Differential Effects of DNA Supercoiling on Chlamydia Early Promoters Correlate with Expression Patterns in Midcycle

Abstract

Changes in DNA supercoiling levels during the chlamydial developmental cycle have been proposed as a global mechanism to upregulate midcycle genes, but the effects on early genes are not known. We examined the promoters for 10 Chlamydia trachomatis early genes and found that they could be separated into two subsets based on their responses to DNA supercoiling in vitro. Furthermore, the type of supercoiling response correlated with the in vivo expression pattern for each early gene. One subset of seven early genes had promoters that were transcribed in a supercoiling-insensitive manner over the physiologic range of supercoiling levels that have been measured in Chlamydia. In vivo transcripts for these genes were detected at similar levels during early-stage and midstage times. In contrast, a second subset, represented in our study by three early genes, had supercoiling-dependent promoters that were transcribed at higher levels from more-supercoiled templates, which is the response observed for midcycle genes. Genes in this subset were expressed at higher levels at midstage times than at early times in vivo. We propose that this second subset represents a novel class of chlamydial developmental genes with features of both early and midcycle genes. We hypothesize that expression of these supercoiling-dependent early genes is upregulated by increased chlamydial supercoiling levels in midcycle via their supercoiling-responsive promoters in a manner similar to that for midcycle genes. Thus, we propose that DNA supercoiling is utilized in Chlamydia as a general mechanism to regulate genes in the midstage of the developmental cycle and not just midcycle genes.

INTRODUCTION

Members of the genus Chlamydia are obligate intracellular bacteria that cause a variety of diseases by infecting human cells. Chlamydia trachomatis is the most common cause of bacterial sexually transmitted disease (5) and the leading cause of infectious blindness (22). A related organism, Chlamydia pneumoniae, causes atypical pneumonia (3). Even though chlamydial infections have a range of clinical manifestations, the main features of the intracellular infection are similar for all chlamydial species and strains. The hallmark of the chlamydial infection is an unusual developmental cycle in which there is conversion between two specialized forms within a eukaryotic cell (1). The infectious elementary body (EB) initiates the infection by binding and entering a host cell, where it resides in the cytoplasm within a membrane-bound vacuole called the chlamydial inclusion. The most prominent event during this early stage of the infection is the conversion of the EB into the larger reticulate body (RB) by 6 to 8 h postinfection (hpi). The RB is the metabolically active form of chlamydiae that replicates repeatedly by binary fission during midcycle. Late in the developmental cycle, RBs convert back into EBs, which are released to infect new host cells.

A number of processes affecting both chlamydiae and the host cell have been described during the early stage of the intracellular chlamydial infection. EB-to-RB conversion is marked by reduction of the disulfide-cross-linked outer membrane coat that provides the EB with protection from the extracellular environment (13). In addition, there is decondensation of chromatin (12), which frees DNA from constraints that may prevent its transcription and replication. The inclusion membrane is formed upon entry, providing a barrier between chlamydiae and the host cytoplasm. This membrane contains chlamydial inclusion membrane proteins, some of which are synthesized early in the infection (25). Proteins encoded by early genes are also important for subverting and usurping a wide range of host processes in order to establish the chlamydial inclusion (7).

During the early stage of the developmental cycle, there is selective transcription of a subset of genes that have been called early genes. For example, Belland et al. performed transcriptional profiling with a DNA microarray and showed that ∼200 of the 894 genes in C. trachomatis are expressed by 3 hpi (2). Examples of these early genes include incD, which encodes an inclusion membrane protein (27), and groEL and dnaK, which are stress response genes that encode heat shock proteins (31, 35). Other early genes include rRNA, which encodes rRNA for the translational machinery (9), and euo, which is a DNA-binding protein (38). Five of the seven transcriptional units on the C. trachomatis plasmid are also transcribed early (24). For example, the early plasmid gene pgp3 (pORF3) (8) encodes an immunogenic protein that is secreted into the host cell cytosol (6, 16).

It is not known how early genes are selectively transcribed at the beginning of the developmental cycle while later temporal classes of genes are not yet expressed. Early, midcycle, and late chlamydial genes are each transcribed by the major form of chlamydial RNA polymerase, σ⁶⁶ RNA polymerase (4, 19, 30, 31), a finding which suggests that there must be specific mechanisms to differentially regulate the temporal classes of genes. Midcycle genes have been proposed to be positively regulated by higher levels of negative DNA supercoiling in midcycle (19). Unpublished results suggest that promoters for late genes may be negatively regulated by the transcriptional repressor EUO, providing a mechanism to prevent their premature expression (25a). A second subset of late genes is regulated by a specialized mechanism in which they are transcribed by an alternative RNA polymerase containing σ²⁸ (36, 37). In contrast, mechanisms for the regulation of early gene expression have not been described.

