s / Placenta 36 (2015) A1eA60 A23 Results: The HTR-8/SVneo cells expressed VEGFA, PGF, KDR and FLT1. VEGFA and FLT1 expression were upregulated in 1% O2 (p1⁄40.001) and 5% O2 (p1⁄40.01) compared to 20% O2. Trophoblast invasionwas increased in 1% O2 (p<0.0001) and 5% O2 (p1⁄40.003) compared to 20% O2. The addition of VEGFA (p<0.05) and PlGF (p<0.05) resulted in a decrease in trophoblast invasion under all three O2 conditions. The addition of VEGF and PlGF did not have any effect on trophoblast proliferation. Conclusion: The biological actions of VEGF and PlGF on first trimester trophoblast cells may differ from their effects on vascular endothelial cells. One possible explanation is that VEGF and PlGF instead promote differentiation of HTR8/SVneo, as VEGF treatment has been shown to promote the formation of endothelial-like tubes on, rather than invasion through, Matrigel. P1.34. THE PLACENTAL MICROBIOME; MICROBIAL PRESENCE IN NATURE’S TRANSPLANT Elise Pelzer , Flavia Hugens , Philip Hugenholtz , Rohan Lourie , Kenneth Beagley , Ross Turner . Queensland University of Technology, Brisbane, Australia; Australian Centre for Ecogenomics, Brisbane, Australia; Mater Pathology, Brisbane, Australia; 4 The Wesley Hospital, Brisbane, Australia Bacterial invasion of the amniotic cavity is a known cause of adverse pregnancy outcomes, and 12% of all pregnancies end preterm, causing significant morbidity for mothers and babies. Bacterial culture, the gold standard for assessing clinical samples for the presence of infectious agents is slow, resulting in a biased over-representation of the most abundant cultivable microbial species, not necessarily those that cause pathology. In this study, placentae were collected for gestations reaching at least 37 weeks gestation from women delivering via: (1) non-labouring elective Caesarean section, (2) emergency Caesarean section (3) spontaneous vaginal delivery, and (4) induced and/or assisted vaginal delivery. Placentae were sampled across maternal and fetal sites. DNA extraction was performed using previously published methods and 16S rRNA screening was performed using the 454 pyrosequencing platform. Preliminary data indicate that the placentae collected from vaginal deliveries exhibit increased community richness when compared to those collected from both elective and urgent Caesarean sections. Lactobacilli were low abundance community members in placentae collected from Caesarean sections, with anaerobic genera including Lachnospiraceae, Propionibacterium and Veillonellaceae dominating these microbial communities. In contrast, the vaginally delivered placentae elaborated microbial communities dominated by lactobacilli, Bacteroidales family members and Gardnerella species. Staphylococcaceae, Streptococcaceae and Corynebacterium species were well represented amongst the most abundant community members in all delivery groups. The placenta in term deliveries in normal healthy pregnancies is not sterile, irrespective of the mode of delivery and sampling site. Further investigations are required to determine the potential significance of individual microbial community variation in placental tissues, where the fetus encounters long-term exposure to a diverse microbial population long before delivery. Results from this study have the potential to improve pregnancy outcomes by improving our understanding of how the healthy term fetus is initially colonized by microorganisms, a defining factor in long-term health. P1.35. IS DIVERSIFICATION OF THE ENDOMETRIAL MICROBIOME SIGNIFICANT FOR REPRODUCTIVE SUCCESS? Elise Pelzer , Dana Willner , Flavia Huygens , Melissa Buttini . Queensland University of Technology, Brisbane, Australia; Australian Centre for Ecogenomics, Brisbane, Australia; 3 The Wesley Hospital, Brisbane, Australia Historically, studies investigating the relationship between infectionrelated pregnancy complications and maternal microbial populations have focused on the lower genital tract vaginal microbiota, due to the accepted belief that in the absence of overt infection, the endometrial cavity is sterile. Technical advancements in molecular microbiology have resulted in significant challenge to the central dogma of a sterile female upper genital tract. This pilot study aimed to investigate the endometrial microbiota of nulliparous women who were also virgo intacta to determine whether the endometrial cavity was sterile, and whether the genital tract microbial community profiles of these women varied compared to sexually active women. We collected paired endocervical and endometrial biopsy samples at the time of operative hysteroscopy and/or laparoscopy. Samples were interrogated for the presence of microbial DNA using a two-step next generation sequencing technology approach to exploit the V8 and V9 regions of the 16S rRNA gene. Pyrosequencing revealed that the endocervix and endometrium share a minor microbial community, but that each site harbours a separate and distinct microbial population (p 1⁄4 0.024). The endometrial microbial community in nulliparous women who were also virgo intacta was not dominated by lactobacilli. Members of the genera Prevotella, Fusobacterium and Jonquetella were the most abundant taxa identified in the endometrium of this group. Conclusion: The female upper genital tract is not sterile, even when the hymen remains intact. The distinct site-specific microbial community profiles between the endometrium and endocervix in these women may offer insight into the significance of the endometrial microbiota in reproductive health. P1.36. TNF-a UPREGULATES SECRETION OF GM-CSF, CCL5 AND IL-10 BY HUMAN FIRST TRIMESTER PLACENTA Monika Siwetz , Amin El-Heliebi , Astrid Blaschitz , Ursula Hiden , Gernot Desoye , Berthold Huppertz , Martin Gauster . Medical University Graz, Institute of Cell Biology, Histology and Embryology, Graz, Austria; Medical University Graz, Department of Obstetrics and Gynaecology, Graz, Austria Objectives: Infectious complications, dysregulated immune responses and placental oxidative stress have been associated with increased TNF-a and pregnancy pathology. The aim of this study was to analyse the secretion profile of inflammation associated molecules in human first trimester placental explants and to determine how this profile is imbalanced in response to TNF-a. Methods: Human first trimester placental explants (n1⁄48) were cultured in presence or absence of recombinant human TNF-a (10ng/ml) under 2.5% oxygen for 48h. After incubation, conditioned culture media and explant homogenates were collected for subsequent inflammation antibody array and ELISA analysis. A proportion of TNF-a treated placental explants and controls were formalinfixedandparaffin-embedded for immunohistochemistry (IHC) and semi-quantitative RNA based in situ expression analysis (padlock probe approach).Moreover, primary humanfirst trimester trophoblasts (n1⁄44)were incubated with or without TNF-a (10ng/ml) and conditioned culture media and cell lysates were analysed by ELISA after 48h culture. Results: Inflammation antibody array analysis of conditioned culture media from placental explants showed that Interleukin (IL)-6, IL-8, chemokine (C-C motif) ligand 2 (CCL2), CCL4 and granulocyte-macrophage colony-stimulating factor (GM-CSF) were the most abundantly secreted inflammation associated molecules by human first trimester placenta under basal conditions. In presence of TNF-a, the secretory profile of placental explants considerably changed and showed increased secretion of GM-CSF (1.7 fold), CCL5 (2.2 fold) and IL-10 (1.8 fold), when compared to control. ELISA analysis confirmed this observation by showing significantly increased synthesis and release of GM-CSF, CCL5 and IL-10 by placental explants in response to TNF-a. Conclusion: Results from this study provide evidence that TNF-a alters the secretion profile of inflammation associated molecules by human first trimester placenta. Whether this altered secretion profile contributes to development of placental pathologies or whether it is rather an attempt to counterbalance adverse effects of TNF-a is subject of further investigation.