Introduction

The immune system plays a crucial role in protecting the human body from various diseases1, infections and injuries. Immune cells produce cytokines in response to different stimuli, which act to direct additional immune responses. Cytokine levels help guide the immune response following infectious and inflammatory exposures, such as COVID-19 and chorioamnionitis. The COVID-19 pandemic has kindled many questions about the in-utero impact of maternal viral infections on the developing fetus. Studies to date have focused on COVID-19 antibody transfer across the placenta1 and postnatal complications in COVID-19 exposed neonates.2 One study characterized the composition of umbilical cord blood mononuclear cells from term infants born to mothers with COVID-19 during the third trimester and found that even in the absence of vertical transmission, COVID-19 infection modulates the fetal immune system.3

The impact of maternal COVID-19 infection during each trimester of pregnancy has been of recent interest. One study demonstrated that third trimester infection was associated with a higher frequency of preterm birth compared to first or second trimester infection combined.4 Meanwhile, no differences have been found between trimesters in the rate of birth defects, which are overall similar to the national baseline estimates.4,5 Pregnant women have been found to mount pronounced inflammatory responses to COVID-19 infections compared to nonpregnant women, which appears to trigger fetal immune activation.6,7 COVID-19 infections during pregnancy have also been shown to cause placental inflammation.8,9 It is unclear, however, if in-utero exposure to COVID-19 results in sustained inflammation during gestation and alteration of immune system function in the early neonatal period.6 Additionally, the impact of a second-hit bacterial infection after COVID-19 exposure in-utero has not been assessed.

As a comparison to maternal COVID-19 exposure, chorioamnionitis represents an inflammatory in-utero environment prior to delivery. Chorioamnionitis most often develops from an ascending bacterial infection from the maternal genital tract. However, recent studies have demonstrated that chorioamnionitis can result from inflammation alone with sterile amniotic fluid samples at the time of clinical chorioamnionitis diagnosis.10 Prior to delivery, chorioamnionitis is suspected in the setting of a maternal fever and one or more of the following criteria: maternal leukocytosis, purulent cervical drainage or fetal tachycardia.11 The diagnosis can be confirmed antenatally with positive amniotic fluid gram stain, glucose level or culture. These tests are rarely performed clinically, however, due to limited time, procedural risks, and patient preference.11 The gold-standard diagnosis of chorioamnionitis is by histopathological examination of the placenta, but this often takes days to weeks to result and cannot be used to make diagnostic decision in real time. Intra-amniotic infections are associated with premature pre-labor rupture of membranes, preterm labor,10 and adverse neonatal outcomes, including sepsis, seizures, hypoxic ischemic encephalopathy, cerebral palsy, and respiratory distress syndrome.12,13,14 Potential long-term effects of early inflammatory exposure remain unclear but may contribute to fetal neurodevelopmental disorders, including autism spectrum disorders, cognitive delay and schizophrenia.15,16,17 Additionally, chorioamnionitis exposure has been shown to alter neonatal immune responses, which may contribute to these morbidities.18,19,20

Neonates are at high risk for bacterial infections, which result in significant morbidity and mortality. Escherichia coli (E. coli), Streptococcus agalactiae (S. agalactiae) and Staphylococcus epidermidis (S. epidermidis) are all neonatal pathogens that can result in sepsis, pneumonia, meningitis, bacteremia, and/or skin and soft tissue infections.21,22,23 This study aimed to evaluate the impact of a clinically relevant bacterial insult19,24 on the neonatal immune system after in-utero exposure to COVID-19 or chorioamnionitis. We chose to use umbilical cord blood mononuclear cells (CBMC) as they have been extensively characterized and demonstrate altered phenotype and function following perinatal inflammation.19,25,26 CBMCs were stimulated with heat-killed E. coli, S. agalactiae, or S. epidermidis and levels of the cytokines and chemokines interleukin-6 (IL-6), interleukin-1 beta (IL-1β), interleukin-8 (IL-8), interleukin-10 (IL-10), interferon gamma inducible protein-10 (IP-10), and monocyte chemoattractant protein-1 (MCP-1) were measured using a multiplexed immunoassay. Flow cytometry was then used to evaluate for cytokine expression in individual immune cell subsets based on in utero inflammatory exposure. Several studies have demonstrated altered fetal/neonatal immune responses following exposure to in utero inflammation.19,25,27 We therefore hypothesized that bacterial stimulation after in utero COVID-19 exposure would result in an upregulation of proinflammatory cytokine levels over unexposed controls, similar to what is seen in chorioamnionitis-exposed neonates.

