Enzymes on microbial pathogens andTrichomonas vaginalis: molecular mimicry and functional diversity

‘Given the pragmatism of molecular evolution, multifunctionality of biological macromolecules may be more common than has been realized, and may affect rates of evolution and even medical therapies targeted at particular molecules with unanticipated secret identities.’ Quote from Piatigorsky and Wistow (1991). This Microreview will attempt to illustrate the importance of relatively new findings: those involving metabolic enzymes on the surface of numerous microbial pathogens. Seemingly remarkable at first, it is now accepted that microorganisms possess surface-associated metabolic enzymes (SAEs). The literature indicates that the SAEs may possess metabolic-enzymatic activity and also alternative functions. Although we believe that the evidence for alternative function(s) is strong, we realize that the physiological significance of SAEs has not yet been clearly established. The challenge to investigators, therefore, is the identification of the different function and elucidation of the in vivo role of the SAE. We will discuss SAEs on pathogens in terms of their functional diversity and contribution to survival of these microorganisms in complex host environments and their contribution to virulence. We have chosen to indicate that SAEs fall into the category of molecular mimicry, defined broadly, taking into account the immunological criteria as well as the more recent inclusion of adaptive and consequential mimicry, and readers are encouraged to read the excellent reviews on this subject (Oldstone, 1987; Damian, 1989; Hall, 1994). In the case of SAEs of microbial pathogens, their existence as host-like molecules may confer an immune evasion strategy (Damian, 1989) and, equally importantly, may mimic or subvert an equivalent host function. Immune (antibody) cross-reactivity with the host equivalent enzyme, such as enolase as discussed below (Fontan et al., 2000), may result in host cytopathology. A novel function for the host-like SAE would still be consistent with molecular mimicry. Precedence for functional diversity of metabolic enzymes comes from one of the most remarkable and best-studied mammalian enzymes, glyceraldehyde-3-phosphate dehydrogenase (GAPDH). Here we will briefly review the functional diversity of GAPDH to set a knowledge baseline for those interested in this field and to illustrate that our overall knowledge with respect to the functional diversity of GAPDH and other enzymes is still in its infancy. What is learned from other experimental systems, including those examining the functional diversity of GAPDH within mammalian cells, will be of importance to the field of cellular microbiology. We will discuss the data showing the surface expression of candidate adhesins that resemble enzymes by the amitochondriate Trichomonas vaginalis, enzymes originally found compartmentalized in vacuoles called hydrogenosomes. Data suggesting an alternative role in cytoadherence for these SAEs and a role for iron in upregulation of gene expression as well as compartmentalization of the SAEs will be discussed. Recent fluorescence and immunoelectron microscopy data will be presented. Investigators examining host–parasite interactions and understanding the molecular mechanisms of pathogenesis could not have predicted that microbial pathogens would have SAEs. What is exciting is the discovery of alternative functions for the SAEs, and a representative listing of microbial pathogens with SAEs and, where known, the function is presented in Table 1. These and other findings of the past decade on metabolic enzymes on microbial surfaces are consistent with the concepts of gene sharing and gene duplication with acquisition of new functions. These concepts were developed by investigators who studied the eye lens. The specialized refactory property of the lens is as a result of the production, in large concentrations, of soluble structural proteins referred to as crystallins (Wistow and Piatigorsky, 1987). It was surprising that these lens proteins with such unusual roles were not specialized structural proteins. These crystallins of the lens, a representative few listed in Table 1, turned out to be metabolic enzymes and/or stress proteins having important roles beyond those in metabolism (Wistow and Piatigorsky, 1987; Piatigorsky and Wistow, 1989, 1991; Wistow, 1993). Moreover, the crystallins were not even lens specific, as various different cell types and tissues synthesized the enzymes. These metabolic enzymes seemed to be directly recruited to new functions by modification of gene expression and not through immediate gene duplication. It was from the unique perspectives of these findings that the researchers developed the concepts of gene sharing and acquisition of new functions that precede gene duplication. Furthermore, since these findings, it has become accepted that proteins reside in compartments and/or locations other than those predicted from structure-function studies. The presence of proteins (SAEs) in unexpected locations is directly related to alternative and diverse functions for these previously characterized proteins (Smalheiser, 1996). ‘As with many things in life, what is thought to be simple and relatively straightforward turns out to be quite complex and elaborate. In this regard, a number of studies, accelerating in the past decade (Meyer-Siegler et al., 1991), have indicated that GAPDH is not an uncomplicated, simple glycolytic protein.’ Quote from the excellent review by Sirover (1999). As indicated above, while this short review focuses on the significance of metabolic enzymes with functional diversity on the surface of microbial pathogens (Table 1), it is essential to the readers to review briefly what is known about one of the most studied mammalian cell metabolic enzymes, GAPDH. There are several reasons for doing this. There are some parallels between what is being found for GAPDH on microbial surfaces with what is known for mammalian cell GAPDH, but more importantly, a great deal of insight as to possible non-enzymatic activities of SAEs on microorganisms might be obtained. The diversity of functions that are now attributed to GAPDH is remarkable, and moreover, the indications are that much more remains to be discovered about the diverse activities of this molecule. Table 2 presents a current listing of the many known functions for GAPDH (Sirover, 1997, 1999). What began as a protein involved in a sequence of substrate-product enzymatic reactions in glycolysis has evolved to a complex picture in which GAPDH plays a role in both normal cellular functions and in cell pathology. About 15 distinct functions have been experimentally ascribed to GAPDH (Table 2), and it is noteworthy that each activity has been examined at the molecular level by numerous laboratories. While the underlying mechanisms for some of the known activities have yet to be delineated, it is clear that there are sequence-specific domains within GAPDH that govern a particular function. For example, nuclear RNA and tRNA export and the properties of DNA replication and repair involve specific nucleotide sequences and GAPDH regions. Interactions between GAPDH with 3′ and 5′ end UTRs appear responsible for transcriptional and translational regulation. A role for GAPDH in neuronal disorders and apoptosis possibly involving the nitric oxide pathway as one mechanism highlights how a multifunctional protein may be involved in disease unique to a particular tissue. More recently, the idea that GAPDH is in an unknown way involved with Alzheimer's disease is reinforced by the findings of specific interactions between the enzyme and β-amyloid precursor protein (Schulze et al., 1993). GAPDH mRNA levels have been found to be greater in neoplastic tissue compared with benign prostate tissue, and it appears that both GAPDH mRNA and glycolytic activity are regulated, at least in part, by androgen exposure. This relationship will provide the impetus to examine the role of GAPDH in other cancers. From a microbial pathogenesis perspective, the interaction of GAPDH with viral RNAs or viral proteins is of particular interest. That these associations may influence viral-mediated disease outcomes is intriguing. Equally importantly, the phosphotransferase and kinase activities of GAPDH that may involve signalling pathways is now important for group A streptococcal interactions with host cells, as will be highlighted in the section below. Interestingly, while it has been suggested recently that the functional diversity of GAPDH may be related to post-translational modification, such as through mono-ADP-ribosylation, this activity is an important attribute of the group A streptococcus surface-associated GAPDH (Pancholi and Fischetti, 1993). Lastly, scientists studying SAEs of both mammalian and microbial cells are challenged by a need to understand how enzyme monomers and tetramers (in this case those of GAPDH) with identical structures are recruited to unexpected cell (either eukaryotic or prokaryotic) sites (Smalheiser, 1996). What are the signals that tell the cells exactly where subsets of identical enzyme monomers (at the nucleotide and amino acid sequence levels) are to be modified for targeting to a particular region (the surface in the case of microorganisms)? Where a monomer may have numerous functions, such as is the case for GAPDH of both mammalian cells and microbial pathogens (Table 2), what determines the particular function to be expressed? It is likely that knowledge gained from parallel studies of enzymes such as GAPDH of mammalian cells and SAEs of microorganisms will be mutually beneficial and synergistic. In summary, Table 1 lists some recent reports that describe metabolic enzymes on microbial surfaces (SAEs). Readers are encouraged to go to the original publications for details. The plasmin receptor for group A streptococci has been purified to homogeneity and identified as GAPDH by two groups (Lottenberg et al., 1992; Pancholi and Fischetti, 1993). The surface location of group A streptococcal GAPDH as well as other SAEs was confirmed on the basis of established cell fractionation and antibody immunoreactivity assays, such as immuno-electron microscopy. It was shown that GAPDH binds to fibronectin, lysozyme, myosin and actin, as well as having ADP-ribosylating and phosphotransferase activity (Pancholi and Fischetti, 1997a). The ability of GAPDH to readily associate with different substrates was demonstrated on the basis of binding of radioiodinated SAE with immobilized lysozyme, myosin and its globular domain (heavy meromyosin), and actin. Indicative of specific interactions, no similar binding was seen with the α-helical domain of myosin (light meromysin), the S-2 fragment of heavy meromyosin, and the streptococcal M protein. Moreover, immobilized streptococcal GAPDH bound fibronectin. It is not known, however, whether the association with fibronectin, as seen in a ligand-blot assay, contributes solely or in a co-operative fashion with other known fibronectin-binding proteins of this microorganism (Hanski and Caparon, 1992; Hanski et al., 1992, 1996) to host colonization. Interestingly, a surface-associated α-enolase of group A streptococci has been identified using approaches and criteria used for GAPDH (Pancholi and Fischetti, 1998). Consistent with the known plasmin-binding activity of surface α-enolase of mammalian cells, this SAE also has a strong plasmin-binding property with higher affinity compared with GAPDH. The presence of two types of plasmin-binding proteins with different affinities has been reported for group G streptococci (Ullberg et al., 1992), although the identity of the receptor is unknown. The true role of these SAEs in group A streptococcal pathogenesis awaits