The innate immune system factors heavily in Parkinson’s disease, but evidence has been thinner for its pickier twin, the adaptive immune system. The latter draws renewed suspicion from a paper in the June 21 Nature. In it, scientists led by David Sulzer of Columbia University, New York, and Alessandro Sette, La Jolla Institute for Allergy and Immunology, California, report that T cells from PD patients readily respond to peptides derived from α-synuclein, a protein that aggregates in Parkinson’s. Specifically, fragments of α-synuclein are displayed on neurons of the substantia nigra by major histocompatibility proteins (also called human leukocyte antigen, or HLA, proteins) whose genes are known to be associated with PD. “We are showing that in PD, there is a response against the patients’ own proteins,” Sulzer told Alzforum. “That opens up the possibility that there’s an autoimmune response in PD.”

“The current study creates a functional link between the genetic associations of the HLA-DRB genes and the pathological changes that are correlated with disease progression,” said Andrew West, University of Alabama, Birmingham. “It’s uncovering interesting biology around α-synuclein.”

Cells chop up proteins into peptides and display them on their surface via MHC proteins. Passing T cells patrolling the brain and body are trained to recognize foreign matter and ignore anything that comes from the “self.” When a T cell comes across a foreign peptide that matches its unique receptor, it binds and proliferates. The resulting cells become helper or cytotoxic T cells that generate an immune response or kill the host cell.

Certain MHC proteins, such as DRB1 and DRB5, had been tied to PD previously, but it was unclear what they have to do with the disease. A clue came with Sulzer and colleagues’ previous report that dopaminergic neurons in the substantia nigra and locus coeruleus of human brains presented antigens on MHC proteins (Cebrián at el., 2014). Most neurons don’t present antigens, but these particular ones do, and they die off in droves in PD. These researchers found that mouse substantia nigra neurons that were made to express foreign antigens died when the corresponding cytotoxic T cells were added. Could this property of antigen presentation play a role in the neurons’ demise in PD?

To find out, the researchers joined forces with the Sette lab to sample blood from 67 patients with PD and 36 age-matched controls. They separated the white blood cells and exposed them to fragments of α-synuclein to see if they would become activated to release either interferon-g (IFNg) or interleukin-5 (IL-5). Two fragments did the job: one at the N terminus known as the Y39 region, and one at the C terminus, known as S129. To react with T cells, the latter required a phosphorylation at S129, a modification found in α-synuclein peptides of Lewy bodies (Fujiwara et al., 2002). Responses were pronounced in T cells from PD patients, but relatively rare in healthy controls.

These specific α-synuclein fragments were posted on two MHC proteins previously tied to PD, i.e. DRB1*15:01 and DRB5*01:01 (Wissemann et al., 2013). Importantly, the alleles for these particular MHC proteins were twice as common in PD patients as in controls. DQB1*03:04 and the MHC class I allele A*11:01 also appeared more frequently in responders, and A*11:01 bound the Y39 α-synuclein fragment. Every participant whose T cells responded to the Y39 α-synuclein peptide carried one of those four HLA variants. The results imply that specific MHC alleles are linked with Parkinson’s because they specialize in displaying α-synuclein fragments.

Patient T cells were also activated in response to protofibrils of α-synuclein, whereas T cells from healthy people had little if any response. The few people whose T cells did respond could be in the early stages of PD, Sulzer said. Overall, his results imply that T cell responsiveness to α-synuclein could be a potential early biomarker of disease, he said.

How might this play out in Parkinson’s? Sulzer hypothesized that as PD takes hold, accumulating α-synuclein proteins are no longer sufficiently degraded via their normal clearance pathways. Neurons instead break it down into peptides that appear on MHC signposts. Alternatively or in addition, α-synuclein may be cut up outside cells and those peptides bind to MHC complexes on neurons; microglia may also digest the protein and themselves display the peptides on MHC molecules, Sulzer said. He noted that he has yet to test whether α-synuclein appears on neuronal MHC complexes.

In all scenarios, since these peptides are not around early in life when the immune system is maturing, T cells are not trained to tolerate them as “self.” Instead, the T cells mistake these MHC-presented α-synuclein peptides for foreign material. T cells then proliferate and may attack neurons, or cause an inflammatory response that causes bystander damage, which may explain why they die, Sulzer speculated. Whether the activated T cells kill neurons in human disease remains to be seen, he cautioned, as his group has only observed that in mouse models.