Methods

Human subjects

The study was approved by the University of Iowa Institutional Review Board. Written informed consent was obtained from 75 birth parents of infants ≥37 weeks gestational age at birth. Participants included healthy patients with positive coronavirus disease (COVID-19) during pregnancy diagnosed either by an antigen or PCR test or diagnosed with chorioamnionitis based on placental histopathology as previously described.9 None of the chorioamnionitis-exposed neonates had funisitis present. None of the patients with COVID-19 during pregnancy were hospitalized during their COVID-19 infection, and timing of COVID-19 infection ranged from the first to the third trimester. Healthy unexposed participants were also included as controls. Samples were collected from June 2022 through November 2022. A total of 23 unexposed, 9 chorioamnionitis-exposed, 36 COVID-19 exposed, and 7 chorioamnionitis and COVID-19 exposed subjects were included in the analysis (75 total subjects).

Isolation of umbilical cord blood mononuclear cells (CBMCs)

Blood was obtained from either the umbilical vein or artery following delivery of the placenta and was kept at 4 °C in non-heparinized syringes until processing within 24 h. Samples were diluted 1:2 with sterile 0.9% saline (Thermo Fisher Scientific, Waltham, MA) and were layered onto Ficoll-Isopaque (Thermo Fisher Scientific, Waltham, MA). The samples were centrifuged at 400 g for 30 min at room temperature to isolate the mononuclear cells. The mononuclear cells were washed once in Rowell Park Memorial Institute (RPMI) 1640 (Gibco, Waltham, MA). The cells were placed in autologous serum with 10% dimethyl sulfoxide (Thermo Fisher Scientific, Waltham, MA) and underwent controlled freezing at −1 °C per minute with a Mr. Frosty Freezing Container (Thermo Fisher Scientific, Waltham, MA). The samples were kept at −80 °C until use, which was less than 12 months from initial processing.

Heat-killed bacterial preparation

Overnight cultures were created of three bacterial species in their appropriate media (Escherichia coli (E. coli, ATCC 700973, Manassas, VA) in Tryptic Soy Broth (Becton Dickinson, Franklin Lakes, NJ), Streptococcus agalactiae (S. agalactiae, ATCC 13813, Manassas, VA) in Brain-Heart Infusion (Becton Dickinson, Franklin Lakes, NJ), and Staphylococcus epidermidis (S. epidermidis, ATCC 12228, Manassas, VA) in Nutrient Broth (Becton Dickinson, Franklin Lakes, NJ)) and incubated at 37 °C for twelve to sixteen hours. Subcultures were then created at dilutions of 1:10, 1:20, 1:50 and 1:100 in the respective media listed. After incubation for the appropriate time, the optical density (OD) 600 was analyzed on a SpectraMax 384 Plus microplate reader (Molecular Devices, San Jose, CA). The OD600 was used to calculate colony forming units (CFU)/mL, which was used to produce 1 × 108 CFU/mL stocks. The S. epidermidis was put in a heat block for one hour at 60 °C, while the E. coli and S. agalactiae were put in a heat block for one hour at 70 °C. Following the hour-long incubation in the heat block, each stock was plated onto its respective agar plate to ensure effective killing. Stocks were then frozen at −80 °C until they were ready to be used.

