Gobs of α-synuclein form the telltale Lewy bodies that wreak havoc in Parkinson’s and other synucleinopathies. Rather than attempting to clean them up, a new therapeutic strategy aims to prevent their formation in the first place. Reporting in Science on September 1, researchers led by Clemens Scherzer at Brigham and Women’s Hospital in Boston reported that β2 adrenergic receptor (β2AR) agonists dampen expression of the α-synuclein gene. These common asthma drugs also forestalled neuronal damage in a mouse model of PD and in neurons from a PD patient. Together with Trond Riise at the University of Bergen, Norway, the researchers uncovered a striking correlation within the well-kept medical records of more than 4 million Norwegians. People taking the asthma drug salbutamol had one-third less risk of developing PD, whereas PD risk more than doubled for those taking a β-blocker.
Based on established links between α-synuclein overabundance and PD, Scherzer and colleagues set out to dampen α-synuclein expression. While people who carry an extra copy of the α-synuclein gene (SNCA) develop an autosomal-dominant form of PD, far more people carry non-coding risk variants in SNCA, some of which modestly elevate expression, Scherzer explained. Prior work has elucidated how α-synuclein expression is controlled, for example by GATA transcription factors, yet these mechanisms have yielded no promising drug targets (see Jul 2008 news).
Trying a different approach, first author Shuchi Mittal and colleagues screened for compounds that lowered SNCA expression in human neuroblastoma cells. They treated the cells with 1,126 compounds, including FDA-approved drugs, vitamins, supplements, and alkaloids. After two days, 35 compounds had lowered synuclein mRNA by more than a third. One was the β2AR agonist metaproterenol. The researchers tried two other β2AR agonists, clenbuterol and salbutamol, in a replication screen. They also lowered synuclein expression. In a final round of screening, four compounds significantly lowered SNCA at both the gene and protein level, and three of them were the β2AR agonists.
Investigating these drugs further, the researchers found that they suppressed synuclein expression in primary rat cortical neurons by about 25 to 30 percent. At a doses of 1 to 10 mg/kg injected intraperitoneally, clenbuterol efficiently crossed the blood-brain barrier and dampened α-synuclein expression in the substantia nigra of wild-type mice. β2AR-knockout mice expressed twice as much α-synuclein as did wild-type, and knocking down this receptor in human cells also boosted expression. In keeping with all this, treating neuroblastoma cells with the β2AR inhibitor propranolol, a common hypertension drug, had the opposite effect, ramping up α-synuclein expression.
What is the β2AR doing here? Because β2ARs have been implicated in acetylation of the SNCA promoter and of intronic enhancers, the researchers investigated this epigenetic modification. Indeed, they found that clenbuterol triggered deacetylation of lysine 27 on histone 3 that was bound to regulatory regions in the SNCA gene in neuroblastoma cells. H3K27 acetylation promotes transcription, and its deacetylation by clenbuterol correlated with suppression of the SNCA gene. Treating cells with the deacetylation inhibitor valproic acid blocked clenbuterol’s effects on SNCA expression, supporting the idea that the β2AR agonist kicked off epigenetic changes.
To test if β2AR agonists might have therapeutic potential, Mittal added it to neurons derived from induced pluripoβent stem cells (iPSCs) taken from a PD patient with an SNCA triplication. Clenbuterol dialed down SNCA expression in the derived neurons by nearly 20 percent. Intriguingly, the drug also protected the cells from the mitochondrial toxin rotenone, which dismantles the electron transport chain, leading the production of free radicals. Synuclein is known to sensitize neurons to rotenone. Clenbuterol also protected rotenone-treated neurons from death. Similarly, the drug spared neurons in an MPTP model of parkinsonism. MPTP also poisons mitochondria and can promote the formation of α-synuclein aggregates. Injecting clenbuterol into MPTP-treated mice prevented loss of dopaminergic neurons in the substantia nigra. Together, the findings suggest that β2AR agonists may protect neurons by reducing α-synuclein, and/or by protecting mitochondria from malfunctions associated with the disease.
Neurons Spared. Compared with control mice (top), MPTP-treated mice lost dopaminergic neurons (bottom left). Clenbuterol protected (bottom right). [Courtesy of Mittal et al., Science, 2017.]
Epidemiology Agrees with Bench Results
β2AR agonists are widely used to treat asthma. Might they modulate risk of PD? Absent a clinical trial, the researchers first took advantage of an epidemiological opportunity. Since 2004, Norway has kept detailed medical records, including prescription drug information, on most of its 4.6 million inhabitants. This enabled the researchers to compare the incidence of PD in people taking salbuterol, the most commonly prescribed asthma drug in that country, with those not taking it. Between 2004 and 2015, salbuterol-takers had a 34 percent lower PD risk. Corticosteroids, a different asthma treatment, did not influence PD risk.
