Skip to main content
NCBI home page
Cerebral Cortex (New York, NY) logoLink to Cerebral Cortex (New York, NY)
. 2024 May 2;34(13):50–62. doi: 10.1093/cercor/bhae082

Differences in mid-gestational and early postnatal neonatal cytokines and chemokines are associated with patterns of maternal autoantibodies in the context of autism

Janna McLellan 1, Lisa A Croen 2, Ana-Maria Iosif 3, Paul Ashwood 4,5, Cathleen Yoshida 6, Kimberly Berger 7, Judy Van de Water 8,9,✉
PMCID: PMC11065110  PMID: 38696596

Abstract

Associations between maternal immune dysregulation (including autoimmunity and skewed cytokine/chemokine profiles) and offspring neurodevelopmental disorders such as autism have been reported. In maternal autoantibody-related autism, specific maternally derived autoantibodies can access the fetal compartment to target eight proteins critical for neurodevelopment. We examined the relationship between maternal autoantibodies to the eight maternal autoantibody-related autism proteins and cytokine/chemokine profiles in the second trimester of pregnancy in mothers of children later diagnosed with autism and their neonates’ cytokine/chemokine profiles. Using banked maternal serum samples from 15 to 19 weeks of gestation from the Early Markers for Autism Study and corresponding banked newborn bloodspots, we identified three maternal/offspring groups based on maternal autoantibody status: (1) mothers with autoantibodies to one or more of the eight maternal autoantibody-related autismassociated proteins but not a maternal autoantibody-related autism-specific pattern, (2) mothers with a known maternal autoantibody-related autism pattern, and (3) mothers without autoantibodies to any of the eight maternal autoantibody-related autism proteins. Using a multiplex platform, we measured maternal second trimester and neonatal cytokine/chemokine levels. This combined analysis aimed to determine potential associations between maternal autoantibodies and the maternal and neonatal cytokine/chemokine profiles, each of which has been shown to have implications on offspring neurodevelopment independently.

Keywords: autism, autoantibody, chemokines, cytokines

Introduction

Immune signaling molecules, such as cytokines and chemokines, are essential factors during pregnancy for healthy development. For example, interleukin (IL)-10 helps to modulate the maternal immune response and promotes placental growth (Roberts et al. 2003; Cheng and Sharma 2015). IL-1β is upregulated during implantation and is a key mediator of the myometrium during parturition (Yockey and Iwasaki 2018; Equils et al. 2020). Notably, both human and animal studies have shown an association between aberrant levels of maternal cytokines and chemokines during pregnancy and altered neurodevelopment in offspring (Dammann and Leviton 1997; Estes and McAllister 2016; Jones et al. 2017). For example, increased mid-gestational levels of IL-4, IL-5, and interferon-gamma (IFNγ) have been associated with an increased risk of having a child with autism (Goines et al. 2011).

Similar to maternal cytokines and chemokines, differences in neonatal levels of cytokines and chemokines have also been associated with neurodevelopmental disorders. Elevated levels of the proinflammatory cytokine IL-6 and cutaneous T-cell-attracting chemokine (CTACK/CCL27) in newborn blood spots have been associated with an increased risk of autism (Heuer et al. 2019; Kim et al. 2022). These findings suggest that immune dysregulation noted at birth could have lasting implications (Saito 2001; Djuardi et al. 2009; Allswede et al. 2020; Smolen et al. 2021).

During pregnancy, maternal antibodies protect the mother and the growing fetus against infection. Maternal immunoglobulin G (IgG) antibodies cross the placenta via the neonatal Fc receptor (FcRn) to provide early immunological protection for the neonate. This process begins late in the first trimester, with maternally derived IgG concentrations increasing as pregnancy progresses (Pyzik et al. 2019; Ciobanu et al. 2020). In cases of maternal autoimmunity, self-targeted maternal IgG autoantibodies (ABs) are equally capable of crossing the placenta and gaining access to their target antigens in the fetus. The association of maternal autoimmunity with neonatal pathogenesis has been documented in cases of maternal systemic lupus erythematosus and Grave’s disease, which can result in congenital heart block and neonatal goiter, respectively (Hon and Leung 2012; Pacheco et al. 2017; Panaitescu and Nicolaides 2018; Ciobanu et al. 2020). The maternal production of specific ABs has also been linked to maternal autoantibody-related (MAR) autism in a subset of children, though the mechanisms of action remain unknown (Braunschweig et al. 2008; Ramirez-Celis et al. 2021).

In cases of MAR autism, mothers produce ABs that bind to eight proteins important for fetal development. These proteins include neuron-specific enolase (NSE), guanine deaminase (GDA), stress-induced phosphoprotein 1 (STIP1), collapsin response mediator proteins 1 and 2 (CRMP1/2), lactate dehydrogenase A and B (LDHA/B), and Y-box binding protein (YBOX) (Braunschweig et al. 2008; Ramirez-Celis et al. 2019). While these protein targets can be found throughout the body, many are crucial for brain development, with CRMP1/2 primarily expressed in the brain (Ravindran et al. 2022). Certain combinations of these ABs, including CRMP1 + CRMP2, CRMP1 + GDA, CRMP2 + STIP1, CRMP1 + STIP1, GDA + Y-BOX1, STIP1 + NSE, LDHA+YBOX-1, and LDHB+YBOX-1, are highly specific to autism (Ramirez-Celis et al. 2021; Angkustsiri et al. 2022; Ramirez-Celis et al. 2022).

Previous research using the Early Markers for Autism (EMA) Study population individually examined the roles of MAR-autism ABs and maternal and neonatal cytokines/chemokines in autism outcomes (Jones et al. 2017; Heuer et al. 2019; DHJ et al. 2023). Within the EMA population, maternal reactivity to one of the MAR-autism AB patterns was observed in 10% of mothers of children with autism versus only 4% of mothers of children with developmental delay and only 1% of mothers of general population controls (Ramirez-Celis et al. 2022). However, how maternal autoimmunity is related to dysregulation of cytokines and chemokines during pregnancy and in the neonate is unclear. To address this knowledge gap, we comprehensively examined the relationship between MAR ABs and maternal and neonatal cytokine/chemokine profiles in the context of autism. These analyses will provide a better understanding of the impact of two paradigms of maternal immune dysregulation in the context of altered neurodevelopment.

Materials and methods

Study population

Samples were obtained from individuals enrolled in the EMA Study (Croen et al. 2008), a population-based, nested case–control study aimed at identifying biomarkers for autism. Study participants were mother–child dyads with available archived prenatal and newborn blood specimens collected originally for the California Department of Public Health prenatal and newborn screening program. Dyads were eligible for inclusion in EMA if mothers participated in the prenatal extended alpha-fetoprotein screening program (XAFP) and delivered a live-born infant between July 2000 and September 2003 and for whom a newborn bloodspot (NBS) was available. The original study was composed of three groups: children with autism (n = 486), children with developmental delay (DD) but not autism (n = 174), and general population controls (n = 397) and their respective mothers. Children with autism or DD were ascertained from regional centers operation by the California Department of Developmental Services, and their diagnostic status was validated by a blinded clinician using the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) criteria. As the presence of MAR ABs is highly associated with autism, with certain patterns present exclusively in mothers of children with autism, for this study we examined only the maternal/offspring data from mothers with neonates later diagnosed with autism. Maternal demographic information, including age, race, ethnicity, and birth country, was obtained through birth certificates. Additional maternal information, including maternal weight and gestational age, were obtained at the time of prenatal sample collection. Neonatal information, including birth weight, birth date, sex, gestational age at birth, and birth type (singleton or multiparous), was obtained through California Vital Statistics. Expanded details of the study population have been previously published (Jones et al. 2017; Heuer et al. 2019; Ramirez-Celis et al. 2022). The institutional review boards of the California Health and Human Services Agency and Kaiser Permanente Northern California approved all study procedures.