CBMC cell culture and stimulation with heat-killed bacteria

The frozen CBMCs were thawed and washed once with RPMI with 10% heat-inactivated human AB- serum (Sigma Aldrich, Burlington, MA). The cells were resuspended in RPMI with 10% heat-inactivated human AB- serum, 400 mM/L L-glutamine (Thermo Fisher Scientific, Waltham, MA), 1% non-essential amnio acids (Thermo Fisher Scientific, Waltham, MA), 1% sodium pyruvate (Thermo Fisher Scientific, Waltham, MA) and 1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA). The cells were then seeded into polypropylene plastic 96-well flat-bottom plates (Corning Incorporated, Tewksbury, MA) at a concentration of 1 × 105 cells/200 μL media. Thawed heat-killed bacterial stocks E. coli, S. agalactiae, and S. epidermidis were added to the cells at a multiplicity of infection of 10:1. An additional 20 μL of complete RPMI was added to a control sample for each specimen. The cell culture plate was incubated at 37 °C with 5% CO2 for 24 h. The contents of the cell culture plates were transferred to 1.5 mL microcentrifuge tubes and centrifuged at 5000 RPM for 5 min at 22 °C to pellet the cells. The supernatant was transferred to new 0.5 mL microcentrifuge tubes and frozen at −80 °C until use, which was within 2 months of freezing.

Determination of cytokine and chemokine concentrations in CBMC supernatants

The levels of cytokines and chemokines in the CBMC supernatants were determined using Bio-Plex Multiplex Immunoassay System technology on a Bioplex 200 (Biorad, Hercules, CA), and repeat freeze-thaw cycles were avoided to minimize protein degradation. The following cytokines and chemokines were measured using a Bioplex Express Kit according to the manufacturer’s directions (Biorad, Hercules, CA): IL-1β, IP-10, MCP-1, IL-6, IL-8, and IL-10.20,28,29

CBMC cell culture for flow cytometry

Frozen CBMCs were thawed and washed once with RPMI with 10% heat-inactivated human AB- serum (Sigma Aldrich, Burlington, MA). The cells were resuspended in RPMI with 10% heat-inactivated human AB-serum, 400 mM/L L-glutamine (Thermo Fisher Scientific, Waltham, MA), 1% non-essential amnio acids (Thermo Fisher Scientific, Waltham, MA), 1% sodium pyruvate (Thermo Fisher Scientific, Waltham, MA), and 1% penicillin-streptomycin (Thermo Fisher Scientific, Waltham, MA). The cells were then seeded into polypropylene plastic 96-well flat-bottom plates at a concentration of 1 × 105 cells/200 μL media. Thawed heat-killed bacterial stocks of E.coli, S. agalactiae, and S. epidermidis were added to the cells at a multiplicity of infection of 10:1. An additional 20 μL of complete RPMI was added to a control sample for each specimen. The cell culture plate was incubated at 37 °C with 5% CO2 for 2 h. 1x monensin (BioLegend, San Diego, CA) diluted in complete RPMI was added to each well. The cell culture plate was incubated at 37 °C with 5% CO2 for 4 h. The contents of the cell culture plate were transferred to a polypropylene plastic 96-well V-bottom plate and centrifuged at 5000 RPM for 5 min at 22 °C to pellet the cells. The pelleted cell cultures were resuspended in phosphate-buffered saline (Thermo Fisher Scientific, Waltham, MA) with 0.002 M ethylenediaminetetraacetic acid (Thermo Fisher Scientific, Waltham, MA) and 1% fetal calf serum (Millipore Sigma, Burlington, MA) (flow buffer). Cells were stained for viability using Zombie NIR Fixable Viability at a dilution of 1:1000 (BioLegend, San Diego, CA, Cat# 423106). Cells were washed twice with flow buffer. Cells were treated with Human TruStain FcX (BioLegend, San Diego, CA, Cat# 422301) at a dilution of 1:200. Cells were labeled for CD3 (clone UCHT1, APC/Fire 50, BioLegend, San Diego, CA, Cat# 300470), CD4 (clone OKT4, PE/Cy7, BioLegend, San Diego, CA, Cat# 317414), CD8 (clone SK1, Pacific Blue, BioLegend, San Diego, CA, Cat# 344718), CD19 (clone HIB19, PE, BioLegend, San Diego, CA, Cat# 302208), CD56 (clone HCD56, PE/Cy5, BioLegend, San Diego, CA, Cat# 318308), CD14 (clone M5E2, Brilliant Violet 785, BioLegend, San Diego, CA, Cat# 301840), CD15 (clone W6D3, Brilliant Violet 605, BioLegend, San Diego, CA, Cat# 323032), CD16 (clone 3G8, Alexa Fluor 700, BioLegend, San Diego, CA, Cat# 302026), and HLA-DR (clone LN3, Alexa Fluor 488, BioLegend, San Diego, CA, Cat# 327010) at a dilution of 1:200. The cells were washed twice with flow buffer. The cells were fixed using Intracellular Fixation Buffer (Thermo Fisher Scientific, Waltham, MA). The cells were permeabilized using 1x permeabilization buffer (Thermo Fisher Scientific, Waltham, MA). The cells were stained for intracellular cytokines IL-1β (clone JK1B-1, Alexa Fluor 647, BioLegend, San Diego, CA, Cat# 508208), IL-6 (clone MQ2-13A5, PerCP/Cy5.5, BioLegend, San Diego, CA, Cat# 501118), and IL-8 (clone BH0814, PerCP, BioLegend, San Diego, CA, Cat# 514606). Flow cytometry was performed using a Cytek Aurora 3-laser configuration (Cytek, Bethesda, MD). The gating strategy is shown in Supplementary Fig 1 and gating was performed as previously published.30 FlowJo version 10.8.2 was used for gating and data visualization.

