Investigating the impact of severe maternal SARS-CoV-2 infection on infant DNA methylation and neurodevelopment

doi:10.21203/rs.3.rs-4562282/v1

Download PDF

Article

Investigating the impact of severe maternal SARS-CoV-2 infection on infant DNA methylation and neurodevelopment

https://doi.org/10.21203/rs.3.rs-4562282/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Maternal infections during pregnancy can increase the risk to offspring of developing a neurodevelopmental disorder. Given the global prevalence and severity of infection with Severe Acute Respiratory Syndrome related Coronavirus 2 (SARS-CoV-2), the objective of this study was to determine if in utero exposure to severe maternal SARS-CoV-2 infection alters infant neurodevelopmental outcomes at 12 months and to identify potential biological markers of adverse infant outcomes. Mother-infant dyads exposed to severe SARS-CoV-2 infection (requiring hospitalization) during pregnancy and age and sociodemographic matched control dyads were recruited from Monash Medical Centre, Australia in 2021/22 and prospectively assessed over 12 months. Maternal serum cytokine levels and Edinburgh Postnatal Depression Scale (EPDS) scores were assessed at birth. DNA methylation was assessed from infant buccal swabs at birth (Illumina EPIC BeadChip). Infant neurodevelopmental outcomes at 12 months were assessed using the Ages and Stages Questionnaire (ASQ-3). Mothers exposed to severe SARS-CoV-2 exhibited elevated serum IL-6 and IL-17A and higher EPDS scores than controls at birth. Infants exposed to severe SARS-CoV-2 in utero demonstrated over 3000 significant differentially methylated sites within their genomes compared to non-exposed (adjusted p-value < 0.05), including genes highly relevant to ASD and synaptic pathways. At 12 months, severe SARS-CoV-2 exposed infants scored lower on the ASQ-3 than non-exposed infants and communication and problem-solving scores negatively correlated with maternal Il-6 levels at birth. DNA methylation changes therefore unveil potential mechanisms linking infection exposure to delayed neurodevelopment and maternal serum IL-6 levels may be a potential biomarker of child developmental delay.

Biological sciences/Neuroscience

Health sciences/Biomarkers/Predictive markers

Biological sciences/Molecular biology

Maternal exposure to viral and bacterial infections during pregnancy is a robustly verified risk factor for future development of neurodevelopmental disorders, including Autism Spectrum Disorder (ASD) ^1–3. For example, in a large-scale epidemiological study following the 1964 rubella pandemic, the incidence of ASD rose from less than 1–13% ². A population-based sample of Danish children born from 1980–2005 (n = 1,418,152) found a threefold (2.98 (CI:1.29–7.15)) increased risk for ASD when the mother was hospitalized for a viral infection during pregnancy ⁴. In a review of epidemiological studies associating infections such as influenza, toxoplasmosis, and genital/reproductive infections with neurodevelopmental disorders, an estimated 33% of ASD cases could be prevented if infection was eliminated from the pregnant population ¹. Understanding how this major risk factor transmutes into ASD is a matter of considerable importance. This is especially pertinent with the COVID-19 pandemic where long-term impacts of COVID-19 infection, especially during prenatal development, are poorly understood but recent studies have found increased risk of neurodevelopmental disorder in children exposed to SARS-CoV-2 in utero^{5, 6}.

Transplacental or vertical transmission of SARS-CoV-2 has been reported, but appears to be rare ⁷. Of considerable concern, however, is the maternal immune response to SARS-CoV-2 and the associated ‘cytokine storm’ ⁸. Concern over this immune response is borne of previous ecological studies, birth cohort studies and animal models that have established key links between the activation of pro-inflammatory pathways in the mother with adverse neurodevelopmental outcomes in the infant ^{2, 4, 9–12}. Indeed, mouse model studies have shown that blocking specific pro-inflammatory cytokines, such as interleukin-6 (IL-6) and Interleukin-17A (IL-17A), following maternal exposure to a viral mimetic can prevent the emergence of adverse behavioural outcomes in offpring^{13, 14}. This suggests altered maternal cytokines are key mediators of infection-induced adverse neurodevelopmental outcomes in offspring in the absence of vertical transmission. Studies in rodents have also shown that one potential mechanism for how maternal immune activation could alter offspring brain development is via DNA methylation changes ^15–17. The objective of this prospective cohort study was to determine if severe maternal SARS-CoV-2 infection influenced neurodevelopmental outcomes at 12 months in infants exposed to SARS-CoV-2 in utero, and whether this correlated with maternal cytokine levels assessed at birth. Secondarily, infant DNA methylation levels were examined at birth to identify potential mechanisms by which exposure to infection with SARS-CoV-2 may disrupt infant neurodevelopment.

Ethics approval and consent: The study was approved by the Monash Health Human Research Ethics Committee (HREC/69258/MonH-20-0000-801A) in accordance with the Australian National Health and Medical Research Council Act 1992 and the National Statement on Ethical Conduct in Human Research (2018). Adult participants provided signed consent prior to participation in the study for themselves and their child. The work described was carried out in accordance with the Code of Ethics of the World Medical Association Declaration of Helsinki – Ethical Principles for Medical Research Involving Human Subjects. The study abides by the STROBE guideline recommended for cohort studies under the EQUATOR Reporting Guidelines.

