key: cord-0825871-e9s3auim
authors: Gee, Sarah; Chandiramani, Manju; Seow, Jeffrey; Modestini, Carlotta; Das, Abhishek; Doores, Katie J; Tribe, Rachel M; Gibbons, Deena L
title: The legacy of maternal SARS-CoV-2 infection on the immunology of the neonate
date: 2021-05-28
journal: bioRxiv
DOI: 10.1101/2021.05.27.446014
sha: d7bf02716341fde9a1f5985402f496b665319078
doc_id: 825871
cord_uid: e9s3auim

Despite extensive and ongoing studies of SARS-CoV-2 and evidence that pregnant women are at increased risk of severe COVID-19, the effect of maternal infection on the developing infant remains unclear. To determine the potential impact of exposure to SARS-CoV-2 in utero on the neonate, we have assessed the immunological status of infants born to mothers with confirmed SARS-CoV-2 infection during gestation. No evidence of vertical transmission of SARS-CoV-2 was observed, but transfer of maternal SARS-CoV-2 specific IgG to infants was apparent, although to a lesser extent in cases of active or recent maternal infection. Infants born to mothers with recent/ongoing infection had elevated circulating pro-inflammatory cytokines and enhanced percentages of innate immune cells compared to that seen in infants born to uninfected mothers. In tandem, higher frequencies of FOXP3+ regulatory T cells and circulating IL-10 demonstrated a further nuance to the neonatal effector response. Interestingly, cytokine functionality was enhanced in infants born to mothers exposed to SARS-CoV-2 at any time during pregnancy. This indicates that maternal SARS-CoV-2 infection influences in utero priming of the fetal immune system.

Despite the ongoing COVID-19 pandemic, the effect of maternal SARS-CoV-2 infection on the 43 immunology of the developing infant is still unclear. Indeed, SARS-CoV-2 infection in pregnancy has 44 been reported to lead to variable outcomes for the mother. The majority of infected women in 45 pregnancy are asymptomatic or only experience mild symptoms 1 2 . Nevertheless, some pregnant 46 women with SARS-CoV-2, especially in the third trimester, appear to be at an increased risk for 47 hospitalization and subsequent intensive care unit admission 3 4 5 and rates of maternal infection 48 increased in the second wave 6 7 . Although the rates of preterm birth did not appear elevated in 49 initial reports, more recent data suggest pregnant women are at a higher risk for subsequent 50 preterm birth albeit much may be due to clinical intervention 8 . 51 52 With respect to the infant, a national UK surveillance study suggested neonatal SARS-CoV-2 infection 53 is uncommon even in babies born to mothers with perinatal infection 9 . Similarly, a study from the 54 US of 116 mothers with confirmed perinatal SARS-CoV-2 infection did not identify any neonatal 55 cases 10 and a systematic review demonstrated no vertical transmission 11 . Set against this, there 56

have been a small number of individual case reports documenting evidence of vertical transmission 12 57 3 60 Whilst vertical transmission of SARS-CoV-2 itself is rare, the potential immunological perturbations 61 induced in the pregnant mother 19,20 may conceivably leave an immunological legacy on the newborn 62 with far reaching consequences. Indeed, our recent study in a preterm baby group found evidence of 63 perinatal inflammation modulating the developing immune system of the infant 21 . It is already 64 appreciated that the immune system of the unborn child can be altered by the presence of human 65 immunodeficiency virus (HIV) or Hepatitis C virus (HCV) in mothers, either with or without vertical 66 transmission 22 23 24 25 ; that metabolites derived from maternal gut microbiota can shape the immune 67 system of the offspring 26 ; and that modulation of the neonatal immune system has been associated 68 with diseases in later life 27 28 . In the limited studies that have assessed the immune status of babies 69 born to SARS-CoV-2 infected mothers, there has been little evidence of impact in the cellular and 70 humoral immunity of the neonate 29 . Similarly, a small study of SARS-CoV-2 infection during 71 pregnancy has been associated with a cytokine response in the fetal circulation (i.e. umbilical cord 72 blood) but with no effect on the cellular immune repertoire 20 . However, to our knowledge, none of 73 the studies have included a comprehensive analysis of the infant cellular immune profile of babies 74 born to SARS-CoV-2 exposed mothers (at any point in their pregnancy) compared to profiles of 75 infants born to unexposed mothers. Indeed, it is increasingly challenging to include an appropriate 76 control group in a global pandemic. Moreover, simultaneous analysis of cytokines and antibody 77 titres in infants and their paired mothers is lacking. In our study, we have assessed the immunologic 78 status of infants born to mothers with SARS-CoV-2 that tested positive either during the two weeks 79 directly prior to birth or earlier on in pregnancy, compared to babies born to mothers never exposed 80 to SARS-CoV-2 to identify if there is a legacy of maternal infection and potential in utero priming of 81 the neonatal immune system. Table 1 , and the group description is shown in Fig.  87 1a. Our neonatal group of infants born to mothers with SARS-CoV-2 exposure (SARS-CoV-2 exposed, 88 SE, n=30) was divided into those with recent or ongoing infection as determined by mothers 89 confirmed positive for SARS-CoV-2 by PCR within 2 weeks of birth (R/O, n=16, median positive swab 90 3 days before birth), or prior to 2 weeks; termed the recovered group (R, n=14, median positive 91 swab 48.5 days prior to birth). 92 93 4

