Temporal Expression of Type III Secretion Genes of Chlamydia pneumoniae

Abstract

Chlamydia pneumoniae has been shown to possess at least 13 genes that are homologous with other known type III secretion (TTS) systems. Upon infection of HEp-2 cells with C. pneumoniae, the expression of these genes was followed by reverse transcriptase PCR throughout the developmental cycle of this obligate intracellular pathogen. In addition, expression was analyzed when C. pneumoniae was grown in the presence of human gamma interferon (IFN-γ). The groEL-1, ompA, and omcB genes were used as markers for the early, middle, and late stages of the developmental cycle, respectively, and the inhibition of expression of the fstK gene was used as a marker for the effect of IFN-γ on the maturation of C. pneumoniae. In the absence of IFN-γ, the TTS genes were expressed as follows: early stage (1.5 to 8 h), yscC, yscS, yscL, yscJ and lcrH-2; middle stage (by 12 to 18 h), lcrD, yscN, and yscR; and late stage (by 24 h), lcrE, sycE, lcrH-1, and yscT. Of the genes expressed early, the lcrH-2 gene was detected the earliest, at 1.5 h. Expression of the yscU gene was not detected at any of the time points examined. Under the influence of IFN-γ, the cluster of TTS genes that were normally not expressed until the middle to late stages of the developmental cycle, namely, lcrD, lcrE, and sycE, as well as lcrH-1, were down-regulated, and expression could not be detected up to 48 h. In contrast, the expression of the other TTS genes appeared to be unchanged in the presence of IFN-γ. The lcrH-1 and lcrH-2 genes differed from one another in both their temporal expression and response to IFN-γ. In other TTS systems, these genes code for proteins that function in regulation of effector protein synthesis as well as serve as chaperones for proteins that provide for the translocation of the effector proteins into the host cell. In summary, the expression pattern of the TTS genes of C. pneumoniae examined suggests that they are temporally regulated throughout the developmental cycle. Furthermore, paralleling the inhibition of the maturation of the reticulate body to the elementary body, TTS genes expressed in the later stages of the cycle appear to be down-regulated when the organism is grown in the presence of IFN-γ.

Chlamydia pneumoniae is an important human pathogen, causing a wide range of respiratory tract infections, including pharyngitis, sinusitis, bronchitis, pneumonia, and asthmatic bronchitis (10, 12). In addition, this organism has been linked to a growing number of chronic diseases, among which are coronary artery disease, strokes, multiple sclerosis, and Alzheimer's (1, 7, 11, 29, 32). Antibiotic treatment appears to be effective in controlling the acute phase of an infection, but there is some in vitro and in vivo evidence suggesting that recurrent and even latent forms of the organism can persist upon resolution of the acute or the initial phase of the infection (4, 13).

Chlamydia is an obligate intracellular bacterium with a unique developmental cycle. The two main forms of this bacterium are the elementary body (EB), an infectious spore-like structure, and the reticulate body (RB), the noninfectious, metabolically active intracellular form. In C. pneumoniae, like other members of this genus, the EB enters the host cell and is contained in an inclusion which continues to grow and develop throughout the developmental cycle. This cycle has been observed to last from 32 to 72 h, depending on the species and growth conditions. Once inside the cell, the EB reorganizes into the RB, which is less condensed and larger in size and has an increased level of gene expression. RBs, after a period of multiplication by binary fission, rearrange back to EBs and are released from the host cell.

