OUP user menu

The intracellular life of Chlamydia psittaci: how do the bacteria interact with the host cell?

Cristina Escalante-Ochoa , Richard Ducatelle , Freddy Haesebrouck
DOI: http://dx.doi.org/10.1111/j.1574-6976.1998.tb00361.x 65-78 First published online: 1 June 1998

Abstract

Throughout the life of any organism interactions with the surrounding environment are always taking place, a process that leads to evolution. Chlamydia psittaci is an obligate intracellular parasite, but it must also be capable of extracellular survival in order to search for new host cells. Therefore, these peculiar prokaryotes have evolved two different particles and a unique developmental cycle that, together with a series of not yet fully understood interactions with their host cells, allow them to fulfil the requirements for their permanence in nature. These interactions are the subject of this paper. Particular attention is paid to the attachment and internalization of the bacteria, the chlamydial vacuole, and the avoidance of lysosomal degradation.

Keywords
  • Chlamydia–host interaction
  • Intracellular permanence
  • Chlamydial vacuole
  • Chlamydia internalization
  • Vacuole membrane
  • Phagolysosome fusion inhibition

1 Introduction

Obligate intracellular parasites like chlamydiae live in what might be called an extreme environment [61], where they have evolved unique fitness traits and become dependent on the living cell. By being inside the host cells, these parasites escape most of the defence mechanisms of the host. Nevertheless, they must adapt to the intracellular environment and establish specific interactions with the host cells that allow them to survive and multiply and so to perpetuate their species in nature.

Chlamydia psittaci, one of the four species described up to now for the genus Chlamydia, is comprised of a heterogeneous group of microorganisms that cause a variety of diseases in animals and can be transmitted to man [59, 71, 72, 85, 96]. Chlamydiae are considered energy parasites since they depend on the eukaryotic ATP for their replication [61]. These prokaryotes have a particular biphasic developmental cycle which involves predominantly two alternating chlamydial forms: the elementary body (EB) and the reticulate body (RB). Infection of susceptible cells is mediated by the EB, a particle of 0.2–0.3 μm in diameter, whereas bacterial multiplication within the cells is carried out by the RB, a particle of 0.5–1.3 μm in diameter [7, 71, 75, 115]. The chlamydial developmental cycle, first described by Higashi [40], may vary according to the chlamydial strain and the host cell involved [40, 9698, 109]. Briefly, the EB attaches to the cytoplasmic membrane of a susceptible host cell inducing its own internalization within a vacuole (inclusion) avoiding the phagosome-lysosome fusion. Once inside the cell, the EB converts itself into the RB, starting then an active process of binary fission. After several divisions RBs start to reorganize into EBs. All this happens inside the vacuole. Eventually, the chlamydial infection may exert a cytocidal effect leading to the release of a new progeny of infectious particles. In some cases, however, the prokaryotes might be extruded from the cell without cellular lysis, or even a persistent infection may take place [20, 22, 47, 62, 98, 105, 106].

Several components of the chlamydial cell wall (for a review see [75]) are known to play a role in the survival of chlamydiae and in the interactions with the host cell (Table 1). Other components are also thought to be important for the interaction with the host cell. For example, early in the developmental cycle, bacterial heat shock proteins of the 60-, 70- and 110-kDa families are synthesized [49, 52, 60, 83, 116]. This could suggest that chlamydiae begin to modify the normal activity of the host cells soon after entry, ensuring themselves a less stressful environment. Chlamydiae are also capable of affecting RNA and DNA [35, 58]. Interestingly, RNA and some protein synthesis appears to be developmental cycle stage specific [18, 52, 73]. It has been observed that when cells are infected with C. psittaci, the rates of synthesis of protein, RNA and DNA in the infected cells never exceed the rates observed with non-infected cells [1]. Therefore, the chlamydiae must somehow inhibit multiplication of the host cell and then exploit the idle synthetic capacity of the cells (biomachinery) for their own macromolecular synthesis.

View this table:
1

Chlamydial cell wall components that interact with the host cell

Cell wall componentFunctionEffectReferences
Lipopolysaccharide (LPS)Stimulate phagocyte cellsDisturbance of immune recognition and immune-mediated cytolysis(?)[75]
Reduction of eukaryotic membrane fluidity
RB projections‘Piercing’ of the chlamydial inclusion membraneFacilitate acquisition of nutrients[51, 54, 93]
Cysteine-rich proteins (CRP):
 (a) Major outer membrane (MOMP)PorinPassive diffusion of hydrophylic compounds into the chlamydial cell[6]
AdhesinAttachment and entry[84, 99102]
 (b) Small CRP or EnvAMurein lipoproteinFunctional equivalent of peptidoglycan(?)[27]
 (c) Large CRP doublet or EnvBMajor structural envelope proteinFunctional equivalent of peptidoglycan(?)[26, 24, 98]

The relationship that chlamydiae establish with host cells is an intimate association that has evolved between two very different forms of life. Once inside the cell, Chlamydia do little damage initially and may even go unnoticed as multiplication progresses. They change their intracellular form of existence and finally they may exert a cytocidal effect on the infected host cell or, under other conditions, persist for extended periods within the surviving host cell [98].