Sample collection

Maternal serum samples were collected between 15 and 19 weeks of gestation. Samples were collected in serum separator tubes by obstetrical care providers and underwent XAFP testing within 7 days of collection at a central laboratory. Leftover specimens were shipped on dry ice to our laboratory and stored at −80°C before use in cytokine/chemokine and AB measurement assays. Expanded details have been previously described (Jones et al. 2017). Newborn capillary blood samples (NBS) were collected within 72 h of birth using the heel prick method and were spotted onto standardized filter paper. Bloodspots were transported without temperature control to a regional screening laboratory for analysis, and the remaining specimens were cataloged and stored at −20°C by the California Department of Public Health. NBS samples from preterm infants (<210 gestational days, n = 7), infants with longer than average gestations (>330 days, n = 2), or infants who had NBS collected >72 h after birth were excluded (n = 39) as these factors can directly influence the neonatal cytokine/chemokine levels. After removing these samples, 438 cases remained.

Measurement of maternal autoantibodies

The presence of maternal ABs to the eight proteins of interest (CRMP1/2, STIP1, LDHA/B, NSE, YBOX1, and GDA) was detected by enzyme-linked immunosorbent assay (ELISA) in mid-gestational maternal sera as previously described (Ramirez-Celis et al. 2022). Briefly, microtiter plates were coated with antigen (2 to 3 μg/μL as optimized for each protein) in carbonate coating buffer (pH 9.6) and incubated overnight at 4°C. The following day, the plates were washed with wash buffer (0.05% phosphate-buffered saline, Tween-20%) and blocked with 2% SuperBlock (Thermo Scientific, Rockford, Illinois). The diluted plasma samples were added in duplicate, followed by a 1.5-h incubation. After washing, the plates were incubated with a secondary antibody (goat anti-human IgG-HRP; Kirkegaard & Perry Laboratories, Inc., Gaithersburg, Maryland), washed with wash buffer, followed by the addition of BD optEIA liquid substrate for ELISA (BD Biosciences, San Jose, California). Following incubation, the reaction was stopped with 2 N HCl. Absorbance was measured at 450 to 595 nm using an iMark Microplate Absorbance Reader (Bio-Rad Laboratories, Hercules, California, USA). Mother–child dyads were categorized into three groups based on maternal AB status: (1) those with individual MAR ABs but not a pattern that is specific to MAR autism (AB+; n = 231); (2) those with at least one or more MAR autism-specific patterns, which include the following combinations of ABs: CRMP1 + CRMP2, CRMP1 + GDA, CRMP2 + STIP1, CRMP1 + STIP1, GDA + Y-BOX1, STIP1 + NSE, LDHA+YBOX-1, or LDHB+YBOX-1 (MAR+; n = 43); and (3) those that did not have any ABs to the 8 proteins of interest (AB−; n = 164). In secondary analysis, dyads were further stratified by child intellectual disability (ID) status (autism+ID, n = 228; autism-noID, n = 210). ID status was based on regional center records with composite scores of <70 on the standardized cognitive and functional tests. An expert clinician then reviewed designations.

Measurement of cytokines and chemokines

For the maternal samples, mid-gestational serum concentrations of 22 cytokines and chemokines were measured using a Millipore Multiples bead-based kit (Milliplex MAP Human Cytokine/Chemokine Kit; Millipore, Billerica, Massachusetts, USA). Further details on assay methods have been previously published (Jones et al. 2017). For the NBS samples, dried bloodspots were received as three 3-mm punches per subject and were stored at −80°C until elution. For elution, each sample was placed in 200 μL of elution buffer (0.5% BSA in 50 mL PBS with 1 tablet of Roche Complete Protease Inhibitor Cocktail; Roche Applied Science, Indianapolis, Indiana) and was placed on a shaker overnight at 4°C. The sample eluate was used to measure 42 cytokines and chemokines using a Luminex Multiplex magnetic bead assay (Bio-Rad Laboratories). Two single beads (IL-12p70 and IL-13) were mixed with 40-plex beads of the Bio-Plex Pro Human Chemokines kit (Bio-Rad Laboratories), and the assay was run according to the manufacturer’s directions. A bicinchoninic acid assay (ThermoScientific, Rockford, Illinois) was used to determine total protein for normalization of cytokine/chemokine levels against protein variation. Further details on the NBS sample preparation and assay methods are described in previous publications (Jones et al. 2017; Heuer et al. 2019).

Statistical analysis

All maternal and neonatal cytokine/chemokine data were natural log-transformed. Descriptive statistics (frequencies, means, standard deviations, medians, quartiles) summarized the sociodemographic and clinical variables and cytokine/chemokine concentrations.

Multiple linear regression models were used to assess the association of AB status with maternal and neonatal cytokines/chemokines after adjusting for covariates. Separate models were fit with each maternal and neonatal cytokine/chemokine as the outcome and maternal AB group (MAR+, AB+, or AB−) as the main predictor of interest. Models were adjusted for factors known to influence cytokine/chemokine levels and for those broadly associated with the AB group (P-value < 0.2). Specifically, models for maternal cytokines/chemokines were adjusted for maternal ethnicity (Hispanic or Non-Hispanic), race (White, Asian, or Other), country of birth (United States, Mexico, or Other), and maternal weight (lb) and gestational age (days) at the time of maternal blood collection. Models for neonatal samples were adjusted for maternal ethnicity (Hispanic or Non-Hispanic), race (White, Asian, or Other), country of birth (United States, Mexico, or Other), and weight (lb) at the time of bloodspot collection, birth type (singleton or multiparous), child birth year (2000 to 2003), birth season (Spring, Summer, Fall, or Winter), birth weight (g), sex (Male or Female), gestational age (days) at birth, and age (hours) at newborn bloodspot collection. To assess the statistical significance of the differences in cytokine/chemokine concentrations across maternal AB groups, for both maternal and NBS samples we first conducted overall, two-degrees-of-freedom F-tests for AB group using the multiple linear regression models described above. Then, we constructed linear contrasts for each cytokine/chemokine to identify all pairs of maternal AB groups with significantly different cytokine/chemokine concentrations (AB+ vs. AB−, MAR+ vs. AB−, and MAR+ vs. AB+). Previous studies have reported differences in maternal and neonatal cytokine/chemokine profiles based on the child’s ID status. Therefore, in secondary analyses, we repeated the analyses described above, stratifying by ID status to assess whether the effect of MAR ABs was modified by ID status.

Since the cytokines/chemokines had different ranges, we used the multiple linear regression models to calculate standardized effect sizes (Cohen’s d) that accounted for the imbalance in groups and covariates (Nakagawa and Cuthill 2007). We used the following formula:

graphic file with name DmEquation1.gif

where t is the t value obtained for evaluating the difference between groups i and j from the multiple linear regression model, df is the degrees of freedom used for the t value, and ni and nj are the sample sizes for groups i and j in the respective model. Tests were two-sided, with α = 0.05. All analyses were conducted in SAS OnDemand version 9.4 (SAS Institute Inc., Cary, North Carolina).

