Advertisement

No evidence for a placental microbiome in human pregnancies at term

  • Author Footnotes
    1 These authors contributed equally to this work.
    Irene Sterpu
    Footnotes
    1 These authors contributed equally to this work.
    Affiliations
    Department of Clinical Science and Education, and Division of Obstetrics and Gynaecology, Department of Clinical Science, Intervention and Technology, Karolinska Institute, Södersjukhuset, Stockholm, Sweden
    Search for articles by this author
  • Author Footnotes
    1 These authors contributed equally to this work.
    Emma Fransson
    Footnotes
    1 These authors contributed equally to this work.
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Department of Women’s and Children’s Health, Uppsala University, Uppsala, Sweden
    Search for articles by this author
  • Author Footnotes
    1 These authors contributed equally to this work.
    Luisa W. Hugerth
    Footnotes
    1 These authors contributed equally to this work.
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden
    Search for articles by this author
  • Juan Du
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Marcela Pereira
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Liqin Cheng
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Sebastian Alexandru Radu
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Lorena Calderón-Pérez
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Eurecat, Centre Tecnològic de Catalunya, Unitat de Nutrició i Salut, Reus, Spain
    Search for articles by this author
  • Yinghua Zha
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Pia Angelidou
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Alexandra Pennhag
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden
    Search for articles by this author
  • Fredrik Boulund
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden
    Search for articles by this author
  • Annika Scheynius
    Affiliations
    Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden

    Department of Clinical Science and Education, Karolinska Institutet, and Sachs’ Children and Youth Hospital, Södersjukhuset, Stockholm, Sweden
    Search for articles by this author
  • Lars Engstrand
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden
    Search for articles by this author
  • Eva Wiberg-Itzel
    Affiliations
    Department of Clinical Science and Education, and Division of Obstetrics and Gynaecology, Department of Clinical Science, Intervention and Technology, Karolinska Institute, Södersjukhuset, Stockholm, Sweden
    Search for articles by this author
  • Ina Schuppe-Koistinen
    Correspondence
    Corresponding author: Ina Schuppe-Koistinen, PhD.
    Affiliations
    Centre for Translational Microbiome Research, Department of Microbiology, Tumor and Cell Biology, Karolinska Institute, Stockholm, Sweden

    Science for Life Laboratory, Karolinska Institutet, Stockholm, Sweden
    Search for articles by this author
  • Author Footnotes
    1 These authors contributed equally to this work.
Open AccessPublished:August 28, 2020DOI:https://doi.org/10.1016/j.ajog.2020.08.103

      Background

      The placenta plays an important role in the modulation of pregnancy immunity; however, there is no consensus regarding the existence of a placental microbiome in healthy full-term pregnancies.

      Objective

      This study aimed to investigate the existence and origin of a placental microbiome.

      Study Design

      A cross-sectional study comparing samples (3 layers of placental tissue, amniotic fluid, vernix caseosa, and saliva, vaginal, and rectal samples) from 2 groups of full-term births: 50 women not in labor with elective cesarean deliveries and 26 with vaginal deliveries. The comparisons were performed using polymerase chain reaction amplification and DNA sequencing techniques and bacterial culture experiments.

      Results

      There were no significant differences regarding background characteristics between women who delivered by elective cesarean and those who delivered vaginally. Quantitative measurements of bacterial content in all 3 placental layers (quantitative polymerase chain reaction of the 16S ribosomal RNA gene) did not show any significant difference among any of the sample types and the negative controls. Here, 16S ribosomal RNA gene sequencing of the maternal side of the placenta could not differentiate between bacteria in the placental tissue and contamination of the laboratory reagents with bacterial DNA. Probe-specific quantitative polymerase chain reaction for bacterial taxa suspected to be present in the placenta could not detect any statistically significant difference between the 2 groups. In bacterial cultures, substantially more bacteria were observed in the placenta layers from vaginal deliveries than those from cesarean deliveries. In addition, 16S ribosomal RNA gene sequencing of bacterial colonies revealed that most of the bacteria that grew on the plates were genera typically found in human skin; moreover, it revealed that placentas delivered vaginally contained a high prevalence of common vaginal bacteria. Bacterial growth inhibition experiments indicated that placental tissue may facilitate the inhibition of bacterial growth.

      Conclusion

      We found no evidence to support the existence of a placental microbiome in our study of 76 term pregnancies, which used polymerase chain reaction amplification and sequencing techniques and bacterial culture experiments. Incidental findings of bacterial species could be due to contamination or to low-grade bacterial presence in some locations; such bacteria do not represent a placental microbiome per se.

      Key words

      Introduction

      In the last century, it has been assumed that the intrauterine environment in healthy pregnancies is sterile. It was not until the last decade that the notion of a sterile womb was challenged,
      • Perez-Muñoz M.E.
      • Arrieta M.C.
      • Ramer-Tait A.E.
      • Walter J.
      A critical assessment of the “sterile womb” and “in utero colonization” hypotheses: implications for research on the pioneer infant microbiome.
      in particular after the introduction of new molecular approaches, such as next-generation sequencing. In nonpregnant women, there is evidence of bacteria not only in the vagina
      • Ravel J.
      • Gajer P.
      • Abdo Z.
      • et al.
      Vaginal microbiome of reproductive-age women.
      but also in the endometrial cavity, without signs of inflammation,
      • Mitchell C.M.
      • Haick A.
      • Nkwopara E.
      • et al.
      Colonization of the upper genital tract by vaginal bacterial species in nonpregnant women.
      and a microbiome of the endometrium has been described.
      • Moreno I.
      • Garcia-Grau I.
      • Bau D.
      • et al.
      The first glimpse of the endometrial microbiota in early pregnancy.
      • Chen C.
      • Song X.
      • Wei W.
      • et al.
      The microbiota continuum along the female reproductive tract and its relation to uterine-related diseases.
      • Winters A.D.
      • Romero R.
      • Gervasi M.T.
      • et al.
      Does the endometrial cavity have a molecular microbial signature?.
      In general, intrauterine pathogens during pregnancy are suggested to be detrimental to pregnancy outcomes
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      • Hitti J.
      • Riley D.E.
      • Krohn M.A.
      • et al.
      Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in premature labor.
      • DiGiulio D.B.
      • Romero R.
      • Kusanovic J.P.
      • et al.
      Prevalence and diversity of microbes in the amniotic fluid, the fetal inflammatory response, and pregnancy outcome in women with preterm pre-labor rupture of membranes.
      because they may cause clinical chorioamnionitis, preterm premature rupture of fetal membranes, and preterm labor with intact membranes.
      • Kim C.J.
      • Romero R.
      • Chaemsaithong P.
      • Chaiyasit N.
      • Yoon B.H.
      • Kim Y.M.
      Acute chorioamnionitis and funisitis: definition, pathologic features, and clinical significance.
      • Romero R.
      • Miranda J.
      • Kusanovic J.P.
      • et al.
      Clinical chorioamnionitis at term I: microbiology of the amniotic cavity using cultivation and molecular techniques.
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.

      Why was this study conducted?

      The existence of a placental microbiome is debated. This study compared samples taken from women not in labor with those taken from women who underwent cesarean and vaginal deliveries at term to investigate the existence and origin of a microbiome in the placenta.

      Key findings

      Only sporadic bacteria, not representing a microbiome per se, were detected in placental tissue. Placental tissue samples from vaginal deliveries produced more live bacteria in culture experiments than those from cesarean deliveries.

      What does this add to what is known?

      The data are in line with previous studies that indicated lack of evidence for a placental microbiome and suggest that the growth of bacteria entering the uterine cavity is inhibited in normal pregnancies.
      The microbiome has been increasingly included in efforts to understand the mechanisms that interact to maintain a healthy immune system and pregnancy.
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Microbial colonization of the skin, gut, and other mucosal surfaces of the newborn is essential for the development of host metabolism, immunity, and resistance to pathogens.
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      ,
      • Gensollen T.
      • Iyer S.S.
      • Kasper D.L.
      • Blumberg R.S.
      How colonization by microbiota in early life shapes the immune system.
      However, as molecules from the maternal microbiome are transported in the blood and infiltrate every organ of the mother, those maternal microbial molecules influence the fetus long before the newborn acquires its own microbiota.
      • Ganal-Vonarburg S.C.
      • Hornef M.W.
      • Macpherson A.J.
      Microbial-host molecular exchange and its functional consequences in early mammalian life.
      The placenta is a barrier against infections and plays an essential role in the modulation of pregnancy immunity.
      • Burton G.J.
      • Jauniaux E.
      What is the placenta?.
      Placental dysfunctions are linked to complications, such as preeclampsia, intrauterine growth restriction, and stillbirth.
      • Fisher S.J.
      Why is placentation abnormal in preeclampsia?.
      • Ptacek I.
      • Sebire N.J.
      • Man J.A.
      • Brownbill P.
      • Heazell A.E.P.
      Systematic review of placental pathology reported in association with stillbirth.
      • Zhang S.
      • Regnault T.R.H.
      • Barker P.L.
      • et al.
      Placental adaptations in growth restriction.
      In 2014, the sterile womb hypothesis was challenged by Aagaard et al,
      • Aagaard K.
      • Ma J.
      • Antony K.M.
      • Ganu R.
      • Petrosino J.
      • Versalovic J.
      The placenta harbors a unique microbiome.
      who reported the detection of bacterial DNA sequences of multiple taxa in placental samples. A comparison of community patterns suggested the oral microbiome as the body site of origin. In addition, several studies have claimed to detect a distinct placental microbiome.
      • Aagaard K.
      • Ma J.
      • Antony K.M.
      • Ganu R.
      • Petrosino J.
      • Versalovic J.
      The placenta harbors a unique microbiome.
      • Bassols J.
      • Serino M.
      • Carreras-Badosa G.
      • et al.
      Gestational diabetes is associated with changes in placental microbiota and microbiome.
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      • Doyle R.M.
      • Alber D.G.
      • Jones H.E.
      • et al.
      Term and preterm labour are associated with distinct microbial community structures in placental membranes which are independent of mode of delivery.
      • Doyle R.M.
      • Harris K.
      • Kamiza S.
      • et al.
      Bacterial communities found in placental tissues are associated with severe chorioamnionitis and adverse birth outcomes.
      • Gomez-Arango L.F.
      • Barrett H.L.
      • McIntyre H.D.
      • Callaway L.K.
      • Morrison M.
      • Nitert M.D.
      Contributions of the maternal oral and gut microbiome to placental microbial colonization in overweight and obese pregnant women.
      • Parnell L.A.
      • Briggs C.M.
      • Cao B.
      • Delannoy-Bruno O.
      • Schrieffer A.E.
      • Mysorekar I.U.
      Microbial communities in placentas from term normal pregnancy exhibit spatially variable profiles.
      • Prince A.L.
      • Ma J.
      • Kannan P.S.
      • et al.
      The placental membrane microbiome is altered among subjects with spontaneous preterm birth with and without chorioamnionitis.
      • Satokari R.
      • Grönroos T.
      • Laitinen K.
      • Salminen S.
      • Isolauri E.
      Bifidobacterium and lactobacillus DNA in the human placenta.
      • Seferovic M.D.
      • Pace R.M.
      • Caroll M.
      • et al.
      Visualization of microbes by 16S in situ hybridization in term and preterm placentas without intraamniotic infection.
      • Tuominen H.
      • Rautava S.
      • Syrjänen S.
      • Collado M.C.
      • Rautava J.
      HPV infection and bacterial microbiota in the placenta, uterine cervix and oral mucosa.
      • Zheng J.
      • Xiao X.
      • Zhang Q.
      • Mao L.
      • Yu M.
      • Xu J.
      The placental microbiome varies in association with low birth weight in full-term neonates.
      In contrast, others have not confirmed these findings and have attributed observed bacterial signals to background contamination arising from reagents used during sample analysis or from the delivery method (in this case, vaginal delivery).
      • Kuperman A.A.
      • Zimmerman A.
      • Hamadia S.
      • et al.
      Deep microbial analysis of multiple placentas shows no evidence for a placental microbiome.
      • de Goffau M.C.
      • Lager S.
      • Sovio U.
      • et al.
      Human placenta has no microbiome but can contain potential pathogens.
      • Lager S.
      • de Goffau M.C.
      • Sovio U.
      • et al.
      Detecting eukaryotic microbiota with single-cell sensitivity in human tissue.
      • Lauder A.P.
      • Roche A.M.
      • Sherrill-Mix S.
      • et al.
      Comparison of placenta samples with contamination controls does not provide evidence for a distinct placenta microbiota.
      • Leiby J.S.
      • McCormick K.
      • Sherrill-Mix S.
      • et al.
      Lack of detection of a human placenta microbiome in samples from preterm and term deliveries.
      • Leon L.J.
      • Doyle R.
      • Diez-Benavente E.
      • et al.
      Enrichment of clinically relevant organisms in spontaneous preterm-delivered placentas and reagent contamination across all clinical groups in a large pregnancy cohort in the United Kingdom.
      • Theis K.R.
      • Romero R.
      • Winters A.D.
      • et al.
      Does the human placenta delivered at term have a microbiota? Results of cultivation, quantitative real-time PCR, 16S rRNA gene sequencing, and metagenomics.
      Such contradictions are expected, because samples from tissue with bacterial load close to the limit of detection, even with modern detection techniques, are particularly prone to revealing background signals, such as unspecific binding of probes or residual molecules left in the analytical apparatus after disinfection.
      • Salter S.J.
      • Cox M.J.
      • Turek E.M.
      • et al.
      Reagent and laboratory contamination can critically impact sequence-based microbiome analyses.
      Furthermore, many of the studies claiming that microbiota are present in the placenta have included results from preterm or other complicated pregnancies.
      • Doyle R.M.
      • Alber D.G.
      • Jones H.E.
      • et al.
      Term and preterm labour are associated with distinct microbial community structures in placental membranes which are independent of mode of delivery.
      ,
      • Prince A.L.
      • Ma J.
      • Kannan P.S.
      • et al.
      The placental membrane microbiome is altered among subjects with spontaneous preterm birth with and without chorioamnionitis.
      ,
      • Tuominen H.
      • Rautava S.
      • Syrjänen S.
      • Collado M.C.
      • Rautava J.
      HPV infection and bacterial microbiota in the placenta, uterine cervix and oral mucosa.
      In addition, studies that have reported mainly negative findings and a small number of microbial species have included placentas from healthy pregnancies, and the load has generally been considered too low to be characterized as a microbial community.
      • Seferovic M.D.
      • Pace R.M.
      • Caroll M.
      • et al.
      Visualization of microbes by 16S in situ hybridization in term and preterm placentas without intraamniotic infection.
      ,
      • de Goffau M.C.
      • Lager S.
      • Sovio U.
      • et al.
      Human placenta has no microbiome but can contain potential pathogens.
      ,
      • Leiby J.S.
      • McCormick K.
      • Sherrill-Mix S.
      • et al.
      Lack of detection of a human placenta microbiome in samples from preterm and term deliveries.
      In summary, there is currently no consensus regarding the existence of a placental microbiome in healthy full-term pregnancies.
      Amniotic fluid (AF) is another essential factor in the maintenance of intrauterine homeostatic conditions that has been considered sterile,
      • Prevedourakis C.N.
      • Strigou-Charalabis E.
      • Kaskarelis D.B.
      Bacterial invasion of amniotic cavity during pregnancy and labor.
      but recent studies have had conflicting results.
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      ,
      • Lim E.S.
      • Rodriguez C.
      • Holtz L.R.
      Amniotic fluid from healthy term pregnancies does not harbor a detectable microbial community.
      Both the AF and vernix caseosa, covering the fetal skin during the last trimester of pregnancy, contain potent substances with broad-spectrum antimicrobial effect.
      • Yoshio H.
      • Tollin M.
      • Gudmundsson G.H.
      • et al.
      Antimicrobial polypeptides of human vernix caseosa and amniotic fluid: implications for newborn innate defense.
      The aim of the current study was to investigate the existence and origin of a placental microbiome. Comprehensive samples were collected from the amniotic sac and placenta to determine potential bacterial content. In addition, samples from the oral cavity, gut, and vagina were collected from each woman to identify a possible source of the bacteria. We compared the specimens from cesarean and vaginal deliveries using polymerase chain reaction (PCR) amplification and sequencing (quantitative PCR [qPCR] of the 16S rRNA gene, bacterial taxa-specific qPCR, metabarcoding based on 16S rRNA gene sequencing) and bacterial culture experiments, and we collected detailed information about the mothers’ clinical backgrounds to avoid possible confounding factors in our results.