Currently, curative and prophylactic measures for the control of chlamydial infections often yield disappointing results. Understanding the mechanisms that chlamydiae have evolved for their survival within vertebrate host cells may be essential for the future development of tools to control these infections. The purpose of this paper is to review the currently available knowledge on the interaction of C. psittaci with the host cell that allows their intracellular survival.

2 Attachment and internalization of C. psittaci

The initial interaction of C. psittaci with the host cell begins with the attachment of the EB to the cell. There has been considerable controversy about the mechanism underlying this attachment. In L cells attachment has been suggested to involve a direct interaction between trypsin-labile sites on the host and heat-sensitive sites on the parasite [11].

Byrne [12] demonstrated that the rates of attachment and ingestion of C. psittaci by the host cell are determined by the multiplicity of infection, being higher when low infectious doses are used. The phagocytosis of chlamydiae may itself be injurious to host cells, so that the ingestion of a critical number of chlamydial cells may alter the rates at which additional parasites are attached and ingested. C. psittaci is negatively charged, probably by means of carboxyl groups, and it is hydrophobic [108]. One function of the negative charge might be to prevent the bacteria from forming large aggregates that cannot be phagocytosed by host cells. Moreover, it is known that chlamydial attachment depends on the physical integrity of the plasma membrane and requires homeostasis of intracellular Ca2+[12, 53]. Attachment of C. psittaci strains to eukaryotic cells is significantly increased by the divalent cations Ca2+ and Mg2+ which may decrease electrostatic repulsion [37]. In addition, hydrophobic forces contribute to the ease at which C. psittaci is phagocytosed by L cells.

Other studies have highlighted a mechanism of molecular complementarity as the means of chlamydial attachment. Byrne and Moulder [13] showed that chlamydiae are ingested by non-professional phagocytes at faster rates than those observed for the phagocytosis of Escherichia coli and polystyrene latex spheres. The uptake of chlamydial organisms by non-professional phagocytes was named ‘parasite-specified phagocytosis’ in order to differentiate it from host-specified phagocytosis, and to suggest that specific structures on the bacterium play a role in the process. Several research teams have obtained results that support this theory.

Ultrastructural studies of C. psittaci infection of intestinal epithelial cells [22], L cells [41, 115] and BGM cells [109], and of C. trachomatis in McCoy cells [42] have shown that the bacteria preferentially bind to cell surface microvilli. This might be advantageous for an obligate intracellular parasite such as Chlamydia since the membrane regions at the bases of the microvilli are areas of active transport of extracellular materials into the cells, so attachment to these sites might assist a rapid and efficient entry [41]. The attachment of both C. psittaci and C. trachomatis to pre-existing cell surface microvilli suggests a common structure of adherence. There is, nevertheless, poor efficiency of binding of C. trachomatis at both 37°C and 4°C compared with that of C. psittaci under the same conditions, which may indicate differences in the initial reaction [41, 42].

Glycoproteins on the surfaces of both bacteria and host cell have been found to be involved in the binding of Chlamydia to the host cells [85, 100, 101, 110]. Furthermore, ultrastructural studies have revealed that coated pits, coated vesicles and endosomes may play a role in C. psittaci entry [41, 42, 115]. Therefore, the initial interaction between chlamydiae and host cells should theoretically involve a receptor-ligand binding, which may be of the lectin type. During experiments concerning the influence that carbohydrates exert on the adhesion and penetration of C. psittaci into L cells [94], it was observed that addition of certain sugars reduces chlamydial infectivity. Inhibition was greater with monosaccharides than with disaccharides and trisaccharides. This suggested that these bacteria may possess carbohydrate-sensitive adhesion properties which could affect their subsequent incorporation into susceptible host cells. It must be highlighted that a glycosaminoglycan-dependent adherence step that may be operative in C. psittaci has been demonstrated [67].

More recently, the major outer membrane protein (MOMP) has been implicated in the entry of chlamydiae into their host cells. It promotes non-specific electrostatic interactions with the cell followed by more specific hydrophobic interactions, acting as an adhesin [99]. One possibility is that the amino acid residues within the hydrophobic region of the variable domain IV of the MOMP confer receptor specificity for binding. Amino acid substitutions occur in the nonapeptide region (TTLNPTIAG, amino acids 296–304) of this domain between C. trachomatis and C. psittaci strains [99] (and probably between C. pneumoniae and C. pecorum, too) which could provide differences in host tropism exhibited by members of these chlamydial strains. Other proposed chlamydial elements probably involved in attachment and ingestion are shown in Table 2.