Results

Maternal and child demographic and clinical characteristics stratified by maternal AB group (AB+, MAR+, and AB−) are presented in Table 1. The only variables that were statistically different across the three AB groups were birth weight and birth year. Neonates born to MAR+ women had lower birthweights than neonates born to AB+ and AB− women. In addition, MAR+ women had a higher percentage of births in the years 2000 and 2003 compared to AB+ and AB− mothers. Summaries of the maternal and neonatal cytokines, including mean values and lower and upper quartiles for each maternal AB group, are shown in Supplementary Tables 1 and 2, respectively.

Table 1.

Characteristics of the participants. We included children from the early markers for autism study who were later diagnosed with autism spectrum disorder and their mothers.

Characteristic AB+
(n = 231)		 MAR+
(n = 43)		 AB−
(n = 164)		 P-valuea
Maternal birth country, n (%) 0.80
 United States 105 (45%) 19 (44%) 84 (52%)
 Mexico 60 (26%) 11 (26%) 40 (24%)
 Other 66 (29%) 13 (30%) 40 (24%)
Maternal race, n (%) 0.50
 White 171 (74%) 28 (65%) 128 (78%)
 Asian 39(17%) 9 (21%) 24 (15%)
 Other 21 (9%) 6 (14%) 12 (7%)
Maternal ethnicity, n (%) 0.77
 Hispanic 97 (42%) 17 (40%) 63 (38%)
 Non-Hispanic 143 (58%) 26 (60%) 101 (62%)
 Maternal age (yr), mean (SD) 30.0 (5.5) 30.2 (5.5) 29.9 (5.6) 0.94
 Gestational age at XAFP blood draw (d), mean (SD) 119.2 (8.4) 118.5 (8.3) 119.5 (9.3) 0.56
 Maternal weight at XAFP blood draw (lb), mean (SD) 151.4 (34.2) 155.8 (41.7) 151.5 (38.3) 0.76
Child sex, n (%) 0.44
 Male 182 (79%) 36 (84%) 137 (84%)
 Female 49 (21%) 7 (16%) 27 (16%)
Birth type—plurality, n (%) 0.49
 Singleton 227 (98%) 41 (95%) 160 (98%)
 Multiple 4 (2%) 2 (5%) 4 (2%)
 Gestational age at birth (d), mean (SD) 275.0 (13.1) 274.3 (16.1) 274.9 (13.2) 0.94
Birth season, n (%) 0.67
 Winter (Dec–Feb) 52 (23%) 9 (21%) 29 (18%)
 Spring (Mar–May) 68 (29%) 15 (35%) 50 (30%)
 Summer (Jun–Aug) 66 (29%) 11 (25%) 42 (26%)
 Fall (Sep–Nov) 45 (19%) 8 (19%) 43 (26%)
Birth year, n (%) 0.06
 2000 41 (18%) 11 (25%) 25 (15%)
 2001 65 (28%) 5 (12%) 45 (28%)
 2002 96 (42%) 15 (35%) 69 (42%)
 2003 29 (12%) 12 (28%) 25 (15%)
Age at NBS blood draw (h), mean (SD) 31.8 (11.1) 29.6 (7.1) 30.7 (12.1) 0.32
Birth weight (g), mean (SD) 3,459.7 (529.5) 3,261.0 (494.8) 3,494.4 (505.1) 0.03
Child diagnosis, n (%) 0.51
 Autism with intellectual disability 118 (51%) 26 (60%) 84 (51%)
 Autism without intellectual disability 113 (49%) 17 (40%) 80 (49%)
Open in a new tab

aGroup differences were assessed using chi-square tests for categorical variables and analysis of variance for continuous variables.

Abbreviations: AB+, women with autoantibodies related to MAR autism, but not a known specific pattern; MAR+, women with MAR autism-specific patterns of autoantibodies; AB−, women without autoantibodies to any of our tested antigens; SD, standard deviation; NBS, newborn bloodspot; XAFP, extended alpha-fetoprotein screening program.

Presence of any maternal AB and maternal cytokine/chemokine profile

After assessing overall ABs group differences in maternal cytokines and chemokines at 15 to 19 weeks of gestation, we examined the pairwise group differences. We first report the differences in maternal cytokine/chemokine levels between AB+ mothers and AB− mothers. AB+ mothers had significantly higher levels of pre-B and pre-T cell growth factor, IL-7, than AB− mothers (Fig. 1A and Supplementary Table 3). No other comparisons had a significant overall test for group. We next assessed if previously identified patterns of MAR ABs known to be associated with mothers of children with autism (MAR+) correlated with specific changes in maternal cytokines and chemokines concentrations. We compared maternal cytokine/chemokine levels among MAR+ mothers to AB− mothers and separately to AB+ mothers. We found that MAR+ mothers had significantly lower levels of IL-7 compared to AB+ mothers (Fig. 1C and Supplementary Table 3).

Fig. 1.

Fig. 1

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in mid-gestational maternal cytokine and chemokine concentrations using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in AB+ mothers when compared to AB− mothers are shown to the right of the center. Those with higher concentrations in AB− mothers than AB+ mothers are shown to the left of the center. B) Cytokine/chemokines with higher concentrations in MAR+ mothers are shown to the right of the center, and those with higher concentrations in AB− mothers are shown to the left of the center. C) Cytokine/chemokines with higher concentrations in MAR+ are shown to the right of the center, and those with higher concentrations in AB+ mothers are shown to the left of the center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*).

Effect size comparisons between AB+, AB−, and MAR+ mothers

We examined effect size differences to further assess trending differences between maternal groups. When evaluating the effect size of the differences between AB+ and AB− mothers, we found that AB+ mothers had higher levels (d > 0.15) of the macrophage activation factor, interferon-gamma (IFNγ), the B-cell growth and differentiation cytokine, IL-13, and the IL-12 regulator, IL12-p40, with lower levels (d < −0.15) of the innate inflammatory cytokines IL-6, IL-1β, and tumor necrosis factor alpha (TNFα) (Fig. 1A).

MAR+ mothers also had higher levels of IFNγ than AB− mothers (Fig. 1B), lower levels of the B-cell activating and Th2-associated cytokine IL-4 compared to AB− and AB+ mothers (Fig. 1B and C), as well as higher levels of the inflammatory cytokines IL-1β and TNFα than AB+ mothers (Fig. 1C).

Presence of maternal AB (AB+) and neonatal cytokine/chemokine patterns

We next sought to determine if there were any significant neonatal cytokine/chemokine pairwise differences between AB groups. When comparing the neonates from AB+ mothers (AB+ group) to those from AB− mothers (AB− group), we observed significantly lower levels of eotaxin-1, IL-13, MIP-1α, and IL-12p70 in the AB+ group (Supplementary Table 4, Fig. 2A).

Fig. 2.

Fig. 2

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in neonatal cytokines and chemokines concentration using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in neonates from AB+ mothers when compared to neonates from AB− mothers are shown to the right of the center. Those with higher concentrations in neonates from AB− mothers compared to neonates from AB+ mothers are shown to the left of the center. B) Cytokines/chemokines with higher concentrations in neonates from MAR+ mothers are shown to the right of the center, and those with higher concentrations in neonates from AB− mothers are shown to the left of the center. C) Cytokines/chemokines with higher concentrations in neonates from MAR+ are shown to the right of the center, and those with higher concentrations in neonates from AB+ mothers are shown to the left of the center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*).