Evidence has been mounting for the existence and role of a type III secretion (TTS) system in Chlamydia (2, 8, 16, 18, 26, 28, 30, 31, 33, 34, 35). The first genetic evidence for this was presented by Hsia et al. in 1997 (16) when they described four genes homologous to structural and regulatory components of a contact-dependent or TTS apparatus. Subsequently, work on the Chlamydia genome revealed structural and regulatory genes that share high homology with TTS systems of other bacterial pathogens (18, 28, 31, 33). It has been shown that Chlamydia trachomatis can express some of the genes identified as belonging to a TTS system (8, 30). Genes coding for Inc proteins in Chlamydia have been shown to be secreted by a heterologous TTS system (34). Proteomic analysis of Chlamydia has identified proteins known to comprise a TTS system (35). Matsumoto and others, using electron microscopy, demonstrated surface projections on EBs and RBs, similar to the TTS apparatus identified in other bacteria (21, 24). Ultrastructural descriptions of these projections have shown that they appear to localize to one side of the RB or EB and, in the case of the RB, to pierce the inclusion membrane within infected cells. It has been suggested by Bavoil et al. (2) that these comprise a TTS secretory apparatus of Chlamydia. Another structural characteristic of the chlamydial inclusion that is suggestive of a TTS system is the close interaction of the RBs with the inclusion membrane (21, 26). Therefore, existence of contact-dependent or TTS in Chlamydia is strongly suggested by the microscopic observations, genomic and proteomic data, and parallels Chlamydia has with other intracellular bacterial pathogens of animals and plants that use this system in their interaction with host cells. Since TTS systems have been shown with other pathogens to play a major role in the pathogen-host interaction, this may also prove to be a key virulence mechanism of Chlamydia (17).

One could postulate a role for TTS both in the initial stages of infection where Chlamydia first comes into contact with the host cell as well as in the intracellular phase of Chlamydia development. With regard to this later point, it has been observed that under the influence of exogenous gamma interferon (IFN-γ) and select antibiotics, the progression of the RB to the EB appears inhibited, leaving Chlamydia in an intracellular, persistent, or latent phase (3).

In this study the temporal expression of 13 Chlamydia TTS genes was examined during the normal developmental cell cycle. In addition, expression was also followed when the pathogen was grown in cells treated with human IFN-γ to determine if the expression pattern was changed when the progression of the maturation of a RB to an EB was altered.

MATERIALS AND METHODS

Organisms and cell culture.

C. pneumoniae CM-1, obtained from the American Type Culture Collection (ATCC) (Manassas, Va.), was propagated in HEp-2 cells (ATCC). HEp-2 cells were grown in Eagle's minimal essential medium (MEM) (Irvine Scientific, Santa Ana, Calif.) supplemented with 10% fetal bovine serum (Invitrogen Life Technologies, Carlsbad, Calif.), 2 mM l-glutamine (Irvine Scientific), 50 μg of gentamicin/ml (Irvine Scientific). C. pneumoniae was prepared by infecting HEp-2 cells in 1-dram glass vials which had been pretreated for 20 min at room temperature with DEAE-dextran (30 μg/ml). After infection, the monolayers were centrifuged at room temperature for 1 h at 800 × g, followed by the addition of media containing cycloheximide (1 μg/ml). Cultures were incubated for 72 h at 37°C, at which time the media was removed, and the cultures were sonicated in sucrose-phosphate-glutamate (SPG; pH 7.4) and centrifuged (500 × g, 10 min), and the supernatant was stored at −80°C.

Hep-2 cells and the C. pneumoniae stocks were determined to be free of Mycoplasma contamination by PCR with primers to the Mycoplasma 16S rRNA, 5′-GGG AGC AAA CAG GAT TAG ATA CCC T and 5′-TGC ACC ATC TGT CAC TCT GTT ACC CTC, which have been previously described (25).

Nucleic acid isolation.

