Chlamydia trachomatis Infection Induces Replication of Latent HHV-6

Human herpesvirus-6 (HHV-6) exists in latent form either as a nuclear episome or integrated into human chromosomes in more than 90% of healthy individuals without causing clinical symptoms. Immunosuppression and stress conditions can reactivate HHV-6 replication, associated with clinical complications and even death. We have previously shown that co-infection of Chlamydia trachomatis and HHV-6 promotes chlamydial persistence and increases viral uptake in an in vitro cell culture model. Here we investigated C. trachomatis-induced HHV-6 activation in cell lines and fresh blood samples from patients having Chromosomally integrated HHV-6 (CiHHV-6). We observed activation of latent HHV-6 DNA replication in CiHHV-6 cell lines and fresh blood cells without formation of viral particles. Interestingly, we detected HHV-6 DNA in blood as well as cervical swabs from C. trachomatis-infected women. Low virus titers correlated with high C. trachomatis load and vice versa, demonstrating a potentially significant interaction of these pathogens in blood cells and in the cervix of infected patients. Our data suggest a thus far underestimated interference of HHV-6 and C. trachomatis with a likely impact on the disease outcome as consequence of co-infection.


Introduction
Human herpes virus 6 (HHV-6) is a ubiquitous pathogen with more than 90% seroprevalence in healthy adults. HHV-6 is commonly associated with self-limited childhood illness roseola infantum, and rarely with more severe syndromes. Chronic viral replication mostly occurs in salivary glands whereas HHV-6 persists in a latent stage in various tissue types including monocytes and bone marrow progenitor cells. Latent HHV-6 resides in the host cell as nuclear episome [1,2] or integrated into telomeric ends of host cell chromosomes. The later form of latent HHV-6 can be vertically transmitted in the germ line [3,4,5]. Chromosomal integration of HHV-6 is described to be the most common form of latent HHV-6 [3]. After acquiring latency, HHV-6 persists in a dormant state with minimal viral transcription or translation in human host cells without any clinical complications. However, under various physiological conditions, latent HHV-6 is reactivated to form infectious viral particles (for further details see reviews [6,7,8]). In immune compromised patients, reactivation of viral activity may lead to severe limbic encephalitis [9]. HHV-6 reactivation has been correlated with relapsing and secondary progressive multiple sclerosis (MS) [10,11,12,13] and Chronic Fatigue Syndrome (CFS) [11,14]. Reactivation of latent HHV-6 can also be achieved by treatment with chemical stimulators like Trichostatin A (TSA), 12-O-tetradecanoyl-phorbol-13-acetate (TPA), calcium ionophore A23187 and hydrocortisone leading to viral replication in an extra-chromosomal state, viral gene transcription and formation of new infectious viral particles [3,5,15].
The Gram-negative, obligate intracellular bacterium Chlamydia trachomatis is the leading infectious cause of blindness and the third most frequent sexually transmitted infection worldwide. Epidemiological studies have connected HHVs and Chlamydia in several conditions. Herpes simplex virus 2 (HSV-2) is associated with C. trachomatis in endometritis and acute salpingitis [16]. HHV-6 DNA is frequently detected in different grades of cervical lesions [17,18]. C. trachomatis infection has also been separately linked to cervical cancer [19] but it is unclear, whether both these pathogens contribute to the development of cervical cancer. Chlamydia pneumoniae, a highly prevalent pathogen with up to 80% serum antibody positivity in adults, is the cause of pneumonia in humans, but has also been associated with chronic diseases like atherosclerosis, progressive neurological disorders and lung cancer. In CFS patients, HHV-6 has been identified together with C. pneumoniae [20]. Although still controversially discussed, support for a role of C. pneumoniae and HHV-6 co-infection in the development and progression of MS has been provided [21,22]. Although C. trachomatis and C. pneumoniae differ in their tissue tropism, they have a similar infection physiology. In a previous study, we have shown that co-infection with HHV-6 influences the development of infectious progeny of both, C. trachomatis and C. pneumoniae [23]. Moreover, different herpes viruses like herpes simplex virus (HSV)-1 and HSV-2 and human cytomegalovirus influence chlamydial infectivity in a similar way [23].
Here, we studied the influence of chlamydial infection on latent HHV-6 and its activation. Our data show that Chlamydia infection induces replication of chromosomally integrated HHV-6 in both immortalized human cells as well as in freshly isolated blood cells.
To get a better insight into such interaction in vivo, we investigated cervical swabs from patients with C. trachomatis-infection for the HHV-6 and C. trachomatis load by qPCR. Our data provide statistically significant correlations supporting an interaction between HHV-6 and C. trachomatis.

Cell Lines, Patients and Samples
For the study of latent CiHHV-6 activation, established CiHHV-6 cell lines like HSB-ML (a tetraploid T-cell line derived from HBS-2 cells with 2-5 copies of CiHHV-6), JL-LCL and PL-LCL (cell lines established from CiHHV-6 individuals with 1-2 copies of HHV-6A per cell) were kindly provided by the HHV-6 foundation, USA (www.hhv-6foundation.org/, www. bioworldantibodies.com). Fresh blood samples from 5 individuals with CiHHV-6 were also provided by the HHV-6 foundation, USA. Viral load was re-verified by qPCR, which confirmed the CiHHV-6 status of these cells. Cervical swabs and blood samples were available from 73 female patients attending the outpatient Sexually Transmitted Infection (STI) clinic of Helsinki University Central Hospital, Helsinki, Finland. The patients were recruited to the study because of symptoms suggestive of C. trachomatis infection, or because notified by an infected partner. Cervical swabs were collected in 2 different sample buffers. Cervical swabs collected in Gen-Probe Aptima specimen collection media (Gen-Probe Inc., USA) were tested for C. trachomatis ribosomal RNA by Aptima Combo 2 Assay (Gen-Probe). The cervical samples collected in Copan UTM-RT media (Copan Diagnostics Inc., USA) were used for culture, then stored at -80uC and later used for total DNA extraction and detection of C. trachomatis and HHV-6 viral load. C. trachomatis antibodies were determined by micro-IF test.