Effect size comparisons between neonates from AB+, AB−, and MAR+ mothers

As with the maternal samples, we also assessed trends in the neonatal cytokine/chemokine differences by examining their effect sizes. We noted lower levels (d < −0.15) of the chemokines 6CKINE (CCL21), which is involved in dendritic cell chemotaxis, eotaxin-3 (CCL26), Fractalkine (CX3CL1), which is involved in monocyte chemoattraction, stromal-derived factor (SDF)-1A/B that attracts B-cell precursors to the bone marrow, MIP-3α, and thymus-expressed chemokine (TECK/CCL25) in the AB+ group compared to the AB− group (Fig. 2A).

We did not observe any statistically significant differences in neonatal cytokine and chemokine levels between neonates from MAR+ women and either AB+ or AB− groups. However, we observed lower neonatal concentrations of B-cell attracting chemokine (BCA)-1, eotaxin-1, I-309 (CCL1), monocyte chemoattractant protein (MCP)-4, the regulatory cytokine IL-10, and the inflammatory cytokine TNFα and higher concentrations (d > 0.15) of the CD4+ T-cell chemoattractant IL-16 in the MAR+ group compared with the AB− group (Fig. 2B). Similarly, the MAR+ group had lower levels of neonatal MCP-4 and higher levels of neonatal IL-16 in addition to higher levels of the B-cell growth and differentiation cytokine, IL-13, compared to the AB+ group (Fig. 2C).

Stratified analysis by intellectual disability status

When we examined AB+, MAR+, and AB− samples stratified by ID status, we found that only in mothers of neonates later diagnosed with autism+ID were maternal IL-12p40 levels significantly higher in AB+ mothers compared to AB− mothers and, separately, to MAR+ mothers (Fig. 3A). In contrast, in mothers of neonates later diagnosed with autism-noID, MAR+ mothers had significantly higher levels of IL-12p40 than both AB− and AB+ mothers (Fig. 4B and C). In addition, while IL-7 showed significant group differences in our overall model, when we stratified by ID status, we found that this significance was driven by the autism-noID dyads and that group differences were not significant in the autism+ID maternal comparisons (Fig. 3 and Fig. 4).

Fig. 3.

Fig. 3

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in mid-gestational maternal cytokine and chemokine concentrations stratified for autism with intellectual disability (autism+ID) using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in AB+ mothers when compared to AB− mothers are shown to the right of the center. Those with higher concentrations in AB− mothers compared to AB+ mothers are shown to the left of center. B) Cytokine/chemokines with higher concentrations in MAR+ mothers are shown to the right of the center, and those with higher concentrations in AB− mothers are shown to the left of center. C) Cytokine/chemokines with higher concentrations in MAR+ are shown to the right of the center, and those with higher concentrations in AB+ mothers are shown to the left of center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*). Autism+ID: N = 118 AB+, n = 84 AB−, n = 26 MAR+.

Fig. 4.

Fig. 4

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in mid-gestational maternal cytokine and chemokine concentrations stratified for autism without intellectual disability (autism-noID) using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in AB+ mothers when compared to AB− mothers are shown to the right of the center. Those with higher concentrations in AB− mothers compared to AB+ mothers are shown to the left of the center. B) Cytokine/chemokines with higher concentrations in MAR+ mothers are shown to the right of the center, and those with higher concentrations in AB− mothers are shown to the left of the center. C) Cytokine/chemokines with higher concentrations in MAR+ are shown to the right of the center, and those with higher concentrations in AB+ mothers are shown to the left of the center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*). Autism-noID: N = 113 AB+, and n = 80 AB−, n = 17 MAR+.

In the neonates, after stratifying by ID status, we found no significant differences across maternal AB groups in neonatal cytokine/chemokine levels among offspring with autism+ID (Fig. 5). However, MCP-4 was significantly lower in the MAR+ group compared to both AB− and AB+ groups only among offspring with autism-noID (Fig. 6B and C).

Fig. 5.

Fig. 5

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in mid-gestational newborn cytokine and chemokine concentrations for autism with intellectual disability (autism+ID) using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in newborns from AB+ mothers when compared to newborns from AB− mothers are shown to the right of the center. Those with higher concentrations in newborns from AB− mothers compared to newborns from AB+ mothers are shown to the left of the center. B) Cytokine/chemokines with higher concentrations in newborns from MAR+ mothers are shown to the right of the center, and those with higher concentrations in newborns from AB− mothers are shown to the left of the center. C) Cytokine/chemokines with higher concentrations in newborns from MAR+ are shown to the right of the center, and those with higher concentrations in newborns from AB+ mothers are shown to the left of the center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*). Autism+ID: N = 118 AB+, n = 84 AB−, n = 26 MAR+.

Fig. 6.

Fig. 6

Open in a new tab

Estimated effect sizes for adjusted pairwise group differences in mid-gestational maternal cytokine and chemokine concentrations stratified by autism without intellectual disability (autism-noID) using multiple linear regression models. A) Cytokine/chemokines with higher concentrations in newborns from AB+ mothers when compared to newborns from AB− mothers are shown to the right of the center. Those with higher concentrations in newborns from AB− mothers compared to newborns from AB+ mothers are shown to the left of the center. B) Cytokine/chemokines with higher concentrations in newborns from MAR+ mothers are shown to the right of the center, and those with higher concentrations in newborns from AB− mothers are shown ito the left of the center. C) Cytokine/chemokines with higher concentrations in newborns from MAR+ are shown to the right of the center, and those with higher concentrations in newborns from AB+ mothers are shown to the left of the center. Comparisons with significant group differences (P < 0.05) are marked with asterisks (*). Autism-noID: N = 113 AB+, and n = 80 AB−, n = 17 MAR+.

Discussion

Maternal inflammation, autoimmunity, and dysregulation of the early postnatal cytokine and chemokine environment have been independently associated with a neurodevelopmental disorder diagnosis in children. However, examination of the relationship between autoantibodies specific to proteins in the developing brain and associated with autism (MAR ABs) and cytokine/chemokine profiles has not yet been addressed. While questions about the maternal–fetal interface remain, this study provides insight into how the different arms of the maternal immune system might influence each other and potentially impact the immune development of their offspring. Our results demonstrate that the presence of MAR ABs is associated with distinct maternal and neonatal cytokine/chemokine profiles and that mothers with MAR autism-specific patterns of ABs (MAR+) had a unique cytokine/chemokine profile compared to AB+ and AB− mothers.

Compared to AB− and MAR+ mothers, AB+ mothers had significantly higher levels of the T-cell and B-cell development cytokine IL-7, which has been shown to promote a proinflammatory environment during pregnancy. This condition has been associated with pregnancy loss, and inflammation that has negative impacts on offspring neurodevelopment (Wu et al. 2016; Vilsmaier et al. 2021).

Due to our limited sample sizes, we also considered changes in effect sizes to examine trending but non-significant differences. In doing so, we found that AB+ mothers had higher levels of the type I helper T cell (Th1) cytokines interferon-gamma (IFNγ) and IL12-p40 and higher levels of the Th2 cytokine IL-13 than AB− mothers. This suggests that mothers with ABs to known MAR proteins but not a MAR-specific pattern could have excess T-cell activation in combination with inflammatory signals that, combined with the increased IL-7, might negatively impact the developing fetus.