Statistical analysis

Cytokine and chemokine concentrations were calculated using Bio-Plex Multiplex Immunoassay System and expressed in pg/mL. The cytokine values were log-transformed as they were non-normally distributed and were highly skewed. Statistical differences between measured values in stimulated and unstimulated conditions were analyzed using Brown-Forsythe and Welch ANOVA test. For the flow cytometry data, differences between groups were assessed with Welch ANOVA testing for normally distributed data and Welch ANOVA testing following log transformation of the data for non-normally distributed data. P values less than 0.05 were considered statistically significant. Statistical tests were performed using Prism 9 (Graphpad Prism, LLC, San Diego, CA).

Results

Baseline maternal and neonatal characteristics were compared between exposure groups. Differences between groups included higher maternal white blood cell count upon admission to labor and delivery in both the maternal COVID-19 and chorioamnionitis-exposed groups, and an increased likelihood of having a neonatal sepsis evaluation due to clinical illness in the 24 h after birth in the COVID-19 and chorioamnionitis-exposed groups (Table 1). Unsurprisingly, the chorioamnionitis-exposed group had a higher incidence of prolonged rupture of membranes and maternal fever (Table 1).

Table 1 Study group characteristics.
Full size table

Cytokine profiles following chorioamnionitis or COVID-19 exposure

Unstimulated chorioamnionitis-exposed CBMCs demonstrated decreased IL-10 expression compared to unexposed controls (Fig. 1a). More pronounced differences were noted following heat-killed bacterial stimulation, including increased IL-1b levels in all of the stimulation conditions (Fig. 1b–d) and increased IL-6 following heat-killed E. coli stimulation in the chorioamnionitis-exposed samples (Fig. 1b). No differences were noted in levels of IP-10 or MCP-1 between exposure groups. COVID-19 exposed neonatal CBMCs demonstrated several baseline differences in cytokine expression, including increased IL-1b, IL-6 and IL-8 and decreased IL-10 compared to unexposed controls (Fig. 2a). Similar findings were observed following stimulation with heat-killed bacteria, with COVID-19 exposed samples demonstrating elevated levels of IL-1b in all of the stimulation conditions, increased IL-6 following heat-killed E. coli stimulation and elevated IL-8 following stimulation with heat-killed E. coli or S. epidermidis (Fig. 2b–d). Levels of IP-10 and MCP-1 were not different between COVID-19 exposed and unexposed groups.

Fig. 1: Chorioamnionitis-exposed CBCMs have elevated IL-1β and IL-6 levels after heat-killed bacterial stimulation compared to unexposed CBMCs.
figure 1

Cord blood mononuclear cells were isolated from umbilical cord blood after delivery. CBMCs were then cultured and stimulated with heat-killed E. coli, S. agalactiae, and S. epidermidis at an MOI of 10:1. Cytokine levels were evaluated using multiplexed protein assay for unexposed CBMCs and chorioamnionitis-exposed CBMCs. ● = unexposed CMBC (n = 23), ○ = chorioamnionitis-exposed CBMC (n = 9). a IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels with no stimulation, (b) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after E. coli stimulation, (c) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after S. agalactiae stimulation, (d) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after S. epidermidis stimulation. Bars represent the mean cytokine level (pg/mL) and error bars represent SEM with individual data points shown. *p < 0.05, **p < 0.01. Data was analyzed using Welch ANOVA test following log transformation of the data.

