Parkinson’s disease (PD) is a devastating illness. It gradually robs patients of motor control and can ultimately cause dementia. Kumar Narayanan, a neurologist at the University of Iowa, treats people with PD. “Every week, I have to watch them get worse,” he said. 

As more people live longer, cases continue to rise because the disease’s biggest risk factor is aging (1). The root cause of PD varies from patient to patient, but many researchers have come to believe that a failure in the body’s ability to provide adequate fuel to neurons plays a central role across the board (2). “As we age, we think that all this gets a little bit worse — our ability to fuel our brain,” said Tim Ryan, a biochemist at Weill Cornell Medicine.

In a recent Science Advances study, Ryan and his collaborators showed that increasing expression of phosphoglycerate kinase 1 (PGK1) — one of 10 enzymes that cells use to harvest energy from glucose via glycolysis — restored function in neurons with a PD-linked mutation in an unrelated gene (3). Their results highlight the outsized importance of PGK1 in the cellular energy production process and validate the enzyme as a worthwhile therapeutic target for certain forms of PD.

Ryan became interested in PGK1 and PD while following the work of Lei Liu, a neuroscientist at Capital Medical University. In 2014, Liu and a team of scientists found that terazosin, a medication for prostate enlargement, improved the survival rates of flies experiencing widespread organ death (4). Further analysis revealed that terazosin’s beneficial effects arose from its off-target interactions with PGK1, whose function the drug appeared to enhance.

Liu, in collaboration with his former PhD advisor Michael Welsh, a pulmonologist at the University of Iowa, soon realized that terazosin boosted energy availability in neurons, so Welsh walked over to Narayanan’s office to ask if that could be important. “Yeah, it could be important,” replied Narayanan. 

In 2019, the three scientists showed that terazosin slowed the progression of PD-like symptoms in rodents and flies (5). Further, there were fewer cases of PD in patients who had taken terazosin for prostate enlargement in the previous two decades compared to those who had treated their prostate enlargement with drugs that did not interact with PGK1.

Initially, Ryan wasn’t sure what to think. “I had my doubts, but I had the hope that, well, if they’re right, then it’ll be something cool that we’ll be good at studying” he said. He and his team had lots of experience investigating how neurons make and use energy.

Around the time that Liu published his first study on terazosin, Ryan developed a method to track levels of ATP — the molecule that powers essential chemical reactions throughout the body — at the synapses where neurons communicate with each other (6). Ryan’s 2014 study showed that neurons break down glucose at their synapses to maintain enough local ATP to produce signals consistently when stimulated. Specifically, neurons consume a lot of ATP to recycle synaptic vesicles, which hold and release the neurotransmitters that serve as messages from one cell to another.

If you waited a day to make it, it’s sort of game over. 
- Tim Ryan, Weill Cornell Medicine

Ryan had doubts about the neuroprotective effects of enhancing PGK1 because it’s just one of 10 enzymes involved in glycolysis. If enhancing PGK1 alone could boost rates of ATP synthesis, then PGK1 had to be the one enzyme that set the rate for the entire 10-step process. 

“That seemed far-fetched,” said Ryan. “It was never thought to be the rate-limiting step, although there wasn’t really any good data.”

In the present study, Ryan and his colleagues addressed this question by testing the capabilities of cultured neurons with sufficient or limited access to glucose. The scientists used fluorescent imaging to monitor how many times a neuron could respond to an electrical stimulus and recycle its synaptic vesicles, and they tracked how ATP levels at the synapse changed before and after a stimulus. Neurons without adequate fuel availability failed to replenish their ATP between stimuli and exhausted their ability to recycle their synaptic vesicles promptly. For neurons in the brain, that delay in ATP production is especially problematic.

“That process is going to get completely mucked up,” said Ryan. “If you waited a day to make it, it’s sort of game over.”