DNA supercoiling is a potential mechanism for the selective expression of early genes. In bacteria, DNA supercoiling has been shown to modulate promoter activity directly by altering DNA structure and melting energy and indirectly by affecting the binding of transcription factors (14). Microarray studies in Escherichia coli and Haemophilus influenzae have demonstrated that changes in negative DNA supercoiling can coordinately regulate the expression of several hundred genes (11, 21). Chlamydial DNA supercoiling levels have been shown to be regulated during the developmental cycle, with low levels at early times increasing to a peak in midcycle (19). One early promoter, for the type III secretion (T3S) gene CT863, has been found to be transcribed in a supercoiling-dependent manner (4), but the response of other early promoters to supercoiling is not known.

To examine if early genes can be regulated by DNA supercoiling, we tested promoters for 10 early genes to determine if they are transcribed in a supercoiling-dependent manner. Seven of 10 early promoters that we tested were not sensitive to changes in supercoiling levels. However, we found a subset of early genes that have supercoiling-dependent promoters that resemble promoters for midcycle genes in their supercoiling response. We propose that this subset of early genes with supercoiling-dependent promoters represents a novel class of chlamydial genes with features of both early and midcycle genes.

MATERIALS AND METHODS

Construction of in vitro transcription plasmids.

The promoter regions for each of the genes listed in Table 1 were amplified by PCR from Chlamydia trachomatis genomic or plasmid DNA. Each PCR product was subcloned into plasmid pMT1125, upstream of a promoterless, G-less cassette transcription template, as previously described (33). This G-less cassette is a synthetic transcription template that does not encode any guanosine residues. In the absence of added GTP, only transcripts that do not contain guanosine residues will accumulate. Transcription start sites for the genes on the chlamydial plasmid were based on the study by Ricci et al. (24). The PCR primers, the specific Chlamydia serovar or strain used as the source of DNA, the size of each plasmid, and the names of the resulting transcription plasmids generated in this study are listed in Table 1. The plasmid genes were named according to the recent convention, with the open reading frame (ORF) immediately downstream of the plasmid origin of replication designated pORF1 (16, 32). The transcription plasmids for dnaK (pMT1161) (33) and groEL (pMT1178) (34) were generated in previous studies.

Table 1.

Plasmids and cloning primers generated in this study

Plasmid	Gene	Primer sequences	Region	Serovar	Size (kb)
pMT1606	incD	5′-GATCGAATTCGTTGCTCCGAGTTAAACGGAGC 5′-AGAAGGTAACATTTTCCCAACAA	−150 to +5	L2	3.2
pMT1482	rRNA	5′-CACGAATTCGCAAAAAAAAAGAATTTCGACAAAAAAATCGAAGAG 5′-AAAGTTACATCTTATGCTACCCCCTATTTTTTATGCACC	−205 to +5	L2	3.2
pMT1484	euo	5′-CCCGAATTCTTAGTGGTTTCGATTGGGTTTTTA 5′-AGAAGGTAACATTTTCCCAACAA	−176 to +5	D	3.2
pMT1547	pORF1	5′-GATCGAATTCGGTAGACTTTGCAACTCTTGGTGG 5′-TGGAGCTCTAAGATTTTAAGAAATTTTTCAAC	−119 to +5	L2	3.1
pMT1610	pORF3	5′-GATCGAATTCGAAGCTTTTCATGCGTTTCCAATAGGATTCT 5′-TTAATAACCTGATTATTTCACTAATCAGGACATTTTAC	−148 to +5	L2	3.2
pMT1554	pORF5	5′-GATCGAATTCGCTTGGTTGGTTATCTAGAAGAAAACGC 5′-TTTTTATTTTGAGCTTTAAATAAATTAGGTTTTTAGTTTCAAG	−200 to +7	L2	3.3
pMT1611	pORF6	5′-GATCGAATTCGAAAGCGGTGTGGTATGGGTTAATGC 5′-TTTTTCTAAATATAAAACCTATAAGAAAAATCCAATAAAAATTGTTTAAG	−151 to +5	L2	3.2
pMT1612	pORF7	5′-GATCGAATTCGTAACTTTCTTAAATTCTCCCTTAGAACTCTACC 5′-TTTTTTAGCTTTAAACTGTTATCCTCTAATTTTTCAAGAACAG	−147 to +3	L2	3.2

Transcription plasmid topoisomers.

