Recently, our group has developed an ultrasensitive technique to detect serum IgM to PC. We demonstrated that almost 90% of the patients had IgM-PC in serum at the onset of the disease1, being the most sensitive and standardizable technique for the early diagnosis of MS1. The presence of antibodies to lipids in serum are related with tissue damage5, and the detection of oligoclonal IgM bands to lipids in the cerebrospinal fluid is a prognosis marker of poor evolution in MS2. In addition, we observed IgM co-localizing with markers of oligodendrocyte and axonal damage4 in brain samples from MS patients. These immunoglobulins can trigger the lysis of the cells, because they induce the activation of the complement cascade3. Regarding this, IgM is pentameric and activates the complement 1000 times more efficiently than IgG13.
IgM to lipids, DNA and carbohydrates form part of the termed “natural antibodies”14, the first line of defense to pathogens. As natural antibodies recognize phylogenetically conserved structures, they raise the elimination of a broad variety of microorganism, virus, bacteria, fungi, and parasites. The structure of IgM promotes the neutralization and agglutination of the microorganisms, preventing their dissemination. These antibodies trigger the pathogen destruction, activating the complement cascade and promoting phagocytosis, which in turn boasts the adaptive immune response13.
Natural antibodies can also bind to autoantigens, such as oxidized-LDL present in atherosclerotic plaques13, and phospholipids expressed by apoptotic cells. It is supposed that IgM facilitates the clearance of the latter, avoiding inflammatory processes14. Regarding this, the most accepted hypothesis in autoimmunity claims that the exposition to a microorganism would trigger the formation of autoantibodies capable to recognize both, microbial and self-antigens15.
The presence of IgM-PC indicates the elevated activity of B1 cells, marginal zone, or peritoneal B lymphocytes. These lymphocytes are only stimulated by their specific antigen and always produce IgM.
The efficiency of therapies focused on B lymphocytes also demonstrate the relevance of these cells in the physiopathology of MS16. Treatments delay the disease progression, but have significant side effects9. In addition, the response to treatment is very variable, and 40% of the cases do not respond to treatment and a change to another therapy is required6. Thus, numerous studies aimed to characterize biomarkers of response to treatment. This is essential to make clinical decisions, when to start the treatment or change to another drug. It is also important for the future, to develop more personalized therapies and improve patients’ response.
Considering these data, we hypothesized that disease modifying therapies could modulate the production of IgM to lipids, and we aimed to study the relation between the IgM-PC levels, detected using our highly sensitive ELISA, and the response to different treatments.
We could observe that patients treated with natalizumab showed higher basal IgM-PC levels than patients receiving Copaxone or interferon-β. This treatment is indicated when first-line drugs are not efficient and patients have a high number of relapses and a rapid EDSS progression17, reflection of a high inflammatory activity18. Moreover, most patients with a high IgM-PC concentration (those in the fourth quartile) responded to natalizumab. In this line, previous results indicate that the decrease of IgM in the cerebrospinal fluid is a biomarker of response to natalizumab19. Our results draw attention to the role of serum IgM-PC level as a biomarker of response to natalizumab.
In addition, most patients treated with natalizumab showed a significant decrease in the levels of IgM-PC despite of they responded or not to the treatment. These results are consistent with those demonstrating reduced levels of both, IgG and IgM in serum20 after treatment with this humanized antibody, because 80% of the circulating IgM consists of “natural antibodies” synthetized by B1 lymphocytes14. VLA-4 mediates the homing of B lymphocytes to peripheral lymph nodes21, splenic white pulp22, mesenteric lymph nodes, and Peyer’s patches22. Therefore, the administration of natalizumab increases the number of B cells in the blood more than other cells23. Different studies reported that the subpopulations affected are pre-B23, memory and marginal zone-like cells24. Unfortunately, there are not experimental evidences describing the effect of natalizumab on B1 lymphocytes. We hypothesize that anti-VLA inhibits the activation of B1 lymphocytes25, or their homing to lymphoid organs, where they have their niches and differentiate to plasmablasts26, main effectors in MS27.
