Ebb and Flow of EBV

Epstein-Barr virus (EBV), the nearly ubiquitous herpesvirus that can cause infectious mononucleosis (or “mono”), seems inextricably tied to multiple sclerosis. But the nature of this link remains mysterious. Now, a new study published on 11 April in PLoS Pathogens (Angelini et al., 2013) shows that in patients with relapsing-remitting MS (RRMS), immune responses to active EBV cycle hand in hand with relapse, correlating the virus' activity with MS activity for the first time.

The study examined cytotoxic (CD8+) T cells—immune cells that kill infected or abnormal cells by recognizing specific antigens carried by those cells—and found an increased T cell response to lytic, or active, EBV in the bloodstream of MS patients during disease relapses. The response receded when MS quieted down. Although several different immune cells play a role in responding to viral infections, CD8+ T cells are particularly important in controlling such infections and bind directly to infected, antigen-presenting cells, meaning their response is highly specific to a single viral antigen. These results suggest that viral flare-ups may prompt MS inflammation. Moreover, this correlation could lead to a new blood-based biomarker for MS relapse and points to the possibility that the right antiviral drugs might dampen MS.

“The question with Epstein-Barr virus is, does it do anything to cause the disease or is it just some kind of epiphenomenon?” said J. William Lindsey, a neurologist at the University of Texas Medical School at Houston, who was not involved in the study. These new findings imply that “there are more CD8+ cells that recognize the lytic antigens in active MS, and that would go along with Epstein-Barr virus pushing the disease activity.”

EBV infects about 95% of healthy adults, but 99% of MS patients harbor the virus (see Viral Villain). Although MS patients don’t have abnormally high blood levels of the virus (Lünemann et al., 2006), they do have higher levels of anti-EBV antibodies (Almohmeed et al., 2013) and of T helper cells specific for the virus (Lünemann et al., 2006) than do healthy people infected with EBV, suggesting that the immune response to the virus goes askew in MS. Like many chronic infections, EBV initially enters the body in a lytic, or actively infectious, phase and then eventually enters a quiescent state called latency inside B cells. During reactivation, the virus awakens and begins to divide and infect new cells. Distinct sets of viral proteins are expressed during these two phases. EBV can reactivate in healthy people, sometimes without causing any symptoms, and blood tests indicate that it does not seem to reactivate more often in people with MS than in the healthy (Lindsey et al., 2009). The virus' possible activity in the brain is another story, however, and a cryptic one. Researchers have postulated that immune cells attempting to clear reactivated EBV from the brain might erroneously attack brain tissue and cause MS' signature inflammation.

To test whether immune activity against reactivated EBV might correlate with disease activity, neuroimmunologists Luca Battistini of the Fondazione Santa Lucia and Francesca Aloisi of the Istituto Superiore di Sanità, both in Rome, Italy, and their colleagues looked at CD8+ T cell responses to lytic and latent EBV proteins in the blood of MS patients who were either in active or inactive phases of RRMS. Over the course of 4 years, the researchers took blood samples from 79 untreated RRMS patients who had been seen at three different Roman hospitals, and from 43 healthy control volunteers. Their ability to analyze immune responses in fresh blood samples was key to the sensitivity of the particular assay they used to look for antigen-specific CD8+ T cells, Battistini said. The technique allowed them to analyze distinct groups of cells that target several different antigens and to quantify EBV-specific T cells as a percentage of total T cells.

Their study was the first to stratify MS patients' immune responses to EBV by disease phase, Aloisi said, and it turned out that made all the difference. When the researchers looked at all RRMS patients together, they saw no changes in T cell response to EBV compared to that in healthy volunteers. But those patients in active phases of MS as determined by magnetic resonance imaging (MRI) of the brain were more likely to have detectable levels of T cells specific to lytic EBV antigens, and they had higher levels of these cells than healthy controls. Patients in remission between relapses tended to have higher levels of T cells responding to latent EBV than relapsing patients or controls.

“Other groups … have shown that there are some differences between patients and controls in the immune response to this virus,” Aloisi said, “but have not been able to pick up this aspect, how the immune response moves in relation to the activity of the disease.”

Matthias von Herrath, an immunologist at the La Jolla Institute for Allergy & Immunology in California, who was not involved in the study, agreed. “Quantitating EBV-specific immune responses in general and associating viral presence with MS has shown association but no detailed mechanistic insight,” he wrote in an email. “The expansion of lytic-antigen specific EBV CD8+ responses during active disease is especially noteworthy.”

The researchers then asked whether viral peaks and troughs in the same patient over time correlate with disease activity. They followed two untreated RRMS patients, one for 7 months and one for 27. In both volunteers, T cell responses to lytic EBV antigens spiked in conjunction with clinical relapse and decreased during remission. One patient had very little detectable response to latent EBV, while the other's response to quiescent virus decreased when the volunteer entered the active phase of disease.