      Materials and Methods

      Study design and participants

      This study used a cross-sectional design to investigate whether there is a placental microbiome in human pregnancies at term and the potential reasons for the placenta being sterile or not by comparing samples from 2 groups of term births: women not in labor with elective cesarean deliveries and women with vaginal deliveries. The inclusion criteria for both groups were as follows: full-term pregnancies, maternal age more than 18 years, Swedish or English speaking. The exclusion criteria were as follows: knowledge of fetal pathology and acute cesarean delivery (Figure 1). All the participants were attending the maternity clinic at Södersjukhuset, Stockholm, Sweden, between March 2017 and October 2017 and were included in the study after receiving information and signing a consent form on the day of the delivery or the day earlier. Data on background and health status were collected from the participants’ medical records. The study was approved by the independent regional Research Ethics Committee, Karolinska Institutet, Stockholm, Sweden (2015/2043-31/2) and complied with the World Medical Association Declaration of Helsinki regarding ethical conduct of research involving human subjects. All the study participants received oral and written information about the study and the sample protocol or analysis and signed a written consent form.
      Figure thumbnail gr1
      Figure 1Study design and workflow for sample collection and analyses
      Inclusion and exclusion criteria for the participants are indicated in the upper panel. The number of samples collected from each fluid or tissue source, how they were preserved, and the analysis performed are listed in the lower panel. aOne of the mothers had a dichorionic diamniotic twin delivery; bStorage temperature.
      qPCR, quantitative polymerase chain reaction.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.

      Sample collection

      An overview of the sample collection, storage, and analyses performed is shown in Figure 1. All samples were collected by 2 obstetricians (I.S. and E.W.I.). The samples were collected in the exact order described below.

      Cesarean delivery group

      In the elective cesarean delivery group, saliva samples were collected before the surgery using a SalivaGene Collector (STRATEC Molecular GmbH, Birkenfeld, Germany) containing lyophilized DNA stabilization buffer according to the manufacturer’s instructions. The vaginal and rectal swab samples were taken in the operating theater, before the start of surgery, by inserting the swab (FLOQSwabs, Copan Flock Technologies, Brescia, Italy) 2 to 3 cm into the vagina or anus and swirling the swab for 10 seconds. Swabs were directly put into FluidX tubes (Brooks Life Sciences, Chelmsford, MA) containing 0.8 mL DNA/RNA Shield (Zymo Research, Irvine, CA). Approximately 1.5 mL of AF was collected by aspirating with a sterile syringe inserted at the incision site directly following the uterotomy. AF was then immediately placed into empty, prebarcoded FluidX tubes (Brooks Life Sciences) and frozen at −80°C.
      Placenta samplings were performed by 2 experienced obstetricians according to a previously used protocol that was slightly adapted to match the purpose of our study.
      • Aagaard K.
      • Ma J.
      • Antony K.M.
      • Ganu R.
      • Petrosino J.
      • Versalovic J.
      The placenta harbors a unique microbiome.
      ,
      • Burton G.J.
      • Sebire N.J.
      • Myatt L.
      • et al.
      Optimising sample collection for placental research.
      The placentas were collected in sterile containers using sterile gloves and instruments for the dissections, which took place in a room (open room environment) adjacent to the operating theater. Three 1×1–cm cross-sectional tissue samples were circumferentially excised, each at about 4 cm from the cord insertion site. Each sample was then dissected into 3 layers: maternal, middle, and fetal. Each layer was cut into 1 set of 3 biopsies, corresponding to 1 cross-section, and immediately stored in 1 of the following conditions to allow potential detection of bacteria in placental tissue: (1) 0.8 mL DNA/RNA Shield for microbiome analyses (Zymo Research), (2) freezing medium for sensitive bacteria MIK1106 (Karolinska University Hospital substrate unit; containing albumin, bovine serum albumin 25 g, sucrose 74 g, potassium dihydrogen phosphate 0.5 g, potassium hydrogen phosphate 1.2 g, sodium glutamate hydrate 0.6 g, in 1 L with a pH of 7.1). Culture experiments are shown in Figure 1.
      The vernix caseosa samples were obtained by rubbing a FLOQSwab (Copan Flock Technologies) in the armpit or groin of the newborns within the first 10 minutes after delivery, in the operating theater. The swab was immediately placed in a FluidX tube (Brooks Life Sciences) with 0.8 mL DNA/RNA Shield and frozen at −80°C.

      Vaginal delivery group

      The placentas were delivered vaginally and collected in sterile containers and covered with a sterile cloth. The placentas were then directly transported to an adjacent room. All dissections were performed using sterile gloves and surgical instruments in the manner described above. AF was collected vaginally with a sterile syringe after rupture of the amniotic membranes, before the baby was born. Samples of AF that were mixed with blood were excluded (n=6). All other samples (vernix caseosa, saliva, vaginal, and rectal samples) were collected during delivery as described above. Samples were kept at −80°C until analysis.

      Deoxyribonucleic acid extraction, amplification, and sequencing

      DNA extraction was performed using the Quick DNA Magbead Plus Kit (Zymo Research, Irvine, CA) according to manufacturer’s instructions, with the following modifications: placental specimens were homogenized with lysing matrix A from MP Biomedicals, LLC (Valiant, China), 3 times for 2 minutes. Vernix caseosa and AF were homogenized with ZR BashingBeads (lysis matrix used to lyse the bacteria cell wall; Zymo Research, Irvine, CA) for 1 minute. Furthermore, 200 μL of the pretreated placental specimens were incubated with 20 μL of lysozyme (100 mg/mL) at 37°C for 60 minutes. For the vernix caseosa and AF, 650-μL samples were incubated with 40 μL of lysozyme solution. Samples were then incubated with 40 μL proteinase K (20 mg/mL), using double the amount of solid tissue buffer II (advised by the manufacturer) at 55°C for 90 minutes (placenta) or double the amount of biofluid and cell buffer II at 55° for 30 minutes (vernix caseosa and AF). DNA purification was performed in Freedom EVO (Tecan, Männedorf, Switzerland) according to the Zymo protocol, and samples were eluted in an EB buffer (Qiagen, Hilden, Germany). Positive controls (ZymoBIOMICS Microbial Community Standard, D6300, Zymo Research, Irvine, CA; a commercial mixture of bacterial colonies containing gram-positive and gram-negative species) and negative controls (DNA/RNA Shield only) were processed with each extraction plate. The resulting sequencing data have been submitted to the European Nucleotide Archive under project PRJEB38528, accession numbers ERX4191918 to ERX4192266.

