|Home | About | Journals | Submit | Contact Us | Français|
Chlamydia species are obligate intracellular pathogens that are important causes of human genital tract, ocular, and respiratory infections. The bacteria replicate within a specialized membrane-bound compartment termed the inclusion and require host-derived lipids for intracellular growth and development. Emerging evidence indicates that Chlamydia has evolved clever strategies to fulfill its lipid needs by interacting with multiple host cell compartments and redirecting trafficking pathways to its intracellular niche. In this review, we highlight recent findings that have significantly expanded our understanding of how Chlamydia exploit lipid trafficking pathways to ensure the survival of this important human pathogen.
Chlamydiae are obligate intracellular, Gram-negative pathogens that have a significant impact on public health. Two human-adapted species, C. trachomatis and C. pneumoniae, are responsible for the majority of Chlamydia infections. C. trachomatis is the leading cause of sexually transmitted disease and non-congenital blindness (Mandell et al., 2010). C. pneumoniae is responsible for ~10% of upper and lower respiratory tract infections and has been implicated as a risk factor in chronic diseases, such as atherosclerosis (Blasi et al., 2009). One key aspect of Chlamydia pathogenesis is its ability to acquire host cell lipids and other nutrients while at the same time avoid immune detection. Understanding the molecular mechanisms of host-Chlamydia interactions may lead to new approaches for the development of novel therapeutics, diagnostics, and preventative strategies.
All Chlamydia species share a dimorphic developmental cycle in which they alternate between an extracellular, spore-like infectious elementary body (EB) and an intracellular, metabolically active but non-infectious reticulate body (RB) (Scidmore, 2011). Following binding and entry, the EB is sequestered within a unique membrane bound compartment (inclusion) where the EB differentiates into an RB and replicates by binary fission. The inclusion rapidly segregates from the endocytic pathway where it avoids fusion with lysosomes and is transported via microtubules to the peri-Golgi region. During this time, the inclusion interacts with multiple host cell compartments and trafficking pathways to acquire host-derived nutrients. Interactions between the inclusion and the host cell are likely mediated by a set of Chlamydia inclusion membrane proteins (Incs) that are inserted into the inclusion membrane by type III secretion or by Chlamydia proteins that are directly secreted into the host cytosol (Valdivia, 2008). A large number of Inc proteins are encoded in the Chlamydia genome and are thought to act as scaffolds for recruiting host cell proteins to the inclusion, where they can modulate host cell trafficking pathways, signal transduction, or the host immune response. In some strains of Chlamydia, infection of cells by more than one EB leads to the formation of multiple individual inclusions that undergo homotypic fusion to form one larger inclusion in a process that requires at least one bacterial membrane protein, IncA (Hackstadt et al., 1999, Delevoye et al., 2004) and likely other host and/or bacterial factors (Paumet et al., 2009). After 24–72 hours (hrs) of replication, RBs redifferentiate back to EBs and are released from the host cell by cell lysis or active extrusion (Hybiske et al., 2007). Under some circumstances, such as exposure to interferon-gamma, penicillin, or tryptophan starvation, Chlamydiae can enter into a reversible, altered growth state called persistence which is characterized by the presence of non-replicating, aberrant RBs and which may contribute to chronic infection (Wyrick, 2010).
Chlamydiae rely on host-derived lipids for survival and have evolved efficient methods to acquire glycerophospholipids, sphingolipids, and cholesterol from the host cell (see below). These lipids are not uniformly distributed throughout the host cell and help to define the identity and function of each unique membrane-bound organelle (Lev, 2010). The Golgi complex plays a central role in vesicular transport of lipids along the endocytic and exocytic pathways and between organelles. Non-vesicular transport also plays an important role in lipid trafficking between organelles and is thought to be facilitated by cytosolic lipid transfer proteins (LTPs) (Lev, 2010). LTPs likely act at membrane contact sites (MCS), which are sub-regions (10–20 nm) where two organelle membranes are arranged in close proximity. Recent studies have begun to shed some light on how Chlamydia actively modulates its lipid composition within bacteria and at the inclusion membrane, what lipids are present, and what function they perform. Here we review the mechanisms of lipid acquisition by Chlamydiae and highlight recent studies that reveal novel strategies by which this important pathogen exploits both vesicular and non-vesicular pathways to acquire host cell lipids.