Full size image
Fig. 2: COVID-19 exposed CBMCs demonstrate elevated IL-1β, IL-6, and IL-8 levels after heat-killed bacterial stimulation compared to unexposed CBMCs.
figure 2

Cord blood mononuclear cells were isolated from umbilical cord blood after delivery. CBMCs were then cultured and stimulated with heat-killed E. coli, S. agalactiae, and S. epidermidis. Cytokine levels were evaluated using multiplexed protein assay for unexposed CBMCs and COVID-19 exposed CBMCs in each trimester of pregnancy. = unexposed CMBCs (n = 23), = COVID-19 exposed CBMCs in any trimester (n = 36). a IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after no stimulation, (b) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after E. coli stimulation, (c) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after S. agalactiae stimulation, (d) IL-1β, IL-6, IL-8, IL-10, IP-10, and MCP-1 levels after S. epidermidis stimulation. Bars represent the mean cytokine level (pg/mL) and error bars represent SEM with individual data points shown. *p < 0.05, **p < 0.01. Data was analyzed using Welch ANOVA test following log transformation of the data.

Full size image

Direct cytokine profile comparison between chorioamnionitis and gestational COVID-19 exposure

Given that exposure to chorioamnionitis and COVID-19 resulted in similar, although distinct, cytokine profiles following pathogen stimulation, we next directly compared the cytokine profiles between these exposure groups. We included an additional exposure groups as well, which included exposure to both gestational COVID-19 and chorioamnionitis to evaluate if multiple exposures to perinatal inflammation further altered neonatal immune responses. There were no differences in baseline cytokine expression between exposure groups (Fig. 3a). There were also no differences in cytokine expression following heat-killed bacterial stimulation between CBMCs exposed to chorioamnionitis, gestational COVID-19 or both chorioamnionitis and gestational COVID-19 (Fig. 3b–d).

Fig. 3: There are no differences in cytokine profiles between chorioamnionitis and COVID-19 exposed CBMCs.
figure 3

Cord blood mononuclear cells were isolated from umbilical cord blood after delivery. CBMCs were then cultured and stimulated with heat-killed E. coli, S. agalactiae, and S. epidermidis. Cytokine levels were evaluated using multiplexed protein assay for chorioamnionitis-exposed CBMCs, COVID-19 exposed CBMCs in any trimester of pregnancy, and both chorioamnionitis-exposed and COVID-19 exposed CBMCs = chorioamnionitis-exposed CBMCs (n = 9), = COVID-19 exposed CBMCs in any trimester (n = 36), = chorioamnionitis and COVID-19 exposed CBMCs in any trimester (n = 7). a IL-1β, IL-6, IL-8, IL-10, IP-10, MCP-1 levels after o stimulation, (b) IL-1β, IL-6, IL-8, IL-10, IP-10, MCP-1 levels after E. coli stimulation, (c) IL-1β, IL-6, IL-8, IL-10, IP-10, MCP-1 levels after S. agalactiae stimulation, (d) IL-1β, IL-6, IL-8, IL-10, IP-10, MCP-1 levels after S. epidermidis stimulation. Bars represent the mean cytokine level (pg/mL) and error bars represent SEM with individual data points shown. Data was analyzed using the Brown-Forsythe and Welch ANOVA test following log transformation of the data.