      Quantitative analysis of bacterial communities by quantitative polymerase chain reaction and 16S ribosomal ribonucleic acid gene sequencing

      Universal 16S rRNA gene primers covering the V6-V8 region (described by Huys et al
      • Huys G.
      • Vanhoutte T.
      • Joossens M.
      • et al.
      Coamplification of eukaryotic DNA with 16S rRNA gene-based PCR primers: possible consequences for population fingerprinting of complex microbial communities.
      ; Integrated DNA Technologies [IDT], Coralville, IA) were used because of their good characteristics when amplifying samples with very low microbial content and high human background (V6-V8 primers are more specific to bacterial DNA, whereas V3-V4 primers can amplify human mitochondrial DNA). Therefore, when aiming to quantify a total bacterial load in a sample rich in human content, the V6-V8 primer shows more specificity and should be the primer of choice (Supplemental Figure 1). In addition to the extraction controls, PCR-positive controls (ZymoBIOMICS Microbial Community DNA Standard, D6305, Zymo Research, Irvine, CA; a commercial mixture of DNA from different bacterial colonies) and PCR-negative controls (PCR-grade water, W4502, Sigma-Aldrich, St. Louis, MO) were amplified and sequenced together with the samples. Real-time PCR experiments were performed in a LightCycler 480 using the SYBR Green assay from Bio-Rad (1725270, Bio-Rad Laboratories, Hercules, CA). The qPCR settings for the V6-V8 amplification were as follows: 98°C preincubation for 3 minutes, 98°C melting for 10 seconds, 57°C annealing for 15 seconds, 72°C extension for 40 seconds, and a total of 35 cycles. As a positive control and for quantification of bacterial content, the DNA standard from Zymo (ZymoBIOMICS Microbial Community DNA Standard, D6305, Zymo Research; a commercial mixture of DNA from different bacterial colonies) was expressed in a plasmid using the TOPO TA Cloning Kit (Invitrogen, K457501, Carlsbad, CA), purified using the Plasmid Miniprep Kit I (VWR, 732-2780, Radnor, PA), and quantified using the Qubit system (dsDNA high sensitive, Q32854, Thermo Fisher Scientific, Waltham, MA). After quantification, the DNA was normalized and used as a standard for the quantification of the samples. As a negative control for the reaction, PCR-grade water (W4502, Sigma-Aldrich) and pure human DNA (Sigma 11691112001, Sigma-Aldrich, St. Louis, MO) were used. For background subtraction (because reagents used in the extraction may contain trace amounts of bacterial DNA), negative DNA extraction controls were also submitted to qPCR.
      To further investigate the bacterial content representing the positive results from the qPCR of the 16S rRNA gene, selected taxa were amplified using hydrolysis probes designed for specific bacterial targets. The experiments were conducted using the same equipment as the total 16S rRNA sequencing experiments, and more information regarding the probes and PCR conditions can be found in Supplemental Table 1. For this experiment, we used a gBlock (IDT) sequence specific to each bacterium as the standard for quantification. Sequences and references are presented in Supplemental Tables 2 and 3. The probe selection was limited by the amount of remaining DNA and by the characteristics of the 16S rRNA gene for each taxon (specificity, size of the amplified fragment, guanine-cytosine (GC) content, etc.).
      For the sequencing procedure, we applied a 2-step PCR approach. We selected the maternal side of the placenta for sequencing because it had the highest sequencing signal during qPCR analysis (Figure 2). Women given antibiotics during delivery were excluded from this analysis (n=6). The input DNA used was 75 ng, and the first PCR was performed under conditions similar to those used for qPCR, except for the polymerase used (High-Fidelity Master Mix, F-565L, Thermo Fisher Scientific, Stockholm, Sweden) and the number of cycles (25 cycles). The samples were then barcoded using the Nextera XT Kit following Illumina’s standard protocol (12 PCR cycles; Illumina, 15052163, San Diego, CA) and sequenced on an Illumina MiSeq (Illumina, San Diego, CA) with V3 chemistry and 2×300 bp reads. Samples generated 1020 to 207,009 reads per sample (median, 23,303). Sequenced placental samples are available from the European Nucleotide Archive under project number PRJEB38528, accession numbers ERX4191918-ERX4192266.
      Figure thumbnail gr2
      Figure 2AF and placental specimens may contain bacterial DNA, at concentrations undistinguishable from negative controls
      Quantification of total bacterial abundance by 16S rRNA gene copies by qPCR (log 10) on placental, AF, and vernix caseosa samples. The boxes represent median and interquartile range; blue boxes represent cesarean deliveries, and yellow boxes represent vaginal deliveries. Because samples were randomized for extraction, the negative extraction controls (gray) cannot be separated by delivery type. The thin dotted horizontal line shows the level of bacterial DNA detected in pure commercial human DNA.
      AF, amniotic fluid; qPCR, quantitative polymerase chain reaction; rRNA, ribosomal ribonucleic acid.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.

      Microbial profiling of maternal saliva, vaginal, and fecal samples using shallow shotgun sequencing

      DNA extraction was performed using the Quick DNA Magbead Plus Kit (Zymo Research) according to manufacturer’s instructions, with few modifications. Each sample type was bead-beaten with a different matrix: saliva (600 μL) was homogenized with lysing matrix B (MP Biomedicals, LLC, Valiant, China); vaginal swabs with ZR BashingBead Lysis matrix (Zymo Research); and rectal swabs with matrix E (Nordic Biolabs, Täby, Sweden). Each sample was bead-beaten at 1600 revolutions per minute for 1 minute before DNA extraction using a FastPrep-96 homogenizer (SKU 116010500, MP Biomedicals, Santa Ana, CA). After bead beating, samples were treated with lysozyme (100 mg/mL, at 37°C for 60 minutes) and proteinase K (20 mg/mL, at 55°C for 30 minutes) previous to extraction using Freedom EVO (Tecan). Extracted DNA was shipped to CoreBiome (OraSure, Bethlehem, PA) and processed with their BoosterShot shallow shotgun-sequencing technology. The resulting sequencing data have been submitted to the European Nucleotide Archive under project PRJEB38528, accession numbers ERX4192267 to ERX4192489.

      Bioinformatics

      Cutadapt (version 1.4)
      • Martin M.
      Cutadapt removes adapter sequences from high-throughput sequencing reads.
      was used to trim 3’-bases with a Phred score of <15, remove primer sequences, and discard reads not containing the primers. The resulting quality-trimmed reads were processed with the R package DADA2 (version 1.11.3)
      • Callahan B.J.
      • McMurdie P.J.
      • Rosen M.J.
      • Han A.W.
      • Johnson A.J.A.
      • Holmes S.P.
      DADA2: high-resolution sample inference from Illumina amplicon data.
      to correct errors, remove chimeras, and produce amplified sequence variants (ASVs). The resulting ASV data were assessed for contamination using the Decontam package (version 1.1.2)
      • Davis N.M.
      • Proctor D.M.
      • Holmes S.P.
      • Relman D.A.
      • Callahan B.J.
      Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data.
      using the function isNotContaminant, which is suitable for samples with low bacterial load, and the prevalence method.

      Placental tissue culture and bacterial deoxyribonucleic acid extraction

      Here, 3 pieces of the placental tissue (around 1×1 cm, maternal, middle, and fetal) from each participant (n=76) collected in freezing medium for sensitive bacteria MIK1106 (Karolinska University Hospital substrate unit) were inoculated by smearing the tissue on GC agar plates (Karolinska Hospital, MIK0346, Sweden) under a sterile cell culture biosafety cabinet. The plates were then incubated anaerobically at 37°C for 48 hours. GC plates with open lids during the inoculation period were used as negative controls. After incubation, present colonies were enumerated and collected for DNA extraction, using the same DNA extraction protocol applied for saliva. The DNA material from the colonies was sequenced using 16S rRNA gene sequencing, as described above.

      Bacterial growth inhibition assay

      Escherichia coli (ATCC 25922) was used in the study of growth inhibition. In total, 12 frozen placental specimens (maternal and fetal sides from 6 cesarean deliveries) were selected on the basis of the qPCR results (3 high vs 3 low bacterial content in each side). E coli was cultured overnight aerobically in 2 mL Luria-Bertani (LB) broth (Sigma L3522) at 37°C. Furthermore, the E coli culture was diluted to 100 colony-forming units (CFUs)/mL on the basis of their optical density measurement to mimic the low bacterial load in the uterine cavity. From the diluted cultures, 10 CFUs in 100 μL were added into each well of the 96-well plate (Thermo Fisher Scientific, 167008). Subsequently, a piece of frozen placental tissue (around 5 mm in diameter) was added into each well for the inhibition experiment. At the same time, 100 μL culture media and placental tissues without bacteria were used as negative controls. In addition, LB broth and LB broth inoculated with 10 CFUs of E coli with and without 10 μg/mL gentamycin (Sigma G1272) were also carried out as controls. All these conditions, including each of the 12 frozen placentas and corresponding controls, were performed in triplicate. After 24 hours’ co-culture according to the bacteria growth condition described above, CFUs were enumerated by serial dilution as described previously.
      • Hu Y.O.O.
      • Hugerth L.W.
      • Bengtsson C.
      • et al.
      Bacteriophages synergize with the gut microbial community to combat Salmonella.

      Statistical analyses and figures

      For the descriptive data, chi-squared tests were used to compare categorical variables between the study groups, and Mann-Whitney tests were applied for the continuous variables, using Statistical Product and Service Solutions (SPSS) (version 25.0, IBM, Armonk, NY). R libraries RColorBrewer (version 1.1-2), vioplot (version 0.2), and treemap (version 2.4-2) were used to create the figures. For qPCR, a 1-sided Kruskal-Wallis test with alternative “greater” was applied using R (version 3.5.2). Bacterial culture comparisons between samples from cesarean and vaginal deliveries in each CFU range group were computed using Fisher’s exact test. The Mann-Whitney test was performed for comparison between all placenta groups (regardless of maternal or fetal side and of high or low bacterial content in the qPCR) with E coli LB control in the bacterial growth inhibition assay using SPSS (version 23.0; IBM). The level of statistical significance was set at P<.05. When triplicates were used, the average value of each triplicate was inserted. In addition, the Kruskal-Wallis H test, followed by posthoc tests using the Dunn multiple comparisons test, was carried out for pairwise comparison between the groups, with the significance level set at P<.05.

      Results

      Characteristics of participants

      Maternal and infant characteristics are presented in Table 1. In total, 76 women with normal pregnancies and 77 infants were included. There were no significant differences regarding background characteristics between the 2 groups (elective cesarean delivery and vaginal delivery) except regarding parity (P=.04). All newborns had normal birthweights, except 1 baby from the twin pregnancy, who was small for gestational age. None of the newborns needed neonatal care. As expected, onset of labor, mode of delivery, and gestational length differed between the groups.
      Table 1Demographics of the 76 participants and their 77 infants
      VariablesElective cesarean delivery (n=50)Vaginal delivery (n=26)P value
      Chi-squared tests were used for categorical variables and Mann-Whitney tests were applied for the continuous variables
      Maternal characteristics
       Age (y)34.5 (23.0–47.0)31.0 (21.0–49.0).4
       BMI in early pregnancy23.1 (19.1–45.2)23.3 (18.5–36.8).8
       Smoking in early pregnancy1 (2.0)0 (0)1.0
       History of psychiatric disease7 (14.0)1 (3.8).2
       IVF5 (10.0)1 (3.8).6
      Parity.04
       Nulliparous17 (34.0)18 (69.2)
       Multiparous33 (66.0)8 (38.2)
      Twins
      Dichorionic diamniotic pregnancy
      1 (2.0)0 (0)
      Complications of pregnancy
      Diagnoses included preeclampsia (n=1), hypertension (n=2), and cholestasis of pregnancy (n=2)
      2 (4.0)3 (11.5).3
      GBS in pregnancy2 (4.0)1 (3.8)1.0
      Preexisting comorbidities
      Diagnoses included asthma, hypothyroidism, inflammatory bowel disease, and diabetes
      10 (20)4 (15.4)1.0
      Antibiotics during pregnancy
      Treatment with penicillin in all cases except 3 that received broad-spectrum antibiotics
      .2
       Total9 (18.0)3 (11.5)
       First trimester21
       Second trimester40
       Third trimester32
      Antibiotics during delivery
      Elective cesarean deliveries treated with broad-spectrum antibiotic (cefuroxime) and vaginal deliveries treated with benzylpenicillin
      3 (6.0)3 (11.5).4
      Onset of labor<.01
       Spontaneous0 (0)18 (69.2)
       Induced0 (0)8 (30.8)
       Planned cesarean delivery50 (100.0)0 (0)
      Mode of delivery<.01
       Spontaneous vaginal0 (0)23 (88.5)
       Vacuum0 (0)3 (11.5)
       Cesarean50 (100.0)0 (0)
      Fetal characteristics
       Gestational age (d)
      Planned cesarean deliveries were performed around 273 days of gestation.
      272 (262–282)283 (262–296)<.01
       Birthweight (g)3680 (2225–4450)3620 (3140–5282).9
       Female sex20 (39.2)15 (57.7).1
      Continuous variables are presented as median (minimum-maximum); categorical variables are presented as number of participants (percentages).
      BMI, body mass index; GBS, group B Streptococcus; IVF, in vitro fertilization.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      a Chi-squared tests were used for categorical variables and Mann-Whitney tests were applied for the continuous variables
      b Dichorionic diamniotic pregnancy
      c Diagnoses included preeclampsia (n=1), hypertension (n=2), and cholestasis of pregnancy (n=2)
      d Diagnoses included asthma, hypothyroidism, inflammatory bowel disease, and diabetes
      e Treatment with penicillin in all cases except 3 that received broad-spectrum antibiotics
      f Elective cesarean deliveries treated with broad-spectrum antibiotic (cefuroxime) and vaginal deliveries treated with benzylpenicillin
      g Planned cesarean deliveries were performed around 273 days of gestation.