Chlamydiae are capable of synthesizing various lipids commonly found in bacteria, such as phosphatidylethanolamine (PE), phosphatidylglycerol (PG), and phosphatidylserine (PS) (Hatch et al., 1998, Wylie et al., 1997). However, analysis of the lipid composition of C. trachomatis membranes reveals the presence of lipids typically found in eukaryotic cells, including phosphatidylcholine (PC), phosphatidylinositol (PI), sphingomyelin (SM), and cholesterol (Hackstadt et al., 1995, Hackstadt et al., 1996, Wylie et al., 1997, Hatch et al., 1998, Carabeo et al., 2003) Approximately 3.69% and 6.47% of total lipids from C. trachomatis EBs is composed of SM and cholesterol, respectively (Carabeo et al., 2003, Hatch et al., 1998, Wylie et al., 1997). Since the Chlamydia genome does not encode the enzymes required for the synthesis of these lipids (Stephens et al., 1998), the bacteria must obtain them from the host.
The inclusion is a dynamic lipid-containing compartment. Initially, the inclusion membrane is small and contains only a single bacterium, however the inclusion must expand during infection to accommodate the replicating bacteria. Defining the lipid composition of the inclusion membrane has been challenging, as it is difficult to isolate this fragile compartment in vitro. Immunofluorescence studies, however, have demonstrated the presence of some host lipids, including SM, ceramide (SM precursor), cholesterol, phosphatidic acid (PA), PI, PE, and PS in the inclusion membrane (Beatty, 2008, Hackstadt et al., 1996, Hackstadt et al., 1995, Carabeo et al., 2003, Elwell et al., 2011). Emerging evidence reveals that a number of host proteins that regulate lipid trafficking and biosynthesis also localize to the inclusion, possibly by binding to one or more Inc proteins. While the potential functions of a few Inc proteins have been revealed and are discussed throughout this review, the majority of Incs have no known function and are attractive candidates for bacterial factors that may facilitate lipid acquisition.
Seminal studies by Hackstadt and colleagues over 15 years ago paved the way to understanding one of the mechanisms by which Chlamydiae acquire SM (Hackstadt et al., 1996). These investigators found a close association between the inclusion and the Golgi complex where SM is synthesized prior to its transport to the plasma membrane. Using a fluorescently labeled ceramide analogue (C6-NBD-ceramide), they showed that this lipid is converted to SM in the Golgi and subsequently accumulates within the inclusion membrane as early as 2 hrs post infection, with up to ~50% of SM synthesized from C6-NBD-ceramide ultimately accumulating within the bacteria (Hackstadt et al., 1995, Hackstadt et al., 1996). Acquisition of SM by C. trachomatis is partially blocked when cells are treated with Brefeldin A (BFA), an inhibitor of vesicular trafficking, and this block correlates with a decrease in inclusion size (Hackstadt et al., 1996). These observations led to the hypothesis that C. trachomatis intercepts an exocytic pathway by fusing with SM-containing vesicles exiting the trans-Golgi in transit to the plasma membrane, and that SM is important for growth of the inclusion membrane. Subsequent studies demonstrate that interception of SM by this pathway is a common feature of several Chlamydia species (Wolf et al., 2001, Rockey et al., 1996). Interestingly, glucosylceramide, a lipid closely related to SM that is also synthesized in the Golgi, is not delivered to the inclusion, suggesting that the inclusion specifically interacts with a subset of exocytic vesicles (Moore et al., 2008). Further specificity for this interaction is demonstrated by the observation that SM is intercepted from a basolaterally directed pathway in polarized cells (Moore et al., 2008). In addition to SM, cholesterol appears to be co-transported to the inclusion in vesicles in a partially BFA-dependent manner where it is incorporated into the inclusion membrane as well as replicating bacteria (Carabeo et al., 2003).