View this table:
2

Chlamydial elements involved in attachment and internalization into host cells

ElementSpeciesStrain/serovarReferences
70-kDa proteinC. trachomatisBiovar LGV, serovar E[76, 86, 102]
62-kDa proteinC. psittaciAvian 6BC[117]
58-kDa proteinC. psittaciAvian 6BC[117]
32-kDa proteinC. trachomatisSerovar L2[100]
30-kDa proteinC. psittaciAvian 6BC and VR-125[4]
28-kDa proteinC. psittaciMammalian serovar 2 LW-613[4]
24-kDa proteinC. psittaciMammalian serovar 1 strains B577 and Fitz-9[4]
18-kDa proteinC. psittaciMammalian B577, Fitz-9, LW-613, meningo-pneumonitis, GPIC strains[4, 29]
C. trachomatisSerovars L1, L2, B, E[30, 44, 45, 100, 112]
MOMPC. trachomatisSerovar L2[99, 101]
EnvBC. psittaciGPIC strain[104]
Glycosaminoglycan-like moleculeC. trachomatisTrachoma and LGV biovars[16, 119]

Host cells which are persistently infected with C. psittaci become almost completely resistant to superinfection as attachment of exogenous EBs is blocked. This feature indicates surface structural differences between infected and non-infected cells [64, 65, 70]. However, the relationship between the resistance to superinfection and the maintenance of the persistently infected state remains unknown.

Although still controversial, the mechanism of entry used by chlamydiae may influence the further survival of these prokaryotes. Ingestion of EBs depends on both membrane integrity and continued metabolic activity of the host cell [12], and as it occurs the host cell expends oxidative and glycolytic energy but the EB neither expends energy nor synthesizes proteins [63].

Phagocytosis, a microfilament-dependent process, and receptor-mediated endocytosis, a microfilament-independent process, have been widely implicated in the entry of chlamydiae into non-professional phagocytic target cells [13, 28, 41, 42, 48, 68, 69, 77, 78, 90, 111]. Inducible non-receptor-mediated endocytosis and a combination of the two types of endocytic mechanisms have also been mentioned for C. trachomatis[77]. Most of the studies have used cytochalasin D to differentiate between endocytosis (cytochalasin D-resistant) and phagocytosis (cytochalasin D-sensitive). Inhibition of phagocytosis by cytochalasin D and prevention of receptor-mediated endocytosis by cytosol acidification have little effect on the internalization of C. psittaci. This suggests that both mechanisms occur independently. In consequence, commitment to one or the other entry pathways must take place at some stage during attachment [78]. When the modes of entry of C. psittaci guinea pig inclusion conjunctivitis strain and C. trachomatis L2 were compared, it was shown that, irrespective of the mode of entry, chlamydial strain characteristics are the predominant influence on productive infection when both mechanisms of uptake are used at the same time by the bacteria [74, 78]. It is possible that differences in entry among the chlamydial species and biovars reflect, at least in part, differences in the chlamydial binding proteins [65].

3 The chlamydial vacuole

3.1 Importation of host cell metabolites through the vacuole membrane

Residence within a host cell enables a microorganism to escape host humoral immune defence mechanisms and to avoid nutrient and site competition with other microorganisms. However, the host cell growth factors may not be in an optimally available form or concentration, and the variety of intracellular pools offers many opportunities for serious metabolite antagonisms. Moreover, the parasite must compete with the host cell for the acquisition of nutrient resources, some of which may require active membrane transport systems in order to reach the bacterial cytoplasm [58]. Chlamydiae reduce the competition with the host cells by exploiting the molecular biosynthetic machinery of the host cell to obtain ATP and by reducing RNA and DNA host synthesis. It has been observed that when HeLa cells are infected with C. psittaci at a multiplicity of infection of about one EB per host cell, the time from the end of mitosis to the start of DNA synthesis and that from the ending of DNA synthesis to the onset of mitosis are prolonged. The duration of DNA synthesis is doubled with a proportional decrease in the rate of DNA synthesis. When extreme multiplicities of infection are used, e.g. leading to 90% or more infected host cells, DNA synthesis is prevented or indefinitely prolonged [19]. The degree at which division of the host cell is reduced seems to be dependent on both the type of host cell and the dose of infection of the microorganism. In C. psittaci-infected L cells the mean generation time can be twice as long as that of uninfected cells, and it can become even longer as the rate of infection increases [1]. Nevertheless, in C. trachomatis-infected HeLa cells no inhibition of host DNA synthesis or of host cell division is observed until 2–3 days after infection has taken place [10, 15, 44]. It has also been noticed that the larger the inclusion, i.e. the more chlamydial particles within the cell, the more drastic the effect on cell division [19, 43].