      16S ribosomal ribonucleic acid gene quantitative amplification and sequencing could not conclusively distinguish any microbial content of placenta from background signals

      There was a high degree of variability in the gene counts observed by qPCR from the negative extraction controls (Figure 2). No significant difference in copy number was observed between any of the sample types from the placenta, AF, or vernix caseosa and the negative controls. The median gene counts were highest for the maternal side of the placenta and for AF, regardless of the mode of delivery.
      To further characterize the bacterial signal observed in the maternal side of the placenta, samples from women who had not taken antibiotics during pregnancy or delivery (n=45 cesarean-delivered infants; n=19 vaginally delivered infants) were used for 16S rRNA gene sequencing. Of 964 observed ASVs, 835 were flagged as contaminants by the Decontam software, corresponding to 3.7% to 67% of each sample (median, 57%) (Figure 3). A large fraction of the remaining tags came from the genus Massilia (0%–45% of total sample counts; median, 35%), also a known laboratory contaminant.
      • Salter S.J.
      • Cox M.J.
      • Turek E.M.
      • et al.
      Reagent and laboratory contamination can critically impact sequence-based microbiome analyses.
      Another common laboratory contaminant observed in our samples was the genus Escherichia or Shigella.
      • Stinson L.F.
      • Keelan J.A.
      • Payne M.S.
      Identification and removal of contaminating microbial DNA from PCR reagents: impact on low-biomass microbiome analyses.
      Finally, regardless of the mode of delivery, we observed a variety of typical vaginal bacteria and opportunistic pathogens (Supplemental Figure 2).
      Figure thumbnail gr3
      Figure 3Most of the detected taxa in placenta are likely contaminants
      The maternal side of the placentas from women who had not taken antibiotics during pregnancy or delivery (n=45 for cesarean delivery; n=19 for vaginal delivery) was taxonomically profiled by sequencing of the V6-V8 region of the 16S rRNA gene. Samples from each individual are placed in the same order in each panel. “P” stands for participant and “t” for twin. A, Total 16S rRNA gene copies for each sample. B, Taxonomic profile of the placentas, expressed as proportion of total bacteria. No correlation is observed between the total bacterial quantification and the observed microbial profile. C is the same as B but excluding the most likely contaminants.
      rRNA, ribosomal ribonucleic acid.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      To further validate these findings, selected taxa were amplified by qPCR with taxon-specific primers and probes. Because the maternal side of the placenta gave the strongest signal with universal primers, the same samples were used for the amplification of 10 specific bacterial taxa suspected to be present in the placenta on the basis of sequencing data and previous studies (Supplemental Tables 2 and 3). In total, 21 of 48 cesarean-delivered and 11 of 22 vaginally delivered placentas were positive for at least 1 probe, yielding no statistical difference in the rate of positive samples (Fischer’s exact test; odds ration=0.78; P=.8) (Figure 4). There was also no difference between cesarean-delivered and vaginally delivered samples with regard to the number of positive signals per positive sample nor on the strength of these signals (Figure 4).
      Figure thumbnail gr4
      Figure 4Selected bacterial taxa could be detected in low numbers in the maternal side of a few placentas
      Probe-based, taxon-specific qPCR was used to detect DNA from bacterial taxa suspected to be present in the placenta on the basis of sequencing data and previous studies. and provide a list of probe-based qPCR primers, probes, and gBlocks. Bacterial concentration is expressed as log10 of 16S copy numbers per nanogram of total DNA, from 10-4 (blue) to 106 (red). “P” stands for participant and “t” for twin. The varying estimated copy numbers for each probe in negative samples depend on that probe’s specific binding characteristics and limits of detection. In addition to placental samples, a randomly selected negative extraction control from each plate was amplified (marked “Neg.Ext”). A negative amplification control is marked “Neg.Amp.” The 2 columns on the left depict the delivery type and extraction plate for each sample. Although the negative amplification control presents no signals, all the negative extraction controls have at least a few signals, mostly associated with typical skin bacteria. These weak signals correlate poorly to the signals detected in placentas extracted in the same plate, highlighting the high degree of variability observed when working in this concentration range.
      qPCR, quantitative polymerase chain reaction.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.

      The lack of a specific placental microbiome did not allow for a comparison of the bacterial content of saliva, vaginal, and fecal samples

      Vaginal swabs, rectal swabs, and saliva samples collected from women at the time of delivery were analyzed to identify the origin of the placental microbiome. Because we did not detect a specific placental microbiome, no meaningful comparison of these samples could be made. However, the bacterial composition of saliva, vaginal, and fecal samples is presented in Supplemental Figure 3. We also display the overlap between these body sites and bacteria detected in the placenta by at least 2 independent methods in Supplemental Table 4.

      Bacterial culture of placental specimens yielded typical vaginal and skin bacteria

      As shown in Figure 5, A and B , after culturing placental tissue on rich medium plates, all the plates were classified into 4 groups according to the number of bacterial colonies (CFUs): no bacteria, 1 to 5 CFUs, 6 to 30 CFUs, and >30 CFUs. Most placental specimens from the cesarean deliveries did not show any CFUs (n=87 of 152 [57.2%]), with only a small portion of tissues containing more than 30 colonies (n=10 of 152 [6.6%]) (Table 2). In contrast, a quarter of the placental tissues from vaginal delivery presented more than 30 CFUs (n=21 of 78 [26.9%]) (Table 2). In 3 of the placenta sections, there were significantly more bacteria observed from vaginally delivered samples than from cesarean-delivered samples (maternal side, P=.007; middle side, P=.041; fetal side, P<.001). Overall, significantly more placental specimens from the cesarean delivery mode had no bacterial colonies (P≤.001), whereas more specimens from the vaginal delivery mode had bacterial colonies over 30 CFUs (Table 2). For both modes of deliveries, no significant correlation was observed between the estimated bacterial load by qPCR and the number of colonies found in culture, except for a weak correlation on the fetal side (Pearson’s correlation; r=0.290; P=.041).
      Figure thumbnail gr5
      Figure 5Bacteria grown from placentas are predominantly typical skin and vaginal taxa
      A, GC agar plates showing the bacterial growth from placental tissues after 48 hours. The placenta cultures are represented in ranges according to the number of CFUs. “P” stands for participant. B, Histogram showing frequency (percentage) distributions of CFUs according to the placental sample type and the delivery mode. Chi-squared tests (Fisher exact tests) were performed with significance level at P=.05. The comparison was between vaginal delivery (white bars) and cesarean delivery (black bars) in each CFU range group. C, Treemaps showing the relative proportion of the taxa that grew in culture by location in the tissue and mode of delivery. Each area is colored according to the bacterial order, as shown in the legend, and the genus is overlaid on the boxes themselves.
      CFU, colony-forming unit; GC, guanine-cytosine.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Table 2Bacterial growth of placental tissues according to the delivery mode
      CFU rangeDelivery modeCesarean delivery vs vaginal delivery
      Cesarean delivery (N=152)
      One mother delivered after a dichorionic diamniotic pregnancy.
      Vaginal delivery (N=78)
      Maternal (n=51)Middle (n=50)Fetal (n=51)Maternal (n=26)Middle (n=26)Fetal (n=26)P value
      028 (54.9)29 (58.0)30 (58.8)8 (30.8)9 (34.6)4 (15.4)<.001
      1–517 (33.4)13 (26.0)10 (19.6)7 (26.9)5 (19.2)5 (19.2).787
      6–304 (7.8)3 (6.0)8 (15.7)3 (11.5)6 (23.1)10 (38.5).996
      >302 (3.9)5 (10.0)3 (5.9)8 (30.8)6 (23.1)7 (26.9)<.001
      Data are expressed as absolute and relative values for total plate count number (percentages). P value shows the differences between delivery groups at the level of the CFU range. The significance level was set at P<.05. The Fisher exact test was used for comparisons.
      CFU, colony-forming unit.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      a One mother delivered after a dichorionic diamniotic pregnancy.
      The 16S rRNA sequencing data on the collected colonies revealed that most of the bacteria that grew on the plates from cesarean-delivered placentas were genera typically found on human skin, such as Propionibacterium, Streptococcus, and Staphylococcus (Figure 5, C), except for Gardnerella, which was retrieved in 1 fetal side tissue, and Bifidobacterium, which was found in 1 piece from the maternal side. In contrast, placentas delivered vaginally had a high prevalence of Lactobacillus and Gardnerella, in addition to the same genera presented in cesarean-delivered samples (Figure 5, C; Supplemental Figure 4). In general, cultures obtained from vaginal deliveries had higher bacterial richness (number of unique ASVs; median 10 [interquartile range (IQR), 6–33] vs median 3 [IQR, 0–9] unique sequences or plate; Kruskal-Wallis, P=10-5) (Supplemental Figure 4).

      Lack of agreement between sequencing and culturing data

      Very few genera detected by 16S sequencing could be confirmed by culturing. These are Pelomonas, which could be found by 16S sequencing in 7 of 8 samples where it could be cultured, Massilia (10 of 11), Leifsonia (5 of 5), and Escherichia or Shigella (12 of 13). All of these were detected by sequencing in >40 samples each, and confirmed by culture in a much smaller set. The overlap between clades detected by qPCR and sequencing is typically 1 sample per genus investigated. In particular, Escherichia of Shigella, although abundant in sequencing data and cultures, was detected in very few samples by qPCR.
      A direct comparison of the 16S gene sequencing data from cesarean-delivered placentas to the metagenomic sequencing data of saliva, feces, and vaginal samples is not possible. However, it is noteworthy that there are a few signals that were confirmed by at least 2 methods and not recovered in any other body site, such as Acinetobacter (n=2), Cupriavidus (n=3), Gemmatirosa (n=1), and Pelomonas (n=7). All of these are typically free-living bacteria.

      Bacterial growth inhibition was observed with placental tissue

      We did not find a strong bacterial signal in cesarean-delivered samples, but we observed a large variation of qPCR quantification in placental tissues, some of which had a much higher bacterial load than others. Thus, we tested whether the placenta could inhibit bacterial growth and whether this effect is influenced by the bacterial load according to our qPCR analysis (Figure 6, A). We selected frozen placental tissues (both maternal [PM] and fetal [PF] side) from 3 cesarean deliveries with relatively high 16S rRNA gene load (PM high and PF high) according to our qPCR results and 3 deliveries with low 16S rRNA gene load (PM low and PF low). Growth inhibition analysis showed that co-culture with placental tissues resulted in significantly less E coli growth than bacteria that grew without placental tissue (P<.01) (Figure 6, B).
      Figure thumbnail gr6
      Figure 6The placenta presents antimicrobial activity
      A, Total 16S rRNA gene counts of specimens from the fetal, middle, or maternal side of each individual placental sample (51 maternal, 50 middle, and 49 fetal), as determined by qPCR. Each dot represents 1 sample. The individuals whose samples were later used for inhibition experiments in (B) are labeled with their participant number. B, Twelve frozen placental specimens from 6 cesarean deliveries were selected on the basis of the qPCR results; 3 high vs 3 low bacterial content in the fetal and maternal sides (as shown in A). The inhibition effect on the proliferation of E coli cultured in LB alone was assessed by co-culturing the selected placental tissues with 10 CFUs of E coli/well in a 96-well plate. Controls were LB broth inoculated with 10 CFUs of E coli and 10 μg/mL gentamycin, LB broth alone, and placental tissues alone. All culture conditions were performed in triplicate. After incubation for 24 hours at 37°C, E coli CFUs from each well were enumerated. Data are presented as median±interquartile range in a log 10 scale, and each dot represents the average of the triplicates. Statistical comparisons were performed between E coli grown in LB and each placenta group and between E coli grown in LB and all placenta groups together. The asterisk indicates P<.05, whereas the double asterisk indicates P<.01.
      CFU, colony-forming unit; LB, Luria-Bertani broth; PF, the fetal side of placenta; PM, the maternal side of placenta; qPCR, quantitative polymerase chain reaction.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      When comparing the E coli growth between each group (PM high, PF high, PM low, PF low, and no placenta control), using a Kruskal-Wallis H test, an overall significant difference was observed (P=.033). In addition, posthoc tests comparing each placenta group with the control group showed that only placental tissues with high 16S rRNA gene load had a significant inhibition effect compared with E coli in LB broth (E coli in LB vs PM high; P=.042; E coli in LB vs PF high; P=.048). Other comparisons did not reach statistical significance but showed the following trend: the placental tissues (both maternal and fetal sides) with high bacterial gene load had a stronger inhibition effect than the tissues (both maternal and fetal sides) with low bacterial gene load.

