Background
Parkinson’s disease (PD) is a common neurodegenerative disorder and a leading cause of long-term disability. Although symptomatic treatments are available and effective, at least partially, there is currently no therapy known to reverse, arrest, or slow its progressive course. Multiple genetic and environmental factors contribute to the development of PD; among them, alpha-synuclein (αSyn, encoded by
SNCA) plays a central role in PD genetics and pathogenesis [
1,
2]. Mutations in
SNCA can cause PD, and accumulation and aggregation of αSyn within Lewy bodies and Lewy neurites in the nervous system are a pathological hallmark of PD. Various cellular events including proteinopathy, neuroinflammation, and oxidative stress contribute to the degenerative process, leading to the eventual loss of dopaminergic neurons of the nigrostriatal dopaminergic pathway of the brain, another pathological hallmark of PD [
3]. As proteostasis, the redox system, and inflammatory processes in PD can be orchestrated by the nuclear factor erythroid 2-related factor 2 (Nrf2), activation of Nrf2 is a promising therapeutic approach for neurodegenerative disease [
4,
5].
Melanocortin 1 receptor (
MC1R) is the major genetic determinant of hair color. Binding of its ligand alpha-melanocyte stimulating hormone (α-MSH) to MC1R on melanocytes activates the cAMP pathway and facilitates brown/black eumelanin synthesis and increases the ratio of yellow/red pheomelanin to eumelanin [
6,
7]. Severe loss-of-function polymorphisms of
MC1R contributes to red hair/fair skin and are associated with skin aging, and melanoma risk [
8‐
10]. More recent studies identify a critical role of MC1R in regulating physiological functions in the skin including the immune response, DNA repair, and cell differentiation and proliferation, which can be pigmentation-dependent or -independent [
10,
11]. In addition to its cutaneous expression and function, MC1R is expressed in other tissue and cell types, including immune and endothelial cells, and can modulate the immune system and inflammatory response [
12,
13]. α-MSH or its synthetic analog Nle
4,D-Phe
7-α-MSH (NDP-MSH), a tanning agent and drug approved by the European Medicines Agency for treating the photosensitive skin condition erythropoietic porphyria [
14], exerts protective effects in models of ischemic stroke, traumatic brain injury, spinal cord injury, Alzheimer’s disease, and neuroinflammatory disease [
15‐
18], with MC1R engagement shown to mediate this protection in the latter model.
Prompted by well-documented epidemiological associations between MC1R, pigmentation, and melanoma, and between melanoma and PD as well as possible associations between red hair and PD, and between MC1R variants and PD [
19], we previously demonstrated the presence of MC1R in dopaminergic neurons in the mouse substantia nigra (SN) and its influence on dopaminergic neuron survival [
20]. Here, we report MC1R-specific protection against αSyn oligomerization and related inflammation and dopaminergic neurotoxicity. In vitro analyses revealed that MC1R counteracted αSyn oligomerization by activating Nrf2. Further, we demonstrate that PD patients exhibited reduced levels of MC1R in the SN.
Methods
Study design
The objective of this study was to characterize the protective role of MC1R in the nigrostriatal dopaminergic pathway and to elucidate responsible downstream mediators. Complementary genetic and pharmacological approaches were employed to manipulate MC1R in vivo and in vitro. αSyn pathologies, neuroinflammation, and related dopaminergic neurotoxicity were induced by overexpressing human wild-type (WT) αSyn in mice and in HEK cells and primary neuronal cultures. Human studies entailed assessment of postmortem brain tissue from PD patients and controls.
Sample sizes for animal experiments were determined based on our previous studies [
20,
21] in which significant differences in primary outcome measures (nigral dopaminergic cell counts and striatal dopamine content) were observed. For all animal experiments involving genetic modification of
MC1R and quantification of the outcome measures, littermates were used as controls. For animal experiments using commercially obtained mice, grouping was randomized. Cell experiments were repeated at least three times with at least three replicates within each condition. Investigators were blind to treatment assignments and/or sample group information wherever practical. All animal and human study protocols were approved by the responsible authorities at Massachusetts General Hospital.
