Chlamydia trachomatis genes whose products are related to energy metabolism are expressed differentially in active vs. persistent infection

Microbes and Infection 4 (2002) 13–22 www.elsevier.com/locate/micinf

Original article

Chlamydia trachomatis genes whose products are related to energy metabolism are expressed differentially in active vs. persistent infection Hervé C. Gérard a, Julia Freise a,b, Zhao Wang a, George Roberts a, Debbi Rudy c, Birgit Krauß-Opatz b, Lars Köhler b, Henning Zeidler b, H. Ralph Schumacher d, Judith A. Whittum-Hudson a,c, Alan P. Hudsona,e* a

Department of Immunology and Microbiology, Wayne State University School of Medicine, Gordon H. Scott Hall, 540 East Canfield Avenue, Detroit, MI 48201, USA b Medizische Hochschule Hannover, Abteilung Rheumatologie, Carl Neuberg Straße 1, 30625 Hannover, Germany c Department of Internal Medicine, Division of Rheumatology, and Department of Ophthalmology, Wayne State University School of Medicine, Lande Research Building, 550 East Canfield Avenue, Detroit, MI 48201, USA d Department of Medicine, Division of Rheumatology, University of Pennsylvania School of Medicine and Medical Research, Department of Veterans Affairs Medical Center, Philadelphia, PA 19104, USA e Medical Research, Department of Veterans Affairs Medical Center, Detroit, MI 48201, USA Received 10 January 2001; accepted 5 October 2001

Abstract The Chlamydia trachomatis genome encodes glycolysis and pentose phosphate pathway enzymes, two ATP/ADP exchange proteins, and other energy transduction-related components. We asked if and when chlamydial genes specifying products related to energy transduction are expressed during active vs. persistent infection in in vitro models and in synovia from Chlamydia-associated arthritis patients. Hep-2 cells infected with K serovar were harvested from 0–48 h post-infection (active infection). Human monocytes identically infected were harvested at 1, 2, 3, 5 days post-infection (persistent). RNA from each preparation and from synovial samples PCR-positive/-negative for Chlamydia DNA was subjected to RT-PCR targeting (a) chlamydial primary rRNA transcripts and adt1 mRNA, (b) chlamydial mRNA encoding enzymes of the glycolysis (pyk, gap, pgk) and pentose phosphate (gnd, tal) pathways, the TCA cycle (mdhC, fumC), electron transport system (cydA, cydB), and σ factors (rpoD, rpsD, rpoN). Primary rRNA transcripts and adt1 mRNA were present in each infected preparation and patient sample; controls were negative for chlamydial RNA. In infected Hep-2 cells, all energy transduction-related genes were expressed by ≈11 h post-infection. In monocytes, pyk, gap, pgk, gnd, tal, cydA mRNA were present in 1–2-day-infected cells but absent at 3 days and after; cydB, mdhC, fumC were expressed through 5 days post-infection. RT-PCR targeting mRNA from σ factor genes indicated that lack of these gene products cannot explain selective transcriptional down-regulation during persistence. Analyses of RNA from synovial tissues mirrored those from the monocyte system. These data suggest that in the first phase of active chlamydial infection, ADP/ATP exchange provides energy required for metabolism; in active growth, glycolysis supplements host ATP. In persistence host, rather than bacterially produced, ATP is the primary energy source. Metabolic rate in persistent C. trachomatis is lower than in actively growing cells, as judged from assays for relative chlamydial primary rRNA transcript levels in persistent vs. actively growing cells. © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. Keywords: Chlamydia trachomatis; ATP production; Glycolysis; Pentose phosphate pathway

1. Introduction Chlamydia trachomatis is an obligate intracellular bacterium responsible for several human diseases. Certain * Corresponding author. Tel.: +1-313-993-6641; fax:+1- 313-577-1155. E-mail address: [email protected] (A.P. Hudson). © 2002 Éditions scientifiques et médicales Elsevier SAS. All rights reserved. PII: S 1 2 8 6 - 4 5 7 9 ( 0 1 ) 0 1 5 0 4 - 0

strains of the organism cause blinding trachoma, for example, while others infect the urogenital tract (see [1] for a review). Active genital chlamydial infections are significant in themselves, but they can also have serious sequelae in both men and women [1–3]. In the latter, such sequelae include often severe fertility problems and possibly cervical cancer (e.g. [3–5]). Another potential outcome of genital

14

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

chlamydial infection is development of reactive (inflammatory) arthritis [6–9]. Some patients with this arthritis develop only acute disease, but in other individuals the acute arthritis becomes chronic, lasting for decades. Chlamydiaassociated arthritis has been diagnosed most often in men, but this is probably an artifact of the diagnostic criteria for the disease; that is, diagnosis of Chlamydia-associated arthritis requires a documented prior infection with C. trachomatis [10]. However, chlamydial infection in women is often asymptomatic, another aspect contributing to the genesis of significant reproductive sequelae [1,3,11]. The C. trachomatis developmental cycle has two phases. In the first, the extracellular, metabolically inactive but infectious elementary body (EB) binds to and is internalized within an appropriate host cell; internalization occurs into a membrane-bound vesicle. In the second phase, intracellular EB reorganize into the active growth form of the organism, the reticulate body (RB), which undergoes several rounds of cell division within the inclusion. At the end of the intracellular phase, RB reorganize back to EB; these are released from the host cell via lysis or exocytosis to perpetuate the infection [12,13]. Recent work has demonstrated that the sequelae of genital chlamydial infection often involve a persistent form of the organism, and these persistent forms reside at anatomic locations distant from that of the primary infection. That is, under some circumstances, C. trachomatis cells disseminate from the urogenital tract to the joint or elsewhere and establish an infection in which RB-like forms persist over long periods in a metabolically and morphologically aberrant state within the host cytoplasmic inclusion [6–9,14,15]. Some aspects of the persistent state have been studied for synovial C. trachomatis. We have supplied evidence that chlamydiae within the joint are metabolically active, although they show an unusual transcriptional profile that distinguishes them from RB during active infections [6,16]. For example, persistent C. trachomatis cells express low levels of omp1 mRNA [11,14,15,17–19]; this gene encodes the major outer membrane protein of the organism, and actively growing C. trachomatis cells transcribe the gene at high level [20]. Some groups have also provided data suggesting that persistent chlamydiae express hsp60, encoding a strongly immunogenic protein, at levels higher than those seen during normal active growth [14,17,18]. Until recently, C. trachomatis was considered an obligate energy parasite on its host [21], with uptake of ATP mediated by bacterially encoded ATP/ADP exchange proteins [22,23]. However, the genome of C. trachomatis includes genes specifying enzymes for glycolysis and a pentose phosphate pathway, as well as genes encoding components of a truncated TCA cycle and an electron transport system [24–26]. Recent work from another group showed that during active growth, transcripts from genes encoding glycolytic and pentose phosphate pathway enzymes in C. trachomatis are produced; this study also showed that the chlamydial glycolytic and pentose phos-