In order to influence the access to nutrients, these microorganisms must overcome three membranes, the inclusion membrane and their own outer and inner cytoplasmic membranes. In spite of this handicap, Chlamydia are energy parasites capable of acquiring the adenine nucleotides of their host by a carrier-mediated transport system, hydrolyzing host ATP to generate a proton motive force or establishing an energized membrane (membrane potential) [33].

The vacuoles filled with Chlamydia are surrounded by a membrane whose convex face is almost entirely smooth, while the opposite concave face shows numerous particles [51]. RBs are found to circumscribe almost exclusively the inner margin of the inclusion with few or no organisms observed free within its center. Using electron microscopy on isolated C. psittaci inclusions, Matsumoto [54, 55] observed that the fine particles on the inclusion membrane are RB surface projections cylindrical in shape (10–13 nm diameter), closely connected to it. C. trachomatis RBs seem to associate strongly with the inclusion membrane showing an apparent thickening of this membrane at the point of contact [31, 56, 105]. This feature can also been observed in vacuoles that contain C. psittaci (Vanrompay et al., personal communication). As a result, RBs may interchange information with the host cells and influence the transfer of substances at different stages of the chlamydial cycle.

3.2 Export of Chlamydia products through the vacuole membrane

Chlamydia genus-specific lipopolysaccharide (LPS) antigen has been observed associated with the cytoplasmic membrane of C. psittaci- and C. trachomatis-infected cell lines by using silver-methenamine staining and electron microscopy [21] as well as indirect immunofluorescence with polyclonal and monoclonal antibodies [79, 80, 113]. Additionally, EB envelope antigens have also been detected on the surface of the infected host cell using indirect immunofluorescence and polyclonal antibodies [24]. In contrast, in other investigations of cells infected with C. trachomatis involving immunofluorescence [5] and with C. psittaci utilizing immunofluorescence [2], immunoelectron microscopy and flow cytometry [109], no chlamydial LPS antigen was detected in the plasma membrane of the host cell. These results seem difficult to reconcile. In one of these studies [5], LPS was detected when infected cells were fixed with methanol but not when paraformaldehyde-glutaraldehyde was used. This led to the conclusion that the labeling of the plasma membrane could be a fixation artifact. However, in those investigations where LPS was detected, a variety of different fixatives was used. Furthermore, Karimi et al. [46] showed the presence of chlamydial LPS in viral envelopes after C. trachomatis-infected cells were superinfected with a virus, demonstrating that the LPS of Chlamydia is indeed incorporated into the host plasma membrane, since the virus acquired its envelope by budding from the plasma membrane. In contrast, no MOMP was detected on virion progeny. Additionally, chlamydial LPS has been observed on bleb-like structures on the host cell surface on proximal processes of neighboring uninfected cells [46]. The presence of the chlamydial LPS in the host plasma membrane thus could be a transit step in ‘exporting’ this antigen, therefore not always detectable. It could also be an event dependent on the infecting strain. Karimi et al. [46] suggested that the LPS destined to appear in the plasma membrane may originate during the morphological transformation of RBs into EBs since during condensation large amounts of outer membrane are probably shed as intracellular vacuoles. This hypothesis was recently confirmed by Vanrompay et al. [109] who observed this antigen in membranous vesicles within the chlamydiae-containing vacuole, which were apparently pinched off from the outer membrane of reticulate bodies. Accumulation of LPS within cellular membranes decreases the membrane fluidity, possibly having significant consequences for the infected cell and host, including probably endocytic processes and lysosome-endosome fusion.

3.3 Composition and function of the chlamydial vacuole membrane

The development and composition of the membrane that surrounds the chlamydial inclusion in the host cell has for a long time been enigmatic. One remarkable aspect of the chlamydial replication cycle indeed is that even if the RNA and DNA synthesis is decreased in the host cell as a consequence of the chlamydial development, the inclusion membrane keeps growing as the infection proceeds. There is evidence from the early 1970s that the chlamydial inclusion membrane differs from the cell membrane since it resists autolysis during storage (at least up to 44 h) [9]. Further evidence was indirectly gained when it was observed that chlamydial inclusions were lysed when placed in 0.25% sucrose [55], which is widely used for the isolation of organelles from various types of cells and tissues. Furthermore, the inclusion membrane seems to be modified in its nature during the multiplication of Chlamydia[55, 92].

In early studies monitoring protein incorporation with radiolabeled amino acids, Stokes [91] proposed that the inclusion membrane might be made from either precursors already present in the host cell at the time of infection, or pre-existing membranous structures. According to this study, in the first 20 h post infection there were no marked changes in the synthesis of new membranous structures by the host cell or in the stability of the membranes already present at the time of infection. After that time, however, when transformation of RBs to EBs is about to begin, protein synthesis decreases until incorporation of radiolabeled amino acids into the membranous organelles of infected cells ceases, and pre-existing structures are broken down or altered much more rapidly than in uninfected cells. More than two decades would be necessary to prove the first proposal of Stokes right in the case of lipids incorporated in the inclusion membrane [31], and to reject his second proposal in the case of incorporated proteins [103] (see below).

