Metformin use in pregnancy is increasingly common, even beyond treatment of diabetes. Metformin crosses the placenta and bioaccumulates at equimolar concentrations near maternal circulation. Concerningly, metformin can suppress mitochondrial complex I activity in adults, limiting oxidative metabolism and ATP production. Examine the effect of metformin on fetal growth and muscle metabolism in Rhesus macaques. Twice-daily metformin (MET, 10 mg/kg) or placebo was given at 30 days of pregnancy with a chow (mCD) or Western diet (mWD). Fetal (n = 25F/17M) muscle was collected at g145 (of g164). Respiration was measured with carbohydrate or lipid substrates in permeabilized fiber bundles (PFB) from gastrocnemius and soleus or isolated mitochondria (mitos) from rectus femoris. Female mCD+MET offspring had reduced weight compared to mWD+MET (p = 0.02), with no differences in males. Muscle cross-sectional area was reduced with MET, independent of diet, in gastrocnemius but not soleus, suggesting fiber type-specific effects. In mCD, MET decreased CI reliance (p = 0.03) in gastrocnemius and soleus PFB. Increased fat oxidation and respiratory capacity in mWD was blunted by MET in PFB. Lipid oxidation (p = 0.03) was lower in mWD+MET mitos concomitant with reduced ATP production. CI and CI+II carbohydrate-respiration was higher in mCD+MET vs. mCD mitos (p = 0.04) but without greater ATP production. Metformin resulted in subtle changes in fetal weight and was principally associated with reduced glycolytic muscle fiber size, mitochondrial respiration and ATP generation. As muscle mass correlates with functional capacity and insulin sensitivity throughout life, caution is warranted in using metformin during pregnancy. (Supported by NIH DK128187)

Metformin is used in adults with diabetes and by pregnant women. However, metformin crosses the placenta and may restrict fetal growth. In primary fetal hepatocytes, we demonstrated that metformin decreases O2 consumption and glucose production, consistent with its effects in adults, yet induces fetal-specific metabolic stress. Determine whether metformin, in combination with Western diet (WD), impairs oxidative metabolism, injuring the fetal liver. Pregnant Rhesus macaques received vehicle (VEH) or metformin (MET) and a healthy control diet (CD) or WD (n = 10/group) from 0.2 to 0.85 gestation. After cesarean delivery, cord blood samples, fetal weights, and isolated fetal hepatocytes were obtained. Hepatocytes from MET vs. VEH fetuses had higher O2 consumption rates with greater maximal (P = 0.015) and spare respiratory capacity (P = 0.05). There were tendencies for increased MFN2 protein (P = 0.09) and PGC1A gene expression (P = 0.06) in liver tissue, indicating increased mitochondrial fusion and biogenesis. MET-exposure did not affect WD-induced liver triglyceride accumulation nor cord blood glucose or O2 levels. However, histology showed increased collagen with MET and WD exposure. Furthermore, WD+MET fetuses had increased blood lactate ­concentrations. Fetal weight was similar across groups; however, CD+MET versus CD+VEH female fetuses weighed less (interaction, P = 0.02). Metformin is associated with fetal metabolic stress, resulting in mitochondrial remodeling and increased oxygen consumption capacity. Further, increased lactate may underlie ­collagen deposition and hepatocellular injury. Ongoing studies are underway to understand if liver metabolic effects are linked with sex-dependent growth effects. (Supported by NIH-R01-DK128187)