Participant recruitment:

The study design follows our previously published protocol ¹⁸. Briefly, pregnant women with a positive SARS-CoV-2 PCR or rapid antigen test and control women matched for age, sociodemographic factors and gestational age at birth were recruited from Monash Medical Centre in-patient records from 2021–2022. While recruitment for this prospective longitudinal cohort study is ongoing, the data set presented in this study represent a subset of women who were severely infected with SARS-CoV-2 during pregnancy, using the World Health Organisation (WHO) 7-point ordinal scale to grade severity of infection. Here, a score of 3 or above constitutes need for hospitalisation, while a score of 4 and above indicates oxygen therapy was required to varying degrees. As previous epidemiological findings suggest that severe infections requiring hospitalisation are associated with an increased incidence of neurodevelopmental disorders such as ASD^{3, 4} we specifically focused on this severe group in the current analysis. This group consisted of 16 women and their infants, with 11 women scoring 3 on the WHO scale, three women scoring 4, and two women scoring 5 (Table 1). Women with a history of substance dependence, family violence or psychiatric illness were excluded (1 participant from the control group). Participants attended Monash Medical Centre for birth (within 4 weeks of giving birth) and 3month assessments. Participants were contacted for follow-up assessments at 12 months via email or telephone in 2023. Follow-up assessments consisted of parent-completed questionnaires delivered online. Maternal demographic data and past medical and obstetric history were collected through face-to-face interview at the first birth assessment. All assessments were performed blind to patient grouping.

Table 1

Cohort characteristics
Cohort characteristics	Control group (N = 14)	SARS-CoV-2 exposed (N = 16)
Age (years), median (SD)	33 (3.63)	31 (6.63)
Gestation of infection n (%)
1st trimester		0 (0%)
2nd trimester		4 (25%)
3rd trimester		12 (75%)
WHO 7-point ordinal scale score n (%)
0	14 (100%)
3		11 (75%)
4		3 (18.75%)
5		2 (12.5%)
Vaccination status
None	0 (0%)	9 (56%)
1 dose	1 (7%)	5 (31.25%)
2 doses	7 (50%)	3 (18.75%)
3 doses	6 (43%)	1 (6.25%)
Mode of delivery
Normal vaginal	8 (61.5%)	7 (46.6%)
Instrumental birth	2 (15.4%)	0 (0%)
Caesarean	3 (23.1%)	9 (60%)
Not recorded	1	0
Pregnancy complications	11 (78.5%)	8 (50%)
Gestational diabetes	7 (50%)	1 (6.25%)
Hypertension	1 (7%)	0
Low iron status	2 (14.2%)	0
Preeclampsia	0	1 (6.25%)
Type 2 diabetes	0	1 (6.25%)
Gall bladder stone	0	1 (6.25%)
Macrosomia	0	1 (6.25%)
Compression of cord	0	1 (6.25%)
Hyperthyroidism	0	1 (6.25%)
High blood pressure	0	1 (6.25%)
2 complications or more	3 (21.4%)	1 (6.25%)
Infant biological sex
Male: Female	8: 6	7: 10

Maternal outcome measures

To assess maternal mental health outcomes at birth and 3 months, the Edinburgh Postnatal Depression Scale (EPDS) ¹⁹ was administered. To assess the mother-infant relationship, the Maternal Postnatal Attachment Scale was administered at birth and 3 months ²⁰. Maternal serum was collected at the birth assessment time point and cytokines known to be altered by SARS-CoV-2 ^{8, 21–23}, including interleukin-6 (IL-6), IL-1β, IL-7, IL-17A, interferon-γ (IFN-γ), Monocyte chemoattractant protein-1 (MCP-1) and tumor necrosis factor α (TNFα) levels were assessed using ProQuantum Immunoassay kits (Thermo Fisher Scientific: IFN-γ cat# A35576, IL-6 cat# A35573. IL-1β cat# A35574, IL-7 cat# A35589, IL-17A cat# A35611, MCP-1 cat# A35598, TNFα cat# A35601). Saliva kits (Salimetrics SalivaBio Passive Drool Method cat# 5004.06) to assess cortisol levels were mailed to participants prior to the birth assessment, and on the morning of their assessment maternal participants were asked to provide a saliva sample at waking, and 30 minutes and 45 minutes after waking. Saliva cortisol levels were measured using salivary cortisol ELISA kits (Salimetrics cat# 1-3002).

Infant outcomes at birth and 3 months

At the birth (within 4 weeks of birth) and 3-month timepoints, infant weight, length and head circumference data were collected. Health data record books provided by the participants provided further anthropometry data points at 2, 4 and 8 weeks, as well as birth details, including Apgar scores, mode of delivery and delivery complications. Paediatric physiotherapists administered the Hammersmith Neonatal Neurological Examination (HNNE) and the Hammersmith Infant Neurological Examination (HINE) at birth and 3 months respectively. At the birth timepoint, a buccal swab was collected for DNA extraction and DNA methylation analysis. Buccal swab samples were stored at -20°C until extraction. Samples were de-identified at collection and allocated a study code.

DNA extraction, bisulphite conversion and DNA methylation analysis

DNA extraction and bisulphite conversion were performed as previously described ²⁴. Methylation profiling was conducted using Infinium Methylation EPIC BeadChip (Illumina, San Diego, CA) and imaged using Illumina iScan System.