Since the study of Stokes different theories concerning the enlargement of the chlamydial inclusion and its physiological implications have been proposed, but only recently more solid evidence has been reported.

C. psittaci vacuoles do not fuse among themselves [42, 115]. When a host cell becomes simultaneously infected with more than one EB, several mature inclusions can be observed within the cell [75, 109, 115]. In contrast, C. trachomatis-containing vacuoles fuse one to another to form one large mature inclusion by 8–12 h post infection [8, 15, 87]. This event is microtubule-dependent up to halfway through the developmental cycle [87].

Observations of fusion between chlamydial inclusions that contain microorganisms of the same species but not between inclusions containing different species [57] have led to the suggestion that the properties of the inclusion membrane of C. psittaci and C. trachomatis (and probably of C. pneumoniae and C. pecorum, too) are different, and that these properties are positively controlled by the organisms in the inclusion. Zeichner [117] demonstrated that the chlamydial vacuole is different from other ‘conventional’ vacuoles since the membrane protein composition of C. psittaci-containing vacuoles appears to differ from the protein composition of the host cell plasma membrane. He performed polyacrylamide gel electrophoresis on chlamydial vacuoles isolated by dextran gradient centrifugation visualizing 10 major proteins instead of a whole set of the host cell (L cells) cytoplasmic membrane proteins. Three main mechanisms that could account for these differences between the chlamydial inclusion and the L cell plasma membrane were then proposed: (1) chlamydiae may enter the host cell at a site on the plasma membrane that has a distinct limited protein composition, (2) entry at any site of the plasma membrane surface specifically excluding certain proteins from the newly formed vacuole, and (3) entry at any site, bringing a full set of the cytoplasmic membrane proteins and then selectively removing or destroying some of them. In this respect, an 11-kDa proteinase necessary for the inclusion formation has been detected in C. psittaci and C. trachomatis[95]. In addition, a C. pneumoniae cysteine proteinase has also been demonstrated [107].

Heat-inactivated chlamydiae can enter host cells via a host-specified route [118]. Comparing inclusions containing infectious EBs with inclusions containing inactivated EBs, which are brought into the host cell by two different endocytic pathways, it was shown that some selection of host plasma membrane proteins occurs during formation of the vacuole. Thus, a 25-kDa protein was only found on the inactivated EB vacuole while a 70-kDa protein was unique to the non-inactivated EB vacuole. Whether these differences are involved in the expression of the special properties of the vacuoles containing infectious EBs (which do not fuse with lysosomes) has not yet been elucidated.

Proteinaceous chlamydial components which may be released during intracellular growth and incorporated in the vacuole membrane have been identified by Rockey and Rosquit [81]. These proteins of 22, 34 and 52 kDa, of which the first two are subspecies-specific antigens of C. psittaci guinea pig inclusion conjunctivitis, were identified in infected cells in vivo and in vitro but not in purified EB lysates. When adsorbed sera from the infected guinea pigs were used for immunofluorescence detection, a fluorescent pattern was observed associated with the surface of the chlamydial inclusions but not with the developmental forms within them. Probably such proteins modify the intracellular environment in favor of the parasite. In further experiments, Rockey et al. [83] cloned and characterized a gene, incA (inclusion membrane protein A), that expresses a 39-kDa protein, IncA, detectable by immunoblotting from 18 h post infection onwards in lysates of infected cells and of RBs but not in those of purified EBs. The concentration of IncA and heat shock protein 60 increased in parallel to a maximum at 30–36 h post infection, which could indicate that IncA is produced like heat shock proteins, as a response to cope with an unkind environment. An antibody raised against IncA reacted in immunofluorescence test with the membrane of the mature chlamydial inclusion (36 h post infection) but did not appear to recognize chlamydial developmental forms, suggesting that this protein is secreted and exported from chlamydial developmental forms. IncA was localized within discrete vacuoles, apparently distributed along host cell cytoskeletal structures. Additionally, it has also been demonstrated that this protein is exposed at the cytoplasmic face of the chlamydial inclusion, and that it is phosphorylated by the host cell [82]. In recent experiments, using a polyclonal antibody against avian C. psittaci, labeling of the inclusion membrane was detected with immunoelectron microscopy [109], demonstrating that chlamydial antigen(s) is (are) incorporated into this membrane, although the nature of the antigen has not been elucidated. Both research teams proposed that these incorporated antigens may play a role in nutrient acquisition, organelle trafficking within the infected cell, membrane osmotic disturbances, and/or lysis of the membrane. It certainly represents a wide and interesting field of research for a better understanding of the interactions between this obligate intracellular parasite and the host cell.