HEp-2 cell monolayers were infected as described with C. pneumoniae at a multiplicity of infection of 1 to 3 (IFN-treated cells). After centrifugation, infected cells were rinsed twice with 0.01 M phosphate-buffered saline (PBS) (pH 7.2). Subsequently, MEM without cycloheximide was added with or without human recombinant IFN-γ (40 ng/ml) (rhIFN-γ; Promega Corporation, Madison, Wis.). Monolayers used to monitor the infection were fixed in methanol at various times after infection and stained as previously described by using a mixture of monoclonal antibodies (MAbs) raised in our laboratory to C. pneumoniae (27). Uninfected controls were treated exactly as the infected monolayers, but SPG alone was used to inoculate the monolayers. Cells were collected and processed for nucleic acid analysis at 1.5, 4, 8, 12, 18, 24, 32, 48, 60, and 72 h postinfection (p.i.). For the time zero determination, the DNA and RNA of the C. pneumoniae stock that was used to infect the monolayers was purified and analyzed along with the samples from time points after infection. Therefore, the results obtained at time zero from the stock taken directly from the portion stored at −70°C is referred to as the basal level of mRNA, since it presumably contains mRNA present when the original stock was harvested. At the designated times, cell monolayers were washed with PBS. TRIzol (Invitrogen Life Technologies) was used in accordance with the manufacturer's instructions for isolating total DNA and RNA. RNA preparations were treated with RNase-free DNase RQ1 (Promega Corporation) by following manufacturer instructions and were confirmed to be DNA-free by performing a PCR for the Chlamydia groEL-1 gene. The concentration of isolated DNA and RNA samples was established by spectrophotometry, and preparations were stored frozen (DNA at −20°C and RNA at −80°C).

PCR.

Optimal amplification conditions for each gene examined were established. Semiquantitative PCR assays were performed with Taq polymerase (Promega). In general, the reaction mixtures included Taq reaction buffer, 2 mM MgCl₂, 0.5 μM primers, 0.5 U of Taq DNA polymerase, 0.2 mM dNTPs, 1% dimethyl sulfoxide, and serial dilutions of total DNA preparations; diethyl pyrocarbonate-treated water was added up to 25 μl. The hot start procedure was applied to attain maximum efficiency of the PCR. PCR conditions were denaturation at 94°C for 2.5 min, primer annealing at 59°C for 1 min, and primer extension at 72°C for 1.5 min followed by 34 cycles of 94°C for 40 s, 59°C for 1 min, and 72°C for 1 min, followed by a final extension for 10 min at 72°C. PCR products were resolved and visualized on 2% agarose gels with ethidium bromide.

Analysis of gene expression.

To quantitate C. pneumoniae DNA for each time point examined, PCR was employed targeting the Chlamydia groEL-1 gene. Briefly, by using the serial dilutions of the total DNA preparation, the minimum amount of DNA that could be amplified by PCR was determined. The dilutions of pure C. pneumoniae DNA (50, 25, 5, and 1 copy/reaction) that were used as standards were amplified in parallel with the experimental DNA preparations.

Transcription of RNA was assayed by RT-PCR. Specific messages were amplified with the Access reverse transcriptase (RT)-PCR kit (Promega). For each sample analyzed, the number of Chlamydia genome equivalents was established, and for each time point, material corresponding to 5 × 10³ Chlamydia genomes was used. In addition, the amount of RNA in each RT-PCR mixture was standardized to 30 ng of RNA by using RNA from uninfected Hep-2 cells that had been shown to be free of DNA. Genes of interest and sequence-specific primers used in this work are shown in Table 1.

TABLE 1.

