Prof. Laurence C. Eisenlohr

There are five major interests of research in the laboratory.

*Processing and presentation for class I-restricted T cell recognition.
Our goals are to identify the cellular elements that participate in antigen processing and how antigen can be modified to alter the efficiency with which it is processed and, consequently the quantity/quality of the CD8+ response. These issues are being addressed with three different approaches: 1) Precise mutation of antigens known to contain class I-restricted epitopes. Our usual source of antigens is influenza virus, and our usual means of expressing these proteins is via recombinant vaccinia viruses. 2) Modification of antigen presenting cells either through treatment with chemicals (such as protease inhibitors) or mutation. 3) Over-expression of proteins suspected of participating in antigen processing. All three approaches are amenable to in vitro and in vivo assays of T cell activation although, for obvious reasons, most in vivo assays to date have been performed in conjunction with the first approach (modification of antigen). Our work has demonstrated the profound effects that sequences adjacent to and far removed from an epitope can have on processing/presentation, that conventional ubiquitination is not required for processing of large polypeptides, that proteases other than the proteasome may participate in cytosolic processing, and that subcellular location of antigen can have a profound and varying effect on the presentation of epitopes. Current efforts for the most part continue along these lines.

*Processing and loading compartments involved in presentation of MHC class II-restricted antigenic peptides.
We have been focusing upon the processing and presentation of epitopes within the two glycoproteins of influenza virus, hemagglutinin (HA) and neuraminidase (NA). HA is responsible not only for attachment of the virion to the cell surface but also for initiating fusion between viral and endosomal membranes, a critical step for successful infection of the cell. This latter function of HA depends upon its undergoing a radical change in conformation triggered by the lowering of pH that occurs within the maturing endosome. We hypothesize that this change in conformation, a mechanism used by many different viruses to trigger fusion, has a profound effect upon the processing of the antigen and can be exploited to uncover features of the processing not easily revealed with standard nominal antigens. To address this, we are focusing upon two epitopes, termed S1 and S3. Both are restricted to the same class II molecule (I-Ed), but reside in regions of HA that are differentially affected by the acid-induced change. By several different criteria, such as sensitivity to protease inhibitors, thesetwo epitopes have very distinct processing/presentation phenotypes. Current efforts are aimed at determining where within the endosomal series these epitopes become available for binding to I-Ed, where they actually bind, and the extent to which each relies upon recycling of surface class II molecules to the early endosomal compartment for its presentation. Recently we have begun to ask how the processing/presentation phenotype of these two epitopes varies with different APC.
NA is of interest because it contains at least one epitope that is not presentable by a B cell lymphoma when the protein is provided as part of UV-inactivated virions, even at very high antigen doses. In contrast, the epitope is efficiently presented when NA is biosynthesized within the APC as a result of infection, even when the multiplicity of infection is low. We hypothesize that endogenous NA reaches peptide-receptive class II molecules through a route that is not accessible to exogenous NA which may be rapidly proteolyzed upon uptake, leading to the destruction of the epitope before it reaches peptide-receptive class II molecules.

*The expression and response to 'cryptic' MHC class I-restricted epitopes.
MHC class I-restricted cytotoxic T lymphocytes (CTL) are exquisitely sensitive to low levels of antigens and we and others have demonstrated that they can be triggered by the products of aberrant gene expression, such as alternative splicing, exon translation, alternative start codon usage and frameshifting. We hypothesize that these aberrant translation products play an important role in defining the world of 'self' and may be involved in triggering autoimmunity. In a collaboration with the laboratory of Dr. Raymond Gesteland (U. of Utah) we are currently focusing on two of these mechanisms, ribosomal frameshifting and stop codon readthrough, and their roles in T cell activation. At the same time, we are determining the utility of T cell activation as a readout for such aberrations and for assessment of compounds that modulate frameshifting and readthrough, particularly in vivo.

