Sequelae of Fetal Infection in a Non-human Primate Model of Listeriosis

Abstract

Listeria monocytogenes (Lm) is a common environmental bacterium that thrives on vegetation and soil matter, but can infect humans if contaminated food products are ingested, resulting in severe disease in immunosuppressed populations, including pregnant women and newborns. To better understand how the unique immunological milieu of pregnancy increases susceptibility to infection, we study listeriosis in cynomolgus macaques, a non-human primate that closely resembles humans in placentation and in the physiology, and immunology of pregnancy. Non-human primates are naturally susceptible to Lm infection, and spontaneous abortions due to listeriosis are known to occur in outdoor macaque colonies, making them ideal models to understand the disease pathogenesis and host-pathogen relationship of listeriosis. We have previously shown that Lm infection in the first trimester has a high rate of miscarriage. This study expands on our previous findings by assessing how the quantity of Lm as well as stage of pregnancy at the time of exposure may influence disease susceptibility. In the current study we inoculated a cohort of macaques with a lower dose of Lm than our previous study and although this did not result in fetal demise, there was evidence of in utero inflammation and fetal distress. Animals that were reinfected with an equivalent or higher dose of the same strain of Lm resulted in approximately half of cases continuing to term and half ending in fetal demise. These cases had inconsistent bacterial colonization of the fetal compartment, suggesting that Lm does not need to directly infect the placenta to cause adverse pregnancy outcomes. Timed surgical collection of tissues following inoculation demonstrated that transmission from mother to fetus can occur as soon as 5 days post-inoculation. Lastly, third trimester inoculation resulted in pregnancy loss in 3 out of 4 macaques, accompanied by characteristic pathology and Lm colonization. Collectively, our studies demonstrate that common laboratory culture tests may not always recover Lm despite known maternal ingestion. Notably, we also find it is possible for maternal infection to resolve in some cases with no discernible adverse outcome; however, such cases had evidence of a sterile intrauterine inflammatory response, with unknown consequences for fetal development.

Keywords: cytokines; fetal infection; histopathology; listeriosis; non-human primate; pregnancy.

Figure 1

Experimental timelines for each cohort showing animal ID, sample collections, and pregnancy outcome. The x-axis displays post-inoculation days. The following symbols indicate events on the timeline: positive blood culture (red asterisk), negative blood culture (black asterisk), positive tissue culture (green bacteria), negative tissue culture (negative sign), and positive fecal culture (green box). Fecal samples were collected on each date shown on the timeline; dates without a green box denote negative fecal cultures, and absent dates indicate that no samples were collected. Overlapping symbols indicate concurrent events. (A) First trimester cohort: 10⁶ CFUs Lm initial dose; 10⁷ CFUs Lm reexposure. (B) First trimester cohort: 10⁷ CFUs Lm initial dose; 10⁷ CFUs Lm reexposure. (C) Third trimester cohort: 10⁷–10⁸ CFUs Lm dose.

Figure 2

Third trimester histology. (A) Maternal-fetal interface. Left: A focus of avascular villi entrapped in fibrin (asterisk) in a placenta from a dam exposed to Lm but without discernible infection. Cy29.1 H&E, 2x. Center: Normal placenta with incidental multifocal mineralization (arrowhead). Cy33. H&E, 40x. Right: Lm-infected placenta with acute subchorionic neutrophilic (PMN) inflammation and intervillous microabscesses (M). Cy31. H&E, 2x. (B) Representative tissue findings in Lm-inoculated animals who did not show clinical signs of infection. Left: Fetal liver with mild diffuse subcapsular neutrophil infiltration, intravascular and perivascular neutrophils. Cy27.1. H&E 40x. Center: Fetal lung with rare squamous cells in alveoli. Cy26.1. H&E 40x. Right: Fetal cerebrum with no significant findings Cy26.1. H&E, 10x. (C) Representative tissue findings in Lm-infected animals. Left: Lm+ Fetal liver with intralesional bacteria (arrow). Cy31. H&E, 10x. Center: Lm+ fetal lung with intravascular bacteria (arrow) and alveolus with luminal bacteria, neutrophils, squamous cells, and cellular debris. Cy31. H&E, 40x. Right: Lm+ fetal cerebrum with an intravascular bacterial embolus (asterisk). Cy31. H&E, 10x. (D) Representative tissue findings in control animals. Left: Control fetal liver. Cy33. H&E, 40x. Center: Control fetal lung. Occasional interalveolar squamous cells are an incidental finding commonly seen in cesarean deliveries. Cy33. H&E, 40x. Right: Control fetal cerebrum. Cy33. H&E, 10x.

Figure 3

Histopathology scores. Each dot represents the final score from a single case. Final scores were calculated from the sum of abnormal histological findings noted in both fetal and maternal-fetal interface tissues. Central black lines indicate the average score of a group.

Figure 4

Cytokine cluster analysis of third trimester placenta samples. (A) Principle component analysis based on the variation between subgroups. Original values were ln(x + 1)-transformed. Singular value decomposition with imputation was used to calculate principal components. X and Y axes show principal component 1 and principal component 2 that explain 56.4 and 37.2% of the total variance, respectively. (B) Heat map showing relative expression of all 30 cytokines assayed. Original values were ln(x + 1)-transformed. Pareto scaling was applied to rows. Both rows and columns were clustered using Manhattan distance and average linkage. Dendrograms indicate tightness of clusters.

Figure 5

Tissue bacterial burden Bacterial burden of Lm from individual pregnancies. The y-axis shows tissues and the x-axis shows the bacterial burden in colony forming units per gram (CFUs/g) in log scale for each tissue. Central black lines indicate the mean CFUs/g for replicate samples.

Figure 6

Histology of the first trimester Lm infection. (A) Umbilical cord. Neutrophilic umbilical arteritis. Arrow points to clustered neutrophils. Cy30.2. H&E, 20x. (B) Control umbilical cord. Cy26c H&E, 20x. (C) Amniotic membranes with Gram-positive bacteria and neutrophilic infiltrate (asterisk). Cy30.2. H&E, 4x. (D) Control amniotic membranes. Cy26c H&E, 20x. (E) Fetal lung with listerial infiltration of pulmonary interstitium between alveoli. Arrows point to bacterial foci. Cy30.2 H&E, 20x. (F) Control fetal lung. A indicates alveolus. Cy26c H&E, 20x.

Figure 7

Fetal bacterial distribution with Lm infection in the first trimester (Cy25.1). (A) Fetal cranium with large clusters of Gram-positive rods within the ocular musculature, diffusely within the nasal sinus and nasal cartilage, multifocally within the neural tissue, periosteum of the skull, and within the dermis, and subcutaneous tissues. Gram stain, 2x (inset 10x). (B) Fetal lung with multiple foci of intravascular bacteria. Gram stain, 2x (inset 40x). (C) Fetal liver, vertebrae, intestines, kidney, and adrenal gland, with multiple foci of bacterial colonization of the kidney, adrenal gland, intestinal serosa, hepatic capsule and interstitium, periosteum, and perichondrium. Gram stain, 2x (inset liver 40x). (D) Fetal membranes (chorioamniotic membranes) with multiple foci of Gram-positive bacteria. Gram stain 2x (inset 10x). Inset (Lo4): Immunohistochemical staining of fetal liver confirming Lm-positive foci.