phate pathway enzymes are functional [27]. This suggests that the organism can produce ATP during active infection, in addition to procuring this resource from its host. The chlamydial TCA cycle and electron transport system are also probably functional (see [28] for a review). We showed that during persistent synovial infection of patients with Chlamydia-associated arthritis, C. trachomatis cells display unusual transcriptional characteristics for some genes in addition to omp1 and hsp60 ([29] and see below). Much of this unusual pattern of gene expression was also observed during persistent fallopian tube infection in patients with ectopic pregnancy [11], and in an in vitro model of chlamydial persistence (e.g. [19,30]). Because this organism shows differential expression of some genes during persistence, we asked whether transcription of C. trachomatis genes encoding components of the glycolytic and pentose phosphate pathways, and the TCA cycle and electron transport system, differs between active and persistent infection. We show that mRNA from glycolysis- and pentose phosphate pathway-related genes are highly attenuated during persistence. adt1, encoding one of the two chlamydial ADP/ATP exchange proteins [24–26], is expressed during persistence, as are most genes required for the TCA cycle and the electron transport system.

2. Materials and methods 2.1. Tissue culture An in vitro model system of persistently infected human monocytes [31] was used that reflects all biochemical and metabolic characteristics of the organism known from long-term synovial infection in vivo (e.g. [19,30]). Briefly, normal human peripheral monocytes were isolated from healthy volunteer donors, infected at multiplicity of infection (MOI) 1:1 with C. trachomatis (serovar K), and allowed to grow for 1, 2, 3, and 5 days; approximately 15% of monocytes are infected in such experiments. At designated times, cells were harvested and the cell pellets snap-frozen at –80 °C until processed. As a control for normal chlamydial growth and gene expression, Hep-2 cells were infected at MOI 1:1 with the same serovar and harvested at 2, 4, 8, 11, 16, 18, 24, 36, and 48 h postinfection, as described [29,32]; 35–40% of cells are usually infected. Cell pellets were snap-frozen at –80 °C until processed for nucleic acids. 2.2. Patient samples Synovial tissue samples were procured from patients attending the Arthritis Clinics at the Philadelphia D.V.A. Medical Center and the University of Pennsylvania Hospital, Philadelphia, PA. Synovial biopsies were procured by the Parker–Pearson technique [33], and samples were immediately snap-frozen at –80 °C until used for preparation

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

of nucleic acids. Four patients studied were included simply because they were PCR-positive for chromosomal DNA from C. trachomatis. Three patients in this PCR-positive group had diagnoses of Chlamydia-associated arthritis, and one was diagnosed with undifferentiated oligoarthritis; diagnoses were made according to criteria of the American College of Rheumatology [10]. Two Chlamydia PCRpositive patients studied were female, two were male; overall disease durations ranged from 0.2 to 14 years. One patient with psoriatic arthritis was included as control because he was PCR-negative for the organism. 2.3. Preparation of nucleic acids Total nucleic acids were prepared from infected and uninfected (control) monocyte cell pellets, pellets from infected and uninfected Hep-2 cells, and from synovial tissue samples by homogenization in 65 °C buffered phenol, then extraction in phenol/chloroform (24:1) as described [16,19,29,32]. Pure RNA was made from total nucleic acids by treatment with RNase-free DNaseI (RQ1; Promega Biotech, Madison, WI, USA). Each preparation was confirmed to be DNA-free by PCR targeting the host actin gene in the absence of reverse transcription (RT; [32]). The level of degradation of RNA was assessed both visually on ethidium bromide-stained agarose gels and by RT-PCR targeting the actin cDNA (see below) [19,32,34]. 2.4. Standard RT-PCR assays Each RT reaction was performed using 5 µg total RNA primed with random hexamers, and the MuLV enzyme (Bethesda Research Labs., Bethesda, MD, USA) (e.g. [32]). cDNAs were treated with RNaseA, RNaseT1, and RNase H, extracted with phenol/chloroform several times, and recovered by ethanol precipitation [32,34]. RT-PCR for C. trachomatis transcripts used ≈10% of each cDNA preparation and targeted mRNA from chlamydial genes encoding components of the glycolysis and pentose phosphate pathways, the TCA cycle, cytochrome oxidase, adt1, and each chlamydial σ factor. The mRNA targeted and the primer sequences used are shown in Table 1. Primers were designed from chlamydial genome sequence data ([25]; www.stdgen.lanl. gov), using GeneRunner software (Hasting Software, Hastings, NY). Testing of primers showed that each amplified only the C. trachomatis sequences intended but nothing in the host genome (Fig. 1), and that each primer set amplifies its target sequence with approximately the same efficiency. Conditions for both rounds of the nested reactions were: 4 min/95 °C; then 35 cycles of 40 s/95 °C, 40 s/annealing temperature, 40 s/72 °C; then 10 min/72 °C; annealing temperatures varied among primer sets. Five percent of the first-round reaction product were used as template in the second amplification round. To confirm viability of infecting Chlamydia, RT-PCR targeting primary transcripts from the bacterial rRNA operons was carried out; this system,