      Comment

      Principal findings

      This study was designed to investigate the potential presence of a placental microbiome in samples from women not in labor with elective cesarean delivery vs samples from women who underwent vaginal deliveries at term. Using qPCR and 16S rRNA gene sequencing, we found that bacterial signals from placental tissue were not distinguishable from background signals, except in some individual cases. Samples with detectable bacterial signals were present in both cesarean and vaginal deliveries and were not associated with any specific health conditions. The apparently random nature of the bacterial signal between different methods (culturing, qPCR, 16S rRNA gene sequencing), combined with the weakness of these signals, suggests that any bacteria present in the human placenta are at the limit of detection of most available technologies. Other investigations of the same placental tissue, using probe-based qPCR targeting commonly reported bacteria from the reproductive tract
      • Ravel J.
      • Gajer P.
      • Abdo Z.
      • et al.
      Vaginal microbiome of reproductive-age women.
      and commonly reported contaminants,
      • Salter S.J.
      • Cox M.J.
      • Turek E.M.
      • et al.
      Reagent and laboratory contamination can critically impact sequence-based microbiome analyses.
      ,
      • Stinson L.F.
      • Keelan J.A.
      • Payne M.S.
      Identification and removal of contaminating microbial DNA from PCR reagents: impact on low-biomass microbiome analyses.
      did not provide clear evidence of a group difference or of a specific microbiome.
      To obtain a comprehensive picture of potential bacterial colonization in utero, we also performed bacterial gene sequencing of the AF and vernix caseosa samples collected from the newborn. As for the placental samples, 16S rRNA gene quantitative amplification sequencing could not conclusively distinguish any microbial content of the AF and vernix caseosa from background signals. Thus, our conclusion is that bacterial presence in pregnancy at term is low and cannot be considered proof of the presence of a microbiome.

      Results: in the context of what is known

      Our findings are supported by several previous reports that were unable to distinguish between sample signals and contamination introduced during DNA purification.
      • Lager S.
      • de Goffau M.C.
      • Sovio U.
      • et al.
      Detecting eukaryotic microbiota with single-cell sensitivity in human tissue.
      ,
      • Lauder A.P.
      • Roche A.M.
      • Sherrill-Mix S.
      • et al.
      Comparison of placenta samples with contamination controls does not provide evidence for a distinct placenta microbiota.
      ,
      • Leon L.J.
      • Doyle R.
      • Diez-Benavente E.
      • et al.
      Enrichment of clinically relevant organisms in spontaneous preterm-delivered placentas and reagent contamination across all clinical groups in a large pregnancy cohort in the United Kingdom.
      A recent study of 500 placentas from human pregnancies concluded that the placenta has no microbiome but can contain pathogens.
      • de Goffau M.C.
      • Lager S.
      • Sovio U.
      • et al.
      Human placenta has no microbiome but can contain potential pathogens.
      Although there are studies that corroborate our findings, the literature regarding the existence of a microbiome in the placenta is controversial, and other studies in the field have presented opposite conclusions. Aagaard et al
      • Aagaard K.
      • Ma J.
      • Antony K.M.
      • Ganu R.
      • Petrosino J.
      • Versalovic J.
      The placenta harbors a unique microbiome.
      reported the detection of bacterial DNA sequences in placental samples from multiple taxa that were suggested to originate from the oral cavity. A possible reason for the contradictory findings is that some studies have included samples from preterm or other complicated pregnancies.
      • Doyle R.M.
      • Harris K.
      • Kamiza S.
      • et al.
      Bacterial communities found in placental tissues are associated with severe chorioamnionitis and adverse birth outcomes.
      ,
      • Prince A.L.
      • Ma J.
      • Kannan P.S.
      • et al.
      The placental membrane microbiome is altered among subjects with spontaneous preterm birth with and without chorioamnionitis.
      ,
      • Tuominen H.
      • Rautava S.
      • Syrjänen S.
      • Collado M.C.
      • Rautava J.
      HPV infection and bacterial microbiota in the placenta, uterine cervix and oral mucosa.
      Other circumstances that might differ between studies are the time between the delivery of the placenta and the sampling. One study investigated the effects of the start of the ischemic process (after the placenta has been parted from the maternal circulation) on the presence of bacteria in the placenta and concluded that placental samples should be processed within 10 minutes of the delivery to minimize the impact of the degradation process in the tissue.
      • Yung H.W.
      • Colleoni F.
      • Atkinson D.
      • et al.
      Influence of speed of sample processing on placental energetics and signalling pathways: implications for tissue collection.
      In summary, the sampling procedure, the time between delivery and sample conservation in the medium, and the type of preservation medium can impact the results.
      • Wolfe L.M.
      • Thiagarajan R.D.
      • Boscolo F.
      • et al.
      Banking placental tissue: an optimized collection procedure for genome-wide analysis of nucleic acids.
      In bacterial cultures from placental tissues examined in this study, some samples produced live bacterial colonies. However, only a few samples resulted in cultures with a high number of colonies. Notably, substantially fewer colonies were found in cesarean-delivered placental tissue cultures. In fact, most live colonies observed in cultures of placental tissue from cesarean deliveries were common skin bacteria (Propionibacterium, Streptococcus, and Staphylococcus),
      • Byrd A.L.
      • Belkaid Y.
      • Segre J.A.
      The human skin microbiome.
      which likely represent contaminations that occurred during surgery. Placental tissue cultures from vaginally delivered placentas also contained common vaginal microbiota species (Lactobacillus spp., Gardnerella, and Bifidobacterium),
      • Mei C.
      • Yang W.
      • Wei X.
      • Wu K.
      • Huang D.
      The unique microbiome and innate immunity during pregnancy.
      a plausible sign of contamination from vaginal delivery. This is supported by other studies that have reported data on culture or in situ hybridization experiments on placenta.
      • Satokari R.
      • Grönroos T.
      • Laitinen K.
      • Salminen S.
      • Isolauri E.
      Bifidobacterium and lactobacillus DNA in the human placenta.
      ,
      • Seferovic M.D.
      • Pace R.M.
      • Caroll M.
      • et al.
      Visualization of microbes by 16S in situ hybridization in term and preterm placentas without intraamniotic infection.
      The placental tissue has no lumen, and colonizing bacteria would be subject to elimination by circulating immune cells and other immune effectors, such as antimicrobial peptides and immunoglobulins from the blood.
      • Gensollen T.
      • Iyer S.S.
      • Kasper D.L.
      • Blumberg R.S.
      How colonization by microbiota in early life shapes the immune system.
      ,
      • Ganal-Vonarburg S.C.
      • Hornef M.W.
      • Macpherson A.J.
      Microbial-host molecular exchange and its functional consequences in early mammalian life.
      Our findings indicate that placental tissue has the capacity to inhibit bacterial growth. This effect was more pronounced in the tissue specimen with a higher 16S rRNA gene signal. It can be speculated that a higher number of bacteria in the tissue may also trigger a stronger immune reaction, resulting in a higher antibacterial inhibition effect. Thus, our findings reinforce the notion that the uterus during healthy pregnancy is sterile or at least a low bacterial milieu.
      Previous studies have used placental extracts and proven its anti-inflammatory capacity.
      • Goswami S.
      • Sarkar R.
      • Saha P.
      • et al.
      Effect of human placental extract in the management of biofilm mediated drug resistance - a focus on wound management.
      ,
      • Sharma K.
      • Mukherjee C.
      • Roy S.
      • De D.
      • Bhattacharyya D.
      Human placental extract mediated inhibition of proteinase K: implications of heparin and glycoproteins in wound physiology.
      Moreover, antimicrobial peptides, like human β-defensins, have been detected in AF.
      • Para R.
      • Romero R.
      • Miller D.
      • et al.
      Human β-defensin-3 participates in intra-amniotic host defense in women with labor at term, spontaneous preterm labor and intact membranes, and preterm prelabor rupture of membranes.
      • Soto E.
      • Espinoza J.
      • Nien J.K.
      • et al.
      Human β-defensin-2: a natural antimicrobial peptide present in amniotic fluid participates in the host response to microbial invasion of the amniotic cavity.
      • Varrey A.
      • Romero R.
      • Panaitescu B.
      • et al.
      Human β-defensin-1: a natural antimicrobial peptide present in amniotic fluid that is increased in spontaneous preterm labor with intra-amniotic infection.
      Similarly, fetal membranes have shown antimicrobial properties
      • Zare-Bidaki M.
      • Sadrinia S.
      • Erfani S.
      • Afkar E.
      • Ghanbarzade N.
      Antimicrobial properties of amniotic and chorionic membranes: a comparative study of two human fetal sacs.
      ,
      • Mao Y.
      • Hoffman T.
      • Singh-Varma A.
      • et al.
      Antimicrobial peptides secreted from human cryopreserved viable amniotic membrane contribute to its antibacterial activity.
      and the capacity to suppress group B Streptococcus.
      • Boldenow E.
      • Jones S.
      • Lieberman R.W.
      • et al.
      Antimicrobial peptide response to group B Streptococcus in human extraplacental membranes in culture.
      Furthermore, vernix caseosa possesses anti-inflammatory properties in vivo.
      • Yoshio H.
      • Tollin M.
      • Gudmundsson G.H.
      • et al.
      Antimicrobial polypeptides of human vernix caseosa and amniotic fluid: implications for newborn innate defense.
      ,
      • Marchini G.
      • Lindow S.
      • Brismar H.
      • et al.
      The newborn infant is protected by an innate antimicrobial barrier: peptide antibiotics are present in the skin and vernix caseosa.
      Taken together, the physiology of the placenta and that of the uterus seem to discourage bacterial colonization—unlike tissues that are known to be colonized (such as the gut, skin, mouth, and vagina). In our study, we used multiple methods, including various molecular methods (total qPCR of the 16S rRNA gene, probe-specific qPCR for selected taxa, and 16S rRNA gene sequencing), to investigate the placental tissue and bacterial cultures derived from the tissue. Comparing the results obtained from qPCR with those obtained from 16S rRNA gene sequencing or culturing, the overlap among the taxa was very low. These highly diverse findings may have resulted from the extremely low signals, which were close to the limit of detection, creating stochastic observations of specific bacteria in samples. Some of our results, such as the Massilia findings, were not always algorithmically flagged as a contaminant in this study. However, Massilia is a well-known laboratory contaminant,
      • Salter S.J.
      • Cox M.J.
      • Turek E.M.
      • et al.
      Reagent and laboratory contamination can critically impact sequence-based microbiome analyses.
      which is often observed in our negative extraction controls. Many species in this genus were initially isolated from air samples, suggesting that they easily spread by this route and can therefore be difficult to completely eliminate. These previous observations, combined with the even distribution of Massilia spp. across almost all samples, make it highly unlikely that it is a true signal. Massilia spp. are found in soils or fresh water. However, various species have been isolated from patient material, such as blood, bone, cerebrospinal fluid, and intraocular fluid.
      • Kämpfer P.
      • Lodders N.
      • Martin K.
      • Falsen E.
      Massilia oculi sp. nov., isolated from a human clinical specimen.
      Massilia are not typically reported as a human commensal bacterium.

