As if the neurodegenerative brain didn’t have enough troubles, a study published in Nature on January 8 reports that it may be swarming with highly trained cellular henchmen. Researchers led by Tony Wyss-Coray at Stanford University found cytotoxic CD8+ T cells in the cerebrospinal fluid and brain tissue of people with Alzheimer’s and Parkinson’s disease. Far from merely surveying the brain, they had multiplied—clonally expanded, in immunology parlance—suggesting they were fighting specific antigens. To what were they responding? For most of the clones, this remains a mystery. However, a small proportion of T cells in three AD patients recognized a snippet of Epstein-Barr virus, a herpesvirus known to infect cells of the CNS.

  • Cytotoxic memory T cells detected in blood, brain, and CSF of people with AD.
  • In CSF, they had multiplied, and appeared activated.
  • In three Alzheimer’s patients, the cells targeted Epstein-Barr virus.

“This beautiful study shows, for the first time, a close association between T cells, cognition, and neurodegenerative disease in humans,” wrote Jonathan Kipnis of University of Virginia in Charlottesville. “These elegant findings on expansion of Epstein-Barr virus-reactive CD8 T cells in the CSF of AD patients are extremely interesting and intriguing,” he added.

Neuroinflammation has moved to center stage in neurodegenerative disease research, but most work focuses on the innate, rather than adaptive, immune system. Still, evidence implicating the adaptive immune system has been quietly trickling in over the years. For example, scientists detected T cells specific for Aβ and α-synuclein in the blood of AD and PD patients, respectively (Bongioanni et al., 1997; Monsonego et al., 2003; Jun 2017 news). While dogma holds that T cells are scarce in the brain, a recently discovered lymphatic drainage system there carries T cells, making it clear that adaptive immune cells can, and do, frequent the CNS (Louveau et al., 2015; Oct 2017 news; Oct 2019 news). Some studies detected more T cells in the brains of people with neurodegenerative disease than in healthy brains (Togo et al., 2002; Town et al., 2005; Sept 2009 news). 

In the new study, first author David Gate and colleagues set out to clarify what the adaptive immune system might do in Alzheimer’s. First, they took stock of immune cell populations in the blood, reporting that people with MCI or AD had more of a subtype of CD8+ T cells than did cognitively unimpaired controls. Specifically, high numbers of a subset of T effector memory cells expressing CD45RA, turned up in AD and MCI. Also known as T-EMRA cells, they are renowned for their killing efficiency, Gate told Alzforum. They rapidly dispense with any cells they recognize and also secrete a slew of proinflammatory cytokines.

In a separate cohort, the researchers found that the more circulating T-EMRA cells a person with AD or MCI had, the worse he or she performed on cognitive tests. Numbers of circulating T-EMRA cells predicted MCI or AD with 80 percent accuracy, suggesting that this particular subset of T cells in the blood was somehow tied to the neurodegenerative process in the brain.

Infiltrating the AD Brain. CD8+ T cells (green) congregate in blood vessels laden with cerebral amyloid angiopathy, and around Aβ plaques (red). [Courtesy of Gate et al., Nature, 2020.]

Could these cells also be in the brain, then? Immunohistochemistry on postmortem brain samples identified more CD8+ T cells in hippocampal parenchyma and adjacent leptomeninges of seven AD patients than of seven controls. The cells congregated around Aβ plaques and neuronal processes. Gate also found numerous CD8+ T cells associated with the blood vessels affected by cerebral amyloid angiopathy in three AD patients, but found nary a T cell around the vasculature of controls. Gate did not measure other markers to determine whether these were T-EMRA cells.

Caught in the Act? A CD8+ T cell (red) in the hippocampus of an AD brain contains cytotoxic granules loaded with Granzyme A, a protein that kills target cells. [Courtesy of Gate et al., Nature, 2020.]

Gate spotted adaptive immune cells in the CSF as well. Most were T cells, but he was surprised to find 20 percent were CD8+, and of those, again 20 percent were T-EMRA. Were they just passing through, or carrying out an immunological hit? To find out, the researchers sequenced T cell receptor genes of individual T cells, which are unique due to the recombination that generates the TCRs. The existence of even two cells with identical receptor sequence indicates clonal expansion, and this occurs only when the killers encounter their cognate antigen (aka prey).

