As Parkinson’s researchers push beyond symptom management to disease modification, they need far better tools to gain a deeper understanding of how α-synuclein misbehaves in the brain. Topping their wish list is a PET tracer to visualize Lewy bodies in living people, and at the first Advances in Alzheimer’s and Parkinson’s Therapies Focus Meeting (AAT-AD/PD), held March 15–18 in Turin, Italy, there finally was some anticipatory buzz based on a candidate that will head to human trials this year. Researchers also need better animal models, and here, too, new data showed promise when a Dutch group presented a marmoset model that appears to mimic PD better than rodents. Other researchers explored how pathology, risk factors, and symptoms interact, with one reporting that chronic stress accelerates the spread of injected α-synuclein through the rodent brain, while another tied cognitive decline in PD to a loss of dopamine signaling in the precuneus.
- The first candidate PET tracer for α-synuclein will enter clinical trials this year.
- Marmoset monkeys may make better behavioral PD models than rodents do.
- Chronic stress exacerbates α-synuclein pathology, and deficits in the precuneus correlate with cognitive impairment.
“The field will be completely transformed by new technology,” Walter Koroshetz of the National Institute of Neurological Disorders and Stroke (NINDS) predicted at the conference.
Such is the need for an α-synuclein PET tracer that the Michael J. Fox Foundation, after funding research into it for years, has offered a $2 million prize to the first group to produce one. So far, AC Immune in Lausanne, Switzerland, seems to have a leg up on the contest. Last year, company researchers described a compound that selectively bound α-synuclein fibrils, but not Aβ amyloid, and had good brain uptake in animals (May 2017 conference news). Tracer selectivity for α-synuclein has been a big challenge, because the human brain contains far less α-synuclein than Aβ, hence many candidate tracers also tend to bind Aβ.
In Turin, Jan Stöhr of AC Immune reported that the company has anointed a new lead compound, dubbed AA, which reportedly has improved properties over the earlier candidate. AA has sub-nanomolar affinity for Lewy bodies in brain sections of people who died with PD, and 500-fold selectivity for α-synuclein over Aβ, Stöhr said. In radio-binding assays, the compound bound to recombinant α-synuclein fibrils and more importantly bound to PD brain-derived α-synuclein aggregates with high affinity, again demonstrating target engagement on patient-derived protein. The researchers have now radiolabeled AA with fluorine 18 and tested uptake in animal brains. In rodents, 6.5 percent of the tracer reached the brain, with uptake in one minute and washout in 45, Stöhr reported. In primates, 4 percent reached the brain, with uptake in five minutes and washout in 20. The company, together with its partner Biogen, is planning the first human study for later this year.
“These data are the most impressive I’ve seen [for α-synuclein tracers],” said John Trojanowski of the University of Pennsylvania, Philadelphia. Jamie Eberling of MJFF agreed. “Selectivity has been a big stumbling block for α-synuclein tracer development. The AC Immune compounds are the first I’ve seen that looked to be extremely selective, along with other favorable characteristics,” she wrote to Alzforum. Another researcher at the AAT-AD/PD talk wondered whether the compound would enter cells to bind α-synuclein deposits. Stöhr said the fact that it passes through the blood-brain barrier so quickly suggests that it will also cross the cell membrane.
PET tracers will identify α-synuclein pathology directly. In the meantime, researchers gauge this pathology indirectly by imaging dopaminergic degeneration with 123I-FP-CIT SPECT, which detects dopamine transporters (e.g., Jung et al., 2018). Can this type of scan relate degeneration to symptoms? At AAT-AD/PD, Andrea Pilotto of the University of Brescia, Italy, described how he used CIT SPECT to investigate the relationship between dopaminergic transmission and cognition. One in two Parkinson’s patients decline cognitively, but researchers cannot predict who has the more severe prognosis.
Working in the lab of Alessandro Padovani, Pilotto scanned 67 PD patients, 34 of whom had mild cognitive impairment. Overall, he saw no link between striatal binding and whether the person was impaired. Pilotto replicated previous work correlating low uptake in the right caudate and right putamen with low MMSE and verbal fluency, respectively, but saw no difference between MCI and controls in this regard. Where he did find a difference was in the precuneus, a piece of parietal cortex hit early on by amyloid plaques in preclinical Alzheimer’s disease. Low CIT binding there did correlate with MCI. Dopamine helps activate the precuneus to perform executive and attention tasks, Pilotto noted, saying his data suggest that a lack of dopamine in the precuneus could underlie cognitive deficits.