Primers used for PCR and RT-PCR assays

C. pneumoniae ORF	Gene	Positiona	Primers (5′→3′)	Product size (bp)
CPn 0322	yscU	F	AATTTTCTTTTTGATTGTTGCGA	259
		R	ATATTTTTCAGGCATGTAGCCAA
CPn 0323	lcrD	F	AGCTCTCAAGAGTTTTTGGGAAT	285
		R	TTGTCCTTGAGAGAACTTGAAGC
CPn 0324	lcrE	F	GGAATCCAAACTCCTTCTGATCT	285
		R	TGCTTTGGGATAATCTTCATTGT
CPn 0325	sycE	F	CCCTGTGCAAATAGAAGAAGATG	283
		R	TCTTAGGGATTCCATCCAAATCT
CPn 0702	yscC	F	ATTCCCTTGGTTTCCGATAATAA	290
		R	AGCAGCTCCTGAAGTCCTTAGAT
CPn 0707	yscN	F	GCTAGTGCATTGGAGAGAACAAT	298
		R	CAGAGTATTTTCGTGATCAAGGG
CPn 0811	lcrH-1	F	TGCCAATAACTGGAAGAGTCCTA	298
		R	CAATCAACCTCAAAAACCTTCAG
CPn 0823	yscT	F	ACTATGGATCGGAATGACTTCAA	262
		R	GAAGGAGATGATCATAGGCATTG
CPn 0824	yscS	F	GCAAAGCGTAAAATCATATTGCT	252
		R	CGTGTTAGCATTTTTCGCAAC
CPn 0825	yscR	F	GAAGGGAAGGTTTTCTGTGAGAT	286
		R	ACGCCTTAGGAGTACAACAAACA
CPn 0826	yscL	F	GAGTTCAGGACGACTTTTCTCAA	300
		R	GCACAAATTCGTCAAGAAGCTAA
CPn 0828	yscJ	F	TACCCCAAACAGAAACATAATCG	268
		R	GTCGATGCCTCAGTACAGATTTC
CPn 1021	lcrH-2	F	TCATTATAGAGGGGCTTGTGTTG	272
		R	TATGCTAAAAGCATCACCGTTTT
CPn 0134	groEL-1	F	ACATAGCTTTTCTTCTGTCACCG	268
		R	CTCTCCAGCTACTTCTCCACAAA
CPn 0557	omcB	F	CTACAGTTACGTTACGGGCAAT	274
		R	CAAGTGATGGGAAATTAGTCTGG
CPn 0695	ompA	F	GAACCCTTCTGATCCAAGCTTAT	259
		R	GTGTAAATGCTTATTGTAGGCCG
CPn 0880	ftsK	F	TTGCTAAAATCAGATGAATCCCTAC	300
		R	GTGATTACCGAATCAAGAGAAGTGT

^aF, forward; R, reverse.

RT reactions were carried out in accordance with the manufacturer's instructions. Briefly, the RT extension was carrying out for 45 to 60 min at 42 to 48°C followed by 2 min at 94°C and PCR for 40 cycles consisting of denaturation at 94°C for 45 s, annealing at 59°C for 1 min, and extension at 69°C for 1.5 min, followed by a final extension at 68°C for 10 min. This was followed by denaturation of the mixture of RNA templates and specific primers (70°C for 5 min) before the PCR. PCR products were separated on 2% agarose gels by ethidium bromide staining. Each RT-PCR was accompanied by a reaction without added RT to control for DNA contamination. Amplification of the 323-bp product with the specific 1.2-kb positive control RNA transcript and gene-specific primers provided in the Access RT-PCR kit was used as a positive control for RT-PCR.

To avoid mRNA degradation and DNA contamination, DNase and RNase free plastic and glassware and diethyl pyrocarbonate (Sigma Chemical Company, St. Louis, Mo.)-treated water were used throughout the experiment. RNasin RNase inhibitor (Promega) (1 U/1 μl of reaction mixture) was used during the RQ1 DNase treatment, and 0.1 U/1 μl was used during the RT-PCR procedure, to inhibit the mRNA degradation.

Characterization of the IFN-γ-treated cultures.

To establish the dose of IFN-γ to be used throughout the experiments as well as to determine the infective progeny from the IFN-γ-treated cultures, monolayers of HEp-2 cells were infected with C. pneumoniae, as described above, without the addition of cycloheximide. Immediately after infection, infected cells were treated with IFN-γ at doses ranging from 0 to 100 ng/ml. To quantitate the viable Chlamydia progeny, infected monolayers were sonicated 72 h p.i. and resuspended in ice-cold SPG. Serial dilutions of the infected cells were transferred to fresh monolayers of HEp-2 cells that were infected as described above. In addition, cycloheximide (1 μg/ml) was included in the media for each of the transferred cultures. All cultures were fixed at 72 h p.i. and stained as previously described. Mature inclusions were counted, and the number of progeny was calculated.

