What does the neuropathology of MS tell us about its pathogenesis?

Multiple sclerosis (MS) is a common, chronic central nervous system (CNS) disease characterized pathologically by inflammation, demyelination, and axonal loss. CNS pathology of MS suggests an immune-driven reaction to a CNS antigen. In addition to mononuclear inflammatory cell infiltration, the majority of active MS lesions contain antibodies, complement, and soluble immune mediators such as cytokines, chemokines, and free radicals. CNS pathology reveals not only injury to myelin but injury to axons, neurons, and oligodendrocytes. The cellular composition of MS lesions comprises primarily macrophages and T lymphocytes; both CD4 (helper) and CD8 (cytotoxic) T cells are present. To a lesser degree, B lymphocytes, plasma cells, and other types of cells such as gamma-delta T cells are also found within lesions and within the normal-appearing white matter. Polymorphonuclear cells are conspicuously absent. The neuropathology of MS suggests an autoimmune pathogenesis. However, thus far, no CNS component has been found to be the primary self-target of MS pathogenesis. If MS is an autoimmune disease, whether all people with the disease would also have the same autoantigen target is probably unlikely. Moreover, the pathology in active MS lesions is heterogeneous, suggesting either variation in the immune responses and/or the inciting events among individual patients.

Gray matter pathology

For many years, it has been known that the gray matter, especially the deep gray structures such as the thalamus, can be affected by MS. However, a surprising new finding is that the cortical gray matter is affected in MS, often to a great extent. These gray matter lesions tend to be smaller than white matter MS lesions and are more difficult to detect with standard clinical imaging techniques. Demonstration of cortical MS lesions by histology is particularly difficult, but cortical demyelination is clearly apparent using specific myelin stains. Most studies indicate that cortical MS lesions are less inflammatory than white matter lesions. A study of biopsies of white matter lesions that incidentally included cortical gray in the specimen found that 40% had cortical gray MS lesions upon closer inspection. Given that the cortical gray components of these biopsies were small and random, it is plausible that a far greater proportion of the patients would have harbored cortical pathology if more regions had been examined.

Multiple sclerosis as an autoimmune disease

Much evidence supports that MS is a disorder in which the immune response aberrantly targets CNS antigens, leading to CNS pathology. This viewpoint is based on several factors. Genetic associations almost exclusively involve immune system genes, including a strong association of MS with certain HLA class II genes. HLA class II-bearing cells process and present antigens to T cells for the CD4⁺ subtype. Thus, the manner in which HLA class II might increase risk of MS would presumably occur via its role in the processing and presentation of self-antigens to autoreactive T cells. The beneficial effect of drugs that alter the immune system, such as natalizumab and fingolimod, which affect trafficking of T lymphocytes, also supports an immune mechanism. Similarities of pathology, clinical course, and response to therapies of MS and the commonly used animal model experimental autoimmune encephalomyelitis (EAE) also support an autoimmune mechanism in MS. The EAE model is induced by immunization with any of several different myelin proteins. Despite the convergence of several lines of evidence in favor of an abnormal immune response in the pathogenesis of MS, formal proof that MS is an autoimmune disease is lacking.

Candidates for the self-antigen in multiple sclerosis

Because of the circumstantial evidence favoring an autoimmune etiology of MS, investigators have sought to identify a self-antigen that is the target of MS pathogenesis. The greatest focus has been on myelin proteins. T cells reactive to self-myelin proteins, including myelin basic protein (MBP), myelin oligodendrocyte glycoprotein (MOG), and myelin proteolipid protein (PLP), are readily found in MS patients’ blood. However, the peripheral blood of healthy controls harbors T cells reactive with the same myelin proteins, and in frequencies similar to MS patients.

Thus, differences in the properties of myelin-reactive T cells in MS versus controls have been sought. Studies using several different methodologies have found that myelin-reactive T cells have been currently or previously activated in MS compared with controls. For example, increased numbers of T cells that recognized MBP and PLP and that expressed interleukin (IL)-2 receptors, a sign of activation, were reported in peripheral blood of MS patients compared with controls. Myelin-reactive T cells in peripheral blood harbor more mutations, a sign of prior proliferation, in MS than control subjects. T cells reactive with MBP, PLP, or MOG from MS patients expressed more Kv1.3 potassium channels per cell, a marker of effector memory T cells, than did T cells from control subjects.

Antigen spreading

Although these studies support that MS patients harbor more previously activated T cells directed against myelin antigens than do controls, they do not necessarily indicate autoimmunity as the primary pathogenic mechanism. Targeting of the immune responses to self-antigens may be the consequence of antigenic spreading. Here, the concept is that an initial CNS insult results in the liberation of CNS components and subsequently leads to secondary immune responses to these self-antigens, including myelin proteins. For example, MBP is not expressed on surface of myelin sheaths and would only be accessible to immune cells upon myelin destruction. Even when responses to several distinct self-antigens are found in an individual with MS, the initiating event might still have been to a single CNS antigen.