Ethics Statement
The CiHHV-6 blood samples were collected under written informed consent under IRB# CI001-HHV6 approved by The Essex Institutional Review Board Committee. Informed written consent was also obtained from all participants of the Chlamy-Trans Study, and the ethics committee approved the consent procedure. A study permit was obtained from Helsinki University Central Hospital, Laboratory Division (HUSLAB

Peripheral Blood Mononuclear Cell (PBMC) Isolation
Fresh PBMCs were first separated from whole blood using Histopaq1077 solution (Sigma, St. Louis, MO, USA) according to manufacturer's instructions. Briefly, freshly collected blood was diluted 2 times with PBS and layered carefully onto 1 volume of Histopaq1077 solution without mixing the solutions. PBMCs were collected after centrifugation at 6006g for 30 min at room temperature. Cells were washed 3 times with 25 volumes of PBS and kept in RPMI 1640 media with 5% fetal bovine serum (FBS) at 37uC and 5% CO 2 . Isolated PBMCs were cultured on a plastic culture dish (NUNC, USA) for 48 h to separate monocytes, which later differentiate to macrophages. Separated adherent monocytes were further cultured for 5-7 days in presence of 50 ng/ml of macrophage colony stimulating factor (M-CSF) (Sigma, USA) to allow complete differentiation into macrophages. Both the adherent macrophage fraction and the PBMC fractions were infected with C. trachomatis separately.

Chlamydia Trachomatis Infection
CiHHV-6 containing cells and PBMCs (both monocyte derived macrophages and leukocytes) were infected with C. trachomatis at an MOI of 5 either in an Eppendorf tube or on a culture plate and were centrifuged for 1 h at 37uC at 12006g as described previously [23]. Cells were washed 3 times with PBS and grown in fresh RPMI 1640 media with 5% FBS.

DNA Extraction from Cervical Swabs
Total DNA was extracted from the cervical swabs using MagNA Pure LC Total Nucleic Acid Isolation Kit (Roche Diagnostics, Germany) with minor modifications. Briefly, 1 ml of the cervical swab sample stored in Copan UTM-RT media was mixed with 100 ml of magnetic beads. To this, 450 ml of lysis/ Binding binding buffer was added together with 150 ml of Proteinase-K and 1 ml of RNAse A. Samples were mixed thoroughly and incubated for 5 min at room temperature. Beads were separated by a magnet and subsequently 850 ml of wash buffer I was added. Beads were again separated and then washed Table 1. Detection of Chlamydia and HHV-6 in cervical swab samples and total blood. Negative (n = 25) n = 7 n = 1 n = 5 n = 12 Positive (n = 48) n = 11 n = 11 n = 13 n = 13 73 cervical swab and total blood samples were tested by qPCR for Chlamydia and HHV-6. doi:10.1371/journal.pone.0061400.t001 with 450 ml of wash buffer II and then followed by another wash in 450 ml of wash buffer III. Bead bound DNA was eluted in 100 ml of elution buffer at 70uC. Removal of RNA from the DNA preparations was confirmed by analyzing random samples with DNAse treatment and reverse transcription PCR. Amount of DNA was quantified using NanoDrop 1000 Spectrophotometer (Thermo Scientific, Germany) and was stored at 220uC until further analysis.

DNA Extraction From Blood
Total DNA was extracted from whole blood samples using QIAamp DNA Blood Mini Kit (Qiagen, Germany) following manufacturer's protocol.

Preparation of Plasmids for Standard Curve Analysis
HHV-6A (U1102), HHV-6B (Z29) [23], U94 ORF, C. trachomatis type III secretion chaperon LcrH/SycD, PI15 ORF were amplified using primer pairs described in Table 1. Amplified PCR products were agarose gel purified and were cloned into TOPO pCR 2.1 vector (Life Technologies, Germany). Cloned DNA vectors were transformed into E. coli DH5a and propagated to isolate ample amount of vectors for qPCR analysis. Cloned ORFs were verified by DNA sequencing.

Quantitative Real time PCR (qPCR)
For qPCR, PerfeCTa TM qPCR SuperMix (Quanta Biosciences) was used and PCR amplifications were done on a StepOnePlus TM real time PCR platform (Applied Biosciences) using manufacturer's protocol and SYBRH Green chemistry. qPCR reactions were performed using 100 ng of DNA per sample per well. Amplified data were analyzed using StepOne TM Software v2.1. Primer details are provided in table 1.