In both the recent/ongoing and recovered group, regardless of the time between the positive swab 94 and birth or maternal IgM levels, SARS-CoV-2 specific IgM was not detected in cord plasma, 95 suggestive of a lack of vertical transmission (Fig. 1b) . This was true for IgM directed against the spike 96 protein (S), the receptor binding domain within the spike protein (RBD) as well as the nucleoprotein 97 (N) for which background IgM reactivity has been shown to be higher 30 . By contrast, SARS-CoV-2 98 specific IgG against all three antigens was detected in infants born to SARS-CoV-2 exposed mothers 99 (Fig. 1b) . Whilst levels of SARS-CoV-2 specific IgG in the mother-infant dyad were comparable in the 100 recovered group, there were significantly lower levels of SARS-CoV-2 specific IgG in infants born to 101 mothers with recent/ongoing infection (Extended Data Fig. 1a) . Thus, when the ratio of infant Ig to 102 their paired maternal Ig (transfer ratio) was calculated for each antigen, the mean transfer ratio of 103 all 3 antigens was 1.04 in the R group and 0.79 in the R/O group (Fig. 1c) . This was despite the 104 presence of high levels of maternal IgG in at least some mothers in the R/O group (Extended Data 105 Fig. 1a) . Indeed, when only comparing IgG levels in infants born to seropositive mothers, transfer of 106 SARS-CoV-2 specific IgG to the infant was still significantly lower in the R/O group (Fig. 1d ) and this 107 did not appear to differ with the sex of the infant (Extended Data Fig. 1b) . 108 109 Elevated plasma cytokines in mothers with recent/ongoing infection and their infants 110 SARS-CoV-2 infection is known to be associated with marked elevation of several plasma cytokines 111

including Interferon gamma-induced protein 10 (IP-10), interleukin (IL)-1β, CXCL8, IL-6 and IL-10 31,32 112 33 34 . To assess the impact of this on the neonate, plasma cytokine concentrations were assessed in 113 paired maternal and cord blood using a multiplex assay. IP-10 and IL-1β levels in the plasma of 114 mothers with recent/ongoing SARS-CoV-2 infection were significantly elevated when compared to 115 that of recovered mothers whilst IL-10, CXCL8 and IL-6 were similar between the SE groups ( Fig. 2a) . 116