Using fluid phase markers, membrane markers, viral constructs, fluorescent plant lectins and antibodies against membrane components [103], it has been shown that the late chlamydial inclusion membrane is not derived from the endocytic pathway of the host cell, and that the expansion of this vacuole does not occur by the addition of host proteins resulting from either de novo host synthesis or by fusion with pre-existing membranes. Indeed, no evidence of the presence of endoplasmic reticulum, Golgi apparatus, late endosomal or lysosomal proteins in the chlamydial inclusion membrane was found. The relative absence of host proteins and the ability to selectively stain the exterior surface of the inclusion using anti-chlamydial sera led to the conclusion that the growing vacuole membrane is derived from bacterial-specified components.

Hackstadt et al. [31, 32], and Heinzen et al. [39] have demonstrated that C. trachomatis interrupts an exocytic pathway to acquire endogenously synthesized sphingomyelin from the Golgi apparatus that is trafficked to the chlamydial inclusion by means of a microtubule-independent vesicle, and not from an indirect trafficking via the host plasma membrane through endocytic vesicles. It was also shown that the lipid is transiently incorporated into the inclusion membrane and from there it is rapidly acquired by the bacteria within the vacuole. Additionally, it was shown that other Golgi-derived lipids may contribute to the growth of the inclusion membrane [31]. Further evidence that the chlamydial vacuole interacts with the endocytic pathway of the host has been highlighted by Ooij et al. [68]. They observed that when cells are infected with Chlamydia a redistribution of the trans-Golgi network, where secreted and lysosomal proteins are sorted, takes place. Such redistribution can be seen once the prokaryotes have begun to multiply. The chlamydial vacuole has been seen to interface with the trans-Golgi network, although it does not appear to fuse with it.

Interactions with the early and/or late endosomal compartments, as well as with the Golgi apparatus, may provide a source of nutrients for the replicating microorganisms. Early and late endosome markers and transferrin receptor [103] and mannose-6-phosphate receptor [68] have been observed in close association with the chlamydial vacuole. It could not be determined, however, whether these markers are present on the vacuole membrane or surround it. These features are consistent with the hypothesis that chlamydiae enter the host cell through an endocytic pathway, continue to interact with the endocytic pathway, interacting additionally with the exocytic pathway. In contrast, Heinzen et al. [39], using the same experimental conditions but with a low multiplicity of infection of the microorganism, could not detect mannose-6-phosphate, which is known to enrich late endosomes and prelysosomal vesicles, in the chlamydial vacuolar membrane. In any case, no lysosomal markers on the chlamydial vacuole membrane were seen.

Finally, two cell organelles have been seen closely associated with the chlamydial vacuole: mitochondria [55, 57, 97, 109] and the Golgi apparatus [31, 32, 87, 109]. The association of mitochondria occurs in infections with C. psittaci and not with C. trachomatis nor C. pneumoniae[57]. It initiates at about 10–12 h after chlamydial inoculation, at the beginning of the reproductive stage, and it increases along with the reproductive activity of RBs. This close association could facilitate the transfer of ATP and ADP from the mitochondria to the bacterial cell [51, 98]. The reason that this happens exclusively with C. psittaci and not with other members of the Chlamydia genus could reflect inherent properties or needs of C. psittaci that warrants further research. The Golgi apparatus has been observed in the vicinity of both C. psittaci and C. trachomatis inclusions, and its association has been implicated in the progression of the chlamydial growth through the prokaryotic acquisition of Golgi-derived lipids [31]. It has been suggested that chlamydiae secrete an early gene product that determines the cellular interactions of the inclusion [89]. It directs the transport of endocytosed EBs to the peri-Golgi region, and the vacuole interaction with the exocytic pathway of the host cell. Interestingly, if the endocytic vesicle is not actively modified, chlamydiae are eventually degraded.

It has been noticed in ultrastructural studies [22, 50, 109] that chlamydial particles do not always remain within the intracellular vacuole during the whole developmental cycle. In some cases, and in apparent correlation with a high pathogenicity of strains [109], the vacuole membrane seems to degrade during the active multiplication of chlamydiae, liberating the bacteria into the cytoplasm where they stay without any alteration detected in their life cycle. The chlamydial products could account for the lysis of the vacuole membrane as a result of a cumulative process, releasing the bacteria into the cytoplasm, and by doing so favoring their access to nutrients and their further multiplication. From the point of view of the bacteria, this is advantageous since, on the one hand, when compared with low-pathogenic strains, the yield of the new progeny can be much larger, up to fourfold [109], and, on the other hand, there is no longer any need to avoid lysosomal degradation as it would be if continuing within the vacuole. More research needs to be done to identify precisely how such a phenomenon takes place.