15

Table 1 C. trachomatis genes targeted and primer sequences for RT-PCR analyses Chlamydial protein for import of ATP: adt1 (ATP/ADP exchange protein) 5’-cgacaacaacggctagcattagtggtg-3’ (Outer) 5’-aacctaagatgagcatgggcgaaagc-3’ (Outer) 5’-tggggaacttctctttctggacaggagt-3’ (Inner) 5’-cccaaagcagcaactaaacctgtagcat-3’ (Inner) Enzymes involved in glycolysis: pyk (pyruvate kinase) 5’-cccttatgttagagaacgagctcctg-3’ (Outer) 5’-gggctcctaaggcagtttctccagac-3’ (Outer) 5’-atcgctgcttcgttcgtcagatgtaatg-3’ (Inner) 5’-tcggctcgtgtaggaagggggttgcg-3’ (Inner) gap (glyceraldehyde-3-phosphate dehydrogenase) 5’-ctgtcgaagccgggataatattctg-3’ (Outer) 5’-gattaatggttttggacggattgggcg-3’ (Outer) 5’-ggccatcaaccacactctgcgtagctg-3’ (Inner) 5’-gagttaaccatcagcagtttgacccagc-3’ (Inner) pgk (phosphoglycerate kinase) 5’-tggtaagccatgtaggacgcccaaagg-3’ (Outer) 5’-ctccctccaatcctctccagagtcacac-3’ (Outer) 5’-agggaggcgtatttgaagaggcat-3’ (Inner) 5’-taccccatacccccagctaatacgaga-3’ (Inner) Enzymes involved in the pentose phosphate pathway: gnd (6-phosphogluconate dehydrogenase) 5’-gggagatattctcatcgatggggggaa-3’ (Outer) 5’-gccttaatagcgtcctctgcgacccaa-3’ (Outer) 5’-gttgggatgggagtctctggaggggaag-3’ (Inner) 5’-cccaacgcccagtccccttctgtccagc-3’ (Inner) tal (transaldolase) 5’-ccgaaatacgctctctgacaactccaca-3’ (Outer) 5’-gcctaagtaccagtcgatgctgacgg-3’ (Outer) 5’-gcagcgatccaccaatcataaatccgaca-3’ (Inner) 5’-ctgggacttgggaaggtatctgtgct-3’ (Inner) Enzymes involved in electron transport: cydA (cytochrome oxidase subunit I) 5’-ggtagcgagggaatgtttgctttc-3’ (Outer) 5’-ccattactccccataacatgacca-3’ (Outer) 5’-gttgttcttggcgcttggttatctggg-3’ (Inner) 5’-tggccattcgtctttagggaactg-3’ (Inner) cydB (cytochrome oxidase subunit II) 5’-gtgactaggtgttgtaaggttcgcg-3’ (Outer) 5’-ccgatcgctccagatacttcgtatag-3’ (Outer) 5’-aaactacagaagggctgcacgaacg-3’ (Inner) 5’-gagcgccttcaatacgacagcacactcct-3’ (Inner) Tricarboxylic acid cycle enzymes: fumC (fumarate hydratoase) 5’-ggagctcagacgggacgttcgcaagag-3’ (Outer) 5’-cccctaatccacaacgaggtcctgacccc-3’ (Outer) 5’-ggcagacaggaagcggcactcaaag-3’ (Inner) 5’-gtctcccttctcaaatactgaatcacc-3’ (Inner) mdhC (malate dehydrogenase) 5’-ggacggtcattaatcagagcttgcg-3’ (Outer) 5’-gggctttatctggtgtgcgcatggag-3’ (Outer) 5’-gtgatacagccgataaaggtacttctgctc-3’ (Inner) 5’-agaggccccaggaatggagagaagag-3’ (Inner) Sigma factor genes: rpoD 5’-gtgcggatgttacagcgttggatg-3’ (outer) 5’-gggacgaccatcaagaagacca-3’ (outer) 5’-aggctgttgagaagtttgagtatcg-3’ (inner) 5’-agtaaagcccaattcctctccga-3’ (inner) rpsD 5’-cttcaggaatgccttctcatgta-3’ (outer) 5’-ctgatgacatcgtatccggtttcc-3’ (outer) 5’-ccacgcagtgtttatcaaagagc-3’ (inner) 5’-ctagaggaggaaaaccaatggg-3’ (inner) rpoN 5’-gcgccttccctacagtcctac-3’ (outer) 5’-gaaatagcgatcgcatagggagta-3’ (outer) 5’-gcatgcaaccacaacaga-3’ (inner) 5’-ggatgagaactcgaaggatagcca-3’ (inner) .

16

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

region from nucleotide –87 5’ to the first coding base of the 16S rRNA gene through nucleotide +26 within that coding sequence; if non-coding sequences 5’ to the 16S rRNA sequence are absent (i.e. have been processed away) from the primary rRNA transcripts, no amplification product is generated. Primers for this assay are: 5’-gagatagaatgcaggccagt-3’ and 5’-caatctgaaccaagatcaaat-3’ and were designed using software from PE Biosystems (Foster City, CA). Values from primary transcript assays were normalized to processed 16S rRNA levels, as described by others [36], using the primers 5’-gcatccgagtaacgttaaagaagg-3’ and 5’-agacttccgtccattgcgaa-3’ in parallel assays run simultaneously with primary transcript assays. Assays were performed several times, each in triplicate, using a PE Biosystems 7700 sequence detector with the SYBR green method [37]. Data from real time assays were calculated using the v1.7 Sequence Detection Software from PE Biosystems.

3. Results

Fig. 1. Representative RT-PCR analyses targeting C. trachomatis genes whose products are required for glycolysis and the pentose phosphate pathway, as a function of time during active infection. Hep-2 cells were infected with K serovar, harvested at the times indicated post-infection, and RNA was prepared from each such harvested culture, all as given in Materials and methods. RT-PCR analyses were performed using primers given in Table 1 and ref. [16]. Panels are RT-PCR analyses targeting: A, primary rDNA transcripts; B, the adt1 transcript; C, the pyk transcript; D, the gap transcript; E, the pgk transcript; F, the gnd transcript; G, the tal transcript; H, the host actin transcript. In each panel, the C+ lane is a positive PCR control for the relevant primer set, using pure C. trachomatis DNA as amplification template; the C– lane is a negative RT-PCR control, using cDNA from uninfected Hep-2 cells as amplification template. The RT– lane shows the control amplification from PCR assays targeting host actin mRNA in the absence of reverse transcription; a different RNA preparation was assayed in each panel, but RNA from the 2-h-infected cells was not assayed. Input into each assay was normalized to the host actin transcript. Amplification product sizes are: primary rRNA transcripts, 429 bp; adt1, 193 bp, pyk, 317 bp; gap, 171 bp; pgk, 496 bp; gnd, 427 bp; tal, 166 bp; host actin, 510 bp.

including the rationale for using it, have been described [16,19]; primers and amplification conditions for the C. trachomatis omp1 mRNA have also been described (e.g. [16,19]). Amplifications were carried out in a PTC-100 MJ thermocycler (MJ Research, Watertown, MA, USA), and products were visualized on agarose gels via ethidium bromide staining [34]. 2.5. Real-time RT-PCR assays To make an assessment of the relative metabolic rate in actively growing vs. persistent C. trachomatis cells, a realtime RT-PCR assay was designed to define relative levels of primary transcripts from the chlamydial rRNA operons [16,35]. The primers amplify a sequence spanning the

