Constipation of the vesicular transport system in neurons is a core feature of Parkinson’s disease, and two new studies describe small-molecule approaches to get this cellular digestion pathway moving again. On October 16 in Science Translational Medicine, scientists led by Dimitri Krainc at Northwestern University Feinberg School of Medicine in Chicago reported that an activator of β-glucocerebrosidase (GCase) boosted lysosomal function and clearance of α-synuclein aggregates. In a separate study published on October 21 in Neuron, scientists led by Joseph Mazzulli, also at Northwestern, ramped up the lysosomal trafficking system more broadly by unleashing the SNARE protein ykt6, whose activity had been squelched by α-synuclein aggregates.

  • An activator of β-glucocerebrosidase boosted lysosomal flow, reduced α-synuclein accumulation.
  • The SNARE protein ykt6 helps lysosomal trafficking, but is inhibited by α-synuclein.
  • Relieving this inhibition with a farnesyl transferase inhibitor revived lysosomes, cleared α-synuclein.

“These are exciting papers, because they are targeting the cellular compartments that may be at the core of the deficits we want to change early in neurodegenerative disease,” commented Gerhard Koenig of Arkuda Therapeutics in Cambridge, Massachusetts. He added that lysosomal dysfunction worsens in neurons with age, so giving these organelles a boost could theoretically have benefits across proteinopathies.

Lysosomes play a critical role in maintaining proteostasis in the cell, and myriad studies have implicated dysfunction in this clearance system in PD (e.g., Mar 2014 conference news and Feb 2017 news). When α-synuclein, the main component of Lewy bodies, accumulates, it disrupts vesicular trafficking (Gitler et al., 2008). PD-linked mutations in GBA1, which encodes the lysosomal enzyme β-glucocerebrosidase (GCase), also cause vesicular traffic jams when unprocessed GCase substrates clog the lysosome. Notably, Mazzulli and Krainc co-authored a report that interactions between α-synuclein and unprocessed GCase substrates triggered a vicious cycle, in which GCase substrates stabilize α-synuclein oligomers, overloading lysosomes even more (Jun 2011 news). Because of the bidirectional interactions between these pathways, researchers have proposed that turning up trafficking, expression, and/or GCase function could be a therapeutic strategy for cases of PD with or without mutations in GBA1.

Lagging Lysosomes in PD. Numerous genes associated with Parkinson’s disease converge on defunct cellular clearance pathways as a core feature of PD pathogenesis. [Courtesy of review by Klein and Mazzulli, Brain, 2018.]

In their STM paper, co-first authors Lena Burbulla and Sohee Jeon and colleagues tested the effect of the small-molecule activator of GCase, S-181, on neurons derived from people with PD. The group had previously reported that another GCase modulator made GCase more active and slowed α-synuclein accumulation. It also made lysosomes more functional in induced pluripotent stem cell (iPSC)-derived dopaminergic neurons from PD patients harboring mutations in GBA1, SNCA, or PARK9, and from people with idiopathic PD (Mazzulli et al., 2016). In the current study, the researchers identified S-181, and extended their investigations to more patient lines and into mouse models.

They reported that in iPSC-derived midbrain dopaminergic neurons from two PD patients with GBA1 mutations, GCase was less active than in controls, and that the neuronal lysates faced a glut of the GCase substrate glucosylceramide (GluCer). Lysosomal proteolysis was impaired, and the researchers detected a buildup of oxidized dopamine—a pathological feature they had previously pinned as contributing to lysosomal dysfunction (Sep 2017 news on Burbulla et al., 2017). Treating the cells with 5 μM S-181 fully or partially corrected all these deficits.

What if there isn’t a GBA mutation? In lysates from neurons with pathogenic PD mutations in LRRK2, Parkin, or DJ-1, or from idiopathic PD, the researchers also found subpar GCase, accumulating oxidized dopamine, and lysosomal dysfunction. S-181 rescued these deficits, and also tamped down levels of α-synuclein, which had accumulated in the cells. Overall, the findings suggested that activating wild-type GCase protein might be therapeutic in multiple types of PD, including the common sporadic form, the authors noted.

