Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
Molecular Human Reproduction Vol.11, No.10 pp. 761–766, 2005 Advance Access publication October 27, 2005 doi:10.1093/molehr/gah234 Universal DNA primers amplify bacterial DNA from human fetal membranes and link Fusobacterium nucleatum with prolonged preterm membrane rupture R.J.Cahill1, S.Tan2, G.Dougan3, P.O’Gaora1, D.Pickard1, N.Kennea2, M.H.F.Sullivan4,5, R.G.Feldman2 and A.D.Edwards2 1 Department of Biological Sciences, Centre for Molecular Microbiology and Infection, Flowers Building, 2Department of Paediatrics, Imperial College London, Hammersmith Campus, Du Cane Road, London, 3The Wellcome Trust Sanger Institute, Wellcome Trust Genome Campus, Hinxton, Cambridge and 4Department of Obstetrics & Gynaecology, Imperial College London, Hammersmith Campus, Du Cane Road, London, UK 5 To whom correspondence should be addressed at: Department of Obstetrics & Gynaecology, Institute of Reproductive & Developmental Biology, Imperial College London, Hammersmith Hospital, Du Cane Road, London W12 0NN, UK. E-mail: [email protected] A large number of bacterial species have been identified in fetal membranes after preterm labour (PTL) associated with intrauterine infection by microbiological culture. In this study, we have investigated a molecular and bioinformatic approach to organism identification which surmounts the need for specific and diverse microbiological culture conditions required by conventional methods. Samples of fetal membranes were taken from 37 preterm infants, and 6 normal term controls delivered by caesarean section, in which bacteria had been detected by in situ hybridization of 16S ribosomal RNA using a generic probe. Degenerate primers were designed to amplify bacterial 16S ribosomal DNA by PCR and used to amplify bacterial DNA from human fetal membranes. Amplicons were cloned, sequenced and bacteria were identified bioinformatically by comparison of sequences with known bacterial DNA genomes. In situ hybridization using an organism specific probe was then used to confirm the presence of the commonest identified organism in tissue samples. Bacterial DNA amplified from 15/43 samples, all from preterm deliveries, and the bioinformatic approach identified organisms in all cases. Multiple bacteria were identified including Mycoplasma hominis, Pasturella multocida, Pseudomonas PH1, Escherichia coli and Prevotella bivia. The commonest organism Fusobacterium nucleatum was found in 9/15 (60%) of samples. Ten of the 12 samples obtained after prolonged membrane rupture were positive for bacterial DNA, and 7 of these (70%) contained DNA from F. nucleatum. Bacteria from fetal membranes may be identified by molecular and bioinformatic methods. Further work is warranted to investigate the apparent linkage between F. nucleatum, fetal membrane rupture and preterm delivery. Key words: 16S rDNA/fetal membranes/F. nucleatum/PTL Introduction Many studies have associated intrauterine infection with preterm labour (PTL) and a series of fetal pathologies. The wide range of organisms found suggests a general inflammatory response to any organisms rather than a specific effect of pathogenic species (Romero et al., 1989a) although recent studies have identified a link between poor oral hygiene and poor pregnancy outcome and implicated oral pathogens in PTL (Hill, 1998; Offenbacher et al., 1998; Carta et al., 2004; Goepfert et al., 2004). Pregnancies from which organisms are detected by microbiological culture often also show chorioamnionitis (inflammatory changes in the fetal membranes) to the extent that the latter has been used as a surrogate marker of the former (Steinborn et al., 1996). Our recent data show that this simple relationship is not tenable, as bacteria may be visualized by fluorescence in situ hybridization with a generic DNA probe to bacterial 16S ribosomal RNA in the majority of term and preterm fetal membranes (Steel et al., 2005). The term tissues showed no evidence of chorioamnionitis (Steel et al., 2005), indicating that the mere presence of bacteria may not be sufficient to cause an inflammatory response, suggesting that the link between bacteria and intrauterine inflammation is more complex than previously suspected. Different organisms may have different effects on fetal membranes, but our study (Steel et al., 2005) provided no information on the species