3.1. Expression of C. trachomatis genes encoding glycolysis and pentose phosphate pathway enzymes during active infection Within Hep-2 cells, C. trachomatis cells undergo normal EB to RB reorganization, growth of RB, and normal RB to EB dedifferentiation at the end of the developmental cycle; the cycle requires ≈50 h for completion in this host cell type. As a control for expression of chlamydial glycolysisand pentose phosphate pathway-related genes during persistence (see below), standard RT-PCR was used to assess the time of appearance post-initiation of active infection of Hep-2 cells for transcripts from representative genes in each pathway in C. trachomatis. The genome project identified chlamydial orthologs of genes encoding pyruvate kinase (pyk), glyceraldehyde-3-phosphate dehydrogenase (gap), and phosphoglycerate kinase (pgk), all of which function in glycolysis [25]. That genome also encodes a 6-phosphogluconate dehydrogenase (gnd) and a transaldolase (tal), both pentose phosphate pathway enzymes [25]. adt1 specifies an ATP/ADP exchange protein [22,23,25]. In the representative RT-PCR results shown in Fig. 1, primary rRNA transcript synthesis began within ≈4 h of EB–host cell binding and continued through the end of the developmental cycle (Fig. 1A), indicating metabolic activity. adt1 was expressed over the same time course (Fig. 1B). Expression of C. trachomatis genes encoding enzymes involved in glycolysis (pyk, gap, pgk; Fig. 1C–E) and the pentose phosphate pathway (gnd, tal; Fig. 1F, G) began at ≈11 h post-infection, consistent with data from two previous reports [27,38], and continued through the end of the developmental cycle. These data suggest that in the initial phase of host cell infection, C. trachomatis cells derive the ATP required for EB to RB transformation, and for initial metabolic needs, from the host via ATP/ADP

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

exchange. The data further suggest that during active growth, chlamydial energy requirements are met by both import of ATP from the host and production of this resource via biochemical pathways intrinsic to the organism. 3.2. Expression of C. trachomatis glycolysis- and pentose phosphate pathway-related genes during persistent infection in vitro Normal human monocytes infected in vitro with C. trachomatis reflect the molecular and morphologic characteristics of persistently infecting chlamydiae and their host cells, as established in in vivo disease contexts [19,29,30]. In the representative RT-PCR experiments shown in Fig. 2, primary rRNA transcripts were identifiable from 1 day post-infection onward, indicating chlamydial metabolic activity over the 5-day course of this experiment (Fig. 2A); this is consistent with previously published data [16]. adt1 mRNA was present from 1–5 days post-infection (Fig. 2B),

17

long after a normal life cycle would have been completed in infected Hep-2 cells. C. trachomatis genes encoding enzymes involved in glycolysis and the pentose phosphate pathway were expressed at 1 and 2 days post-infection, but expression had ceased by 3 days, when the characteristics of chlamydial persistence were well established (Fig. 2C–G), i.e. when aberrant morphologic forms of the organism were present and omp1 expression (Fig. 2H) had been attenuated [15,19,29,30]. Thus, during the first 1–2 days of monocyte infection, C. trachomatis cells appeared to undergo the initial portion of the developmental cycle normally, expressing energy transduction-related genes and primary rRNA transcripts, as in Hep-2 cell infection. However, after persistence had been established at 3 days post-infection and thereafter, expression of genes encoding glycolysis and pentose phosphate pathway enzymes was down-regulated; adt1 and primary rRNA transcripts were still produced. This suggests that persistent C. trachomatis cells take ATP from their hosts during persistent infection, but they do not produce enzymes required for ATP synthesis. 3.3. Expression of C. trachomatis TCA cycle- and electron transport-related genes during active and persistent infection in vitro

Fig. 2. Representative RT-PCR analyses targeting C. trachomatis genes whose products are required for glycolysis and the pentose phosphate pathway, as a function of time during persistent infection. Normal human peripheral monocytes were infected with K serovar, harvested at the times indicated post-infection, and RNA was prepared from each such harvested culture, all as given in Materials and methods; RT-PCR analyses were performed using primers given in Table 1 and ref. [16]. Panels are RT-PCR analyses targeting: A, primary rDNA transcripts; B, the adt1 transcript; C, the pyk transcript; D, the gap transcript; E, the pgk transcript; F, the gnd transcript; G, the tal transcript; H, the omp1 transcript; I, the host actin transcript. In each panel, the C+ lane is a positive PCR control for the relevant primer set, using pure C. trachomatis DNA as amplification template; the C– lane is a negative RT-PCR control, using cDNA from uninfected human monocytes as amplification template. The RT– lane shows the control amplification from PCR assays targeting host actin mRNA in the absence of reverse transcription; a different RNA preparation was assayed in each panel. Input into each assay was normalized to the host actin transcript. Amplification product sizes are as given in the legend to Fig. 1; amplicon size for omp1 is 745 bp.

C. trachomatis has no glyoxylate cycle, but the genome encodes components for an abridged TCA cycle and an electron transport system [24–26]. We asked whether, and if so when, chlamydial genes encoding two cytochrome oxidase subunits (cydA, cydB), and two TCA cycle enzymes (malate dehydrogenase, mdhC; fumarate hydratase, fumC) are expressed during active infection of Hep-2 cells. As shown by the representative RT-PCR results in Fig. 3, cydA and cydB were expressed by 8 h post-infection (Fig. 3A, B), slightly earlier than are genes whose products are required for glycolysis and the pentose phosphate pathway. Transcripts from the fumC and mdhC genes appeared at about the same time as did those from glycolysis- and pentose phosphate pathway-related enzymes and continued through 48 h post-infection (Fig. 3C, D). Thus, assuming the products encoded by these genes are functional, C. trachomatis cells use their electron transport system and TCA cycle during active infection, as would be expected. We assessed transcripts from the bacterial genes encoding the two TCA cycle enzymes, and those specifying the cytochrome oxidase subunits, in monocytes infected with C. trachomatis. Representative results from these assays are also shown in Fig. 3. cydB, fumC, and mdhC all were expressed at 1 day post-infection (Fig. 3F–H), and those transcripts continued to be present through day 5. However, we could identify no transcripts from the cydA gene at any time assayed during monocyte infection (Fig. 3E). Thus, while transcripts from the TCA cycle enzymes and one cytochrome oxidase subunit are produced during persistent infection, expression of cydA appears to be abrogated or severely attenuated during persistent chlamydial infection.