Topoisomers for each transcription plasmid were generated as previously described (4). Briefly, CsCl gradient-purified DNA for each plasmid was incubated for 3 h at 37°C with 5 U of wheat germ topoisomerase I (Promega) and ethidium bromide. Topoisomers of different superhelical densities were produced by varying the ethidium bromide concentration from 0 to 60 μM. After incubation, the ethidium bromide was removed by phenol-chloroform and chloroform extractions, and plasmid DNA was recovered by ethanol precipitation and resuspended in RNase-free Tris-EDTA buffer. The topoisomers were resolved on a 1.4% agarose gel with ethidium bromide, and the average difference in linking number (ΔLK) was determined by the band-counting method of Keller et al. (15). Superhelical density (σ) was calculated using the formula σ = −10.5(ΔLK/n), where n is the total number of base pairs in the plasmid.

Chlamydia and cell culture.

C. trachomatis serovar L2 (strain L2/434/Bu) was grown in L929 mouse fibroblast cells at 37°C with 5% CO₂ in RPMI 1640 (Cellgro, Manassas, VA) supplemented with 5% heat-inactivated fetal bovine serum (FBS) (Omega Scientific, Tarzana, CA).

Purification of chlamydial RNA polymerase.

L929 cells were infected with C. trachomatis at a multiplicity of infection (MOI) of 3 and harvested at 20 hpi. Chlamydial RNA polymerase was partially purified using heparin-agarose chromatography as previously described (30).

Supercoiling sensitivity transcription assay.

For each promoter, a set of topoisomers of various superhelical densities was used as the transcription template in individual in vitro transcription reactions. Transcription experiments were performed as described previously (19) with minor changes. One microliter of purified chlamydial RNA polymerase with 25 nM concentrations of each topoisomer plasmid DNA was used in each transcription reaction (30). Radiolabeled transcripts were resolved on a urea-polyacrylamide gel, exposed to a phosphor screen, and quantified with Quantity One software (Bio-Rad, Hercules, CA). For each promoter, the relative promoter activity was calculated by normalizing the amount of activity obtained from each topoisomer to the maximal promoter activity (defined as 100%) observed over the range of superhelical densities tested. For each topoisomer, the mean and standard deviation of the relative promoter activity are reported. Transcriptions for each set of topoisomers were repeated as three independent experiments.

Quantitative RT-PCR.

Chlamydial DNA was isolated at 1, 3, 8, 16, and 24 hpi from L929 cells infected with C. trachomatis at an MOI of 3. Total DNA was isolated using the QIAamp tissue kit (Qiagen, Valencia, CA) per the manufacturer's protocol. The number of chlamydial genomes in each DNA sample was determined by amplifying and measuring the rRNA gene by quantitative reverse transcriptase PCR (RT-PCR) (20). In parallel, total RNA was extracted with RNA Stat-60 (Tel-Test, Inc., Friendswood, TX) per the manufacturer's protocols. Residual DNA was removed by resuspending the RNA in 100 μl DNase I buffer (10 mM Tris-Cl, 50 mM NaCl, 10 mM MgCl₂, 1 mM dithiothreitol [DTT]) and incubating with 20 U RNase-free DNase I (Roche) for 30 min at 37°C, followed by purification using RNeasy RNA columns (Qiagen) as described previously (17). cDNA was synthesized and amplified by RT-PCR using the iScript one-step RT-PCR kit (Bio-Rad) on a Bio-Rad iCycler iQ per the manufacturer's protocols. The primer sets used for the RT-PCRs are listed in Table 2. The amount of RT-PCR product was quantified by converting mean critical threshold values to ng of DNA using standard curves generated with the Bio-Rad iCycler iQ Optical System software (Bio-Rad). The calculated number of transcripts was normalized to the genome copy number so that the relative expression of each gene could be compared between time points.

Table 2.