of bacteria present. For many years, identification of bacteria was a complex process; a multitude of specific reagents would be needed for immunohistochemistry, and microbiological culture can be complicated by the different growth requirements of bacteria and also by the difficulty in growing some bacteria resulting in false-negative data. A series of papers have shown that PCR-based methods, in which DNA primers are selected to amplify defined parts of the bacterial genome can be used to identify organisms in a range of tissues and samples (Greisen et al., 1994; Jalava et al., 1996; Hitti et al., 1997; Gardella et al., 2004; Rinttila et al., 2004). The first aim of this study was to identify a pair of primers that would amplify DNA from many species of bacteria, which would be a more generic approach than species-specific primers. We then aimed © The Author 2005. Published by Oxford University Press on behalf of the European Society of Human Reproduction and Embryology. All rights reserved. For Permissions, please email: [email protected] 761 R.J.Cahill et al. to apply the method to archive fetal membrane tissue, identify the bacteria present, and attempt to relate the data to pregnancy outcome. Methods Design of degenerate 16s rRNA PCR primers The GenBank® nucleotide sequence database was searched using the Ovid platform (Ovid, New York, NY, USA) for complete 16s rRNA sequences. These sequences were aligned using ClustalW (Thompson et al., 1994). Regions of variation were identified and marked within the sequences. The conserved DNA sequences flanking these regions of DNA variation were then used to design degenerate primers. Primers were designed with lengths ranging from 14 to 25 bp. The primers designed were then analysed using the BLAST programme and those primers sequences that recognized over 100 16s rRNA sequences in the BLAST database were generated for use by ThermoHybaid (MA, USA). Control non-degenerate 16s rRNA primers designed against conserved regions of 16s rRNA and used previously in clinical studies were also generated (Parkhill et al., 2001). Previously validated and published primers for human β-actin were also used in this study (Ace and Okulicz, 1995). PCR methodology The PCR primers used in the study are shown in Figure 1 and the exact sequences in Table I. To standardize the location of the selected 16s rRNA primers within the 16s rRNA, each primer sequence was mapped to the Escherichia coli 16s rRNA gene. All PCR were optimized for the primer set and carried out according to standard protocols in 50 ul reactions, using a reaction mixture of 1.5 mM MgCl2, 50 mM KCl, 10 mM Tris–HCl (pH 8.3), 200 μM of each deoxynucleotide triphosphate and 2 U of Amplitaq DNA polymerase (PE Applied Biosystems). Amplification reactions were carried out using an ABgene® thermocycler (Epsom, Surrey, UK) employing the following conditions; 95°C × 1´, 35 cycles of 95°C × 30´, 50°C × 30´ and 72°C × 1´ and a final elongation of 72°C × 5´. Control reactions were carried out for each PCR with double distilled H2O replacing DNA to assess bacterial DNA contamination. Cloning and sequencing of PCR amplicons The PCR products were resolved by agarose gel electrophoresis and the amplicon of required size was extracted using the Qiaex II gel extraction kit (Quiagen) according to the manufacturers protocols. The resulting DNA was ligated into the PCR®-TOPO® TA vector (Invitrogen) and transformed into One Shot® Top 10 chemically competent E. coli (Invitrogen). Twelve clones per sample were sequenced using ABI 3100 sequencing service at the Advanced Biotechnology Centre (ABC, Imperial College, London, UK); previous experience indicates that this is sufficient to identify the main bacteria present. To confirm this, 50 clones from one sample were sequenced, and no additional sequences were found other than those in the first 12 samples sequenced. All sequencing results obtained were analysed using the BLAST programme and the resulting alignments analysed (Altschul et al., 1990). Isolation of bacterial DNA Bacterial-type strains were obtained from the NCTC, and clinical isolates were provided by Microscience, Reading, UK, and Mr Michael Quail of the Pathogen Sequencing Unit at The Wellcome Trust Sanger Institute, Cambridge, UK. Each bacterial strain was cultured as per individual requirements. DNA was isolated using the DNeasy Tissue