Epigenetic signatures in children and adolescents at familial high risk: linking early-life environmental exposures to psychopathology

doi:10.21203/rs.3.rs-4722934/v1

Download PDF

Research Article

Epigenetic signatures in children and adolescents at familial high risk: linking early-life environmental exposures to psychopathology

https://doi.org/10.21203/rs.3.rs-4722934/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background This study investigates the relationship between environmental risk factors and severe mental disorders using genome-wide methylation data. Methylation profile scores (MPS) and epigenetic clocks were utilized to analyze epigenetic alterations in a cohort comprising 211 individuals aged 6–17 years. Participants included offspring of schizophrenia (n = 30) and bipolar disorder (n = 82) patients, and a community control group (n = 99). The study aimed to assess differences in MPS indicative of intrauterine stress and epigenetic aging across familial risk groups, and their associations with cognition, prodromal psychotic symptoms, and global functioning through statistical models.

Results Individuals at high familial risk demonstrated significant epigenetic alterations associated with pre-pregnancy maternal overweight/obesity, pre-eclampsia, early preterm birth and higher birth weight (p.adj ≤ 0.001) as well as decelerated epigenetic aging in the Horvath and Hannum epigenetic clocks (p.adj ≤ 0.005). Among offspring of schizophrenia patients, more severe positive and general prodromal psychotic symptoms correlated with MPS related to maternal pre-pregnancy BMI and overweight/obesity (p.adj ≤ 0.008) as well as with accelerated epigenetic aging across all examined epigenetic clocks (p.adj ≤ 0.012).

Conclusions These findings underscore the potential of methylation analysis to quantify persistent effects of intrauterine events and their influence on the onset of psychotic symptoms, particularly in high-risk populations. Further research is essential to elucidate the underlying biological mechanisms during critical early stages of neurodevelopment.

Understanding the etiopathogenesis of severe mental disorders requires untangling the intricate relationship between genetic predisposition and environmental influences. Although both factors play essential roles, neither can fully explain the complex phenotypes observed (1). The study of environmental influences poses significant challenges due to the diverse range of factors involved and their dynamic effects across different life stages. Comprehensive measurements of environmental exposures, referred to as the "exposome", present notable methodological challenges due to their inherent subjectivity and the difficulty in accurately assessing their impact (2). Schizophrenia and bipolar disorder, exemplify this complexity, as their onset and prognosis are strongly affected by environmental events during critical periods of intrauterine neurodevelopment (3, 4). Exploring the biological impact of early-life environmental factors is essential for elucidating the mechanisms underlying the manifestation of mental disorders and for identifying individuals at risk.

Understanding the link between environmental risk factors and the onset of severe mental disorders presents a significant challenge due to the limited knowledge of the biological processes that may mediate this association (5, 6). Epigenetic modifications, defined as DNA changes that do not alter sequence, are promising candidates for studying the biological effects of environmental stressors and the mechanisms by which organisms cope with external inputs that may disrupt physiological homeostasis. Among various epigenetic processes, CpG methylation stands out due to its association with the modulation of gene expression (7) and its responsiveness to environmental inputs, making it essential for understanding the biological effects of these factors. Efforts to identify the risk of severe mental health disorders through methylation data have highlighted methylation abnormalities in genes implicated in the pathophysiology of schizophrenia and bipolar disorder as well as in genes implicated in the immune system and inflammatory responses (8).

Transitioning from candidate gene approaches to genome-wide methodologies offers a broader scope for studying complex phenotypes (9). Various methods exist for investigating genome-wide changes, with the earliest approach involving the selection of CpGs associated with undergoing methylation changes over time. These estimators of biological age are known as epigenetic clocks and can be used to detect discrepancies with the individual’s chronological age. The accelerated aging hypothesis emerged due to the prevalence of age-related comorbidities among individuals with mental disorders (10, 11); however, inconsistent results have been reported regarding epigenetic age acceleration in these patients (12–14), possibly attributed to complexity and variety of mechanisms involved in aging processes (6). Based on findings in epigenome-wide association studies (EWAS), epigenetic profile score (MPS) analysis offers a novel method to summarize the methylation changes associated with a specific condition, such as environmental factors or health conditions. Although their construction resembles that of polygenic risk scores (PRS), two key distinctions emerge: MPS may exhibit variability across tissues and over time (15), and its causal relationship with the environmental factors cannot be straightforwardly presumed (16). Several studies have shown associations between MPS and metabolic, inflammatory and mental health outcomes (17–21).

The objectives of this study stem from previous findings within the study sample, which consists of child and adolescent individuals who are offspring of patients diagnosed with schizophrenia and bipolar disorder. Previous research has shown elevated rates of psychopathology (22–24), an increased genetic predisposition to schizophrenia (25) and epigenetic age deceleration (26) among individuals at familial high risk compared to offspring of community controls. These findings underscore the interconnected nature of familial antecedents of severe mental disorders, biological factors and subthreshold clinical features. The principal objective of this study was to analyze the epigenetic profile of the sample by employing the two aforementioned epigenetic constructs: MPS and epigenetic clocks. First, our goal was to summarize the epigenetic imprint resulting from exposure to stressful prenatal conditions. Second, we aimed to employ epigenetic clocks to quantify asynchronicities between chronological and biological ages, as previously investigated in a subset of this sample (26). We hypothesized that individuals at familial high risk, particularly those reporting more elevated rates of psychopathology subclinical features (i.e., offspring of schizophrenia patients), would exhibit greater MPS and epigenetic age deceleration. Our hypothesis posited that these abnormal epigenetic patterns would be associated with cognition, prodromal psychotic symptoms and the global functioning.

The present study is part of the Bipolar and Schizophrenia Young Offspring Study (BASYS), which is a multicenter, longitudinal, naturalistic study that aims to compare the clinical, neuropsychological, neuroimaging, genetic and epigenetic characteristics of child and adolescent offspring of patients diagnosed with schizophrenia or bipolar disorder and of a community control group. This study was conducted in the Child and Adolescent Psychiatry Departments of two hospitals in Spain: the Hospital Clinic in Barcelona and the Hospital Gregorio Marañón in Madrid. The methodology and the clinical and cognitive characteristics of the sample have been described previously in detail (27).

2.1 Sample

The individuals at familial high risk were identified through their parents, who were recruited from the adult psychiatry units of both hospitals. The inclusion criteria were (a) age between 6 and 17 years and (b) a parent diagnosed with schizophrenia or bipolar disorder. The exclusion criteria were (a) intellectual disability with an impact on functioning and (b) significant head injury or a current medical or neurological condition. Community control parents were recruited through advertisements posted in primary health care centers and other community locations in the same geographical area as the patients. The only inclusion criterion for the offspring of the community controls was an age between 6 and 17 years, while the exclusion criteria were the same as those for the offspring of schizophrenia or bipolar disorder patients plus a family history of psychotic disorders in first- or second-degree relatives. BASYS included 69 offspring of parents with schizophrenia (SZoff), 143 offspring of parents with bipolar disorder (BDoff) and 155 offspring of community controls (CCoff). Given the focus on epigenetic data in this study, only the 211 individuals who had provided biological samples for DNA methylation analyses and had passed the quality controls (30 SZoff, 82 BDoff and 99 CCoff) were included.

2.3 Assessments

2.3.1 Cognitive assessment

The intelligence quotient was assessed using the Spanish version of the Wechsler Intelligence Scale for Children - Fourth Edition (WISC-IV) (28) and Fifth Edition (WISC-V) (29), which evaluates intellectual abilities in children and adolescents aged between 6 and 16 years. The WISC-IV provides four composite scores: the Verbal Comprehension Index, the Perceptual Reasoning Index, the Working Memory Index and the Processing Speed Index. Previous research has shown that the Working Memory Index and Processing Speed Index may be impaired in SZoff (30) and BDoff (31, 32). To avoid the influence of each of these indices on the full-scale IQ, the General Ability Index (GAI), derived from the Verbal Comprehension Index and Perceptual Reasoning Index, was used as an index of cognition (33).