To establish the IFN-γ dose with which C. pneumoniae could be rescued from a persistent state, at 72 h p.i., IFN-γ-containing medium was removed from infected monolayers, and the monolayers were washed with PBS. MEM containing 20 μg of l-tryptophan (Sigma)/ml was added to each vial, and cells were incubated at 37°C. At 72 h p.i., cultures were fixed and stained, and a set was transferred to fresh monolayers as described above.

RESULTS

Expression of TTS genes during the developmental cycle of C. pneumoniae.

In order to determine the expression of the TTS genes throughout the developmental cycle of C. pneumoniae, HEp-2 cells were harvested at various times after infection with C. pneumoniae, and total RNA and DNA were purified. By PCR with primers to the Chlamydia groEL-1 gene, the DNA samples were serially diluted to determine the lowest dilution at which PCR products could be detected. In addition, purified Chlamydia DNA was used as a standard. From this semiquantitative analysis, there was an approximately 160-fold increase in C. pneumoniae DNA over the 72-h observation period (Fig. 1). The largest increase in DNA was seen from 24 to 60 h after infection. During the initial stages of infection, up to 4 h, Chlamydia DNA accounted for 0.02% of the total DNA in infected cultures, and this increased to 1.3% by 72 h.

FIG. 1.

HEp-2 cells infected with C. pneumoniae CM-1 were harvested at various times after infection. The DNA was isolated, and serial dilutions of the DNA preparation were tested by PCR by using primers to the groEL-1 gene. Purified Chlamydia DNA was used as a standard to estimate the genomic copies present in the cultures at each time point.

For each time point examined, RT-PCR was performed with purified nucleic acid containing 5,000 genome equivalents of C. pneumoniae. The overall results of these assays for the 13 TTS genes can be seen in Fig. 2. The genes used as markers for the different phases of the developmental cycle—early, middle, and late—were expressed by 1.5 to 4 h, 12 h, and 24 h for groEL, omcB, and omp A, respectively. At time zero, mRNA for groEL, ompA, and omcB could be detected in the original inoculum, as well as mRNA for four of the TTS genes, yscN, yscL, yscJ, and lcrH-2. This basal level of mRNA seen in the inoculum was depleted by 4 h for most of the genes examined. Despite optimizing amplification of the primers for the yscU gene by PCR with purified C. pneumoniae, mRNA could not be detected at any of the time points examined. A faint signal could be seen for yscT at 18 h p.i. which remained weak for all of the subsequent time points examined. Both of these genes code for structural protein elements of the TTS machinery.

FIG. 2.

The expression of 13 known TTS secretion genes was analyzed throughout the developmental cycle of C. pneumoniae by RT-PCR, as was that of three chlamydial genes, groEL-1, ompA, and omcB, that have been shown to correlate with the early, middle, and late developmental cycle, respectively (30). HEp-2 cells infected with C. pneumoniae CM-1 were harvested at various times after infection. The RNA was isolated, and the equivalent of 5,000 genomic copies of C. pneumoniae was used in a RT-PCR with primers to the genes shown (Table 1). Each RNA preparation was treated with RQ1 RNase-free DNase and was shown to be free of C. pneumoniae DNA by performing a PCR with the primers used for RT-PCR (data not shown). RT-PCR products were separated on a 2% agarose gel and stained with ethidium bromide. RNA from uninfected HEp-2 cells (UI) was used as a negative control, and phiX174 DNA/HaeIII (Promega) served as base pair markers (M). The genomic map at the bottom was modified from the Chlamydia genomic project website (http://chlamydia-www.berkeley.edu:4231), and the genes of interest are highlighted in black.