Full size image

Comparison of cytokine production based on immune cell type

Flow cytometry was next employed to determine the source(s) of differential cytokine expression between exposure groups. Chorioamnionitis-exposed samples demonstrated no alterations in baseline cell populations, but did have increased IL-1b in CD4 + T cells following stimulation (heat-killed S. epidermidis), increased IL-6 in baseline CD4 + T cells and following stimulation (heat-killed S. epidermidis) and increased IL-6 in CD16hi monocytes following stimulation (heat-killed E. coli) compared to unexposed samples (Table 2). Chorioamnionitis-exposed samples also demonstrated decreased IL-6 in dendritic cells and decreased IL-8 in CD8 + T cells following stimulation (heat-killed S. agalactiae) compared to unexposed samples (Table 2). COVID-19 exposed samples demonstrated decreased numbers of CD16hi NK cells but increased numbers of CD16lo monocytes compared to the other exposure groups (Table 2). COVID-19 exposed samples also demonstrated increased IL-1b in dendritic cells following stimulation (heat-killed E. coli and S. agalactiae), increased IL-6 in unstimulated CD16lo monocytes and following stimulation (heat-killed E. coli and S. epidermidis), increased IL-8 in CD16hi NK cells following stimulation (heat-killed E. coli), increased IL-8 in dendritic cells following stimulation (heat-killed S. epidermidis) and increased IL-8 in baseline CD16lo monocytes and following stimulation (heat-killed E. coli) compared to unexposed and chorioamnionitis-exposed samples (Table 2). It is interesting that differences in cell-level cytokine expression were noted between the chorioamnionitis-exposed and COVID-19 exposed groups despite the lack of difference detected between these groups in total cytokine expression (Fig. 3).

Table 2 Comparison of cell proportions and intracellular cytokine expression measured by flow cytometry between exposure groups.
Full size table

Discussion

This study presents a comprehensive analysis of the cytokine profiles of term neonatal CBMCs following exposure to different sources of perinatal inflammation, including chorioamnionitis and gestational COVID-19. We chose to stimulate CBMCs with heat-killed neonatal-specific pathogens, including E. coli, S. agalactiae and S. epidermidis, as these are the leading causes of both early and late onset neonatal sepsis and are the most likely bacterial species a neonate will encounter after birth. Exposure to in-utero inflammation, including chorioamnionitis or gestational COVID-19, led to decreased baseline IL-10 expression and variably increased IL-1β, IL-6, and IL-8 expression following stimulation with a secondary pathogen. These cytokine changes were found to be immune-cell specific, with cytokine responses varying based on immune subset studied. This suggests that the fetal/neonatal response to in utero inflammation differs based on immune cell type, but ultimately results in a globally enhanced pro-inflammatory response, regardless of the etiology of perinatal inflammation.

Subjects in this study were well matched based on most maternal and neonatal characteristics. However, it was interesting that maternal WBC upon admission to labor and delivery was higher in the gestational COVID-19 group compared to unexposed controls, despite the fact that many of the cases of COVID-19 occurred remote from delivery. This suggests some degree of ongoing maternal inflammation following gestational COVID-19 despite the absence of an active infection. This is in line with several studies demonstrating ongoing inflammation in a subset of patients with post-acute sequelae of SARS-CoV-2, also known as long COVID.31,32 Neonates exposed to gestational COVID-19 were also more likely than unexposed neonates to undergo an evaluation for neonatal sepsis for abnormal clinical signs or symptoms in the 24 h following delivery. This suggests that exposure to in utero COVID-19 has long-lasting effects on the fetus/neonate, even if the exposure was far from delivery. This is consistent with previous studies showing that maternal upper respiratory tract infections during pregnancy can adversely impact the health and neurodevelopment of the offspring.33 Overall, these differences between exposure groups suggests that gestational COVID-19 exposure has effects on both the mother and fetus that persist long after the acute infection has resolved.