Statistical Analysis
For categorically scaled measures, results are presented as numbers and frequencies. In case of metric data, mean, median and the standard deviation is given unless otherwise stated. To assess the relationship between two categorical measures, the chisquare test was used. When comparing a metric measure between two independent samples the normality assumption was checked first. If normality was given, the t-test was performed, otherwise the Wilcoxon test was used. For the comparison of more than two independent groups the Kruskal-Wallis test was conducted. The binomial test was done for the comparison of two proportions. All tests were two-tailed and differences were considered statistically significant if P,0.05. The software package used was IBM SPSS for Windows, version 19.0.

Activation of HHV-6
To study HHV-6 reactivation, CiHHV-6 cell lines and fresh blood samples from individuals with CiHHV-6, were infected with C. trachomatis serovar L2 at an MOI of 1-5. Chlamydial infection was monitored by phase contrast microscopy. In parallel, CiHHV-6 cells were cultured in RPMI medium 1640 supplemented with 10% FCS together with 1610 26 M hydrocortisone or 80 ng/ml Trichostatin A (TSA) (Sigma, USA) for 3-5 days, which are known to induce HHV-6 reactivation. After 2-3 days of infection, cells were grown in fresh media supplemented with 1 mg/ml of doxycycline, which allowed Chlamydia infected cells to recover. As HSB-2 and Molt-3 cells (T-cell derived cells) allows productive HHV-6 infection, 1610 4 HSB-2 or Molt-3 cells were added to 10 6 PBMCs and co-cultured for 10-14 days in order to recover the reactivated viruses. Reactivation of HHV-6 was monitored by analysis of cytopathic effects, Gardella gel analysis and detection of viral genomes by qPCR.

Gardella Gel Analysis
Gardella gel analysis allows the detection of large episomal DNA as it is formed as a consequence of HHV-6 reactivation. This method was used to separate episomal HHV-6 DNA from cellular DNA as described by Gardella et al. [24] with minor modifications. Briefly, 2.5610 6 CiHHV-6 cells co-cultured with HSB-2 or Molt-3 cells were directly lysed inside each well of a vertical polyacrylamide gel in a sample buffer containing 15% Ficol, 2 Kunitz units of RNAase type1A (Sigma, USA) per ml, and 0.01% bromophenol blue in TBE buffer and were layered with lysis buffer containing 5% Ficoll, 1% SDS, 1 mg of pronase (Sigma, USA) per ml and 0.05% xylene cyanol green. For lysis of isolated chlamydial elementary bodies (EB), samples were lysed in above mentioned lysis buffer containing lysozyme (Sigma, USA) at a final concentration of 7,500 U/ml. Gels were run at 4uC overnight and then stained in ethidium bromide for visualization of DNA. Subsequently, gels were used for southern transfer and hybridization.

Southern Hybridization
Gardella gels were incubated in 0.125 M HCl for 10 min and then in DNA denaturation buffer (1.5 M NaCl, 0.5 M NaOH) for 30 min. DNA was transferred to Nylon-XL membrane (Amersham Hybond-XL, GE life sciences) by capillary transfer using denaturation buffer for transfer. After transfer, membrane was washed with neutralization buffer (3 M NaCl, 0.3 M Tri-sodium citrate, 0.5 M Tris, pH 8.0) for 15 min and was subsequently preincubated in hybridization buffer (GE life sciences, USA). After 1 h of pre-incubation, either random primed C. trachomatis probes (GE life sciences, USA) or end labeled HHV-6 oligo probes (Table  S1) were added to the hybridization buffer and incubated overnight at 65uC or at 42uC, respectively. Membranes were washed and exposed overnight to phosphor storage screens (Fujifilm), which were then scanned by Typhoon 9200 imager (GE Healthcare). PCR amplified Ctr LcrH/SycD gene product of 136 bp was used for random priming and subsequent as probe.
SEM of three independent experiments. (F) qPCR analysis of viral DNA in freshly isolated blood cells from CiHHV-6 individuals after 5 days of C. trachomatis infection. Fresh PBMCs were isolated from total blood and were infected with C. trachomatis for 2 days at an MOI of 5 after which cells were washed thoroughly and grown in presence of fresh media with 1 mg/ml of doxycycline for another 3 days. In parallel, the same numbers of PBMCs were cultured in presence of 80 ng/ml of TSA for 5 days. Total leukocytes were separated from monocyte-derived macrophages and were infected with C. trachomatis in parallel. Heat inactivated C. trachomatis (hiCtr) were added to total PBMCs in a separate well and were included as negative control. Data represent the mean 6 SEM of results from blood samples of 5 different CiHHV-6 individuals. (G) qPCR analysis of viral DNA in total leukocytes and monocyte-derived macrophages from CiHHV-6 individuals after 5 days of C. trachomatis infection. Total leukocytes were separated from monocyte-derived macrophages and were infected with C. trachomatis for 2 days at an MOI of 5 after which cells were washed thoroughly and grown in presence of fresh media with 1 mg/ml of doxycycline for another 3 days. Data represent the mean 6 SEM of three independent experiments performed at the same time from one CiHHV-6 individual. doi:10.1371/journal.pone.0061400.g001