4 Avoidance of lysosomal degradation

A virulence factor of obligate intracellular parasites is the ability to promote their own ingestion by non-professional phagocytes, which enables the parasite to occupy an unexploited ecological niche and to minimize their interaction with the cellular defences of the host [11, 61, 115]. Nevertheless, an antiparasitic capability shared by both professional and non-professional phagocytes is fusion of their lysosomes with parasite-containing vacuoles and release of acid hydrolases into the resulting phagolysosome. One way of avoiding digestion is to avoid contact with the lysosomal enzymes. Chlamydia has evolved a mechanism of doing this by preventing the fusion of the lysosomes with the vacuole where it is contained. From this point onwards we shall refer to this phenomenon as inhibition of phagosome-lysosome fusion. It has been shown that this particular inhibition is restricted to chlamydiae-laden vacuoles, since in mixed infections with C. psittaci and Saccharomyces cerevisiae or E. coli phagolysosomal fusion takes place in those vacuoles not containing chlamydiae, and neither EBs or RBs can protect the coinfecting organism from degradation [23]. In addition, in vitro analyses show that not every chlamydiae-containing phagosome escapes fusion with lysosomes, depending on the kind of host cell, chlamydial strain, what has been done to that agent and the conditions under which it has been ingested [63].

The mechanism of inhibition of the phagosome-lysosome fusion is enigmatic, and in consequence it has been the subject of a great deal of investigation. Lysosomal markers are known to be distributed within infected cells but excluded from the chlamydial vacuole [24, 39, 68, 103]. It is also known that C. psittaci and C. trachomatis avoidance of phagolysosomal fusion is not due to a general repression of lysosome function, as fusion with phagocytosed zymosan in C. psittaci-infected macrophages occurs normally [23, 106].

The microenvironment within the chlamydiae-containing vacuole is also modified by these microorganisms. It is now known that at least after the initial internalization event the chlamydial inclusion is not acidified and that it is isolated from the endosomal-lysosomal pathway [39, 88]. Without acidification there is no formation of the endosomal carrier vesicle [17], a large and spherical molecule which has been proposed to mediate the passage from early to late endosomes [3]. The neutral pH of the vacuole seems to result from the activity of the ion pumps Na+,K+-ATPase [88] and not from the absence of at least a portion of the vacuolar-type H+-ATPase as has been suggested [39]. The ion pumps are internalized with ligands during endocytosis creating an interior-positive membrane potential in the endosome that limits acidification by the H+-ATPase. Inactivation of this ion pump inhibits chlamydial growth; however, it must be highlighted that the acidification of epithelial cell vesicles that contain heat-killed inactivated EBs requires more than 2 h [88], a feature that could indicate that an active maintenance of the pH within the inclusion may not be needed until EB converts into metabolically active RB.

Many theories have been evoked that attempt to explain the mentioned inhibition on the basis of the chlamydial-host cell interactions previously described in this paper. Those theories are summarized below for further comments.

  1. Chlamydia modify the endosomal membrane by insertion of a chlamydial component causing alterations in its physicochemical properties, in membrane signals or in recognition sites so that fusion with lysosomes is no longer possible [23, 24, 82, 83, 106].

  2. Chlamydia can recognize and exploit specific receptor sites on the host cytoplasmic membrane that trigger formation of a type of endosome that normally does not fuse with lysosomes (coated-pit or receptor-mediated endocytosis is not programmed for lysosomal fusion) [23, 41, 117].

  3. Chlamydia interact with the exocytic pathway of the host cell to modify the chlamydial vacuole and evade lysosomal fusion [32, 89].

  4. EB envelope factor(s) might accumulate in lysosomes, rendering them unable to recognize or to fuse with phagosomes [24].

  5. EB envelope factor(s) might mask or enzymatically degrade recognition molecule(s) on the phagosome surface that normally marks it as an organelle with which the lysosome can fuse [24, 117].

  6. EB components interact with signal or regulatory molecules in the phagosome membrane which in turn act as a shut-off mechanism for phagosome-lysosome fusion [24].

  7. As EB enters the host cell, during formation of the vacuole membrane it specifically includes or excludes some component of the plasma membrane, and this act alters the ability of the chlamydial vacuole to fuse with lysosomes [118].

  8. Accumulation of chlamydial LPS into the vacuole membrane affects its fluidity, and this may affect endocytic processes and lysosome-endosome fusion [46].

  9. The chlamydial vacuole retains the characteristics of an early endosome that is unable to fuse with lysosomes [68, 88, 103].

The events and arguments that have given rise to these theories have been exposed in the course of this review. Nevertheless, some important remarks must be pointed out. (a) It has been widely believed that the EB is the chlamydial particle in charge of the circumvention of the phagosome-lysosome fusion on the basis that while inactivated EBs introduced into a host cell are efficiently digested by lysosomes, internalized EB envelopes are not degraded [24]. However, while the EB elements required for this evasion have not been defined, the permanence of the EB wall along the chlamydial cycle has been proven not to be necessary to maintain the inhibition throughout it [114]. This could indicate that either the EB causes a permanent blockage, or that other structures and factors play also a role in this process. (b) Receptors on the host cell surface can influence the final destination of endocytosed material. Inhibition or avoidance of the phagolysosomal fusion thus may not necessarily be an aggression by the microbe against the host, but an exploitation of host cell receptors intended for regulatory molecules with a physiological function [23]. (c) In receptor-mediated endocytosis certain requisite nutritional and regulatory proteins attach at receptor-coated pits on the eukaryotic cell surface; the resultant coated vesicles have different intracellular destinations, so some may fuse with lysosomes while others may not [23].