Experimental animals
MC1R extension (
MC1Re/e) mice carrying an inactivating frameshift mutation of
MC1R in a C57BL/6 J background [
20,
22] were backcrossed with C57BL/6 J mice from the Jackson Laboratory (Bar Harbor, ME). Offspring heterozygous breeders were crossed with each other to generate
MC1Re/e and littermate WT mice.
MC1R transgenic (Tg) mice in an e/e background (
MC1Re/eTg) were originally generated and characterized at University of Edinburgh, UK [
23].
MC1Re/eTg mice express the human
MC1R under the transcriptional control of its human promoter, yielding a physiological expression pattern similar to that in humans. The transgene rescues the
MC1R deficiency dermal phenotype to give
MC1Re/eTg mice a WT-like dark coat.
MC1Re/eTg mice were crossed with
MC1Re/e mice to generate
MC1Re/eTg and littermate
MC1Re/e mice.
To test the effects of the MC1R agonist BMS-470539, 3-month-old male C57Bl/6 J mice were purchased from the Jackson Laboratory. To test the effects of the MC1R agonist NDP-MSH, MC1Re/e mice and their WT littermates were used.
Mice were maintained in home cages at a constant temperature with a 12-h light/dark cycle and free access to food and water.
Viral vectors and intra-SN infusion
Vector production and stereotaxic virus intra-SN infusion were described previously [
24]. The vectors used were: (1) p adeno-associated virus (AAV)-CBA-human αSyn-WPRE (αSyn AAV), (2) pAAV-CBA-WPRE empty vector (vector), (3) pAAV-CBA-venus1-human αSyn-WPRE and pAAV-CBA-human αSyn-venus2-WPRE bimolecular fluorescence complementation (BiFC αSyn AAV), and (4) pAAV-CBA-Venus-WPRE (venus).
Viral vectors were infused at a volume of 2 µl into the left SN at the following coordinates: AP + 0.9 mm, ML + 1.2 mm, and DV -4.3 mm relative to lambda.
MC1R agonist treatments
NDP-MSH and BMS‐470,539 dihydrochloride were purchased from Tocris Bioscience (Bristol, UK). BMS-470539 (20 mg/kg) or vehicle saline was administered subcutaneously daily starting 1 day after αSyn AAV or empty vector infusion for 4 weeks. A total dose of 3 nmol NDP-MSH in 2 µl PBS was injected intracranially at 30 µl/60 min into the left striatum (coordinates: AP + 0.9 mm, ML + 2.2 mm, and DV -2.5 mm relative to bregma). Control mice received PBS injection. αSyn AAV was infused into the SN immediately after NDP-MSH or vehicle administration.
Mice were sacrificed, and their ventral midbrain and striatum were dissected. Protein sequential extraction and immunoblotting of αSyn were conducted as previously reported [
24] with modifications. Briefly, tissues were homogenized in 1% Triton X-100 buffer and centrifuged. The supernatant was designated as the “Triton X-100-soluble” fraction. The pellet was resuspended in lysis buffer containing 2% SDS and designated as the “SDS-soluble” fraction. Protein concentrations were determined by BCA protein assay. Protein from each Triton X-100-soluble (50 µg) and SDS-soluble (80 µg) sample were run on NuPAGE 4–12% SDS-PAGE gel and transferred to PVDF membranes following fixation with 0.4% paraformaldehyde for 30 min. Primary antibody against human αSyn (clone Syn211, ThermoFisher Scientific, AHB0261) was added at 1:700 and incubated overnight at 4 °C. Membranes were then incubated with a secondary antibody. Signals were detected using enhanced chemiluminescence. Band densities were determined using ImageJ and normalized to ponceau staining.