Ryan’s team then checked how neurons performed if the cells overexpressed each of the 10 glycolytic enzymes. Increasing PGK1 expression, and only PGK1 expression, allowed sugar-starved neurons to perform like their counterparts that had adequate access to fuel. Treating neurons with terazosin had a similar effect. These interventions did not affect the resting levels of ATP in neurons that had not yet been stimulated. Rather, the treatments allowed sugar-starved neurons to make ATP on time. Ryan’s team concluded that PGK1 is the rate-limiting step in glycolysis.

Satisfied with that answer, Ryan and his colleagues checked if increasing PGK1 activity could really be a PD treatment strategy. Although PD has many different genetic and environmental causes, the researchers focused on PD-linked mutations that previous research had connected to issues with synaptic vesicle recycling. They zeroed in on two genes: synaptojanin 1  and Parkinsonism associated deglycase (DJ-1) .

Once again, the researchers tracked the response of neurons to stimuli by monitoring fluorescent molecules in the cells’ synaptic vesicles. Overexpression of PGK1 or  treatment with terazosin rescued the capabilities of neurons with a faulty version of synaptojanin 1, indicating that enhancing PGK1 activity could be an effective therapeutic intervention for patients with this mutation. 

However, overexpressing PGK1 in neurons with impaired DJ-1 genes failed to rescue the capabilities of those cells. Surprised, Ryan and his team decided to probe the relationship between PGK1 and DJ-1. They found a weak association between the two proteins when they pulled them out of solution, and they found a feedback loop whereby overexpression of one protein compensated for reduced expression of the other. Their results indicated that DJ-1 facilitates the function of PGK1, indicating that enhancing the latter’s activity would not be an effective treatment for patients with mutations in the former. Although that finding takes a potential treatment strategy off the table for a subset of patients with PD, Narayanan, who was not involved in this study, was excited to see the researchers identify a previously unknown functional link between the two proteins.

“That was an absolutely wild, very creative advance from this story,” he said.

In science, it’s gratifying to make a discovery. … But when someone independently replicates something, that’s when we get really excited. 
- Kumar Narayanan, University of Iowa

Ryan’s team wrapped their study with one more model of PD. The scientists injected mice with a neurotoxin, monitored the animals for changes in movement patterns, and ultimately took brain samples to determine how much neurodegeneration had occurred. Although the neurotoxin-induced PD-like symptoms in the mice and caused corresponding neurodegeneration, prophylactic overexpression of PGK1 protected mice from these severe symptoms. In other words, Ryan’s team presented evidence supporting Liu, Welsh, and Narayanan’s claim that terazosin protects against PD by enhancing PGK1 activity.

“In science, it’s gratifying to make a discovery,” said Narayanan. “But when someone independently replicates something, that’s when we get really excited.”

Still, neither Narayanan nor Ryan view the new study as an explicit endorsement for treating or preventing PD with terazosin. Instead, the results help strengthen the theory that PD, despite its disparate environmental and genetic causes, is fundamentally a disorder of neuronal energetics. Further, Ryan and his team’s work lays the foundation for designing new treatments specifically for PD, which continues to increase in worldwide prevalence.

“This study is a step towards the larger goals,” said Narayanan.

References

1. Dorsey, E. et al. The emerging evidence of the Parkinson pandemic. J Parkinsons Dis  8, S3-S8 (2018).
2. Blaszczyk, J. The emerging role of energy metabolism and neuroprotective strategies in Parkinson’s disease. Front Aging Neurosci  10, 301 (2018).
3. Kokotos, A. et al. Phosphoglycerate kinase is a central leverage point in Parkinson’s disease-driven neuronal metabolic deficits. Sci Adv  10, eadn6016 (2024).
4. Chen, X. et al. Terazosin activated Pgk1 and Hsp90 to promote stress resistance. Nat Chem Biol   11, 19-25 (2015).
5. Cai, R. et al. Enhancing glycolysis attenuates Parkinson’s disease progression in models and clinical database. J Clin Invest  129, 4539-4549 (2019).
6. Rangaraju, V. et al. Activity-driven local ATP synthesis is required for synaptic function. Cell  156, P825-835 (2014).