The scientists report evidence of numerous expanded clones in the CSF of 12 people with either MCI or AD (see image below). Clonal expansion was far less common among 10 controls. While one out of 10 healthy controls had a highly expanded clone—i.e., one that comprised at least 3 percent of all TCR sequences—four of six AD patients and one of five people with MCI did.

Moreover, individual clones detected in AD CSF had expanded to a greater degree than those from controls. One AD patient had a CD8+ T cell clone that comprised a whopping 44 percent of all CD8+ T cells in their CSF. Roughly two-thirds of all expanded CD8+ T cell clones belonged to the T-EMRA subset; the other third were T effector memory (T-EM) cells.

Clonal Expansion. Doughnuts show the T-cell pool in the CSF. While a majority of T cells from healthy controls were unique, i.e. not expanded (white), a majority of T cells in AD CSF had clonally expanded two to four times (orange), five to nine times (blue), or more than 10 times (green). For expanded clones, each colored wedge represents the percentage of total TCRs in the pool. Being too numerous, unique clones (white) are not represented by wedges. [Courtesy of Gate et al., Nature, 2020.]

The researchers also found more clonal expansions in the CSF of six people with PD, including two people with highly expanded clones, than in 10 controls. This suggested the phenomenon is not limited to AD.

The cells did not appear benign. Using single-cell RNA sequencing, the researchers found that the most highly expanded T-cell clones in the CSF expressed a bevy of pro-inflammatory cytokine and cytotoxic proteins, including Granzyme A, also made by CD8+ T cells in AD patients. The findings imply that these CNS T cells were killing cells that waved their specific antigenic flags in the brain, commented Terrence Town of the University of Southern California in Los Angeles. “Granzyme A is the smoking gun,” Town added.

To which antigens were these clones responding? For most of the clones, this question is unanswered. Employing TCR sequencing with a large helping of computational wizardry, the researchers identified T-cell clones in two people with AD and one with MCI that carried the T-cell receptors recognizing the same antigen—an epitope from the EBNA3 protein in Epstein-Barr virus. EBV is a herpes virus known to infect the brain. Using an unbiased machine-learning approach, the researchers also picked up T cells in two other patients that reacted to yet another epitope from EBV—this one came from the EBV trans-activator protein BZLF1. No common expanded TCRs were found among controls.

“The presence of the EBV-specific T cells is intriguing, but in no way establishes a link between EBV and AD,” wrote Howard Weiner and Dennis Selkoe of Brigham and Women’s Hospital in Boston in a joint comment to Alzforum. Weiner and Selkoe previously identified T cells specific for Aβ and α-synuclein in the blood of AD and PD patients.

The findings don’t implicate EBV in the pathogenesis of AD. However, Gate speculated that stress and inflammation in the diseased brain could trigger reactivation of latent EBV, which could lead to recruitment of EBV-specific T cells and the slaughter of any infected neurons. Alternatively, or perhaps additionally, changes at the blood-brain barrier could let more T cells into the brain, Gate added. Town wondered whether damage to the integrity of the blood-brain barrier in vessels wracked with cerebral amyloid angiopathy could explain how CD8+ T cells gained access to those regions.

Guillaume Dorothée of INSERM in Paris said that the T-cell clones in the CSF could reflect clonal expansion that took place outside the brain. Perhaps the cells were responding to peripheral antigens, and then were unselectively recruited to the CSF of inflamed brain, he said. A direct comparison of T-cell clones in the blood and CSF of the same patients could help clarify this point, he added. Defining whether parenchymal T cells share antigen specificities with the clones patrolling the CSF will also be important, Dorothée noted. Regardless of what drove the cells into the CSF and brain, their presence supports the idea that T cells play some part in the neurodegenerative process, and that targeting them could alter the course of disease, he said. Dorothée’s previous studies in AD mice suggest that amplifying regulatory T cells, which control immune responses including T-cell immunity, slowed disease progression (Dansokho et al., 2016).

Gate did not test how many of the patients had been infected with EBV, nor directly measure EBV infection in brain samples. In a commentary accompanying the paper, Michael Heneka of the German Center for Neurodegenerative Diseases in Bonn urged caution in interpreting the findings. He noted that more than 95 percent of people are infected by EBV in their lifetime, but a previous study detected EBV DNA in only 6 percent of AD brains (Carbone et al., 2014). 

