key: cord-0008820-2m8sgu6z
authors: Bliska, James B.
title: Crystal structure of the Yersinia tyrosine phosphatase
date: 2000-08-01
journal: Trends Microbiol
DOI: 10.1016/s0966-842x(00)88898-8
sha: a1d15f416a198b3c6460d2d365b4a4e3f017482c
doc_id: 8820
cord_uid: 2m8sgu6z

nan

against respiratory syncytial virus or the parainfluenza viruses, and genetic-engineering methods should be very valuable in developing effective vaccines against the diseases caused by these viruses. Similar approaches may lead to improved vaccines against influenza virus and measles virus. For influenza viruses, and now for rabies virus, these negative-strand RNA viruses have been shown to be able to express additional protein sequences or transcriptional units. Thus, these viruses may act as vectors to express foreign proteins, which may broaden their use as vaccines for both prophylactic and therapeutic purposes. Finally, these transfectant viruses may be useful in gene therapy. Transient expression of genes may be helpful in therapy of, for example, cystic fibrosis or cancer, and targeting cells using an RNA virus may have advantages over that using viruses with a DNA phase.

Much effort will probably go into the exploitation of these systems, and much will be learned about virus replication, virus-cell interactions and the biological properties of viruses. Furthermore, these systems promise to be important in the development of medically useful biological agents. for infection8 through its ability to suppress intracellular signaling in cells of the immune system9.

The Yersinia PTPase domain consists of an eight-stranded p sheet surrounded by seven cc helices that form a prominent substrate-binding cleft'. At the base of the cleft lies a phosphate-binding loop (the P loop) that is composed of amino acids 403-410 (CRAGVGRT). The P loop provides a framework of hydrogen bonds that initially stabilize the negatively charged thiolate of the catalytic Cys403 residue. On binding substrate, the hydrogenbonding array subsequently positions the bound anion for nucleophilic attack by the thiol group. PTPlB and the low-molecular-mass mammalian PTPase have similar active-site clefts and P-loop structures, although their topologies differ significantly outside this central core2,4. The specificity of the PTPase anion-binding and transfer reaction may be controlled by the depth of the substrate-binding cleft, which appears to be inaccessible to the shorter side chains of phosphoserine and phosphothreonine.

A substrate-induced conformational change may be an essential step in the catalysis reaction. By analyzing crystals of the protein complexed with the phosphate analog tungstate, Stuckey et al. l showed that a second loop of amino acids (residues 350-360) shifts towards the catalytic cleft, effectively trapping the bound anion (Fig. 1) . The conserved Asp356, an important catalytic residue, moves approximately 6 A into the active site. This conformational change would place the side chain of the aspartic acid in an ideal position for proton transfer during hydrolysis, assuming that a similar movement occurs when phosphotyrosine binds. In unpublished work, the same conformational change has been observed in crystals of the inactive Cys403+Ser Yersinia PTPase domain complexed with sulfate'. Thus, this movement may prove to be a general mode of PTPase action as the structures of further PTPasesubstrate complexes are resolved. Several observations showing that PTPase substrates are tightly associated with inactive Cys+Ser or Cys+Ala PTPase mutants support the occurrence of this substrateinduced conformational changelo-i2.

YopH is unique in that it is the only tyrosine-specific phosphatase that has been identified in a prokaryote. However, a distinct class of protein phosphatases with dual specificity has been found in a baculovirus and in several orthopoxviruses13. The prototype of these enzymes is the vaccinia-virus VHl protein, which hydrolyzes both phosphotyrosine and phosphoserine14. Dual-specificity phosphatases contain the conserved P-loop motif (Fig. 2 ), but otherwise have little sequence similarity with PTPases.

The dual-specificity viral phosphatases and the Yersinia PTPase all appear to act on targets within infected eukaryotic cells. It has been suggested that these virulence determinants were acquired by lateral gene transfer from eukaryotic organisms 3,9J5. However, two recent observations challenge the notion that these types of enzyme evolved initially in eukaryotes. Stuckey et al. l have found that the P-loop motif is conserved in rhodanese, a mitochondrial sulfur transferase (Fig. 2) . As it is generally accepted that mitochondria evolved from prokaryotes that were internalized by primitive eukaryotic cells, the P loop may be an evolutionarily conserved motif for anion binding and hydrolysis. Furthermore, Potts

et ~1.'~ have genetically and biochemically characterized a dualspecificity phosphatase (IphP) in the cyanobacterium Nostoc commune. As N. commune is free living, IphP probably evolved directly from prokaryotic ancestry. In addition, a potential target for IphP has been found in N. commune, a rare example of protein tyrosine phosphorylation in a prokaryote. These results raise the possibility that tyrosine phosphorylation and its associated enzyme functions arose in evolution before the divergence of prokaryotes and eukaryotes. As eukaryotic organisms evolved and began to use protein tyrosine phosphorylation as a major mechanism to activate cellular responses in the immune system, microorganisms such as Yersinia seem to have acquired new genetic traits to subvert this process.

Throughout the meeting, state-ofthe-art speakers reviewed themes in the current research on coronaand related viruses. The opening lecture was an inspiring outside view of studies aiming to define host genes involved in susceptibility and resistance to various infections. The applications of this technology to coronaviruses have so far been undeservedly limited, although a good example has been the identification 0 1995 Elsevier Science Ltd of a murine-hepatitis susceptibility gene of the fibrinogen family (Emil Skamene, Montreal General Hospital, Quebec, Canada).

The pathogenesis of coronavirus infections has been studied mainly with the murine coronavirus mouse hepatitis virus (MHV), a common mouse pathogen. Neurotropic strains of MI-IV cause demyelinating diseases of rodents that provide an animal model of human central nervous system (CNS) disorders, such as multiple sclerosis. The JHM strain of MHV has been used to identify determinants of tropism on both the virus and the target cells (Samuel Dales, University of Western Ontario, London, Ontario, Canada).

Several cellular receptors used by coronaviruses to enter target cells are now known. These include members of the carcinoembryonic antigen family for MI-IV, the aminopeptidase N for the 229E strain of human coronavirus and porcine transmissible gastroenteritis virus, and 9-0-acetylated neuraminic acid for bovine coronavirus. Binding domains on viral proteins are now starting to be identified, and it is

Proc. Nut1 Acad. Sci. USA

Pierre Talbot and Gary Levy T he 6th International Symposium on Corona-and Related Viruses discussed progress in the understanding of the molecular biology, immunology and pathogenesis of corona-, toro-and arterivirus infections. These large, enveloped animal viruses are responsible for a variety of common acute and chronic diseases in birds and mammals, including humans', and mainly cause infections of the respiratory, gastrointestinal and nervous systems2. In humans, coronaviruses cause lo-35% of common colds, have been implicated in some diarrhea1 diseases, and may be involved in multiple sclerosis, an inflammatory, autoimmune neurological disorder of multifactorial etiology3. In the veterinary field, corona-and related viruses cause economically very important losses in cattle, pigs and chickens2.Coronaviruses have the longest known RNA genome (27-31 kb), which is of positive polarity. Replication of the viral genome occurs by the production of a characteristic 3'-coterminal nested set of several subgenomic RNAs4. This replication strategy is also characteristic