2.3.2 Clinical assessment

A trained psychiatrist or psychologist performed a mental health assessment of all the parents using the Spanish version of the Structured Clinical Interview for DSM-IV Disorders (SCID-I) (34, 35). Parents or primary caregivers were also interviewed about their children. Psychopathology was evaluated by child psychiatrists who were blinded to the parental diagnoses using the Spanish version of the Schedule for Affective Disorders and Schizophrenia for School-Age Children – Present and Lifetime version (K-SADS–PL) (36, 37).

Participants were assessed with the Scale of Prodromal Symptoms (SOPS) within the Structured Interview for Prodromal Symptoms (38). The reliability of the SOPS was calculated by the team members who performed the clinical assessments (Kappa statistic for both the SOPS total score and the subscales > 0.8). The SOPS is a 19-item scale that contains four subscales for positive, negative, disorganization and general symptom constructs. Higher scores indicate more severe prodromal psychotic symptoms.

Measures of global functioning capture the severity of psychotic symptoms and the level of occupational and social functioning with the Children’s Global Assessment Scale (39). This measurement consists of a scale of 1–100, as evaluated by a clinician. Higher scores indicate better global functioning.

2.3.3 Assessment of obstetric complications

Information about obstetric complications (OC) was collected using the Lewis-Murray scale (40). This scale rates 15 OC as absent or definitely present, while 9 of the exposures can also be rated as equivocally present. OC can be grouped into three categories: complications of pregnancy (class A), abnormal fetal growth and development (class B), and difficulties in delivery (class C). All variables were categorized as dichotomous variables (present/absent), considering a positive history of OC when at least one exposure was definitely present.

2.4 Biological samples

Blood samples were collected in EDTA tubes (K2EDTA BD Vacutainer EDTA tubes; Becton Dickinson, Franklin Lakes, New Jersey, USA), and genomic DNA was extracted with the MagNA Pure LC DNA Isolation Kit I and a MagNA Pure LC 2.0 instrument (Roche Diagnostics GmbH, Mannheim, Germany). Saliva samples were collected using an Oragene DNA Saliva Collection Kit (OG-500, DNA Self-Collection Kit, Genotek, Ottawa, Ontario, Canada), and DNA was extracted according to the manufacturer's instructions. DNA concentration and quality were measured spectrophotometrically using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Epsom, Surrey, UK). DNA methylation β-values were obtained at GenomeScan using the Illumina Infinium MethylationEPIC BeadChip Kit (n = 117) and the Illumina Infinium MethylationEPIC v2.0 Kit (n = 94) (Illumina, San Diego, CA, USA).

2.5 Methylation data collection

Raw intensity data (.IDAT) files were generated and parallel bioinformatics analyses were conducted in-house using the Chip Analysis Methylation Pipeline (ChAMP) Bioconductor package (41). We processed methylation data separately for each type of methylation platform and biological sample: MethylationEPIC blood (n = 79) and saliva (n = 38), as well as for MethylationEPIC v2.0 blood (n = 61) and saliva (n = 33). In total, four parallel analyses were conducted for each platform and tissue. The raw .IDAT files were used to load the data into the R environment with the champ.import function, which also enabled the undertaking of probe quality control and removal steps. Probes with weak signals (p < 0.010), cross-reactive probes, non-CpG probes, probes with < 3 beads in at least 5% of the samples per probe, probes that bind to SNP sites and sex chromosomes were considered problematic for the accurate detection of downstream methylation and were therefore removed with the champ.filter function. β-values were normalized using the champ.norm function, specifically with the beta-mixture quantile method (BMIQ function). Next, the singular value decomposition (SVD) method was performed with champ.SVD to assess the amount and significance of the technical batch components in our dataset. Using the champ.runCombat function, ComBat algorithms were applied to correct for slides and arrays (significant components detected by the SVD method). The epigenetic measures discussed in the upcoming sections were derived from methylation sites found in all MethylationEPIC platforms and tissues, totaling 709,670 CpGs available for analysis.

2.2 Methylation profile score calculation

Seven EWAS were selected for the construction of MPS reflecting stressful prenatal conditions. The constructed MPS included pre-conception maternal body-mass index (BMI) (42), pre-conception maternal overweight/obesity (42), hypertensive disorders of pregnancy (43), pre-eclampsia (43), gestational diabetes, early preterm birth (44) and birth weight (45). For their construction, we selected methylation sites that (a) had reported p ≤ 0.050 in the EWAS summary statistics and (b) were quantified and passed the quality control in the study sample. The individual MPS were calculated as the sum of all methylation site methylation values, weighted by the estimated effect associated with the environmental exposure:

$$\:{MPS}_{i}={\sum\:}_{j}^{{m}_{MPS}}\widehat{{\text{b}}_{j}}x\:{CpG}_{ij}$$

where:

$\:i$ represents the subject
j represents the methylation probe
$\:{m}_{MPS}$ represents the total number of probes for the MPS
$\:\widehat{{\text{b}}_{j}}$ represents the estimated effect size for probe $\:j$
$\:{CpG}_{ij}$ represents the methylation value of the probe $\:j$ in the individual $\:i$

The MPS were computed using the package methylscore for R, created by our research group and available on GitHub at the following URL: https://github.com/agonse/methylscore. The MPS were standarized by subtracting the mean and dividing by the standard deviation (SD). Further details about the MPS and the referenece EWAS can be found in Table S1.

2.3 Epigenetic clock calculation

The methylclock R package (46) was used to estimate the intrinsic epigenetic age acceleration (IEAA) for the epigenetic clocks used in the present study: Horvath (47), Hannum (48), Levine (49), PedBE (50) and Wu (51) (Table S2). The detailed procedure for IEAA estimation can be found in our previous study (26).

2.4 Polygenic risk score calculation

The genotyping data was processed through the Michigan Imputation Server (52), adhering to stringent quality control measures and utilizing reference panels tailored to European populations. GWAS summary results were employed to construct PRS for schizophrenia, bipolar disorder, major depressive disorder, neuroticism, intelligence, educational attainment, and cognitive performance. Detailed methods can be found in our previous study (25).

2.5 Statistical analysis

Familial risk group differences in sociodemographic, cognitive, prodromal psychotic symptoms and functioning scales at study entry were calculated by generalized linear mixed-effects models, using family ID as a random effect. The correlation between epigenetic constructs was tested using Pearson’s product-moment correlation.

Differences in epigenetic constructs among the familial risk groups were evaluated using generalized linear mixed-effect models. Family ID was included as a random effect, and the analysis was adjusted for biological tissue, methylation array, and sex. When MPS was a fixed effect, age was also included as a covariate.

The associations between the epigenetic constructs and measures of cognition, prodromal psychotic symptoms (including subscales) and functioning were examined using independent linear mixed-effects models for each familial risk group. Family ID was included as a random effect, and the analysis was adjusted for the same covariates as in the previous analysis. To assess the association between epigenetic constructs and OC in the entire sample, generalized linear mixed-effect models were employed, utilizing the same random effects and covariates.

For the analyses including PRS, a genetic principal component analysis (PCA) was performed to control for population stratification (53) by means of the SNPRelate package. Associations between PRS and epigenetic constructs were examined using linear mixed-effects models, where family ID was treated as a random effect. The analysis was adjusted for familial risk group, biological tissue, methylation array, sex and the first 10 components of the genetic PCA. Additionally, for MPS, age was included as a covariate.

All the analyses were performed in the RStudio version 4.3.1 (54). The variance explained was defined by the marginal pseudo-R² (R²). The reported R² values reflect the variance of the PRS on the outcome variable and were calculated as the difference between the R² of the full model (including the epigenetic constructs and the covariates) minus the model excluding the epigenetic constructs. Multiple testing correction was applied in all the analyses by means of the FDR method, and the threshold of significance of the adjusted p value (p.adj) was set at α < 0.05.