An infection could also incite autoimmunity within the CNS. That such a phenomenon is possible has been conclusively shown using a viral animal model, Theiler’s murine encephalomyelitis. This virus-induced CNS demyelinating model is initiated by intracerebral inoculation with Theiler’s virus in mice. Initial myelin destruction due to virus and virus-specific T cells is followed by a chronic progressive phase, in which autoreactive T cells that target myelin proteins are the cause of destruction. Presumably, the latter T cells are activated due to myelin destruction in which their target proteins (self-antigens) are accessible for processing and presentation to T cells.

Epitope spreading

Epitope spreading is a phenomenon that represents a subset of antigen spreading. Conclusive evidence for epitope spreading in mammals derives from EAE and other autoimmune models of MS. For example, after EAE initiation by immunization with a small peptide component of the CNS myelin antigen, MBP, different regions of the same protein (MBP) become major targets of the ongoing autoimmune response. During the relapsing–remitting course of EAE, T cells specific not only for the initiating MBP peptide but for additional regions of the same protein (epitope spreading) can become activated and mediate relapses in the model. Presumably, a similar phenomenon could occur in MS patients. It has been speculated that one reason why early treatment seems to be most effective in control of MS disease activity is that early treatment can prevent relapses and the tissue destruction that leads to epitope and antigen spreading. More than just CNS tissue destruction is needed to induce chronic, relapsing demyelination, as evidenced by multiple well-performed studies refuting any association of head trauma or stroke with subsequent MS.

Outside-in versus inside-out?

The initial pathogenic events remain unknown. Two opposing mechanistic viewpoints exist, one stating that the initiation of the MS lesion is via an outside-in and the other upholding an inside-out mode of onset. The opposing mechanisms each have support based on neuropathology studies in MS. However, it should be kept in mind that studies of MS pathology can vary, depending upon the timing, sites, and preservation of tissue sampling; thus, interpretations may be erroneous if samples are not representative. The outside-in hypothesis suggests that the fundamental abnormality begins outside the CNS and proceeds into the CNS for lesion development. The inside-out mode implies that a fundamental abnormality exists within the CNS, behind the blood–CNS barrier, which subsequently recruits inflammatory cells from the peripheral blood. Such a paradigm is seen in the disease adrenoleukodystrophy (ALD). In ALD, a genetic mutation in the ABCD1 gene resulting in abnormal white matter causes secondary CNS inflammation and demyelination. One proponent of the inside-out version has described early loss of oligodendrocytes without concomitant lymphocyte or macrophage invasion in some acute MS lesions, interpreting these findings as support that the MS disease process began in the CNS, with secondary immune cell recruitment.

Imaging studies have been inconclusive regarding the inside-out versus outside-in theories. In careful studies using monthly MRIs, gadolinium enhancement was the very earliest evidence of new lesions in almost 100% of new T2-weighted white matter lesions. Because gadolinium enhancement indicates loss of the blood–CNS barrier integrity and moreover has been shown in limited numbers of studies to correlate closely with cellular inflammation in MS lesions, enhancement at lesion onset best supports the outside-in idea. Imaging support for outside-in is by no means airtight. Studies using several different nonstandard imaging modalities, such as MR spectroscopy and magnetization transfer, have indicated that abnormalities in CNS may be present months prior to the development of actual gadolinium enhancement and lesions seen by T2-weighted MRI. These studies lend some support to an inside-out mechanism. Current imaging techniques cannot fully answer the question of whether the MS lesion begins within the CNS itself or is initiated from the periphery.

Lymphocyte trafficking into the CNS

In either the outside-in or the inside-out mechanism, immune cells abnormally enter the CNS. Focal changes in vascular permeability including increased expression of cell adhesion molecules by the endothelium are believed to mediate the increased leukocyte trafficking into the CNS. Leukocyte migration through the blood–CNS barrier represents an important step in MS pathogenesis. This multistep process occurs in sequential interactions at postcapillary venules of different adhesion molecules and chemokines (see the following text) expressed on endothelial cells and immune cells. A key adhesion molecule implicated in leukocyte trafficking into the CNS is the α4β1 integrin (VLA-4, very late activating antigen), which is expressed on the lymphocyte surface and interacts with an adhesion molecule (vascular cell adhesion molecule-1, VCAM-1) expressed on the endothelium. The importance of this interaction is supported by the profound effect in decreasing MRI and clinical activity of natalizumab, a monoclonal antibody directed against VLA-4.

Chemokines are small cytokines that regulate migration of immune cells including leukocyte migration into the brain. When present on the surface of the brain endothelium, chemokines mediate leukocyte arrest through binding to specific chemokine receptors on the leukocytes. They also drive leukocyte transendothelial migration and locomotion within the tissue along chemoattractant gradients. Altered levels of chemokines and their receptors have been reported in MS peripheral blood cells, in MS cerebrospinal fluid (CSF), and in CNS lesions of MS. Taken together, these findings support a role of chemokines in MS pathogenesis.

Lymphocyte trafficking is also the presumed mechanistic target of the first oral medication available for MS patients, fingolimod. Fingolimod is a sphingosine-1-phosphate receptor modulator, which functionally acts as an antagonist and which leads to retention of lymphocytes within lymphoid tissue. This results in a profound decrease in circulating lymphocytes and thus the inability of the cells to gain access to the CNS.