In utero therapies allow for prevention of irreversible fetal morbidity. While preferential transplacental delivery for fetal treatment is challenging, lipid nanoparticles (LNPs) offer an established safety profile, and lipoprotein-rich fetal blood provides a compatible environment. Since modifications in LNP composition allow for fetal tissue target specificity, we hypothesized that a specialized LNP delivery vehicle with near infrared (NIR) fluorescence detection would enable targeted fetal delivery which could be readily visualized in vivo after peripheral administration to a gravid mouse. DiR, a NIR fluorescent dye, and other lipids dissolved in EtOH were rapidly mixed with buffer and sonicated to form DiR LNPs at 6 mM lipid concentration. 1.7 mg/uL DiR LNPs were injected into the tail veins of e14.5 gravid mice (n=2). Images were obtained at 4 timepoints before removing the maternal uterus, liver, kidneys, and viscera for imaging. The fetuses and placentae were subsequently imaged as well. As anticipated, maternal in vivo images demonstrated strong fluorescent LNP signal in the liver at all timepoints (Fig1A). Initial faint LNP signal in the maternal uterus became concentrated in the distribution of the L uterine horn (Fig1B). Excitingly, the fluorescent LNP signal was observed in every placenta and fetus (Fig1C), and honed to the fetal liver. At high resolution, revealing there was preferential LNP signal in the fetal live over the other abdominal viscera (Fig1D). These data show, for the first time, successful in vivo trans-placental LNP delivery to fetal mice via intravenous injection of a gravid mouse, with preferential uptake in the fetal liver. These findings are an exciting next step in the development of a novel, maternal peripherally delivered yet transplacental LNP delivery of cargo to the fetal liver for in utero fetal therapies.

Gut microbes moderate and/or produce metabolites like serotonin and short-chain fatty acids (SCFAs) that are crucial for normal behavior. In a Japanese macaque model, offspring exposed to maternal Western-Style diet (mWSD) display anxiety, despite being weaned onto a control diet. We hypothesized that maternal mWSD alters the gut microbiome’s ability to moderate the production of serotonin and SCFA pathway metabolites, contributing to the anxiety phenotype. Dams were fed a Chow control diet or mWSD during pregnancy and lactation, then offspring were weaned onto Chow or Western-Style diet. Voided stool, colon contents, and intestinal mucosa were sampled throughout development ranging from fetal (gestational day 130) to pre-puberty (36-40 months) (Fig. 1A). DNA from these tissues (total n=1223 + n=27 kit negative controls) underwent whole-genome sequencing, taxonomic classification with MetaPhlAn4, functional classification with HUMAnN3, and differential abundance determination by MaAsLin2. mWSD drove differential abundance of n=37 taxa in voided stool, independent of offspring sex, offspring diet, or repeat sampling. Presented at the family level, mWSD was associated with depleted Campylobacteraceae (p=0.002) and Lactobacillaceae (p=0.018), and enriched Lachnospiraceae (p=0.034) and Ruminococcaceae (p=0.028) (FDR adjusted p values) (Fig. 1B). At the functional level, mWSD was associated with suppression of gene expression, with n=395 genes down-regulated vs n=122 up-regulated (2-fold change threshold). Genes responsible for tryptophan (serotonin’s precursor) availability and SCFA phosphorylation were persistently downregulated at all time points in stool from offspring of dams fed a mWSD (Fig. 1C). Gut microbiome composition and bacterial gene function were found to be significantly altered by mWSD exposure, even when offspring were weaned onto and fed a control diet for 2.5 years. Downregulation of serotonin’s precursor and SCFA-related gene functions may promote the increased anxiety observed in these offspring via altered microbe-to-host signaling in the gut.

Maternal Western-Style diet (mWSD) feeding and obesity during pregnancy/breastfeeding have been linked to increased offspring anxiety. Anxiety is the result of decreased serotonin, an important neurotransmitter in mood regulation. Serotonin is synthesized from tryptophan and degraded into 5-hydroxyindoleacetic acid (5-HIAA). We hypothesized that mWSD feeding in a primate model would alter her offspring’s serotonin pathway metabolites, leading to persistent and evident offspring anxiety at 3 years of age (peripuberty). Japanese macaque dams were fed a Chow control diet or mWSD during gestation and lactation, then offspring were weaned onto Chow or Western-Style diet (Fig. 1A). Anxiety was tested by multiple validated measures at 34 months and scored by blinded reviewers. Serum was collected at 36-40 months and analyzed by targeted liquid chromatography-mass spectrometry of all known serotonin pathway metabolites and other metabolites known to affect behavioral neuroactivity (gamma-aminobutyric acid and short-chain fatty acids). Anxiety was calculated as the sum of four components: high energy outbursts and reactive, ritualized, and inactive behaviors (Fig. 1B). Total anxiety scores display a right shift (more frequent occurrence) in males (Z-test p= 0.005, Fig. 1C). We detected 18 neuroactive metabolites in serum which withstood confirmatory factor analysis, revealing statistically significant inverse relationships between 5-HIAA levels and ritualized anxiety (p=0.004) and high energy outbursts (p=0.048, Fig. 1D). We demonstrated that mWSD feeding programs male offspring with persistently increased anxiety, despite 2.5 years of control diet feeding. Analysis of serotonin metabolites showed an inverse correlation between 5-HIAA and anxiety, indicating low levels of serotonin inferred from its degradation product that correspond with higher levels of anxiety. These data underscore the importance of dietary counseling during pregnancy and lactation to potentially prevent unintended programming of offspring neuroactivity and anxiety behavior.