DNA methylation analysis

Methylation analysis was carried out as previously described ²⁴ and following the R package “ChAMP”, which includes a pre-processing pipeline of quality checks (using minfi R package)²⁵, normalisation (using Beta MIxture Quantile dilation (BMIQ) method²⁶ and batch effect corrections (using sva R package)^{27, 28}. R package “limma” ^{29, 30} was used to identify differentially methylated CpG sites in SARS-CoV-2 samples compared to controls. The “Bumphunter” method ³¹ was employed to further assess for differentially methylated regions (DMRs), using all nominally significant probes to determine segments within the genome which display consistently hyper- or hypomethylated CpGs. Differentially methylated sites with a log2 fold-change (logFC) greater than 0.1 or less than − 0.1 and an adjusted p-value of less than 0.05 were submitted to enrichment analysis performed using Gene Ontology (GO) annotations. The enrichment analysis was visualised using clusterprofiler R package ^{32 33}.

Infant outcome measures at 12 months

Parent-reported questionnaires were completed online at the 12-month timepoint. These included the Ages and Stages Questionnaire (ASQ-3) ³⁴, Vineland Adaptive Behaviour Scale – third edition (VABS-3) ³⁵, and Sensory Profile-2 (SSP2) ³⁶. The ASQ is a widely used screening tool that is easy to use and provides an indication of potential neurodevelopmental delays ³⁴.

Statistical analysis

Data were assessed for normality using D’Agostino, Pearson or Shapiro-Wilk tests. Data that were not normally distributed, were assessed using Mann-Whitney U tests. Normally distributed, data were assessed using unpaired t-tests or 2-way ANOVAs. Mixed effects analysis was used when multiple time points were assessed. Grubb’s test was used to identify outliers due to small sample size. Identified outliers are described in figure legends. Pearson r tests were used to assess correlations.

Cohort characteristics

Cohort characteristics are described in Table 1. The two groups were not significantly different in terms of maternal age. The majority of the SARS-CoV-2 exposed group were infected during the third trimester (75%), with 25% infected in the second trimester and none infected in the first trimester. 93% of the control group mothers but only 25% of the SARS-CoV-2 exposed group received 2 or more doses of the SARS-CoV-2 vaccination. A larger proportion of the SARS-CoV-2 exposed group had caesarean deliveries (60%) compared to controls (23.1%). Pregnancy complications were common among both groups (78% controls and 50% SARS-CoV-2 exposed). However, the control group had a larger proportion of women with gestational diabetes. Approximately 57% of the control group infants and 41% of the SARS-CoV-2 exposed group births were male. One participant in the SARS-Co-V-2 group had twins, therefore there were 17 infants and 16 mothers in this group. At the follow-up 12-month assessment the control group dropped to 7 and the severe SARS-CoV-2 exposed group dropped to 9.

Maternal outcomes

Maternal serum IL-6 levels were significantly increased in SARS-CoV-2 exposed mothers (median (IQR) = 0.037(0.15) vs control 0.006(0.05), p = 0.025, Fig. 1A). Levels of IL-17A were also significantly elevated in SARS-CoV-2 exposed (8.97(33.1) vs control 1.32(17.7), p = 0.045, Fig. 1B.) There was no significant difference in serum levels of IL-1β, TNFα, IFN-γ, IL-7 and MCP-1 between the groups (Fig. 1C-G). Levels of IL-6 correlated significantly and positively with WHO illness severity scores (r² = 0.29, p = 0.006, Fig. 1H) but no other cytokines correlated with illness severity. Cortisol levels significantly increased in both groups 30 minutes after waking (main effect of time, F (F (1.322, 29.09) = 4.099, p = 0.04) and returned to waking levels 45 minutes post-waking with Tukey’s post-hoc comparisons showing significant differences 30 vs. 45 minutes (p = 0.003) but not between 0 and 45 minutes (0.75). However, there was no main effect of group and no group x time interaction for cortisol levels, suggesting cortisol levels were unchanged in SARS-CoV-2 exposed mothers at birth (Fig. 1I). EPDS scores were significantly higher in mothers exposed to SARS-CoV-2 (Mean 7.813) compared to controls (Mean 4.154, 95% CI 0.05–7.29, p = 0.046, Fig. 1J) with 5 women scoring over 10 and 3 women scoring over 13, indicative of clinically significant symptom severity. EPDS scores did not significantly correlate with maternal serum cytokine levels. No group differences were found for scores from the maternal postnatal attachment scale (data not shown).

Infant DNA methylation outcomes at birth

DNA methylation data are from 14 controls and 17 infants exposed to SARS-CoV-2. DNA extracted from buccal swabs collected at birth showed 4306 differentially methylated probes on 3398 protein coding genes in infants exposed to severe SARS-CoV-2 compared to controls, with an adjusted p-value less than 0.05 (See Supplementary table 1 for full list). This is from a total array analysis of ~ 850,000 probes. 577 protein coding genes were hypermethylated (positive log fold change from control), while 2821 were hypomethylated (-ve log fold change from control) (Supplementary table 1). Gene ontology (GO) analysis showed that a number of biological processes involved in neurodevelopment were altered, including ‘neurogenesis’, ‘generation of new neurons’ and ‘neuron projection development’ (Fig. 2A). This was also true for cellular component pathways, with ‘neuron projection’ and ‘synapse’ ranked as the top 2 pathways altered in infants exposed to SARS-CoV-2 compared to controls (Fig. 2B). Altered molecular function pathways included ‘adenyl nucleotide binding’, ‘cytoskeletal protein binding’ and ‘kinase activity’ (Fig. 2C). Of the 3398 differentially methylated probes, 267 of the genes on which these probes target overlapped with the Simon’s Foundation Autism Research Institute (SFARI) gene list, which consist of a total of ~ 1140 genes (Fig. 2D, Supplementary table 2). Of these 267 overlapping genes, synapse organisation was the top pathway altered with 31 genes involved in synapse assembly, organisation and structure found to be differentially methylated, including SHANK3, RELN, NRXN1 and ANK3 (Fig. 2E). Several genes contained multiple methylation sites that were differentially altered in SARS-CoV-2 exposed infants, including ARID1B (x3), AUTS2 (x6), CASZ1 (x5), DIP2C (x4), FOXP1 (x3), GLIS1 (x5), GRIK4 (x3), HDAC4 (x4), and RORA (x5), (Supplementary table 2). In addition, many significant DMPs were within translational start sites (TSS), suggesting they are likely to alter gene expression. This included hypermethylated genes HOMER1, NLGN1, and CNTNAP2 and hypomethylated genes ANK3, GAS2, GRIP1, PDCD1, RELN and WASF1 (bold and italicised in Supplementary table 2).