Primer sets for quantitative RT-PCR

Primer name	Primer sequence
incD-Forward	5′-CGCATAGCATTCAGCAAGAG
incD-Reverse	5′-TGCCGCTACAACAGAAACAG
rRNA-Forward	5′-GGAGAAAAGGGAATTTCACG
rRNA-Reverse	5′-TCCACATCAAGTATGCATCG
euo-Forward	5′-TTATTCCGTGGGACAAGTGG
euo-Reverse	5′-TGCAAGACTTTTCCCTTTGC
groEL-Forward	5′-GCGAAAGAAGTTGAGCTTGC
groEL-Reverse	5′-ACCTCGTTTGAGGTCCATTG
dnaK-Forward	5′-ATTGAGCGAACCAAACAACC
dnaK-Reverse	5′-TAGCAGCTCCAATCGCTACA
pORF1-Forward	5′-ACTTGGCCCAATTTTTAGGG
pORF1-Reverse	5′-TGATGCAGGAATTAGGTCCA
pgk-Forward	5′-GCGCAACTTTCTCCTGGTAG
pgk-Reverse	5′-TGTGGCACCCGATATACAGA

RESULTS

A subset of early promoters is responsive to changes in superhelical density.

To examine whether early genes are regulated by DNA supercoiling levels, we tested promoters for 10 early genes with an in vitro supercoiling sensitivity transcription assay. The only other early promoter that has been shown to be transcriptionally active in vitro is the CT863 promoter, which we previously showed to be transcribed in a supercoiling-dependent manner (4). We cloned the promoter of each gene on a plasmid-based transcription template. We then generated topoisomers of each plasmid to cover the physiological range of superhelical densities from −0.03 to −0.07 that have been measured at different stages of the chlamydial developmental cycle (19). A representative gel depicting how the superhelical density of each topoisomer was calculated is shown in Fig. 1A. By measuring the amount of transcript produced from the same promoter on different topoisomers, we were able to assay the effect of DNA supercoiling on promoter activity.

Fig 1.

Supercoiling sensitivity transcription assays of early promoters that are not responsive to DNA supercoiling levels. For each gene, the promoter was cloned on a transcription plasmid. Then a series of plasmid topoisomers was generated over a range of superhelical densities (σ) from 0 (completely relaxed) to −0.07 (negatively supercoiled) by varying the ethidium bromide (EtBr) concentration used in the in vitro topoisomerase reaction. (A) Representative 1.4% agarose gel showing the resolution of pORF1 topoisomers in 0.04 μg/ml EtBr. Additional gels were run at different EtBr concentrations, as no single gel could resolve the entire series of topoisomers. For each individual topoisomer, the concentration of EtBr used in the topoisomerase reaction, the average difference in linking number (ΔLK) as determined by counting the ladder of topoisomer bands for the predominant band (arrow), and the superhelical density (σ) calculated from the ΔLK are shown. More details are provided in Materials and Methods. (B) Each topoisomer was transcribed using chlamydial σ⁶⁶ RNA polymerase in an individual in vitro reaction to measure the promoter activity at a specific superhelical density. The difference between the maximal and minimal promoter activities over this span of superhelical densities was calculated as a fold change and is reported as the average of three independent experiments. The seven supercoiling-insensitive early promoters are grouped according to the location of their gene on the C. trachomatis chromosome or plasmid.

We found that early promoters divided into two groups depending on their responses to DNA supercoiling. In the first group, the promoters showed a less-than-2-fold change in activity when transcribed over a range of superhelical densities (Fig. 1B). These early genes with supercoiling-insensitive promoters were located on the chlamydial chromosome (incD, rRNA, and euo) or the chlamydial plasmid (pORF3, pORF5, pORF6, and pORF7). In contrast, we found that three other early promoters showed a differential response to altered DNA supercoiling levels. DNA supercoiling had the greatest effect on the groEL promoter, which showed a 200-fold range in activity depending on the level of superhelicity (Fig. 2). The dnaK promoter and the early plasmid promoter pORF1 showed 20-fold and 3.8-fold responses to DNA supercoiling levels, respectively.

Fig 2.

A subset of early promoters is transcribed in a supercoiling-dependent manner. Supercoiling sensitivity transcription assay showing that three early promoters were transcribed by chlamydial σ⁶⁶ RNA polymerase at higher levels from topoisomers of increased superhelical density (σ). The fold change in promoter activities over the range of superhelical densities tested is shown for each promoter and is the average of three independent experiments. groEL and dnaK are present on the chlamydial chromosome, while pORF1 is located on the chlamydial plasmid.

The response of supercoiling-dependent early promoters resembles the pattern for midcycle promoters.