      Clinical implications

      It can be hypothesized that bacteria that enter the placenta are killed or inhibited during normal healthy gestation and that this function is strengthened during the course of gestation, because the placenta serves as a tool with which the immune system protects the fetus from microbes.
      • Mor G.
      • Kwon J.Y.
      Trophoblast-microbiome interaction: a new paradigm on immune regulation.
      In an experimental study using Salmonella enterica serovar Typhimurium to infect human placentas, viable bacteria were recovered from 100% of the placental explants exposed to the strain, but the bacterial numbers obtained from first trimester of pregnancy tissues were markedly higher than those obtained from second trimester and term placental of pregnancy tissues,
      • Perry I.D.
      • Nguyen T.
      • Sherina V.
      • et al.
      Analysis of the capacity of Salmonella enterica Typhimurium to infect the human placenta.
      indicating an evolving immune function of the placenta during the course of pregnancy.
      • Gotsch F.
      • Romero R.
      • Kusanovic J.P.
      • et al.
      The fetal inflammatory response syndrome.
      Moreover, a recent study suggested that conditions in the fetal gut strongly limit bacteria.
      • Rackaityte E.
      • Halkias J.
      • Fukui E.M.
      • et al.
      Viable bacterial colonization is highly limited in the human intestine in utero.
      A possible explanation is the continuous exposure of the fetal gut to AF. Another important piece of evidence pointing toward a bacterial colonization of the infant only after birth is the difference in early microbial colonization between infants born vaginally and those born through cesarean delivery.
      • Dominguez-Bello M.G.
      • Costello E.K.
      • Contreras M.
      • et al.
      Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns.
      Finally, the ability to breed germ-free mammals is strong evidence that the fetus is sterile. Germ-free neonates have been generated for mice, rats, rabbits, guinea pigs, cats, dogs, lambs, pigs, calves, goats, baboons, chimpanzees, marmosets, and humans (reviewed in Perez-Muñoz et al
      • Perez-Muñoz M.E.
      • Arrieta M.C.
      • Ramer-Tait A.E.
      • Walter J.
      A critical assessment of the “sterile womb” and “in utero colonization” hypotheses: implications for research on the pioneer infant microbiome.
      ). There are cases in which, because of the expected severe immune deficiencies, human infants have been delivered in sterile conditions and kept in isolators where they remained sterile.
      • Kirk R.G.
      “Life in a germ-free world”: isolating life from the laboratory animal to the bubble boy.
      ,
      • Barnes R.D.
      • Bentovim A.
      • Hensman S.
      • Piesowicz A.T.
      Care and observation of a germ-free neonate.
      These facts are convincing evidence, if not for the sterile womb hypothesis, for the low bacterial presence in the womb. Thus, if the physiology of the placenta does not encourage bacterial colonization, it might be possible for bacteria to enter the uterine cavity but not to colonize it. This could explain the random presence of bacteria in a few samples during analysis, but it should not be considered a placental microbiome.
      In summary, despite the presence of a small number of bacteria in some samples, our interpretation of the results is that the placenta does not harbor a distinct microbiome. In accordance with the literature,
      • Kuperman A.A.
      • Zimmerman A.
      • Hamadia S.
      • et al.
      Deep microbial analysis of multiple placentas shows no evidence for a placental microbiome.
      • de Goffau M.C.
      • Lager S.
      • Sovio U.
      • et al.
      Human placenta has no microbiome but can contain potential pathogens.
      • Lager S.
      • de Goffau M.C.
      • Sovio U.
      • et al.
      Detecting eukaryotic microbiota with single-cell sensitivity in human tissue.
      • Lauder A.P.
      • Roche A.M.
      • Sherrill-Mix S.
      • et al.
      Comparison of placenta samples with contamination controls does not provide evidence for a distinct placenta microbiota.
      • Leiby J.S.
      • McCormick K.
      • Sherrill-Mix S.
      • et al.
      Lack of detection of a human placenta microbiome in samples from preterm and term deliveries.
      • Leon L.J.
      • Doyle R.
      • Diez-Benavente E.
      • et al.
      Enrichment of clinically relevant organisms in spontaneous preterm-delivered placentas and reagent contamination across all clinical groups in a large pregnancy cohort in the United Kingdom.
      • Theis K.R.
      • Romero R.
      • Winters A.D.
      • et al.
      Does the human placenta delivered at term have a microbiota? Results of cultivation, quantitative real-time PCR, 16S rRNA gene sequencing, and metagenomics.
      our data indicate that healthy placental tissue facilitates the inhibition of bacterial growth. However, because bacteria are present everywhere in the environment, it is very difficult to prove the absolute absence of bacteria in the samples.

      Strengths and limitations

      The most important strength of the study is the comprehensive and controlled sample collection from women not in labor undergoing cesarean delivery at term and from women delivering vaginally, in which case contaminations from the passage through the birth channel could be expected. Sampling was performed exclusively by the 2 clinicians in the group and included placental tissue from 3 parts of the placenta divided into 3 layers and AF and infant vernix caseosa. Samples from the oral cavity, gut, and vagina were collected to allow for the comparison with potential findings from the placental tissue. Furthermore, detailed clinical data were obtained from each study subject and reported to allow for a careful evaluation of each case. Multiple methods for the analysis of bacterial DNA and bacterial culture and growth inhibition were combined in this study to investigate whether the amniotic cavity contains bacteria in normal pregnancies at term.
      Limitations of our study include the fact that some of the mothers had been treated with antibiotics during pregnancy or at delivery, which may have influenced the results. However, only 18% of the women who underwent cesarean deliveries and 11% of the women who delivered vaginally received antibiotics, and no differences in copy number or bacterial growth pattern could be seen for those cases. Another limitation of the study is that the AF was collected in very different ways from women undergoing cesarean or vaginal deliveries. Therefore, the bacterial signals from these 2 sample types could not be effectively compared. Furthermore, in the culture experiments of placental tissue, freezing of biologic samples before culture may have influenced the recovery of some bacterial clones. In addition, for the culture experiments, only 1 growth medium and anaerobic culture conditions were used. Thus, this may not be the optimal culture conditions for all bacterial species. However, our study used multiple methods, and results are therefore not only dependent on 1 experimental condition.
      The technical limitations of our study are related to the difficulty of accurately characterizing microbial communities in low microbial biomass niches, especially because there is no clear understanding of how many bacteria constitute a tissue microbiome. The inclusion of controls with placental tissue samples that have been spiked with known numbers of bacteria would help to define a limit of DNA detection in our study. Moreover, controls of human tissue with very low confirmed biomass, such as muscle, could have been used as controls but were not available to us.

      Conclusions

      Our data, together with recent reports and current immunologic understanding, do not support the existence of a placental microbiome. We conclude that the healthy fetus develops in utero in an environment enclosed by the amniotic membrane that is free of bacterial colonization.

      Acknowledgments

      We are grateful to the participants who donated the biologic samples. We also want to acknowledge the laboratory manager at the Centre for Translational Microbiome Research, Marica Hamsten, and Kristin Wannerberger from Ferring Pharmaceuticals (Saint-Prex, Switzerland) for planning part of the data collection.