The top 20 differentially methylated regions (DMRs) that contained protein coding genes are listed in Supplementary table 3. Here, eight of the 23 genes within the top DMRs are medium to highly expressed in the brain (according to the National Centre for Biotechnology Information NCBI genome browser), and several of these genes are associated with intellectual disability, developmental delay, ASD and schizophrenia. These include ABHD16A, PGAP3 and ALDH3A2^37–44, RPTOR, which is a regulator-associated protein of mTOR associated with ASD^{45, 46}, and AFT6B⁴⁷ and SV2A (synaptic vesicle glycoprotein) ⁴⁸, which are associated with schizophrenia.

Infant outcomes from birth to 12 months

At the birth timepoint (within 4 weeks of birth), mothers and their infants attended Monash Medical Centre for biospecimen collection and neurological examination. Birth data collected from Community maternal and child health service records were recorded, including mode of delivery, birth complications, Apgar scores and anthropometry (length, weight and head circumference). Apgar scores were available for 11 controls and 13 SARS-CoV-2 exposed infants at 1 and 5 minutes post-delivery, whereby there was a significant effect of time (F (1, 44) = 8.702, p = 0.005), a significant effect of group (F (1, 44) = 4.250, p = 0.0452), and a significant group x time interaction (F (1, 44) = 4.250, p = 0.0452). Here, multiple comparisons showed that infants exposed to severe SARS-CoV-2 infection in utero had lower Apgar scores at the 1-minute timepoint (Mean diff 1.266, 95% CI of difference 0.02–2.5, p = 0.011), but not at 5 minutes, indicative of reduced tolerance of the birthing process but healthy performance outside the womb (Fig. 3A). No group differences were found for HNNE scores assessed at birth and HINE scores at 3 months (Fig. 3B). Parents completed the ASQ at 12 months and the VBAS-3 and SSP2 at birth, 3 months and 12 months. No significant group differences were found on the VBAS-3 or SSP2 at any time point (data not shown). However, at 12-months there was a significant effect of group on ASQ scores for communication, problem solving and personal-social domains (Difference between predicted means = 9.259, 95% CI of difference = 2.64–15.88, F (1, 39) = 8.005, p = 0.0073), whereby SARS-CoV-2 exposed infants scored lower than non-exposed (Fig. 3C). No group differences were found for general and fine motor skills assessed by the ASQ (data not shown). Maternal IL-6 levels correlated significantly and negatively with child communication (r² = 0.399, p = 0.009, Fig. 3D) and problem-solving scores (R² = 0.322, P = 0.02, Fig. 3E), but not personal-social scores. IL-17A levels did not correlate significantly with child outcomes (data not shown). Maternal EPDS scores did not correlate significantly with child communication, problem solving or personal social scores (data not shown). ASQ scores were not significantly influenced by pregnancy complications or mode of delivery (Supplementary Fig. 1). Males tended to score lower than female infants (Supplementary Fig. 1), however, there are less males than females in the infection-exposed group and when including sex as a variable, there was no significant sex x group interaction for ASQ scores. Weight (Fig. 3F), length (Fig. 3G) and head circumference (Fig. 3H) were not affected by group.

Overall, the above data showed that mothers who were hospitalised due to severe infection with SARS-CoV-2 during pregnancy showed elevated levels of pro-inflammatory cytokines IL-6 and IL-17A several months after recovery from infection (average ~ 3 months post-infection) and increased scores on the EPDS. IL-6 levels correlated significantly with illness severity, but not with EPDS scores suggesting SARS-CoV-2 infection may likely underpin the prolonged elevation in pro-inflammatory cytokines. Infants exposed to severe maternal SARS-CoV-2 show altered DNA methylation of pathways implicated in ASD, particularly synaptic organisation and regulation. Severe SARS-CoV-2 infection exposed infants scored lower on communication, problem solving and personal-social domains of the ASQ at 12 months. Communication and problem-solving scores correlate significantly and negatively with maternal IL-6 levels, suggesting a potential association between elevated maternal serum IL-6 and poorer child development. No significant correlations were found between infant ASQ outcomes and maternal EPDS