“This provides further evidence for a causal role of adaptive immunity in the etiology of PD,” wrote Heidi McBride, McGill University, Montreal, to Alzforum. “This opens new possibilities for strategic therapeutic intervention to inhibit the immune system to halt progression of PD.”

West noted that the cohort was small and cross-sectional, and said researchers need to test if the finding generalizes to larger patient samples. He also wondered if the immune state is a lifelong one, or reflects a late-life reaction to disease. Since only 40 percent of all PD patients in this study had T cells reactive to α-synuclein, West suggested there might be disease subgroups that respond differently to anti-inflammatory therapies.

The results align with some previous reports that α-synuclein can cause a T cell response in mice and rats (see Theodore et al., 2008; Benner et al., 2008). T cells were previously found to infiltrate the substantia nigra in mouse models of PD (Brochard et al., 2009). Sulzer and colleagues will try to replicate these data in an independent set of patients and controls. They are also looking for ways to better model this response in the lab.—Gwyneth Dickey Zakaib


  1. The elegant studies by Sulzer and Sette are a significant step forward for the neurodegeneration field, and a compelling example of the great things that can happen when neuroscientists and immunologists join forces. These findings advance our understanding of how the innate and adaptive immune systems might orchestrate a response to neuronal signals resulting from accumulation of aggregation-prone endogenous proteins like α-synuclein that have been post-translationally modified inside the cell. The authors demonstrated that a phosphorylated site on synuclein was a high-affinity binding site. It also would have been interesting to test whether nitrosylated α-synuclein, a post-translational modification of synuclein shown to be antigenic, might have elicited a different or more robust response relative to phosphorylation of α-synuclein.

    Although these findings need to be replicated in a larger independent cohort, the results by Sulzer and Sette have immediate implications for the development of immunomodulatory clinical interventions for Parkinson’s disease. Selective targeting of cytotoxic T cell subsets and boosting protective T cell subsets early in the course of PD may be a feasible therapy based on success of such approaches in the multiple sclerosis (MS) clinic. One promising example is that reported by Gendelmann et al. (2017) where GM-CSF was administered to PD patients to boost regulatory T cells.

    In addition, the findings by Sulzer and Sette compel follow-up studies to identify the location— brain versus peripheral lymph nodes—where antigen presentation to T cells is occurring; one safe bet is in the deep cervical lymph nodes. Also of interest is whether other antigen-specific T-cell subsets exist, as subsequent sequencing of their T-cell receptors (TCRs) will aid in identification of ligands that trigger their differentiation and proliferation in the Parkinson’s state.

    Given that the No. 1 risk factor for PD is aging, one outstanding question is the extent to which this immune response changes as the immune system itself ages through a process called immunosenescence. One might envision that a young innate immune system response might include production of neurotrophic factors to boost neurons in trouble and/or to promote phagocytosis and removal of extracellular aggregates that might leak out of the cell. In contrast, an aging innate immune system may less efficiently remove protein aggregates and instead present processed antigens to naïve T-cells in ways that may be deemed as foreign antigens.

    Finally, these studies raise the intriguing possibility that an orchestrated response of the innate and adaptive immune systems may play an important role in other neurodegenerative diseases, including Alzheimer’s and frontotemporal dementia, which are the most common dementias affecting older and middle-aged individuals. In support of this idea, common genetic variants in an antigen-presenting gene implicated in the findings by Sulzer and Sette, DRB5.1, have also been identified as risk factors in sporadic AD. Furthermore, mutations and variants in genes that encode proteins highly expressed in innate and adaptive immune cells, such as PU.1/ SPI1, TREM2, CD33, progranulin (GRN), TMEM106B, and C9ORF72, have been reported to be risk factors for FTD and AD. Investigations to elucidate the role of these proteins in orchestrating the innate-adaptive immune system cross-talk that keeps neurons healthy, and how that goes awry to increase the risk of neurodegeneration, will provide a mechanism-based rationale for interventional immunomodulatory therapies in these diseases.