DNA extraction kit (Quiagen) modified for Gram-negative and Gram-positive bacteria as per manufacturers’ instructions. Figure 1. Schematic of 16S ribosomal DNA gene, showing location of primers and generic probe relative to the genome of Escherichia coli. 762 Patient samples Approval for this study was obtained from the Hammersmith Hospitals Trust Research Ethics Committee, and all parents gave consent. Fetal membranes (amnion, chorion and adherent decidua) from 43 women (6 women after uncomplicated term deliveries by caesarean section, 37 women after preterm delivery) were obtained within 30 min of delivery and placed in phosphatebuffered saline (PBS) supplemented with penicillin-streptomycin to limit further bacterial growth. Tissue samples (∼200 mg) were frozen in liquid nitrogen and kept at –80°C until further study, and matched samples were fixed in formalin and stored in 70% ethanol : 30% water. Data from these pregnancies are summarized in Table II. The preterm delivery groups are defined as follows: preterm not in labour (PNIL), infants delivered by caesarean section, for worsening maternal or fetal health due to pre-eclampsia (with or without fetal growth restriction); PTL, infants delivered from women in PTL but no evidence of fetal membrane rupture and prolonged rupture of fetal membranes (PROM), infants delivered from women in PTL at least 24 h after rupture of fetal membranes. Women in PTL were all treated with augmentin, and women with PROM were treated with erthyromycin. A single fixed fetal membrane sample was processed for histological assessment, with a positive diagnosis of chorioamnionitis based on the presence of at least 10 polymorphonuclear leukocytes per 400-power field in the fetal membrane sample. All methods to isolate, amplify and sequence bacterial DNA were carried out using aseptic technique, filtered tips and sterile reagents to avoid contamination with environmental bacteria. DNA was extracted from the membrane tissue using the DNeasy tissue extraction kit (Quiagen) according to the manufacturers’ instructions. The DNA was resuspended in 200 μl of double distilled H2O and stored at –20°C. The integrity of the human DNA from these extractions was confirmed by amplification of β-actin using specific primers (Table I), and this was successful in all cases. Bacterial DNA was amplified by PCR using the degenerate generic primers identified in the first part of the study: the Forward primer was CCTACGGRSGCAGCAG (16s2F) and the Reverse primer was GGACTACCMGGGNTATCTAATCCKG (16s4R). PCR products were ligated into the PCR®-TOPO® TA vector (Invitrogen), cloned and sequenced as described above. All sequencing results obtained were analysed using the BLAST programme and the resulting alignments analysed (Altschul et al., 1990). Fluorescent in-situ hybridization for bacteria Flourescent in-situ hybridization (FISH) was used to demonstrate the presence of bacteria in fetal membranes, using the method described previously (Steel et al., 2005). In brief, fixed paraffin-embedded tissues were cut using a sterile approach and the sections mounted on poly-L-lysine coated slides. Sections were treated with proteinase K before incubation with a generic probe to 16S ribosomal RNA (5´-F-ACTGCTGCCTCCCGTAGGAGTTTATTCCTT). A subset of fetal membrane samples were also used for FISH with a specific probe to the most common organism, Fusobacterium nucleatum. This probe was designed using the ARB software and ARB database (ARB—a software environment for sequence data, http://www.arb-home.de). The sequence used was CGCAAAGCTCTCTCACAG, and the technique was the same as with the generic probe (Steel et al., 2005). Results The primer sequences identified from the databases were tested on bacterial DNAs, with the results summarized in Table III. Only the combination of 16s2F and 16s4R amplified DNA from the 12 species tested and showed no additional bands. This pair of primers was used to test DNA from a further 25 species, all of which were amplified successfully (Bacteroides fragilis, Citrobacter rodentium, Clostridium perfringens, Escherichia coli K12, F. nucleatum, Mycoplasma bovis, Niesseria meningitides, Salmonella paratyphi, Streptococcus pneumoniae, Bordellata pertussis, Burkholderia cepacia, Campylobacter jejuni, Chlamydia trachomitis, Clavibacter michagenanesis, Erwinia amylovara, Haemophilus influenzae, Helicobacter