BACKGROUND:Maternal periodontal disease is associated with preterm and low-birthweight deliveries, but randomized trials of likely efficacious treatments (e.g., dental scaling and root planing) during pregnancy have not reduced these adverse outcomes. As an alternative, we hypothesized that periconception initiation of xylitol chewing gum would reduce the occurrence of preterm or low-birthweight deliveries among a historical high-prevalence population in Malawi. METHODS:We conducted an open-label, parallel-enrollment, matched-pair, cluster-randomized, controlled clinical trial across eight health centers (sites) in and around Lilongwe, Malawi. Sites were paired by anticipated delivery volume and randomized to prenatal and oral health education alone (active control) or with twice-daily xylitol chewing gum (intervention) throughout the periconception and antenatal periods. For the primary prevention of preterm (<37 weeks) and low-birthweight (<2,500 g) deliveries (co-primary outcomes), comparison by allocation group was performed using generalized linear mixed models for each outcome as a fixed factor and the site(s) as a random factor. FINDINGS:10,069 participants were enrolled (n = 4,549 at intervention sites, n = 5,520 at active control sites), with >95% available for analyses. Initiation of xylitol chewing gum resulted in significant reductions in the co-primary outcomes: preterm birth (12.6% [549/4,349] vs. 16.5% [878/5,321]; relative risk [RR] 0.76, 95% confidence interval [CI] 0.57-0.99) and <2,500-g neonates (8.9% [385/4,305] vs. 12.9% [679/5,260]; RR 0.70, 95% CI 0.49-0.99). Xylitol chewing gum use also led to fewer neonatal demises (0.2% [8/4,305] vs. 0.4% [22/5,260]; RR 0.41, 95% CI 0.19-0.89). CONCLUSIONS:Periconception initiation and ongoing use of xylitol chewing gum significantly reduced the occurrence of preterm and low-birthweight deliveries in Malawi. FUNDING:E.W. Al Thrasher Foundation (to K.A.) and USAID Saving Lives at Birth Grand Challenges Grant AID-OAA-G-11-00062 (to K.A.). Additional financial and in-kind support was graciously provided by Texas Children's Hospital and Baylor Foundation Malawi.