When assessing neonatal cytokine levels, IL-10 was significantly elevated in the cord plasma from 117 babies born to mothers with recent/ongoing infection compared to those born to recovered 118 mothers. CXCL8 levels were also numerically higher in the recent/ongoing group although this did 119 not reach significance, largely driven by three infants with undetectably levels (Fig. 2b) . However, 120 concentrations of this chemokine was significantly higher in infants than in their paired mothers (Fig.  121 2c) which was not seen with any of the other cytokines tested. Conversely, infant IP-10 was 122 significantly lower than their paired mothers (Extended Data Fig. 2a) . The majority of babies born to 123 recovered mothers that showed elevated levels of CXCL8 were born by vaginal delivery (Fig. 2d) , 124 known to elevate several cytokines 35 . However, there were still notable increases of CXCL8 in babies 125 born via caesarean section (CS) in the recent/ongoing group, compared to the recovered group, 126 suggesting this was indeed a fetal response to maternal SARS-CoV-2 infection. 127 5 Spearman rank test identified significant (p<0.05) correlations in cytokine levels both within and 128 between maternal and cord blood in both groups. Nonetheless, this was more evident in mothers 129 and infants in the R/O group compared with the recovered group suggestive of a greater degree of 130 immune co-regulation in recent/ongoing SARS-CoV-2 infection (Fig. 2e) : there was a correlation 131 between CXCL8 and IP-10 in infants born to mothers with recent/ongoing infection (Fig. 2f) ; 132 maternal IL-1β levels correlated with CXCL8 and IP-10 in the infants from this group (Fig. 2g-h) , and 133 interestingly, maternal IL-1β levels negatively correlated with days between a positive COVID swab 134 and birth (Fig. 2i) the combined (R and R/O) SARS-CoV-2 exposed group (SE; n=30) were compared to those from term 145 infants born to healthy mothers collected prior to the pandemic (Non SARS-CoV-2 exposed, NSE; 146 n=15) but measured contemporaneously. We performed tSNE dimensionality reduction (Fig. 3a) on 147 91 individual flow cytometry immune parameters and observed that infant immune profiles in the SE 148 group clustered away from the normal immune profiles of the NSE group. Correlations between 149 immune parameters appeared significantly different in infants born to SE mothers compared to NSE 150 mothers ( Fig. 3b-c) . For example, a weaker correlation was seen between IFN-γ producing cells and a 151 stronger correlation occurred between proliferating TEMRAs (Extended Data Fig. 4a-b) . Indeed, 152 when focusing on the immune parameters that drive the biggest differences, we identified that 153 infants born to mothers in the R/O group separated furthest away from the NSE group, using 3-154 dimensonal PCA (Fig. 3d) , and that all three groups tended to segregate based on their maternal 155 SARS-CoV-2 status upon unbiased hierarchal clustering analysis (Fig. 3e) parameters, including those known to be perturbed in the adult response to SARS-CoV-2, were not 160 different between groups. Infant T cell lymphopenia was not observed and moreover the relative 161 6 frequencies of major adaptive lymphocyte subsets (e.g. CD4 and CD8 αβ T cells, γδ T cells and B cells) 162 were unaffected by maternal exposure to SARS-CoV-2 (Fig. 3f) . Despite preserved composition of cell 163 frequencies, changes in adaptive cell populations were still evident, including an increase in the 164 percentage of CD161 expressing CD8 T cells in babies born to mothers with recent/ongoing infection 165 ( Fig. 3g ) as well as increased CD25 + FOXP3 + T REGS (Fig. 3h) , the latter of which positively correlated 166 with the percentage of Vδ2 T cells (Fig. 3i) . Although we did not directly assess neonates for the presence of SARS-CoV-2, we did assess SARS-212

CoV-2 specific IgM levels and could find no evidence of vertical transmission in any of the 30 infants 213 born to SARS-CoV-2 exposed mothers. SARS-CoV-2 specific IgG was, however, transferred to the 214 neonates from their mothers suggestive of the transfer of protective immunity. There was a 215 correlation between maternal and infant SARS-CoV-2 IgG levels in mother-infant dyads in both 216 groups as previously suggested 44 . However, even though for many pathogens, umbilical cord titers 217 of IgG at normal term delivery are higher than in maternal blood 45 46 47 48 , there were reduced levels 218 of SARS-CoV-2 specific IgG in infants born to mothers with recent/ongoing infection compared to 219 their paired mothers. This did not appear to be a threshold issue, as many mothers exhibited high 220 levels of SARS-CoV-2 specific IgG which was not transferred efficiently to their infant. Reduction of 221 SARS-CoV-2 specific Ig transfer via the placenta has been suggested to occur in the third trimester 222 due to altered glycosylation 49 and reduced maternal SARS-CoV-2-specific antibody titers and 223 impaired placental antibody transfer were also noted in pregnancies with a male fetus 50 , although 224 there did not appear to be any sex bias in this data set. It is currently unclear whether antibodies 225 We thank the mothers and their infants for blood collection and all the midwives at STH for sample 427 collection. We also thank Shraddha Kamdar and those staff involved with COVID-IP at King's College 428

London 32 for some sample processing and Thomas Lechmere for assistance with ELISAs. We thank 429

Iva Zlatareva for assistance with the cytokine multiplex analysis. We would also like to acknowledge 430

Evolve Biosystems for funding (RT and DG) towards the non SARS-CoV-2 exposed (NSE infants born to SARS-CoV-2 exposed mothers. a,b, Representative flow cytometry plots of % TNF-α + 694

Universal Screening for SARS-CoV-2 in

Women Admitted for Delivery

Clinical Characteristics of Pregnant Women with Covid-19 in Wuhan

Clinical course of severe and critical coronavirus disease 2019 466 in hospitalized pregnancies: a United States cohort study

Neonatal Early-Onset Infection With SARS-CoV-2 in a Newborn Presenting 495

Analysis of SARS-CoV-2 vertical transmission during pregnancy

Persistence of SARS-CoV-2 in the first trimester placenta leading to 499 transplacental transmission and fetal demise from an asymptomatic mother

Peripheral immunophenotypes in children with multisystem inflammatory 594 syndrome associated with SARS-CoV-2 infection

COVID-19: immunopathogenesis and Immunotherapeutics

Comprehensive mapping of immune perturbations associated with 598 severe COVID-19

Phenotypical and functional alteration of unconventional T cells in severe 600 COVID-19 patients

SARS-CoV-2 infection in infants under 1 year of age in Wuhan City

Reference intervals for lymphocyte subsets in preterm and term 604 neonates without immune defects

