Positive correlation between striatal MMPs and ROS with overall LID severity
The present study revealed an increase of ROS in the striatum of parkinsonian rats presenting LID. The neuroprotective effect of doxycycline has been attributed to its ability to scavenge ROS (Antonio et al., 2014). After chronic doxycycline treatment with L-DOPA, a robust correlation with ROS levels and LID intensity was established. Oxidative stress has been found to be one of the factors responsible for the initiation and progression of PD (Charvin et al., 2018; Zhou et al., 2008). An increase in MMPs may be a consequence of ROS production (Gottschall and Yu, 1995; Koli et al., 2008) and the presence of reactive astrocytes (Muir et al., 2002; Ogier et al., 2006). MMP-2/MMP-9 gelatinolytic activity and MMP-3 expression increased in the striatum of rats presenting L-DOPA dyskinesia. Alterations of MMP-2/MMP-9 activity and MMP-3 expression were positively correlated with LID. Also, our results showed that MMP-2/MMP-9 activity and MMP-2 and MMP-9 gel detection also increase after 6-OHDA lesion. Doxycycline is an inhibitor of metalloproteinases activity (Bahrami et al., 2011; Garcia-Martinez et al. 2010; Nikodemova et al., 2006). The doxycycline treatment attenuated the MMP-2/MMP-9 activity increase when given either acutely or chronically to rats expressing LID. Contrastingly, the MMP-2 and MMP-9 gel detection was decreased, primarily after chronic treatment. We cannot explain the former contrasting results, but it is important to note that gel detection permits the separated detection of each enzyme and provides distinct in situ detection of the general gelatinases activities.
Lorenzl et al. (2004), Annese et al. (2015), and De Stefano and Herrero (2017) described the up‐regulation of MMP-2/MMP‐9 activity in neurons and microglia in the striatum and SNc in the MPTP model of PD. In the MPTP model, Chung et al. (2013) found increased MMP-3 protein and activity levels. In the postmortem brains of patients with PD, alpha-synuclein, and MMP-3 are co-localized in Lewy bodies (Choi et al. 2014, 2011). In contrast, Lorenzl et al. (2002) detected a 50% reduction in MMP-2/MMP-9 activity levels in the SNc of patients with PD. Accordingly, our findings of alterations in MMP-2, MMP-9, and MMP-3 levels in the striatum of PD rats submitted to L-DOPA treatment consistent with the possibility that alterations in MMPs may contribute to LID pathogenesis.
Antimicrobial drugs in Parkinson’s disease
There is evidence that antimicrobial drugs have the potential to exert neuroprotective effects on central nervous system neurodegenerative diseases, such as PD (Amor et al., 2010; 2014; Bassani et al., 2015; Bortolanza et al., 2018; Champagne-Jorgensen et al., 2019; Santa Cecília et al., 2019).
The penicillin derivative ceftriaxone (beta-lactam antibiotic) has been found to prevent and reverse behavioral and neuronal deficits in the MPTP model of PD (Chotibut et al., 2014; Hsieh et al., 2017; Hsu, 2015; Huang et al., 2015; Weng et al., 2016). It increases the expression of glutamate transporter 1, which plays a major role in glutamate clearance in the central nervous system (Yimer et al., 2019). Ceftriaxone has been reported to slow the development of LID, but it does not change previously established LID (Chotibut et al. 2017; Kelsey and Neville 2014). The antibiotic poorly crosses the blood-brain barrier, and the concentrations used might be even higher than those for infectious disease treatment, thus exerting potent, selective pressure on commensal and pathogenic microbiota (Mattappalil and Mergenhagen, 2014).
Another first-line agent for the treatment of tuberculosis, rifampicin, readily crosses the blood-brain barrier and has also been used for the treatment of central nervous system infections (Yulug et al., 2014). Rifampicin reduces microglial inflammatory responses and neurodegeneration induced in vitro by α-synuclein fibrillary aggregates (Acuña et al., 2019). Rifampicin pre-treatment inhibits the rotenone toxicity induced in the PC12 cells (Lin et al., 2017; Wu et al., 2018) and rotenone-induced microglial inflammation (Liang et al., 2017). Paradoxically, rifampicin possesses strong pro-inflammatory properties (Yuhas et al., 2009). It increases inducible nitric oxide synthase expression, NF-κB activation, and decreases PPAR-g expression.