These longitudinal findings are important but need to be expanded, von Herrath said. “A larger study of this kind would not only allow for clearly defining a direct pathogenic link between EBV and MS, but would also allow the development of individual biomarkers related to the presence of EBV and the immune response, and the development and maybe severity of MS,” he wrote. Battistini and Aloisi say they are currently following six additional untreated MS patients to track their immune responses and disease activity over time, with plans to enroll six more untreated patients this year, as well as 20 treated MS patients in a new study this summer.

Battistini envisions a possible biomarker for MS disease activity that measures T cell response to lytic EBV. Such an assay could be used in addition to brain MRI, he said, although because these techniques need to be performed on fresh blood samples, not every clinic is set up for this type of immune analysis. Although only 39% of MS patients in the active phase of their disease showed responses to the two lytic antigens the researchers tested, adding more lytic proteins to the test might boost that percentage high enough to yield a biomarker of disease activity with real clinical use, Battistini said.

This type of work cannot determine whether EBV actually causes MS in the first place or just exacerbates it, because the volunteers in the study already had the disease, Aloisi said. “We have to think that by studying the response to the virus during MS, we can get information that can strengthen the association,” she said. “If all the evidence that we get goes in that direction, of course that does not allow us to answer whether EBV alone is the cause, but tells us that the immune response to the virus during MS is strongly associated with the biology of disease.”

In support of their theory that EBV reactivation in the brain is linked to MS activity, the researchers also found the lytic EBV protein BZLF-1 in postmortem tissue samples from the brains of five MS patients who died in the progressive phase of their disease and donated their bodies to research. BZLF-1, one of the lytic proteins to which relapsing MS patients showed T-cell responses, seemed to appear in inflamed regions of the MS patients' brains, and it was not present in brains from controls without MS. Aloisi and her colleagues, as well as another research group, have found evidence of EBV in postmortem brain tissue from MS patients in past studies (Tzartos et al., 2012; Serafini et al., 2013), but she acknowledges that these results are controversial, as other researchers have not been able to replicate these findings (Willis et al., 2009; Sargsyan et al., 2010; Peferoen et al., 2010; Torkildsen et al., 2010). The next step to connect EBV reactivation in the brain more directly to MS relapse would be to search for EBV-specific T cells in these brain samples, Aloisi said.

The researchers also looked at RRMS patients undergoing treatment with either interferon β (IFN-β; see interferon β-1a and interferon β-1b) or natalizumab and found that patients' response to their treatment correlated with their T-cell response to EBV. Those patients responding well to their treatment (that is, those in remission) had little to no response against lytic EBV antigens. All 14 patients on natalizumab in the study were in remission. Some on IFN-β relapsed during their treatment; these patients showed a response to both lytic and latent EBV antigens as compared to IFN-β-treated patients in remission, who had no detectable response to EBV.

“When patients are under therapy and they are controlled in a stable phase of the disease, the quality of the immune T-cell response against EBV is also supporting the idea that the infection is under control,” Battistini said. As a possible clinical application, “the response to EBV could be a good way to monitor a therapy's efficacy or nonefficacy,” Aloisi added.

While the numbers of patients in different subgroups in this study are small, the results are intriguing, Lindsey said. If the link between lytic viral activity and MS disease state plays out, that could point to antiherpes drugs as potential therapies for MS. Although antiherpes drugs for MS have met with mixed results (Lycke et al., 1996; Bech et al., 2002; Friedman et al., 2005), the right treatment might still be out there. In fact, IFN-β has known antiviral activity (Kieseier, 2011). A clinical trial testing the effect of raltegravir, an antiviral drug used for HIV treatment that may have antiherpes activity as well, was just initiated earlier this year. “It's possible that you could treat the virus, and the disease would quiet down,” Lindsey said.

Key open questions

• If EBV reactivation does trigger MS relapse, what is the mechanism? Are the increased numbers of CD8+ T cells specific for lytic EBV in MS patients somehow responsible, perhaps by erroneously attacking human myelin (Lünemann et al., 2008), or is some other aspect of the immune system responsible?
• Is EBV actually reactivating in the brain? Can this be proven? What might cause such reactivation?
• Are T cell responses to EBV prevalent enough in MS patients to serve as a true biomarker of relapse?
• If EBV reactivation causes disease activity, why haven't antiviral treatments for MS shown much success in past trials? Could a different type of antiherpes medication better treat MS than existing therapies?

Image Credit

Thumbnail image on landing page. “Tidal Bore” from the Geograph project (http://commons.wikimedia.org/wiki/Commons:Geograph_Britain_and_Ireland) collection, by Arnold Price via Wikimedia Commons. Licensed for reuse under the Creative Commons Attribution-ShareAlike 2.0 Generic license.