|
||||||||
|
IMMUNOLOGY Unexpected Immune System Pathway Linked to Rheumatoid ArthritisMechanism Proposed for Joint-specific Arthritis Arising from Systemic Disease It begins in the third or fourth week with swelling in symmetrically situated joints. By nine weeks, once healthy paws curl like lotus flowers. The disease, spontaneously arising in a transgenic mouse, plays out in fast-forward the cartilage deterioration and bone attrition of rheumatoid arthritis (RA). In many details, it mimics the excesses of the human disorder at the cellular level: leukocytes invade and macrophages and fibroblasts divide, all in a potent brew of cytokines and other soluble mediators. Complement is active in the joint at the early stages of arthritis. Enhanced phagocyte and protease activities associated with articular inflammation are reported by an optical probe developed by Umar Mahmood and Ralph Weissleder. Fluorochrome tags on a polylysine backbone are quenched when in the high-density polymer configuration. Cathepsin B cleaves the backbone and releases the tag. Soft tissue is transparent to the resulting increase in infrared fluorescence at the excitation and emission wavelengths of the probe, allowing imaging in vivo. The paw of a healthy control mouse (left) and a mouse that received arthritic serum (right) were injected with the optical probe. Though clinical signs of the disease were negligible 20 hours after serum transfer, a strong fluorescent signal was apparent. Complement-deficient mice (not shown) resembled the control, consistent with the finding that these mice do not develop serum-induced arthritis. In a report published in the Feb. 20 Immunity, researchers at Joslin Diabetes Center, in collaboration with four HMS labs, pinpoint a set of factors contributing to joint destruction in the mouse model. In doing so they assign a new and unexpected role for a well-known immunological reaction, the alternative pathway of complement activation. This result, added to data accrued from five years of study, allows the researchers to draw a blueprint for the effector phase of arthritis. They map out, at least in the murine model, the steps leading to disease from its inception and provide the most detailed explanation yet of how an autoimmune disease, systemic in origin, can localize to a specific tissue. Because the alternative pathway can be specifically targeted, the work signals the way to future therapeutic agents. The Critical AlternativeThe classic pathway of complement activation has always been regarded as the primary effector for antibody action. A second complement cascade, one that is activated on microbial surfaces, is called the alternative pathway. Until now, it was thought to function primarily in the absence of antibody-antigen complexes. Christophe Benoist and Diane Mathis, HMS professors of medicine who head the Section on Immunology and Immunogenetics at Joslin, working with HMS researchers Hong Ji, Koichiro Ohmura, Isao Matsumoto, Umar Mahmood, and David Lee, discovered that these accepted roles are not necessarily followed. Using a mouse model for rheumatoid arthritis, they found that antibody-antigen complexes also could stimulate the alternative complement pathway, rather than the classic, and initiate the cascade to joint destruction.
Diane Mathis, Christophe Benoist, and Koichiro Ohmura (l to r) believe that immune complexes that accumulate on the cartilage of arthritic mice turn on the alternative pathway of complement activation and cause inflammation. Photo by Steve Gilbert. Iamge courtesy of Umar Mahmood and Ralph Weissleder. "It was no surprise that complement was involved," said Mathis. "There are lots of indications for this in human disease and also in animal models. But it was a really big surprise that it went via the alternative path." Earlier research from the Benoist- Mathis lab using the murine model demonstrated that arthritis was a result of immune complex formation between an antibody and the protein glucose-6-phosphate isomerase (GPI), which is present in all cells. This suggested that at least in some cases of arthritis, a joint-specific disease arises after an immune response against an antigen present throughout the body. In the current study, the researchers injected serum from arthritic mice into healthy mice. The latter consistently developed arthritis within days following transfer of the anti-GPI:GPI complexes. Unlike other murine models of autoimmune disorders, disease developed in a broad variety of strains. This has allowed Mathis and Benoist to utilize mutant strains lacking genes encoding specific complement molecules (in particular, strains created in the Michael Carroll and Craig Gerard labs at HMS). They found that proteins from the classic complement pathway were completely dispensable for arthritis development. However, when complement proteins along the alternative pathway were eliminated, the mice remained healthy. Additional experiments showed that a molecule involved in binding immunoglobulin molecules, called the Fc receptor, was also vital for the processes that led to joint disease. In vivo imaging indicated that both Fc receptors and complement are acting early during the disease process (see image). Active by DefaultRheumatoid arthritis has a variety of causes and clinical courses, complicating attempts to determine whether the disease arises primarily through autoimmune or inflammatory mechanisms. Regardless of the initiating event, Benoist and Mathis argue that inflammation and wasting of the joint--events that occur at a later stage--are brought about through a common set of molecular reactions. Indeed, the observation that the alternative pathway is essential for disease is no mere rule-breaking curiosity. It may hold the key to the question: how does a systemic immune reaction cause a disease localized to the joint?"The alternative pathway follows a reverse logic," said Benoist. Unlike T and B cells that detect the presence of non-self, he said, the alternative pathway recognizes the absence of self. The pathway is active unless turned off by enzymes that are present on all mammalian cells. "And that's the problem," said Mathis. "The articular surface is not cellular." Mathis and Benoist think that because cartilage lacks deactivating enzymes, the alternative pathway may be slightly more active in the joint, more so than on kidney surfaces or muscle. "Recently, Isao and David have observed that GPI is deposited as a thin film on cartilage, even in normal mice. Binding of anti-GPI may be the critical step causing the already primed pathway to spin out of control." Mathis and Benoist propose the following scenario: Anti-GPI:GPI complexes form a latticework on the cartilage surface, stabilizing the complement fragment, C3b. C3b then recruits two essential alternative pathway players, factors B and D, the initiators of the alternative pathway. According to the researchers, the decorated complexes are machines of perpetual destruction, since, once present, a variety of built-in amplification loops kick in. Critical for this process is another complement molecule, C5. One of its cleavage products attracts neutrophils, which secrete properdin, itself an amplifier of the alternative complement pathway. Elimination of any one of the factors C5, C3, factor B, or neutrophils prevents arthritis in serum transfer experiments. As Mathis explains, "Once the IgG-GPI complexes line up along the noncellular proteoglycan interface, there's nothing to block the whole autocatalytic process." Antigen SearchHow good is the murine model for arthritis? "RA is a very heterogeneous disease and probably there are different ways to arrive at the downstream phase," said Mathis. "We think this model [reactivity to an antigen stuck on the joint surface] is one generic mechanism of getting there. In some humans, it may well be that GPI is the antigen involved--and some RA patients have anti-GPI antibodies or GPI deposits in their joints--but in other individuals it could be a different antigen. We're trying to identify them."The Benoist-Mathis lab moved from Strasbourg, France to Joslin two years ago. "Coming to Harvard has been very, very beneficial because we've been able to start great collaborations with clinical and imaging experts and others," Benoist said. "Our collaboration with [co-authors and HMS faculty] Michael Carroll, Craig Gerard, Michael Brenner, and Ralph Weissleder was unusually important to the present study. The breadth of expertise and the collegial spirit at HMS have made this exciting and fun." --Anne Mahon |