It is now evident that the ability of the host cell to synthesize SM is absolutely essential for multiple steps of Chlamydia development (van Ooij et al., 2000, Robertson et al., 2009). C. trachomatis serovar L2 replication is severely impaired in cells that carry a temperature sensitive mutation in Serine Palmitoyltransferase (SPT), the first enzyme in the biosynthetic pathway for SM biosynthesis (van Ooij et al., 2000). Pharmacological inhibition of SM biosynthesis in C. trachomatis serovar E-infected cells results in loss of inclusion membrane stability, premature differentiation of RBs to EBs, early release of EBs from host cells, and disruption of homotypic fusion of multiple inclusions (Robertson et al., 2009). Inhibition of SPT also severely impairs the ability of C. trachomatis serovar B RBs to recover from a persistent state induced by interferon-gamma (Robertson et al., 2009). Taken together, these studies reveal that SM is essential not only for bacterial replication but also for both biogenesis of the inclusion membrane and reactivation from persistence.
Recent studies have yielded further mechanistic insight into BFA-sensitive vesicular trafficking of SM from the Golgi to the inclusion and reveal that only a subset of Golgi trafficking proteins are co-opted by C. trachomatis, including Arf1 GTPase (Moorhead et al., 2010, Mehlitz et al., 2010) and its cis-Golgi localized guanine exchange factor (GEF), GBF1 (Robertson et al., 2009, Elwell et al., 2011). Arf GTPases are key players in vesicular trafficking as they mediate the formation of vesicle carriers by recruiting COPI coat proteins (Bui et al., 2009). Arf1 activation in the Golgi is regulated through differential localization of specific GEFs, GBF1 and BIG1/BIG2, which are the direct targets of BFA (Manolea et al., 2008). GBF1 is restricted to the cis-Golgi and is required for the assembly and maintenance of the Golgi stack whereas BIG1/2 are found in the trans-Golgi and are required for maintenance of the trans-Golgi network (Manolea et al., 2008). Arf1 localizes to the inclusion membrane of C. trachomatis, although conflicting roles in infection have been reported (Moorhead et al., 2010, Mehlitz et al., 2010, Elwell et al., 2011). Depletion of GBF1, but not BIG1/2, reduces SM transport to the inclusion with a concomitant loss of inclusion membrane integrity and decrease in inclusion size (Elwell et al., 2011). In addition, inhibition of GBF1 results in a slight disruption of homotypic fusion (Robertson et al., 2009). These results indicate that Chlamydia acquires SM at least in part through GBF1-dependent vesicular trafficking and that this source of SM contributes to inclusion membrane stability and growth as well as fusion. Despite the reduction in SM transport following loss of GBF1 activity, there is no effect on production of infectious progeny (Elwell et al., 2011). These observations are consistent with the original report that BFA treatment does not block replication (Hackstadt et al., 1996), and unexpectedly, indicate that Chlamydia requires only one of the BFA-sensitive targets (GBF1) for SM acquisition. Since host SM biosynthesis is necessary for replication and inhibiting GBF1-dependent trafficking is not sufficient to impair replication, SM must also be acquired through GBF1-independent pathways.
SM synthesis is a multistep process (Hanada et al., 2003, Huitema et al., 2004, Hanada et al., 2009). The SM precursor, ceramide, is synthesized on the cytosolic surface of the Endoplasmic Reticulum (ER). It is subsequently transported to the trans-Golgi, where it is converted to SM by SM synthases, SMS1 or SMS2 (Huitema et al., 2004). It was previously thought that transport of ceramide from the ER to the trans-Golgi for de novo SM biosynthesis followed canonical vesicular trafficking through the Golgi stacks. However, Hanada and colleagues discovered that the major pathway for de novo SM biosynthesis requires a novel ceramide transfer protein, called CERT (Hanada et al., 2003, Hanada et al., 2009). CERT binds to the ER membrane proteins VAP-A and VAP-B by a short motif containing two phenylalanines in an acidic tract (FFAT), where it extracts newly synthesized ceramide using its steroidogenic acute regulatory protein-related lipid transfer (START) domain (Hanada et al., 2009). CERT then transfers ceramide by non-vesicular trafficking to the trans-Golgi where it binds phosphatidylinositol 4-phosphate (PI4P) through its pleckstrin homology (PH) domain and releases ceramide (Hanada et al., 2009). CERT-mediated transfer of ceramide is thought to occur at ER-Golgi MCS.