Genes clustered from CPn 0323 to 0325 (Fig. 2) appeared to be expressed in a temporal sequence from 12 h to 32 h after infection. Homologues of the protein coded by the first gene, lcrD, have been shown in other systems to be located in the bacterial inner membrane and are thought to contribute to the formation of a channel that conducts proteins. The product of lcrE (YpoN) is part of the needle or valve regulating the transport of proteins that eventually will be translocated to the host cell (36). The last gene in this operon, sycE (CPn 0325), is thought to act as a chaperone, protecting the translocation domain of effector proteins, e.g., YopE in Yersinia. Expression of the other two clusters of TTS genes, which encode the injection needle of the TTS system and consist of CPn 0823 to 0826, CPn 0828, CPn 0702, and CPn 0707, in general appeared about 8 h after infection. Expression of these genes was detected throughout the developmental cycle, with peak expression during the exponential stage, 32 to 48 h p.i. The exception to this was yscC, which appeared earlier, at 1.5 h after infection.

Of interest are the two chaperone homologues, lcrH-1 (CPn 0811) and lcrH-2 (CPn 1021), which, unlike other TTS systems, are not clustered with other known homologues of TTS genes. In addition, these two genes are clearly expressed at different times during the Chlamydia developmental cycle. lcrH-2 is expressed early, by 1.5 h after infection, and continuously during the cycle and appears to diminish towards the last 24 h of development. In contrast, lcrH-1 is not expressed until late in the developmental cycle, suggesting a different role for each of the gene products.

IFN-γ influence on the development of C. pneumoniae.

In order to study the effect that IFN-γ had on the expression of the TTS genes, the amount of IFN-γ to use was established for the cell culture system employed. Here the IFN-γ effect was measured by the microscopic appearance of the inclusions, the effect on infectious progeny, and the ability to recover C. pneumoniae from IFN-γ-treated cultures. As can be seen in Fig. 3, the effect on progeny recovered at 72 h from infected monolayers was inversely correlated with the amount of IFN-γ added. At 40 ng of IFN-γ/ml, this effect appeared to level off. In addition, at 40 ng/ml, in contrast to the lower concentrations used, abnormal or smaller inclusions were predominant in the monolayer (Fig. 4). However, monolayers to which no IFN-γ was added, as well as those with concentrations up to 20 ng/ml, had a predominance of normal appearing inclusions. At 72 h of infection, IFN-γ was removed and replaced with fresh media supplemented with tryptophan (0.1 mM). Upon incubation for 72 h, these monolayers were assessed for the number of infectious inclusion-forming units, i.e., progeny. There was an increase of more than 2 log units in the number of infectious progeny over the number in monolayers under the continuous influence of IFN-γ. Therefore, for the TTS gene expression experiments, 40 ng of IFN-γ/ml was employed.

FIG. 3.

Monolayers of HEp-2 cells were infected with C. pneumoniae CM-1 in the presence of IFN-γ at concentrations ranging from 3 to 100 ng/ml. Infected cells were incubated at 37°C, and at 72 h, the IFN-γ-containing medium was removed. The monolayers were washed with PBS, sonicated in SPG, transferred to fresh HEp-2 monolayers, and allowed to incubate in media free of IFN-γ. The infected cells were fixed at 72 h and stained with MAbs to C. pneumoniae (27). The number of inclusions were counted, and the total yield of infectious progeny was calculated for each IFN-γ concentration.

FIG. 4.

Monolayers of HEp-2 cells were infected with C. pneumoniae CM-1 without (A) and with (B) the addition of IFN-γ (40 ng/ml). Cultures were incubated at 37°C, and at 72 h, monolayers were fixed with methanol and stained with MAbs to C. pneumoniae (27). The abnormal, small inclusions that develop when the infected cells are incubated in the presence of IFN-γ are indicated by an arrow.

IFN-γ influence on gene expression.

Untreated and IFN-γ- treated infected monolayers were harvested at 4 and 48 h after infection. Expression of the genes groEL-1, ompA, and omcB appeared slightly up-regulated by IFN-γ (Fig. 5). However, the gene, ftsK, which was included as a control for monitoring IFN-γ treatment, as expected, was not expressed at 48 h when IFN-γ was present. The genes in clusters 2 and 4, which were expressed early to midcycle in the untreated cultures, appeared unaffected by IFN-γ treatment. In contrast, the genes in cluster 1, expressed mid- to late cycle in untreated cultures, were not expressed when C. pneumoniae was grown in the presence of IFN-γ. Genes coding two of the chaperones, lcrH-1 and lcrH-2, were affected differently by IFN-γ treatment. lcrH-1, which was expressed mid- to late cycle in the normal developmental cycle, was not detected when IFN-γ was present. In contrast, lcrH-2, expressed throughout the normal developmental cycle, did not appear to be affected by IFN-γ treatment.