MAR+ mothers showed a lower level of the Th2 cytokine IL-4 than AB− and AB+ mothers. IL-4 typically helps maintain the type II helper T cell (Th2) phenotype often associated with pregnancy, protecting the immunologically foreign fetus from the maternal immune response (McFadden et al. 2015). Therefore, a decrease in the levels of maternal IL-4 in MAR+ mothers could indicate a skewing toward a more type I helper T cell (Th1) or inflammatory T-cell environment. This is further supported by the elevated IFNγ concentrations in MAR+ mothers compared to AB− mothers and higher serum concentrations of the inflammatory cytokines IL-1β and TNFα compared to AB+ mothers. Skewing toward a Th1 cell phenotype during pregnancy has been linked to an increased rate of fetal death and spontaneous abortion, and a gestational increase in the Th1-type cytokine IFNγ has been previously associated with an increased risk for offspring development of autism (Goines et al. 2011; Wang et al. 2020). Changes to the Th1/Th2 dichotomy during gestation in MAR+ mothers could disrupt the balance crucial for fetal protection from the maternal immune response during pregnancy.

There is research demonstrating that in utero, immune system exposures could have a life-long impact, thereby influencing many non-infectious diseases related to immune dysfunction, such as asthma and autoimmunity. In healthy pregnancies, multiple overlapping mechanisms exist to maintain tolerance at the maternal–fetal interface (Erlebacher 2013; Vento-Tormo et al. 2018). There is also strong evidence suggesting the fetal innate immune system can be “trained” during pregnancy, by which maternal immune activation induces changes to the function of the fetal innate immune system with the potential for life-long impacts (Levy and Wynn 2014; Netea et al. 2016; Barrat et al. 2019; Csaba 2020; Netea et al. 2020; Eades et al. 2022; Dominguez-Andres et al. 2023). The immune response of the offspring develops during gestation and the early neonatal period, although maturation is not fully complete until months after birth. Furthermore, in neonates, the innate immune response is the main defense mechanism against infection as the adaptive immune response becomes more fully developed with postnatal exposures (Beloosesky et al. 2010).

When we examined the potential effects of maternal ABs on the offspring’s immune profile at birth, neonates from AB+ mothers had significantly lower levels of proinflammatory chemokines eotaxin-1, MIP-1α, as well as significantly lower IL-12p70 and IL-13 compared to neonates from AB− mothers, which could have implications on the developing brain. IL-13 can induce T and B cells and microglia to produce eotaxin-1 (Nazarinia et al. 2022). While the function of eotaxin-1 in the developing brain is not entirely clear, microglia and other cells in the brain, including neurons, express the receptor for eotaxin-1 (Sirivichayakul et al. 2019). Thus, a decrease in IL-13 could lead to a subsequent lack of eotaxin-1 in the brain. While some studies have reported that elevated eotaxin-1 levels are associated with impaired neurogenesis (Teixeira et al. 2018), in a rodent model of multiple sclerosis, it was found that eotaxin-1 yielded a protective effect and promoted an ant-inflammatory environment (Adzemovic et al. 2012). Given the maternal inflammation observed in AB+ mothers, a lack of eotaxin-1 in their neonates may be more detrimental due to a lack of compensatory protection from maternal inflammation.

Thus, the lower cytokine and chemokine levels in the neonate, including the proinflammatory molecules IL-12p70 and MIP-1α, could be a response to the increased inflammatory and T-cell signaling molecules in the AB+ mothers.

The above findings are further supported by the trending differences we observed in the neonatal immune profile, in which neonates from AB+ mothers also trended toward having lower levels of several chemokines, such as 6CKINE, fractalkine, eotaxin-3, and TECK, compared to the neonates from AB− mothers. Lower levels of 6CKINE, which is essential for immune cell homing to lymphoid tissues, and the T-cell chemokines eotaxin-3 and TECK could suggest a deviation in the neonatal immune function, while decreases in fractalkine could suggest impaired neurodevelopment as it has been previously show to work in concert with its microglial receptor, CX3CR1, to mediate synapse development (Ferro et al. 2021).

Although we observed no significantly different levels of cytokines/chemokines in neonates from MAR+ mothers compared to neonates from AB− and AB+ mothers after adjusting for covariates, we observed several notable differences when assessing their effect sizes. Neonates from MAR+ mothers had lower levels of the chemokines I-309 (CCL1) and eotaxin-1, as well as IL-4 when compared to neonates from AB− mothers, all of which are considered to be involved in the Th2 immune response (Miller and Krangel 1992; Teran et al. 1999). Notably, when we compared the effect sizes of neonates from AB+ mothers to those of neonates from MAR+ mothers, we found that the MAR+ group had higher levels of the proinflammatory cytokine IL-16, which preferentially activates Th1 cells (Wilson et al. 2004). It has been previously documented that during pregnancy, similarly to maternal immune cells, neonatal immune cells are biased toward a Th2 response (McFadden et al. 2015). However, when considered in conjunction with the decrease in levels of IL-4 in the MAR+ mothers, these data suggest that the lack of the Th2 phenotype noted in MAR+ mothers during pregnancy might imprint this immune phenotype onto the developing immune system of the offspring (lower IL-4), as detected in the early postnatal period (Semmes et al. 2021; Smolen et al. 2021). This suggests that skewing away from the traditional neonatal Th2 phenotype is unique to neonates from MAR+ mothers in the EMA study population. While these differences did not reach statistical significance due to our modest sample sizes, especially in the MAR+ group, studies with increased sample sizes are underway to explore these findings further.

Our stratified analyses highlighted several differences in maternal and neonatal cytokines/chemokines based on maternal AB group between those neonates later diagnosed with autism with or without ID. More specifically, MAR+ mothers whose offspring were later diagnosed with autism-noID had significantly higher levels of the Th1 associated cytokine IL-12p40 during gestation than AB− and AB+ mothers. However, when we examined the serum levels of MAR+ mothers of neonates later diagnosed with autism+ID, the opposite was true, where MAR+ mothers had significantly lower gestational levels of IL-12p40 than AB+ mothers.

This finding is supported by previous studies that found that elevated IL-12p40 was associated with children who received milder autism diagnoses (Keqin et al. 2023).

Similarly, in the neonates from these mothers, we identified that MCP-4 was significantly lower in neonates from MAR+ mothers who were later diagnosed with autism-noID but not autism+ID. Although the exact role of MCP-4 in the central nervous system is currently unknown, elevated levels of the related chemokine MCP-1 have been associated with more severe behavioral scores in children with autism (Ashwood et al. 2011). These results suggest an underlying heterogeneity both within the autism population and based on AB status, in addition to specific cytokine/chemokine profiles that may contribute to the ID outcome.

While there remains active investigation regarding how changes in the maternal immune system directly impact the fetus, this study provides insight into the associations between MAR autism-specific maternal AB production and the gestational and neonatal cytokine and chemokine profiles. MAR autism has been well documented in various human populations and addressed as a mechanism for altered neurodevelopment in preclinical studies (Jones et al. 2020; Ramirez-Celis et al. 2021; Bruce et al. 2023). Although it is unclear why some women produce autoantibodies to these targets, previous studies have identified that only reactivity to certain epitopes of the identified proteins yields pathogenicity (Edmiston et al. 2018; Ramirez-Celis et al. 2019). One reason for this could be due to maternal exposure to commonly hidden or “cryptic” epitopes of the protein that are only visible to the immune system during cases of inflammation and cell death (Root-Bernstein and Fairweather 2014). However, antigen-specific reactivity, in other words, the protein target and/or specific region of the protein, could be one explanation for the MAR-specific cytokine/chemokine changes we observe in the maternal and neonatal cytokines and chemokines.