Maternal serum IL-6 correlated significantly and negatively with child communication and problem solving. Whilst this appears to be driven mainly by one IL-6 value at 0.417 pg/mL, this value was not identified as a statistical outlier. Nevertheless, if it was removed the correlation with communication became non-significant at 0.09, but the correlation with problem solving becomes highly significant at p = 0.0006, r² = 0.6. Replication of this finding would, therefore, be of value. The relationship between maternal IL-6 levels and offspring neurodevelopment is consistent with animal model studies that demonstrate IL-6 as a critical mediator of MIA induced neurodevelopmental disruptions ¹³. In addition, Graham et al. reported that higher maternal IL-6 concentrations during pregnancy are associated with larger amygdala volume and stronger connectivity (a brain region critical to communication and problem solving) as well as lower impulse control ⁴⁹. The same group also reported an inverse correlation of maternal IL-6 concentrations and offspring nonverbal fluid intelligence and pars triangularis volume ⁵⁰. Further follow up of the current cohort with detailed imaging may be informative. In retrospective studies of patient survival following SARS-CoV-2 infection, it has been reported that IL-6 levels are the strongest predictor of mortality ⁵¹, sparking interest in the use of anti-IL-6 agents for the treatment of COVID-19 ^{8, 52}. Indeed, anti-IL6R treatments tend to be effective in severe disease, but within specific windows of illness progression ⁵³. Once again, further studies in mouse models such as the SARS-CoV-2 mouse adapted model ⁵⁴ would be highly beneficial in determining if IL-6 blockade following severe infection during pregnancy may prevent neurodevelopment disruptions in the offspring, and could inform the appropriate timing of intervention.

We demonstrate here that offspring exposed to severe maternal infection with SARS-CoV-2 in utero show, at birth, altered methylation of gene pathways relevant to ASD, including synaptic organisation and assembly. In addition, many of the differentially methylated regions we identified contained genes implicated in neurodevelopmental disorders ^37–44, ^{45, 46} and schizophrenia ^{47 48}. Of the differentially methylated genes that overlapped with the SFARI gene list, many genes had multiple differentially methylated sites, including the ASD candidate genes Autism Susceptibility Candidate 2 (AUTS2) and RAR Related Orphan Receptor α (RORA). Increased methylation and reduced expression of RORA has been found in lymphoblastoid cells of individuals with ASD compared to sibling controls⁵⁵. However, we identified hypomethylation (reduced methylation) of RORA in infants exposed to SARS-CoV-2 from samples taken at birth, which aligns with a previous study that showed hypomethylation of RORA in peripheral blood DNA of people with ASD⁵⁶. This would be predicted to increase expression of RORA. Similarly, AUTS2 showed 6 differentially hypomethylated sites in SARS-CoV-2 exposed offspring and is associated with a number of neurodevelopmental disorders, including ASD, intellectual disability, schizophrenia and epilepsy, and plays a key postnatal role in regulating excitatory synapses ⁵⁷. Downstream targets of AUTS2 that are involved in synapse formation, such as RELN and CACNA1C, were also hypomethylated in SARS-CoV-2 exposed infants with the DMP for RELN located in the translational start site, suggesting it is likely to alter gene expression.

DNA methylation is a dynamic process that changes substantially according to the stage of development. Global demethylation occurs during early embryonic development ⁵⁸, therefore the significant hypomethylation profile in our sample of infants exposed to SARS-CoV-2 in utero may indicate failure to methylate demethylated DNA, whereby genes remain unmethylated into postnatal development (when the sample was collected). Furthermore, the persistence of elevated maternal IL-6 levels post-delivery suggests that the window within which a maternal immune response could affect the trajectory of methylation during foetal development is expanded well beyond the period of infection itself. Whether the altered methylation identified in infants exposed to SARS-CoV-2 persists throughout child development or is occurring specifically during this window of infancy is an important question that we plan to address through our prospective study design. It is also important here to reiterate that DNA samples are from peripheral buccal swabs, so we do not know if altered DNA methylation reflects brain transcript changes. Here, once again, preclinical models would provide much insight. Indeed, in a human brain organoid model, it was shown that SARS-CoV-2 similarly affected synaptic pathways, causing a loss of excitatory synapses in neurons ⁵⁹.

A major limitation of this study is the sample size, which limited our ability to co-vary for factors such as pregnancy complications and mode of delivery, although we did assess the effects of these factors individually on ASQ outcomes and found no significant effects (Supplementary Fig. 1). Here, the authors suggest caution when interpreting the ASQ outcomes and encourage replication of the above assessments of both DNA methylation and child behaviour using the ASQ on larger cohorts of mother-infant dyads exposed to severe SARS-CoV-2 infection during pregnancy. In addition, while our control group reported no known infection with SARS-CoV-2 during their pregnancy, it is possible they were infected asymptomatically. However, it is important to note we compared controls with severely infected participants. Another limitation is that our infected group had one participant with twins. Ideally, our control group would have a triad with twins to control for this additional strain on the mother and infants.

In conclusion, we report infants exposed to severe maternal SARS-CoV-2 infection show signs of delayed neurodevelopment at 12 months that correlates with maternal serum IL-6 levels taken at birth, suggesting maternal serum IL-6 levels may be a potential candidate biomarker of future adverse child developmental outcomes. We furthermore demonstrate significant DNA methylation changes in infants exposed to severe SARS-CoV-2 in utero, particularly within genes involved in synaptic organisation and assembly. Further planned longitudinal assessments of this cohort will help to delineate the relative risk of SARS-CoV-2 exposure on child mental health outcomes and test the validity of infant DNA methylation markers and maternal serum markers as potential biomarkers of adverse neurodevelopmental outcomes.

Acknowledgments

The authors would like to acknowledge One in Five Philanthropic organisation for providing financial support for this study as well as Illumina and the Australian Neuroscience Society for providing Rachel Hill with a Neurogenomics award to complete the DNA methylation experiment.

Conflicts of interest and data sharing statement

The authors declare no conflicts of interest. The authors are open to sharing data upon request.