      Appendix

      Figure thumbnail fx1
      Supplemental Figure 1V3-V4 universal bacterial primers also amplify human DNA
      Amplification of mostly bacterial vs pure human DNA by universal 16S primers V3-V4 or V6-V8. Purple lines depict ZymoBIOMICS Microbial Community DNA Standard (artificial microbial consortium). Blue lines depict pure human DNA. Green lines depict negative template control.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Figure thumbnail fx2
      Supplemental Figure 2Abundant taxa detected in placenta and negative controls
      The maternal side of the placentas from women who had not taken antibiotics during pregnancy or delivery (n=45 for cesarean delivery; n=19 for vaginal delivery) was profiled by sequencing of the V6-V8 region of the 16S rRNA gene. “P” stands for participant and “t” for twin. The Neg.Ext labels are negative controls from each of the extraction plates, whereas the Neg.Amp label is a negative amplification (PCR) control, as described in the methods.
      PCR, polymerase chain reaction; rRNA, ribosomal ribonucleic acid.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Figure thumbnail fx3
      Supplemental Figure 3Bacterial profile of vaginal swabs, saliva, and rectal swabs at the time of delivery
      Bar plots depict the relative abundance of each taxon in each sample, taken at the time of delivery from each participant and analyzed by shallow shotgun sequencing, as described in the methods. The taxa that appear in more than 1 dataset have consistent colors. Each participant’s samples are stacked in each dataset. “P” stands for participant. Bacterial DNA from vaginal and rectal samples and saliva were characterized by metagenomic sequencing.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Figure thumbnail fx4
      Supplemental Figure 4Bacteria from cesarean-delivered placentas are similar to skin, whereas vaginal origin is common for normal deliveries
      Richness (number of identified amplified sequence variants) and taxonomic composition of bacterial isolates grown from placental specimens, based on sequencing of the V6-V8 region of the 16S rRNA gene. For each set of panels, richness is represented at the top and composition at the bottom. Study participants are aligned in all plots, and their study number is depicted under each bar plot. Where a column is left blank, no bacterial colonies were obtained. The colored legend is common to all 3 panels. Top indicates the fetal side of the placenta, middle indicates the middle part, and the bottom indicates the maternal side.
      rRNA, ribosomal ribonucleic acid.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Supplemental Table 1Probe-based qPCR primers and probes used as standard
      CladeForwardReverseMelting temperatureProbeTmFluorophore
      CutibacteriumGGAAGTGTAATCTTGGGGCTTACTGGTGTTCCTCCTGATATCTG59CCACCGCTCCACCAGGAATT65FAM (498-580)
      EnterococcaceaeCGGTTTCTTAAGTCTGATGTGATCCTCCATATATCTACGCATTTC59ATTCCACTCTCCTCTTCTGC65Red_610 (533-610)
      GardnerellaGGTTGTAAACCGCTTTTGATTGCAAGCTCTTTACGCCCAATAATTC59AGTGTACCTTTCGAATAAGC65Cy5/Cy5-5 (618-660)
      BacteroidesATGGGGATGCGTTCCATGCACGGAGTTAGCCGAT58CTGAACCAGCCAAGTAG63FAM (498-580)
      Escherichia coliGAGGAAGGGAGTAAAGTTAATACGGGATTTCACATCTGACTTAAC58CTCATTGACGTTACCCGCAG63Red_610 (533-610)
      UreaplasmaGAACGATGAAGGTCTTATAGATTGACGCTTGCATCCTATGTATTAC58CATAGTTAGCCGATACTTATTCAA63Cy5/Cy5-5 (618-660)
      StreptococcusTAACGCGTAGGTAACCTTACTGCTGCCTCCCGTAGGA57GGACCTGCGTTGTATTA62Cy5/Cy5-5 (618-660)
      AnaerococcusAACGCGTGAGTAACCTGCCTTCACTGCTGCCTCCCGTAGGAGT57GTGTACGGCCACATTGGG62Red_610 (533-610)
      PrevotellaceaeGGCGGGGTAACGGCCCACGGAATTAGCCGGTCCTT57ACCAGCCAAGTAGCGTG62FAM (498-580)
      StaphylococcusGGTACCTAATCAGAAAGCCACGGCGCGCTTTACGCCCAATAATTC57CGGATAACGCTTGCCACCTAC63FAM (498-580)
      FAM, fluorescein amidite; qPCR, quantitative polymerase chain reaction.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Supplemental Table 2Bacterial taxa-specific gBlock sequences used as standard
      CladegBlock sequence
      CutibacteriumCGCATTTATTGGGCGTAAAGGGCTCGTAGGTGGTTGATCGCGTCGGAAGTCTAATCTTGGGGCTTAACCCTGAGCGTGCTTTTGATACGGGTTGACTTGAGGAAGGTAGGGGAGAATGGAATTCCTGGTGGAGCGGTGGAATGCGCAGATATCAGGAGGAACACCGGTGGCGAAGGCGGTTCTTCGGGCCTTTCCTGACGC
      EnterococcaceaeCGRCAGTTACAGACCAGAGAGTCGCCTTCGCCACTGGTGTTCCTCCATATATCTACGCATTTCACCGCTACACATGGAATTCCACTCTCCTCTTCTGCACTCAAGTTTCTCAGTTTCCAATGACCCTCCCCGGTTAAGCCGGGGGCTTTCACATCAGACTTAAGAAACCGCCTGCGCTCGCTTTACGCCCAATAAATCCGGACAACGCTTGCCACCTACGT
      GardnerellaTTCACACCAGACGCGACGAACCGCCTACAAGCTCTTTACGCCCAATAATTCCGGATAACGCTTGCGCCCTACGTATTACCGCGGCTGCTGGCACGTAGTTAGCCGGCGCTTATTCGAAAGGTACACTCACCCGAAAGCTTGCTCCCAATCAAAAGCGGTTTACAACCCGAAGGCCTTCATCCCGCACGCGGCGTCGCTGCGTCAGGGTTTCCCCCATTGCGCAATATTCCCCACTGCTGCCTCC
      BacteroidesCATCTTGAGAAAGTTAAAGATTTATTGGTTATGGATGGGGATGCGTTCCATTAGATAGTTGGTGAGGTAACGGCTCACCAAGTCTTCGATGGATAGGGGTTCTGAGAGGAAGGTCCCCCACATTGGTACTGAGACACGGACCAAACTCCTACGGGAGGCAGCAGTGAGGAATATTGGTCAATGGGCGAGAGCCTGAACCAGCCAAGTAGCGTGAAGGATGAAGGTCCTATGGATTGTAAACTTCTTTTATAGTAGAATAAAGTGACCCACGTGTGGGTTTTTGTATGTATACTATGAATAAGGATCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGATCCGAGC
      Escherichia coliTGCCGCGTGTATGAAGAAGGCCTTCGGGTTGTAAAGTACTTTCAGCGGGGAGGAAGGGAGTAAAGTTAATACCTTTGCTCATTGACGTTACCCGCAGAAGAAGCACCGGCTAACTCCGTGCCAGCAGCCGCGGTAATACGGAGGGTGCAAGCGTTAATCGGAATTACTGGGCGTAAAGCGCACGCAGGCGGTTTGTTAAGTCAGATGTGAAATCCCCGGGCTCAACCTGGGAACTGCATCT
      UreaplasmaGGGAATTTTTCACAATGGGCGAAAGCCTTATGAAGCAATGCCGCGTGAACGATGAAGGTCTATAAGATTGTAAAGTTCTTTTATTTGGGAAGAACCACTAAAATAGGAAATGATTTTAGTTTGACTGTACCATTTGAATAAGTATCGGCTAACTATGTGCCAGCAGCCGCGGTAATACATAGGATGCAAGCGTTATCCGGATTTACTGGGCGTAAAACGAG
      StreptococcusCACTCACGCGGCGTTGCTCGGTCAGGGTTCCCCCCATTGCCGAAGATTCCCTACTGCTGCCTCCCGTAGGAGTCTGGGCCGTGTCTCAGTCCCAGTGTGGCCGATCACCCTCTCAGGTCGGCTATGTATCGAAGCCTTGGTAAGCCGTTACCTTACCAACTAGCTAATACAACGCAGGTCCATCTTTAAGTGGTGCACTTGCACCTTTTAAGTAGCTGACATGTGTCCGCCACTATTATGCGGTATTAGCTATCGTTTCCA
      AnaerococcusCAGGGTTTCCCCCATTGTGCAAAATTCCCCACTGCTGCCTCCCGTAGGAGTCTGGGCCGTGTCTCAGTCCCAATGTGGCCGTACACTCTCTCAAGCCGGCTACTGATCGTTGCCTTGGTGAGCTGTTATCTCACCAACTAGCTAATCAGACGCAAGTCCATCTTAGAGCGATAAATCTTTGACCAGCACTTCATGCGAGGTGTTGGTTTCATAGGGTATTATTCTTCGTTTCCAAAGGCTATCCCCTTCTCTAAGGCAGGTTACTCACGCGTTACTCACCCGTCCGCCACTAATCCATCTAATTTCACTCCGAAGAGATCAATTAGGTTTCATCGTTCGACTTGCATGTGTTATGCACGCCGCCAGCGT
      PrevotellaceaeATCCGATTTGGACCAAAGGCTTAGCGGTAAAGGATGGGGATGCGTCCGATTAGCTTGACGGCGGGGTAACGGCCCACCGTGGCAACGATCGGTAGGGGTTCTGAGAGGAAGGTCCCCCACACTGGAACTGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTGAGGAATATTGGTCAATGGGCGTAAGCCTGAACCAGCCAAGTAGCGTGCAGGATGACGGCCCTATGGGTTGTAAACTGCTTTTATGCGGGGATAAAAGAGCCCACGTGTGGGTTTTTGCAGGTACCGCATGAATAAGGACCGGCTAATTCCGTGCCAGCAGCCGCGG
      StaphylococcusTGAGACACGGTCCAGACTCCTACGGGAGGCAGCAGTAGGGAATCTTCCGCAATGGGCGAAAGCCTGACGGAGCAACGCCGCGTGAGTGATGAAGGTCTTCGGATCGTAAAACTCTGTTATTAGGGAAGAACAAATGTGTAAGTAACTGTGCACGTCTTGACGGTACCTAATCAGAAAGCCACGGCTAACTACGTGCCAGCAGCCGCGGTAATACGTAGGTGGCAAGCGTTATCCGGAATTATTGGGCGTAAAGCGCGCGTAGGCGGTTTCTTAAGTCTGATGTGAAAGCCC
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Supplemental Table 3A comparison of observed bacterial clades with literature data
      CladeObserved in our study (16S gene sequencing)Observed in our study (culture)Previously observed in other studiesComment
      CutibacteriumYesYesAagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Formerly known as Propionibacterium
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Collado et al
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      EnterococcaceaeYesNoBassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Genus Enterococcus was the intended target, but no genus-level probe with good characteristics could be designed
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      GardnerellaYesYesDiGiulio et al
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Doyle et al
      • Ganal-Vonarburg S.C.
      • Hornef M.W.
      • Macpherson A.J.
      Microbial-host molecular exchange and its functional consequences in early mammalian life.
      BacteroidesYesYesAagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Escherichia coliYesYesAagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Genera Escherichia and Shigella were the desired target, but a species-level assay was chosen to avoid unspecific binding
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Collado et al
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      UreaplasmaNoNoDiGiulio et al
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      StreptococcusYesYesDiGiulio et al
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Collado et al
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      Gomez-Arango et al
      • Burton G.J.
      • Jauniaux E.
      What is the placenta?.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      AnaerococcusYesNo
      PrevotellaceaeYesYesDiGiulio et al
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      Genus Prevotella was the intended target, but no genus-level probe with good characteristics could be designed
      Aagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Gomez-Arango et al
      • Burton G.J.
      • Jauniaux E.
      What is the placenta?.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      StaphylococcusYesYesDiGiulio et al
      • DiGiulio D.B.
      • Romero R.
      • Amogan H.P.
      • et al.
      Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
      Aagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Collado et al
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      GemellaYesNoBassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Excluded because of very high background signal
      LactobacillusYesYesAagaard et al
      • Murtha A.P.
      • Edwards J.M.
      The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
      Excluded because of unspecific binding to human DNA
      Bassols et al
      • Fox C.
      • Eichelberger K.
      Maternal microbiome and pregnancy outcomes.
      Collado et al
      • Dominguez-Bello M.G.
      • Godoy-Vitorino F.
      • Knight R.
      • Blaser M.J.
      Role of the microbiome in human development.
      Zheng et al
      • Collado M.C.
      • Rautava S.
      • Aakko J.
      • Isolauri E.
      • Salminen S.
      Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.
      Supplemental Table 4Number of specimens from cesarean deliveries with bacteria detection by different methods and in different body sites
      GenusPositive culture samples (n=51)Postive sequencing samples (n=48)Positive qPCR samples (n=48)Overlap culture vs sequencingOverlap culture vs qPCRPositive vaginal swabs (n=50)Positive saliva samples (n=50)Positive rectal swabs (n=50)
      Acinetobacter59NA2NA000
      Aerococcus22NA0NA494748
      Anaerococcus4070021047
      Bacteroides30700494150
      Blastococcus14NA0NA000
      Bradyrhizobium23NA0NA000
      Burkholderia or Paraburkholderia15NA0NA494742
      Cupriavidus435NA3NA000
      Cutibacterium230702003
      Dialister12NA0NA274646
      Escherichia or Shigella134531202031
      Gardnerella6370128025
      Gemella2030074711
      Gemmatirosa210NA1NA000
      Lactobacillus75NA1NA494750
      Lawsonella12NA0NA9042
      Leifsonia542NA5NA002
      Massilia1143NA10NA20125
      Megasphaera12NA0NA114624
      Mesorhizobium116NA0NA000
      Methylobacterium12NA0NA000
      Microbacterium32NA0NA105
      Mycobacterium14NA0NA484640
      Pelomonas843NA7NA000
      Prevotella40601304750
      Pseudomonas1513NA7NA493138
      Staphylococcus15562113123
      Streptococcus163601494750
      Veillonella41NA0NA244738
      Only bacteria detected in a placental biopsy by at least 2 methods are included.
      NA, not applicable.
      Sterpu et al. No evidence for a placental microbiome in full-term pregnancies. Am J Obstet Gynecol 2021.