18

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

Fig. 3. Representative RT-PCR analyses targeting C. trachomatis genes whose products are required for the TCA cycle and the electron transport system, as a function of time during active infection (panels A–D) and persistent infection (panels E–H). Hep-2 cells and normal human peripheral monocytes were infected with K serovar, harvested at the times indicated post-infection, and RNA was prepared from each such harvested culture, all as given in Materials and methods. RT-PCR analyses were performed using primers given in Table 1 and ref. [16]. Panels show RT-PCR analyses targeting: A, the cydA transcript; B, the cydB transcript; C, the fumC transcript; D, the mdhC transcript, all in infected Hep-2 cells; E, the cydA transcript; F, the cydB transcript; G, the fumC transcript; H, the mdhC transcript, all in infected monocytes. The same RNA/cDNA preparations analyzed in Figs. 1 and 2 were used for the reactions shown here. In each panel, the C+ lane is a positive PCR control for the relevant primer set, using pure C. trachomatis DNA as amplification template; the C– lane is a negative RT-PCR control, using cDNA from uninfected Hep-2 cells or monocytes as amplification template. The RT– lanes show PCR amplifications using the cydB primers but without reverse transcription of the 11-, 18-, 24-, and 36-h Hep-2-derived samples (panels A–D), and the 1-, 2-, 3-, and 5-day monocyte-derived samples (panels E–H). Input into each assay was normalized to the host actin transcript. Amplification product sizes are: cydA, 384 bp; cydB, 147 bp; fumC, 479 bp; mdhC, 215 bp.

3.4. Expression of C. trachomatis glycolysis-, pentose phosphate pathway-, TCA cycle-, and electron transportrelated genes during persistent infection in vivo Previous studies from this group showed that in patients with chronic Chlamydia-associated arthritis, the organisms exist in synovial tissue primarily in the persistent form [6,15]. To determine whether the attenuated expression for gap, pyk, and other genes observed in the in vitro model of persistence also occurs in vivo, we used the standard RT-PCR systems to assess transcripts encoding components of these systems in total RNA from synovial biopsies of individuals with Chlamydia-associated arthritis. Represen-

Fig. 4. Representative RT-PCR analyses of RNA/cDNA prepared from arthritis patient samples and targeting C. trachomatis genes whose products are involved in the glycolytic and pentose phosphate pathways, as well as the TCA cycle and the electron transport system. Synovial tissue samples were procured and processed to RNA, then cDNA, as described in Materials and methods, and analyses were performed using primers given in Table 1 and ref. [16]. Panels A–D show analyses for the transcripts indicated in four patients with Chlamydia-associated arthritis. Panel E shows parallel analyses for a control patient with psoriatic arthritis whose synovial tissue sample was PCR-negative for C. trachomatis DNA. The RT– lane in each panel shows the result of a PCR amplification with pure RNA (no reverse transcription) as template and using the fumC primers. The ACT lane in each panel shows the amplification product from an RT-PCR assay targeting host actin mRNA. Input into each assay was normalized to the host actin transcript. Amplification product sizes are as given in the legends to Figs. 1–3.

tative results from these assays are shown in Fig. 4. All samples from patients with Chlamydia-associated arthritis showed chlamydial primary rRNA transcripts, as expected [16], as well as mRNA from adt1, cydB, fumC, and mdhC (Fig. 4A–D). As in the in vitro system of persistence, however, transcripts from cydA were not identifiable. Moreover, most patient samples lacked transcripts from gap, pyk, tal, and other chlamydial genes encoding products involved in the glycolytic and pentose phosphate pathways, and when transcripts from these genes were present at all they were attenuated. The synovial sample from the control patient with psoriatic arthritis was negative in all assays targeting chlamydial RNAs, as expected (Fig. 4E). Thus, transcript data from persistent synovial chlamydial infection generally reflect those from the in vitro model of persistence. 3.5. Expression of C. trachomatis σ factor genes during persistent infection The sequencing project identified genes specifying three σ factors for C. trachomatis, designated rpoD, rpsD, and rpoN [25]. Because the orthologs of these proteins govern

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

expression of somewhat different groups of genes in Escherichia coli and other organisms, we asked whether abrogated production of one or more chlamydial σ factors during persistence could explain the apparent lack of mRNA for glycolysis- and pentose phosphate pathway-related proteins observed in infected monocytes. The representative RT-PCR analyses presented in Fig. 5 indicate that this cannot be the case, since amplification products from the rpoD (Fig. 5A), rpsD (Fig. 5B), and rpoN (Fig. 5C) cDNA

19

all were apparent from day 1–5 in C. trachomatis-infected monocytes. These results are congruent with data from analyses of RNA from synovial tissue samples of patients with Chlamydia-associated arthritis. Amplicons from each σ factor cDNA were present in the samples analyzed (Fig. 5F, G) but absent in samples from the control patient (Fig. 5E). Consistent with results from previous studies [19,30], mRNA from the chlamydial gene encoding the major outer membrane protein (omp1) was present for about 2 days p.i. in Chlamydia-infected monocytes but absent after persistence was established (Fig. 5D); this messenger was absent in synovial samples from patients with Chlamydiaassociated arthritis, as expected (Fig. 5F, G; [16]). Thus, lack of expression of σ factor coding sequences cannot explain differential gene expression in actively growing vs. persistent chlamydiae, as judged by availability of mRNA from these genes. 3.6. Metabolic rate in actively growing vs. persistent C. trachomatis

Fig. 5. Representative RT-PCR assays targeting mRNA from the C. trachomatis rpoD, rpsD, and rpoN genes as a function of time during persistent infection. Normal human peripheral monocytes were infected with K serovar, harvested at the times indicated post-infection, and RNA was prepared from each such harvested culture and relevant patient synovial samples, as given in Materials and methods; analyses were performed using primers given in Table 1 and ref. [16]. Panels are RT-PCR analyses targeting: A, rpoD transcripts; B, rpsD transcripts; C, rpoN transcripts; D, omp1 transcripts, all in infected monocytes; the same cDNA preparations used in Figs. 1–3 were employed for these analyses. Panel E, similar analyses of RNA/cDNA prepared from synovial tissue of a patient with psoriatic arthritis who was PCR-negative for C. trachomatis; panels F and G, analyses of RNA/cDNA from synovial tissues of each of two patients with Chlamydia-associated arthritis. In panels A–D, the C+ lane is a positive PCR control for the relevant primer set, using pure C. trachomatis DNA as amplification template; the C– lane is a negative RT-PCR control, using cDNA from uninfected human monocytes as amplification template. The lane marked RT– in each panel is a negative control showing a PCR amplification for the relevant gene using RNA from day 1-, 2-, 3-, and 5-infected monocytes (panels A–D, respectively), or targeting the rpoD gene in RNA from patient samples (panels E–G). The ACT shows the amplification product from an RT-PCR assay targeting host actin. Amplification product sizes are: rpoD, 230 bp; rpsD, 149 bp; rpoN, 338 bp.