3.1 Descriptive statistics

The sample consisted of 211 children and adolescents, 30 SZoff (14.2%), 82 BDoff (38.9%) and 99 CCoff (46.9%). Table 1 provides information on sociodemographics, cognition, prodromal psychotic symptoms, global functioning and OC as well as a comparison between familial risk groups. Consistent with previous studies in this sample, the SZoff and BDoff participants reported higher prodromal psychotic symptom scores and poorer global functioning as well as a greater proportion of class B OC (abnormal fetal growth and development) (p.adj < 0.05 for all analyses). Additional information on these analyses can be found in Table S3.

Table 1

Sociodemographic data, cognitive measures, prodromal psychotic symptom severity, functioning measures, obstetric complications and comparisons among familial risk groups.
Feature	SZoff n = 30	BDoff n = 82	CCoff n = 99	comparison*
Feature	mean(SD) or n(%)	mean(SD) or n(%)	mean(SD) or n(%)	comparison*
Age	10.4(3.2)	12.6(3.2)	12.8(3.3)
Sex - male	16(53.3%)	37(45.1%)	43(43.4%)
Cognition	96.7(14.1)	105.3(12.5)	107.3(12.7)
Total prodromal psychotic symptoms	7.2(9.6)	4.9(7.7)	1.8(2.4)	SZoff > BDoff > CCoff
positive prodromal psychotic symptoms	1.9(3.2)	1.1(2.2)	0.5(0.9)	SZoff > CCoff; BDoff > CCoff
negative prodromal psychotic symptoms	2.0(3.9)	1.3(2.8)	0.4(0.8)	SZoff > CCoff
disorganized prodromal psychotic symptoms	1.6(1.5)	1.2(1.9)	0.4(0.8)	BDoff > CCoff
general prodromal psychotic symptoms	1.6(3.7)	1.3(2.3)	0.5(1.1)
Global functioning	76.9(12.2)	80.2(10.5)	87.1(7.3)	SZoff < CCoff; BDoff < CCoff
Class A OC present	2(7.1%)	5(6.3%)	4(4.2%)
Class B OC present	6(21.4%)	9(11.4%)	5(5.1%)	SZoff > CCoff; BDoff > CCoff
Class C OC present	6(21.4%)	14(17.7%)	15(15.5%)
^{SZoff: offspring of schizophrenia patients; BDoff: offspring of bipolar disorder patients; CCoff: offspring of community controls; OC: obstetric complications}
^{* p.adj < 0.050}

The correlation analyses revealed strong positive associations within MPS, except pre-pregnancy BMI MPS and the other MPS. Similarly, positive associations were observed among all IEAA. Furthermore, pre-pregnancy BMI MPS was correlated with the Hannum and Levine IEAA, while early preterm birth was negatively correlated with the Levine IEAA (Fig. 1).

3.2 Epigenetic profile of the familial high risk groups

In SZoff, MPS reflecting pre-pregnancy overweight/obesity, pre-eclampsia, early preterm birth and birth weight were greater than in CCoff (p.adj < 0.001 for all analyses). The BDoff group had elevated MPS for pre-pregnancy overweight/obesity, hypertensive disorders of pregnancy, gestational diabetes and higher birth weight, compared with the CCoff group (p.adj < 0.001 for all analyses). Regarding the IEAA, SZoff demonstrated a deceleration in the Horvath epigenetic clock compared to CCoff (p.adj < 0.001) and acceleration compared to BDoff (p.adj = 0.005). BDoff showed a deceleration in Hannum epigenetic clock compared to CCoff (p.adj < 0.001) (Table 2).

Table 2

Comparison of MPS and IEAA among familial risk groups. Significant results are marked in bold.
	Epigenetic construct	SZoff vs CCoff		BDoff vs CCoff		SZoff vs BDoff
	Epigenetic construct	beta	p.adj	beta	p.adj	beta	p.adj
MPS	pre-pregnancy BMI	2.055	0.513	0.347	0.835	-0.078	0.990
	pre-pregnancy overweight/obesity	1.018	0.000	8.642	0.000	-0.365	0.990
	hypertensive disorders of pregnancy	3.690	0.511	9.565	0.000	0.033	0.990
	pre-eclampsia	4.432	0.000	0.579	0.835	-0.248	0.990
	gestational diabetes	9.404	0.551	9.526	0.000	-0.162	0.990
	early preterm birth	0.064	0.000	0.224	0.835	-0.251	0.990
	birth weight	3.778	0.000	10.190	0.000	-0.360	0.990
epigenetic clock	Horvath IEAA	0.100	0.000	0.538	0.922	-2.364	0.005
	Hannum IEAA	0.218	0.967	0.469	0.000	-0.357	0.982
	Levine IEAA	-0.333	0.967	-0.071	0.922	-0.156	0.982
	PedBE IEAA	0.045	0.967	0.069	0.922	-0.027	0.982
	Wu IEAA	0.322	0.967	0.080	0.922	0.138	0.982
^{SZoff: offspring of schizophrenia patients; BDoff: offspring of bipolar disorder patients; CCoff: offspring of community controls; MPS: methylation profile score; IEAA: intrinsic epigenetic age acceleration}

3.3 Association of epigenetic constructs with cognition, prodromal psychotic symptoms and global functioning

Prodromal psychotic symptoms in the SZoff group were positively associated with pre-pregnancy BMI and overweight/obesity MPS (p.adj = 0.023, p.adj = 0.001; respectively) and with accelerated Horvath (p.adj = 0.007), Hannum (p.adj = 0.007), Levine (p.adj = 0.038), PedBE (p.adj = 0.016) and Wu (p.adj = 0.018) IEAA (Table 3). Conversely, no epigenetic construct showed associations with cognition, prodromal psychotic symptoms or global functioning for the BDoff or CCoff groups (p.adj > 0.050 for all analyses).