      References

        • Perez-Muñoz M.E.
        • Arrieta M.C.
        • Ramer-Tait A.E.
        • Walter J.
        A critical assessment of the “sterile womb” and “in utero colonization” hypotheses: implications for research on the pioneer infant microbiome.
        Microbiome. 2017; 5: 48
        • Ravel J.
        • Gajer P.
        • Abdo Z.
        • et al.
        Vaginal microbiome of reproductive-age women.
        Proc Natl Acad Sci U S A. 2011; 108: 4680-4687
        • Mitchell C.M.
        • Haick A.
        • Nkwopara E.
        • et al.
        Colonization of the upper genital tract by vaginal bacterial species in nonpregnant women.
        Am J Obstet Gynecol. 2015; 212: 611.e1-611.e9
        • Moreno I.
        • Garcia-Grau I.
        • Bau D.
        • et al.
        The first glimpse of the endometrial microbiota in early pregnancy.
        Am J Obstet Gynecol. 2020; 222: 296-305
        • Chen C.
        • Song X.
        • Wei W.
        • et al.
        The microbiota continuum along the female reproductive tract and its relation to uterine-related diseases.
        Nat Commun. 2017; 8: 875
        • Winters A.D.
        • Romero R.
        • Gervasi M.T.
        • et al.
        Does the endometrial cavity have a molecular microbial signature?.
        Sci Rep. 2019; 9: 9905
        • DiGiulio D.B.
        • Romero R.
        • Amogan H.P.
        • et al.
        Microbial prevalence, diversity and abundance in amniotic fluid during preterm labor: a molecular and culture-based investigation.
        PLoS One. 2008; 3e3056
        • Hitti J.
        • Riley D.E.
        • Krohn M.A.
        • et al.
        Broad-spectrum bacterial rDNA polymerase chain reaction assay for detecting amniotic fluid infection among women in premature labor.
        Clin Infect Dis. 1997; 24: 1228-1232
        • DiGiulio D.B.
        • Romero R.
        • Kusanovic J.P.
        • et al.
        Prevalence and diversity of microbes in the amniotic fluid, the fetal inflammatory response, and pregnancy outcome in women with preterm pre-labor rupture of membranes.
        Am J Reprod Immunol. 2010; 64: 38-57
        • Kim C.J.
        • Romero R.
        • Chaemsaithong P.
        • Chaiyasit N.
        • Yoon B.H.
        • Kim Y.M.
        Acute chorioamnionitis and funisitis: definition, pathologic features, and clinical significance.
        Am J Obstet Gynecol. 2015; 213: S29-S52
        • Romero R.
        • Miranda J.
        • Kusanovic J.P.
        • et al.
        Clinical chorioamnionitis at term I: microbiology of the amniotic cavity using cultivation and molecular techniques.
        J Perinat Med. 2015; 43: 19-36
        • Murtha A.P.
        • Edwards J.M.
        The role of mycoplasma and ureaplasma in adverse pregnancy outcomes.
        Obstet Gynecol Clin North Am. 2014; 41: 615-627
        • Fox C.
        • Eichelberger K.
        Maternal microbiome and pregnancy outcomes.
        Fertil Steril. 2015; 104: 1358-1363
        • Dominguez-Bello M.G.
        • Godoy-Vitorino F.
        • Knight R.
        • Blaser M.J.
        Role of the microbiome in human development.
        Gut. 2019; 68: 1108-1114
        • Gensollen T.
        • Iyer S.S.
        • Kasper D.L.
        • Blumberg R.S.
        How colonization by microbiota in early life shapes the immune system.
        Science. 2016; 352: 539-544
        • Ganal-Vonarburg S.C.
        • Hornef M.W.
        • Macpherson A.J.
        Microbial-host molecular exchange and its functional consequences in early mammalian life.
        Science. 2020; 368: 604-607
        • Burton G.J.
        • Jauniaux E.
        What is the placenta?.
        Am J Obstet Gynecol. 2015; 213: S6.e1-S6.e8
        • Fisher S.J.
        Why is placentation abnormal in preeclampsia?.
        Am J Obstet Gynecol. 2015; 213: S115-S122
        • Ptacek I.
        • Sebire N.J.
        • Man J.A.
        • Brownbill P.
        • Heazell A.E.P.
        Systematic review of placental pathology reported in association with stillbirth.
        Placenta. 2014; 35: 552-562
        • Zhang S.
        • Regnault T.R.H.
        • Barker P.L.
        • et al.
        Placental adaptations in growth restriction.
        Nutrients. 2015; 7: 360-389
        • Aagaard K.
        • Ma J.
        • Antony K.M.
        • Ganu R.
        • Petrosino J.
        • Versalovic J.
        The placenta harbors a unique microbiome.
        Sci Transl Med. 2014; 6: 237ra65
        • Bassols J.
        • Serino M.
        • Carreras-Badosa G.
        • et al.
        Gestational diabetes is associated with changes in placental microbiota and microbiome.
        Pediatr Res. 2016; 80: 777-784
        • Collado M.C.
        • Rautava S.
        • Aakko J.
        • Isolauri E.
        • Salminen S.
        Human gut colonisation may be initiated in utero by distinct microbial communities in the placenta and amniotic fluid.
        Sci Rep. 2016; 6: 23129
        • Doyle R.M.
        • Alber D.G.
        • Jones H.E.
        • et al.
        Term and preterm labour are associated with distinct microbial community structures in placental membranes which are independent of mode of delivery.
        Placenta. 2014; 35: 1099-1101
        • Doyle R.M.
        • Harris K.
        • Kamiza S.
        • et al.
        Bacterial communities found in placental tissues are associated with severe chorioamnionitis and adverse birth outcomes.
        PLoS One. 2017; 12e0180167
        • Gomez-Arango L.F.
        • Barrett H.L.
        • McIntyre H.D.
        • Callaway L.K.
        • Morrison M.
        • Nitert M.D.
        Contributions of the maternal oral and gut microbiome to placental microbial colonization in overweight and obese pregnant women.
        Sci Rep. 2017; 7: 2860
        • Parnell L.A.
        • Briggs C.M.
        • Cao B.
        • Delannoy-Bruno O.
        • Schrieffer A.E.
        • Mysorekar I.U.
        Microbial communities in placentas from term normal pregnancy exhibit spatially variable profiles.
        Sci Rep. 2017; 7: 11200
        • Prince A.L.
        • Ma J.
        • Kannan P.S.
        • et al.
        The placental membrane microbiome is altered among subjects with spontaneous preterm birth with and without chorioamnionitis.
        Am J Obstet Gynecol. 2016; 214: 627.e1-627.e16
        • Satokari R.
        • Grönroos T.
        • Laitinen K.
        • Salminen S.
        • Isolauri E.
        Bifidobacterium and lactobacillus DNA in the human placenta.
        Lett Appl Microbiol. 2009; 48: 8-12
        • Seferovic M.D.
        • Pace R.M.
        • Caroll M.
        • et al.
        Visualization of microbes by 16S in situ hybridization in term and preterm placentas without intraamniotic infection.
        Am J Obstet Gynecol. 2019; 221: 146.e1-146.e23
        • Tuominen H.
        • Rautava S.
        • Syrjänen S.
        • Collado M.C.
        • Rautava J.
        HPV infection and bacterial microbiota in the placenta, uterine cervix and oral mucosa.
        Sci Rep. 2018; 8: 9787
        • Zheng J.
        • Xiao X.
        • Zhang Q.
        • Mao L.
        • Yu M.
        • Xu J.
        The placental microbiome varies in association with low birth weight in full-term neonates.
        Nutrients. 2015; 7: 6924-6937
        • Kuperman A.A.
        • Zimmerman A.
        • Hamadia S.
        • et al.
        Deep microbial analysis of multiple placentas shows no evidence for a placental microbiome.
        BJOG. 2020; 127: 159-169
        • de Goffau M.C.
        • Lager S.
        • Sovio U.
        • et al.
        Human placenta has no microbiome but can contain potential pathogens.
        Nature. 2019; 572: 329-334
        • Lager S.
        • de Goffau M.C.
        • Sovio U.
        • et al.
        Detecting eukaryotic microbiota with single-cell sensitivity in human tissue.
        Microbiome. 2018; 6: 151
        • Lauder A.P.
        • Roche A.M.
        • Sherrill-Mix S.
        • et al.
        Comparison of placenta samples with contamination controls does not provide evidence for a distinct placenta microbiota.
        Microbiome. 2016; 4: 29
        • Leiby J.S.
        • McCormick K.
        • Sherrill-Mix S.
        • et al.
        Lack of detection of a human placenta microbiome in samples from preterm and term deliveries.
        Microbiome. 2018; 6: 196
        • Leon L.J.
        • Doyle R.
        • Diez-Benavente E.
        • et al.
        Enrichment of clinically relevant organisms in spontaneous preterm-delivered placentas and reagent contamination across all clinical groups in a large pregnancy cohort in the United Kingdom.
        Appl Environ Microbiol. 2018; 84 (e00483-18)
        • Theis K.R.
        • Romero R.
        • Winters A.D.
        • et al.
        Does the human placenta delivered at term have a microbiota? Results of cultivation, quantitative real-time PCR, 16S rRNA gene sequencing, and metagenomics.
        Am J Obstet Gynecol. 2019; 220: 267.e1-267.e39
        • Salter S.J.
        • Cox M.J.
        • Turek E.M.
        • et al.
        Reagent and laboratory contamination can critically impact sequence-based microbiome analyses.
        BMC Biol. 2014; 12: 87
        • Prevedourakis C.N.
        • Strigou-Charalabis E.
        • Kaskarelis D.B.
        Bacterial invasion of amniotic cavity during pregnancy and labor.
        Obstet Gynecol. 1971; 37: 459-461
        • Lim E.S.
        • Rodriguez C.
        • Holtz L.R.
        Amniotic fluid from healthy term pregnancies does not harbor a detectable microbial community.
        Microbiome. 2018; 6: 87
        • Yoshio H.
        • Tollin M.
        • Gudmundsson G.H.
        • et al.
        Antimicrobial polypeptides of human vernix caseosa and amniotic fluid: implications for newborn innate defense.
        Pediatr Res. 2003; 53: 211-216
        • Burton G.J.
        • Sebire N.J.
        • Myatt L.
        • et al.
        Optimising sample collection for placental research.
        Placenta. 2014; 35: 9-22
        • Huys G.
        • Vanhoutte T.
        • Joossens M.
        • et al.
        Coamplification of eukaryotic DNA with 16S rRNA gene-based PCR primers: possible consequences for population fingerprinting of complex microbial communities.
        Curr Microbiol. 2008; 56: 553-557
        • Martin M.
        Cutadapt removes adapter sequences from high-throughput sequencing reads.
        EMBnet J. 2011; 17: 10-12
        • Callahan B.J.
        • McMurdie P.J.
        • Rosen M.J.
        • Han A.W.
        • Johnson A.J.A.
        • Holmes S.P.
        DADA2: high-resolution sample inference from Illumina amplicon data.
        Nat Methods. 2016; 13: 581-583
        • Davis N.M.
        • Proctor D.M.
        • Holmes S.P.
        • Relman D.A.
        • Callahan B.J.
        Simple statistical identification and removal of contaminant sequences in marker-gene and metagenomics data.
        Microbiome. 2018; 6: 226
        • Hu Y.O.O.
        • Hugerth L.W.
        • Bengtsson C.
        • et al.
        Bacteriophages synergize with the gut microbial community to combat Salmonella.
        mSystems. 2018; 3 (e00119-18)
        • Stinson L.F.
        • Keelan J.A.
        • Payne M.S.
        Identification and removal of contaminating microbial DNA from PCR reagents: impact on low-biomass microbiome analyses.
        Lett Appl Microbiol. 2019; 68: 2-8
        • Yung H.W.
        • Colleoni F.
        • Atkinson D.
        • et al.
        Influence of speed of sample processing on placental energetics and signalling pathways: implications for tissue collection.
        Placenta. 2014; 35: 103-108
        • Wolfe L.M.
        • Thiagarajan R.D.
        • Boscolo F.
        • et al.
        Banking placental tissue: an optimized collection procedure for genome-wide analysis of nucleic acids.
        Placenta. 2014; 35: 645-654
        • Byrd A.L.
        • Belkaid Y.
        • Segre J.A.
        The human skin microbiome.
        Nat Rev Microbiol. 2018; 16: 143-155
        • Mei C.
        • Yang W.
        • Wei X.
        • Wu K.
        • Huang D.
        The unique microbiome and innate immunity during pregnancy.
        Front Immunol. 2019; 10: 2886
        • Goswami S.
        • Sarkar R.
        • Saha P.
        • et al.
        Effect of human placental extract in the management of biofilm mediated drug resistance - a focus on wound management.
        Microb Pathog. 2017; 111: 307-315
        • Sharma K.
        • Mukherjee C.
        • Roy S.
        • De D.
        • Bhattacharyya D.
        Human placental extract mediated inhibition of proteinase K: implications of heparin and glycoproteins in wound physiology.
        J Cell Physiol. 2014; 229: 1212-1223
        • Para R.
        • Romero R.
        • Miller D.
        • et al.
        Human β-defensin-3 participates in intra-amniotic host defense in women with labor at term, spontaneous preterm labor and intact membranes, and preterm prelabor rupture of membranes.
        J Matern Fetal Neonatal Med. 2019; ([Epub ahead of print])
        • Soto E.
        • Espinoza J.
        • Nien J.K.
        • et al.
        Human β-defensin-2: a natural antimicrobial peptide present in amniotic fluid participates in the host response to microbial invasion of the amniotic cavity.
        J Matern Fetal Neonatal Med. 2007; 20: 15-22
        • Varrey A.
        • Romero R.
        • Panaitescu B.
        • et al.
        Human β-defensin-1: a natural antimicrobial peptide present in amniotic fluid that is increased in spontaneous preterm labor with intra-amniotic infection.
        Am J Reprod Immunol. 2018; 80e13031
        • Zare-Bidaki M.
        • Sadrinia S.
        • Erfani S.
        • Afkar E.
        • Ghanbarzade N.
        Antimicrobial properties of amniotic and chorionic membranes: a comparative study of two human fetal sacs.
        J Reprod Infertil. 2017; 18: 218-224
        • Mao Y.
        • Hoffman T.
        • Singh-Varma A.
        • et al.
        Antimicrobial peptides secreted from human cryopreserved viable amniotic membrane contribute to its antibacterial activity.
        Sci Rep. 2017; 7: 13722
        • Boldenow E.
        • Jones S.
        • Lieberman R.W.
        • et al.
        Antimicrobial peptide response to group B Streptococcus in human extraplacental membranes in culture.
        Placenta. 2013; 34: 480-485
        • Marchini G.
        • Lindow S.
        • Brismar H.
        • et al.
        The newborn infant is protected by an innate antimicrobial barrier: peptide antibiotics are present in the skin and vernix caseosa.
        Br J Dermatol. 2002; 147: 1127-1134
        • Kämpfer P.
        • Lodders N.
        • Martin K.
        • Falsen E.
        Massilia oculi sp. nov., isolated from a human clinical specimen.
        Int J Syst Evol Microbiol. 2012; 62: 364-369
        • Mor G.
        • Kwon J.Y.
        Trophoblast-microbiome interaction: a new paradigm on immune regulation.
        Am J Obstet Gynecol. 2015; 213: S131-S137
        • Perry I.D.
        • Nguyen T.
        • Sherina V.
        • et al.
        Analysis of the capacity of Salmonella enterica Typhimurium to infect the human placenta.
        Placenta. 2019; 83: 43-52
        • Gotsch F.
        • Romero R.
        • Kusanovic J.P.
        • et al.
        The fetal inflammatory response syndrome.
        Clin Obstet Gynaecol. 2007; 50: 652-683
        • Rackaityte E.
        • Halkias J.
        • Fukui E.M.
        • et al.
        Viable bacterial colonization is highly limited in the human intestine in utero.
        Nat Med. 2020; 26: 599-607
        • Dominguez-Bello M.G.
        • Costello E.K.
        • Contreras M.
        • et al.
        Delivery mode shapes the acquisition and structure of the initial microbiota across multiple body habitats in newborns.
        Proc Natl Acad Sci U S A. 2010; 107: 11971-11975
        • Kirk R.G.
        “Life in a germ-free world”: isolating life from the laboratory animal to the bubble boy.
        Bull Hist Med. 2012; 86: 237-275
        • Barnes R.D.
        • Bentovim A.
        • Hensman S.
        • Piesowicz A.T.
        Care and observation of a germ-free neonate.
        Arch Dis Child. 1969; 44: 211-217