Diversity in cytokine response to bacteria associated with preterm birth by fetal membranes Ramkumar Menon, PhD; Morgan R. Peltier, PhD; Judith Eckardt, MD; Stephen J. Fortunato, MD OBJECTIVE: This study compared cytokine and prostaglandin (PG) re- sponses by fetal membranes stimulated with 4 different bacterial spe- cies associated with preterm birth (PTB). STUDY DESIGN: Fetal membranes (n = 13 from normal term cesarean sections [not in labor]) in an organ explant system were stimulated with heat-killed Ureaplasma parvum, Gardanerella vaginalis, Esche- richia coli, group B Streptococcus (GBS), or lipopolysaccharide (LPS). Cytokines (interleukin [IL]-1, IL-6, IL-8, IL-10, tumor necrosis factor [TNF]-, and interferon-) and PG (PGF 2 and PGE 2 ) concentrations were quantitated and compared. RESULTS: LPS and E coli increased all cytokine and PG productions compared with controls. Cytokine profiles were similar after G vaginalis and GBS stimulation. G vaginalis increased PGE 2 , whereas GBS increased PGF 2 . U parvum demonstrated the mildest response with only IL-10 and TNF- concentrations being higher with no detectible effect on PGs. CONCLUSION: Fetal membrane cytokine signatures of 4 different bac- teria associated with PTB are distinct, suggesting that infection as a potential cause of PTB is not homogeneous in its presentation. Key words: amniochorion, inflammation, intraamniotic infection, preterm birth, prostaglandins Cite this article as: Menon R, Peltier MR, Eckardt J, et al. Diversity in cytokine response to bacteria associated with preterm birth by fetal membranes. Am J Obstet Gynecol 2009;201:306.e1-6. S pontaneous preterm birth (PTB; birth before 37 weeks’ gestation) is a major complication of pregnancy, and infection is associated with approxi- mately 50% of cases. 1-3 These infections are often asymptomatic and consist of bacteria that have ascended from the va- gina through the cervix to colonize the tissues at the maternal-fetal interface 4-6 Microbial factors are thought to evoke a series of events that compromise the immunologic privileges that the fetus enjoys from conception until labor by stimulating the production of proin- flammatory cytokines and chemokines. Cytokines and chemokines, in turn, up- regulate the downstream effectors of specific clinical events of labor. For ex- ample, prostaglandins (PGs) stimulate uterine contractions and cervical dila- tion and matrix metalloproteinases de- grade extracellular matrix in fetal mem- branes, cervix, and placenta 7-11 Many preterm labors are resistant to tocolytics. 11 This suggests that the bio- molecular pathways of preterm labor re- sulting in PTB are complex and that pa- tient-specific interventions targeted to a particular molecular mechanism of pre- term birth may be necessary. Bacteria in- teract with host cells through toll-like re- ceptors (TLRs) that can produce different but overlapping arrays of cyto- kines upon recognition of conserved biochemical motifs on various species of bacteria. For example, TLR-4 recognizes the lipid A component of lipopolysac- charide from Gram-negative species; TLR-2 also recognizes peptidoglycan from Gram-positive bacteria and li- poproteins from different species of Mycoplasma. 12,13 Previously we reported different in vitro inflammatory cytokine responses to lipopolysaccharide (LPS) from Gram- negative bacteria and peptidoglycan polysaccharide from Gram-positive bac- teria by fetal membranes, suggesting that initiation of labor may depend on the type of immune response to the patho- gen present. 14 Other tissue and bacterial factor–specific cytokine response have also been reported. 15-17 Not all cyto- kines, however, seem to activate the downstream mediators that cause labor and PTB. 18,19 Nonhuman primate stud- ies have shown the induction of preterm labor by interleukin (IL)-1 and tumor necrosis factor (TNF)- but not IL-6 or IL-8. 20 Therefore, we hypothesized that the immune response by fetal membranes to bacteria associated with preterm birth differs between specific organisms. We tested this hypothesis using an in vitro system in which clinically normal mem- branes were stimulated with heat-inacti- vated suspensions of Escherichia coli, Streptococcus agalactiae (group B Strep- tococcus [GBS]), Ureaplasma parvum, From the Perinatal Research Center, Centennial Women’s Hospital (Drs Menon, Eckardt, and Fortunato), and Maternal-Fetal Group PediatriX (Dr Fortunato), Nashville, TN; the Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA, and the Department of Obstetrics and Gynecology and Reproductive Medicine, Yale University School of Medicine, New Haven, CT (Dr Menon); and the Departments of Obstetrics and Gynecology and Pediatrics, Winthrop University Hospital, Mineola, NY (Dr Peltier). Presented at the 29th Annual Meeting of the Society for Maternal-Fetal Medicine, San Diego, CA, Jan. 26-31, 2009. Received Feb. 18, 2009; revised May 11, 2009; accepted June 11, 2009. Reprints: Ramkumar Menon, PhD, Department of Epidemiology, Rollins School of Public Health, Emory University, Atlanta, GA 30322. fortnat@edge.net/rmenon3@emory.edu. This study was supported in part by the Maternal-Fetal Group, Nashville, TN. 0002-9378/$36.00 • © 2009 Mosby, Inc. All rights reserved. • doi: 10.1016/j.ajog.2009.06.027 SMFM Papers www. AJOG.org 306.e1 American Journal of Obstetrics & Gynecology SEPTEMBER 2009