Table 3

Stratified analysis of the association between MPS and IEAA with measures of cognition, prodromal psychotic symptoms, and global functioning in familial risk groups. Analyses were stratified for the SZoff, BDoff, and CCoff groups.
SZoff		Epigenetic construct	Cognition			Prodromal psychotic symptoms			Global functioning
		Epigenetic construct	beta	R²	p.adj	beta	R²	p.adj	beta	R²	p.adj
	MPS	pre-pregnancy BMI	0.090	0.003	0.676	0.863	0.630	0.023	-0.456	0.351	0.370
		pre-pregnancy overweight/obesity	0.294	0.014	0.629	10.262	0.403	0.001	-0.198	0.084	0.976
		hypertensive disorders of pregnancy	0.584	0.055	0.629	0.154	-0.006	0.984	-0.015	0.068	0.976
		pre-eclampsia	0.424	0.025	0.629	1.630	0.007	0.750	0.021	0.074	0.976
		gestational diabetes	0.462	0.029	0.629	0.033	-0.007	0.984	-0.104	0.078	0.976
		early preterm birth	0.227	0.022	0.629	0.573	0.128	0.078	-0.256	0.042	0.976
		birth weight	0.275	0.010	0.629	9.423	0.127	0.078	-0.052	0.075	0.976
	epigenetic clock	Horvath IEAA	0.004	0.000	0.984	0.543	0.269	0.007	-0.156	0.097	0.907
		Hannum IEAA	0.015	0.000	0.984	0.548	0.345	0.007	-0.145	0.156	0.907
		Levine IEAA	0.006	0.000	0.984	0.394	0.136	0.038	0.017	0.009	0.931
		PedBE IEAA	-0.083	0.006	0.984	0.474	0.207	0.016	-0.081	0.066	0.907
		Wu IEAA	-0.036	0.001	0.984	0.452	0.187	0.018	-0.068	0.045	0.907
BDoff		Epigenetic construct	Cognition			Prodromal psychotic symptoms			Global functioning
		Epigenetic construct	beta	R²	p.adj	beta	R²	p.adj	beta	R²	p.adj
	MPS	pre-pregnancy BMI	0.075	0.001	0.984	-0.092	-0.008	0.953	-0.166	0.000	0.275
		pre-pregnancy overweight/obesity	-0.027	0.011	0.984	0.018	0.010	0.953	-0.303	0.006	0.275
		hypertensive disorders of pregnancy	-0.081	0.001	0.984	-0.071	0.013	0.953	-0.214	-0.004	0.396
		pre-eclampsia	-0.058	0.010	0.984	0.036	0.009	0.953	-0.357	0.008	0.275
		gestational diabetes	-0.005	0.008	0.984	-0.033	0.009	0.953	-0.302	0.012	0.275
		early preterm birth	-0.064	-0.008	0.984	-0.186	0.004	0.953	0.125	-0.004	0.407
		birth weight	-0.072	0.012	0.984	0.013	0.010	0.953	-0.290	0.006	0.284
	epigenetic clock	Horvath IEAA	-0.023	-0.009	0.897	0.007	-0.002	0.953	0.174	0.122	0.342
		Hannum IEAA	-0.014	-0.009	0.897	-0.022	-0.006	0.953	-0.010	-0.007	0.942
		Levine IEAA	0.029	-0.010	0.897	0.178	0.025	0.733	-0.045	-0.015	0.942
		PedBE IEAA	-0.027	-0.012	0.897	0.104	0.036	0.854	-0.128	0.045	0.386
		Wu IEAA	-0.127	0.002	0.897	0.039	0.000	0.953	-0.007	-0.006	0.942
CCoff		Epigenetic construct	Cognition			Prodromal psychotic symptoms			Global functioning
		Epigenetic construct	beta	R²	p.adj	beta	R²	p.adj	beta	R²	p.adj
	MPS	pre-pregnancy BMI	0.307	0.022	0.247	0.188	0.017	0.335	-0.081	-0.005	0.705
		pre-pregnancy overweight/obesity	0.475	0.012	0.247	0.345	0.017	0.335	-0.331	-0.004	0.401
		hypertensive disorders of pregnancy	0.487	0.048	0.247	0.392	0.019	0.335	-0.389	-0.016	0.401
		pre-eclampsia	0.402	0.020	0.300	0.310	0.010	0.335	-0.634	-0.001	0.401
		gestational diabetes	0.628	0.053	0.247	0.253	0.009	0.335	-0.338	0.002	0.401
		early preterm birth	0.236	-0.004	0.247	0.177	0.012	0.335	-0.004	-0.003	0.977
		birth weight	0.431	0.015	0.279	0.305	0.011	0.335	-0.472	0.001	0.401
	epigenetic clock	Horvath IEAA	-0.004	0.002	0.972	-0.103	0.010	0.406	0.150	-0.016	0.388
		Hannum IEAA	-0.053	-0.009	0.807	-0.107	0.010	0.406	0.036	0.003	0.906
		Levine IEAA	-0.091	-0.020	0.807	-0.138	0.016	0.406	0.159	0.043	0.388
		PedBE IEAA	-0.058	-0.002	0.807	-0.177	0.030	0.406	-0.010	-0.001	0.921
		Wu IEAA	-0.092	0.009	0.807	-0.084	0.006	0.417	0.065	-0.009	0.898
^{SZoff: offspring of schizophrenia patients; BDoff: offspring of bipolar disorder patients; CCoff: offspring of community controls; MPS: methylation profile score; BMI: body mass index; IEAA: intrinsic epigenetic age acceleration}

In examining the prodromal psychotic symptom subscales in the SZoff group, pre-pregnancy BMI MPS exhibited an association with positive prodromal psychotic symptoms (p.adj = 0.004) and pre-pregnancy overweight/obesity MPS exhibited an association with positive and general prodromal psychotic symptoms (p.adj < 0.001; p.adj = 0.008; respectively). No other MPS showed associations with prodromal psychotic symptoms. Accelerated Horvath, Hannum, Levine, PedBE and Wu IEAA were associated with positive prodromal psychotic symptoms (p.adj < 0.002 for all analyses) and with general prodromal psychotic symptoms (p.adj < 0.012 for all analyses) (Table 4).

Table 4

Association of epigenetic constructs with prodromal psychotic symptom subscales in the offspring of schizophrenia patients. Significant results are marked in bold.
	Epigenetic construct	Positive prodromal psychotic symptoms			Negative prodromal psychotic symptoms			Disorganized prodromal psychotic symptoms			General prodromal psychotic symptoms
	Epigenetic construct	beta	R²	p.adj	beta	R²	p.adj	beta	R²	p.adj	beta	R²	p.adj
MPS	pre-pregnancy BMI	1.082	0.748	0.004	0.308	0.056	0.524	0.723	0.189	0.086	0.552	0.155	0.183
	pre-pregnancy overweight/obesity	11.049	0.464	0.000	4.468	0.017	0.524	8.464	0.442	0.064	9.621	0.337	0.008
	hypertensive disorders of pregnancy	0.307	-0.003	0.820	-0.907	0.037	0.524	-0.313	-0.030	0.954	0.528	0.002	0.701
	pre-eclampsia	3.228	0.047	0.326	-2.151	0.029	0.524	-0.163	0.000	0.954	2.933	0.039	0.406
	gestational diabetes	-1.253	0.016	0.527	1.111	-0.075	0.524	0.872	0.163	0.858	-0.708	0.003	0.701
	early preterm birth	0.447	0.074	0.245	0.349	0.018	0.524	0.279	0.173	0.790	0.665	0.172	0.060
	birth weight	9.712	0.134	0.110	2.474	0.004	0.618	4.074	0.158	0.790	11.440	0.189	0.060
epigenetic clock	Horvath IEAA	0.677	0.427	0.000	0.057	0.005	0.963	0.409	0.430	0.103	0.626	0.359	0.002
	Hannum IEAA	0.696	0.471	0.000	0.210	0.051	0.963	0.296	0.335	0.176	0.644	0.464	0.002
	Levine IEAA	0.580	0.310	0.002	-0.065	0.002	0.963	0.256	0.328	0.218	0.488	0.213	0.012
	PedBE IEAA	0.604	0.344	0.001	0.028	-0.010	0.963	0.452	0.698	0.103	0.542	0.271	0.005
	Wu IEAA	0.609	0.349	0.001	0.009	-0.004	0.963	0.320	0.582	0.176	0.551	0.280	0.005
^{MPS: methylation profile score; BMI: body mass index; IEAA: intrinsic epigenetic age acceleration}

We conducted a similar analysis within the SZoff group, taking into account whether the affected parent was the mother. We observed consistent associations between MPS and the IEAA, mirroring those identified in the previous analysis. Further details are provided in Table S4.

3.4 Epigenetic constructs association with obstetric complications and polygenic risk scores

Pre-pregnancy BMI MPS showed a positive association with class B OC (p.adj < 0.001) and pre-eclampsia MPS demonstrated a positive association with both class B and class C OC (p.adj < 0.001 for both analyses) (Table S5).

No associations were detected between any of MPS or the IEAA and the PRS indicative of susceptibility to psychiatric disorders, neuroticism or cognition in the entire sample (p.adj > 0.050 for all analyses) (Table S6).

This study explored various aspects of genome-wide methylation patterns in a cohort enriched with offspring of schizophrenia and bipolar disorder patients. We utilized methylation data to create two epigenetic constructs – methylation profile scores and epigenetic clocks. Our aim was to examine these constructs, both as a consequence of exposure to early-life stress and as potential modulators of sub-clinical features in children and asolescents. Indeed, these epigenetic constructs provided evidence of a greater impact of stressful intrauterine events and a delay in biological aging in the offspring of patients with schizophrenia and bipolar disorder. Furthermore, these changes in methylation patterns exhibited specific associations with the manifestation of prodromal psychotic symptoms, although solely in the schizophrenia offspring group. Collectively, the findings of this study contribute to a deeper understanding of epigenetic methylation as a potential biological process linking the impact of environmental factors to the emergence of subthreshold clinical manifestations of mental health disorders.

