The Neuroimmunology Laboratory's research interests encompass autoimmunity and tumor immunology. Patients with "paraneoplastic" disorders exhibit a remarkable overlap of autoimmunity and cancer, but cancer itself is usually inapparent when medical attention is sought. These patients have unsuspected thymic epithelial tumors or carcinomas of the lung, ovary, or breast. Their first manifestations of cancer are unusual autoimmune neurological symptoms and signs. Disabling disorders that affect brain, nerve and muscle reflect tumor immune responses limiting the growth and spread of cancer. In some patients, an attack on autonomic neurons causes profound gastrointestinal dysmotility and dysregulation of the cardiovascular system. These disparate clinical presentations result from immune responses initiated against tumor proteins that have normal counterparts in neurons and muscle. The multidisciplinary nature of our basic and clinical research interests has fostered strong longterm collaborative links with investigators in the Enteric Neuroscience Program (Joseph Szurszweski Ph.D., Michael Camilleri M.D., Gianrico Farrugia M.D.), Autonomic Disorders Program (Phillip Low M.D., Eduardo Benarroch M.D. and W. Steven Brimijoin Ph.D.), Neuromuscular Disorders Group(Steven Vernino M.D., Ph.D.and C. Michel Harper)and Multiple Sclerosis and Movement Disorder Divisions of Neurology (Claudia Lucchinetti M.D.and Brian Weinshenker M.D., and Eric Ahlskog M.D.) and Psychiatry (John Black M.D.). Although considered rare clinically, IgG autoantibodies of neuronal or muscle specificity (which are surrogate markers of T lymphocyte activation) are detectable in more than 50% of all patients with small-cell lung carcinoma (SCLC) or thymic epithelial neoplasia, but they are not detected in healthy subjects. By establishing permanent lines of tumor cells from patients who have paraneoplastic autoimmunity, we have begun to define the molecular basis for the immunogenicity of these tumors, and the mechanisms responsible for bypassing self-tolerance.
The central theme of our research program is the antigenicity of proteins that are targets of paraneoplastic autoimmunity. Synaptic plasma membrane cation channels and neurotransmitter receptors are of particular interest because they readily interact with circulating antibodies. Molecules that are used by tumors for autocrine growth, and possibly metastasis, are in many cases identical to the signaling molecules used by nerve and muscle cells in communicating with each other. The specific antigens we are studying are voltage-gated calcium and potassium channels, nicotinic acetylcholine receptors, and related molecules expressed in carcinomas of lung and ovary, and epithelial thymomas. Tumor proteins of this type initiate the helper T lymphocyte-dependent production of autoantibodies that impair synaptic transmission in the central and peripheral nervous systems. The following projects are in progress:
Nicotinic ACh receptors (AChR) expressed in neoplasms
We recently discovered that the alpha-3 subunit of the neuronal AChR responsible for synaptic transmission through autonomic ganglia is expressed exclusively in SCLC cell lines derived from patients with neurological autoimmunity. By immunizing animals with the extracellular domain of that protein, we reproduced the cardinal clinical and electrophysiological signs of autoimmune autonomic neuropathy, thus demonstrating the pertinence of this tumor antigen to a particularly devastating human neurological disorder. In addition to implications for cancer therapy, this animal model allows new treatment strategies to be developed for autoimmune gastrointestinal and cardiovascular disorders.Myasthenia gravis (MG) is a postsynaptic disorder of neuromuscular transmission that occurs in approximately 35% of patients who have an epithelial thymoma. AChR-binding autoantibodies are pathogenic in this disorder. Their production is thought to reflect immune responses initiated by muscle autoantigens expressed in immunogenic form in thymomas. We are endeavoring to develop neoplastic thymic epithelial cell lines to test this hypothesis. However, the less common association of MG with neoplasms other than thymoma appears not to be fortuitous. A SCLC line that we established from a patient with MG aberrantly expresses muscle-type AChR. In common with other SCLC tumors, our novel SCLC line has morphologic and cytogenetic markers characteristic of SCLC. It also secretes neuropeptides and has high-voltage-activated calcium channels of neuronal type. Agonist stimulation of this tumor's aberrantly expressed AChR induces an influx of Na+ that is inhibitable by curare and by alpha-bungarotoxin (alpha-BTx), which are both antagonists of muscle AChR. The alpha-BTx receptors solubilized from the tumor cosediment with authentic muscle AChR (i.e., are pentameric) by density gradient centrifugation and they are selectively precipitated by a monoclonal IgG that binds to muscle-type alpha-BTx receptors, but not by a monoclonal IgG that binds to neuronal-type alpha-BTx-receptors. Northern blot reveals mRNA encoding muscle-type AChR subunits. Other SCLC lines are negative. Sequencing of full-length cDNA clones obtained from the MG patient's tumor revealed an mRNA that was derived from the use of a cryptic RNA splice acceptor site. The protein encoded by this mRNA would be truncated, ending with 4 missense amino acids. This would yield a mutant autoantigen corresponding to the extracellular domain of the alpha-1 subunit of muscle AChR, which is a major target of pathogenic autoantibodes in MG. The non-self epitope at its C-terminus is potentially stimulatory for helper T-lymphocytes. These data support our hypothesis that paraneoplastic MG can be initiated by a tumor that expresses muscle-type AChR in a highly immunogenic form. Immune responses driven by distinct AChR subtypes expressed in cancer cells may account for the spectrum of autoimmune disorders affecting cholinergic systems that can complicate SCLC, including autoimmune autonomic neuropathies, seizures, dementia, movement disorders, and sensory and motor neuronopathies. A corollary of our hypothesis is that the immune responses responsible for impairing neurological function may also impair tumor growth and metastasis. To address these neurologic and oncologic hypotheses, we are using affinity purified channel proteins, recombinant subunit fragments, synthetic peptide antigens to produce neuronal antibodies of defined specificities. These antibodies are to be tested for effects on: a) viability and regulated ion-flux responses in cultured human neuronal cell lines; b) transmission at neuromuscular and autonomic synapses in rodents; and c) growth of cancer cells in vitro and in immunodeficient mice. We are using DNA vaccines to activate cytotoxic effector T cells.