Mark Cookson of the National Institutes of Health, Bethesda, Maryland, expressed surprise at the apparent benefit of S-181 for idiopathic PD lines, noting that there would be little genetic risk driving dopamine oxidation or α-synuclein accumulation. “It will be important to identify the mechanism by which these reprogrammed idiopathic lines retain these effects, so we can understand how and why GCase directed molecules would benefit this larger group of patients,” he wrote.

Next, Burbulla and colleagues tested S-181 in mice. They injected 50 mg/kg of the compound twice daily into the abdomens of wild-type mice and into mice carrying one copy of GBA1 with the pathogenic D409V mutation. After 15 days of this, GCase activity ramped up in brain tissue from both the wild-type and mutant mice compared with vehicle-treated controls. In addition, S-181 reduced levels of GCase substrates and insoluble α-synuclein in the D409V-GBA1 mutant mice, the scientists report.

Koenig said that the data appear promising, though he also noticed an apparent discrepancy in the activity of the compound reported in the in vitro and in vivo findings. While micromolar-range concentrations of S-181 were required to boost GCase activity presumably in a serum-free assay, the drug reportedly appeared to work at low nanomolar, unbound concentrations in the brain. One would expect the opposite, i.e., lower concentrations needed in enzyme assays and higher concentrations needed to see efficacy in vivo. Koenig said a more detailed presentation of the pharmacological properties of this tool compound will be helpful to account for the ascribed new mechanism of action.

In their Neuron paper, Mazzulli’s group took a different approach. These scientists investigated how α-synuclein accumulation affects lysosomal dysfunction. First author Leah Cuddy and colleagues grew iPSC-derived midbrain dopaminergic neurons from PD patients who carried a triplication in SNCA, or the pathogenic A53T mutation in the gene. Using in situ assays that measure lysosomal function in living cells, the researchers found that activity of GCase, as well as other lysosomal hydrolases, started to sag between 75 to 130 days in culture, just after α-synuclein aggregates appeared in the cell bodies of the neurons. Trafficking of GCase and other proteins to the lysosome slowed.

How might α-synuclein be jamming protein delivery to lysosomes? The researchers hypothesized that α-synuclein could interact with SNARE proteins that facilitate trafficking from the endoplasmic reticulum (ER) to the Golgi. Indeed, they found that α-synuclein latched on to the SNARE protein ykt6. It did so in neurons from idiopathic PD patients as well as those with pathogenic mutations in α-synuclein, GBA1, and PARK9, but not in those from healthy controls.

Ykt6 is known to exist primarily in an auto-inhibited state in the cytosol, where a farnesyl moiety holds it in a closed conformation. In response to cell stress, ykt6 opens and attaches to ER and Golgi membranes, switching on its SNARE activity to crank up the vesicular trafficking machinery (for review, see Daste et al., 2015). Previous work even pegged ykt6 as a modifier of α-synuclein toxicity in yeast (Jun 2006 newsThayanidhi et al., 2010; and Khurana et al., 2017). 

Cuddy and colleagues went on to parse how ykt6 poisons lysosomes in cahoots with α-synuclein. They found that when α-synuclein, which normally resides in synapses, accumulates in the cytosol, it binds ykt6 there, trapping ykt6 in its closed, inactive conformation. Knockdown experiments suggested that without ykt6, GCase and other hydrolases fail to traffic to the lysosome, and a backlog of long-lived proteins accumulate in the cell. Overexpressing ykt6 in A53T-α-syn patient-derived neurons not only restored lysosomal function, but also nudged down α-synuclein accumulation. Crucially, these benefits more than doubled when the researchers overexpressed a mutant form of ykt6 that could not be farnesylated.

Might a small-molecule farnesyltransferase inhibitor (FTI) do the same thing? To find out, the scientists tested LNK-754, an FTI that had been trialed for cancer (Moulder et al., 2004). They found that treatment with 1nM of this inhibitor partially restored GCase activity in A53T-α-syn patient neurons, with slightly higher concentrations rescuing it almost completely. LNK-754 also reduced pathological α-synuclein in the neurons by a third, and they survived longer in culture. LNK-754 had no effect in neurons lacking ykt6, suggesting it worked via this SNARE protein.