Neutral-neutral 2D DNA Electrophoresis
Separation of DNA in neutral-neutral 2D gels was performed as previously described [25]. Briefly, DNA was first separated on 0.4% agarose at low voltage in 0.56 TBE, and the gel was stained with 0.3 mg/ml ethidium bromide. The lane was cut and placed on a clean gel support at 90u to the direction of the first electrophoresis. The positions of the size markers for the first-dimension gel electrophoresis are shown at the top. (E) Gardella gel analysis and subsequent southern hybridization showing extra-chromosomal HHV-6 DNA after C. trachomatis infection. HSB-ML and JL-LCL cells were infected with C. trachomatis (MOI 5) for 2 days after which cells were washed thoroughly and grown in presence of fresh media with 1 mg/ml of doxycycline for another 7 days. Fresh HSB-2 cells were added to these cells and were co-cultured for another 2 weeks. 2.5610 6 of these cells were loaded directly onto a Gardella gel and subsequently stained with ethidium bromide and afterwards transferred to a nylon membrane. HSB-2, HHV-6-infected HSB-2 for productive virus infection (HSB-2+HHV-6), C. trachomatis elementary bodies (Ctr EB) and C. trachomatis-infected HSB-ML cells, which were subsequently not co-cultured with HSB-2 cells (HSB-ML+Ctr) were loaded as controls. Nylon membrane was hybridized first with a HHV-6-specific probe and subsequently stripped and used again for hybridization with C. trachomatis-specific probe. Extra-chromosomal linear HHV-6 DNA is marked with *and possible nicked circular viral DNA is marked with a rectangle. Chlamydial DNA did not run in the gel due to the size of genomic DNA and hence C. trachomatis-specific hybridization was observed only in the wells of lane 4 and 5. A clear band (probably the chlamydial plasmid) in lane 3, which was observed after ethidium bromide staining was not detected by C. trachomatis-specific probe. Instead some additional bands were detected in lane 4 with C. trachomatis-specific probes. (F) Transmission electron microscopy pictures showing appearance of particle like condensed DNA structures (marked by black arrowhead) in CiHHV-6 cell nuclei after Chlamydia infection. doi:10.1371/journal.pone.0061400.g002 Samples with HHV-6 DNA below detection limit (5 copies/10 3 cells) were arbitrarily specified as 4 copies/10 3 cells. Similarly, samples with less than 100 copies/10 3 cells were arbitrarily specified as 50 copies/10 3 cells. Sensitivity of PCR was marked as detection limits. doi:10.1371/journal.pone.0061400.g003 electrophoresis, cast with 1.1% agarose containing 0.3 mg/ml ethidium bromide and run in 0.56 TBE. The first dimension was run overnight at 1 V/cm, and the second at 4 V/cm for 3.5 h, both at room temperature. Southern blot analyses were performed as described above.

C. trachomatis Infection Induces CiHHV-6 DNA Replication and Formation of Extra-chromosomal HHV-6 DNA
To test the effect of C. trachomatis-infection on latent chromosomally integrated HHV-6 (CiHHV-6) DNA replication, we infected HSB-ML cells carrying 1-2 copies of HHV-6A per cell with C. trachomatis (MOI 1) ( Figure 1A). We also infected cervical epithelial HeLa cells carrying latent HHV-6A (100 copies of HHV-6A per 1000 cells) ( Figure 1B) [23]. Total DNA extracted at 2 days post infection (dpi) and 9 dpi was used for quantitative PCR (qPCR) [23] against HHV-6 U94 ORF (Table S1). In parallel, all the cell lines were treated with hydrocortisone (1 mM) or TSA (80 ng/ml), which were previously shown to induce latent HHV-6 DNA replication [3]. DMSO solvent controls were included in separate wells/cultures and all experiments were repeated at least 5 times. We observed increased amount of viral DNA in both cell types at 2 days after C. trachomatis infection ( Figure 1A, 1B). Hydrocortisone treatment did not induce any HHV-6 replication in our system. HHV-6 DNA replication induced by C. trachomatis infection was comparable to that observed upon treatment with the chemical inducers TSA. The increase in the amount of viral DNA, however, was not statistically significant because of high level of variation within the experiments. Interestingly, at 9 dpi the amount of viral DNA was significantly higher than in unstimulated cells (p,0.05) ( Figure 1C). Similar experiments were carried out using 2 other CiHHV-6 cell lines (JL-LCL and PL-LCL), and a 1.8-2.4 fold increase in the amount of viral DNA at 9 dpi was observed ( Figure 1D, 1E).
To test the effect of C. trachomatis on replication of latent HHV-6 in fresh PBMCs, we isolated PBMCs from 5 healthy individuals having CiHHV-6. The PBMC were either directly infected with C. trachomatis for 3 days or used for separation of monocytes from total leukocytes (see Material and Methods), which were further differentiated into macrophages. Total leukocytes and monocytederived macrophages were then separately infected with C. trachomatis (MOI 1-5) for 3 days. Heat inactivated C. trachomatis (hiCtr) was used as a (negative) control. Total DNA was extracted at 3 dpi and analyzed by qPCR. We observed a minor increase (1.3-fold) in the amount of viral DNA in the presence of TSA whereas a 1.7-fold increase was observed when total PBMCs were infected with C. trachomatis ( Figure 1F). In order to check cell type specificity for C. trachomatis infection and subsequent viral activation, we analyzed the amount of viral DNA in total leukocytes as well as in monocyte-derived macrophages. We observed up to 3.6-fold increase in the amount of viral DNA in monocyte-derived macrophages whereas only a 1.6-fold increase was observed in total leukocytes. The increase in the amount of viral DNA in response to C. trachomatis infection was consistent in all the 5 samples tested and was statistically significant (p,0.05).
Since not enough PBMCs from the five CiHHV-6 individuals were available to perform control experiments in non-infected total leukocytes and monocyte-derived macrophages, we repeated these important control experiments with PBMCs from one CiHHV-6 individual ( Figure 1G) and thereby corroborated the results shown in figure 1F.

