Choriodecidual group B streptococcal inoculation induces fetal lung injury without intra-amniotic infection and preterm labor in Macaca nemestrina

Abstract

Background: Early events leading to intrauterine infection and fetal lung injury remain poorly defined, but may hold the key to preventing neonatal and adult chronic lung disease. Our objective was to establish a nonhuman primate model of an early stage of chorioamnionitis in order to determine the time course and mechanisms of fetal lung injury in utero.

Methodology/principal findings: Ten chronically catheterized pregnant monkeys (Macaca nemestrina) at 118-125 days gestation (term=172 days) received one of two treatments: 1) choriodecidual and intra-amniotic saline (n=5), or 2) choriodecidual inoculation of Group B Streptococcus (GBS) 1×10(6) colony forming units (n=5). Cesarean section was performed regardless of labor 4 days after GBS or 7 days after saline infusion to collect fetal and placental tissues. Only two GBS animals developed early labor with no cervical change in the remaining animals. Despite uterine quiescence in most cases, blinded review found histopathological evidence of fetal lung injury in four GBS animals characterized by intra-alveolar neutrophils and interstitial thickening, which was absent in controls. Significant elevations of cytokines in amniotic fluid (TNF-α, IL-8, IL-1β, IL-6) and fetal plasma (IL-8) were detected in GBS animals and correlated with lung injury (p<0.05). Lung injury was not directly caused by GBS, because GBS was undetectable in amniotic fluid (~10 samples tested/animal), maternal and fetal blood by culture and polymerase chain reaction. In only two cases was GBS cultured from the inoculation site in low numbers. Chorioamnionitis occurred in two GBS animals with lung injury, but two others with lung injury had normal placental histology.

Conclusions/significance: A transient choriodecidual infection can induce cytokine production, which is associated with fetal lung injury without overt infection of amniotic fluid, chorioamnionitis or preterm labor. Fetal lung injury may, thus, occur silently without symptoms and before the onset of the fetal systemic inflammatory response syndrome.

Figure 1. Histopathology of the fetal lungs.

Hematoxylin and eosin stained histologic sections of fetal lung are shown for a saline control (A; lung score = 0) and GBS animal with severe fetal lung injury (B; lung score = 4).

Figure 2. Uterine activity and amniotic fluid cytokines.

Temporal relationships among inoculation of GBS or saline, uterine activity, and amniotic fluid (AF) cytokines (TNF-α, IL-8) are shown in a representative animal after saline inoculation (A) and in an animal after GBS inoculation (B), which developed moderate lung injury in the setting of minimal uterine activity. The x-axis represents gestational age in days ranging from the vascular implantation surgery until cesarean section. The y-axis is hourly contraction area (HCA; gray bars), or the level of amniotic fluid TNF-α (red line) or IL-8 (blue line). CD, choriodecidual; AF, amniotic fluid.

Figure 3. Histopathology of chorioamnion.

Hematoxylin and eosin stained histologic sections of chorioamnion (fetal membranes) are shown for a saline control (A) and GBS animal with chorioamnionitis (B). Neutrophils in the chorion are indicated with arrows in panel B, as well as being abundant in the decidua.

Figure 4. Our conceptual model.

1. Bacteria from the lower genital tract ascends into the choriodecidual space. 2. Inflammatory mediators (e.g. IL-8) produced by decidua and/or membranes diffuse into amniotic fluid and the fetal lung. 3. Fetal lung injury is induced by inflammatory mediators. IL, Interleukin; AF, amniotic fluid.