Immunostaining, imaging, and quantification
Mice were sacrificed, and their brains were processed and sectioned coronally as described [
24]. For immunostaining, sections were incubated with primary antibodies overnight at 4ºC and corresponding secondary antibodies for 1 h at 37 ºC. The primary antibodies used were against human αSyn (clone Syn211, ThermoFisher Scientific, AHB0261) at 1:500, phosphorylated αSyn at serine 129 (p-αSyn) (p-syn/81A, BioLegend, 825,701) at 1:500, glial fibrillary acidic protein (GFAP) (clone GA5, Sigma, G3893 and MAB360) at 1:1000, ionized calcium binding adapter molecule 1 (iba1) (clone EPR16588, Abcam, ab178846, and ab107159) at 1:500, tyrosine hydroxylase (TH) (clone TH2, Sigma, T1299) at 1:1000, and Nrf2 (Abcam, ab31163) at 1:500. For fluorescence staining, sections were incubated with goat anti-rabbit or anti-mouse lgG-Alexafluor-546 or -488. For DAB staining, sections were incubated with appropriate secondary antibodies, and the staining was developed by incubating with DAB.
MC1R staining was performed as previously reported [
20] with modifications. Sections were heated for antigen retrieval and incubated with primary anti-MC1R (Santa Cruz, SC‐19,485) at 1:50 or anti-human MC1R (LSBio LS-A1040) at 1:100 overnight at 4 °C [
25]. Sections were then incubated with Alexa Fluor conjugated secondary antibody at 1:200 at 37 ºC for 30 min. After washes, subsequent TH or GFAP or iba1 staining was performed.
Fluorescence images were captured under a Nikon C2s laser scanning microscope. Images from DAB-stained sections were captured under an Olympus BX50 microscope with a DP 70 digital camera system. Posterior (interaural 0.00/bregma -3.80 mm), posterior central (interaural 0.28 mm/bregma -3.52 mm), anterior central (interaural 0.64 mm/bregma -3.16 mm), and anterior (interaural 0.88 mm/bregma -2.92 mm) midbrain sections from each mouse [
26] were selected for quantification unless stated otherwise.
To evaluate αSyn transduction efficiency, midbrain sections were co-labeled with antibodies against human αSyn and TH. To determine the percentage of αSyn-positive dopaminergic neurons in the SN, images were acquired in 488- and 546-nm channels with 40 × magnification. ImageJ software was used to count TH-positive cells and cells that were both TH- and αSyn-positive.
Quantification of p-αSyn staining was performed using the optical fractionator method at 40 × magnification (Olympus BX51 microscope and Olympus CAST stereology software) [
27] to count positively stained particles in the SN.
For astrogliosis and microgliosis analyses, GFAP staining and the morphology of iba1-positive cells in the SN pars compacta (SNpc) were analyzed as previously described [
24].
For proteinase K digestion, sections were mounted onto slides, dried overnight, and cover-slipped [
24]. After image acquisition, cover slips were carefully removed, and sections were rehydrated. Sections were incubated with 50 μg/ml proteinase K at 55 °C for 120 min, and images were recaptured. Reconstituted venusYFP intensity was quantified using ImageJ.
For thioflavin-S staining and quantification, sections were incubated with 0.05% thioflavin S solution for 8 min [
28]. Images were recaptured in 488-nm channel at 40 × magnifications. Thioflavin-S fluorescence intensity was quantified using ImageJ.
To assess nuclear-to-cytoplasmic Nrf2 ratio, sections were counterstained by DAPI to reveal the nucleus, and nuclear Nrf2 signal was defined within DAPI regions. Sections were imaged at excitations of 488 nm for Nrf2 and 568 nm for TH. The mean fluorescence intensities of nuclear and cytoplasmic Nrf2 per cell were measured, and the ratio was determined using ImageJ’s “Intensity Ratio Nuclei Cytoplasm Tool” as previously described [
29]. Only TH-positive neurons were measured. Five to ten cells were randomly picked on each side of the SN from each section and a total of 30 cells on each side from each animal were analyzed.
Protein oxidation
Protein carbonyls in ventral midbrain tissue were detected using an Oxyblot protein oxidation detection kit (Millipore, S7150) according to the manufacturer’s instructions. Band density was analyzed using ImageJ and normalized to ponceau staining density.