The researchers next evaluated LNK-754 in wild-type and PD mouse models. Injecting 0.9 mg/kg of it into the abdomens of wild-type mice for 26 days boosted the proportion of ykt6 associated with SNARE complexes on membranes, enhanced GCase activity and trafficking to the lysosome, and reduced levels of total α-synuclein. The same regimen rescued pathological deficits in these phenotypes in 12- to 14 month-old DA-SYN53 mice, which express human A53T-α-syn in dopaminergic neurons. Also, LNK-754 partially restored levels of tyrosine hydroxylase, a key enzyme in dopamine synthesis, and corrected slight balance deficits seen in these transgenic mice.

Vikram Khurana of Brigham and Women’s Hospital in Boston, who had previously reported that ykt6 sways the toxicity of α-synuclein in yeast, noted that the new findings have roots in deeply conserved biology. That Cuddy and colleagues extrapolated the role of ykt6 to the complex biology of human neurons and the mouse brain supports this SNARE protein as a promising therapeutic target, he said.

However, Khurana has reservations about whether FTIs will be the best way to bolster ykt6, noting that these drugs have multiple targets. He prefers gene therapy to raise expression of target genes but acknowledged that such approaches are further from the clinic.

Koenig noted the many targets of FTIs, including the Ras proteins that formed the basis of their development for cancer (for review, see Wang et al., 2017). He said that while previous clinical studies established maximum tolerated doses for these drugs in cancer, these doses would need to be much lower to be acceptable for indications such as Parkinson’s, which would require long-term treatment in people without a terminal illness.

Still, the effects of FTIs on other targets have suggested benefits in proteinopathies (Cheng et al., 2013). Ten years ago, LNK-754 underwent a Phase 1 pharmacokinetic trial in healthy volunteers with an eye toward development as an autophagy booster in Alzheimer’s, though clinical development ended there (see Peter Lansbury of Harvard Medical School, a co-author on the current paper and chief scientific officer of Link Medicine Corporation at the time, used another FTI to target a farnesylated, membrane-bound form of the ubiquitin C-terminal hydrolase-L, which promoted accumulation and neurotoxicity of α-synuclein (Mar 2009 news). That compound, FTI-277, worked by blocking association of UCH-L1 with membranes.

More recently, researchers led by Kenneth Kosik at the University of California, Santa Barbara, reported that the FTI lonafarnib revved up lysosomes in tauopathy models, preventing accumulation of tau tangles and behavioral deficits (Mar 2019 news). It did so primarily by targeting Rhes, a member of the Ras family of GTPases. This compound is currently in trials for hepatitis and for progeria, an accelerated aging syndrome for which it is available under an expanded-access program ( 

“While FTIs have now been implicated as a potential therapeutic option for two neurodegenerative inclusions, it is important to point out the difficulty of pinpointing the precisely relevant farnesylation substrates because this post-translational modification is so widely used,” Kosik wrote to Alzforum. The different FTIs affect their substrates differently. In the case of Rhes, lonafarnib triggered loss of its association with membranes, while LNK-754 had the opposite effect on ykt6.

Regardless of the specific approach, will stimulating lysosomal function become a therapeutic strategy for neurodegenerative proteinopathies? Most commentators contacted by Alzforum agreed that helping lysosomes clear accumulating proteins in the aging brain could have benefits across proteinopathies.

Case in point, a paper published today in Nature describes a small molecule that relegated mutant Huntingtin protein (mHtt) to this disposal system. Scientists led by Chen Ding of Fudan University in Shanghai and Shouqing Luo of University of Plymouth, England, U.K, identified a small molecule that latched onto the polyglutamine tracts of mHtt as well as to LC3b, a membrane component of autophagosomes, essentially tethering the two together. Ultimately, the compound led to lysosomal digestion of mHtt but not the wild-type protein. This was reported to work in iPSC-derived neurons from Huntington’s disease patients and in an HD mouse model, where it assuaged behavioral deficits. The researchers report that their compound did the same for other proteins harboring polyglutamine expansions, including ataxin-3 (Li et al., 2019).—Jessica Shugart


  1. These two papers nicely complement each other in saying that lysosomal dysfunction and accumulation of α-synuclein are related, and are a potentially tractable therapy for PD. However, the two papers take different mechanistic approaches and therefore are worth comparing with each other. In some ways, the Burbulla et al. approach is more straightforward—if we know that loss of GCase activity is associated with PD, then increasing activity would be predicted to be therapeutically beneficial.