Brown AS. Epidemiologic studies of exposure to prenatal infection and risk of schizophrenia and autism. Developmental neurobiology 2012; 72(10): 1272–1276.
Patterson PH. Immune involvement in schizophrenia and autism: etiology, pathology and animal models. Behavioural brain research 2009; 204(2): 313–321.
Atladottir HO, Thorsen P, Ostergaard L, Schendel DE, Lemcke S, Abdallah M et al. Maternal infection requiring hospitalization during pregnancy and autism spectrum disorders. Journal of autism and developmental disorders 2010; 40(12): 1423–1430.
Atladottir HO, Thorsen P, Schendel DE, Ostergaard L, Lemcke S, Parner ET. Association of hospitalization for infection in childhood with diagnosis of autism spectrum disorders: a Danish cohort study. Archives of pediatrics & adolescent medicine 2010; 164(5): 470–477.
Edlow AG, Castro VM, Shook LL, Kaimal AJ, Perlis RH. Neurodevelopmental Outcomes at 1 Year in Infants of Mothers Who Tested Positive for SARS-CoV-2 During Pregnancy. JAMA Netw Open 2022; 5(6): e2215787.
Edlow AG, Castro VM, Shook LL, Haneuse S, Kaimal AJ, Perlis RH. Sex-Specific Neurodevelopmental Outcomes Among Offspring of Mothers With SARS-CoV-2 Infection During Pregnancy. JAMA Netw Open 2023; 6(3): e234415.
Hosier H, Farhadian SF, Morotti RA, Deshmukh U, Lu-Culligan A, Campbell KH et al. SARS-CoV-2 infection of the placenta. J Clin Invest 2020; 130(9): 4947–4953.
Copaescu A, Smibert O, Gibson A, Phillips EJ, Trubiano JA. The role of IL-6 and other mediators in the cytokine storm associated with SARS-CoV-2 infection. J Allergy Clin Immunol 2020; 146(3): 518–534 e511.
Anderson G, Berk M, Dodd S, Bechter K, Altamura AC, Dell'osso B et al. Immuno-inflammatory, oxidative and nitrosative stress, and neuroprogressive pathways in the etiology, course and treatment of schizophrenia. Progress in neuro-psychopharmacology & biological psychiatry 2013; 42: 1–4.
Brown AS, Begg MD, Gravenstein S, Schaefer CA, Wyatt RJ, Bresnahan M et al. Serologic evidence of prenatal influenza in the etiology of schizophrenia. Archives of general psychiatry 2004; 61(8): 774–780.
Meyer U, Feldon J. Epidemiology-driven neurodevelopmental animal models of schizophrenia. Progress in neurobiology 2010; 90(3): 285–326.
Zerbo O, Qian Y, Yoshida C, Grether JK, Van de Water J, Croen LA. Maternal Infection During Pregnancy and Autism Spectrum Disorders. Journal of autism and developmental disorders 2015; 45(12): 4015–4025.
Smith SE, Li J, Garbett K, Mirnics K, Patterson PH. Maternal immune activation alters fetal brain development through interleukin-6. The Journal of neuroscience: the official journal of the Society for Neuroscience 2007; 27(40): 10695–10702.
Choi GB, Yim YS, Wong H, Kim S, Kim H, Kim SV et al. The maternal interleukin-17a pathway in mice promotes autism-like phenotypes in offspring. Science 2016; 351(6276): 933–939.
Richetto J, Massart R, Weber-Stadlbauer U, Szyf M, Riva MA, Meyer U. Genome-wide DNA Methylation Changes in a Mouse Model of Infection-Mediated Neurodevelopmental Disorders. Biological psychiatry 2017; 81(3): 265–276.
Basil P, Li Q, Dempster EL, Mill J, Sham PC, Wong CC et al. Prenatal maternal immune activation causes epigenetic differences in adolescent mouse brain. Transl Psychiatry 2014; 4: e434.
Palacios-Garcia I, Lara-Vasquez A, Montiel JF, Diaz-Veliz GF, Sepulveda H, Utreras E et al. Prenatal stress down-regulates Reelin expression by methylation of its promoter and induces adult behavioral impairments in rats. PloS one 2015; 10(2): e0117680.
Hill RA, Malhotra A, Sackett V, Williams K, Fahey M, Palmer KR et al. A prospective, longitudinal, case-control study to evaluate the neurodevelopment of children from birth to adolescence exposed to COVID-19 in utero. BMC Pediatr 2023; 23(1): 48.
Cox J. Use and misuse of the Edinburgh Postnatal Depression Scale (EPDS): a ten point 'survival analysis'. Arch Womens Ment Health 2017; 20(6): 789–790.
Condon JT, Corkindale CJ. The assessment of parent-to-infant attachment: Development of a self-report questionnaire instrument. Journal of Reproductive and Infant Psychology 1998; 16(1): 57–76.
Karaba AH, Zhou W, Hsieh LL, Figueroa A, Massaccesi G, Rothman RE et al. Differential Cytokine Signatures of Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) and Influenza Infection Highlight Key Differences in Pathobiology. Clin Infect Dis 2022; 74(2): 254–262.
Potere N, Batticciotto A, Vecchie A, Porreca E, Cappelli A, Abbate A et al. The role of IL-6 and IL-6 blockade in COVID-19. Expert Rev Clin Immunol 2021; 17(6): 601–618.
Gubernatorova EO, Gorshkova EA, Polinova AI, Drutskaya MS. IL-6: Relevance for immunopathology of SARS-CoV-2. Cytokine Growth Factor Rev 2020; 53: 13–24.
Hill RA, Gibbons A, Han U, Suwakulsiri W, Taseska A, Hammet F et al. Maternal SARS-CoV-2 exposure alters infant DNA methylation. Brain Behav Immun Health 2023; 27: 100572.
Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics 2014; 30(10): 1363–1369.
Teschendorff AE, Marabita F, Lechner M, Bartlett T, Tegner J, Gomez-Cabrero D et al. A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data. Bioinformatics 2013; 29(2): 189–196.
Leek JT, Johnson WE, Parker HS, Jaffe AE, Storey JD. The sva package for removing batch effects and other unwanted variation in high-throughput experiments. Bioinformatics 2012; 28(6): 882–883.
Morris TJ, Butcher LM, Feber A, Teschendorff AE, Chakravarthy AR, Wojdacz TK et al. ChAMP: 450k Chip Analysis Methylation Pipeline. Bioinformatics 2014; 30(3): 428–430.
Wettenhall JM, Smyth GK. limmaGUI: a graphical user interface for linear modeling of microarray data. Bioinformatics 2004; 20(18): 3705–3706.
Smyth GK. Linear models and empirical bayes methods for assessing differential expression in microarray experiments. Stat Appl Genet Mol Biol 2004; 3: Article3.
