The role of the chlamydial effector CPAF in the induction of genomic instability Free

The role of CPAF in chlamydial virulence is justifiably a tangle of confusion. CPAF is a potent and promiscuous cysteine protease capable of cleaving many proteins (Zhong, 2011). Over the 13 years since the original characterization of CPAF in the laboratory of Guaming Zohng (Zhong et al., 2001) chlamydial researchers have contributed a long list of identified CPAF cleavable target proteins of both chlamydial and host origin (A Conrad et al., 2013). However, careful studies by the Tan and Sutterlin group convincingly demonstrated that the majority of CPAF activity in infected cells is actually sequestered in the chlamydial inclusion and that biochemical assays relying on cell homogenization in many cases released the sequestered CPAF allowing degradation of ‘targets’ that would not interact with CPAF under biological conditions (Chen et al., 2012). Thus, the question becomes not what host proteins does CPAF cleave but what host proteins does CPAF have access to and when does it gain this access.

New data from the investigation of growth and virulence phenotypes of two mutant strains of chlamydia null for CPAF activity isolated by Emily Snavely in the Valdivia laboratory have added critical observations that have evolved our understanding of the role of CPAF in pathogenesis (Snavely et al., 2014). The Valdivia laboratory observed that CPAF null strains follow the same developmental cycle and produce normal appearing inclusions when compared to isogenic CPAF positive chlamydial strains (Snavely et al., 2014). The CPAF mutants, however, produce approximately threefold less infectious progeny. It is not yet clear at what stage in development this defect lies (Snavely et al., 2014).

We had previously demonstrated that Chlamydia trachomatis infection causes genomic instability by inducing the over duplication of centrosomes, causing premature mitotic exit and restricting reorganization of the centrosome/microtubule network (Knowlton et al., 2011; Brown et al., 2012). These phenotypes contribute to chromosome segregation errors leading to cytokinesis failure and multinucleation (Brown et al., 2012). We demonstrated biochemically that both cyclin B1 and securin (two proteins involved in mitotic check point control) could be degraded by CPAF and theorized CPAF to be involved in the induction of genomic instability (Brown et al., 2012). However, the published results from the Tan laboratory demonstrated that the majority of detected cyclin B1 cleavage was due to inclusion lysis during sample handling (Chen et al., 2012). We verified this result, and additionally found that securin degradation was also only observed during sample handling.

With the availability of the isogenic CPAF mutant isolates generously shared by the Valdivia laboratory, we directly tested the role of CPAF in genomic instability. Our studies published in the June edition of PlosOne showed that CPAF null strains were deficient in inducing centrosome amplification and early mitotic exit, two critical phenotypes involved in causing genomic instability (Brown et al., 2014). As a result, these strains caused dramatically less multinucleation (Brown et al., 2014). By investigating genomic instability in the GspE chlamydial mutant [isolated and described by the Valdivia laboratory (Nguyen & Valdivia, 2012)] we further demonstrated that secretion through the type II secretion system is required for the CPAF dependent phenotypes of centrosome amplification and early mitotic exit (Brown et al., 2014).

Although we have demonstrated a clear phenotype dependent on CPAF production and secretion from chlamydia, critical questions remain. We have not yet identified the relevant targets of CPAF that control centrosome over duplication and early mitotic exit. At this stage, it is difficult to predict how CPAF could target these cellular control points so precisely. One possible explanation could be the effect of the relative concentrations of CPAF and its targets. Although the details involved in regulation of centrosome duplication are not fully elucidated, it is clear that the licensing scheme controlling duplication is regulated by just a few proteins that tether the mother and daughter centrioles (Brownlee & Rogers, 2013). The fact that cells have only one or two centrosomes would result in extremely low levels of these control proteins. At these concentrations, it would not take much CPAF activity to impact this regulatory system. Mitotic exit is similarly controlled by a small number of proteins that physically tether the sister chromosomes (Uhlmann et al., 1999). While there are more chromosomes then centrosomes, the concentrations of these critical regulatory proteins are still very small on a per cell basis.

Although small quantities of CPAF are potentially all, that is, required to effect these regulatory pathways, the mechanism by which CPAF is secreted from the chlamydial inclusion is unclear. The inclusion membrane acts as a barrier between the bacteria and host cytosol and effectively hides the bacteria from host innate immune surveillance (Scidmore et al., 2003). Intriguingly, this barrier also acts to inhibit access to nutrients in the cell cytosol and restricts the delivery of effector proteins. The balance between these two opposing properties likely reflects important trade offs between access and protection.

There is significant evidence that although the inclusion membrane restricts diffusion between the host cytosol and the intrainclusion space, the two compartments maintain some level of communication. Multiple proteins secreted through the classical type II secretion system (T2SS) into the intrainclusion space are found in the host cytosol. The T2SS is unique in that it allows for the secretion of folded proteins from the periplasm across the outer membrane (Connie Lu et al., 2013b). In Chlamydia, this results in fully folded proteins secreted into the intrainclusion space. In order for these proteins to reach the host cytosol, a second secretion step must facilitate transport across the inclusion membrane. In addition to CPAF other classical Sec secretion signal containing proteins Tsp, cHtrA, CT795, CT311 and GlgA have all been reported to be present in the cytosol using immunofluorescent microscopy or functional assays (Lei et al., 2011; Qi et al., 2011; Zhong, 2011; Lu et al., 2013a). Currently, this system is completely uncharacterised and likely has no evolutionary equivalent in the more commonly studied model organisms.

Although these recent studies have helped to focus our understanding the role CPAF plays in chlamydial biology, the picture is far from complete, critical questions remain. The mechanism and timing of CPAF secretion is a critical gap in our understanding. Our recent data show that there are CPAF dependent phenotypes induced in infected cells starting as early as 18–24 h postinfection. Additionally, the Valdivia laboratory has definitively identified late targets of CPAF suggesting CPAFs role may change over the developmental cycle. Going forward, the key goals toward understanding the role CPAF plays in chlamydial biology is to determine how and when CPAF gains access to the cell cytosol and to identify the targets that produce genuine phenotypes impacting chlamydial growth and development or that impact the cell resulting in disease.

References