Biotecnologia Aplicada Elfos Scientiae ISSN: 0684-4551 Vol. 17, Num. 3, 2000, pp. 193

Biotecnología Aplicada, Volume 17, July-September 2000, p. 193

ldentification of MHC-associated Peptidesfrom Trypanosoma cruzi Infected Cells by Liquid Chromatography/Tandem Mass Spectrometry

Ron Orlando, Adam H Brockman, Rick L Tarleton

Departments of Biochemistry and Molecular Biology, and Cellular Biology University of Georgia, Athens, GA 30602 USA, E-mail: orlando@ccrc.uga.edu

Selection of papers from Biotecnología Habana`99 Congress.
November 28-December 3, 1999.

Trypanosoma cruzi is an intracellular protozoan parasite and the causative agent of Chagas. disease. This devastating disease affects an estimated 15 to 20 million persons in Latin America, and the chronic form of the infection can manifest itself as a fatal cardiomyopathy [1]. T. cruzi amastigotes reside in the cytoplasm of host cells where they release proteins that can enter the major histocompatibility complex (MHC) processing and presentation pathway [2]. MHC class I-bound peptides and the CD8+ cytotoxic T lymphocytes (CTLs) that recognize the MHC-peptide complex are critical players in cell-mediated immunity to T. cruzi [3]. For this reason, the identification of parasite peptides that are presented by the MHC on the surface of infected host cells is an important prelude to the development of vaccines.

We have previously used a variety of techniques to identify targets of the CTL response in T. cruzi, including the screening of synthetic peptides encoded by previously cloned genes for MHC binding and CTL targeting activity, and the cloning of additional genes that we suspected would be CTL targets [4, 5]. While these considerable efforts were rewarded with the identification of three CTL target molecules, the discovery process involved long and expensive investigations. We sought to circumvent the necessity of this biologically intensive empiracle approach by the use of alternative analytical methods.

Hunt and coworkers have pioneered a method of identification of CTL target peptides [6, 7]. In this method, the fractionation of MHC-associated peptides using high-pressure liquid chromatography (HPLC), the identification of fractions containing relevant target peptides using standard CTL assays, and the sequencing of the target peptides using MS/MS are combined. The approach is facilitated by the iterative use of preparative reversed-phase HPLC in conjunction with the CTL assay to lower the complexity of the peptide mixture introduced to the mass spectrometer for sequencing [6, 7]. This method works well in instances where the number of potential disease-specific peptides is low and thus only a relatively few CTL lines must be generated and maintained for the CTL assays. However, because T. cruzi exhibits a higher level of polymorphism and a lower concentration of pathogen-related MHC peptides than myeloma cell lines, the effectiveness of CTL assay is limited. In fact, we were unable to stimulate CTLs with naturally derived peptides such that a significant signal-to-noise level could be achieved. However, we were able to detect and identify peptides that effectively stimulate CTLs at concentrations achieved using synthetic materials.

We have developed an alternative strategy for identifying targets of the CTL response that use information from the pathogen's genome to avoid the necessity of generating CTL lines for peptide identification [8]. In this approach, genome sequences from infecting agents are scanned for stretches of amino acids that match a particular MHC binding motif. The identified peptide sequences are then tabulated, and their masses are calculated and compared to those of peptides isolated from pathogen-infected host cells. Peptides in the pool with masses matching those in the database can be sequenced by MS/MS to determine their identity. Using this approach we were able to confirm the processing and presentation of two T. cruzi proteins by the MHC class I pathway. The approach does not necessitate that the parent protein of the peptide be known in advance, or that genome sequence information be complete in nature. In the example presented here, the genome sequence was far from complete. Experimentally, the method does demand that genetically characterized models with a known MHC binding motif be used and that the pathogen of interest is also known. These data suggest that a rigorous automated sequencing approach employing two-dimensional separations in conjunction with MS/MS and bioinformatics is a feasible approach to identify pathogen gene products of immunological interest when CTL assay is rendered experimentally impossible. Such an approach may provide a systematic method to identify parent proteins (vaccine candidates), independent of biological assays, for a variety of infectious disease systems.

References

Copyright Elfos Scientiae 2000