BACKGROUND:In recent years, pragmatic metformin use in pregnancy has stretched to include prediabetes mellitus, type 2 diabetes mellitus, gestational diabetes mellitus, and (most recently) preeclampsia. However, with its expanded use, concerns of unintended harm have been raised. OBJECTIVE:This study developed an experimental primate model and applied ultrahigh performance liquid chromatography coupled to triple-quadrupole mass spectrometry for direct quantitation of maternal and fetal tissue metformin levels with detailed fetal biometry and histopathology. STUDY DESIGN:Within 30 days of confirmed conception (defined as early pregnancy), 13 time-bred (timed-mated breeding) Rhesus dams with pregnancies designated for fetal necropsy were initiated on twice-daily human dose-equivalent 10 mg/kg metformin or vehicle control. Pregnant dams were maintained as pairs and fed either a control chow or 36% fat Western-style diet. Metformin or placebo vehicle control was delivered in various treats while the animals were separated via a slide. A cesarean delivery was performed at gestational day 145, and amniotic fluid and blood were collected, and the fetus and placenta were delivered. The fetus was immediately necropsied by trained primate center personnel. All fetal organs were dissected, measured, sectioned, and processed per clinical standards. Fluid and tissue metformin levels were assayed using validated ultrahigh performance liquid chromatography coupled to triple-quadrupole mass spectrometry in selected reaction monitoring against standard curves. RESULTS:Among 13 pregnancies at gestational day 145 with fetal necropsy, 1 dam and its fetal tissues had detectable metformin levels despite being allocated to the vehicle control group (>1 μmol metformin/kg maternal weight or fetal or placental tissue), whereas a second fetus allocated to the vehicle control group had severe fetal growth restriction (birthweight of 248.32 g [<1%]) and was suspected of having a fetal congenital condition. After excluding these 2 fetal pregnancies from further analyses, 11 fetuses from dams initiated on either vehicle control (n=4: 3 female and 1 male fetuses) or 10 mg/kg metformin (n=7: 5 female and 2 male fetuses) were available for analyses. Among dams initiated on metformin at gestational day 30 (regardless of maternal diet), significant bioaccumulation within the fetal kidney (0.78-6.06 μmol/kg; mean of 2.48 μmol/kg), liver (0.16-0.73 μmol/kg; mean of 0.38 μmol/kg), fetal gut (0.28-1.22 μmol/kg; mean of 0.70 μmol/kg), amniotic fluid (0.43-3.33 μmol/L; mean of 1.88 μmol/L), placenta (0.16-1.00 μmol/kg; mean of 0.50 μmol/kg), fetal serum (0.00-0.66 μmol/L; mean of 0.23 μmol/L), and fetal urine (4.10-174.10 μmol/L; mean of 38.5 μmol/L) was observed, with fetal levels near biomolar equivalent to maternal levels (maternal serum: 0.18-0.86 μmol/L [mean of 0.46 μmol/L]; maternal urine: 42.60-254.00 μmol/L [mean of 149.30 μmol/L]). Western-style diet feeding neither accelerated nor reduced metformin bioaccumulations in maternal or fetal serum, urine, amniotic fluid, placenta, or fetal tissues. In these 11 animals, fetal bioaccumulation of metformin was associated with less fetal skeletal muscle (57% lower cross-sectional area of gastrocnemius) and decreased liver, heart, and retroperitoneal fat masses (P<.05), collectively driving lower delivery weight (P<.0001) without changing the crown-rump length. Sagittal sections of fetal kidneys demonstrated delayed maturation, with disorganized glomerular generations and increased cortical thickness. This renal dysmorphology was not accompanied by structural or functional changes indicative of renal insufficiency. CONCLUSION:Our study demonstrates fetal bioaccumulation of metformin with associated fetal growth restriction and renal dysmorphology after maternal initiation of the drug within 30 days of conception in primates. Given these results and the prevalence of metformin use during pregnancy, additional investigation of any potential immediate and enduring effects of prenatal metformin use is warranted.

Intrahepatic cholestasis of pregnancy (IHCP) is a poorly understood disease known to have hormonal and metabolic links. In a discovery approach, we sought to naively assess a large database of pregnancies using machine learning and natural language processing (NLP) for novel insights into pathology and risk. Peribank, a research database of 46,641 pregnancies containing 12,017 data entry fields, was natural language processed using spaCy pretrained models, with Boolean and scalar data retained. Medications were corrected against MEDI-Ensemble using Levenshtein distance fuzzy matching. Post processing filtering left ∼450,000 unique features that were assessed for predictive feature selection with scikit-learn. Logit regression was used for select features. There were 1374 IHCP pregnancies discovered in the database. Feature selection revealed top scoring predictors of IHCP (positive or negative) in decreasing score as T2DM as indication for antenatal testing, T2DM, metformin use at conception, white maternal race, white paternal race (both inclusive of Hispanic ethnicity), no medications at conception, tetanus vaccination, high alkaline phosphatase, high glucose, flu vaccination, and GA at admittance. Diabetes and metformin use were then assessed independently for risk. T2DM patients had increased risk for IHCP (OR 1.44, 95% CI 1.05-1.98, p=0.023) as did GDM (1.54, CI 1.31-1.81, p=3.5x10-7), while there was no association with Type I diabetes or metformin use independently. The associations were confirmed by Χ2 for T2DM (p=0.0029), GDM (p=1.3 x10-7) and metformin use (p=0.056) (Fig. A). A sub analyses of only diabetic patients of all types (n=5110) found metformin treated patients more likely to develop IHCP than others (14.0% vs. 10.8%) though it did not reach significance (p=0.17, Χ2) (Fig. B). The novel association of risk for IHCP with T2DM adds to the known association of GDM. Given the absence of association with T1DM, and the possible influence of metformin, the results support the theory that IHCP pathology is linked to hepatic insulin resistance.