The degree of success with which the normal host deals with microorganisms is determined largely by the properties of the microbial surface. Thus, relatively small modifications in the microbial envelope profoundly affect one or all of the aspects of its interaction with the host cell, including recognition, ingestion, killing and digestion [25]. The biphasic developmental cycle represents in itself a number of changes. For instance, EBs are about one third the diameter of RBs, and so transition between these two particles is accompanied by a fourfold change in surface area, a variation that will probably affect both the composition and the permeability properties of the outer membrane of RBs [14, 58]. Furthermore, soon after entry reduction of disulfide bonds of the outer membrane proteins into free sulfhydryl side chains takes place as EBs convert into RBs. This generates the ‘opening’ of transmembrane pores through which nutrients may diffuse and uptake of ATP may occur. It also restores the ‘normal’ plasticity of the outer membrane allowing cell expansion and growth [14, 34, 38, 66]. Chlamydial growth stops as ATP and reducing power diminish, inducing a decrease of metabolic activities, oxidation of free sulfhydryls into disulfides (that leads to crosslinking of the cysteine-rich proteins), closing of outer membrane pores, and outer membrane rigidification as RBs transform to EBs [6, 14, 75]. In addition, it has been shown that different proteins are expressed at different times during the chlamydial cycle [18, 36, 39, 52, 73].

5 Conclusion

Currently, curative and prophylactic measures for the control of chlamydial infections often yield disappointing results; therefore, understanding the mechanisms which chlamydiae have evolved for their survival within vertebrate host cells may be essential for the future development of new tools for the control of these infections.

In the light of the information gathered here, the attachment and internalization, the inhibition of the phagosome-lysosome fusion and the further intracellular survival of chlamydiae are quite likely a multifactorial process where EBs, RBs and the host cell might be involved at different stages during the developmental cycle. For both the understanding of the biology of obligate intracellular parasites and the continuous search for better prophylactic and therapeutic measures to control chlamydial infections, not only the virulence factors of Chlamydia but also the series of interactions that take place between these microorganisms and the host cell are of great relevance. The challenge to bring completely to the light of common knowledge how all this happens is still there.

Acknowledgements

The National Council of Science and Technology (CONACYT), Mexico City, Mexico, is acknowledged for providing a grant to C. Escalante-Ochoa (Register No. 93110).

References

  1. [1]
  2. [2]
  3. [3]
  4. [4]
  5. [5]
  6. [6]
  7. [7]
  8. [8]
  9. [9]
  10. [10]
  11. [11]
  12. [12]
  13. [13]
  14. [14]
  15. [15]
  16. [16]
  17. [17]
  18. [18]
  19. [19]
  20. [20]
  21. [21]
  22. [22]
  23. [23]
  24. [24]
  25. [25]
  26. [26]
  27. [27]
  28. [28]
  29. [29]
  30. [30]
  31. [31]
  32. [32]
  33. [33]
  34. [34]
  35. [35]
  36. [36]
  37. [37]
  38. [38]
  39. [39]
  40. [40]
  41. [41]
  42. [42]
  43. [43]
  44. [44]
  45. [45]
  46. [46]
  47. [47]
  48. [48]
  49. [49]
  50. [50]
  51. [51]
  52. [52]
  53. [53]
  54. [54]
  55. [55]
  56. [56]
  57. [57]
  58. [58]
  59. [59]
  60. [60]
  61. [61]
  62. [62]
  63. [63]
  64. [64]
  65. [65]
  66. [66]
  67. [67]
  68. [68]
  69. [69]
  70. [70]
  71. [71]
  72. [72]
  73. [73]
  74. [74]
  75. [75]
  76. [76]
  77. [77]
  78. [78]
  79. [79]
  80. [80]
  81. [81]
  82. [82]
  83. [83]
  84. [84]
  85. [85]
  86. [86]
  87. [87]
  88. [88]
  89. [89]
  90. [90]
  91. [91]
  92. [92]
  93. [93]
  94. [94]
  95. [95]
  96. [96]
  97. [97]
  98. [98]
  99. [99]
  100. [100]
  101. [101]
  102. [102]
  103. [103]
  104. [104]
  105. [105]
  106. [106]
  107. [107]
  108. [108]
  109. [109]
  110. [110]
  111. [111]
  112. [112]
  113. [113]
  114. [114]
  115. [115]
  116. [116]
  117. [117]
  118. [118]
  119. [119]
View Abstract