FIG. 5.

HEp-2 cells were infected with C. pneumoniae CM-1 with and without IFN-γ (40 ng/ml). Infected monolayers were incubated at 37°C, and at 4 and 48 h p.i., cells were harvested and RNA was isolated. The equivalent of 5,000 chlamydial genomes were used in each RT-PCR in the presence (+) and absence (−) of the RT. A positive control included with the RT-PCR kit (see Material and Methods) was tested in parallel with all RNA preparations. phiX174 DNA/HaeIII DNA served as base pair markers.

DISCUSSION

TTS systems are potent pathogenic tools used by several bacteria to alter the host to ensure their own survival. As with other virulence factors, in general, TTS genes are clustered in a pathogenicity island, presumably being transferred horizontally through evolution (6, 19, 23). TTS systems have been described in several bacteria, including Yersinia, Salmonella, Shigella, and Pseudomonas (17). These systems are known to consist of several elements, including the following: structural elements within the bacterial membrane as well as those that are assembled upon contact with the host cell and that extend into the host cytoplasm (e.g., ysc genes); effector proteins which are released into the host cell to modulate host cell function (e.g., yop genes); a translocator apparatus that provides a pore within the host cell membrane for delivery of effectors (e.g., translocator genes yopB, yopD, and their chaperone lcrH); and specialized chaperones which function to stabilize and assure efficient secretion of translocator proteins and also to regulate expression of some of the TTS genes (e.g., syc genes). Depending on the pathogen, modulation of the host cell is accomplished in a number of ways. This can vary from host cell apoptosis, which is seen with Shigella, to the interference with host cell signal transduction with Yersinia which ultimately results in this pathogen evading an immune response (5, 17). Therefore, this type of system would benefit intracellular pathogens such as Chlamydia by allowing them not only to invade but to grow and possibly persist in the host cell.

The characteristics that seem common to most TTS systems described to date are the clustering of the approximately 20 genes that comprise the TTS system and a relatively low G+C content of this cluster relative to the rest of the genome. Chlamydia appears to differ in both of these traits in that the corresponding putative TTS genes are scattered throughout the chromosome and the G+C content of its genome does not vary greatly from that of the TTS genes (33). In addition, in Chlamydia, there appear to be two homologues, lcrH-1 and lcrH-2, of the lcrH gene, which codes for a chaperone in the other systems. In Yersinia, lcrH codes for a small acidic cytoplasmic protein (19 kDa) with a predicted pI of 4.5 which is thought to act as a chaperone for the translocator apparatus proteins, YopB and YopD, and is thought to have an additional regulatory role for yop expression (17). From our experimental evidence, it appears that the expression of these genes is distinct, with lcrH-2 being expressed early in the developmental cycle while lcrH-1 is expressed late in the developmental cycle.

Temporal regulation of expression of the TTS system by a pathogen is not unique (17). One of the TTS pathogenicity islands in Salmonella, SPI1, codes for a number of proteins that appear to be required for the initial stages of establishing a systemic infection (9). These include proteins that depolymerize the host cell actin filaments, others that bind to tyrosine phosphatase and disrupt the actin cytoskeleton, and proteins that target host cell transduction pathways. On the other hand, the other TTS system in Salmonella, SPI2, codes for effector proteins that interfere with normal cellular vesicle trafficking. A number of factors have been described that affect the regulation of both SPI1 and SPI2. It appears that expression of genes in SPI1 and SPI2 is highly regulated and responds to environmental conditions and that they are inversely regulated (14, 15, 37). Unlike Salmonella, C. pneumoniae has only one set of structural TTS genes. The finding of two homologues of the chaperone lcrH in C. pneumoniae that are expressed at different times in the developmental cycle entertains the possibility of a single-two phase TTS system employed by this obligate intracellular bacteria.