The groups at familial high risk reported greater epigenetic scores indicative of maternal overweight/obesity, hypertensive disorders of pregnancy, pre-eclampsia, gestational diabetes, early preterm birth and higher birth weight. These altered methylation patterns align with large epidemiological studies that have demonstrated an elevated frequency of perinatal complications in mothers diagnosed with schizophrenia, schizoaffective and bipolar disorders (55–58). Within this cohort, 73.3% of SZoff individuals were born to affected women, and despite adjusting for this confounding factor, we still observed associations of MPS and IEAA with prodromal psychotic symptoms. However, we found no associations between genetic susceptibility to several psychiatric disorders and MPS reflecting intrauterine stress. Research indicates that women with mental disorders are more likely to experience greater incidence and severity of perinatal events due to unhealthy lifestyles and inadequate monitoring of pregnancy (59), rather than due to biological traits related to the disorder itself. In contrast, Ursini and colleagues proposed that genetic variants associated with schizophrenia risk may influence early neurodevelopmental processes through the placental response to stress, suggesting a shared genetic susceptibility for schizophrenia and intrauterine complications (60). These findings suggest that adverse events during pregnancy may be more prevalent in families with mental disorders, independent of the genetic background or the polygenic basis of the disorder. Noticeably, individuals at familial high risk exhibited epigenetic patterns associated with increased birth weight. Although this finding may seem contradictory, given that low birth weight is interpreted as a proxy for unspecific complications during pregnancy that restrict fetal growth (4), we observed a strong positive correlation between MPS in the prenatal environment and birth weight in our sample. The epigenetic patterns in individuals at high familial risk suggest that intrauterine sustained exposure to higher glucose levels, such as in cases of maternal obesity and gestational diabetes, may not only contribute to preterm deliveries but also potentially result in higher birth weight (61–63). Notably, gestational diabetes stands as one of the risk factors with higher odds ratios for schizophrenia (3, 4).

Epigenetic age acceleration, estimated through epigenetic clocks, provides insights into the biological aging pace of individuals exposed to various intrauterine and early-life environmental factors. In the current study, encompassing a larger sample size that included individuals from the previous study, we replicated previous findings indicating deceleration of the Horvath and Hannum epigenetic clocks among individuals at familial high risk (26). Our findings further evidence the complexity and dynamism of aging mechanisms, suggesting not only that each epigenetic clock may capture different aspects of aging, but also their sensitivity to external inputs (6, 64, 65). Notwistanding, the deaccelerations in Horvath and Hannum clocks challenge the accelerated aging hypothesis of schizophrenia, which is often inferred from the higher prevalence of age-related conditions in younger schizophrenia patients (10, 11). A recent review proposed that prenatal events may prompt abnormal fetal development, intricately associated with a slower pace of biological aging in epigenetic clocks estimating chronological age (e.g., Horvath) and a faster pace in those capturing mortality-associated phenotypes (e.g., Levine), ultimately linked to schizophrenia (64). Age acceleration may vary throughout the lifespan (66–68), adding complexity to the understanding of epigenetic aging dynamics, its interplay with early-life stressors and its role in the manifestation of mortality-associated phenotypes (69) and clinical symptoms later in life (70).

Analyses examining the associations between epigenetic constructs and clinical outcomes revealed that pre-pregnancy BMI and overweight/obesity MPS, as well as accelerated epigenetic aging were linked to the manifestation of positive and general prodromal psychotic symptoms exclusively in SZoff individuals. These findings suggest a potential connection between specific prenatal conditions and later-life prodromal psychotic features, mediated by epigenetic alterations and consistent with the developmental hypothesis of schizophrenia (71). Notably, this association appears unique to individuals raised in a family with a parent diagnosed with schizophrenia, emphasizing the interplay of the pre- and postnatal environment in triggering prodromal psychotic symptoms within this specific group. Previous studies have found associations between MPS for C-reactive protein and tobacco smoking with cognitive performence (20, 21, 72). Two studies found increased MPS for schizophrenia in schizophrenia patients (73) and but not with age at onset, clozapine use, cognitive status or global functioning (74). Regarding epigenetic age acceleration, it is intriguing to note that SZoff individuals exhibited deaceleration compared to controls, yet acceleration was associated with more severe prodromal psychotic symptoms. Studies on this subject are scarce, and their results are conflicting, reporting epigenetic age acceleration in schizophrenia patients with more severe prodromal psychotic symptoms or no acceleration in in psychiatric and healthy populations (20, 64, 67, 75–77).

The findings of this study should be interpreted within the scope of its limitations. First, the sample size, particularly when conducting independent analyses for the three study groups, may constrain the statistical power of association analyses (78). Additionally, epigenetic methylation is a dynamic process modulated by factors such as time, environmental conditions and sample tissue (16, 79). Therefore, the range of ages in our sample and the two tissues used to obtain methylation data may have contributed to heterogeneity in the epigenetic constructs. Finally, this study focused solely on prodromal psychotic symptoms and did not encompass prodromal symptoms of other mental conditions, such as affective disorders, thereby limiting its scope. Despite these limitations, we included a study sample of individuals at familial high risk in a critical stage for the development of mental disorders, complemented by a comprehensive battery of clinical assessments. The simultaneous use of these two epigenetic constructs is highly innovative, representing a promising approach to capturing long-lasting epigenetic patterns associated with severe mental disorders. The reference data used for constructing MPS were sourced from publicly available EWAS datasets, which, although unlikely to perfectly match the methylation array, tissue, age, and ethnicity of the study sample, encompass greater sample sizes. Genome-wide epigenetic constructs, particularly MPS, currently lack standardized methods and complementary procedures to optimize the technique, such as the imputation of CpG methylation sites (80). In this study, we implemented a thresholding method for CpG selection, building upon previous studies (19, 81, 82) and publicly shared the code for replication and refinement, in https://github.com/agonse/methylscore.

Genetic and environmental factors critically influence the onset and prognosis of severe mental disorders; however, neither is sufficient. Building upon previous genetic and epigenetic findings in this sample (25, 26), our study suggested that changes of specific methylation patterns may pose as key biological mechanisms linking external stress with the clinical manifestation of these disorders. Epigenetic constructs offer a promising solution to certain limitations of retrospective assessments, such as the Lewis-Murray scale for obstetric complications (2, 83), by quantifying the long-lasting biological repercussions of environmental inputs in peripheral tissues. If validated, methylation constructs could yield novel insights into the etiopathological mechanisms underlying the manifestation of severe mental disorders. Integrating genetic and epigenetic measures could provide a more comprehensive understanding of the dynamic interplay between the genetic architecture of disorders and environmental exposures (15). Ultimately, the integration of epigenetic data into prediction models in personalized psychiatry could enhance the detection of individuals at high risk, thereby facilitating the formulation of preventive policies.

Ethics approval and consent to participate: All procedures contributing to this work complied with the ethical standards of the relevant national and institutional committees on human experimentation and with the Helsinki Declaration of 1975, as revised in 2013. The study was approved by the Ethical Review Board of each participating hospital. Written informed consent was obtained from one of the parents or legal guardians, with the other parent been informed, together with written assent from the participant if 12 years of age or older.

Consent for publication: not applicable

Availability of data and materials: All data supporting the findings of this study are available within the paper and its Supplementary Information. EWAS data used for the calculation of MPS are publicly available and can be downloaded from https://www.ewascatalog.org/.

Competing interests: The authors declare no competing interests.