Our results should be considered with the following limitations: First, we only measured maternal ABs to the antigens associated with MAR autism. Mothers could have other antibodies reactive to fetal antigens not measured in this study population. Given the cross-sectional nature of this study, we are also unable to determine if maternal AB status directly influences the maternal cytokine and chemokine profile or if the cytokine/chemokine profile increases susceptibility to autoimmunity and, thus, the generation of ABs to the proteins of interest. Second, it remains unclear how the dysregulation of maternal cytokines and chemokines impacts the placental cytokine/chemokine production, which could directly affect the fetal compartment (Dealtry et al. 2000; Raghupathy 2013). As we only measured the maternal cytokines/chemokines at one mid-gestational timepoint, we also have no insight into how these may have changed throughout pregnancy. Current studies are underway to address this issue. Similarly, it is important to consider that we assessed only the neonates’ peripheral cytokine/chemokine profiles via the newborn bloodspots. Although this can yield some insight into cytokine/chemokine changes in the brain, it does not fully elucidate potential brain-specific changes. Therefore, while we may be able to hypothesize implications for neurodevelopment, we cannot make definitive conclusions about the changes ongoing in the brain due to the altered peripheral cytokine/chemokine levels. As we limited our analyses to children diagnosed with autism and their mothers in this pilot examination, our findings may not be generalizable to a broader population. In this study, we only had 43 MAR+ mothers available to include in the analysis. Thus, some observed trends did not reach statistical significance, and we used effect sizes to assess the magnitude of the group differences even in the absence of statistical significance. Furthermore, due to the limited sample size, we did not control for multiple comparisons in this analysis of maternal and neonatal cytokine/chemokine levels. Therefore, we cannot rule out the possibility of a higher type I error. Thus, our findings should be regarded as tentative, but given the potential broader implications, we hope they will provide a basis for larger cohort studies. A larger sample size of MAR+ mothers is needed to examine the relationship between specific MAR patterns and maternal and neonatal cytokine/chemokine profiles more thoroughly and with adequate adjustment for multiple comparisons.

In summary, this novel study provides early insights into the relationships between the maternal gestational immune profile and the fetal immune profile. As we further explore how changes in the maternal immune system during gestation can have profound, lasting impacts on offspring outcomes, expanding our knowledge of these relationships will allow us to elucidate the mechanisms by which maternal ABs and cytokines/chemokines impact the developing fetus. Future studies may be aimed at developing intervention strategies to protect the fetus from aberrant maternal immune molecules, including autoantibodies and dysregulated immune signaling molecules. Additional questions could address when, how, and why maternal autoantibodies against the MAR-specific proteins develop and if skewing of the maternal cytokines and chemokines plays a role in the MAR autoimmune sequela.

Supplementary Material

EMA_ABSupp_TablesandFigs_FinalforUpload_021524_bhae082
ema_absupp_tablesandfigs_finalforupload_021524_bhae082.docx (58.2KB, docx)

Acknowledgments

We would like to thank the families that participated in the EMA study and the staff members at the Kaiser Permanente Division of Research, Autism Research Program. The findings and conclusions here are those of the authors and do not necessarily represent the official position of the California Department of Public Health. All authors read and approved the final manuscript.

Contributor Information

Janna McLellan, Department of Internal Medicine, Division of Rheumatology, Allergy, and Clinical Immunology, University of California Davis, 451 Health Sciences Drive, Suite 6505C, Davis, CA 95616, United States.

Lisa A Croen, Division of Research, Kaiser Permanente Northern California, 2000 Broadway, Oakland, CA 94612, United States.

Ana-Maria Iosif, Department of Public Health Sciences, Division of Biostatistics, University of California Davis, Medical Sciences 1C, Davis, CA, 95616, United States.

Paul Ashwood, MIND Institute, University of California Davis, 2805 Wet Lab Building, Sacramento, CA 95817, United States; Department of Medical Microbiology and Immunology, University of California Davis, 3146 One Shields Avenue, Tupper Hall, Davis, CA 95616, United States.

Cathleen Yoshida, Division of Research, Kaiser Permanente Northern California, 2000 Broadway, Oakland, CA 94612, United States.

Kimberly Berger, Sequoia Foundation, 741 Addison Suite B, Berkeley, CA 94710, United States.

Judy Van de Water, Department of Internal Medicine, Division of Rheumatology, Allergy, and Clinical Immunology, University of California Davis, 451 Health Sciences Drive, Suite 6505C, Davis, CA 95616, United States; MIND Institute, University of California Davis, 2805 Wet Lab Building, Sacramento, CA 95817, United States.

Author contributions

Janna McLellan (Conceptualization, Formal analysis, Writing—original draft, Writing—review & editing), Lisa Croen (Conceptualization, Data curation, Funding acquisition, Writing—review & editing), Ana-Maria Iosif (Formal analysis, Writing—review & editing), Paul Ashwood (Writing—review & editing), Cathleen Yoshida (Conceptualization, Data curation, Formal analysis), Kimberly Berger (Writing—review & editing), and Judy Van de Water (Conceptualization, Data curation, Funding acquisition, Investigation, Project administration, Writing—review & editing).

Funding

This work was supported by the National Institutes of Health through awards R01ES016669 (EMA study), P50HD103526, and P50MH106438.

 

Conflict of interest statement: None declared.

Ethics approval and consent to participate

Study activities were approved by the Institutional Review Board at Kaiser Permanente, the State Committee for the Protection of Human Subjects, and the Institutional Review Board at the University of California, Davis. Informed consent was not required for the EMA study; consent forms for screening programs were distributed during blood collection, which stipulated that specimens and results from prenatal and neonatal testing could be used for research purposes, given IRB approval.