Background: Zika virus congenital infection evades double-stranded RNA detection and may persist in the placenta for the duration of pregnancy without accompanying overt histopathologic inflammation. Understanding how viruses can persist and replicate in the placenta without causing overt cellular or tissue damage is fundamental to deciphering mechanisms of maternal-fetal vertical transmission. Objective: Placenta-specific microRNAs are believed to be a tenet of viral resistance at the maternal-fetal interface. We aimed to test the hypothesis that the Zika virus functionally disrupts placental microRNAs, enabling viral persistence and fetal pathogenesis. Study design: To test this hypothesis, we used orthogonal approaches in human and murine experimental models. In primary human trophoblast cultures (n=5 donor placentae), we performed Argonaute high-throughput sequencing ultraviolet-crosslinking and immunoprecipitation to identify any significant alterations in the functional loading of microRNAs and their targets onto the RNA-induced silencing complex. Trophoblasts from same-donors were split and infected with a contemporary first-passage Zika virus strain HN16 (multiplicity of infection=1 plaque forming unit per cell) or mock infected. To functionally cross-validate microRNA-messenger RNA interactions, we compared our Argonaute high-throughput sequencing ultraviolet-crosslinking and immunoprecipitation results with an independent analysis of published bulk RNA-sequencing data from human placental disk specimens (n=3 subjects; Zika virus positive in first, second, or third trimester, CD45(-) cells sorted by flow cytometry) and compared it with uninfected controls (n=2 subjects). To investigate the importance of these microRNA and RNA interference networks in Zika virus pathogenesis, we used a gnotobiotic mouse model uniquely susceptible to the Zika virus. We evaluated if small-molecule enhancement of microRNA and RNA interference pathways with enoxacin influenced Zika virus pathogenesis (n=20 dams total yielding 187 fetal specimens). Lastly, placentae (n=14 total) from this mouse model were analyzed with Visium spatial transcriptomics (9743 spatial transcriptomes) to identify potential Zika virus-associated alterations in immune microenvironments. Results: We found that Zika virus infection of primary human trophoblast cells led to an unexpected disruption of placental microRNA regulation networks. When compared with uninfected controls, Zika virus-infected placentae had significantly altered SLC12A8, SDK1, and VLDLR RNA-induced silencing complex loading and transcript levels (-22; adjusted P value <.05; Wald-test with false discovery rate correction q<0.05). In silico microRNA target analyses revealed that 26 of 119 transcripts (22%) in the transforming growth factor-beta signaling pathway were targeted by microRNAs that were found to be dysregulated following Zika virus infection in trophoblasts. In gnotobiotic mice, relative to mock controls, Zika virus-associated fetal pathogenesis included fetal growth restriction (P=.036) and viral persistence in placental tissue (P=.011). Moreover, spatial transcriptomics of murine placentae revealed that Zika virus-specific placental niches were defined by significant up-regulation of complement cascade components and coordinated changes in transforming growth factor-beta gene expression. Finally, treatment of Zika virus-infected mice with enoxacin abolished placental Zika virus persistence, rescued the associated fetal growth restriction, and the Zika virus-associated transcriptional changes in placental immune microenvironments were no longer observed. Conclusion: These results collectively suggest that (1) Zika virus infection and persistence is associated with functionally perturbed microRNA and RNA interference pathways specifically related to immune regulation in placental microenvironments and (2) enhancement of placental microRNA and RNA interference pathways in mice rescued Zika virus-associated pathogenesis, specifically persistence of viral transcripts in placental microenvironments and fetal growth restriction.