mustalae, Photorhabus asymbiotica, Proteus mirabilis, Rhizobium leguminosarum, Shigella sonnei, Staphylococcus aureus MSSA, Streptomyces scabies, Vibrio salmoncida and Yersinnia enterocolitica). To confirm Molecular identification of organisms in fetal membranes Table I. Summary of primers investigated in this study (reference) Primer name Location Sequence 5´–3´ Actin F (15) Actin R (15) 16sRNA1 (14) Salmonella typhi 16sRNA2 (14) Salmonella typhi 16s1F 16s2F 16s4R 16s5R 16s6R 16s7R – – GCACACTTTAGCCGTGTTATTTGCACTTCT AGGCGTACAGGGATAGCACAGCCTGGATAG GTGAAATGCGTAGAGATCTGGAGG CGAGCTGACGACAGCCATGC GTGCSAGCANGCCGCGGTAATAC CCTACGGRSGCAGCAG GGACTACCMGGGNTATCTAATCCKG CATTYCTTTRAGTT GAGCTGACGACAGCCATGC GACGTCNTCCCCACCTTCC 648–671 1011–1031 58–78 299–320 741–764 866–880 1011–1030 1131–1151 Location is relative to Escherichia coli 16S rRNA gene. Table II. Summary of clinical information from pregnancies included in this study Gestation at Delivery (weeks) Birthweight (g) Delivery type Pre-eclampsia IUGR Membrane Rupture (h)† Dexamethasone Chorioamnionitis Term (6) PNIL (11) PTL (14) PROM (12) Term* Normal* CS (6) 0 0 26.2–32.0 602–1668 CS (11) 11 (4 with IUGR) 0 23.3–31.0 478–2040 CS (4), V (10) 2 (1 with IUGR) 1 23.5–31.2 652–1530 CS (7), V (5) 0 0 0 0 0/6 0 11 0/11 0 14 6/14 30–816 12 11/12‡ CS, caesarean section; IUGR, intrauterine growth restriction; PNIL, preterm delivery not in labour; PROM, prolonged rupture of fetal membranes; PTL, preterm labour; V, vaginal delivery. Ranges are given for gestational ages and birthweights of preterm deliveries. *Gestational age and birthweight were not specifically recorded from term deliveries but known to be within normal ranges. †The period of time from membrane rupture until delivery of the infant. ‡P < 0.05 by χ2 compared with other preterm delivery groups. Table III. Comparison of primer sets designed to amplify bacterial 16s rDNA using a range of bacterial species Bacterial species Source RNA1 + RNA2 16s1F + 16s5R 16s2F + 16s5R 16s1F + 16s4R 16s2F + 16s4R 16s2F + 16s6R 16s2F + 16s7R Enteropathogenic Escherichia coli Enterohaemorrhagic Escherichia coli Salmonella typhi Citrobacter rodentium Helicobacter pylori Group A Streptococcus (NC8191) Neisseria meningitidis (MC38) Pasteurella multocida Salmonella enteriditis Staphylococcus epidermis (SL61) Group B Streptococcus (A909) Staphylococcus aureus (MH20) Control (no DNA) DNA DNA DNA DNA DNA DNA DNA DNA DNA Culture Culture Culture – + + + + (+) + (+) + + (++) + (++) + (++) + (++) + + (++) – – – – – – – – – – – – – – – – – – – – – – – – – – – + – – – + + + + – + – + – + + + + + + + + + + + + – + + – – + + + + – + + + – – + – – + + + + + + – + – The results are expressed as + or – based on the presence of an amplicon of predicted size. The number of non-specific amplicons is shown within parentheses. +, 1; ++, 2 non-specific bands. that the amplified DNA was of the correct sequence, two separate clones from five bacteria (Citrobacter rodentium, Enterohaemorrhagic E. coli, F. nucleatum, Pasteurella multocida and Salmonella enterica serovar Typhimurium) were sequenced. All sequences obtained showed 100% homology with the database sequences, indicating that correct amplification had been achieved. DNA was extracted from 43 fetal membrane samples as indicated above. PCR with the β-actin primers succeeded for all samples, indicating that the human DNA had not been damaged during extraction. Bacterial 16S rDNA sequences were amplified from 15 of the 43 samples investigated, as summarized in Table IV. Most of the samples from which bacterial DNA was amplified were from pregnancies with PROM (Table IV). Reorganization of the data revealed that there was a close correlation between chorioamnionitis and the success of PCR (Table IV). F. nucleatum was the most commonly detected organism, being present in 9 of the 15 samples (60%) that contained bacterial DNA (Table IV). DNA from other organisms (Mycoplasma hominis, P. multocida, 763 R.J.Cahill et al. Table IV. Histological and 16s rDNA PCR amplification results from human fetal membrane tissues Study group n Chorioamnionitis positive [n (%)] Positive 16s rDNA [n (%)] Term/preterm no labour Preterm labour no PROM 17 14 0 (0) 7 (50) 1 (6) 4 (28) Preterm labour PROM 12 11 (92) 10 (83)* Histological chorioamnionitis 18 18 (100) 13 (72)* No histological chorioamnionitis 24 0 (0) 2 (8) Bacteria identified