Two new studies now suggest that Chlamydia hijacks CERT to facilitate the transfer of ceramide from the ER to the inclusion membrane, possibly for subsequent conversion to SM (Derre et al., 2011, Elwell et al., 2011). As early as 2 hrs post infection, CERT, as well as VAP-A and VAP-B, localize to the inclusion membrane in discrete domains. Ultrastructural studies confirm direct recruitment of CERT and further reveal ER tubules containing VAP-B in close apposition to the inclusion, suggesting the creation of ER-inclusion MCS (Derre et al., 2011). Indeed, an intimate association of the ER with the chlamydial inclusion was previously reported (Peterson et al., 1988, Majeed et al., 1999, Giles et al., 2008).
While the PH domain of CERT is necessary and sufficient for binding to the inclusion (Derre et al., 2011), surprisingly, the regions that target CERT to the ER (FFAT) and to the Golgi (PI4P binding) are not required for CERT localization to the inclusion (Elwell et al., 2011). Previous studies demonstrated that SM acquisition by the inclusion required bacterial protein synthesis (Hackstadt et al., 1996), suggesting involvement of a chlamydial protein. Immunoprecipitation and mass spectrometry analysis of C. trachomatis-infected cells identified a bacterial inclusion membrane protein, IncD, as a binding partner for CERT (Derre et al., 2011). IncD was found to interact with the PH domain of CERT and to partially colocalize with CERT at the inclusion membrane (Derre et al., 2011). A strain that infects guinea pigs, C. caviae, lacks IncD and does not recruit CERT to the inclusion (Derre et al., 2011), thus making IncD an exciting candidate mediating CERT recruitment. Despite the lack of IncD and CERT recruitment, C. caviae acquires SM from host cells (Rockey et al., 1996), suggesting that C. caviae relies on other pathways for SM acquisition.
Immunofluorescence analysis reveal an enrichment of ceramide at the inclusion and to the adjacent region, suggesting that CERT may function to mobilize this lipid not only to the Golgi but also to the inclusion (Elwell et al., 2011). While SMS1 and SMS2 are found in close proximity to the inclusion, only SMS2 localizes at the inclusion membrane (Elwell et al., 2011). Interestingly, inclusion-localized SMS2 partially overlaps with CERT (Elwell et al., 2011). Depletion of CERT, VAPs, SMS1, and SMS2 significantly reduces the production of infectious progeny (Derre et al., 2011, Elwell et al., 2011). Furthermore, inhibition of CERT activity results in a significant defect in SM acquisition by the inclusion (Elwell et al., 2011). Based on these observations, it is possible that CERT mediates the transfer of ceramide (a) from the ER to the inclusion where it may be converted to SM by SMS2, and/or (b) from the ER to the Golgi where it is converted to SM by SMS1/SMS2 and subsequently transferred to the nearby inclusion. Consistent with either of these possibilities, diacyglycerol (DAG), a product generated during SM synthesis, accumulates at the inclusion (Tse et al., 2005). It is also possible that CERT-mediated ceramide transfer may function at ER-inclusion MCS to generate specialized metabolic and/or signaling platforms necessary for replication (Derre et al., 2011). It will be interesting to determine whether these specialized lipid platforms are similar to the recently described microdomains located at the inclusion membrane enriched in cholesterol, Src family kinases, and a subset of Incs (Mital et al., 2010).
The ability of Chlamydia to exploit both vesicular and non-vesicular trafficking pathways for SM acquisition has provided a possible explanation for a longtime paradox as to why inhibitors of vesicular trafficking have no effect on replication despite an absolute requirement for host SM biosynthesis in Chlamydia development (Elwell et al., 2011). This finding raises the possibility that vesicular and non-vesicular pathways contribute to different aspects of Chlamydia development, with GBF-dependent vesicular trafficking being essential for inclusion membrane stability and CERT-dependent transport being necessary for replication.
The ability of Chlamydia to redirect specific components of host SM biosynthetic machinery, such as CERT, VAPs, and SMS2, directly to the inclusion provides a novel strategy in which a pathogen creates its own nutrient factory. Future work will likely focus on whether Chlamydia co-opts additional LTPs to obtain other lipids, such as cholesterol and glycerophospholipids.