Jaffe AE, Murakami P, Lee H, Leek JT, Fallin MD, Feinberg AP et al. Bump hunting to identify differentially methylated regions in epigenetic epidemiology studies. Int J Epidemiol 2012; 41(1): 200–209.
Yu G, Wang LG, Han Y, He QY. clusterProfiler: an R package for comparing biological themes among gene clusters. OMICS 2012; 16(5): 284–287.
Hill RA, Gibbons A, Han U, Suwakulsiri W, Taseska A, Hammet F et al. Maternal SARS-CoV-2 exposure alters infant DNA methylation. Brain, Behavior, & Immunity-Health 2023; 27: 100572.
Squires J, Bricker D. Ages and Stages Questionnaire, Third Edition. Paul H. Brookes Publishing Co., Inc2009.
Sparrow SS, Cicchetti DV, Saulnier CA. Vineland adaptive behavior scales: San Antonio, TX, 2016.
Dunn W. Sensory Profile 2. User's manual. Bloomington: Pearson 2014.
Rajeshwari M, Karthi S, Singh R, Efthymiou S, Gowda VK, Varalakshmi P et al. Novel ALDH3A2 mutations in structural and functional domains of FALDH causing diverse clinical phenotypes in Sjogren-Larsson syndrome patients. Human mutation 2021; 42(8): 1015–1029.
Abidi KT, Kamal NM, Bakkar AA, Alotaibi M, Asseri H, Bokari KA. Sjogren-Larsson Syndrome: A case series of five members from an extended family with a novel mutation. Mol Genet Genomic Med 2020; 8(11): e1487.
Bezuidenhout H, Bayley S, Smit L, Kinnear C, Moller M, Uren C et al. Hyperphosphatasia with mental retardation syndrome type 4 in three unrelated South African patients. Am J Med Genet A 2020; 182(10): 2230–2235.
Abdel-Hamid MS, Issa MY, Otaify GA, Abdel-Ghafar SF, Elbendary HM, Zaki MS. PGAP3-related hyperphosphatasia with mental retardation syndrome: Report of 10 new patients and a homozygous founder mutation. Clin Genet 2018; 93(1): 84–91.
Knaus A, Awaya T, Helbig I, Afawi Z, Pendziwiat M, Abu-Rachma J et al. Rare Noncoding Mutations Extend the Mutational Spectrum in the PGAP3 Subtype of Hyperphosphatasia with Mental Retardation Syndrome. Human mutation 2016; 37(8): 737–744.
Howard MF, Murakami Y, Pagnamenta AT, Daumer-Haas C, Fischer B, Hecht J et al. Mutations in PGAP3 impair GPI-anchor maturation, causing a subtype of hyperphosphatasia with mental retardation. American journal of human genetics 2014; 94(2): 278–287.
Lemire G, Ito YA, Marshall AE, Chrestian N, Stanley V, Brady L et al. ABHD16A deficiency causes a complicated form of hereditary spastic paraplegia associated with intellectual disability and cerebral anomalies. American journal of human genetics 2021; 108(10): 2017–2023.
Miyake N, Silva S, Troncoso M, Okamoto N, Andachi Y, Kato M et al. A homozygous ABHD16A variant causes a complex hereditary spastic paraplegia with developmental delay, absent speech, and characteristic face. Clin Genet 2022; 101(3): 359–363.
Chen CJ, Sgritta M, Mays J, Zhou H, Lucero R, Park J et al. Therapeutic inhibition of mTORC2 rescues the behavioral and neurophysiological abnormalities associated with Pten-deficiency. Nat Med 2019; 25(11): 1684–1690.
Sato A. mTOR, a Potential Target to Treat Autism Spectrum Disorder. CNS Neurol Disord Drug Targets 2016; 15(5): 533–543.
Xue X, Wu X, Liu L, Liu L, Zhu F. ERVW-1 Activates ATF6-Mediated Unfolded Protein Response by Decreasing GANAB in Recent-Onset Schizophrenia. Viruses 2023; 15(6).
Yoon JH, Zhang Z, Mormino E, Davidzon G, Minzenberg MJ, Ballon J et al. Reductions in synaptic marker SV2A in early-course Schizophrenia. Journal of psychiatric research 2023; 161: 213–217.
Graham AM, Rasmussen JM, Rudolph MD, Heim CM, Gilmore JH, Styner M et al. Maternal Systemic Interleukin-6 During Pregnancy Is Associated With Newborn Amygdala Phenotypes and Subsequent Behavior at 2 Years of Age. Biological psychiatry 2018; 83(2): 109–119.
Rasmussen JM, Graham AM, Gyllenhammer LE, Entringer S, Chow DS, O'Connor TG et al. Neuroanatomical Correlates Underlying the Association Between Maternal Interleukin 6 Concentration During Pregnancy and Offspring Fluid Reasoning Performance in Early Childhood. Biol Psychiatry Cogn Neurosci Neuroimaging 2022; 7(1): 24–33.
Zhou F, Yu T, Du R, Fan G, Liu Y, Liu Z et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet 2020; 395(10229): 1054–1062.
Crisafulli S, Isgro V, La Corte L, Atzeni F, Trifiro G. Potential Role of Anti-interleukin (IL)-6 Drugs in the Treatment of COVID-19: Rationale, Clinical Evidence and Risks. BioDrugs 2020; 34(4): 415–422.
Zizzo G, Tamburello A, Castelnovo L, Laria A, Mumoli N, Faggioli PM et al. Immunotherapy of COVID-19: Inside and Beyond IL-6 Signalling. Front Immunol 2022; 13: 795315.
Bader SM, Cooney JP, Sheerin D, Taiaroa G, Harty L, Davidson KC et al. SARS-CoV-2 mouse adaptation selects virulence mutations that cause TNF-driven age-dependent severe disease with human correlates. Proceedings of the National Academy of Sciences of the United States of America 2023; 120(32): e2301689120.
Nguyen A, Rauch TA, Pfeifer GP, Hu VW. Global methylation profiling of lymphoblastoid cell lines reveals epigenetic contributions to autism spectrum disorders and a novel autism candidate gene, RORA, whose protein product is reduced in autistic brain. FASEB J 2010; 24(8): 3036–3051.
Homs A, Codina-Sola M, Rodriguez-Santiago B, Villanueva CM, Monk D, Cusco I et al. Genetic and epigenetic methylation defects and implication of the ERMN gene in autism spectrum disorders. Transl Psychiatry 2016; 6(7): e855.
Hori K, Shimaoka K, Hoshino M. AUTS2 Gene: Keys to Understanding the Pathogenesis of Neurodevelopmental Disorders. Cells 2021; 11(1).
Tremblay MW, Jiang YH. DNA Methylation and Susceptibility to Autism Spectrum Disorder. Annu Rev Med 2019; 70: 151–166.
Mesci P, de Souza JS, Martin-Sancho L, Macia A, Saleh A, Yin X et al. SARS-CoV-2 infects human brain organoids causing cell death and loss of synapses that can be rescued by treatment with Sofosbuvir. PLoS Biol 2022; 20(11): e3001845.