We used a graphical analysis to compare these different responses of early promoters to supercoiling levels. For each promoter, we calculated the relative promoter activity by defining their maximal level of transcription at any superhelical density as 100% and normalizing other transcription levels to this value. We then graphed the effect of superhelical density on the activity of each promoter. For the seven supercoiling-insensitive early promoters, the relative promoter activity remained above 50% for all superhelical densities tested (Fig. 3A). This pattern was similar to the published supercoiling-insensitive responses of six chlamydial late promoters, including the omcAB promoter (4, 19). In contrast, the three supercoiling-dependent early promoters resembled the promoter of the CT863 operon, which encodes type III secretion genes that are expressed early in the chlamydial developmental cycle (4). These four supercoiling-sensitive early promoters were each transcribed at low levels from a relaxed template but at higher levels with increasing superhelical density (Fig. 3B). Intriguingly, this supercoiling pattern is similar to the supercoiling-dependent response characteristic of promoters for chlamydial midcycle genes, such as pgk (19). Midcycle genes are upregulated in the midstage of the developmental cycle when chlamydial supercoiling levels are highest, and thus, this resemblance raised the question of whether early genes with supercoiling-responsive promoters are also regulated by DNA supercoiling in midcycle.

Fig 3.

Graphs depicting the relationships between promoter activity and superhelical density for early promoters. (A) Graphs for seven supercoiling-insensitive early promoters were divided into two groups for clarity of presentation (chromosomal genes on the left, plasmid genes in the middle graph). For comparison, the graph on the right shows that the supercoiling-insensitive promoter of the late operon omcAB has a similar response (adapted from reference 19). (B) Graphs of three supercoiling-sensitive early promoters (left). For comparison, the supercoiling-dependent responses for the promoters of the early T3S gene CT863 (adapted from reference 4) and the midcycle gene pgk (adapted from reference 19) are shown in the middle and right graphs, respectively. Relative promoter activity was calculated as a percentage of the maximal promoter activity (defined as 100%) over the range of superhelical densities tested. The dashed horizontal line on each graph indicates a relative promoter activity of 50%. All reactions were performed as three independent experiments, and the error bars represent standard deviations.

Correlation between the supercoiling response of early promoters and the in vivo expression pattern in midcycle.

We examined the in vivo expression patterns of early genes in midcycle and compared them to the pattern for a known midcycle gene. We infected L929 cells with C. trachomatis and measured transcript levels for individual genes by quantitative RT-PCR at 1, 3, 8, 16, and 24 hpi. Expression of three early genes that have supercoiling-insensitive promoters showed little change from 1 to 24 hpi, with a maximum 2.2-fold difference in transcript levels over this time period (Fig. 4A). In contrast, three early genes with supercoiling-dependent promoters were transcribed at higher levels after the early stage of the developmental cycle (compare transcript levels at 16 and 24 hpi with levels at 1, 3, and 8 hpi). We measured increases in transcript levels of 8.8-fold for groEL, 7.8-fold for dnaK, and 20.9-fold for pORF1 at 24 hpi (Fig. 4B). This in vivo expression pattern was similar to the upregulation of the known midcycle gene pgk at 16 and 24 hpi (Fig. 4B). These results demonstrate that these early genes with supercoiling-dependent promoters are upregulated in midcycle. They thus appear to have a distinctive expression pattern with early expression followed by increased transcription in midcycle. By analogy to midcycle genes, we propose that this subset of early genes is upregulated in the midstage of the developmental cycle by increased chlamydial supercoiling via their supercoiling-dependent promoters.

Fig 4.

Differences in expression patterns for the two subsets of early genes. In vivo expression levels for each gene were measured by quantitative RT-PCR assays. (A) Transcript levels remained fairly constant from early stage to midstage of the developmental cycle (1 to 24 hpi) for early genes with supercoiling-insensitive promoters. (B) Transcript levels for early genes with supercoiling-sensitive promoters increased at 16 to 24 hpi. For comparison, the expression pattern for the midcycle gene pgk was also measured. All reactions were performed as three independent experiments, and the error bars represent standard deviations. hpi, hours postinfection.

DISCUSSION

In this study, we found that early genes can be separated into two subsets on the basis of their response to DNA supercoiling and their in vivo expression pattern in midcycle. One subset of early genes has supercoiling-insensitive promoters and appears to be transcribed at similar levels during the early stage and midstage of the developmental cycle. In contrast, a second subset of early genes has supercoiling-dependent promoters, and expression is increased during midcycle when chlamydial supercoiling levels have been shown to be highest (19).