    There are some theoretical uncertainties about this assumption, as there are, for example, variants in the encoding GBA gene that are not associated with Gaucher’s disease but increase risk of PD. However, this is less important in the context of the paper; so long as the compounds are helpful in models, the potential benefit for PD patients should be the important thing to focus on.

    One thing that surprised me in the paper was the apparent benefit for idiopathic PD lines, as there is presumably little genetic risk to drive the oxidation of dopamine and accumulation of α-synuclein. It will be important to identify the mechanism by which these reprogrammed idiopathic lines retain these effects, so we can understand how and why GCase-directed molecules would benefit this larger group of patients.

    The Cuddy et al. paper is a more complicated mechanism and a more complex pharmacological approach. The identification of ykt6 as a SNARE protein that interacts with α-synuclein to mediate lysosomal dysfunction is intriguing. Although there is exploration of ER-Golgi trafficking, which is logical given the prior literature in this area, it would have been interesting to also examine ykt6 at the autophagosome, where it is also implicated as being important.

    The targeting of ykt6 via farnesylation is, however, a little bit more complicated than the simple approach used by Burbulla et al. Farnesyl transferases should prevent addition of lipid moiety to multiple cellular signaling proteins. Therefore, the compound used certainly affects ykt6 but also likely some other proteins, potentially adding to its effects on cells. Whether this would be helpful to human PD remains to be established, although it is potentially of interest that farnesyltransferase inhibitors have been assessed clinically in premature aging syndromes.

  2. Increasingly, the lysosome is becoming the centerpiece for understanding the cell biology of inclusion body neurodegenerative diseases. The fascinating paper by Cuddy et al. adds an intricate layer of regulation over lysosomal activation in the context of the synucleinopathies.

    The study is now the second report to implicate farnesyl transferase inhibition in the activation of the lysosome and consequently amelioration of pathology—in one case, the inhibition of farnesylation prevented tau pathology (Hernandez et al., 2019), and now a mouse model that expresses human A53T α-syn within dopaminergic neurons (DASYN53) was rescued by intraperitoneal injection of a farnesyl-transferase inhibitor (FTI).

    The starting point for the study was the observation that lysosomal dysfunction in the context of induced pluripotent stem cell-derived neurons harboring the A53T α-syn mutation results from disrupted protein maturation (Cooper et al., 2006). The impaired maturation pointed to ER-Golgi traffic, and additional experiments pinpointed a mediator of this traffic, ykt6, which co-IP’ed with α-syn.

    The α-syn-ykt6 complexes occurred in the cytosol, and this complex inhibited ykt6 membrane association. The intellectual jump to a therapeutic was the observation that farnesylation-incompetent ykt6 augmented lysosomal activity compared to the wild-type, which prompted experiments using the selective FTI LNK-754. Treatment with the FTI increased ykt6’s membrane association, and lysosomal activity.

    The counterintuitive effect of an FTI on ykt6 has been explained previously (Wen et al., 2010): The farnesyl group keeps the protein in its water-soluble conformation by sequestering the lipid group inside a hydrophobic pocket. Palmitoylation of ykt6 increases the partition coefficient of the double-lipidated protein to membranes, thereby shifting some pools of the protein from the cytosol to cellular membranes. These data support scrutiny of all the complex regulatory components involved in double-lipidated proteins and their attachment to specific membrane compartments as an entrée for the development of novel pharmaceuticals.

    FTIs have now been implicated as a potential therapeutic option for two neurodegenerative inclusions. That said, it is important to point out the difficulty of pinpointing the precisely relevant farnesylation substrates, because this post-translational modification is so widely used. The relevant substrate for an FTI proposed for tau clearance was Rhes, and in this case, it is ykt6. In the former case, an FTI will result in loss of membrane association; in the case of ykt6, an FTI will increase membrane association. Among the bases for this difference is probably the contrasting ability of these proteins to be palmitoylated.