The fetal inflammatory response syndrome (FIRS) is defined as elevated levels of IL-6 in amniotic fluid or umbilical cord blood following exposure to intraamniotic infection or inflammation (i.e., chorioamnionitis).34 Our findings of elevated IL-1ß and IL-6 in chorioamnionitis-exposed subjects and elevated IL-1ß, IL-6 and IL-8 in COVID-19 exposed subjects is in line with the concept of FIRS and other studies demonstrating elevated levels of pro-inflammatory cytokines in neonates exposed to chorioamnionitis or gestational COVID-19.20,35,36,37,38 Our findings do contrast other studies that demonstrate a hyporesponsive phenotype of chorioamnionitis-exposed neonatal monocytes.19,26,39 This discrepancy is likely related to differences in the cell type studied (purified monocytes versus CBMCs), the type of stimulation used (toll-like receptor agonists versus bacteria), the difference in gestational ages between the studies (preterm versus term), a difference in severity of fetal response to maternal inflammation (funisitis versus no funisitis) or the severity and length of exposure to the perinatal inflammation. Our study further drilled down to uncover the cellular source of these cytokine differences using intracellular flow cytometry. Interestingly, maternal inflammation had varying effects on different immune cell subsets, with some subsets demonstrating enhanced pro-inflammatory cytokine expression (dendritic cells and CD16lo monocytes following COVID-19 exposure) but others demonstrating decreased cytokine expression (CD8 + T cells following either chorioamnionitis or COVID-19 exposure). Gestational COVID-19 exposure also changed the proportion of baseline CBMC immune populations, resulting in decreased CD16hi NK cells but increased CD16lo monocytes. These findings are in line with a recent publication demonstrating that CBMCs exposed to gestational COVID-19 had increased NK cell exhaustion with more activated monocytes and dendritic cells.3 There were also cell-specific differences in cytokine expression between chorioamnionitis-exposed and COVID-19 exposed subjects that were masked when global cytokine expression was evaluated rather than cell-specific expression. This suggests that the fetal/neonatal response to maternal inflammation differs based on the cause and overall strength of the maternal inflammatory response and differentially impacts immune cell subsets. These findings likely explain some of the conflicting studies regarding the impact of maternal inflammation on neonatal immune responses, as the results are likely to vary based on the cell population studied and the type of assay used to measure immune responses.

An additional noteworthy finding from our study was decreased IL-10 expression in unstimulated chorioamnionitis-exposed and COVID-19 exposed subjects with baseline elevations of the pro-inflammatory cytokines IL-1ß, IL-6 and IL-8 in COVID-19 exposed subjects. This suggests that exposure to in utero inflammation alters the baseline status of fetal/neonatal immune cells, resulting in a more pro-inflammatory phenotype regardless of pathogen exposure. This could explain some of the inflammation-associated morbidities experienced by chorioamnionitis-exposed neonates, including increased rates of bronchopulmonary dysplasia and an increased incidence of childhood atopic diseases.40,41 Although less well studied, gestational COVID-19 exposure has also been associated with inflammation-associated morbidities during early life, including neonatal respiratory distress and an increased likelihood of being admitted to the NICU.6,42 It is also worth noting that both chorioamnionitis and gestational COVID-19 exposed subjects demonstrated an upregulation of pro-inflammatory cytokines that was not pathogen-specific. This suggests that fetal/neonatal exposure to maternal inflammation will result in dysregulated immune responses to a broad variety of stimuli, with a heightened inflammatory profile that is likely to have detrimental effects. The similar global cytokine profiles seen in chorioamnionitis and COVID-19 exposed subjects also suggests that the fetal/neonatal response to maternal inflammation shares common underlying mechanisms, regardless of the etiology of the inflammation.

The study has several strengths, including the use of a well-characterized cohort of neonates and an investigation into the impact of multiple inflammatory events on the neonatal immune system. However, we recognize that our study also has several limitations. We acknowledge that the sample size is relatively small, making it difficult to identify small differences between exposure groups. The majority of the mothers with gestational COVID-19 were vaccinated against COVID-19 prior to the infection, which likely resulted in a less severe infection and subsequent inflammatory response. Additionally, no subjects included in this study were born to women who required hospitalization for severe COVID-19 infection, so our findings may not be representative of neonates born to severely ill mothers. However, we did observe differences in maternal WBC around time of delivery with increased rates of neonatal sepsis evaluations in the gestational COVID-19 group, suggesting a significant and long-lasting effect of the COVID-19 infection on both the mother and infant. Finally, the nature of human pregnancy and delivery involves many variables that cannot be completely controlled for, such as chronic or acute maternal illness, labor complications, and genetic susceptibility. Despite these limitations, we believe our results are a much needed first step in understanding the impact of gestational COVID-19 exposure on postnatal immune responses and subsequent susceptibility to infection.

Overall, the findings of this study suggest that fetal/neonatal exposure to maternal inflammation leads to upregulated pro-inflammatory responses in the immediate neonatal period. Our study highlights the importance of investigating the impact of multiple inflammatory events on the neonatal immune system and may have implications for the prevention and treatment of neonatal infections following exposure to in utero inflammation. Further studies are needed to investigate the persistence of these effects outside of the immediate neonatal period, and to better understand what role they play in neonatal and infant susceptibility to infection.