    . Evaluation of the safety and immunomodulatory effects of sargramostim in a randomized, double-blind phase 1 clinical Parkinson’s disease trial. npj Parkinson's Disease 3, Article number: 10 (2017 March 27)

  2. "T cells may be tricked into thinking dopamine neurons are foreign due to the build-up of damaged α-synuclein." The key point here is the bioaccumulation of a damaged protein. So in this case the damaged protein came before the immune response. This response is therefore similar to the formation of a hapten in which drugs form covalent bonds with endogenous proteins/peptides, which in turn induces a T cell-mediated hypersensitivity reaction (see Pavlos et al., 2015). The modified peptides are then processed by antigen-presenting cells. By contrast, if the system attacked normal α-synuclein, which is abundant in the brain, it would be an aberrant immune response and thus, more akin to what we generally define as an auto-immune reaction. 


    . T cell-mediated hypersensitivity reactions to drugs. Annu Rev Med. 2015;66:439-54. Epub 2014 Oct 27 PubMed.

  3. Recent evidence has confirmed the importance of adaptive immune responses in neurodegeneration in general, and the emergence of a chronic neuroinflammatory state in PD. This is supported by evidence that mutations affecting cytokine gene loci, such as IL-1β or TNF-α, increase incidence rates of PD by threefold. This new paper by Sulzer and co-workers adds an intriguing piece to this puzzle by revealing: i) that 2 α-synuclein fragments (Y-39 and S-129) act as antigenic epitopes for triggering an adaptive immune response; ii) that immunogenicity is not susceptible to nitration or phosphorylation, suggesting that α-synuclein post-transcriptional modification is detrimental for cytoplasmic cell functions, but not for eliciting the immune response; iii) that HLA class II variants, DRB1*15:01 and DRB5*01:01, bind the immunogenic portions of α-synuclein, indicating a strong genetic link between the immune response and PD in the study cohort.

    However, numerous questions still remain. They include, for instance: Is the priming of T-helper lymphocytes prodromal? Does it occur during their maturation in the thymus at the embryonic stage? Or is priming a later event and an indicator of disease progression? The latter would provide a unique opportunity to diagnose disease status and provide personalized therapy. Moreover, T cell priming may be exploited in clinical trials for patient stratification, and may be linked with the haplotype for HLA gene. Not addressed in this context were the immunological mechanisms that cause the innate immune response. Special consideration needs to be given to the functionality of the blood-brain barrier and Virchow-Robin spaces as putative drainage systems because failure to clear α-synuclein could precipitate the immune response.

    Nevertheless, the identification of several N- and C-terminal antigen regions within α-synuclein confirms their direct activation of the adaptive immune system and that HLA class I and II are instrumental for the antigen presentation mechanism and CD4+ T-lymphocyte activation. These results should be exploited to investigate the existence of similar mechanisms in other neurodegenerative proteinopathies. 

  4. The idea that PD as well as AD neurodegeneration might be mediated by autoimmunity has been lingering in the field since the 1980s (Rogers et al., 1988; Barker and Cahn, 1988). In AD, Aβ peptide-reactive CD4 T cells have been described by Monsonego et al., 2003, and found to be more numerous in peripheral blood from patients than control.

    The present paper from Sulzer et al. now offers, for the first time, similar evidence of “autoimmune” responses in peripheral blood from PD patients.

    Whether this phenomenon has clinical relevance is hard to conclude in both AD and PD. We propose the following points for further discussion:

    First, α-syn-reactive T cells seem to be extremely rare, and an extended incubation of 14 days with peptide was necessary to obtain a positive signal with ELISPOT. Even under these extreme conditions, the authors report that only 0.2 percent of CD3+ T cells responded to peptides (Extended data figure 2).

    Second, it is still not clear whether these α-syn-reactive T cells are the ones that infiltrate the brains of PD patients, and, if so, what their activation state is once they are there (Th1, Th2, TH17, T reg, anergic?). In AD, a growing body of literature suggests that infiltrating T cells are not overly active (Ferretti et al., 2016), and that brain-infiltrating T cells with regulatory phenotype (induced with transient peripheral depletion and then re-bound) could actually be beneficial in animal models (Baruch et al., 2015).

    Finally, the data presented in the paper seem to indicate that the response to α-syn is not specific to T cells from PD patients, since a signal was often observed also in T cells from healthy control individuals. What seems to be driving the response is the occurrence of certain MHC alleles, two already linked to PD risk (DRB1*15:01 and DRB5*01:01) and two new (DQB1*03:04 and A*11:01). Carrying at least one of these alleles was linked to immune response in both PD (p<0.00007) and healthy controls (p<0.009). 