Western blotting
Ventral midbrain tissues were lysed, and proteins were extracted and electrophoresed. The blot was probed with anti-Nrf2 (Abcam, ab31163) at 1:1000 or anti-MC1R (Santa Cruz, SC-19485) at 1:1000 or anti-TH (ENZO BML-SA497-0100) at 1:1000. Band density was analyzed using ImageJ and normalized by actin. All the original full-blot images are in Supplemental files, Fig.
S6 and Fig.
S7.
Quantitative polymerase chain reaction (qPCR) for cytokines and Nrf2 target genes
Total RNA was isolated using TRI reagent (Invitrogen) and reverse-transcribed into cDNA using a superscript III kit (Invitrogen). qPCR was performed in a 96-well plate using SYBR Green PCR Master Mix in an Applied Biosystem 7500. GAPDH was used to normalize expression levels of the target genes. The 2
–∆∆Ct method was employed for data analysis [
30]. The primers used are provided in Supplementary Table
1.
Amphetamine-induced rotational behavior
Amphetamine-induced (5 mg/kg, intraperitoneal) rotational behavior was assessed by an automated rotometry system (San Diego Instruments) for 60 min as previously described [
24,
31].
Striatal dopamine measurement
Mice were sacrificed, and the striatum was dissected. Dopamine content was determined by high-performance liquid chromatography (HPLC) coupled with electrochemical detection as previously described [
24,
31].
Stereological analysis of SN dopaminergic neurons
A complete set of serial midbrain sections were collected and immunostained for TH and counterstained for Nissl to reveal dopaminergic neurons and total neurons. Unbiased stereological counting was performed as previously described [
24,
31].
HEK293T cell transfection, transduction, immunostaining, immunoblotting, and chromatin immunoprecipitation (ChIP)-qPCR
HEK293T cells were purchased from Clonetech. Cells were maintained in DMEM with 10% FBS in a humidified incubator at 37 °C with 5% CO
2. pcDNA3.1( +)-human WT αSyn and control plasmid were provided by a former colleague Dr. Joseph Mazzulli (Northwestern University, IL) [
32]. MC1R-Tango expressing human MC1R tagged with FLAG and vector control GPRC5A-Tango were gifts from Bryan Roth (Addgene plasmid #66,427 and #66,382). αSyn and MC1R and their respective controls were transfected into cells using Lipofectamine® 2000 (ThermoFisher Scientific) according to the manufacturer’s instructions.
Human shNrf2 was purchased from Dharmacon RNAi Consortium (RHS4533-EG4780). pLKO.1- scrambled RNA (scRNA) (Sigma, SHC016-1EA) was used as a control. Plasmids were packaged in lentivirus with packaging plasmid psPAX2 and envelope plasmid pMD2.G (Addgene plasmid #12,259 and #12,260, gifts from Didier Trono). Lentiviral particles were produced in HEK293T cells. For scRNA or shNrf2 transduction, cells were incubated in medium containing lentiviral particles in the presence of polybrene for 16 h.
Cells were harvested 48 h after transfection with or without viral transduction and lysed for immunoblotting or qPCR. For αSyn immunoblotting, in-cell crosslinking was performed using disuccinimidyl suberate ligand (ThermoFisher Scientific) as previously described [
33]. Primary antibodies used were anti-human αSyn (clone Syn211, ThermoFisher Scientific, AHB0261) at 1:500, anti-Nrf2 (Abcam, ab31163) at 1:1000, anti-FLAG (Sigma, F1804) at 1:1000, anti-phosphorylated cAMP response element-binding protein (pCREB) (Ser133) (Cell Signaling, 9198S) at 1:1000, anti-CREB binding protein (CBP) (Cell Signaling, 7389S) at 1:1000, and anti- MC1R (LSBio LS-A1040) at 1:1000. Densities of bands were analyzed using ImageJ and normalized by actin.