During C. trachomatis infection, the Golgi is fragmented into mini-stacks that surround the inclusion and enhance C. trachomatis replication (Heuer et al., 2009). Golgi fragmentation correlates with cleavage of the Golgi matrix protein, Golgin-84, which is mediated by host proteases and the Chlamydia protease, CPAF (Heuer et al., 2009, Christian et al., 2011). Inhibiting fragmentation of the Golgi by blocking Golgin-84 cleavage reduces progeny production and decreases SM uptake. Conversely, inducing Golgi fragmentation enhances progeny formation (Heuer et al., 2009). It is speculated that Golgi fragmentation enhances SM transport to the inclusion during infection. Whether the enhanced SM uptake observed following Golgi fragmentation occurs via vesicular trafficking or whether altering the Golgi structure somehow promotes non-vesicular trafficking to the inclusion is not known. One attractive hypothesis is that fragmentation of the Golgi may increase the number of ER-Golgi-inclusion MCS for efficient lipid transfer. Although SM acquisition is observed as early as two hours post infection, Golgi fragmentation occurs during later stages of infection (Hackstadt et al., 1996, Heuer et al., 2009). This finding suggests that Chlamydia may utilize different mechanisms of lipid transport at early and late times of development.
Rab GTPases are critical regulators of membrane trafficking between specific intracellular compartments (Stenmark, 2009). A subset of Rabs are recruited to the inclusion in a species-dependent and independent manner, including Rab1, Rab6, and Rab10, which are involved in ER-Golgi trafficking, and Rab4, Rab11, and Rab14, which are associated with endosomes (Rzomp et al., 2003, Capmany et al., 2010). Recruitment of Rabs to the inclusion may be mediated by specific Inc proteins. Rab4 interacts with C. trachomatis CT229 (Rzomp et al., 2003) while C. pneumonia Cpn0585 interacts with Rab1, Rab10, and Rab11 (Cortes et al., 2007). Interaction of specific Rabs with the inclusion likely allows for selective fusion with host vesicles and access to essential lipids. Thus far, Rab6, 11, and 14 have been shown to play a role specifically in SM acquisition (Rejman Lipinski et al., 2009, Capmany et al., 2010). Rab6 and Rab11 regulate C. trachomatis-induced Golgi fragmentation following Golgin-84 cleavage (Rejman Lipinski et al., 2009). Depletion of Rab6 or Rab11 impairs progeny formation, SM uptake, and Golgi fragmentation (Rejman Lipinski et al., 2009). Remarkably, progeny formation and SM uptake are rescued when Golgi fragmentation is induced by depletion of p115, another Golgi matrix protein (Rejman Lipinski et al., 2009). These results suggest that Rab6 and Rab11 may increase the efficiency of SM acquisition by mobilizing the Golgi fragments. Rab14 has also been implicated in SM acquisition by C. trachomatis (Capmany et al., 2010). Interestingly, Rab14-positive vesicles are present within the lumen of the inclusion, suggesting that the inclusion may directly fuse with Rab14 vesicles containing SM. In addition to the recruitment of Rabs, Chlamydia may regulate fusion with host vesicles for delivery of essential lipids by recruiting Golgi-associated host soluble N-ethylmaleimide-sensitive factor attachment protein receptors (SNAREs) (Moore et al., 2011, Pokrovskaya et al., 2012), which are key components of intracellular fusion machinery, and/or expression of Inc proteins that contain SNARE-like motifs (Delevoye et al., 2008).