We propose that the supercoiling-dependent subset of early genes is a novel class of chlamydial developmental genes with features of both early and midcycle genes. Like other early genes, these genes are selectively transcribed at the beginning of the developmental cycle when midcycle and late genes are not yet expressed. However, like midcycle genes, these genes are upregulated in the midstage of the developmental cycle and have promoters that are transcribed at higher levels from more-supercoiled templates. One explanation for this amalgam of features is that this new class is regulated by mechanisms that control expression of early as well as midcycle genes. In other words, these genes may be selectively expressed as early genes by a mechanism that has still not been elucidated and then upregulated by increased DNA supercoiling together with the midcycle genes.

Another possibility is that these early genes with supercoiling-responsive promoters (groEL, dnaK, and pORF1) are actually midcycle genes that have been misclassified. Midcycle genes have been identified in RT-PCR and microarray studies as genes whose transcripts were not detected until the midstage of the developmental cycle (2, 28). The groEL, dnaK, and pORF1 transcripts have each been detected at early times in multiple studies (2, 18, 28), and thus the expression of these genes at early times is not in doubt. However, it is possible that the defining feature of a midcycle gene is not the absence of expression at early times followed by transcription in midcycle but rather the upregulation of expression in midcycle. Thus, groEL, dnaK, and pORF1 may represent highly expressed midcycle genes that are transcribed at lower but detectable levels at early times before being upregulated in midcycle. In support of this possibility, C. trachomatis dnaK has a strong promoter that is highly transcribed in vitro and whose promoter sequence is an almost perfect match to those of the consensus chlamydial σ⁶⁶ and E. coli σ⁷⁰ promoters (26, 29).

Our finding that many early promoters are supercoiling insensitive provides support for a model to explain how early genes are selectively expressed at the beginning of the developmental cycle. At these early times, chlamydial supercoiling levels are low (19), which should not prevent the expression of early genes with supercoiling-independent promoters. In contrast, representative midcycle genes have supercoiling-dependent promoters that require higher supercoiling levels for transcriptional activity (4, 19), and thus, these low supercoiling levels may not be sufficient for expression of most midcycle genes, with the possible exception of the supercoiling-dependent early genes. Although late genes also have supercoiling-independent promoters (4, 19) and, therefore, the potential to be expressed at early times, unpublished results suggest that late promoters are negatively regulated by a transcriptional repressor called EUO (25a). Thus, we propose that the selective transcription of early genes at early times is due to their constitutive expression at low supercoiling levels while midcycle and late-cycle genes are regulated to prevent their expression at early times. We do not propose that DNA supercoiling is utilized as an active mechanism to upregulate the expression of early genes at early times because we did not identify early promoters with higher activity at low superhelical density.

Our study is also the first to examine the transcriptional regulation of genes on the chlamydial plasmid. We tested the promoters for the seven transcriptional units on the C. trachomatis plasmid, including the antisense gene pAS (10, 23). The promoters for each of the five early genes/operons on the plasmid could be assigned to one of two subsets that we observed for early genes. Specifically, four transcriptional units (the pORF3/pORF4 operon, pORF5, pORF6, and the pORF7/pORF8 operon) had supercoiling-insensitive promoters and similar expression levels at early stage and midstage of the developmental cycle, while one early gene (pORF1) with a supercoiling promoter was upregulated in midcycle. We found that the remaining two late transcriptional units on the plasmid (pORF2 and pAS) had supercoiling-insensitive promoters (data not shown), a pattern which is consistent with the supercoiling responses of other late chlamydial genes. Thus, it appears that genes on the chlamydial plasmid are regulated by the same mechanisms that control the temporal expression of chromosomal genes.

In summary, our results indicate a nuanced role for DNA supercoiling in the regulation of early chlamydial genes. We propose that a newly identified subset of early genes is upregulated by increased chlamydial supercoiling, but this increased expression occurs in midcycle. Thus, we propose an expanded role for DNA supercoiling in upregulating midcycle genes, as has been previously proposed (19), as well as supercoiling-dependent early genes during the midstage of chlamydial development. We also propose that supercoiling has a role in the selective expression of early genes at the beginning of the intracellular infection because the low supercoiling levels at early times inhibit the expression of midcycle genes while allowing early genes with supercoiling-independent promoters to be transcribed. These findings demonstrate how functional analyses of promoters can supplement the descriptive information provided from chlamydial expression studies and add to our understanding of temporal gene regulation in Chlamydia.