Data provided above suggest that during active infection, C. trachomatis cells both procure ATP from their hosts and produce it via their own pathways, but that during persistence the organisms rely primarily on host energy resources. In turn, this suggested to us that the overall metabolic rate of persistent chlamydiae might be lower than that of organisms during active growth. The rate of production of primary rDNA transcripts is known to reflect the general metabolic rate in all cells; because these transcripts have no functional half-life [16,35], the relative level of those primary transcripts in any RNA sample should reflect the overall level of metabolism in the cells from which the RNA is prepared. We developed a real-time RT-PCR assay targeting chlamydial primary rRNA transcripts and used it to assess relative levels of these transcripts at two equivalent time points in infected Hep-2 cells and human monocytes. In these experiments, values obtained for primary transcripts from the bacterial rRNA operons were normalized to processed 16S rRNA molecules in each sample, as described by others [36]. Representative results from several such assays are shown in Fig. 6. The data demonstrate that relative levels of primary chlamydial rRNA transcripts are more than ten-fold lower in infected monocytes than in infected Hep-2 cells at both 24 and 48 h post-infection, the only congruent time points available in these analyses. Additional assays demonstrated decline in relative levels of primary chlamydial rRNA transcripts from day 3–5 postinfection in the monocyte system (data not shown). These results support the contention that overall metabolic rate in C. trachomatis is significantly lower during monocyte infection than during active infection of Hep-2 cells, and that this attenuated metabolic rate is established even before persistence is fully developed in this in vitro model system.

20

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

Fig. 6. Representative real-time RT-PCR assays targeting primary rRNA transcripts from C. trachomatis in infected Hep-2 cells and human monocytes at 24 and 48 h post-infection. Hep-2 cells and normal human peripheral monocytes were infected with K serovar, harvested at the times indicated post-infection, and RNA, then cDNA, was prepared from each such harvested culture, all as given in Materials and methods; analyses were performed using primers given in that section. Assays of these transcripts from infected Hep-2 cells and monocytes were run simultaneously in triplicate and normalized to levels of processed chlamydial 16S rRNA, as described previously [36]. Several repeats of the same determination on these and independently prepared RNA preparations gave virtually identical results.

4. Discussion Infection of the urogenital system by C. trachomatis is associated with development of reactive arthritis, and it was initially thought that synovial pathogenesis in this disease was a sterile immune-mediated process [6]. Studies from this and other laboratories, however, have indicated that chlamydiae infecting the synovium over long periods are viable and metabolically active (see [6,7] for a review). Moreover, evidence suggests that the metabolic characteristics of persistent synovial chlamydiae are not identical to those of actively growing organisms [14,15,17,18,29,30]. The results reported here indicate that during the initial phase of active infection, C. trachomatis cells derive the energy required for EB to RB transformation, and for metabolism, from host cells via ATP/ADP exchange. During active growth of RB, the organisms obtain ATP not only from the host, but also via their own glycolytic and pentose phosphate pathways. Our data also suggest that during the first phase of monocyte infection, before the full establishment of persistence, C. trachomatis cells utilize both ATP/ADP exchange and their own pathways to support metabolic needs, even though overall metabolic rate in the organisms is relatively low. However, once persistence has been established the primary source of ATP appears to be the host. That is, mRNA for glycolytic and pentose phosphate pathway enzymes are absent or severely attenuated in this context, suggesting that these systems are virtually, if not completely, shut down during persistence. We have not quantitated transcript levels from the chlamydial glycolysis and pentose phosphate pathway genes assessed here, nor have we extensively defined the

sensitivity of the standard RT-PCR assays used. However, it seems unlikely that the pyk, gap, gnd, and other transcripts sought in the monocyte system were missed at late times after infection. In this model, amplification products from all glycolytic and pentose phosphate pathway cDNA were easily identifiable at day 1 post-infection. Moreover, the adt1 cDNA product was visible on gels after the first round of PCR at all times assessed through day 5 post-infection. Primary rRNA transcripts were also easily identified in this system at all times assayed, and the RT-PCR system employed to assess them is not nested. That no amplification products from pyk, gnd, and other related cDNA were found even following use of nested primer sets after 2 days post-infection indicates that transcripts from these genes are at extremely low levels or absent in persistent chlamydiae. Elegant work published by Iliffe and McClarty [27] showed that the chlamydial gene products related to glycolysis and the pentose phosphate pathway whose transcripts are assessed here are functional; moreover, our RT-PCR results regarding production of those transcripts during active growth of chlamydiae are consistent with those of that and one other study [38]. Thus, C. trachomatis cells appear to be only partial energy parasites on their hosts during active growth, but during persistent infection the organisms seem to be fully dependent on the host for ATP. This suggested that the overall metabolic rate in persistent chlamydiae might be lower than that observed in actively growing organisms, since availability of energy resources required to support metabolism is circumscribed. The realtime analyses targeting relative primary chlamydial rRNA transcript levels support this contention, since those transcript levels were several-fold lower in infected monocytes than in infected Hep-2 cells at the two congruent time points assayed post-infection. This suggests that while the initial phase of monocyte infection is relatively normal in terms of bacterial transcripts produced, and relatively normal therefore in terms of EB to RB development, some host–parasite interaction initiated during the first 24 h of infection attenuates bacterial metabolism, possibly influencing the elicitation of persistence. It will be of interest to elucidate what that interaction entails, since obviation of the interaction(s) may allow effective more antibiotic treatment of persistent C. trachomatis infections. Current thinking holds that C. trachomatis cells do possess a functional, if abridged, TCA cycle [28], and the results presented here indicate that genes encoding products required for assembly of that system are expressed during active infection. It is not surprising that these same genes are expressed during persistence, since the TCA cycle is important for generation of NADH2, as well as precursor molecules for intermediary metabolism. Similarly, C. trachomatis cells possess an electron transport system [25,28] whose components are produced during active growth. We were surprised, however, that expression of the C. trachomatis cydA gene is down-regulated during monocyte infection, while expression of cydB is not. Both peptides are