Phosphoinositides are short-lived phosphorylated derivatives of PI that regulate the localization of PI-binding proteins and thus contribute to host cell lipid transport and signaling (D’Angelo et al., 2008). Rabs control the levels of phosphoinositides on membranes through recruitment of lipid phosphatases, such as oculocerebrorenal syndrome of Lowe protein (OCRL1), which interacts with Rab1, 6, 8, and 14 (Behnia et al., 2005). OCRL hydrolyzes PI(4,5)P2 to produce PI4P and is recruited to the inclusion along with another PI4P producing enzyme, PI4KIIα (Moorhead et al., 2010). Pools of PI4P, which are normally highly enriched at the Golgi, are present at the inclusion (Moorhead et al., 2010). Depletion of OCRL or PI4KIIα reduces production of infectious progeny and inclusion formation, suggesting that PI4P generation at the inclusion is important for infectivity (Moorhead et al., 2010). The exact role of PI4P at the inclusion is unclear. Although PI4P is not required for CERT recruitment to the inclusion (Derre et al., 2011, Elwell et al., 2011), this lipid may recruit other PI4P-binding host proteins that help define a specialized compartment that is necessary for survival and/or regulate lipid trafficking to the inclusion. Indeed, the PH domain of OSBP, a sterol transporter, which binds to PI4P at the Golgi, is recruited to the inclusion in a PI4P-dependent manner (Moorhead et al., 2010), however it is not clear whether this is necessary for Chlamydia infection. Interestingly, SM and sterol metabolism are co-regulated at the Golgi through CERT and OSBP (Lev, 2010), therefore Chlamydia may co-opt CERT and OSBP to coordinately regulate the local lipid composition at the inclusion.
PI4P production at the Golgi membrane can be modulated by Arf1 GTPases, through their ability to recruit enzymes involved in the biosynthesis of PI4P, such as PI4KIIIβ (Shin et al., 2004) Although Arf1 is not required for recruitment of PI4KIIα to the inclusion, simultaneous depletion of OCRL, PI4KIIα, and Arf1 results in a greater reduction in infectivity than single or double depletions (Moorhead et al., 2010). These results suggest that PI4P production and Arf1 play non-redundant roles in infection.
Multivesicular bodies (MVBs) are late endocytic organelles which have been proposed as another source of SM and cholesterol, possibly acting as a post-Golgi trafficking intermediate (Beatty, 2006, Beatty, 2008, Robertson et al., 2009). Several MVB constituents, including CD63, MLN64, and lysobisphosphatidic acid (LBPA) localize to the inclusion, and vesicles positive for CD63 have been observed within the inclusion (Beatty, 2006, Beatty, 2008), suggesting direct fusion of MVBs with the inclusion. Despite the presence of CD63 within the inclusion, CD63 itself is not required for the association of MVBs with the inclusion (Beatty, 2008). Treatment of infected cells with U18666A, a pharmacological inhibitor of MVB biogenesis, disrupts inclusion maturation, decreases SM and cholesterol acquisition, blocks homotypic inclusion fusion, and inhibits reactivation from a persistent infection (Beatty, 2006, Beatty, 2008, Robertson et al., 2009), suggesting that MVBs may be an important source of SM and cholesterol lipids for multiple steps in C. trachomatis development (Robertson et al., 2009). It should be noted that inhibitors of MVB biogenesis might have pleiotropic effects on other cellular functions (Cenedella, 2009). In addition, others have reported no effect of U18666A on SM uptake (Ouellette et al., 2010). While there is a clear interaction of the Chlamydia inclusion with MVBs, the precise role of this organelle in Chlamydia infection and lipid transport remains an open question. MLN64 localization at the inclusion is interesting since it contains a StAR-related lipid transfer (START) domain that binds cholesterol acquired from endocytosed LDL and may play a role in cholesterol transport to the inclusion, as Chlamydia acquires cholesterol from an LDL pathway and from de novo synthesis (Carabeo et al., 2003).
Lipid droplets (LD) have been proposed as a source of neutral lipids, such as long chain fatty acids (LCFAs), for C. trachomatis (Kumar et al., 2006, Cocchiaro et al., 2008). LDs accumulate near the periphery of the inclusion and are translocated into the lumen at IncA-rich sites (Kumar et al., 2006, Cocchiaro et al., 2008). At least 2 Chlamydia proteins, Lda1 and Lda3, localize with LDs adjacent to the inclusion (Kumar et al., 2006), and Lda3 causes disassociation of adipocyte differentiation related protein (ADRP), a host protein that inhibits lipases that degrade LDs (Cocchiaro et al., 2008). This has led to the hypothesis that Lda3 docks LDs to the inclusion and removes ADRP so that lipases can degrade LD lipids following translocation into the lumen (Cocchiaro et al., 2008). Since LCFAs are important precursors in phospholipid synthesis, it has been postulated that Chlamydia utilizes the LCFA released from LDs for membrane biosynthesis (Cocchiaro et al., 2008). Inhibition of neutral lipid biosynthesis impairs Chlamydia development, suggesting an important function of LDs in Chlamydia pathogenesis (Cocchiaro et al., 2008). Recent studies have revealed that Arf1 and GBF1 also play important roles in lipid droplet biogenesis (Soni et al., 2009, Guo et al., 2008); thus it will be interesting to determine whether these protein play dual roles in SM acquisition and LD mobilization during Chlamydia infection.