*Dynamics of CD8+ T cell responses.
Using tools generated for our studies of MHC class I-restricted antigen processing, we are investigating the impact that antigen dose has upon the magnitude of the T cell response and the character of the resulting memory population. In order to vary expression levels of the antigen while maintaining the same level of viral infection, we have developed a system involving insertion of thermostable duplexes (hairpins) of varying length between the vaccinia promoter and the open reading frame that encodes the antigen under investigation. These duplexes serve to impede the progress of scanning ribosomes, thus limiting translation of the antigen. The longer the hairpin, the lower the level of translation of the antigen. The longer the hairpin, the lower the level of translation. With this system we can achieve a range from 200 to 60,000 copies of a particular epitope per cell surface. We have shownthat the magnitude of an in vivo T cell response is directly related to the amount of epitope expressed except at the highest (60,000 copies) level. At this highest level, the numbers of epitope-specific gamma interferon-producing, cytolytic CD8+ T cells decreases dramatically. More recent work with class I/peptide tetramers indicates the presence under these priming conditions of a large population of cells that appears to have been driven to an anergic or altered effector state. Current efforts are aimed at characterizing this population and at determining how varying epitope affects the lefspan and stability of the resulting memory pouplation.

*Cancer and autoimmunity.
For several years, we have been interested in applying what we have learnedin pursuing the areas described above to the prevention and treatment of cancer. While this goal remains a considerable challenge, the existence of various autoimmune diseases, featuring the attack of very specific tissue types, indicates that, in principle, this goal is attainable. Our accomplishments in this area include: 1) The generation and extensive characterization of a recombinant vaccinia virus that expresses the human GM-CSF gene at high levels. This virus is now being tested in a clinical trial headed by Dr. Michael J. Mastrangelo (Medicine Department, Division of Medical Oncology) involving melanoma patients. The virus is injected directly into accessible melanoma metastases, with the idea that the GM-CSF will recruit a high number of dendritic cells to the area of the tumor, enhancing the level with which tumor/tissue-specific antigens are presented. Initial results suggest that repeated injections of the virus can lead to significant tumor regression in some cases. 2) In collaboration with Dr. Edmund C. Lattime (currently Associate Director of the Cancer Institute of New Jersey), the production of cytokine-producing vaccinia recombinans designed to combat tumor in a murine bladder cancer model. Initial experiments with recombinants expressing influenza antigens demonstrated that inoculation of the bladder was easily achieved and that preimmunity to vac does not prevent expression of the recombinant protein. 3) Chemical analysis of hapten-modified autologous tumor cells. Dr. David Berd (Medicien Department, Division of Medical Oncology) has worked for several years on an alternative approach to treating melanoma. This involves removal of tumor from the patient and haptenation of the tumor with dinitrophenyl (DNP). These modified cells are irradiated and injected into the patient, elicting a vigorous immune response that can also be effective against resident, unmodified metastases. The hypothesis to explain the effect is that haptentation leads to chemical modification of MHC/peptide complexes that elicit a strong T cell response capable of partial cross-reactivity against the unmodified complexes on resident tumor cells. In collaboration with Dr. Berd and his associate Dr. Takami Sato, we have worked toward identifying the peptide(s) that might provide the basis for this cross-reactivity.

Our current interests are centered upon the fundamental and vital issues of tolerance, autoimmunity and tumor-specific immunity. Utilizing the sytems adopted or developed for the projectsdescribed above, we have begun experiments designed to determine how expression level, intracellular location, and tissue tropism influence the establishment and vulnerability of tolerance with respect to viral infection and tumor cell challenge. The tumor models that we have begun working in are thyroid and colon cancer. For the thyroid model, we are collaborating with Dr. Jay Rothstein (Department of Otolaryngology-HNS) and for colon cancer we are collaborating with Dr. Scott Waldman (Director, Division of Pharmacology, Medicine Department). Interestingly, co-incidence of autoimmunity and neoplasia has been noted in human populations for both types of cancer.

We anticipate that progress in these five areas will contribute important information to the areas of vaccine design, tissue transplantation, autoimmune therapy, and immune-based anti-cancer strategies.