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

required for a functional electron transport component, and we assume that this system is necessary during persistence to generate reducing equivalents in the bacterial cell. We do not understand why expression of one but not both cytochrome oxidase subunits is attenuated under these circumstances. As with the assays targeting pyk, gnd and other transcripts discussed above, control data indicate that the RT-PCR primers targeting cydA and cydB are of roughly equal efficiency in amplification of their target cDNA. Moreover, in the Hep-2 system we had no difficulty showing amplicons from cydA and cydB cDNA, even after the first of the two nested reactions. Thus, more study, and probably additional assay systems, will be required to elucidate this issue. Results of analyses of cydA, cydB, fumC, and mdhC transcripts, as well as those encoding the glycolysis- and pentose phosphate pathway-related enzymes, in synovial tissue samples from chronic arthritis patients PCR-positive for C. trachomatis generally reflect those from the in vitro model of chlamydial persistence. It is of interest that we did find low levels of gap and pgk mRNA in a minority of patient RNA preparations. It is not clear why some but not all glycolysis-related messengers were present in a few such samples, but this may be related to clinical aspects of the disease or to individual patient characteristics, e.g. disease duration, possible reinfection with C. trachomatis, etc. It may also be the case that, while the monocyte system provides uniformly persistent organisms, patient-derived materials include C. trachomatis cells in both the active and persistent states. More study will be required to elucidate this discrepancy between the monocyte model and in vivo synovial infection by the organism. Previous work has shown that the MOMP and the mRNA from the gene encoding it (omp1) are attenuated during persistent infection, both in in vitro models of persistence and in vivo in synovial tissue from patients with Chlamydiaassociated arthritis [11,16–18]. The results presented here indicate that expression of C. trachomatis genes encoding glycolytic and pentose phosphate pathway enzymes is similarly attenuated during persistence. In observations presented elsewhere, we further demonstrate that transcription of several genes whose products are required for cytokinesis in this same organism is strongly downregulated during persistent infection of human monocytes and in relevant patient samples [29]. We do not understand in detail the mechanism(s) by which transcript initiation is governed in C. trachomatis during EB to RB reorganization, active growth, or RB to EB dedifferentiation. Importantly for the present topic, we do not understand the mechanism(s) controlling the selective transcriptional downregulation of the energy transduction system-related C. trachomatis genes during persistence. Data presented here indicate, though, that mRNA from each of the three σ factor genes is present in persistent cells both in vivo and in the in vitro system, suggesting that lack of the proteins they encode cannot explain differential expression of glycolysis-

21

and pentose phosphate pathway-related genes. A recent study showed that rpoD and rpsD are expressed early, and rpoN relatively late, in the C. trachomatis developmental cycle [36], and that the level of expression of these genes is not identical. It is possible that extremely low levels of one or more σ factors contributes to abrogated/attenuated expression of pyk and the other genes studied here. We are using real-time RT-PCR assays to assess this possibility. Further, we have performed an initial analysis of the 5’ portion of the coding sequence and 5’ flanking regions for omp1, pyk, gnd, and other chlamydial genes whose transcription is down-regulated in persistence, but we could identify no common DNA sequence which might function as a repressor binding site. We also could identify no sequence 5’ to genes expressed during persistence which might function in a trans-activation system for transcript initiation. Thus, more study is required to define the details of selective transcriptional regulation in persistent vs. actively growing C. trachomatis. Other in vitro models of chlamydial persistence have been described in the literature, in addition to the one used here. For example, HeLa or other cell types infected with C. trachomatis and treated with IFN-γ display the aberrant morphology characteristic of both synovial samples from patients with Chlamydia-associated arthritis and the monocyte model (see [14] for a review). In collaboration with another group, we are assessing transcripts from the chlamydial metabolic system genes assayed here in IFN-γtreated C. trachomatis-infected cells to determine whether mRNA from these genes is present. It should be of significant value to compare the results of such analyses among the various in vitro models of chlamydial persistence.

Acknowledgements This work was supported by NIH grants AR-42541 (A.P.H.), AI-44493 (J.A.W.-H.), and AR-47186 (H.C.G.), grants from the Department of Veterans Affairs Medical Research Service (H.R.S., A.P.H.), a grant from the Arthritis Foundation (J.A.W.-H.), and a grant from the German Ministry of Technology (01 GI 9950 to L.K.).

References [1] J. Schachter, J.T. Grayston, Epidemiology of human chlamydial infections, in: R.S. Stephens, G.I. Byrne, G. Christiansen, I.N. Clarke, J.T. Grayston, R.G. Rank, G.L. Ridgway, P. Saikku, J. Schachter, W.E. Stamm (Eds.), International Chlamydia Symposium, San Francisco, Chlamydial Infections, 1998, pp. 3–10. [2] B.A. Greendale, S.T. Haas, K. Holbrook, B. Walsh, J. Schachter, R.S. Phillips, The relationship of Chlamydia trachomatis infection and male infertility, Am. J. Pub. Health 83 (1993) 996–1001. [3] W. Cates, J.H. Wasserheit, Genital chlamydial infections: epidemiology and reproductive sequelae, Am. J. Obstet. Gynecol. 164 (1991) 1771–1781.