A number of host signaling molecules have been implicated in the transport of glycerophospholipids and SM to the inclusion. Host glycerophospholipids, such as PC and PI, are trafficked to the inclusion in a vesicle-independent manner (Hatch et al., 1998, Wylie et al., 1997). Unlike SM and cholesterol, glycerophospholipids are selectively modified by Chlamydia which replace the straight chain fatty acid with a bacteria-derived branched chain fatty acid (Wylie et al., 1997), in a process mediated by host Ca2+-dependent cytosolic phospholipase A2 (cPLA2) that is activated by ERK1/2 (Su et al., 2004).
A role for serine/threonine kinases in SM uptake by Chlamydia has been suggested by the observation that treatment with rottlerin, an inhibitor of several kinases, including protein kinase C delta (PKC δ), calmodulin-dependent protein kinase III, and p38-regulated/activated kinase, reduces SM transport to the inclusion (Shivshankar et al., 2008). This reduction in SM transport correlates with inhibition of chlamydial growth (Shivshankar et al., 2008). While it is unclear how rottlerin exerts its effect, one of the major targets of rottlerin, PKCδ is selectively recruited to the inclusion by binding to DAG produced at the inclusion and plays a role in protecting infected cells from apoptosis during infection (Tse et al., 2005). PKC also plays a role in regulating vesicular trafficking at the Golgi (Diaz Anel et al., 2005) and could direct SM transport to the inclusion.
A recent siRNA screen to identify host proteins required for SM acquisition by C. trachomatis has also implicated Src family kinases (SFK), specifically Fyn, in SM acquisition (Mital et al., 2011b). Activated Fyn and Src kinases localize in discrete microdomains at the inclusion that are enriched in cholesterol and overlap with a subset of Incs, including IncB, Inc101, Inc222, and Inc850 (Mital et al., 2010). Depletion of Fyn decreases C6-NBD-sphingolipid retention by both the inclusion and EBs, however, depletion of Fyn alone does not impair production of infectious progeny, indicating that SM trafficking mediated by Fyn signaling functions redundantly with other lipid trafficking pathways (Mital et al., 2011b). SFK regulate microtubule- and dynein-dependent trafficking of the inclusion to the microtubule organizing center (MTOC) (Mital et al., 2011a), therefore, it is speculated that Fyn may mediate linkage of the inclusion to the microtubule network and thereby intersect SM containing vesicles trafficking along microtubules. Whether Fyn regulates vesicle-mediated trafficking from the Golgi and/or MVBs and whether Fyn plays a role in cholesterol acquisition is not known.
Chlamydiae employ a surprising variety of strategies to acquire essential host cell lipids (Figure 1), including co-opting vesicular trafficking from the Golgi and MVBs (1), hijacking non-vesicular trafficking from Lipid droplets (2) and the ER (3), recruiting lipid modifying enzymes (4), and activating signaling transduction pathways (5). Subversion of multiple transport mechanisms helps to ensure the survival of this highly successful pathogen within the hostile intracellular environment. With the recent development of new genetic tools to study Chlamydiae (Binet et al., 2009, Wang et al., 2011, Nguyen et al., 2012, Kari et al., 2011), future studies will likely focus on identifying the bacterial factors that mediate lipid acquisition and understanding how these factors target specific pathways. Ideally, finding ways to disrupt lipid acquisition by Chlamydiae without disrupting overall host lipid synthesis may prove to be a useful strategy to combat Chlamydia infections. These studies will likely yield new insights into mechanistic details of host lipid trafficking.
We kindly thank Dr. Kathleen Mirrashidi for comments on the manuscript. We apologize to those investigators whose work was not cited due to space constraints. Work in the laboratory is supported by grants from NIH RO1-AI073770.