Increased Viral DNA Replication is not Sufficient to Produce Infectious Viral Particles
To test the long-term influence of C. trachomatis infection on HHV-6 DNA replication, we repeated infections in HSB-ML cells and kept the cells in culture till 35 days in presence of 1 mg/ml of Doxycycline. The amount of viral DNA was increased up to 173fold within 35 days of first C. trachomatis infection (Figure 2A), suggesting a strong induction of HHV-6 DNA replication under these conditions. Southern hybridization of total DNA from these cells verified the replication and presence of extra-chromosomal HHV-6 DNA in 35 dpi HSB-ML cells, which represents the activated HHV-6 DNA ( Figure 2B). We repeated this experiment in fresh PBMCs from one of the CiHHV-6 individuals and observed appearance of full-length extra chromosomal HHV-6 DNA ( Figure 2C) after 16 days of C. trachomatis infection. Further long-term culture was not possible with isolated PBMCs because of their limited life span in cell culture. We detected small viral DNA (smaller than 48 kb) in mock-infected PBMCs without having any full-length HHV-6 DNA. To differentiate extra-chromosomal HHV-6 DNA between mock-infected and C. trachomatis-infected PBMCs, we performed neutral-neutral 2D DNA electrophoresis ( Figure 2D), which differentiates DNA in 2 dimensions, by mass and by conformation. We observed clear differences between both the DNAs thus confirming our hypothesis that C. trachomatis infection induces CiHHV-6 DNA replication and formation of extra-chromosomal HHV-6 DNA. Hence our results show that C. trachomatis infection promotes formation of extra-chromosomal HHV-6 DNA and its replication in various cell types.
It has been previously shown that chemically induced latent HHV-6 activation may or may not lead to formation of new viral particles [3,26,27,28]. To test whether C. trachomatis-induced HHV-6 DNA replication results in the formation of new infectious viral particles, latently infected JL-LCL or HSB-ML cells were cocultivated with HSB-2 cells, which are permissive for the production of HHV-6 viral particles. HSB-ML and JL-LCL were infected with C. trachomatis (MOI 1-5) for 2-3 days, and then grown in doxycycline (1 mg/ml) containing media for additional 7 days to avoid lysis of the cells due to chlamydial replication. These cells were then co-cultured with fresh HSB-2 cells in order to allow HHV-6 infection and propagation. After 2 weeks of co-culture, 2.5610 6 cells were collected and used for Gardella gel analysis (to look for extra-chromosomal viral genomes). The co-culture experiments were repeated at least 3 times. We detected low amounts of both linear as well as circular HHV-6 DNA molecules in all experiments ( Figure 2E, S1). However, we could not detect any infectious viral particles in the co-cultured cells by transmission electron microscopy (data not shown). Interestingly, we observed slow migrating, slightly higher molecular weight HHV-6 extra-chromosomal DNA in co-cultured HSB-ML and JL-LCL cells pointing to the possibility of additional confirmations for HHV-6 DNA or presence of additional viral or non-viral sequences than the known linear double stranded DNA ( Figure 2E, S1). In addition, we observed either nicked circular or concatemeric HHV-6 DNA in all the co-cultured cells ( Figure 2E, S1), suggesting circularization of HHV-6 DNA. In line with these observations, electron microscopic analysis of C.  Table 5. Two-way Chi-square test to study the association between Chlamydia and HHV-6 load in cervical smear of all patients. trachomatis-infected HSB-ML, JL-LCL and PL-LCL cells could not show presence of any viral particles in these cells. Only condensed particle like DNA structures were detected in their nuclei ( Figure 2F). Thus, our data shows that C. trachomatis infection reactivates replication of HHV-6 in CiHHV-6 cells without leading to infectious virus particle formation.
qPCR Assays for Quantification of HHV-6 and C. trachomatis DNA Our previous [23] as well as the present study using cell culture models indicated that C. trachomatis infection has an effect on active as well as latent HHV-6 infection. To investigate the nature of C. trachomatis and HHV-6 co-infection in a clinically relevant situation, we established methods to detect these microbes in patient samples. Latent HHV-6 DNA can be detected in various body parts [29,30] including cervical epithelial tissues [31,32] where they influence human papillomavirus (HPV) gene regulation [2]. C. trachomatis is a mucosal pathogen and its DNA can be detected in various grades of cervical neoplasia as well as in asymptomatic patients [33,34]. Hence, the possibility of an interaction between Chlamydia and HHV-6 in cervical epithelial cells is plausible.
We therefore studied cervical swabs from 73 female patients and determined C. trachomatis load and HHV-6 viral load. As the number of cervical epithelial cells in such samples can be quite low, we developed a highly efficient SYBR green-based quantitative real time PCR (qPCR) assay to detect low amounts of microbial DNA in small samples.
For preparation of standard curves, we cloned a PCR-amplified full-length HHV-6A (U1102 strain), HHV-6B (Z29) U94 (1473 bp) ORF and a 136 bp fragment of C. trachomatis type III secretion chaperon LcrH/SycD into TOPO pCR 2.1 (Life Technologies, Germany) vector backbone. Primers for HHV-6 U94 ORF, one of the well-studied transcripts essential for HHV-6 latency, were selected for qPCR and standard curve analysis [23,35]. As the type III secretion chaperon of C. trachomatis is essential for chlamydial survival and infection, it was selected for qPCR as well as standard curve analysis. For normalization of microbial DNA to cellular DNA, we selected PI15 ORF (NM_015886), a gene that exists as a single copy per haploid genome (or 2 copies per human cell) and cloned the full-length PI15 (777 bp) into the same vector backbone. The recombinant plasmid standards were diluted serially 10-fold starting from 1 plasmid copy per ml to 10 6 copies per ml. The standards were then amplified in triplicates (1-10 6 copies per well) in a single run and the data collected were used to generate a linear-log regression plot. To evaluate inter-and intra-experimental variation, we repeated these tests at least 5 times and for 5 different time points and calculated the co-efficient of variation (CV) ( Table S2). For all the 3 standards, mean Ct values from triplicates were plotted against plasmid copy number per reaction, which resulted in a linear graph ( Figure 3A-C). The calculated detection limit was 5 copies for HHV-6, 100 copies for C. trachomatis and 100 copies for PI15 per reaction with a CV value of less than 5%. Detection of HHV-6, C. trachomatis and PI15 was highly efficient over a wide range (5 to 10,000 copies per reaction). Ten randomly selected PCR products from each PCR were verified by sequencing.