References

  1. Adzemovic MZ, Ockinger J, Zeitelhofer M, Hochmeister S, Beyeen AD, Paulson A, Gillett A, Thessen Hedreul M, Covacu R, Lassmann H, et al. Expression of Ccl11 associates with immune response modulation and protection against neuroinflammation in rats. PLoS One. 2012:7(7):e39794. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Allswede DM, Yolken RH, Buka SL, Cannon TD. Cytokine concentrations throughout pregnancy and risk for psychosis in adult offspring: a longitudinal case-control study. Lancet Psychiatry. 2020:7(3):254–261. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Angkustsiri K, Fussell JJ, Bennett A, Schauer J, Ramirez-Celis A, Hansen RL, Van de Water J. Pilot study of maternal autoantibody-related autism. J Dev Behav Pediatr. 2022:43(8):465–471. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Ashwood P, Krakowiak P, Hertz-Picciotto I, Hansen R, Pessah IN, Van de Water J. Associations of impaired behaviors with elevated plasma chemokines in autism spectrum disorders. J Neuroimmunol. 2011:232(1–2):196–199. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Barrat FJ, Crow MK, Ivashkiv LB. Interferon target-gene expression and epigenomic signatures in health and disease. Nat Immunol. 2019:20(12):1574–1583. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Beloosesky R, Maravi N, Weiner Z, Khatib N, Awad N, Boles J, Ross MG, Itskovitz-Eldor J. Maternal lipopolysaccharide-induced inflammation during pregnancy programs impaired offspring innate immune responses. Am J Obstet Gynecol. 2010:203(185):e181–e184. [DOI] [PubMed] [Google Scholar]
  7. Braunschweig D, Ashwood P, Hertz-Picciotto I, Croen L, Hansen R, Pessah IN, Van de Water J. Maternal serum antibodies to fetal brain in autism. Neruotoxicology. 2008:29:226–231. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Bruce MR, ACM C, Grant S, McLellan J, Ku K, Chang C, Bachman A, Matson M, Berman RF, Maddock RJ, et al. Altered behavior, brain structure, and neurometabolites in a rat model of autism-specific maternal autoantibody exposure. Mol Psychiatry. 2023:28(5):2136–2147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cheng SB, Sharma S. Interleukin-10: a pleiotropic regulator in pregnancy. Am J Reprod Immunol. 2015:73(6):487–500. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Ciobanu AM, Dumitru AE, Gica N, Botezatu R, Peltecu G, Panaitescu AM. Benefits and risks of IgG transplacental transfer. Diagnostics (Basel). 2020:10(8):10. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Croen LA, Braunschweig D, Haapanen L, Yoshida CK, Fireman B, Grether JK, Kharrazi M, Hansen RL, Ashwood P, Van de Water J. Maternal mid-pregnancy autoantibodies to fetal brain protein: the early markers for autism study. Biol Psychiatry. 2008:64(7):583–588. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Csaba G. Possible contribution of trained immunity in faulty hormonal imprinting and DOHaD: review and hypothesis. Acta Microbiol Immunol Hung. 2020:67(3):143–147. [DOI] [PubMed] [Google Scholar]
  13. Dammann O, Leviton A. Maternal intrauterine infection, cytokines, and brain damage in the preterm newborn. Pediatr Res. 1997:42(1):1–8. [DOI] [PubMed] [Google Scholar]
  14. Dealtry GB, O'Farrell MK, Fernandez N. The Th2 cytokine environment of the placenta. Int Arch Allergy Immunol. 2000:123(2):107–119. [DOI] [PubMed] [Google Scholar]
  15. Djuardi Y, Wibowo H, Supali T, Ariawan I, Bredius RG, Yazdanbakhsh M, Rodrigues LC, Sartono E. Determinants of the relationship between cytokine production in pregnant women and their infants. PLoS One. 2009:4(11):e7711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Dominguez-Andres J, Dos Santos JC, Bekkering S, Mulder WJM, Meer JWM, Riksen NP, Joosten LAB, Netea MG. Trained immunity: adaptation within innate immune mechanisms. Physiol Rev. 2023:103(1):313–346. [DOI] [PubMed] [Google Scholar]
  17. Eades L, Drozd M, Cubbon RM. Hypoxia signalling in the regulation of innate immune training. Biochem Soc Trans. 2022:50(1):413–422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Edmiston E, Jones KL, Vu T, Ashwood P, Van de Water J. Identification of the antigenic epitopes of maternal autoantibodies in autism spectrum disorders. Brain Behav Immun. 2018:69:399–407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Equils O, Kellogg C, McGregor J, Gravett M, Neal-Perry G, Gabay C. The role of the IL-1 system in pregnancy and the use of IL-1 system markers to identify women at risk for pregnancy complications. Biol Reprod. 2020:103(4):684–694. [DOI] [PubMed] [Google Scholar]
  20. Erlebacher A. Immunology of the maternal-fetal interface. Annu Rev Immunol. 2013:31(1):387–411. [DOI] [PubMed] [Google Scholar]
  21. Estes ML, McAllister AK. Maternal immune activation: implications for neuropsychiatric disorders. Science. 2016:353(6301):772–777. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Ferro A, Auguste YSS, Cheadle L. Microglia, cytokines, and neural activity: unexpected interactions in brain development and function. Front Immunol. 2021:12:703527. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Goines PE, Croen LA, Braunschweig D, Yoshida CK, Grether J, Hansen R, Kharrazi M, Ashwood P, Van de Water J. Increased midgestational IFN-gamma, IL-4 and IL-5 in women bearing a child with autism: a case-control study. Mol Autism. 2011:2(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Heuer LS, Croen LA, Jones KL, Yoshida CK, Hansen RL, Yolken R, Zerbo O, DeLorenze G, Kharrazi M, Ashwood P, et al. An exploratory examination of neonatal cytokines and chemokines as predictors of autism risk: the early markers for autism study. Biol Psychiatry. 2019:86(4):255–264. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Hon KL, Leung AK. Autoimmune Dis 2012. 2012:301274, Neonatal lupus erythematosus, 1, 6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  26. Jones KL, Croen LA, Yoshida CK, Heuer L, Hansen R, Zerbo O, GN DL, Kharrazi M, Yolken R, Ashwood P, et al. Autism with intellectual disability is associated with increased levels of maternal cytokines and chemokines during gestation. Mol Psychiatry. 2017:22(2):273–279. [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Jones KL, Pride MC, Edmiston E, Yang M, Silverman JL, Crawley JN, Van de Water J. Autism-specific maternal autoantibodies produce behavioral abnormalities in an endogenous antigen-driven mouse model of autism. Mol Psychiatry. 2020:25(11):2994–3009. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Keqin, Liu YH, Zhu Y, Zhao Y, Kong X. The role of maternal immune activation in immunological and neurological pathogenesis of autism. J Neurorestoratology. 2023:11(1):8–20. [Google Scholar]
  29. Kim DH, Krakowiak P, Meltzer A, Hertz-Picciotto I, Van de Water J. Neonatal chemokine markers predict subsequent diagnosis of autism spectrum disorder and delayed development. Brain Behav Immun. 2022:100:121–133. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. DHJ K, Iosif AM, Ramirez-Celis A, Ashwood P, Ames JL, Lyall K, Berger K, Croen LA, Van de Water J. Neonatal immune signatures differ by sex regardless of neurodevelopmental disorder status: macrophage migration inhibitory factor (MIF) alone reveals a sex by diagnosis interaction effect. Brain Behav Immun. 2023:111:328–333. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Levy O, Wynn JL. A prime time for trained immunity: innate immune memory in newborns and infants. Neonatology. 2014:105(2):136–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. McFadden JP, Thyssen JP, Basketter DA, Puangpet P, Kimber I. T helper cell 2 immune skewing in pregnancy/early life: chemical exposure and the development of atopic disease and allergy. Br J Dermatol. 2015:172(3):584–591. [DOI] [PubMed] [Google Scholar]