For chromatin immunoprecipitation (ChIP)-qPCR assay [
34,
35] identifying CREB binding to
Nrf2 following MC1R activation, cells (5 × 10
6) were transfected with αSyn and MC1R and harvested 48 h after transfection. ChIP was performed using the ChIP-IT® Express (Active Motif, 53,008) following the manufacturer’s instructions [
36]. In brief, cells were fixed with 37% formaldehyde (12 min) followed by sonication (25% amplitude, pulse for 30 s on and 30 s off for a total of 15 cycles). Immunoprecipitation was carried out using antibodies against anti-pCREB (Ser 133) (Cell Signaling, 9198S) or nonspecific rabbit IgG (Cell Signaling, 2729S) as negative control. Following antibody pull-down and DNA purification, qPCR was conducted. Forward and reverse primers used were CGGGCTGAGCTTCCGAAAAT and AACTCTTTATCTCGCGGGCG. The primers were designed to detect CREB binding site TGACG in the
Nrf2 promoter based on published information [
37,
38], which also matched the predicted CREB binding site from our own in silico promoter analysis of
Nrf2. The program for quantification amplification was 2 min at 95 °C, 15 s at 95 °C, 20 s at 58 °C and 20 s at 72 °C for 40 cycles in 20 μl reaction volume. Data was presented as % input genomic DNA for different treatment groups.
For immunocytochemistry, cells were grown on pre-coated poly-lysine coverslips, fixed 48 h post-transfection with or without viral transduction in 4% PFA for 10 min, and blocked in 5% normal goat serum in PBS/0.3% Triton X-100 for 30 min at 37 ºC. Cells were incubated with anti-Nrf2 (Abcam, ab31163, 1:200) overnight at 4ºC and Alexa-Fluor-488 for 1 h at 37 ºC. Images were acquired under a Nikon C2s laser scanning microscope. The subcellular distribution of Nrf2 fluorescence in nuclear and cytoplasmic regions was quantified using ImageJ as described above and previously in cell cultures [
29].
Primary neuron cultures, viral transduction, NDP-MSH treatment, immunostaining and quantification, and cell assays
Primary cortical neurons were prepared as previously described [
39,
40] from the cerebral cortex of embryonic day 16–17 WT and
MC1Re/e mice. Mouse shNrf2 was purchased from Sigma (SHCLNG, NM 010,902) and packaged with lentivirus. For lentiviral transduction, cells were incubated in 250 μl medium containing lentiviral scRNA or shNrf2 and 250 μl complete neuron growth medium (Neurobasal medium with 2% B27 supplement, 2 mM L-glutamine, 100 U/ml penicillin, and 100 µg/ml streptomycin) in the presence of polybrene at DIV3 for 48 h in a 24-well plate. BiFC αSyn AVV or control venus AAV was added at 1.25 × 10
10 gc to each well for 48 h. Medium was changed to complete neuron growth medium supplemented with or without 10 nM NDP-MSH at DIV5. At DIV12, 50 μl supernatant from each well was taken for lactate dehydrogenase (LDH) release assay using a Pierce LDH Cytotoxicity Assay Kit (ThermoFisher Scientific 88,953, 88,954) according to the manufacturer’s instructions. Nrf2 knock-down efficiency was determined by qPCR.
At DIV9, images of living neurons were obtained under an inverted Leica fluorescence microscope. At DIV12, neurons transduced with αSyn AAV or control venus AAV were fixed and immunostained. Primary antibodies used were against MAP2 (Invitrogen, PA5-17,646, or 13–1500 for double staining) at 1:250, MC1R (Santa Cruz, SC‐19,485) at 1:1000, and human αSyn (Thermo Fisher Scientific, AHB0261) at 1:250. Cell were counterstained by DAPI.
For MAP2-positive cell counting, three images were captured randomly from each coverslip under a 20 × objective in an Olympus BX50 microscope equipped with a DP70 digital camera system. MAP2 and DAPI channels were merged, and MAP2-positive cells in each visual field were counted using ImageJ [
41]. A total of nine images from three replicate coverslips were analyzed.
