22 Apr 2004

Priming the immune system to recognize and clear endogenous antigens from the nervous system has always been a controversial therapeutic strategy. We only have to look to clinical trials of Elan’s Aβ vaccine to see how it can go wrong (see ARF related live discussion). In last week’s Journal of Neuroscience, Phillip Popovich and colleagues at Ohio State University reported that active or passive immunization against myelin basic protein (MBP) offers no help toward recovering from spinal cord injury. These results throw cold saline on previous data, which suggested that vaccination against MBP does have therapeutic value (Hauben et al., 2000).

Myelin-reactive T cells are found in all humans. Usually they don’t attack myelinated neurons because peripheral tolerance mechanisms keep them inactive. In patients with multiple sclerosis, activation of these T cells results in axonal damage with profound neurological consequences, so deliberately activating these cells to prevent neurodegeneration or aid recovery from neuronal trauma seems counterintuitive. Nonetheless, over the past few years, work from Michal Schwartz’s lab, at the Weizman Institute in Israel, has suggested that immunization against MBP might promote recovery from spinal cord injury. Schwartz proposed the concept of “protective autoimmunity” in the context of amyotrophic lateral sclerosis (ALS), as well (see Angelov et al., 2003, with comments).

But does autoimmunity really protect? This is what Popovich and colleagues set out to confirm, using the same protocols and models that Schwartz’s group used. Lead author Bucky Jones and colleagues first tested passive immunization, giving rats 10 million MBP-reactive T cells shortly after they suffered a moderate or severe spinal cord injury. In neither case did the immunization prove any more protective than administering saline. In fact, in animals with severe injury, the T cells actually hampered recovery, as judged by the Basso-Beattie-Brasnahan (BBB) scale for locomotor activity (scores of 5.9 versus 7.1 for rats administered T cells and saline, respectively, pNext, Jones turned to active immunization. When the authors vaccinated rats with MBP plus adjuvant, the animals recovered more slowly from injury than did those given saline plus adjuvant. For example, rats immunized with MBP had BBB scores of around 3.0 10 days after injury, while control animals had scores of around 5.0.

To correlate the effects of immunization with T cell activity, Jones and colleagues measured myelin at the site of injury, finding that both active and passive immunization was accompanied by greater loss of myelin and reduced spinal cord cross-section.

“For over 70 years, scientists have immunized animals with myelin proteins or myelin-reactive T cells to induce EAE, a model of CNS inflammation and demyelination,” write the authors. “Our data support this historical role of myelin-reactive T cells as effectors of CNS pathology.” The authors also suggest that protection previously thought due to enhanced autoimmune T cell activity may be attributable to inconsistent interpretation of the BBB scale. The authors, in consideration of recent suspended trials of vaccines for Alzheimer’s and multiple sclerosis (see Kappos et al., 2000), question the safety of “intentionally activating CNS-reactive T cells as a therapy for spinal cord injury." In fact, in the Elan example, it is now widely thought that T cells, not the antibody or other immune system components, caused the inflammatory side effect that scuttled the trial.—Tom Fagan