Gut iron absorption is regulated to allow for optimal bioavailability & overcome iron losses during menstruation. However, clinical response to iron supplementation is unpredictable. We hypothesized that the gut microbiome may explain iron absorption disparity among healthy, non-anemic menstruating people (pwm).

Limosilactobacillus reuteri is a commensal bacteria and common neonatal probiotic that rarely naturally occurs in humans. We hypothesized that 3rd trimester oral administration of L. reuteri, but not placebo, would modify the maternal and neonatal microbiome composition (community ecology). Healthy gravidae (n=30) planning a scheduled Cesarean were recruited at 28-36 wks from Sahlgrenska University Hospital in Göteborg, Sweden and block randomized to 5 arms. A probiotic in lozenge, capsule or chewing gum formulation containing 200 million CFU of L. reuteri DSM17938 and L. reuteri ATCC PTA5289 (1:1), or a lozenge or capsule placebo, taken twice daily. Maternal oral, vaginal, and stool samples were taken at enrollment (V1), admission/delivery (V2) and 2-6 weeks postpartum (V3); breast milk, neonatal stool & oral, were sampled at V2 & V3. Samples were deeply WGS metagenomic sequenced (Illumina) for strain detection, and analyzed after removal of human reads and contaminants, with taxonomic classification on Kraken2. Classified reads were mapped directly to the probiotic strain L. reuteri reference genome using bowtie2. Mapping directly to the probiotic L. reuteri strains, 1876 reads were found in placebo samples compared to 7595 in probiotic (p=0.00004, χ2) indicating probiotic strain level enrichment for treated groups. Body site specific comparisons were undertaken at each time point, demonstrating significant changes in the community ecology at some but not all maternal & neonatal sites in 3rd trimester probiotic arms (permanova < 0.001, Fig 1). Cooccurrence analysis revealed 120 significant species level interactions in the probiotic group (Fig 2B), and 91 in the placebo group (Fig 2A). Notably, no exclusions or associations were found between vaginal hallmark species L.crispatus, L.iners, nor Gardnerella vaginalis. A third trimester probiotic is sufficient to change the ecology of the maternal vaginal, breast milk and neonatal stool and oral microbiome genera.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

The prevalence of respiratory syncytial virus (RSV), the leading cause of infant hospitalizations, rose >10-fold during the SARS-CoV-2 Delta-variant surge, resulting in unprecedented SARS-CoV-2 and RSV co-infections. Infant susceptibility to RSV and SARS-CoV-2 pediatric inflammatory multisystem syndrome remain crucial gaps in our knowledge. Therefore, we aimed to generate high-resolution single-cell RNA-sequencing (scRNA-seq) profiles of SARS-CoV-2 and RSV infections. We hypothesized synergistic immune evasion mechanisms from each virus would lead to shifts in inflammatory signaling and adaptive immune cell recruitment. Single cells from de-identified bronchioalveolar lavage fluids (BALF) from infant subjects (< 1 year old;N=6) hospitalized with singular SARS-CoV-2 or RSV infections were processed for 10x Genomics 3' scRNA-seq. Reads were aligned to a custom reference transcriptome with CellRanger. Further quality control, integration, dimension reduction, cluster analysis, and differential expression were done using the package Seurat. We identified 27,642 cells in 15 distinct subpopulations defined by canonical marker transcripts (Fig. 1). SARS-CoV-2 transcripts were detected in 80 cells, while RSV was observed in 1,309 cells. Most viral transcription was observed in macrophages (Fig. 2a). The top markers for RSV or SARS-CoV-2 infections based on differential expression (ln(fold-change) >0.693;q< 0.05) identified 625 significantly upregulated genes for SARS-CoV-2 and 271 for RSV. Pathway analyses revealed RSV infection was associated with increased interferon signaling while SARS-CoV-2 had increased antigen presentation and complement signaling (Fig. 2b). To our knowledge, this is the first study to analyze RSV transcription at the single-cell level. Comparisons between SARS-CoV-2 and RSV infections identified virus-specific differences in macrophage polarization, B cell activation, and dendritic cell recruitment. Further analyses from co-infection BALF samples will be critical to understanding synergistic and antagonistic antiviral programs in vulnerable neonates.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