clarification. Immuno-electron microscopy has shown that a neuraminidase is located on the surface of Streptococcus pneumoniae (Camara et al., 1994). While a distinct function has not been ascribed to this surface protein, it is of interest that this gene contains two possible start codons, which may permit synthesis of two forms of the neuraminidase from the single gene. For one form, a putative signal sequence in the N terminus, similar to the consensus for signal peptides of Gram-positive bacteria, would allow surface sequestration. It is known that Actinomyces viscosus neuraminidase has a leader peptide (Yeung, 1993) and that neuraminidase of Bacteroides fragilis is surface exposed (Guzman et al., 1990). Activity of the surface neuraminidase of B. fragilis appears to be related to bacterial attachment to epithelial cells (Guzman et al., 1990). Streptococcus gordonii possesses a surface-associated glucosyltransferase as an et al., showing the diversity of mechanisms by pathogens for host colonization. A surface protein of gordonii also found in in was identified as a candidate to bacterial attachment to cells in This protein glucosyltransferase The of reports showing different microbial pathogens with surface-associated and distinct GAPDH the from mammalian cells on the diversity of this protein. GAPDH is also a on the surface of the et al., It is a protein for and et al., 2000), and its glycolytic activity on the surface of and in cell not also has the to which its enzymatic activity on the surface and 1999). There is strong evidence that GAPDH is an of by binding to and et al., et al., et al., 1999). The surface location was with immuno-electron microscopy using specific and both antibody and purified protein associations with and the purified surface GAPDH bound to immobilized and in a ligand-blot This binding of by GAPDH is distinct from that which through surface and 1999). Moreover, a of GAPDH for and one of this or on the cell and et al., 1992, et al., Interestingly, of expression of may result from different from those important for the other GAPDH studies using a with has in a the functions for GAPDH on group A streptococcus (Table clearly illustrate that it is that this enzyme may to and For example, it was found that group A streptococcus surface GAPDH, also called streptococcal surface dehydrogenase protein kinase as well as protein kinase of the cell (Pancholi and Fischetti, These enzymes in host cell showing the complex host this of the protein streptococcal of the The data that plays an important role in between and the host and such understanding of the of this will be to the pathogenesis of streptococcal While the in vivo of some of the diverse functions of the SAEs may be the existence of not host enzymes, some of which are surface may to to of (Oldstone, 1987; Damian, and of interest will be host antibody such as that seen for the GAPDH in with et al., 1989) will to immune with host enzymes. a appears to result from to α-enolase surface on mammalian cells et al., for antibody from microbial surface expression of this host-like enzyme (Pancholi and Fischetti, 1998). Furthermore, has a cell and α-enolase and et al., and with antibody to It is not therefore, that with streptococcal but not to streptococcal surface α-enolase (Fontan et al., The α-enolase with the α-enolase on the mammalian cell the idea of to microbial SAEs (Fontan et al., is the presence of from from et al., the true role of the SAE be established in vivo through a possible strategy for of be of interest will be the of novel or of SAE function may or and disease of SAEs now also into account possibly to the For example, targeting microbial GAPDH may to the of a cellular function essential to the well being of normal host cells and This may where related host GAPDH other with many alternative functions may be by the specific In it may be that approaches on the microorganisms with SAEs be more to a role for SAEs in virulence. the that these are enzymes may it or to even the SAEs as as has been found for GAPDH of et al., and the candidate adhesins of discussed below. of gene expression and/or function of SAEs may cellular functions that may the These studies both our of knowledge of the mechanisms of pathogenesis and of the of the Trichomonas is a responsible for the et al., the number for and 1996). This and binds to as a in and 1999). The epithelial cells but not to epithelial and cells suggesting host cell in cell targeting is indicated by our that were to binding of and and to of epithelial cells and evidence of in vivo host cell cytoadherence by is as by the signalling for et al., that within attachment to There is in the signalling as cells no similar interactions appear involved between and epithelial cells et al., et al., 1992, 1993). The surface structures on by are unknown. surface proteins to as and on and as were identified as established criteria et al., to that the proteins were candidate adhesins were et al., 1992; et al., et al., The proteins reside on the surface as by surface and There is a relationship between of surface proteins and levels of of the surface proteins by of synthesis and surface by of in the of host cells with purified adhesins attachment by in a of with different of antibody to each of the proteins cytoadherence in a of the adhesins was by binding to epithelial cells et al., 1993) and by iron et al., in with both of proteins and levels of proteins with binding of the proteins to host A identified on with the for binding to host interest was that the but not a cytoadherence by to cells and In of the on the identity and role of the group has found on the surface of as an protein for et al., and have reported on the of enzyme and et al., et al., 1998). More recently, this group has shown their experimental the proteins from to different cells in and to et al., the different cells and were not with to associations with