Fusobacterium nucleatum/Mycoplasma hominis (1) Mycoplasma hominis (2) Pseudomonas (1) Fusobacterium nucleatum/Mycoplasma hominis (1) Fusobacterium nucleatum (6) Pseudomonas (1) Escherichia coli (1) Prevotella bivia (1) Fusobacterium nucleatum/Mycoplasma hominis/Pasturella multocida (1) Fusobacterium nucleatum (6) Mycoplasma hominis (2) Pseudomonas (1) Escherichia coli (1) Prevotella bivia (1) Fusobacterium nucleatum/Mycoplasma hominis (1) Fusobacterium nucleatum/Mycoplasma hominis/Pasturella multocida (1) Fusobacterium nucleatum/Mycoplasma hominis (1) Pseudomonas (1) PROM, prolonged rupture of fetal membranes. Chorioamnionitis positive is defined as a minimum of 10 polymorphonuclear lymphocytes/400 power field. Positive PCR is defined as PCR which resulted in amplification and sequencing of bacterial 16s DNA. *P < 0.05 by χ2 compared with other patient groups. Pseudomonas PH1, E. coli and Prevotella bivia) was detected less frequently as shown. Three samples revealed polymicrobial infection (Table IV); all three contained F. nucleatum and M. hominis, and one also contained P. multocida. Seven of the nine samples (77.8%) which contained DNA for F. nucleatum (alone or with other organisms) were from tissues obtained after prolonged membrane rupture (P < 0.05 by χ2 compared with the incidence in tissues from pregnancies not complicated by prolonged membrane rupture). The presence of bacteria in the fetal membranes was confirmed by FISH with the generic probe used in our recent study.7 All the samples used contained organisms, and typical data are shown in Figure 2a. FISH with a specific probe was also used to localize F. nucleatum in fetal membranes and typical data are shown in Figure 2b. Three different fetal membrane samples obtained after PROM that were positive for F. nucleatum by PCR and sequencing were also positive by FISH. No positive hybridization with the F. nucleatum-specific probe was observed in five different tissue samples (both term and preterm) that were negative for F. nucleatum by PCR and sequencing; these samples were positive for bacteria by generic FISH. Discussion We describe methods that may be used to amplify DNA from many different species of micro-organisms, and use this DNA to identify the organisms present in human fetal membrane tissues. The bioinformatics proved to be extremely reliable; submission of each DNA sequence obtained provided a positive match with a known bacterial 16S rDNA, and in three cases showed polymicrobial infection. All the tissues studied were positive for bacteria by FISH, but PCR succeeded only in 15 of the 43 samples (35%). Eschenbach and coworkers have used similar methods to show the presence of bacteria (Hitti et al., 1997) and identify bacteria (Gardella et al., 2004) in amniotic fluid, but not all their samples were positive by PCR and sequencing. More work seems to be needed to improve the efficiency Figure 2. Fluorescence in-situ hybridization (FISH) with a generic probe to bacterial 16S rRNA (a) or with a probe specific for Fusobacterium nucleatum (b). Organisms stain strongly positive (arrowheads) against background autofluorescence. Scale bar, 10 μM. 764 Molecular identification of organisms in fetal membranes of this stage of the analysis, although reasons for this should be considered. One possibility is a ‘mass-effect’, whereby a minimum number of organisms are needed to obtain a PCR amplicon, in contrast to the visualization of single organisms by FISH. A quantitative study has shown that real-time quantitative PCR has a detection limit of 30–4500 bacterial genomes when specific DNA primers are used (Rinttila et al., 2004), indicating that there may be a limit to this type of approach. The human DNA was of sufficient quality in all samples for PCR with primers for β-actin to succeed, so it seems unlikely that bacterial DNA has been lost. All the organisms we identified (F. nucleatum, M. hominis, P. multocida, Pseudomonas PH1, E. coli and P. bivia) from the sequence databases have been reported previously in tissues obtained after PTL, indicating that our data are consistent with previous data. Bacterial 16S ribosomal DNA was successfully sequenced from PROM and some PTL tissues, but less frequently from preterm no labour (1/11) and term tissues (0/6) (Table IV). The close relationship between identifying bacterial DNA and chorioamnionitis (Table