Zika virus (ZIKV) causes congenital disease in association with placental replication and fetal infection. We have previously demonstrated in a germ-free (GF) mouse model that the absence of microbes in pregnant mice increases ZIKV vertical transmission, leading to greater fetal demise. Here, we hypothesized that reinoculation of GF mice with its natural microbiome would restore endogenous resistance against ZIKV. Our aim was to measure differences in the ZIKV loads and the microbiome in GF, adult recolonized, and control mice. Timed pregnant specific-pathogen-free (SPF) or GF Swiss-Webster mice were mock injected or received 4 doses of 1x104 PFU first passage ZIKV on embryonic days 4 through 7. Adult GF mice that were microbiome recolonized by first co-housing in an SPF cage were also infected with ZIKV. Mouse placentas and fetal pups were extracted via sterile Cesarean on embryonic day E18.5, homogenized, and then analyzed via RT-qPCR ZIKV load. Stool was collected on E18.5 to assess differences in maternal microbiome via V4 16S sequencing. GF mouse placentas did not have significantly higher ZIKV loads or incidence of infection than placentas of SPF mice. However, adult recolonized dams had significantly greater ZIKV loads by several orders of magnitude and incidence of infection (100%) compared to SPF or GF (p< 0.0001, Fig 1). The microbiomes were not significantly different between SPF mock- or ZIKV-infected mice. In contrast, adult recolonized ZIKV+ mice had microbiomes significantly different from both SPF groups (p=0.025, Fig 2). We tested the novel hypothesis that reintroduction of microbes to GF mice would restore normal resistance to ZIKV by restoring the normal microbiome. Interestingly, the microbiome of adult recolonized mice was not restored, and concurrently the endogenous resistance to ZIKV was significantly less with worse rates of ZIKV infection. These observations indicate that adult recolonized mice should not be considered "normal" mice, and the normal microbiome prior to onset of adult life is needed for endogenous congenital viral resistance.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

Oligosaccharides have diverse important roles in pathogen defense like promoting mucosal immune development. Some are exclusively digested by bacteria, while sialylated oligosaccharides are both bacterial and viral pathogen decoys. Here we set out to measure all oligosaccharides from mid trimester amniocentesis and characterize their composition by gestational age at time of collection, and predictive (or not) of PTB. Midtrimester (15-28 weeks) amniotic fluid from n=519 pregnancies underwent comprehensive oligosaccharide profiling. Full chromatograms quantitatively assessed by a nanoHPLC-chip/TOF mass spectrometer. A subset (n=445) were metagenomically sequenced for bacterial DNA for intrauterine exposure. Reads were rigorously filtered against contamination controls and for robust bacterial read mapping. Correlational and multivariate analysis controlling for multiple comparisons was performed. A total of 36 discrete oligosaccharides (subunit size of 3 to 13 monosaccharides) were detected, with multivariate Ward's clustering analysis showing a strong association with gestational age (GA; Fig. 1), of whom a significant number withstood adjustment for multiple comparisons (FDR< 0.05; Fig. 2A-B). There was no association with PTB (spontaneous or overall), race/ethnicity, parity, prior history of PTB, maternal age nor amniocenteses reason. Interestingly, oligosaccharide complexity trended to simplicity with advancing GA: four subunit oligosaccharides increased while 9, 10, or 12 monosaccharides decreased (p< 0.0001, 2-way ANOVA; Fig 2C) as did sialylated oligosaccharides (r=-0.36, p=4x10-17)(Fig. 2D). Detection of bona fide bacterial reads also decreased with GA (r=-0.21, FDR=0.0003). Our novel findings suggest that (1) oligosaccharides are abundant in amniotic fluid and undergo predictable composition changes during mid-trimester, (2) decreasing sialylated oligosaccharides that correspond to decreases in bacterial DNA, suggest these changes be related to a role of amniotic fluid oligosaccharides in resisting intraamniotic pathogens.View Large Image Figure ViewerDownload Hi-res image Download (PPT)