Interleukin-8 (CXCL8) production is a signatory T cell effector function of 606 human newborn infants

Differential patterns of methylation of the IFN-608 gamma promoter at CpG and non-CpG sites underlie differences in IFN-gamma gene expression 609 between human neonatal and adult CD45RO-T cells

Maternal exposure to polycyclic aromatic hydrocarbons and 5'-CpG 611 methylation of interferon-gamma in cord white blood cells

Phenotypic and Epigenetic Adaptations of Cord Blood CD4+ T Cells 614 to Maternal Obesity

The Impact of Milk and Its Components on Epigenetic Programming of

Immune Function in Early Life and Beyond: Implications for Allergy and Asthma

In utero priming of highly functional effector T cell responses to human 619 malaria

Trained immunity in newborn infants of HBV-infected mothers

Altered Memory T-Cell Responses to Bacillus Calmette-Guerin and 623

Tetanus Toxoid Vaccination and Altered Cytokine Responses to Polyclonal Stimulation in HIV-624

RHS: n=16) groups. e-h, Spearman correlation plots with 660 generalised linear model lines and 95% confidence intervals in the R/O group (e-g) (e: n=15, f-g: 661 n=14) and the SE group (h) (n=27). i, Infant CXCL8 levels in babies born via CS (LHS: n=18) or NVD 662 (RHS: n=12). Cytokine scatter plots central line = median value

01) were assessed by Kolmogorov-Smirnov tests (a,b,i) and two-sided paired Wilcoxon tests 664 (c)

| Recent/ongoing maternal SARS-CoV-2 infection influences some, but not all, adaptive 666 immune cell proportions in infant cord blood. a, tSNE dimensionality reduction of infant immune 667 profiles measured from cord blood mononuclear cells (CBMCs) in those born to SARS-CoV-2 Exposed 668 mothers (SE; n=28) and Non SARS-CoV-2 Exposed mothers (NSE; n=13) infants. b,c, Spearman 669 correlation matrix of significant (p<0.05) correlations of infant cellular populations in the NSE (n=15) 670 (b) the SE (n=30) (c) groups. d, 3D PCA dimensionality reduction of

Recovered group (R; n=13) and Non SARS-CoV-2 Exposed infant 672 groups (NSE; n=13)

Dendrogram cluster heatmap of 45 flow cytometry immune populations in infants within the NSE 674 (n=14), R (n=14), R/O (n=15) groups. f, Boxplots displaying the proportions of αβ T cells, CD4 + T cells

CD8 T cells and γδ T cells in the (NSE: n=15, R: n=14, R/O: n=16) groups, and proportion of B cells 676 (NSE: n=14, R: n=14, R/O: n=16). g-h, Boxplots displaying the proportions of CD161 + CD8 + T cells and

NSE: n=15, R: n=14, R/O: n=16). i, Spearman correlation plots in the SE group 678 (n=30) of infant proportions of FOXP3 + CD25 + T REG cells and Vδ2 γδ T cells with generalised linear 679 model lines and 95% confidence intervals

central line = median, upper line = 75 th percentile; lower line = 25 th line; whiskers = 1.5*75 th /25 th 681 percentile

| Elevated proportions of innate immune cells in cord blood from infants born to mothers 683 with recent/ongoing SARS-CoV-2 infection a, Representative flow cytometry plots of % CD56 + NK 684 cells (of CD3 -live cells) in an NSE and R/O infant, measured from cord blood mononuclear cells 685 (CBMCs), and the boxplot for the total infants (NSE: n=15), R: n=14), R/O: n=16). b-c, Boxplots 686 displaying the proportions of NKT cells and Vδ2 γδ T cells

Spearman correlation plots in the SE group t (n=30) of infant CXCL8 levels and the proportions of 688 NKT cells (d) or CD69 + NK cells (e) with generalised linear model lines and 95% confidence intervals

All boxplots follow standard Tukey representations; central line = median, upper line = 75 th 690 percentile; lower line = 25 th line; whiskers = 1.5*75 th /25 th percentile

001) were assessed by two-sided Wilcoxon rank-sum tests

Boxplots displaying the proportions of TNF-α + (a) and IFN-γ + (b) CD4 + T 696 cells

Spearman correlation 699 plots of infant CD4 + T EM cells/TNF-α + CD4 + T (c), CD4 + T EM cells /IFN-γ + CD4 + T cells (d) in the SE group 700 (n=30) with generalised linear model lines and 95% confidence intervals. All boxplots follow standard

Tukey representations; central line = median, upper line = 75 th percentile

*75 th /25 th percentile

001) were 703 assessed by two-sided Wilcoxon rank-sum tests

Table 1 | Maternal and infant clinical characteristics

Non-SARS-CoV-2 Exposed (NSE)