Doxycycline is one of the most extensively studied for properties other than its antimicrobial effects (Aminov, 2013; Chen et al., 2019; Stephenson et al., 2018), and excellent blood-brain barrier permeability (Barza et al., 1975; Liu et al., 2001). The drug is known to have a broad range of actions (Garrido-Mesa et al., 2013) including the modification of the neurotransmitters system known to be involved in the pathophysiology of LID such as the cholinergic (Conti et al., 2018; Schlesinger et al., 2004; Schlesinger et al., 2004), serotonergic (Zhang et al., 2006), glutamatergic (Monte et al., 2013) as well as others (Mansson et al., 2007; Munzar et al., 2002). In addition, González-Lizárraga et al. (2017) reported that doxycycline reshapes α-synuclein oligomers, inhibits α-synuclein aggregation, and the seeding of new oligomers, thus preventing cytotoxicity in dopaminergic cell lines.
The pharmacokinetic properties of doxycycline are advantageous concerning the gut microbiota. Based on their long, excellent safety profiles in humans, following oral application, the rapid rise in the serum levels indicates that it is primarily absorbed by the small intestine. Thus, doxycycline represents a low burden for the colorectal tract (Bartlett et al., 1975). As the motor benefits of L-DOPA were preserved after doxycycline treatment, it is strong evidence that the drug did not interfere with L-DOPA intestinal absorption. However, one cannot eliminate this hypothesis. New experiments are ongoing in our laboratory using the subcutaneous application of L-DOPA. Preliminary results confirm the outcome of oral L-DOPA. To avoid side effects related to the antimicrobial activity and possible bacterial resistance, chemically modified tetracycline COL-3 may have numerous applications without the associated risk of antibiotic resistance.
Finally, despite anecdotal evidence from patients with PD, we do not know to what extent the effects of doxycycline are valid. Although we have hypothesized that neuroinflammation plays a major role in the development of LID, we recognize that a causal link between the two needs to be firmly established. As a final point, repurposing doxycycline against dyskinesia in PD might be a novel tactic for neuroprotection, exploiting this therapeutic potential of a ready-to-use drug (Del Bel et al., 2016; Reglodi et al., 2017; Socias et al., 2018).
SUMMARY and CONCLUSION We demonstrated for the first time that doxycycline is effective in the attenuation of LID without interfering with the anti-akinetic action of L-DOPA. The possible mechanisms underlying the effect of doxycycline decreasing LID are interesting but need to be further investigated to optimize the dose and ensure the safety of long-term treatment. Our findings point to the potential of the repurposing doxycycline or its derivatives as an adjunct treatment to L-DOPA in the therapy of PD to delay the onset of dyskinesia and possibly attenuate neurodegeneration.
Ethics statement: The experiments were conducted according to the principles and procedures described by the guidelines for the care and use of animals in neuroscience and behavioral research (ILAR, USA). The local ethics committee approved the protocol.
Acknowledgments: The authors appreciate Prof. Francisco S Guimarães and Prof. Paul Bolam for their critical review of the paper. The authors would like to thank Célia A. da-Silva, Sara Saltareli, and Vitor Castania for their technical assistance.
Funding statement: The study was supported by the São Paulo State Foundation for the Support of Research (FAPESP, Brazil; Grant 2014/25029-4). EDB is a recipient of grants from the National Council for Scientific and Technological Development (CNPq, Brazil), CAPES, and the Comité Français d’Evaluation de la Coopération Universitaire avec le Brésil (French Committee for the Evaluation of Academic and Scientific Cooperation with Brazil; CAPES-COFECUB Grant #848/15). EDB is a CNPq research fellow. MB and GCN were recipients of FAPESP, Brazil, fellowship 2016-06602-0 and 2015-03053-3, respectively).
Conflict of interests: The authors have no financial or personal conflicts of interest to declare in relation to this study.