Naturally killing rheumatoid arthritis

To get this semester's blogging off and running (no one wants to be the first to post on an empty blog!), here is my take on a new PNAS paper from the Cantor lab:

            Rheumatoid arthritis (RA) is an autoimmune inflammatory disease that affects around 1.3 million people in the United States. Found more commonly in women than men, this disease causes pain and swelling in the joints, and can lead to cartilage and bone loss, and joint deformity. RA is commonly treated with methotrexate, in combination with other drugs, but treatments can be expensive, with lifetime direct costs of up to $180,000 per patient (in 2010 dollars). These treatments can also have adverse side effects. The prevalence of RA in the U.S. is increasing as the population ages, suggesting a need for continued research into the causes of, and potential new therapeutics for, RA.

            At its core, RA is a disease of multiple immune cell types. One component is B cells, which produce antibodies that recognize “self” proteins found in connective tissue. A second component is “helper” T cells (also termed CD4 T cells), which help B cells produce antibody, and which contribute to an inflammatory state in the joints. A study published this week in the Proceedings of the National Academy of Sciences by Leavenworth and colleagues further refines the contribution of different CD4 T cell subsets in RA, and defines a new role for yet a third cell type, natural killer (NK) cells, in suppressing the development of RA. To do this, the authors used an experimental mouse model of RA, induced by injection of type II collagen, a prominent component of connective tissue, under the skin. This model (collagen-induced arthritis; CIA) causes many of the same symptoms as human RA, and has been used to test a number of new treatments for RA (1).

            To begin, the authors evaluated the capacity of different subtypes of CD4 T cells to induce CIA. Several different types of helper T cells have been identified (T_H1, T_H2, T_H17, among others), and contribute to different types of immune responses. Leavenworth et al. isolated different CD4 T cell subsets from mice that had been previously injected with collagen, and transferred them, along with B cells, into mice that couldn’t produce these cell types on their own. They then injected these “recipient” mice with collagen, and evaluated them for the production of antibodies that recognize collagen, and for the development of CIA. As had been previously shown, transfer of the follicular T helper cell subset (T_FH) caused the production of anti-collagen antibodies by the B cells (2). Interestingly, only when T_FH cells were co-transferred with another helper T cell subset, T_H17 cells, did full-blown CIA result. These experiments indicate that both cell types are important for CIA: the T_FH cells for the production of autoantibodies, and the T_H17 cells for the triggering of the inflammatory response.

            Leavenworth and colleagues then evaluated the role of NK cells in this process. NK cells are important for eliminating virus-infected cells, and are regulated by two classes of proteins: activating receptors and inhibitory receptors. Normal cells express a protein, Qa-1, that binds to one of the NK cell inhibitory receptors, NKG2A, and prevents NK cell killing. Many virally-infected cells stop expressing Qa-1, allowing NK cells to lyse them to prevent virus replication. NK cells have also been implicated in suppressing some autoimmune diseases, by killing self-reactive T cells. To test whether NK cells could suppress CIA, the authors evaluated the ability of purified NK cells to lyse helper T cell subsets. NK cells preferentially killed T_H17 and T_FH cells, in comparison to T_H1 and T_H2 subsets. Additionally, the symptoms of CIA were much worse in mice that lacked NK cells, suggesting that these cells have a role in preventing CIA. Most intriguingly, treatment of mice with antibodies that block the interaction between Qa-1 and the NKG2A significantly decreased the production of anti-collagen antibodies and markedly reduced the severity of CIA. These experiments suggest that this strategy might be developed as another potential treatment for RA.

            This work elegantly illustrates the intricacy of interactions between many different immune cell types in the development, and prevention, of autoimmune diseases. Additionally, as antibody-based therapies have been developed for many different diseases including cancers (rituximab), cardiovascular disease (abciximab), and other autoimmune diseases (infliximab), it is conceivable that anti-NKG2A antibody-based therapies for RA could move quickly to clinical trials.  

Reference: Leavenworth, J.W., Wang, X., Schellack Wenander, C., Spee, P., and Cantor, H. 2011. Mobilization of natural killer cells inhibits development of collagen-induced arthritis. Proceedings of the National Academy of Sciences, U.S.A.. 108: 14584-14589. 

Other citations:

(1)  Asquith, D.L., Miller, A.M., McInnes, I.B., and Liew, F.Y. 2009. Animal models of rheumatoid arthritis. European Journal of Immunology. 39: 2040-2044. 

(2) Kim, H.J., et al. 2011. CD8+ T regulatory cells express the Ly49 Class I MHC receptor and are defective in autoimmune prone B6-Yaa mice. Proc. Natl Acad Sci USA. 107: 2562-2567.