Funding: This work was supported by the Spanish Ministry of Health, Instituto de Salud Carlos III «Health Research Fund»/ FEDER funds (PI1800976, PI2100330, FORT23/00002_SUGR_G6), the European Commission (grant number 101057529), Fundació Clínic Recerca Biomèdica (Ajut a la Recerca Pons Bartran) and INVESTIGO-AGAUR (Next Generation Funds, Generalitat de Catalunya).

Authors' contributions: AGS, SM, JCF designed the study. AGS, IMS and LJ analyzed the data. AGS, IMS, EdlS, GS, IB, DMM, PG, NR, AMP, LJ, CT, CGR, SM and JCF interpreted the data. AGS and IMS wrote the manuscript. AGS and LJ prepared figures and tables. IMS, EdlS, GS, IB, MDP, SP, DMM and contributed to the datasets used in the study. All authors reviewed and approved the manuscript prior to submission.

Acknowledgments: We are extremely grateful to all participants and their families. We would also like to thank the authors who participated in the development of this manuscript. We are also grateful for the support of Spanish Ministry of Science, Innovation and Universities, Instituto de Salud Carlos III (ISCIII), CIBER - Consorcio Centro de Investigación Biomédica en Red - (CB/07/09/0023), co-financed by the European Union, ERDF Funds from the European Commission, “A way of making Europe” (PI07/00853, PI11/02283, PI15/00810, PI17/00741, PI18/01119, PI1800976, PI20/00344, PI21/00519, PI2100330, PI21/01694, FORT23/00002_SUGR_G6); financed by the European Union - NextGenerationEU (PMP21/00051), Generalitat de Catalunya (Programa Investigo-AGAUR), Madrid Regional Government (S2022/BMD-7216 AGES 3-CM), EU Seventh Framework Program, H2020 Program and Horizon Europe (101057529); the National Institute of Mental Health of the National Institutes of Health, Marató TV3 Foundation (202234-30, 202232-30-31, 202210-10), Fundació Clínic Recerca Biomèdica (Pons Bartran Grant) (FRCB_IPB2-2023, FCRB_PB1_2018), Familia Alonso Foundation and Alicia Koplowitz Foundation.