HHV-6 and C. trachomatis Load in Cervical Swabs and Total Blood DNA
The established qPCR assays were utilized to detect HHV-6 and chlamydial DNA in cervical swabs as well as in DNA of total blood of patients. Based on qPCR results, the distribution pattern of different patient samples is shown in Table 1. A scatter plot was generated to visualize the correlation of HHV-6 and chlamydial load in cervical swab samples (only samples, which were either positive for HHV-6 or Chlamydia or for both (n = 65) were included) ( Figure 3D).
Interestingly, we observed 2 clearly different groups of patients in terms of the viral and chlamydial load in their cervical swabs. The C. trachomatis negative group (n = 22) comprised patients with symptoms suggestive of genital infection or with partners having C. trachomatis infection, but negative serology as well as negative C. trachomatis NAAT test (Table 1, 2). We could detect chlamydial DNA in cervical samples from two of these patients, whereas 20 were negative also by our qPCR assay. Interestingly, 18 out of these 22 patients had high viral load either in blood or in cervical swabs or both. 12 out of these 18 HHV-6 positive patients had detectable HHV-6 in cervical swab samples and all of them had higher viral load in cervical swabs than their blood.
Another group (n = 21) of patients were positive for C. trachomatis DNA but no HHV-6 DNA was detectable ( Table 1, 3) in their respective cervical swabs. 18 patients had also serological evidence of C. trachomatis infection: 8 had IgM, 16 had IgG and four had both IgG and IgM antibody. In 13 of the 21 patients HHV-6 DNA was detected in the blood.
We detected both HHV-6 and C. trachomatis DNA in cervical swabs of 23 patients (Table 1, 4). In 12 of these 23 patients, higher levels of viral DNA could be detected in cervical swab samples than in the blood. In the blood of the remaining 11 patients, HHV-6 DNA was not measurable. All these patients had detectable chlamydial DNA in the same cervical sample. 20 of these patients (90.9%) were positive for C. trachomatis IgG antibody indicating previous exposure to C. trachomatis or long-term infection, and 6 of them had also IgM antibody. Interestingly, none of these patients had high chlamydial load in their cervical swabs. In conclusion, our qPCR assay detected high HHV-6 load in cervical swab samples from patients with C. trachomatis infection supporting an interaction between these two microbes also in patients.