    The rather frequent A*11:01 reactivity in the PD cohort, but not the control cohort, supports also the notion that this allele is present with increased frequency in PD patients; in fact, within the respective North American population of Caucasian origin, only 1.3 percent of this allele should be found (2n). The lack of response in the control cohort may thus be the result of absence of the respective allele.

    Therefore, the link between these MHC alleles and immune response to α-syn is a very important finding that ought to be further investigated in larger cohorts. It is in our view crucial to establish whether the T cell response to α-syn is a PD-specific phenomenon (as the title of the paper seems to imply) or if it is genetically driven, thus determined by presence/absence of antigen-presenting MHC molecules. In the latter case, it would not be surprising to find a higher response in T cells from PD patients, since these alleles are significantly more frequent in PD (circular argument).

    If, using a larger cohort, different associations are found in healthy controls and PD, it will be intriguing to figure out the additional pathological mechanisms responsible for the PD-specific response. Furthermore, whether any responding healthy controls are on their way to developing PD would be also interesting to see, similar to the amyloid-positive non-cognitively impaired individuals in AD. 

    Professor Thorsten Buch (Institute of Laboratory Animal Science, University of Zurich, Switzerland) is co-author of this comment. 


    . Expression of immune system-associated antigens by cells of the human central nervous system: relationship to the pathology of Alzheimer's disease. Neurobiol Aging. 1988 Jul-Aug;9(4):339-49. PubMed.

    . Parkinson's disease: an autoimmune process. Int J Neurosci. 1988 Nov;43(1-2):1-7. PubMed.

    . Increased T cell reactivity to amyloid beta protein in older humans and patients with Alzheimer disease. J Clin Invest. 2003 Aug;112(3):415-22. PubMed.

    . T-cell brain infiltration and immature antigen-presenting cells in transgenic models of Alzheimer's disease-like cerebral amyloidosis. Brain Behav Immun. 2016 May;54:211-25. Epub 2016 Feb 9 PubMed.

    . Breaking immune tolerance by targeting Foxp3(+) regulatory T cells mitigates Alzheimer's disease pathology. Nat Commun. 2015 Aug 18;6:7967. PubMed.

Make a Comment

To make a comment you must login or register.


Paper Citations

  1. . MHC-I expression renders catecholaminergic neurons susceptible to T-cell-mediated degeneration. Nat Commun. 2014 Apr 16;5:3633. PubMed.
  2. . alpha-Synuclein is phosphorylated in synucleinopathy lesions. Nat Cell Biol. 2002 Feb;4(2):160-4. PubMed.
  3. . Association of Parkinson disease with structural and regulatory variants in the HLA region. Am J Hum Genet. 2013 Nov 7;93(5):984-93. Epub 2013 Oct 31 PubMed.
  4. . Targeted overexpression of human alpha-synuclein triggers microglial activation and an adaptive immune response in a mouse model of Parkinson disease. J Neuropathol Exp Neurol. 2008 Dec;67(12):1149-58. PubMed.
  5. . Nitrated alpha-synuclein immunity accelerates degeneration of nigral dopaminergic neurons. PLoS One. 2008;3(1):e1376. PubMed.
  6. . Infiltration of CD4+ lymphocytes into the brain contributes to neurodegeneration in a mouse model of Parkinson disease. J Clin Invest. 2009 Jan;119(1):182-92. PubMed.

External Citations

  1. DRB1 
  2. DRB5

Further Reading


  1. . Genomics implicates adaptive and innate immunity in Alzheimer's and Parkinson's diseases. Ann Clin Transl Neurol. 2016 Dec;3(12):924-933. Epub 2016 Nov 4 PubMed.
  2. . Dopaminergic Regulation of Innate Immunity: a Review. J Neuroimmune Pharmacol. 2017 Jun 3; PubMed.
  3. . The role of innate and adaptive immunity in Parkinson's disease. J Parkinsons Dis. 2013;3(4):493-514. PubMed.

Primary Papers

  1. . T cells from patients with Parkinson's disease recognize α-synuclein peptides. Nature. 2017 Jun 29;546(7660):656-661. Epub 2017 Jun 21 PubMed.