  33. Meyyazhagan A, Kuchi Bhotla H, Pappuswamy M, Tsibizova V, Al Qasem M, Di Renzo GC. Cytokine see-saw across pregnancy, its related complexities and consequences. Int J Gynaecol Obstet. 2023:160(2):516–525. [DOI] [PubMed] [Google Scholar]
  34. Miller MD, Krangel MS. The human cytokine I-309 is a monocyte chemoattractant. Proc Natl Acad Sci USA. 1992:89(7):2950–2954. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Nakagawa S, Cuthill IC. Effect size, confidence interval and statistical significance: a practical guide for biologists. Biol Rev Camb Philos Soc. 2007:82(4):591–605. [DOI] [PubMed] [Google Scholar]
  36. Nazarinia D, Behzadifard M, Gholampour J, Karimi R, Gholampour M. Eotaxin-1 (CCL11) in neuroinflammatory disorders and possible role in COVID-19 neurologic complications. Acta Neurol Belg. 2022:122(4):865–869. [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Netea MG, Joosten LA, Latz E, Mills KH, Natoli G, Stunnenberg HG, O'Neill LA, Xavier RJ. Trained immunity: a program of innate immune memory in health and disease. Science. 2016:352(6284):aaf1098. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Netea MG, Dominguez-Andres J, Barreiro LB, Chavakis T, Divangahi M, Fuchs E, Joosten LAB, Meer JWM, Mhlanga MM, Mulder WJM, et al. Defining trained immunity and its role in health and disease. Nat Rev Immunol. 2020:20(6):375–388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Pacheco Y, Barahona-Correa J, Monsalve DM, Acosta-Ampudia Y, Rojas M, Rodriguez Y, Saavedra J, Rodriguez-Jimenez M, Mantilla RD, Ramirez-Santana C, et al. Cytokine and autoantibody clusters interaction in systemic lupus erythematosus. J Transl Med. 2017:15(1):239. [DOI] [PMC free article] [PubMed] [Google Scholar]
  40. Panaitescu AM, Nicolaides K. Fetal goitre in maternal Graves’ disease. Acta Endocrinol (Buchar). 2018:14(1):85–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Pyzik M, Sand KMK, Hubbard JJ, Andersen JT, Sandlie I, Blumberg RS. The neonatal Fc receptor (FcRn): a misnomer? Front Immunol. 2019:10:1540. [DOI] [PMC free article] [PubMed] [Google Scholar]
  42. Raghupathy R. Cytokines as key players in the pathophysiology of preeclampsia. Med Princ Pract. 2013:22(Suppl 1):8–19. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Ramirez-Celis A, Becker M, Nuño M, Schauer J, Aghaeepour N, Van de Water J. Risk assessment analysis for maternal autoantibody-related autism (MAR-ASD): a subtype of autism. Mol Psychiatry. 2021:26(5):1551–1560. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Ramirez-Celis A, Croen LA, Yoshida CK, Alexeeff SE, Schauer J, Yolken RH, Ashwood P, Van de Water J. Maternal autoantibody profiles as biomarkers for ASD and ASD with co-occurring intellectual disability. Mol Psychiatry. 2022:27(9):3760–3767. [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Ramirez-Celis A, Edmiston E, Schauer J, Vu T, Van de Water J. Peptides of neuron specific enolase as potential ASD biomarkers: from discovery to epitope mapping. Brain Behav Immun. 2019:84:200–208. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Ravindran E, Arashiki N, Becker LL, Takizawa K, Levy J, Rambaud T, Makridis KL, Goshima Y, Li N, Vreeburg M, et al. Monoallelic CRMP1 gene variants cause neurodevelopmental disorder. Elife. 2022:11:e80793. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Roberts CT, White CA, Wiemer NG, Ramsay A, Robertson SA. Altered placental development in interleukin-10 null mutant mice. Placenta. 200324 Suppl A:S94-99:24:S94–S99. [DOI] [PubMed] [Google Scholar]
  48. Root-Bernstein R, Fairweather D. Complexities in the relationship between infection and autoimmunity. Curr Allergy Asthma Rep. 2014:14(1):407. [DOI] [PMC free article] [PubMed] [Google Scholar]
  49. Saito S. Cytokine cross-talk between mother and the embryo/placenta. J Reprod Immunol. 2001:52(1–2):15–33. [DOI] [PubMed] [Google Scholar]
  50. Semmes EC, Chen J-L, Goswami R, Burt TD, Permar SR, Fouda GG. Understanding early-life adaptive immunity to guide interventions for pediatric health. Front Immunol. 2021:11:595297. [DOI] [PMC free article] [PubMed] [Google Scholar]
  51. Sirivichayakul S, Kanchanatawan B, Thika S, Carvalho AF, Maes M. Eotaxin, an endogenous cognitive deteriorating chemokine (ECDC), is a major contributor to cognitive decline in normal people and to executive, memory, and sustained attention deficits, formal thought disorders, and psychopathology in schizophrenia patients. Neurotox Res. 2019:35(1):122–138. [DOI] [PubMed] [Google Scholar]
  52. Smolen KK, Plotkin AL, Shannon CP, Idoko OT, Pak J, Darboe A, Haren S, Amenyogbe N, Tebbutt SJ, Kollmann TR, et al. Ontogeny of plasma cytokine and chemokine concentrations across the first week of human life. Cytokine. 2021:148:155704. [DOI] [PMC free article] [PubMed] [Google Scholar]
  53. Teixeira AL, Gama CS, Rocha NP, Teixeira MM. Revisiting the role of eotaxin-1/CCL11 in psychiatric disorders. Front Psychiatry. 2018:9:241. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Teran LM, Mochizuki M, Bartels J, Valencia EL, Nakajima T, Hirai K, Schroder JM. Th1- and Th2-type cytokines regulate the expression and production of eotaxin and RANTES by human lung fibroblasts. Am J Respir Cell Mol Biol. 1999:20(4):777–786. [DOI] [PubMed] [Google Scholar]
  55. Vento-Tormo R, Efremova M, Botting RA, Turco MY, Vento-Tormo M, Meyer KB, Park J-E, Stephenson E, Polański K, Goncalves A, et al. Single-cell reconstruction of the early maternal–fetal interface in humans. Nature. 2018:563(7731):347–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Vilsmaier T, Amann N, Lob S, Schmoeckel E, Kuhn C, Zati Zehni A, Meister S, Beyer S, Kolben TM, Becker J, et al. The decidual expression of Interleukin-7 is upregulated in early pregnancy loss. Am J Reprod Immunol. 2021:86(3):e13437. [DOI] [PubMed] [Google Scholar]
  57. Wang W, Sung N, Gilman-Sachs A, Kwak-Kim J. T helper (Th) cell profiles in pregnancy and recurrent pregnancy losses: Th1/Th2/Th9/Th17/Th22/Tfh cells. Front Immunol. 2020:11:2025. [DOI] [PMC free article] [PubMed] [Google Scholar]
  58. Wilson KC, Center DM, Cruikshank WW. The effect of interleukin-16 and its precursor on T lymphocyte activation and growth. Growth Factors. 2004:22(2):97–104. [DOI] [PubMed] [Google Scholar]
  59. Wu L, Li J, Xu HL, Xu B, Tong XH, Kwak-Kim J, Liu YS. IL-7/IL-7R signaling pathway might play a role in recurrent pregnancy losses by increasing inflammatory Th17 cells and decreasing Treg cells. Am J Reprod Immunol. 2016:76(6):454–464. [DOI] [PubMed] [Google Scholar]
  60. Yockey LJ, Iwasaki A. Interferons and proinflammatory cytokines in pregnancy and fetal development. Immunity. 2018:49(3):397–412. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

EMA_ABSupp_TablesandFigs_FinalforUpload_021524_bhae082
ema_absupp_tablesandfigs_finalforupload_021524_bhae082.docx (58.2KB, docx)

Articles from Cerebral Cortex (New York, NY) are provided here courtesy of Oxford University Press

RESOURCES