Discussion
C. trachomatis is a sexually transmitted mucosal pathogen and very frequently detected in cervical epithelial cells of infected patients [36,37]. Yet, role of chlamydial infection in these patients is not well understood. In our previous report, we showed that active HHV-6 infection induces chlamydial persistence in HeLa cells. On the other hand, Chlamydia infection favors HHV-6 infection and survival. As most humans acquire HHV-6 infection early in life and C. trachomatis infection later, we addressed the question whether C. trachomatis infection can activate latent HHV-6 infection. Several chemical stimuli have been shown to activate viral replication in CiHHV-6 cell lines leading to formation of infectious viral particles [3]. Therefore, we tested 3 CiHHV-6 cell lines as well as PBMCs from 5 individuals having CiHHV-6 for their ability to induce viral replication after C. trachomatis infection. We observed an increase in the amount of viral DNA in all the cell types within 5-9 days after C. trachomatis infection. This increase in viral DNA was comparable to that induced by known chemical inducers like TSA [3]. However, unlike reactivation with chemical stimuli, long-term co-culture experiments did not result in infectious virus production. It is currently unclear, why formation of infectious viral particles was not detected after long-term culture. While chemical inducers are easily accessible for every cell in a particular experiment, Chlamydia infection is highly asynchronous in cultured suspension cells. Under most successful infection conditions, only 40-70% of the cells can be infected with C. trachomatis. Also, C. trachomatis has a highly lytic life cycle in cultured cells, which prevents development of long term Chlamydia infection in vitro. Thus, it is possible that any signal for HHV-6 activation resulting from Chlamydia infection is only available for limited time in vitro, preventing complete activation of HHV-6 DNA replication in our co-culture experiments.
Activation of HHV-6 leading to extra-chromosomal HHV-6 DNA formation might be the initial stage for virus activation. We observed formation of extra-chromosomal linear HHV-6 DNA in all the CiHHV-6 cell lines after C. trachomatis infection as well as in total PBMCs from one of the CiHHV-6 individuals. We detected HHV-6 DNA smaller than 48-kb in length in mock infected PBMCs from the CiHHV-6 individual ( Figure 2C, 2D), which might represent spontaneously activated and incomplete HHV-6 DNA in these cells or can originate from community-acquired HHV-6 strains. In addition, we detected slowly migrating HHV-6 DNA in C. trachomatis infected CiHHV-6 cells co-cultured with permissive HSB-2 cells ( Figure 2E, S1), which might represent either nicked circular or concatemeric HHV-6 DNA. In both cases, generation of extra-chromosomal circular HHV-6 DNA can be postulated as reason for this observation. Circular HHV-6 DNA is rarely detected [1,2] in some of the cell lines infected with HHV-6, but the exact role of such circular viral DNA is unclear. Circular viral genome is postulated to be the start of replication for different herpes viruses, which can form precursors for linear viral DNA by rolling circle amplification mechanism [38,39,40,41,42]. Thus our results suggest possible viral activation in these cells in response to C. trachomatis infection, which leads to formation of extra-chromosomal HHV-6 DNA.
To check whether HHV-6 activation is restricted to certain cell types, we infected total blood leukocytes and monocyte-derived macrophages with C. trachomatis. Although we observed increase in the amount of viral DNA in both the cell types, HHV-6 DNA replication was more efficient in human macrophages. This may be due to the fact that macrophages are more susceptible to C. trachomatis infection than in leukocytes [23] thus leading to a better induction signal for HHV-6 DNA replication. It will be interesting in future to see the effect of increased viral DNA accumulation on transcription and translation profile of HHV-6 and its consequence on chlamydial growth.
To obtain evidence for a role of C. trachomatis and HHV-6 infections in clinical settings, we analyzed cervical swabs taken from patients with suspected C. trachomatis infection for the presence of HHV-6 and chlamydial DNA by qPCR. Results obtained with our established C. trachomatis qPCR assay correlated very well with those obtained using a commercial NAAT test with 91.7% of the specimens having similar results by both detection methods. The few discrepant results may be due to the sample collection as separate cervical swabs collected in different media were used for NAAT and the qPCR assay (see Material and Methods), or different analytical sensitivities (rRNA detection vs. DNA detection). Besides being qualitative, our qPCR assay could also quantitate the C. trachomatis load. We detected HHV-6 DNA in 48% of the tested samples. Although the amount of HHV-6 DNA in cervical swabs of Chlamydia-infected patients is comparatively high, HHV-6 or other HHVs have been frequently detected in cervical swabs earlier [17,43,44,45]. We also detected HHV-6 DNA in blood cells of 59% of the patients. Although HHV-6 DNA has previously been detected in PBMCs of 40-90% of healthy individuals [46], the viral DNA copy numbers have not been determined in these studies. Thus, the detection of 10 3 -10 4 copies of HHV-6 DNA per 10 6 cells in total blood of C. trachomatis-infected patients points to a high probability of viral activation in all these patients.
We observed high viral loads either in total blood or cervical swabs from 55 patients. Interestingly, 81.8% of the patients in the C. trachomatis negative group had activated HHV-6 in their blood. HHV-6 levels in the whole blood of healthy individuals range between 5-10 copies per ml of blood [30]. Similarly, in CiHHV-6 patients, HHV-6 DNA levels exceed 5.5 log10 copies/ml [4]. HHV-6 DNA load in total blood, which lies in between these two extremes, indicates an activated state of the virus. Remarkably, 66.7% of the C. trachomatis negative patients with activated HHV-6 in their blood also had high viral titers in cervical epithelial cells ranging up to 5-6 log10 per 10 6 cells in 10 of the samples (Table 2). Cervical epithelial cells are not known to be permissive for productive HHV-6 infection. High viral DNA in these cells could also come from blood cells derived from blood vessels associated with cervical tissue. But this assumption may not be valid as the viral titer was comparatively low in whole blood samples in the respective patients. Thus, we assume that high viral titer in cervical swabs could originate from localized activation of viral replication in cervical epithelial cells or in blood cells that enrich cervical epithelium and thus come in contact with cervical epithelial cells.
We observed coexistence of HHV-6 and Chlamydia in 23 samples (Table 4). Viral load was high in all these samples indicating HHV-6 activation. Similar to the previous observation, the blood HHV-6 load was comparatively low or negative in all these 23 patients in comparison to the viral load in their cervix. This could again be due to localized activation of HHV-6 in cervical epithelial cells in the presence of Chlamydia infection. Presence of chlamydial IgG antibody in 21 of these patients indicates the presence of chronic or long-term C. trachomatis infection, which supports our results from cell culture based HHV-6 reactivation studies.
For a better understanding of our study results, we divided the patients into 4 subgroups according to the viral load (group 1, 2, 3 and 4) and according to the chlamydial load into 3 subgroups (group A, B and C) ( Figure 3D). Using the Kruskal-Wallis test, we observed a negative correlation between viral and bacterial load (Table S3). However, due to high degree of standard deviation, particularly within the samples with high chlamydial DNA load the correlation did not reach statistical significance. Low chlamydial DNA load has been detected in incident C. trachomatis infections in comparison to prevalent infections [36]. This could indicate that an active chlamydial infection might be associated with high chlamydial load whereas persistent or recovered infections would have lower organism loads. Thus, we excluded the samples (n = 22) with high chlamydial load (group A) and analyzed the patients with no C. trachomatis and low to moderate C. trachomatis load (group B and C). A statistically significant association between chlamydial and viral load was observed in these subgroups (Table S4). A two-way chi-square test within these subgroups also showed statistically significant association (Table  S5).
In contrast to Kruskal-Wallis ANOVA test involving total samples, a two-way chi-square test revealed a statistically significant negative association (p,0.05) between viral and bacterial load (Table 5). Samples with high viral titer in cervix had a low chlamydial load whereas samples with low viral titer in cervix had a high chlamydial load ( Figure 3D).
We observed 16 cases where both cervical swab as well as total blood was negative for HHV-6 DNA. In comparison, 22 samples had high viral load in blood whereas no HHV-6 was detected in their respective cervical swabs. As we cannot confirm the source of viral activation in whole blood of these 22 patients, we analyzed the rest of the 35 cases where high HHV-6 DNA titer was observed in cervical swabs (n = 34) in comparison to their whole blood counterpart by a binomial test (Table S6). We observed a statistically significant preference for viral activation in cervical epithelial tissues compared to whole blood, which can be attributed to co-infection of cervical tissue with other pathogens including C. trachomatis.
Our qPCR results show strong correlation between HHV-6 and C. trachomatis infection in these patients. However, currently the role of other pathogens like Herpes simplex virus 1 and 2 (HSV1/ 2) [47] or Neisseria gonorrhoeae [37], which are also regularly detected in cervical epithelial cells, remains elusive. More elaborate studies are required to understand the exact mechanism behind such co-infections and their effect on disease progression. Although HHV-6 is very frequently detected in various grades of cervical lesions, a role of HHV-6 in cervical cancer is still debatable. Because of lack of any molecular mechanism to associate HHV-6 together with other pathogens to cervical cancer, HHV-6 has been so far regarded as a bystander [48] in the oncogenesis of cervical cancer. At the same time, several studies have pointed out a strong role of HHV-6 in cervical pathogenesis [17,31,49,50]. To the best of our knowledge, no in depth study has been carried out so far to look for possible mechanism and role of HHV-6 and Chlamydia in cervical diseases. Our study for the first time links Chlamydia infection to HHV-6 pathogenesis in cervical epithelial cells. More detailed study in this regard is necessary to understand the consequences of such interactions in the context of disease development. Figure S1 Gardella gel analysis and subsequent southern hybridization showing extra-chromosomal HHV-6 DNA after C. trachomatis infection. HSB-ML cells were infected with C. trachomatis (MOI 5) for 2 days after which cells were washed thoroughly and grown in presence of fresh media with 1 mg/ml of doxycycline for another 7 days. Fresh HSB-2 cells were added to these cells and were co-cultured for another 2 weeks. 2.5610 6 of these cells were loaded directly onto a Gardella gel and subsequently stained with ethidium bromide and afterwards transferred to a nylon membrane. HHV-6-infected HSB-2 for productive virus infection (HSB-2+HHV-6), C. trachomatis elementary bodies (purified Ctr EB) and C. trachomatis-infected HSB-ML cells, which were subsequently not co-cultured with HSB-2 cells (HSB-ML+Ctr) were loaded as controls. Nylon membrane was hybridized with HHV-6-specific probe. Extra-chromosomal linear HHV-6 DNA and possible nicked circular viral DNA are marked. (TIF) Table S1 Oligos used for qPCR, southern hybridizations and cloning of plasmids for standard curve generation. (DOCX)