Robinson N, Bergen SE. Environmental Risk Factors for Schizophrenia and Bipolar Disorder and Their Relationship to Genetic Risk: Current Knowledge and Future Directions. Front Genet. 2021 Jun 28;12:686666.
Guloksuz S, Rutten BPF, Pries LK, Ten Have M, De Graaf R, Van Dorsselaer S, et al. The Complexities of Evaluating the Exposome in Psychiatry: A Data-Driven Illustration of Challenges and Some Propositions for Amendments. Schizophrenia Bulletin. 2018 Oct 17;44(6):1175–9.
Cannon M, Jones PB, Murray RM. Obstetric Complications and Schizophrenia: Historical and Meta-Analytic Review. AJP. 2002 Jul;159(7):1080–92.
Davies C, Segre G, Estradé A, Radua J, De Micheli A, Provenzani U, et al. Prenatal and perinatal risk and protective factors for psychosis: a systematic review and meta-analysis. The Lancet Psychiatry. 2020 May;7(5):399–410.
Butera A, Smirnova L, Ferrando‐May E, Hartung T, Brunner T, Leist M, et al. Deconvoluting gene and environment interactions to develop an “epigenetic score meter” of disease. EMBO Mol Med. 2023 Sep 11;15(9):e18208.
Harvanek ZM, Boks MP, Vinkers CH, Higgins-Chen AT. The Cutting Edge of Epigenetic Clocks: In Search of Mechanisms Linking Aging and Mental Health. Biological Psychiatry. 2023 Nov;94(9):694–705.
Moore LD, Le T, Fan G. DNA Methylation and Its Basic Function. Neuropsychopharmacol. 2013 Jan;38(1):23–38.
Teroganova N, Girshkin L, Suter CM, Green MJ. DNA methylation in peripheral tissue of schizophrenia and bipolar disorder: a systematic review. BMC Genet. 2016 Dec;17(1):27.
Murphy TM, Mill J. Epigenetics in health and disease: heralding the EWAS era. The Lancet. 2014 Jun;383(9933):1952–4.
Kirkpatrick B, Messias E, Harvey PhillipD, Fernandez-Egea E, Bowie CR. Is Schizophrenia a Syndrome of Accelerated Aging? Schizophrenia Bulletin. 2008 Aug 20;34(6):1024–32.
Kirkpatrick B, Kennedy BK. Accelerated aging in schizophrenia and related disorders: Future research. Schizophrenia Research. 2018 Jun;196:4–8.
Chrusciel JH. A systematic review and meta-analysis of epigenetic clocks in schizophrenia. Schizophrenia Research. 2022;
Jeremian R, Malinowski A, Chaudhary Z, Srivastava A, Qian J, Zai C, et al. Epigenetic age dysregulation in individuals with bipolar disorder and schizophrenia. Psychiatry Research. 2022 Sep;315:114689.
Wu X, Ye J, Wang Z, Zhao C. Epigenetic Age Acceleration Was Delayed in Schizophrenia. Schizophrenia Bulletin. 2021 Apr 29;47(3):803–11.
Yousefi PD, Suderman M, Langdon R, Whitehurst O, Davey Smith G, Relton CL. DNA methylation-based predictors of health: applications and statistical considerations. Nat Rev Genet. 2022 Jun;23(6):369–83.
Hüls A, Czamara D. Methodological challenges in constructing DNA methylation risk scores. Epigenetics. 2020 Feb 1;15(1–2):1–11.
Barker ED, Cecil CAM, Walton E, Houtepen LC, O’Connor TG, Danese A, et al. Inflammation-related epigenetic risk and child and adolescent mental health: A prospective study from pregnancy to middle adolescence. Dev Psychopathol. 2018 Aug;30(3):1145–56.
Hamilton OKL, Zhang Q, McRae AF, Walker RM, Morris SW, Redmond P, et al. An epigenetic score for BMI based on DNA methylation correlates with poor physical health and major disease in the Lothian Birth Cohort. Int J Obes. 2019 Sep;43(9):1795–802.
Kiltschewskij DJ, Reay WR, Geaghan MP, Atkins JR, Xavier A, Zhang X, et al. Alteration of DNA Methylation and Epigenetic Scores Associated With Features of Schizophrenia and Common Variant Genetic Risk. Biological Psychiatry. 2023 Jul;S0006322323014336.
Raffington L, Tanksley PT, Sabhlok A, Vinnik L, Mallard T, King LS, et al. Socially Stratified Epigenetic Profiles Are Associated With Cognitive Functioning in Children and Adolescents. Psychol Sci. 2023 Feb;34(2):170–85.
Stevenson AJ, McCartney DL, Hillary RF, Campbell A, Morris SW, Bermingham ML, et al. Characterisation of an inflammation-related epigenetic score and its association with cognitive ability. Clin Epigenet. 2020 Dec;12(1):113.
Sánchez-Gutiérrez T, Rodríguez-Toscano E, Llorente C, de la Serna E, Moreno C, Sugranyes G, et al. Neuropsychological, clinical and environmental predictors of severe mental disorders in offspring of patients with schizophrenia. Eur Arch Psychiatry Clin Neurosci. 2020 Sep;270(6):739–48.
De la Serna E, Ilzarbe D, Sugranyes G, Baeza I, Moreno D, Rodríguez-Toscano E, et al. Lifetime psychopathology in child and adolescent offspring of parents diagnosed with schizophrenia or bipolar disorder: a 2-year follow-up study. Eur Child Adolesc Psychiatry. 2021 Jan;30(1):117–29.
Noguera A, Castro-Fornieles J, Romero S, de la Serna E, Sugranyes G, Sánchez-Gistau V, et al. Attenuated psychotic symptoms in children and adolescent offspring of patients with schizophrenia. Schizophrenia Research. 2018 Mar;193:354–8.
Segura AG, De La Serna E, Sugranyes G, Baeza I, Valli I, Martínez-Serrano I, et al. Polygenic risk scores mediating functioning outcomes through cognitive and clinical features in youth at family risk and controls. European Neuropsychopharmacology. 2024 Apr;81:28–37.
Segura AG, De La Serna E, Sugranyes G, Baeza I, Valli I, Díaz-Caneja C, et al. Epigenetic age deacceleration in youth at familial risk for schizophrenia and bipolar disorder. Transl Psychiatry. 2023 May 8;13(1):155.
Sanchez-Gistau V, Romero S, Moreno D, de la Serna E, Baeza I, Sugranyes G, et al. Psychiatric disorders in child and adolescent offspring of patients with schizophrenia and bipolar disorder: A controlled study. Schizophrenia Research. 2015 Oct;168(1–2):197–203.
Wechsler D. Wechsler Intelligence Scale for Children - Fourth Edition.Escala de inteligencia de Wechsler para niños. Manual técnico y de interpretación. 4th ed. 19500 Bulverde Road, San Antonio, TX 78259: Pearson; 2005.
Wechsler D. Wechsler Intelligence Scale for Children - Fifth Edition. Escala de inteligencia de Wechsler para niños. Manual técnico y de interpretación. 4th ed. 19500 Bulverde Road, San Antonio, TX 78259: Pearson; 2015.
Niemi LT, Suvisaari JM, Tuulio-Henriksson A, Lönnqvist JK. Childhood developmental abnormalities in schizophrenia: evidence from high-risk studies. Schizophrenia Research. 2003 Apr;60(2–3):239–58.
Duffy A, Hajek T, Alda M, Grof P, Milin R, MacQueen G. Neurocognitive functioning in the early stages of bipolar disorder: visual backward masking performance in high risk subjects. 2009;
Gotlib IH, Traill SK, Montoya RL, Joormann J, Chang K. Attention and memory biases in the offspring of parents with bipolar disorder: indications from a pilot study. Child Psychology Psychiatry. 2005 Jan;46(1):84–93.
Flanagan DP, Kaufman AS. Essentials of WISC-IV assessment. 2nd ed. Wiley; 2009.
First M, Spitzer R, Gibbon M, Williams J. Structured clinical Interview for DSM-IV axis i disorders, clinician version (SCID-CV). 1997.
First M, Spitzer R, Williams J. Structured clinical Interview for DSM-IV axis i disorders, clinician version (SCID-CV). 1999.
Kaufman J, Birmaher B, Brent D, Rao U, Flynn C, Moreci P, et al. Schedule for affective disorders and schizophrenia for school-age children-present and lifetime version (K-SADS-PL): Initial reliability and validity data. Journal of the American Academy of Child and Adolescent Psychiatry. 1997;36(7):980–8.
Ulloa R, Ortiz S, Higuera F, Nogales I, Fresán A, Apiquian R, et al. Interrater reliability of the Spanish version of Schedule for Affective Disorders and Schizophrenia for School-Age Children--Present and Lifetime version (K-SADS-PL). Actas españolas de psiquiatria. 2006;34(4):36–40.
Miller TJ, McGlashan TH, Rosen JL, Cadenhead K, Ventura J, McFarlane W, et al. Prodromal Assessment With the Structured Interview for Prodromal Syndromes and the Scale of Prodromal Symptoms: Predictive Validity, Interrater Reliability, and Training to Reliability. Schizophrenia Bulletin. 2003 Jan 1;29(4):703–15.
Shaffer D. A Children’s Global Assessment Scale (CGAS). Arch Gen Psychiatry. 1983 Nov 1;40(11):1228.
Lewis S, Owen M, Murray R. Obstetric complications and schizophrenia: methodology and mechanisms. In: Schizophrenia: A Scientific Focus. Schulz SC, Tamminga CA. Oxford University Press; 1989. p. 56–68.
Tian Y, Morris TJ, Webster AP, Yang Z, Beck S, Feber A, et al. ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Valencia A, editor. Bioinformatics. 2017 Dec 15;33(24):3982–4.
Sharp GC, Salas LA, Monnereau C, Allard C, Yousefi P, Everson TM, et al. Maternal BMI at the start of pregnancy and offspring epigenome-wide DNA methylation: findings from the pregnancy and childhood epigenetics (PACE) consortium. Human Molecular Genetics. 2017 Oct 15;26(20):4067–85.
Kazmi N, Sharp GC, Reese SE, Vehmeijer FO, Lahti J, Page CM, et al. Hypertensive Disorders of Pregnancy and DNA Methylation in Newborns: Findings From the Pregnancy and Childhood Epigenetics Consortium. Hypertension. 2019 Aug;74(2):375–83.
Knijnenburg TA, Vockley JG, Chambwe N, Gibbs DL, Humphries C, Huddleston KC, et al. Genomic and molecular characterization of preterm birth. Proc Natl Acad Sci USA. 2019 Mar 19;116(12):5819–27.