Several studies have tied Alzheimer’s to various viruses, particularly herpes, though their connection to AD pathophysiology remains controversial (Itzhaki et al., 1997; Jun 2018 newsJun 2018 news).—Jessica Shugart


  1. We and others have been exploring the role of T cells in cognition and in neurodegenerative diseases and aging and this beautiful study shows, for the first time, a close association among T cells, cognition, and AD in humans. These elegant findings on expansion of Epstein-Barr virus-reactive CD8 T cells in the CSF of AD patients are extremely interesting and intriguing. A role of viral infection as a driving factor in pathogenesis of AD has been proposed by Rudy Tanzi and late Rob Moir (who has just passed away a couple of weeks ago). Their ideas were first met with skepticism and now the current work points in the same direction.

    On a personal note, this work is exciting because it shows a human relevance of neuroimmune interactions that we have been studying in mice. Future works will be aimed at better understanding how and why these antiviral T cells are accumulating in the CSF, the function or dysfunction of meningeal lymphatics in drainage of these T cells into the periphery, and the mechanisms underlying their impact on cognition. 


  2. This article by Gate et al. elegantly shows the presence of clonally expanded CD8 T cells in Alzheimer’s disease. The primary question regarding the finding is the degree to which these cells are secondary to the disease process or actually play a pathogenic or disease-amplifying role in AD. Though they aren’t found in healthy, age-matched individuals, they do not appear to be 100 percent specific to AD, because they are also seen to some extent in Parkinson’s disease. Their cross-reactivity with certain Epstein-Barr virus peptides is intriguing but in no way establishes a link between EBV and AD, as the authors also caution. Their apparent linkage to cognition could simply represent their secondary dependence on CNS damage in AD.

    Antigen-specific T cell responses to Aβ peptides and α-synuclein have been shown in AD and PD, respectively, and it is not yet clear how these pro-inflammatory type CD8+ cells relate to T cell antigen-specific responses to pathogenic proteins that accumulate in these diseases. Demonstrating that these expanded clonal T cells change with treatment in people (which is not a simple task) or identifying them in animal models of AD and showing a true pathogenic role early in mouse disease development by manipulating them specifically, will now be important approaches to establish their relevance.

  3. Overall I think this is a beautiful study.

    It makes a clear point that immune components are essential for our understanding of Alzheimer’s disease. Whether these are causal factors or part of the repair mechanism, or a repair system that went wrong, it suggests that we can use the immune system as an early detection tool for malfunctions of the brain and potentially as a treatment vehicle. The detection of clones for non-self antigens, although stated with caution, suggests that pathogenic factors may be part of the pathology. Overall, I think that CD8 cells still receive limited attention in our perception of the immune compartment in the brain, and I am sure that this study will shift the focus.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Trigger Warning: α-Synuclein Sets Off T Cells in Parkinson’s
  2. Lymphatic Vessels Found in Human Brain
  3. Do Immune Cells in the Meninges Help with … Memory?
  4. ALS: T Cells Step Up
  5. Herpes Triggers Amyloid—Could This Virus Fuel Alzheimer’s?
  6. Aberrant Networks in Alzheimer’s Tied to Herpes Viruses

Paper Citations

  1. . T-cell interferon gamma binding in patients with dementia of the Alzheimer type. Arch Neurol. 1997 Apr;54(4):457-62. PubMed.
  2. . 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.
  3. . Structural and functional features of central nervous system lymphatic vessels. Nature. 2015 Jul 16;523(7560):337-41. Epub 2015 Jun 1 PubMed.
  4. . Occurrence of T cells in the brain of Alzheimer's disease and other neurological diseases. J Neuroimmunol. 2002 Mar;124(1-2):83-92. PubMed.
  5. . T-cells in Alzheimer's disease. Neuromolecular Med. 2005;7(3):255-64. PubMed.
  6. . Regulatory T cells delay disease progression in Alzheimer-like pathology. Brain. 2016 Apr;139(Pt 4):1237-51. Epub 2016 Feb 1 PubMed.
  7. . Herpes virus in Alzheimer's disease: relation to progression of the disease. Neurobiol Aging. 2014 Jan;35(1):122-9. PubMed.
  8. . Herpes simplex virus type 1 in brain and risk of Alzheimer's disease. Lancet. 1997 Jan 25;349(9047):241-4. PubMed.

Further Reading

Primary Papers

  1. . Clonally expanded CD8 T cells patrol the cerebrospinal fluid in Alzheimer's disease. Nature. 2020 Jan;577(7790):399-404. Epub 2020 Jan 8 PubMed.