Human samples
Postmortem frozen SN tissue from pathologically diagnosed PD patients and control individuals with no neurological conditions was obtained from the Massachusetts Alzheimer’s Disease Research Center of the Massachusetts General Hospital. Age and postmortem interval (PMI) for cases used for each of the experiments were provided in figure legends. All subjects were Caucasian. Tissues were cryosectioned at 8 μm and airdried at 37 ºC for 2 min and fixed in 70% ethanol for 2 min. For peroxidase immunostaining, sections were immersed in 1% H2O2 in 100% methanol for 10 min at 37 ºC to quench endogenous peroxidase. Sections were then blocked by 5% normal goat serum and incubated with anti-human MC1R (LSBio, LS-A1040) at 1:50 and/or anti-TH (Sigma, T1299) at 1:500 at 37 ºC for 45 min. After three washes, sections were incubated with either biotin-conjugated secondary antibody at 1:500 for peroxidase immunostaining or Alexa Fluor-conjugated secondary antibodies at 1:200 for fluorescent immunostaining at 37 ºC for 30 min.
For immunoblotting, twenty Sects. (10 μm) containing only SN from frozen tissues were collected in RIPA lysis buffer. Protein sample (120 µg) was loaded and run on NuPAGE 4–12% SDS-PAGE gels at 150 voltage for 80 min and transferred to PVDF membranes at 90 voltage for 75 min at 4 ºC. Blots were then probed by anti-human MC1R (LSBio LS-A1040) at 1:700, anti TH (Sigma, T1299) at 1:1000, and anti Nrf2 (D1Z9C) (Cell Signaling 12721S) at 1:1000, overnight at 4 ºC, followed by an extra one-hr incubation at 37 ºC for MC1R probing to enhance the binding. Actin was used as the loading control. Blots were then incubated with anti-rabbit or anti-mouse secondary antibodies at 1: 1000 for 1 h at 37 ºC. Blots were incubated in enhanced chemiluminescence for 2 min and then scanned by an LI-COR Odyssey Fc Image System.
All human tissue study protocols were approved by the Partners Human Research Committee.
Statistical analysis
All values are presented as the mean ± standard error of mean (SEM). The comparison between two groups was conducted using unpaired two-tail Student’s t test. Comparisons among multiple groups were performed using one-way or two-way ANOVA followed by Tukey's post hoc test. Specific statistical methods for quantitative experiments were indicated in figure legends. P-values ≤ 0.05 were considered statistically significant.
Discussion
We previously reported protective effects of melanoma-related MC1R in nigrostriatal dopaminergic neuron survival under basal conditions and in toxin models of PD [
20]. In the present study, we found exacerbated αSyn pathology following AAV-mediated overexpression of αSyn in the SN of
MC1Re/e mice carrying a loss-of-function mutation. Increased oligomeric αSyn, especially amyloid-like αSyn fibrils and p-αSyn, indicated pathologic αSyn aggregation in
MC1Re/e mice. Existing evidence connecting MC1R and αSyn is scarce and indirect. In vitro studies using melanoma cells, which express MC1R, indicates that αSyn is highly expressed and promotes melanoma cell survival [
54,
55]. In addition, αSyn is shown to reduce UV-induced melanin synthesis in melanoma cells, suggesting a possible inhibitory effect of αSyn on the melanin pathway that is controlled by MC1R [
56].
As a consequence of exacerbated αSyn pathology,
MC1Re/e mice with SN-targeted overexpression of αSyn displayed exacerbated dopaminergic deficits anatomically and neurochemically. This exacerbation cannot be solely explained by a preexisting dopaminergic dysfunction previously demonstrated in these mice around the same age [
20]. The ability of human
MC1R to rescue the dopamine deficits indicates the specificity of this MC1R effect. The human
MC1R transgene, under its human physiological promoter, expresses less but more potent MC1R than the mouse gene and restores pigmentation in
MC1Re/e mice [
23]. Although our results suggest that the SN expression pattern of the transgenic human
MC1R was similar to the endogenous WT mouse
MC1R and exerted a protective effect, they do not indicate whether the rescue was partial or complete or whether the transgene might have had trophic effects, as no Tg and WT littermates were compared. Collectively, our findings of
MC1R disruption-induced impairment of αSyn defense and human
MC1R-mediated rescue strongly support MC1R-specific dopaminergic protection against αSyn.