    Another contrast is the possibly different physico-chemical states of the pathological entities recognized by these two pathways. Rhes can sense the presence of aberrant tau well before aggregates are present, whereas ykt6 appears to sense α-syn after the mutant form begins to harden. This difference points to the importance of identifying the phase state of a disordered protein that activates proteostasis.


    . A farnesyltransferase inhibitor activates lysosomes and reduces tau pathology in mice with tauopathy. Sci Transl Med. 2019 Mar 27;11(485) PubMed.

    . Alpha-synuclein blocks ER-Golgi traffic and Rab1 rescues neuron loss in Parkinson's models. Science. 2006 Jul 21;313(5785):324-8. PubMed.

    . Lipid-Induced conformational switch controls fusion activity of longin domain SNARE Ykt6. Mol Cell. 2010 Feb 12;37(3):383-95. PubMed.

    View all comments by Kenneth Kosik

Make a Comment

To make a comment you must login or register.


News Citations

  1. Protecting Neurons by Ramping Up Waste Disposal?
  2. Lindquist Leaves Behind Parkinson’s Interactome as Her Parting Gift
  3. Feedback Loop—Molecular Mechanism for PD, Gaucher’s Connection
  4. Lysosomes Take Center Stage in Parkinson’s and Frontotemporal Dementia
  5. ER-Golgi Traffic Jam Explains α-Synuclein Toxicity
  6. Striking at Synuclein by Driving Degradation
  7. Compound Stimulates Lysosome, Clears Tau in Human Cells and Mice

Paper Citations

  1. . The Parkinson's disease protein alpha-synuclein disrupts cellular Rab homeostasis. Proc Natl Acad Sci U S A. 2008 Jan 8;105(1):145-50. PubMed.
  2. . Activation of β-Glucocerebrosidase Reduces Pathological α-Synuclein and Restores Lysosomal Function in Parkinson's Patient Midbrain Neurons. J Neurosci. 2016 Jul 20;36(29):7693-706. PubMed.
  3. . Dopamine oxidation mediates mitochondrial and lysosomal dysfunction in Parkinson's disease. Science. 2017 Sep 22;357(6357):1255-1261. Epub 2017 Sep 7 PubMed.
  4. . Structure and function of longin SNAREs. J Cell Sci. 2015 Dec 1;128(23):4263-72. Epub 2015 Nov 13 PubMed.
  5. . Alpha-synuclein delays endoplasmic reticulum (ER)-to-Golgi transport in mammalian cells by antagonizing ER/Golgi SNAREs. Mol Biol Cell. 2010 Jun 1;21(11):1850-63. PubMed.
  6. . Genome-Scale Networks Link Neurodegenerative Disease Genes to α-Synuclein through Specific Molecular Pathways. Cell Syst. 2017 Feb 22;4(2):157-170.e14. Epub 2017 Jan 25 PubMed.
  7. . A phase I open label study of the farnesyltransferase inhibitor CP-609,754 in patients with advanced malignant tumors. Clin Cancer Res. 2004 Nov 1;10(21):7127-35. PubMed.
  8. . New tricks for human farnesyltransferase inhibitor: cancer and beyond. Medchemcomm. 2017 May 1;8(5):841-854. Epub 2017 Feb 16 PubMed.
  9. . Farnesyl Transferase Haplodeficiency Reduces Neuropathology and Rescues Cognitive Function in a Mouse Model of Alzheimer's Disease. J Biol Chem. 2013 Oct 17; PubMed.
  10. . Allele-selective lowering of mutant HTT protein by HTT-LC3 linker compounds. Nature. 2019 Nov;575(7781):203-209. Epub 2019 Oct 30 PubMed.

External Citations


Further Reading


  1. . Is Parkinson's disease a lysosomal disorder?. Brain. 2018 May 30; PubMed.

Primary Papers

  1. . Stress-Induced Cellular Clearance Is Mediated by the SNARE Protein ykt6 and Disrupted by α-Synuclein. Neuron. 2019 Oct 9; PubMed.
  2. . A modulator of wild-type glucocerebrosidase improves pathogenic phenotypes in dopaminergic neuronal models of Parkinson's disease. Sci Transl Med. 2019 Oct 16;11(514) PubMed.