Table S3
Kruskal-Wallis test to demonstrate the association between Chlamydia and HHV-6 load in cervical smear of all patients. Samples have been arbitrarily divided into 4 sub groups (group 1, 2, 3 and 4) depending on the HHV-6 viral load. Respective HHV-6 DNA load is mentioned within brackets. SD, standard deviation. (DOCX)

Table S4
Kruskal-Wallis test to demonstrate the association between Chlamydia and HHV-6 load in cervical smear of patients with moderate to low chlamydial load. Group B: Chlamydial DNA load between 100-25000 chlamydial genome copies/1000 cells; Group C: Chlamydial DNA load below 100 genome copies (not detectable)/1000 cells. *Samples have been arbitrarily divided into 4 sub groups (group 1, 2, 3 and 4) depending on the HHV-6 viral load. Respective HHV-6 DNA load is mentioned within brackets. SD, standard deviation. Significance = 0.048. (DOCX)

Table S5
Two-way Chi-square test to demonstrate the association between Chlamydia and HHV-6 load in cervical smear of patients having moderate to low chlamydial load. *Samples have been arbitrarily divided into 4 sub groups (group 1, 2, 3 and 4) depending on the HHV-6 viral load. Respective HHV-6 DNA load is mentioned within brackets. Similarly, all the samples have been arbitrarily divided into 3 sub groups (group A, B and C) depending on the chlamydial DNA load. **Group B: between 100-25000 DNA chlamydial genome copies/1000 cells; Group C: Below 100 chlamydial genome copies (not detectable)/1000 cells. Significance = 0.024.

(DOCX)
Table S6 Binomial test showing higher preference of HHV-6 activation in cervical epithelial cells than in the whole blood counterpart. Group 1, samples where HHV-6 copy number in cervical smear is more than in total blood. Group 2, samples where HHV-6 copy number in cervical smear is less than in total blood. (DOCX)