The ability and specificity of MC1R to protect against αSyn-induced dopaminergic deficits were further demonstrated by BMS‐470,539, a selective MC1R agonist with modest blood–brain barrier permeability, and by locally delivered NDP-MSH, a broader agonist for melanocortin receptors with no brain penetrance, in WT but not
MC1R mutant mice. Commercially known as Scenesse®, NDP-MSH is pending U.S. Food and Drug Administration review after being approved by the European Medicines Agency to treat patients with erythropoietic protoporphyria. The neuroprotective effects of systemically adminstered NDP-MSH have been reported in models of ischemic stroke, traumatic brain injury, spinal cord injury, and Alzheimer’s disease [
15‐
17]. A more recent study demonstrates the MC1R-dependent neuroprotective effects of intravenously injected NDP-MSH in mouse models of neuroinflammatory disease involving a compromised blood–brain barrier [
18]. Activation of MC1R in the brain locally as well as peripherally, perhaps through improvements in the systemic environment, could theoretically protect the nigrostriatal system and may not necessarily exclude each other. While the involvement of peripheral MC1R cannot be ruled out, especially in the context of BMS-470539 neuroprotection, the efficacy of locally administered NDP-MSH suggests that CNS MCIR has protective functions, which is strongly supported by the similar protective effects observed in primary neuronal cultures. Therefore, our study reveals that neuronal MC1R functions as a protective signaling inducer against αSyn neurotoxicity. Further differentiating peripheral versus CNS MC1R actions using cell- or tissue-specific knock-out (e.g., in tyrosinase- and/or TH- expressing cells) in addition to global
MC1Re/e would be important for further target validation and drug development and for gaining a better understanding of the mechanisms underlying MC1R-mediated dopaminergic defense.
αSyn, especially in its oligomeric form, induces neuroinflammation and oxidative stress, which may contribute to neurodegeneration in PD [
48,
57]. We found that exacerbation and protection of αSyn dopaminergic neurotoxicity by genetic and pharmacological MC1R manipulations were accompanied by altered pro-inflammatory cytokines and microglia activation status in the SN. Immune cells including microphages, monocytes, and endothelial cells express MC1R [
12,
13]. Expression of MC1R has been reported in a human microglial cell line [
58] but not in rat primary microglia [
59]. We could not detect MC1R in microglial cells by immunofluorescence double-staining, suggesting that microglial responses may be mediated by neuronal MC1Rs and alterations in neuronal activities. Together with the reported MC1R-dependent protective effect of NDP-MSH against neuroinflammation [
18], our findings further support the broad role of MC1R in immunomodulation and inflammation, not only in the periphery [
12] but also in the CNS. Despite the previously reported expression of MC1R in a human astrocyte cell line and suggested glial cell MC1R-mediated inhibition of TNF-
α by
α-MSH in a mouse model of brain inflammation [
58,
60,
61], our double-labeling showed rare colocalization of MC1R and GFAP. MC1R manipulation was not associated with the gliosis that we and others have observed following αSyn overexpression, at least at the time points analyzed.