In another study, Pilotto used FDG PET to examine brain metabolism in 54 people with PD. One in three had an atypical scan that resembled the FDG PET pattern of AD and dementia with Lewy bodies (DLB); only these patients developed dementia (Pilotto et al., 2018). Once again, the atypical PD and the AD patterns shared waning metabolism in the precuneus. Precuneus dysfunction could be the key to cognitive impairment in PD, Pilotto suggested in Turin. Intriguingly, other research has correlated tau tangle pathology in the precuneus with cognitive impairment in PD and DLB (Sep 2016 news). Pilotto is currently validating his findings with PPMI data, and is following his original cohort longitudinally.
Besides tracers for imaging PD pathology, researchers also need animal models that more fully recapitulate the human disease. Marmosets are sometimes used to make primate models because these New World monkeys weigh less than 2 pounds, breed quickly, yet have some behaviors that resemble those of humans. Over the last three decades, some groups have used marmosets for PD research by injecting the toxin MPTP to kill dopaminergic neurons, but recent refinements to this approach may bring it closer to human disease.
Ingrid Philippens of VU University, Amsterdam, injects a low, 0.5 mg/kg dose of MPTP weekly. This dosage schedule leads to a slowly progressing form of parkinsonism that shares features with the human disease. In the brain, dopaminergic neuron damage activates compensatory mechanisms and triggers α-synuclein pathology, Philippens told Alzforum. At the level of behavior, this approach models prodromal stages of PD, with subtle motor deficits plus other premotor symptoms. For example, the monkeys develop REM sleep behavior disorder. Just like people with prodromal Parkinson’s, affected males kick their female partners during sleep (yes, marmosets do sleep with their partners, see Verhave et al., 2011). The researchers also see genetic differences in the monkeys’ susceptibility to the disease that correlate with neurotransmitter levels (Franke et al., 2016).
In Turin, Philippens described how her team monitored the behavior of freely moving monkeys in their home cages through telemetry. The marmosets wear necklaces that contain an actimeter, similar to the wrist actimeters people wear. As the animals bustle about their cages, the actimeters send signals to a receiver using Actiwatch software. In addition, the marmosets are implanted with receivers that measure EEG activity along with other physiological parameters such as body temperature. The combination of these two devices detects sleep disturbances. The approach provides continuous data, similar to what some PD trials are now collecting to better track small, day-to-day changes (May 2017 conference news). In addition, the researchers place a touchscreen in the marmoset’s home cage to evaluate changes in its learning and memory. Such tests typically require the animal learn which of two images on the screen will provide a food reward, and then unlearn this association when the reward is paired with the other object (e.g., Kangas et al., 2016). The use of tests more comparable to those used with people may facilitate translation of therapies to the clinic, the scientists noted.
Other researchers are investigating ways to induce parkinsonism in marmosets without a toxin. Fifteen years ago, a human α-synuclein transgene introduced into marmosets was found to cause pathology and dopaminergic neuron loss (Feb 2003 news) and more recently, Japanese scientists triggered Lewy body pathology, spread, and some dopaminergic neuron loss by injecting synthetic α-synuclein fibrils into the basal ganglia of non-transgenic marmosets (Shimozawa et al., 2017).
Koroshetz noted that the value of a given model depends on its purpose. MPTP models were essential in deciphering circuit disruption due to loss of dopaminergic neurons and led to the development of deep-brain stimulation as a therapy. However, they do not address the spread of α-synuclein pathology believed to underlie PD. “The marmoset model from Japan may be very helpful if they have pathology that resembles what we see in PD patients,” he wrote to Alzforum. In Turin, Koroshetz noted that the behavioral features of this model look identical to PD.
To better understand how α-synuclein spreads and affects function, other scientists are incorporating additional drivers of PD into the models they already have in an effort to build more “complete” representation of the human disease. Consider, for example, an attempt to build stress and depression into a mouse model by Johannes Burtscher in the lab of Hilal Lashuel of École Polytechnique Fédérale de Lausanne, Switzerland.
Burtscher noted in Turin that anxiety and depression increase a person’s risk for PD. These emotions are processed in the amygdala, where Lewy bodies are common; what’s more, aggregated α-synuclein injected into mouse striatum spreads to many connected regions, including the amygdala (Popescu et al., 2004; Luk et al., 2012). Burtscher found that aggregated α-synuclein levels in the amygdala peaked 30–90 days after injecting preformed fibrils into the dorsal striata of wild-type mice.