Background: Functional placental niches are presumed to spatially separate maternal-fetal antigens and restrict the vertical transmission of pathogens. We hypothesized a high-resolution map of placental transcription could provide direct evidence for niche microenvironments with unique functions and transcription profiles. Methods: We utilized Visium Spatial Transcriptomics paired with H & E staining to generate 17,927 spatial transcriptomes. By integrating these spatial transcriptomes with 273,944 placental single-cell and single nuclei transcriptomes, we generated an atlas composed of at least 22 subpopulations in the maternal decidua, fetal chorionic villi, and chorioamniotic membranes. Findings: Comparisons of placentae from uninfected healthy controls (n = 4) with COVID-19 asymptomatic (n = 4) and symptomatic (n = 5) infected participants demonstrated that severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) detection in syncytiotrophoblasts occurred in both the presence and the absence of maternal clinical disease. With spatial transcriptomics, we found that the limit of detection for SARS-CoV-2 was 1/7,000 cells, and placental niches without detectable viral transcripts were unperturbed. In contrast, niches with high SARS-CoV-2 transcript levels were associated with significant upregulation in pro-inflammatory cytokines and interferon stimulated genes, altered metallopeptidase signaling (TIMP1), with coordinated shifts in macrophage polarization, histiocytic intervillositis, and perivillous fibrin deposition. Fetal sex differences in gene expression responses to SARS-CoV-2 were limited, with confirmed mapping limited to the maternal decidua in males. Conclusions: High-resolution placental transcriptomics with spatial resolution revealed dynamic responses to SARS-CoV-2 in coordinate microenvironments in the absence and presence of clinically evident disease.

Cells within the fetal cerebrospinal fluid (CSF) at 22-26 weeks' gestation are only now being characterized. During fetoscopic myelomeningocele repair, fetal CSF is drawn from the lesion as part-and-parcel of the repair. We hypothesized most of these cells would be neuronal and neuroepithelial cell-types. We have generated a repository of paired amniotic fluid (AF) and fetal CSF specimens from >50 fetoscopic surgeries consenting for use. We used orthogonal methods to catalog cell populations in the fetal CSF during gestation. We performed flow cytometry to quantify CD34+ hematopoietic stem cells, CD56+/CD45+ NK cells, and CD133+/CD45+ hematopoietic progenitor cells (N=4) and established a neurosphere culture system that could be maintained for 4 weeks (N=3). Lastly, we performed single-cell RNA-sequencing (scRNA-seq) as an unbiased approach to characterize paired fetal CSF and AF cell populations (N=9). By flow, neural precursor cells represented approximately 39% of the CSF cells, leaving many cell-types uncharacterized. Therefore, we utilized scRNA-seq resulting in 22,058 total cells clustered into 18 distinct cell-types (Fig 1a). We then focused on 3,742 transcriptomes of neuronal origin (Fig 1b), revealing fetal CSF neuronal cells were mainly composed of microglia, neuroblasts, and immature neurons (Fig 1c). A cluster of putative ARC capsids, which are endogenous 40nm virus-like capsids carrying neurotransmitter RNA, demonstrated that ARC was expressed in 261 CSF cells from 87% of the samples (Fig 2b). 73 transcriptomes putatively annotated as ARC capsids revealed RNA content payloads associated with cadmium ion response, cell cycle, and transcription regulation (GO pathways, f< 0.05 controlling for multiple comparisons). Neuronal cell-types in the mid-gestation fetal CSF of ongoing pregnancies provide an unparalleled picture of neuronal development. Further analyses of the recently discovered ARC capsid neuronal signaling system will unveil the sophisticated differentiation trajectories of neurogenesis in utero and how neurons may be engineered or repaired.View Large Image Figure ViewerDownload Hi-res image Download (PPT)