Küpers LK, Monnereau C, Sharp GC, Yousefi P, Salas LA, Ghantous A, et al. Meta-analysis of epigenome-wide association studies in neonates reveals widespread differential DNA methylation associated with birthweight. Nat Commun. 2019 Apr 23;10(1):1893.
Pelegí-Sisó D, De Prado P, Ronkainen J, Bustamante M, González JR. methylclock : a Bioconductor package to estimate DNA methylation age. Robinson P, editor. Bioinformatics. 2021 Jul 19;37(12):1759–60.
Horvath S. DNA methylation age of human tissues and cell types. Genome Biology. 2013;14(10):R115.
Hannum G, Guinney J, Zhao L, Zhang L, Hughes G, Sadda S, et al. Genome-wide Methylation Profiles Reveal Quantitative Views of Human Aging Rates. Molecular Cell. 2013 Jan;49(2):359–67.
Levine ME, Lu AT, Quach A, Chen BH, Assimes TL, Hou L, et al. An epigenetic biomarker of aging for lifespan and healthspan. 2018;
McEwen LM, O’Donnell KJ, McGill MG, Edgar RD, Jones MJ, MacIsaac JL, et al. The PedBE clock accurately estimates DNA methylation age in pediatric buccal cells. Proc Natl Acad Sci USA. 2020 Sep 22;117(38):23329–35.
Wu X, Chen W, Lin F, Huang Q, Zhong J, Gao H, et al. DNA methylation profile is a quantitative measure of biological aging in children. Aging. 2019 Nov 22;11(22):10031–51.
Das S, Forer L, Schönherr S, Sidore C, Locke AE, Kwong A, et al. Next-generation genotype imputation service and methods. Nat Genet. 2016 Oct;48(10):1284–7.
Price AL, Patterson NJ, Plenge RM, Weinblatt ME, Shadick NA, Reich D. Principal components analysis corrects for stratification in genome-wide association studies. Nat Genet. 2006 Aug;38(8):904–9.
RStudio Team. RStudio. Viena, Austria; 2017. (R: A language and environment for statistical computing. R Foundation for Statistical Computing).
Mohamed MA, Elhelbawy A, Khalid M, AbdAllatif LA, Lialy HE. Effects of bipolar disorder on maternal and fetal health during pregnancy: a systematic review. BMC Pregnancy Childbirth. 2023 Aug 28;23(1):617.
Safont G, Garriga M, González-Rodríguez A, Amoretti S, Simón O, Solè E, et al. Maternity in women with schizophrenia and schizoaffective disorder. Spanish Journal of Psychiatry and Mental Health. 2023 Sep;S295028532300011X.
Tang W, Zhou LJ, Zhang WQ, Jia YJ, Hu FH, Chen HL. Adverse perinatal pregnancy outcomes in women with schizophrenia: A systematic review and meta-analysis. Schizophrenia Research. 2023 Dec;262:156–67.
Vigod SN, Ray JG, Cohen E, Wilton AS, Saunders NR, Barker LC, et al. Maternal Schizophrenia and the Risk of a Childhood Chronic Condition. Schizophrenia Bulletin. 2022 Nov 18;48(6):1252–62.
King-Hele S, Webb RT, Mortensen PB, Appleby L, Pickles A, Abel KM. Risk of stillbirth and neonatal death linked with maternal mental illness: a national cohort study. Archives of Disease in Childhood - Fetal and Neonatal Edition. 2009 Mar 1;94(2):F105–10.
Ursini G, Punzi G, Chen Q, Marenco S, Robinson JF, Porcelli A, et al. Convergence of placenta biology and genetic risk for schizophrenia. Nat Med. 2018 Jun;24(6):792–801.
Bouvier D, Forest JC, Dion-Buteau E, Bernard N, Bujold E, Pereira B, et al. Association of Maternal Weight and Gestational Weight Gain with Maternal and Neonate Outcomes: A Prospective Cohort Study. JCM. 2019 Nov 27;8(12):2074.
Goldstein RF, Abell SK, Ranasinha S, Misso M, Boyle JA, Black MH, et al. Association of Gestational Weight Gain With Maternal and Infant Outcomes: A Systematic Review and Meta-analysis. JAMA. 2017 Jun 6;317(21):2207.
Liu X, Wang H, Yang L, Zhao M, Magnussen CG, Xi B. Associations Between Gestational Weight Gain and Adverse Birth Outcomes: A Population-Based Retrospective Cohort Study of 9 Million Mother-Infant Pairs. Front Nutr. 2022 Feb 14;9:811217.
Ori APS, Olde Loohuis LM, Guintivano J, Hannon E, Dempster E, St. Clair D, et al. Meta-analysis of epigenetic aging in schizophrenia reveals multifaceted relationships with age, sex, illness duration, and polygenic risk. Clin Epigenet. 2024 Apr 8;16(1):53.
Sumner JA, Gao X, Gambazza S, Dye CK, Colich NL, Baccarelli AA, et al. Stressful life events and accelerated biological aging over time in youths. Psychoneuroendocrinology. 2023 May;151:106058.
deSteiguer AJ, Raffington L, Sabhlok A, Tanksley P, Tucker-Drob EM, Harden KP. Stability of DNA-Methylation Profiles of Biological Aging in Children and Adolescents. bioRxiv. 2023 Nov 1;2023.10.30.564766.
Li Z, Zong X, Li D, He Y, Tang J, Hu M, et al. Epigenetic clock analysis of blood samples in drug-naive first-episode schizophrenia patients. BMC Psychiatry. 2023 Jan 17;23(1):45.
Liu L, Qi X, Cheng S, Meng P, Yang X, Pan C, et al. Epigenetic analysis suggests aberrant cerebellum brain aging in old-aged adults with autism spectrum disorder and schizophrenia. Mol Psychiatry. 2023 Nov;28(11):4867–76.
Barker DJ, Osmond C, Golding J, Kuh D, Wadsworth ME. Growth in utero, blood pressure in childhood and adult life, and mortality from cardiovascular disease. BMJ. 1989 Mar 4;298(6673):564–7.
Garcia-Rizo C, Bitanihirwe BKY. Implications of early life stress on fetal metabolic programming of schizophrenia: A focus on epiphenomena underlying morbidity and early mortality. Progress in Neuro-Psychopharmacology and Biological Psychiatry. 2020 Jul;101:109910.
Murray RM, Bhavsar V, Tripoli G, Howes O. 30 Years on: How the Neurodevelopmental Hypothesis of Schizophrenia Morphed Into the Developmental Risk Factor Model of Psychosis. Schizophrenia Bulletin. 2017 Oct 21;43(6):1190–6.
Corley J, Cox SR, Harris SE, Hernandez MV, Maniega SM, Bastin ME, et al. Epigenetic signatures of smoking associate with cognitive function, brain structure, and mental and physical health outcomes in the Lothian Birth Cohort 1936. Transl Psychiatry. 2019 Oct 7;9(1):248.
Ohi K, Shimada M, Soda M, Nishizawa D, Fujikane D, Takai K, et al. Genome-wide DNA methylation risk scores for schizophrenia derived from blood and brain tissues further explain the genetic risk in patients stratified by polygenic risk scores for schizophrenia and bipolar disorder. BMJ Ment Health. 2024 Jan;27(1):e300936.
Kiltschewskij DJ, Reay WR, Geaghan MP, Atkins JR, Xavier A, Zhang X, et al. Alteration of DNA Methylation and Epigenetic Scores Associated With Features of Schizophrenia and Common Variant Genetic Risk. Biological Psychiatry. 2024 Apr 1;95(7):647–61.
Caspi A, Shireby G, Mill J, Moffitt TE, Sugden K, Hannon E. Accelerated Pace of Aging in Schizophrenia: Five Case-Control Studies. Biological Psychiatry. 2024 Jun 1;95(11):1038–47.
Dada O, Adanty C, Dai N, Jeremian R, Alli S, Gerretsen P, et al. Biological aging in schizophrenia and psychosis severity: DNA methylation analysis. Psychiatry Research. 2021 Feb;296:113646.
Nguyen S, McEvoy LK, Espeland MA, Whitsel EA, Lu A, Horvath S, et al. Associations of Epigenetic Age Estimators With Cognitive Function Trajectories in the Women’s Health Initiative Memory Study. Neurology. 2024 Jul 9;103(1):e209534.
De Prisco M, Vieta E. The never-ending problem: Sample size matters. European Neuropsychopharmacology. 2024 Feb;79:17–8.
Barker ED, Walton E, Cecil CAM. Annual Research Review: DNA methylation as a mediator in the association between risk exposure and child and adolescent psychopathology. Child Psychology Psychiatry. 2018 Apr;59(4):303–22.
Nabais MF, Gadd DA, Hannon E, Mill J, McRae AF, Wray NR. An overview of DNA methylation-derived trait score methods and applications. Genome Biology. 2023 Feb 16;24(1):28.
Elliott HR, Tillin T, McArdle WL, Ho K, Duggirala A, Frayling TM, et al. Differences in smoking associated DNA methylation patterns in South Asians and Europeans. Clin Epigenet. 2014 Dec;6(1):4.
Shah S, Bonder MJ, Marioni RE, Zhu Z, McRae AF, Zhernakova A, et al. Improving Phenotypic Prediction by Combining Genetic and Epigenetic Associations. The American Journal of Human Genetics. 2015 Jul;97(1):75–85.
Ellman LM, Murphy SK, Maxwell SD, Calvo EM, Cooper T, Schaefer CA, et al. Maternal cortisol during pregnancy and offspring schizophrenia: Influence of fetal sex and timing of exposure. Schizophrenia Research. 2019 Nov 1;213:15–22.

No competing interests reported.

supplementarymaterial.xlsx

Download PDF

Editor assigned by journal
12 Jul, 2024
Submission checks completed at journal
12 Jul, 2024
First submitted to journal
11 Jul, 2024

You are reading this latest preprint version

Epigenetic signatures in children and adolescents at familial high risk: linking early-life environmental exposures to psychopathology

Status:

Version 1

Abstract

Figures

1. INTRODUCTION

2. METHODS

2.1 Sample

2.3 Assessments

2.3.1 Cognitive assessment

2.3.2 Clinical assessment

2.3.3 Assessment of obstetric complications

2.4 Biological samples

2.5 Methylation data collection

2.2 Methylation profile score calculation

2.3 Epigenetic clock calculation

2.4 Polygenic risk score calculation

2.5 Statistical analysis

3. RESULTS

3.1 Descriptive statistics

3.2 Epigenetic profile of the familial high risk groups

3.3 Association of epigenetic constructs with cognition, prodromal psychotic symptoms and global functioning

3.4 Epigenetic constructs association with obstetric complications and polygenic risk scores

4. DISCUSSION

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1