Additionally, given the predominant transduction of αSyn and MC1R expression in neurons, the altered inflammation and oxidative damage in the ventral midbrain were likely effects of altered Nrf2 signaling in neurons in response to MC1R manipulation. Nrf2, known as a master regulator of the immune system and oxidative stress, was shown to be regulated by MC1R, with impaired induction and activation following αSyn expression in
MC1Re/e mice; and conversely, with enhanced induction and activation following genetic and pharmacological activation of MC1R in vivo. In cellular models of αSyn pathologies, including primary neuron cultures, the role of Nrf2 was indispensable for the counteractions of MC1R genetic and pharmacological activation. Nrf2 is tightly regulated at multiple levels including transcriptional, post-transcriptional, and post-translational regulation, most importantly in the cytoplasm by its primary negative regulator Kelch-like ECH-associated protein 1 (Keap1) through the canonical and non-canonical mechanisms [
62]. Disassociation from Keap1 renders Nrf2 stabilization and translocation into the nucleus, where it activates transcription of the target genes. MC1R activation by α-MSH has been shown to induce Nrf2 mRNA in human skin [
37]. In addition, MC1R activates PI3K pathway [
63,
64] and PI3K activation in vivo increases Nrf2 mRNA and abundance of Nrf2 protein in nuclear extracts in the liver [
65]. A silico promoter analysis of the human
Nrf2 gene identified putative binding site for CREB [
37]. The cAMP pathway mediates MC1R signaling [
50,
66]. Genome-wide location analysis indicated
Nrf2 as a CREB binding positive gene [
51], supporting CREB on the
Nrf2 promoter. We demonstrated that MC1R overexpression in αSyn expressing HEK293T cells increased Nrf2 mRNA and protein as well as nuclear Nrf2, suggesting that Nrf2 activation and target gene HO-1 expression were likely the results of de novo Nrf2 production and nuclear accumulation. Indeed, ChIP assay demonstrated increased CREB binding to the
Nrf2 promoter following MC1R activation, further supporting transcriptional activation of Nrf2 by MC1R through CREB. Although MC1R overexpression in HEK293T cells appeared to induce CREB binding, mRNA, protein, and nuclear translocation of Nrf2 to the comparable extend, other mechanisms may still contribute to MC1R activation of Nrf2 in the context of αSyn toxicity. Nevertheless, these results provide direct evidence that Nrf2 is a downstream mediator of the protective effects of MC1R activation. Our findings also support the ability of Nrf2 in neurons to be responsive [
53], both in vivo and in vitro, to MC1R manipulation in the context of αSyn toxicity. Nrf2 is altered in Parkinson’s and related neurodegenerative diseases, and Nrf2 activators have progressed to active clinical practice or development as neurotherapeutics [
4,
53,
67‐
69]. Although we were not able to assess cytoplasmic
vs. nuclear Nrf2 due to limited tissue amount, our results indicated reduced Nrf2 in PD postmortem SN that was at the margin of statistical significance. Collectively, these findings provide not only mechanistic insight into MC1R defense against αSyn pathologies but also therapeutic implications for the possible use of Nrf2 activators in PD patients, especially those carrying partial or complete loss-of-function MC1R variants.
The translational significance of our findings is further highlighted by the presence of MC1R in dopaminergic neurons in the human SN and the suggestion of reduced MC1R expression at the tissue level in PD patients
vs. controls. MC1R immunoreactivity in periaqueductal gray neurons is reported in fixed human brain sections [
70]. Mykicki et al. described the expression of MC1R in human neuronal cells differentiated from a progenitor cell line and in neurons from postmortem human brain tissues [
18]. The same study reported reduced brain MC1R expression in multiple sclerosis patients, although no brain regions or neuron types were specified [
18]. The remarkable overlap of MC1R and TH in the SN observed in our study suggests primary, if not exclusive, dopaminergic neuronal functions of MC1R within the SN. Notably, MC1R did not appear to be reduced at the cellular level in surviving dopaminergic neurons. MC1R expression was unaltered in the ventral midbrain in AAV αSyn mice at an earlier timepoint when dopaminergic neuron loss was not yet significant. Although it remains to be determined whether loss of MC1R at the tissue level in PD patients precedes loss of dopaminergic neurons, the evidence presented by this and other studies supports that MC1R is involved mechanistically in the pathophysiology of PD and is a promising therapeutic target for PD and related disorders. The findings encourage development of CNS-penetrant MC1R agonists as well as potential repurposing of existing, primarily peripherally acting MC1R agonists, as candidate disease-modifying therapy for PD. Given the links among MC1R loss of function, red hair, melanoma, and PD, our findings also provide evidence of a possible MC1R basis for the well-established link between PD and melanoma. The similar protection observed in vitro in cells that are considered non-pigmented and lack pigmentation machinery, including HEK293T cells and primary cortical neurons, supports non-pigmentary pathways of MC1R actions. Further studies are needed to determine whether
MC1R as the key pigmentation gene acts through pigmentation and whether neuromelanin or melanin in the periphery is involved in MC1R neuroprotection in vivo.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.