The authors have declared there is NO conflict of interest to disclose

SupplementaryFigure1.tif
Supplementary Figure 1. ASQ-3 scores assessed by A. Pregnancy complications, B. Model of delivery and C. Infant sex. There was no significant effect of pregnancy complications or mode of delivery on ASQ-3 scores. Male infants scored significantly lower than female infants. *, p=<0.05.
Supplementarytable1DMPsHill.xlsx
Supplementary table 1. Lists all differentially methylated probes (DMPs) detected in the severe SARS-CoV-2 exposed infants compared to non-exposed controls. Log FC = log fold change, β values are the ratio of intensity of the methylated probe and the overall intensity (sum of methylated and unmethylated probe intensities) (range from 0-1), CHR = chromosome, feat. cgi = feature CpG island (where it is located on the gene), SNP_ID = single nuclei polymorphisms located within the same region, SNP_Distance = distance from CpG site.
Supplementarytable2DMPsmatchedtoSFARIHill.xlsx
Supplementary table 2. Lists all DMPs that matched to the Simons Foundation Autism Research Institute (SFARI) gene list. Methylated sites located within the translational start site (TSS) are in bold and italics. Log FC = log fold change, β values are the ratio of intensity of the methylated probe and the overall intensity (sum of methylated and unmethylated probe intensities) (range from 0-1), CHR = chromosome, feat. cgi = feature CpG island (where it is located on the gene)
Supplementarytable3top20DMRs.xlsx
Supplementary table 3. Lists the top 20 differentially methylated regions (DMRs), the genes located within these regions, their function, their expression within the brain and their association with psychiatric or neurodevelopmental disorders.

Download PDF

Editorial decision: revise
20 Aug, 2024
Review #1 received at journal
10 Jul, 2024
Review #2 received at journal
08 Jul, 2024
Reviewer #2 agreed at journal
17 Jun, 2024
Reviewer #1 agreed at journal
16 Jun, 2024
Reviewers invited by journal
14 Jun, 2024
Editor assigned by journal
13 Jun, 2024
Submission checks completed at journal
12 Jun, 2024
First submitted to journal
11 Jun, 2024
Unknown event
11 Jun, 2024

You are reading this latest preprint version

Investigating the impact of severe maternal SARS-CoV-2 infection on infant DNA methylation and neurodevelopment

Status:

Version 1

Abstract

Figures

INTRODUCTION

SUBJECTS AND METHODS

Participant recruitment:

Maternal outcome measures

Infant outcomes at birth and 3 months

DNA extraction, bisulphite conversion and DNA methylation analysis

DNA methylation analysis

Infant outcome measures at 12 months

Statistical analysis

RESULTS

Cohort characteristics

Maternal outcomes

Infant DNA methylation outcomes at birth

Infant outcomes from birth to 12 months

DISCUSSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1