To see what the arrival of this pathology might do to the amygdala, Burtscher injected 5–10 μg of preformed α-synuclein fibrils into the striatum and checked behavior one month later. The mice were less afraid of open spaces than controls, but seemed passive, showing less interest in new objects and moving less during a forced swim test. Because an overactive amygdala is linked to anxious behavior, these data suggest the opposite process, namely an underactive amygdala, Burtscher said. He noted that apathy and emotional numbing are also seen in the prodromal stages of PD.
Because chronic stress also increases risk for PD, Burtscher wondered if stress could affect pathology. He slipped corticosterone into the mice’s drinking water for a month to stress them. In response, the mice gained weight and showed signs of depression on classic rodent tests, giving up quickly in forced swim tests and showing no interest in sugar water. Burtscher then injected the aggregated α-synuclein into their striata while continuing with corticosterone for another month. In the stressed mice, α-synuclein pathology spread farther than in controls, reaching the entorhinal cortex, whereas in controls it stopped at the substantia nigra and amygdala. Burtscher also saw more degeneration of dopaminergic neurons in the stressed group. Chronic stress predisposes α-synuclein to spreading, Burtscher concluded.—Madolyn Bowman Rogers
- α-Synuclein Antibodies Enter Phase 2, Sans Biomarker
- Tau Deepens Cognitive Trouble in Lewy Body Diseases
- Do Smartphones Collect Better Clinical Data Than Paper-and-Pencil Tests?
- Viral Transgene Models Parkinson's in Primate
- Jung Y, Jordan LG 3rd, Lowe VJ, Kantarci K, Parisi JE, Dickson DW, Murray ME, Reichard RR, Ferman TJ, Jones DT, Graff-Radford J, Savica R, Machulda MM, Fields JA, Allen LA, Drubach DA, St Louis EK, Silber MH, Jack CR Jr., Knopman DS, Petersen RC, Boeve BF. Clinicopathological and 123 I-FP-CIT SPECT correlations in patients with dementia. Ann Clin Transl Neurol. 2018 Mar;5(3):376-381. Epub 2018 Jan 24 PubMed.
- Pilotto A, Premi E, Paola Caminiti S, Presotto L, Turrone R, Alberici A, Paghera B, Borroni B, Padovani A, Perani D. Single-subject SPM FDG-PET patterns predict risk of dementia progression in Parkinson disease. Neurology. 2018 Mar 20;90(12):e1029-e1037. Epub 2018 Feb 16 PubMed.
- Verhave PS, Jongsma MJ, Van den Berg RM, Vis JC, Vanwersch RA, Smit AB, Van Someren EJ, Philippens IH. REM sleep behavior disorder in the marmoset MPTP model of early Parkinson disease. Sleep. 2011 Aug 1;34(8):1119-25. PubMed.
- Franke SK, van Kesteren RE, Hofman S, Wubben JA, Smit AB, Philippens IH. Individual and Familial Susceptibility to MPTP in a Common Marmoset Model for Parkinson's Disease. Neurodegener Dis. 2016 Mar 22; PubMed.
- Kangas BD, Bergman J, Coyle JT. Touchscreen assays of learning, response inhibition, and motivation in the marmoset (Callithrix jacchus). Anim Cogn. 2016 May;19(3):673-7. Epub 2016 Feb 3 PubMed.
- Shimozawa A, Ono M, Takahara D, Tarutani A, Imura S, Masuda-Suzukake M, Higuchi M, Yanai K, Hisanaga SI, Hasegawa M. Propagation of pathological α-synuclein in marmoset brain. Acta Neuropathol Commun. 2017 Feb 2;5(1):12. PubMed.
- Popescu A, Lippa CF, Lee VM, Trojanowski JQ. Lewy bodies in the amygdala: increase of alpha-synuclein aggregates in neurodegenerative diseases with tau-based inclusions. Arch Neurol. 2004 Dec;61(12):1915-9. PubMed.
- Luk KC, Kehm VM, Zhang B, O'Brien P, Trojanowski JQ, Lee VM. Intracerebral inoculation of pathological α-synuclein initiates a rapidly progressive neurodegenerative α-synucleinopathy in mice. J Exp Med. 2012 May 7;209(5):975-86. PubMed.