In contrast, people on propranolol, a β2AR blocker, had more than double the PD risk of people who took neither drug. Propranolol is prescribed not only to temper high blood pressure, but also to quiet essential tremor, a type of shaking generally not caused by PD. To account for the possibility that some of the tremor patients took propranolol because they already had PD, the researchers excluded all patients with essential tremor from the analysis; they also excluded people who developed PD within two years of initiating treatment. These corrections reduced the elevation in PD risk with propranolol from 2.2-fold to 1.82-fold. Still, overall the findings suggest that β2AR agonists reduce PD risk while antagonists elevate it. Both the researchers and commentators emphasized that these correlations cannot prove causality.
Concerning Correlation. Among 4.6 million Norwegians, those taking low (green), medium (yellow), and high (red) doses of salbutamol (left) had a lower incidence of PD than those never taking it (blue), while those taking propranolol (right, green) had a higher incidence. [Courtesy of Mittal et al., Science, 2017.]
Scherzer and colleagues proposed that the modest reduction in α-synuclein expression exerted through β2AR agonists in vitro over time translates into reduced PD risk. This suggests that lowering expression may benefit even people without extra copies of the gene or expression-boosting variants. Charbel Moussa of Georgetown University in Washington, D.C., considers targeting α-synuclein transcription a feasible approach, but he favors going after post-translational modifications that may promote the protein’s accumulation. That way, potential side effects of lowering normal levels of α-synuclein could be avoided.
In an accompanying editorial, Evan Snyder of the University of California, San Diego, pointed out that simple reduction in α-synuclein transcription might not be the only way β2AR agonists could counteract accumulation of the protein. For example, β2AR activation accelerates autophagy, a pathway known to dispose of α-synuclein aggregates (Farah et al., 2014).
Moussa commented that it will be crucial to define the mechanism. A key experiment would be to test the drugs in mice overexpressing α-synuclein, rather than in MPTP models, he added. Scherzer said that, with extensive behavioral analysis, is in the works.
Moussa cautioned that the proposed mechanism of action— H3K27 deacetylation—may exert broad epigenetic effects. Scherzer told Alzforum that initial genetic analyses indicated β2AR activation does not affect housekeeping genes, but a more extensive genetic analysis is forthcoming.
What is next on the horizon for β2AR agonists? Scherzer wants to team up with medicinal chemists to optimize the drugs for maximal α-synuclein downregulation. In the meantime, the established safety profile of the drugs means that a small, proof-of-concept clinical trial in PD could be started, he said. Of the three asthma drugs that reduced α-synuclein expression, clenbuterol, which has been approved in Europe to treat asthma, would probably make the best choice as it is long-acting and known to cross the blood-brain barrier, Scherzer said. Salbutamol also reportedly crosses the barrier, but it is short-acting, and metaproterenol is not known to enter the brain.
Ullrich Wüllner of the German Center for Neurodegenerative Diseases in Bonn agreed. “Identification of pharmaceuticals already in use, which modulate α-synuclein expression, provides an instant translational benefit to PD patients,” he wrote to Alzforum.
For larger trials to succeed where many have failed, Scherzer emphasized that trial design will be essential. He thinks an optimal trial would select PD patients carrying either SNCA multiplications or certain risk variants, and include biomarkers of α-synuclein expression.
Mark Frasier of the Michael J. Fox Foundation, an early funder of this work, expressed cautious optimism for the approach. “Alpha-synuclein is a high priority Parkinson’s target and we are pleased to have another strategy for intervention,” he wrote. “As with any therapeutic approach, if β2AR agonists will be tested in the clinic, careful consideration to dosing, safety, and pharmacodynamic biomarkers will be essential to design an informative study.”—Jessica Shugart
- Farah BL, Sinha RA, Wu Y, Singh BK, Zhou J, Bay BH, Yen PM. β-Adrenergic agonist and antagonist regulation of autophagy in HepG2 cells, primary mouse hepatocytes, and mouse liver. PLoS One. 2014;9(6):e98155. Epub 2014 Jun 20 PubMed.
No Available Further Reading
- Mittal S, Bjørnevik K, Im DS, Flierl A, Dong X, Locascio JJ, Abo KM, Long E, Jin M, Xu B, Xiang YK, Rochet JC, Engeland A, Rizzu P, Heutink P, Bartels T, Selkoe DJ, Caldarone BJ, Glicksman MA, Khurana V, Schüle B, Park DS, Riise T, Scherzer CR. β2-Adrenoreceptor is a regulator of the α-synuclein gene driving risk of Parkinson's disease. Science. 2017 Sep 1;357(6354):891-898. PubMed.