IV) reflects previous studies linking ‘infection’ with chorioamnionitis (Steinborn et al., 1996). The most common organism identified was F. nucleatum. Many papers have reported that this organism is commonly found in tissues with chorioamnionitis (Altshuler and Hyde, 1985, 1988), as well as after PTL (Romero et al., 1992; Bearfield et al., 2002), but this seems to be the clearest link with PROM reported so far. The haematogenous spread of F. nucleatum can cause PTL in mice (Han et al., 2004), a finding which mirrors the proposed impact of this organism in human pregnancy. The detailed mechanisms through which F. nucleatum causes PTL have not been addressed in this study, and this requires further investigation. It should be noted that F. nucleatum is a Gram-negative organism which will therefore have an outer layer of lipopolysaccharide (LPS). LPS (usually from E. coli) is a potent activator of prostaglandin and cytokine production from human fetal membranes (Rajasingam et al., 1998; Alvi et al., 1999) and from the constituent amnion, chorion and decidua (Romero et al., 1989b; Dudley et al., 1992a,b, 1993; Mitchell et al., 1993), and live bacteria can have similar effects (Rajasingam et al., 1998). We anticipate that live F. nucleatum and LPS from these organisms will induce type-2 cyclo-oxygenase (and hence prostaglandin production), and inflammatory cytokine output from fetal membranes, which will parallel the changes observed in these factors during labour. It is interesting to note that many studies have implicated Ureaplasma urealyticum as a common cause of PTL and subsequent complications (McDonald and Chambers, 2000; Berger et al., 2003; Gerber et al., 2003), but we did not detect it, even though Gerber and coworkers used a very similar molecular approach to ours. We suggest that this is because U. urealyticum is primarily an organism found in fluids (e.g. urine, amniotic fluid), and the cells contacting them, rather than within tissues. Our samples would have mostly organisms within the tissues, rather than those on the surface of the amniotic epithelium, and this may explain the difference between the earlier study and our findings. This new molecular and bioinformatics approach provides a new method to investigate the associations between particular intrauterine organisms and pregnancy outcome, avoiding the need for multiple microbiological culture. F. nucleatum was associated with prolonged fetal membrane rupture before delivery, and this apparent linkage requires further study. Acknowledgements We thank MicroScience for funding this study. References Ace CI and Okulicz WC (1995) Differential gene regulation by estrogen and progesterone in the primate endometrium. Mol Cell Endocrinol 115,95–103. Altschul SF, Gish W, Miller W, Myers EW and Lipman DJ (1990) Basic local alignment search tool. J Mol Biol 215,403–410. Altshuler G and Hyde S (1985) Fusobacteria. An important cause of chorioamnionitis. Arch Pathol Lab Med 109,739–743. Altshuler G and Hyde S (1988) Clinicopathologic considerations of fusobacteria chorioamnionitis. Acta Obstet Gynecol Scand 67,513–517. Alvi SA, Rajasingam D, Brown NL, Bennett PR, Elder MG and Sullivan MHF (1999) The production of interleukin-1β from human fetal membranes is not obligatory for increased prostaglandin output. Immunology 97,249–256. Bearfield C, Davenport ES, Sivapathasundaram V and Allaker RP (2002) Possible association between amniotic fluid micro-organism infection and microflora in the mouth. BJOG 109,527–533. Berger A, Witt A, Haiden N, Kretzer V, Heinze G and Kohlhauser C (2003) Microbial invasion of the amniotic cavity at birth is associated with adverse short-term outcome of preterm infants. J Perinat Med 31,115–121. Carta G, Persia G, Falciglia K and Iovenitti P (2004) Periodontal disease and poor obstetrical outcome. Clin Exp Obstet Gynecol 31,47–49. Dudley DJ, Trautman MS, Araneo BA, Edwin SS and Mitchell MD (1992a) Decidual cell biosynthesis of interleukin-6: regulation by inflammatory cytokines. J Clin Endocrinol Metab 74,884–889. Dudley DJ, Trautman MS, Edwin SS, Lundin-Schiller S and Mitchell MD (1992b) Biosynthesis of interleukin-6 by cultured human chorion laeve cells: regulation by cytokines. J Clin Endocrinol Metab 75,1081–1086. Dudley DJ, Trautman MS and Mitchell MD (1993) Inflammatory mediators regulate interleukin-8 production by cultured gestational