Over the past decade, progress has been made in understanding the molecular mechanisms by which effector proteins mediate their function. Amongst the findings, several lines of evidence suggest that certain effectors have evolved in such a way that they harbor distinct functional domains on either termini of the protein. As an example, the SPI-1 effector protein SipA has been reported to contain two functional domains: the C-terminal domain has been reported to be involved in actin binding (
Mitra et al., 2000), whereas the N-terminal domain has been demonstrated to be essential for inducing transcellular signals that guide PMN transepithelial migration () (
Wall et al., 2007).
SifA is another effector protein that has been reported to have a dual function. This effector plays a critical role in SCV formation and maintenance, especially regarding Sif formation. The C-termini domain of SifA contains a WxxxE motif, a protein motif present in bacterial effectors that directly mimic activated GTPases (
Alto et al., 2006; Ohlson
et al., 2008) while the N-termini binds to SKIP, a cellular protein that binds to the motor protein kinesin and is required for SCV tubulation () (Ohlson
et al., 2008). Therefore, SifA has both a GTPase activity located at its C-termini, as well as a motor function (by binding to SKIP (kinesin)) located at its N-termini. Furthermore, SptP, also functions both as a tyrosine phosphatase and a GTPase activating protein (GAP) (
Fu and Galan, 1999;
Humphreys et al. 2009; Malik-Kale
et al., 2010).
Two interesting hypotheses have recently been postulated regarding the evolution of the dual functionality exhibited by effector proteins: i) “terminal reaasortment” hypothesis postulated by the Guttman lab, and ii) the “Caspase-3 processing” hypothesis postulated and discovered in our laboratory (
Stavrinides et al., 2006;
Srikanth et al., 2010).
Terminal Reassortment in the evolution of effector proteins
One of the most prominent features amongst T3SS effectors is their modular nature wherein these proteins consist of well-defined regions (called domains) that confer a specific function. These domains (within the same effector) usually mediate different and unrelated functions suggesting that these domains evolved independent of each other and combined to form a chimeric effector (
Stavrinides et al., 2006). This process, termed “terminal reassortment”, is a common theme not only among several
Salmonella effectors but also other bacteria effectors as well.
A conventional example is the case of the
Salmonella effector, SptP whose N-terminal domain has sequence homology with ExoS and YopE proteins of
Pseudomonas and
Yersinia spp., respectively, while its C-terminal is highly homologous to another
Yersina protein, YopH (
Kaniga et al., 1996; Dean, 2011). Described as design-by-recombination, “terminal reassortment” explains the high diversity amongst bacterial effectors and together with horizontal gene transfer and genomic shuffling, it results in the fusion of these functional effector modules with the C-termini of the different effectors (Dean, 2011). This establishes the newly formed chimeric protein with a new C-termini and the original N-terminal domain, giving rise to a single step evolution mechanism that yields new effector proteins (Dean, 2011; Stavrinides
et al., 2011). Four
Salmonella effectors SifA, SspH1, SseI and SseJ all share a homologous N-terminal domain but different C-terminal regions, further substantiating the evidence and significance of the terminal reassortement hypothesis (
Hansen-Wester et al., 2002, Dean
et al., 2011). Strengthening this hypothesis is the fact that ~32% of T3SS effector families in bacterial species are chimeric proteins, which is much greater than in any other protein family analyzed to date, and highlights the importance of this process in the evolution of bacteria virulence factors (
Stavrinides et al., 2006; Dean, 2011). Thus, it is likely that terminal reassortment plays a role in effector diversity by the addition of a new function to an already existing effector protein.
Caspase-3 processing of effector proteins
As described above, a critical feature of Salmonella effectors is the presence of functional modules that comprise domains or motifs that confer an array of functions within the eukaryotic cell. These domains/motifs represent a fascinating repertoire of molecular determinants with important roles during infection (Dean, 2011). Here we will discuss some of the recent findings on the understanding of some Salmonella effector motifs and their role in infection, which may be linked to the terminal reassortment hypothesis.
In examining the bi-functional properties of SipA, we discovered that SipA contains a caspase-3 recognition and cleavage site (DEVD) at amino acid position 431, which is precisely located at the junction between the SipAa N-terminal domain and the SipAb C-terminal domain (
Srikanth et al., 2010). This motif is physiologically significant, as a single amino acid substitution to a sequence not recognized by caspase-3 profoundly attenuates the virulence of this pathogen in both
in vitro and
in vivo models of salmonellosis (
Srikanth et al., 2010). Conversely, knocking out the caspase-3 gene in mice resulted in a significantly less virulent
Salmonella infection, and dramatically reduced the severity of intestinal inflammation (
Srikanth et al., 2010). Notably, SipA, itself, was found to be necessary and sufficient for early caspase-3 activation, but in a process independent from the apoptotic cascade. Further analysis of the
S. Typhimurium T3SS revealed the presence of caspase-3 cleavage motifs in other secreted effectors with known bi-functional properties (i.e., SopA, SifA) (
Srikanth et al., 2010), indicating this phenomenon is not isolated to SipA. Remarkably, no caspase-3 cleavage motifs were identified in the structural proteins, chaperones, or transcriptional regulators that together comprise the T3SS.
The identification of caspase-3 cleavage motifs in secreted effectors introduces a novel concept that certain effector proteins of S. Typhimurium exist in a pro-form that requires processing to become functionally active (or perhaps inactive). Since caspase-3 cleavage motifs detected in SPI-1 and SPI-2 T3SS are restricted to secreted effectors, it is tempting to speculate that caspase-3 cleavage maybe a general subversion strategy employed by Salmonella for processing of its secreted effectors. Furthermore, while bacteria have previously been described to interact with caspase-3, our study is the first to document that a pathogenic organism is able to sabotage a major host death pathway in the absence of apoptosis, thus directly exploiting the host enzyme, caspase-3, to facilitate infection and pathogenesis.
Based on these new concepts, we hypothesize that terminal reassortment precedes caspase-3 processing in the generation of new bi-functional effectors. First, through terminal reassortment, a new functional motif is added to an “existing” effector protein to form a chimeric protein (as may have been the case with SipA domains described above). Second, upon infection, the bacteria utilize the host cell protein, caspase-3 to process the chimeric effector yielding the two functional motifs, which may play an important role in bacterial pathogenesis.
WEK(I/M)xxFF motif
Another important motif identified in
Salmonella secreted effectors is the WEK(I/M)xxFF motif. This motif is located on the N-termini of SifA, SopD2, SseJ and SspH2 and is required for Golgi targeting of these effectors (
Brown et al., 2006). Other functional motifs have also been identified in other effector proteins and their putative functions have been suggested (Dean, 2011).