It is known that in vitro, IFN-γ appears to halt the Chlamydia developmental cycle, depending on the time at which it is added to the infected cell, but in general the development of the RB to the EB appears to be inhibited. In the presence of IFN-γ, the lcrH-2 gene expression appeared unaffected by IFN-γ treatment, while the lcrH-1 gene, which had been expressed in the later stages of infection, was not expressed in the IFN-γ-treated cells. In addition, a small cluster of TTS genes, Cpn 0323 to 0325, was also affected by IFN-γ treatment in that while being expressed in the middle of the cycle to late in the cycle under normal growth conditions, they, like lcrH-1, are not expressed when C. pneumoniae is grown in IFN-γ-treated cells. We cannot rule out the possibility that for those genes for which we did not see an effect with IFN-γ, we are detecting persistent mRNA, a reflection of their mRNA stability. However, it is clear that some TTS gene expression is downregulated by IFN-γ. One could then speculate that because TTS genes normally expressed late in the developmental are affected by IFN-γ treatment, a TTS system contributes in some way to the inability of the RB to reorganize and develop normally into an EB, possibly due to the inability of the “second phase” of the TTS system to function properly.

The results we observed for the temporal expression of groEL-1, ompA and omcB with C. pneumoniae were similar to that reported by Shaw et al. (30) for C. trachomatis. Under the conditions we employed groEL-1 expression represented the early phase of infection (4 h), ompA the middle phase (12 h) and omcB the later stages of infection (24 h). Of the TTS genes we studied, Fields and Hackstadt (8) examined one of the clusters during development of C. trachomatis in HeLa cells. They reported some differences from what we observed with C. pneumoniae. They found that sycE (scc1) was expressed by 2 h after infection while we were not able to detect gene products until 24 h into the normal developmental cycle. Similarly, they reported a weak signal from lcrE (copN) as early as 12 h after infection, whereas we were not able to detect this product until 24 h with C. pneumoniae. In both systems, expression of lcrD (cds2) was detected by 12 h. The major difference between the two studies was with the yscU (cds1) gene. We were unable to detect expression whereas Fields and Hackstadt (8) were able to detect it by 12 h after infection. These differences between studies could be due to the fact that C. pneumoniae has a longer developmental cycle than C. trachomatis and/or the sensitivity of the systems employed.

Mathews et al. (20) also followed expression of ompA, omcB and groEL-1 (hsp60) when C. pneumoniae was grown in the presence of IFN-γ. They presented evidence that ompA was up-regulated, a finding that we also observed. They did not find omcB or groEL-1 to be affected; however, although slight, it appeared that in our experimental system they were up-regulated. Byrne et al. (4), in examining the effect of IFN-γ on DNA replication, reported that the ftsK gene was down-regulated when C. pneumoniae was in a persistent state, a finding corroborated by our work. Performing a proteomic analysis of C. pneumoniae under the influence of human recombinant IFN-γ, Molestina et al. (22) reported that the one TTS protein they detected, SctN (yscN), was unaffected by interferon treatment. While we did not analyze the effect of IFN-γ at 24 h, at 48 h we did not detect a significant effect on gene expression from the aspect of mRNA production for yscN.

In summary, we have presented evidence for temporal regulation of the TTS genes of C. pneumoniae. In addition, TTS genes expressed in the mid- to late stage of the developmental cycle appear to be down-regulated by IFN-γ treatment. One could then postulate that C. pneumoniae uses the structure of the TTS system to translocate different effectors into the host cell, depending on the phase of the developmental cycle. In addition, suppression of some of the TTS genes may play a role in maintaining C. pneumoniae in a persistent or altered state within the host cell. Therefore, it will be important to determine what effectors are present in C. pneumoniae and what role each plays in the development and possibly persistence of this important human pathogen.