tissues: evidence for a cytokine network at the chorio-decidual interface. J Clin Endocrinol Metab 76,404–410. Gardella C, Riley DE, Hitti J, Agnew K, Krieger JN and Eschenbach D (2004) Identification and sequencing of bacterial rDNAs in culture-negative amniotic fluid from women in premature labor. Am J Perinatol 21,319–323. Gerber S, Vial Y, Hohlfeld P and Witkin S (2003) Detection of Ureaplasma urealyticum in second-trimester amniotic fluid by polymerase chain reaction correlates with subsequent preterm labor and delivery. J Infect Dis 187,518–521. Goepfert AR, Jeffcoat MK, Andrews WW, Faye-Petersen O, Cliver SP, Goldenberg RL and Hauth JC (2004) Periodontal disease and upper genital tract inflammation in early spontaneous preterm birth. Obstet Gynecol 104,777–783. Greisen K, Loeffelholz M, Purohit A and Leong D (1994) PCR primers and probes for the 16S rRNA gene of most species of pathogenic bacteria, including bacteria found in cerebrospinal fluid. J Clin Microbiol 32,335–351. Han YW, Redline RW, Li M, Yin L, Hill GB and McCormick TS (2004) Fusobacterium nucleatum induces premature and term stillbirths in pregnant mice: implication of oral bacteria in preterm birth. Infect Immun 72,2272–2279. Hill GB (1998) Preterm birth: associations with genital and possibly oral microflora. Ann Periodontol 3,222–232. Hitti J, Riley DE, Krohn MA, Hillier SL, Agnew KJ, Krieger JN and Eschenbach DA (1997) Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in premature labor. Clin Infect Dis 24,1228–1232. Jalava J, Mantymaa ML, Ekblad U, Toivanen P, Skurnik M, Lassila O and Alanen A (1996) Bacterial 16S rDNA polymerase chain reaction in the detection of intra-amniotic infection. Br J Obstet Gynaecol 103,664–669. McDonald HM and Chambers HM (2000) Intrauterine infection and spontaneous midgestation abortion: is the spectrum of microorganisms similar to that in preterm labor? Infect Dis Obstet Gynecol 8,220–227. Mitchell MD, Trautman MS and Dudley DJ (1993) Cytokine networking in the placenta. Placenta 14,249–275. Offenbacher S, Jared HL, O’Reilly PG, Wells SR, Salvi GE, Lawrence HP, Socransky SS and Beck JD (1998) Potential pathogenic mechanisms of periodontitis associated pregnancy complications. Ann Periodontol 3,233–250. Parkhill J, Dougan G, James KD, Thomson NR, Pickard D, Wain J, Churcher C, Mungall KL, Bentley SD, Holden MT et al. (2001) Complete genome sequence of a multiple drug resistant Salmonella enterica serovar Typhi CT18. Nature 413,848–852. Rajasingam D, Bennett PR, Alvi SA, Elder MG and Sullivan MHF (1998) Stimulation of prostaglandin production from intact human fetal membranes by bacteria and bacterial products. Placenta 19,301–306. Rinttila T, Kassinen A, Malinen E, Krogius L and Palva A (2004) Development of an extensive set of 16S rDNA-targeted primers for quantification of pathogenic and indigenous bacteria in faecal samples by real-time PCR. J Appl Microbiol 97,1166–1177. Romero R, Sirtori M, Oyarzun E, Avila C, Mazor M, Callahan R, Sabo V, Athanassiadis AP and Hobbins JC (1989a) Infection and labor. V. Prevalence, 765 R.J.Cahill et al. microbiology, and clinical significance of intraamniotic infection in women with preterm labor and intact membranes. Am J Obstet Gynecol 161,817–824. Romero R, Wu YK, Brody DT, Oyarzun E, Duff GW and Durum SK (1989b) Human decidua: a source of interleukin-1. Obstet Gynecol 73,31–34. Romero R, Gonzalez R, Sepulveda W, Brandt F, Ramirez M, Sorokin Y, Mazor M, Treadwell MC and Cotton DB (1992) Infection and labor. VIII. Microbial invasion of the amniotic cavity in patients with suspected cervical incompetence: prevalence and clinical significance. Am J Obstet Gynecol 167,1086–1091. Steel JH, Malatos S, Kennea N, Edwards AD, Miles L, Duggan P, Reynolds PR, Feldman RG and Sullivan MHF (2005) Bacteria and inflammatory 766 cells in fetal membranes do not always cause preterm labour. Pediatr Res 57,404–411. Steinborn A, Gunes H, Roddiger S and Halberstadt E (1996) Elevated placental cytokine release, a process associated with preterm labor in the absence of intrauterine infection. Obstet Gynecol 88,534–539. Thompson JD, Higgins DG and Gibson TJ (1994) CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, positions-specific gap penalties and weight matrix choice. Nucleic Acids Res 22,4673–4680. Submitted on July 18, 2005; revised on September 19, 2005; accepted on September 22, 2005