these and other proteins. no cytoadherence were to their with levels of associations with host cells and with of the proteins. The that adhesins and may not be surprising the literature on the of both of as of a likely of their is that a or structures common to each of the cells and possibly on is being by the proteins. in the of these structures the cells and may also the of the proteins are what has been reported for cells, for et al., it is possible that the proteins are in of different surface as was recently found for the protein 1 of et al., of the mechanism this group has shown the by of et al., a role for the proteins in these possibly specific Furthermore, it is noteworthy that what was not by this group was the role of these proteins in and binding to More recent that of the adhesins studied to are each of et al., et al., and sequence at both the nucleotide and amino acid levels structural molecular of adhesins with known metabolic enzymes et al., 1996). of the adhesins have sequence identity with enzyme and the of and enzymes compartmentalized to vacuoles called are involved in of from glycolysis of the for the identified the binding domain with the ability to associations with host cells and It was noteworthy that enzymes with identity to the adhesins were of binding of the and adhesins to host cell surfaces and cytoadherence et al., et al., and It is also noteworthy that in the 3′ and 5′ end UTRs for each of the within the 3′ end of some of the were sequences that may be important for possibly expression of surface targeting of the proteins. the N of the adhesins has a short signal sequence et al., that may a role in targeting to hydrogenosomes. it is that a similar N sequence is found for other surface such as the and and that the property of for et al., 1987; et al., showing that this signal sequence may a role in targeting to As will be out the proteins in to iron are found in on the surface and in other cellular This indicates that there other protein sequence-specific or of the proteins within This may not be surprising that many of the SAEs in this review (Table have no leader sequence or As the surface location of these is a for it was important to the adhesins by fluorescence and immuno-electron microscopy of were and to with both and antibody to the which have been characterized et al., et al., 1987; et al., 1991; et al., 1992; and It is noteworthy that both and identical and showing surface expression of and and of the of Trichomonas used et al., 1987; and 1993). were in and or of was with et al., 1992; et al., 1996) and with et al., were and surface or to with This is as defined by with a et al., 1987; et al., 1987; and 1993). at of were in and to to previously with were with in were and with at or with were using with the cells were in for were for with as in the were for 1 with or antibody for of bound The were in and using were examined with an 2 in a Immuno-electron microscopy showing of proteins with in to to the For these of et al., 1992, et al., et al., 1996) were with at as in 1. were in a of and by in and in For were in and in for The were with or to for several in were in the presence of several were with and to in a of the antibody by has demonstrated the of immunoreactivity with and normal et al., 1992, et al., et al., 1999). 1 that surface expression of was for a known to expression of synthesis et al., et al., as seen for both and cells the other a greater of surface fluorescence was for the and as the surface expression of the above, was readily for consistent with an These are consistent with the and reports and the surface expression of the SAEs characterized as Interestingly, on and with in to readily in the other cell surface and the In the immunoreactivity with was and protein within Equally overall immunoreactivity for and also for past findings that synthesis and surface expression of adhesins in et al., These the surface location for the adhesins and illustrate that iron not synthesis but is also involved in compartmentalization of and importantly, the data by but not of adhesins in the various these data illustrate the in which are and the antibody data be that are from reports et al., This will to our understanding of structure-function properties for these functional proteins and of the role of iron in the compartmentalization of the SAEs within In it is not surprising to metabolic enzymes on surfaces of microbial pathogens and to have the SAEs functional diversity as within mammalian The findings of SAEs with function for microorganisms with insight into how these with in the complex host on mammalian cells or the of function that some of these enzymes is remarkable and may result from the of each of which is by the particular at the the evidence on the diversity of function for GAPDH, it may not be surprising to that enzymes function as molecular to host for surface-associated GAPDH on group A streptococci and It is that SAEs may be as by for host proteins and adhesins Furthermore, of their structural to host may to survival through molecular (Oldstone, 1987; Damian, 1989; Hall, 1994). As SAEs be examined in within the of microbial pathogenesis For vaginalis, the that each of the adhesins studied to are of presents in whether some or for the enzymatic and functions. for the gene As above, the presence of on 3′ end UTRs of not the et al., while may also permit for compartmentalization of adhesins in greater by such as that seen in The that in iron the of adhesins within the that other are involved in compartmentalization or of molecules to alternative locations other than vacuoles for metabolic function. it will be to those acid sequences for as well as non-enzymatic function. these clear the importance of antibody to the protein for about the true of SAEs in microbial pathogens. however, gene targeting and of the will be to the function and contribution of each of the proteins to host