22

H.C. Gérard et al. / Microbes and Infection 4 (2002) 13–22

[4] H. Thejls, J. Gnarpe, O. Lundkvist, G. Heimer, G. Larsson, A. Victor, Diagnosis and prevalence of persistent Chlamydia infection in infertile women by tissue culture, direct antigen detection, and serology, Fertil. Steril. 55 (1991) 304–310. [5] R. Anttila, P. Saikku, P. Koskela, A. Bloigu, J. Dillner, I. Ikaheimo, et al., Serotypes of Chlamydia trachomatis and risk for development of cervical squamous cell carcinoma, J. Am. Med. Assoc. 285 (2001) 47–51. [6] R.D. Inman, J.A. Whittum-Hudson, H.R. Schumacher, A.P. Hudson, Chlamydia-associated arthritis, Curr. Opin. Rheumatol. 12 (2000) 254–262. [7] L. Köhler, H. Zeidler, A.P. Hudson, Etiologic agents in reactive arthritis: their molecular biology and phagocyte-host interactions, Baillière’s Clin. Rheumatol. 12 (1998) 589–609. [8] H.R. Schumacher, Chlamydia-associated arthritis, Isr. Med. Assoc. J. 2 (2000) 532–535. [9] D. Taylor-Robinson, C.B. Gilroy, B.J. Thomas, A.C.S. Keat, Detection of Chlamydia trachomatis DNA in joints of reactive arthritis patients by polymerase chain reaction, Lancet 340 (1992) 81–82. [10] H.R. Schumacher, J.H. Klippel, W.J. Koopman (Eds.), Primer on the Rheumatic Diseases, 10th ed., Arthritis Foundation, Atlanta, 1993. [11] H.C. Gérard, P.J. Branigan, G.R. Balsara, C. Heath, S.S. Minassian, A.P. Hudson, Viability of Chlamydia trachomatis in fallopian tubes of patients with ectopic pregnancy, Fertil. Steril. 70 (1998) 945–948. [12] T. Hackstadt, Cell biology, in: R.S. Stephens (Ed.), Chlamydia: Intracellular Biology, Pathogenesis, and Immunity, American Society for Microbiology, Washington DC, 1999, pp. 101–138. [13] P. Wyrick, Cell biology of Chlamydia infection: a journey in the host epithelial cell by the ultimate cellular microbiologist, in: R.S. Stephens, G.I. Byrne, G. Christiansen, et al (Eds.), International Chlamydia Symposium, San Francisco, Chlamydia Infections, 1998, pp. 69–78. [14] W.L. Beatty, R.P. Morrison, G.I. Byrne, Persistent Chlamydiae: from cell culture to a paradigm for chlamydial pathogenesis, Microbiol. Rev. 58 (1994) 686–699. [15] R. Nanagara, F. Li, A.M. Beutler, A.P. Hudson, H.R. Schumacher, Alteration of Chlamydia trachomatis biological behavior in synovial membranes: suppression of surface antigen production in reactive arthritis and Reiter’s syndrome, Arthritis Rheum. 38 (1995) 1410–1417. [16] H.C. Gérard, P.J. Branigan, H.R. Schumacher, A.P. Hudson, Synovial Chlamydia trachomatis in patients with reactive arthritis/Reiter’s syndrome are viable but show aberrant gene expression, J. Rheumatol. 25 (1998) 734–742. [17] W.L. Beatty, G.I. Byrne, R.P. Morrison, Morphologic and antigenic characterization of interferon gamma-mediated persistent Chlamydia trachomatis infection in vitro, Proc. Natl. Acad. Sci (USA) 90 (1993) 3998–4002. [18] M.L. Jones, J.S. Hill Gaston, J.H. Pearce, Induction of abnormal Chlamydia trachomatis by exposure to interferon-γ or amino acid deprivation and comparative antigenic analysis, Microb. Pathogen. 30 (2001) 299–309. [19] H.C. Gérard, L. Köhler, P.J. Branigan, H. Zeidler, H.R. Schumacher, A.P. Hudson, Viability and gene expression in Chlamydia trachomatis during persistent infection of cultured human monocytes, Med. Microbiol. Immunol. 187 (1998) 115–120. [20] R.S. Stephens, Molecular genetics of Chlamydia, in: W.R. Bowie, H.D. Caldwell, M.A. Chernesky, et al (Eds.), Chlamydial Infections, Societa Editrice Esculapio, Bologna, 1994, pp. 377–386. [21] J.W. Moulder, Interaction of chlamydiae and host cells in vitro, Microbiol. Rev. 55 (1991) 143–190.

[22] T.P. Hatch, E. Al-Hossainy, J.A. Silverman, Adenine nucleotide and lysine transport in Chlamydia psittaci, J. Bacteriol. 150 (1982) 662–667. [23] T.P. Hatch, Metabolism of Chlamydia, in: A.L. Baron (Ed.), Microbiology of Chlamydia, CRC Press, Boca Raton, FL, 1988, pp. 97–109. [24] S. Kalman, W. Mitchell, R. Marathe, C. Lammel, J. Fan, R.W. Hyman, et al., Comparative genomes of Chlamydia pneumoniae and C. trachomatis, Nat. Genet. 21 (1999) 385–389. [25] R.S. Stephens, S. Kalman, C. Lammel, J. Fan, R. Marathe, L. Aravind, et al., Genome sequence of an obligate intracellular pathogen of humans: Chlamydia trachomatis, Science 282 (1998) 754–759. [26] T.D. Read, R.C. Brunham, C. Shen, S.R. Gill, J.F. Heidelberg, O. White, et al., Genome sequences of Chlamydia trachomatis MoPn and Chlamydia pneumoniae AR39, Nucl. Acids Res. 28 (2000) 1397–1406. [27] E.R. Iliffe-Lee, G. McClarty, Glucose metabolism in Chlamydia trachomatis: the ‘energy parasite’ hypothesis revisited, Mol. Microbiol. 33 (1999) 177–187. [28] G. McClarty, Chlamydial metabolism as inferred from the complete genome sequence, in: R.S. Stephens (Ed.), Chlamydia: Intracellular Biology, Pathogenesis, and Immunity, American Society for Microbiology, Washington DC, 1999, pp. 69–100. [29] H.C. Gérard, B. Krauβe-Opatz, D. Rudy, J.P. Rao, H. Zeidler, H.R. Schumacher, et al., Expression of Chlamydia trachomatis genes required for DNA synthesis and cell division in active vs. persistent infection, Mol. Microbiol. 41 (2001) 731–741. [30] L. Köhler, E. Nettelnbreker, A.P. Hudson, N. Ott, H.C. Gérard, P.J. Branigan, H.R. Schumacher, J. Drommer, H. Zeidler, Ultrastructural and molecular analysis of the persistence of Chlamydia trachomatis (serovar K) in human monocytes, Microb. Pathogen. 22 (1997) 133–142. [31] C.D. Rothermel, J. Schachter, P. Lavrich, E.C. Lipsitz, T. Francus, Chlamydia trachomatis-induced production of interleukin-1 by human monocytes, Infect. Immun. 57 (1989) 2705–2711. [32] H.C. Gérard, J.A. Whittum-Hudson, A.P. Hudson, Genes required for assembly and function of the protein synthetic system in Chlamydia trachomatis are expressed early in elementary to reticulate body transformation, Mol. Gen. Genet. 255 (1997) 637–642. [33] H.R. Schumacher, J.B. Kulka, Needle biopsy of the synovial membrane: experience with the Parker–Pearson technique, N. Engl. J. Med. 286 (1972) 416–419. [34] J. Harwood (Ed.), Basic DNA and RNA Protocols, Humana Press, Totowa, NJ, 1996. [35] M. Nomura, R. Gourse, G. Baughman, Regulation of the synthesis of ribosomes and ribosomal components, Annu. Rev. Biochem. 53 (1984) 75–117. [36] S.A. Mathews, K.M. Volp, P. Timms, Development of a quantitative gene expression assay for Chlamydia trachomatis identified temporal expression of σ factors, FEBS Lett. 458 (1999) 354–358. [37] M. Hiratsuka, Y. Agatsuma, M. Mizugaki, Rapid detection of CYP2C9*3 alleles by real time fluorescence PCR based on SYBR green, Mol. Genet. Metabol. 68 (1999) 357–362. [38] E.I. Shaw, C.A. Dooley, E.R. Fischer, M.A. Scidmore, K.A. Fields, T. Hackstadt, Three temporal classes of gene expression during the Chlamydia trachomatis developmental cycle, Mol. Microbiol. 37 (2000) 913–925.