It seems contradictory that different evidences support the role of IgM to lipids as a main pathological mechanism, but patients who did not respond to natalizumab showed reduced serum levels of these antibodies after treatment. This is the reflection of the existence of two different compartments, CNS and peripheral system1, and the characteristics of patients treated with this monoclonal antibody. Natalizumab inhibits mainly the egression of CD4 + cells and B cells19 from blood to the CNS28. In addition, natalizumab reduces the intrathecal synthesis of IgG and IgM, but do not eliminate the presence of oligoclonal IgG or IgM bands completely19. We observed plasma cells and IgM deposits4 in chronic demyelinating lesions from MS patients with a long evolution. Moreover, it was detected lymphoid follicle-like structures in the meninges, associated with the damage of the nervous system29. Therefore, natalizumab can not avoid the presence of resident B and plasma cells producers of antibodies against self-antigens of the central nervous system. In this sense, depletion of B-lymphocytes in these patients does not eliminate the presence of oligoclonal bands in the CSF30.
As described above, this is a second-line treatment, administered in those cases in which first-line drugs have failed17 in patients with a poor evolution. The main mechanism of action of first-line drugs, such as Copaxone or interferons, is to inhibit the activity of T-lymphocytes31. These evidences support that in these cases immunosuppressive therapies do not regulate this population, which also plays a main role in the disease32. Moreover, it was suggested that intrinsic neurodegenerative mechanisms could be involved, especially in patients in the progressive phase33, explaining the inefficiency of immunosuppressive therapies in these individuals8.
IgM-PC.b.t levels did not predict the response to Copaxone. Interestingly, this therapy did not drop the IgM-PC levels, neither in non-responders nor in responders. These were expected results, because its main mechanism of action is to inhibit autoreactive T lymphocytes. This copolymer binds to many MHC molecules, inhibiting the response to different antigens34, and prevents the response of T lymphocytes to MBP, a major protein of the myelin. The inhibition of CD4 T lymphocytes could explain the suppression of experimental allergic encephalomyelitis by this polymer34,35. Patients treated with this drug showed increased levels of IL-10, IL-4 in serum, an anti-inflammatory profile35,36 that promotes the skewing towards TH2 responses. Regarding these, Copaxone reduces the number total number of B cells, plasmablast and memory B cells37,38. Probably, the affected subpopulation are the B2-lymphocytes, because B cells obtained from patients treated with the copolymer did not proliferate in response to CD40L39, the pathway used by T lymphocytes to activate this B2 subpopulation. However, B1 lymphocytes, the minority B subset in blood, are T independent14.
The levels of IgM-PC did not decrease in patients who did not respond to interferon-β. However, the immunoglobulin concentration dropped in patients who responded to this treatment. A decrease in these antibodies was observed in almost 80% of the responding patients. These data demonstrate that the study of IgM-PC could be a biomarker of response to treatment with interferon-β. It was described the relation between the concentration of different inflammatory molecules, such as, IL-17A40, IFN-ɣ, TNF-α41 IL-242, TRAIL41 and CXCL1340, and the response to interferon-β. Another possible biomarker of response to treatment is the quantification of neurofilament heavy and light chains, a marker of axonal damage40. Nevertheless, they are not used in daily clinical practice. On the other hand, the detection of IgM-PC as a biomarker of prognosis has large advantages, the technique is cheap and standardizable, the interpretation of results is easy, and serum samples are used. In summary, this assay could be used routinely in most clinical laboratories. Moreover, the diminution of IgM-PC concentration in patients who do not experience relapses or increased disability after treatment could demonstrate that these immunoglobulins are a major pathologic mechanism in MS. These data are in line with other previously published, since interferon-β decreases the number of memory and activated B cells43. This could be mediated by the activation of the FAS-FASL pathway, inducing apoptosis of B cells44. However, the physiology and regulation of B lymphocytes producers of antibodies to lipids, and the effect of the disease modifying therapies in this subpopulation remains unknown. Our results support that interferon-β regulates this subpopulation. This could be of great relevance to characterize new therapeutic targets and also for the basic knowledge of the functioning of the immune system. In the future, it would be of great interest to study the functioning of these cells in patients who respond and do not respond to the treatment, in order to offer more personalized therapies and obtain greater effectiveness in the disease treatment.
In summary, these data could have a significant impact at the clinical level, but further studies are necessary to validate these results in larger cohorts of patients.