, such as cyclooxygenase-2 and its downstream product Prostaglandin E2 along with the cytokines TNF-α, IL-1β and IL-6. TH-immunodetection revealed that vehicle and Doxy-treated mice had similar striatal lesions, excluding that LID improvement by Doxy could result from neurorestorative effects. Importantly, LID inhibition by Doxy was associated with decreased Fos-B and COX-2 expression and reduced levels of PGE₂, TNF-α, and IL-1β in the dorsolateral striatum of dyskinetic mice. We conclude 1) that Doxy has the potential to prevent LID regardless of the intensity of dopaminergic lesioning and 2) that the anti-inflammatory effects of Doxy probably account for LID attenuation. Overall, the present results further indicate that Doxy might represent an attractive and alternative treatment for LID in PD.

Keywords: COX-2; IL-1β; PGE2; Parkinson’s disease; TNF-α; cytokines; neuroinflammation; tetracyclines.

FIGURE 1

Schematic drawing of Doxy pharmacological treatments in hemiparkinsonian mice with established LID. Eighty-five mice were selected after 6-OHDA stereotaxic surgery (solutions of 2.5 or 3.75 µg µl⁻¹ of 6-OHDA were used for Doxy 20 mg kg⁻¹ or 40 mg kg⁻¹ treatment groups, respectively) employing the apomorphine rotational test (0.5 mg.kg-1 s. c.). These mice were then treated chronically with l-DOPA for 21 days (25 mg kg⁻¹ + benserazide 10 mg kg⁻¹ i. p. once/day) and scored for AIMs. Once AIMs had been evaluated, hemiparkinsonian dyskinetic mice were divided into Veh- and Doxy (20 and 40 mg.kg⁻¹ s. c.)-treated groups with equivalent AIMs scores. Then, a sub-chronic treatment with Doxy + L-DOPA or Veh + L-DOPA was administered for the next 5 days. During this period, mice were scored for AIMs on days 1, 3, and 5.

FIGURE 2

Impact of sub-chronic treatment with Doxy on established AIMs and locomotor activity in hemiparkinsonian mice with partial lesions. (A) AIMs observed in hemiparkinsonian mice with mild striatal lesions were significantly reduced at days 3 and 5 of sub-chronic treatment with Doxy (20 mg.kg⁻¹ s. c.). (B) At day 5 of treatment with 20 mg kg⁻¹, AIMs were significantly reduced 30 and 60 min after initiating l-DOPA treatment. (C) When given at 40 mg.kg⁻¹ s. c. to hemiparkinsonian mice with intermediate striatal lesions, Doxy effectively reduced AIMs from day 1 of treatment until day 5. (D) At day 5 of treatment with 40 mg kg⁻¹, AIMs were significantly reduced 30, 60, and 90 min after initiating l-DOPA treatment. (E) The total distance traveled and (F) the average speed of lesioned mice were increased by l-DOPA, and the co-administration of Doxy did not modify l-DOPA’s beneficial effects. a–d: *p < 0.05 vs. Veh (Data represent the mean ± SEM; r-MANOVA followed by Bonferroni post-hoc test, n = 8/group). e–f: *p < 0.05 vs 6-OHDA (Data represent the mean ± SEM; One way-ANOVA followed by Bonferroni post-hoc test, n = 8/group).

FIGURE 3

Box plot charts showing the impact of treatment with 20 and 40 mg kg⁻¹ of Doxy on specific subtypes of AIMs. (A–C) Impact of Doxy 20 mg kg⁻¹ on axial (A) limb (B) and orofacial (C) AIMs compared to the Veh-treated group in hemiparkinsonian mice with moderate striatal lesions. (D–F) Impact of Doxy 40 mg kg⁻¹ on axial (D) limb (E) and orofacial (F) AIMs compared to the Veh-treated group in hemiparkinsonian mice with intermediate striatal lesions. Box plots represent the summary of the dataset for each parameter (minimum, lower quartile, median, upper quartile, and maximum). *p < 0.05 vs. Veh, Mann-Whitney test, n = 8/group. Note that the p-value in b is close to significance (p = 0.07).

FIGURE 4

Analysis of the impact of Doxy treatment on TH, Fos-B, and COX-2 expression within the dorsolateral striatum of mice with partial lesions. (A) Quantitative optical density of striatal TH immunolabelling in coronal sections of the striatum of lesioned mice treated or not with 20 mg kg⁻¹. (B) Photomicrographs of coronal sections of the rostral striatum illustrate that the loss of TH immunosignal is limited to the dorsolateral area of the striatum. (C) Chronic treatment with l-DOPA increased Fos-B expression in the lesioned mice’s dorsolateral area of ​​the striatum, and co-treatment with 20 m kg⁻¹ Doxy significantly reduced the number of Fos-B⁺ cells. (D) Representative photomicrographs of Fos-B-ir cells in the dorsolateral striatum of Veh/l-DOPA and Doxy/l-DOPA-treated mice. (E) Positive correlation between the intensity of AIMs and Fos-B expression in lesioned mice treated with Veh or Doxy + l-DOPA. (F) The number of COX-2⁺ cells increased mainly in the dorsolateral striatum of lesioned mice after l-DOPA treatment and co-treatment with 20 m kg⁻¹ Doxy significantly reduced the number of these cells. (G) Representative photomicrographs describing COX-2-ir cells in the lesioned striatum of Veh/l-DOPA and Doxy 20/l-DOPA-treated animals. (H) Positive correlation between the intensity of AIMs and the number of COX-2⁺ cells in mice treated with Veh or Doxy + l-DOPA. a, c and (F) *p < 0.05 vs unlesioned side; Two-way ANOVA followed by Bonferroni analysis, n = 8/group. e and (H) Spearman correlation. Values ​​are presented as mean ± SEM. cc: corpus callosum. ND: non-detectable. The red dashed line area corresponds to the striatum portion in which TH, Fos-B, and COX-2 were quantified. Scale Bar = 1 mm (B) and 50 µm (D–G).

FIGURE 5

Effect of Doxy co-treatment on the production of PGE₂ and pro-inflammatory cytokines in mice with partial lesions. (A) l-DOPA treatment increased PGE₂ levels in the striatum of lesioned mice, and the co-treatment with Doxy 20 mg kg⁻¹ suppressed this effect. (B) Doxy was also similarly effective in preventing the increase of TNF-α striatal levels induced by l-DOPA treatment. (C) Doxy only partially reduced the elevation of IL-1β levels observed after l-DOPA treatment. (D) The co-treatment with Doxy had no impact on the increase of IL-6 levels caused by l-DOPA treatment. *p < 0.05 vs 6-OHDA; #p < 0.05 vs Veh (One-way ANOVA followed by Bonferroni. n = 6/group). Values (pg/ml) are presented as mean ± SEM.