Abstract
Introduction
An imbalance between pro- and anti-inflammatory factors is implicated in the pathogenesis of preeclampsia (PE). Although IL10 (rs1800896) and IL6 (rs1800795) genotypes are known to influence circulating cytokine levels, their associations with longitudinal inflammatory profiles across pregnancy have not been well characterized.
Methods
We conducted a retrospective case–control study to longitudinally compare nine pro- and four anti-inflammatory cytokines across pregnancy based on IL6 and IL10 genotype. The cohort included 111 women with overweight/obesity (37 with PE; 74 without PE), matched 2:1, and predominantly Black (72%). Separate 3 × 3 mixed multivariate analyses of variance (MANOVAs) assessed interactions between IL6 or IL10 genotype (between-subjects’ factor) and pregnancy trimester (within-subjects factor). Pro- and anti-inflammatory cytokines were analyzed in separate models for each genotype. Due to the exploratory nature of the study and the examination of clinical relevance, statistical significance was set at p = .10.
Results
Pro-inflammatory cytokine levels changed over pregnancy (p < .05) but were not associated with IL6 or IL10 genotype (all p > .10). Anti-inflammatory cytokine levels also varied across pregnancy (p < .05). IL6 genotype was not associated with anti-inflammatory cytokine levels (p > .10). In contrast, IL10 genotype was significantly associated with differences in anti-inflammatory cytokine levels across genotypes (V = 0.22, F (8,162) = 2.48, p = .014, η p 2 = 0.11) and demonstrated an interaction effect with pregnancy trimester (V = 0.28, F (16,154) = 1.56, p = .087, η p 2 = 0.14).
Conclusion
IL10 genotype may influence circulating anti-inflammatory cytokines across pregnancy and, as such, may contribute to preeclampsia by influencing anti-inflammatory cytokine expression across pregnancy.
Introduction
Pregnancy creates a remarkable immunological environment whereby complex inflammatory processes must be carefully regulated to protect both mother and developing child while allowing the pregnancy to progress successfully. Increasing evidence shows that when cytokine expression becomes dysregulated, it can profoundly impact both maternal health and fetal development (Guan et al., 2023; Liang et al., 2025; Meyyazhagan et al., 2023). Among the most serious pregnancy-related complications arising from this delicate imbalance is preeclampsia, a severe condition that impacts around 2-8% of pregnancies globally and results in the deaths of more than 70,000 mothers and 500,000 infants each year (American College of Obstetricians and Gynecologists, 2020; Ayyash et al., 2024).
The underlying mechanisms of preeclampsia (PE) involve a complex network of interactions between inflammatory mediators, with pro-inflammatory cytokines playing a particularly harmful role in both disease development and progression (Deer et al., 2023; Herrock et al., 2023). Extensive research has consistently revealed that dysregulation of inflammatory cytokines, typically elevated pro-inflammatory cytokines and low anti-inflammatory cytokines can contribute significantly to preeclampsia development (Chaiworapongsa et al., 2023; Jancsura et al., 2023, 2025; Meyyazhagan et al., 2023). This delicate equilibrium between pro- and anti-inflammatory responses serves as a cornerstone for maintaining pregnancy stability and preventing adverse outcomes.
IL-6, a multifaceted pro-inflammatory cytokine, and IL-10, an archetypal anti-inflammatory cytokine, have been the subject of extensive investigation in the context of pregnancy complications (Jancsura et al., 2023; Nasri et al., 2022; Nath et al., 2020; Wallace et al., 2021). However, both the IL6 and IL10 genes contain single nucleotide polymorphisms (SNPs) in the promoter region that significantly influence cytokine expression levels, which is why we are limiting evaluation of these two cytokine-related SNPs (Hu et al., 2018; Veisian et al., 2019, 2020). The IL6 gene polymorphism at position −174 G/C (rs1800795) has been associated with differential IL-6 production, where single-nucleotide changes can influence transcriptional activity (Kang et al., 2020). The IL10 gene contains a polymorphism at position −1082 A/G (rs1800896) where the AA, AG, and GG genotypes produce low, medium, and high amounts of IL-10, respectively (Li et al., 2022). Additionally, rs1800896 has been implicated in the development of PE, which is likely related to alterations in circulating cytokine levels (Mora-Palazuelos et al., 2022; Veisian et al., 2020).
Despite the growing recognition of genetic influences on pregnancy-related inflammation, there remains a significant gap in our understanding of how specific cytokine genotypes interact with the dynamic inflammatory changes that naturally occur across pregnancy trimesters. Previous studies have primarily examined single time points during pregnancy, which limits our comprehensive understanding of how genetic influences on inflammation evolve across trimesters. Furthermore, the complex interaction between genetic variations and maternal factors like obesity in affecting inflammatory patterns throughout pregnancy requires further investigation.
The current study addresses this important knowledge gap by examining the interaction effects between IL6 rs1800795 and IL10 rs1800896 genotypes and pregnancy trimester progression on both pro- and anti-inflammatory cytokine profiles. By focusing on a multi-ethnic cohort of pregnant women with overweight and obesity (a population at heightened risk for PE), this research seeks to elucidate the complex relationships between genetic predisposition, inflammatory dynamics, and pregnancy outcomes. Understanding these intricate interactions may ultimately inform personalized approaches to preeclampsia prevention and management, potentially improving maternal and fetal health outcomes in high-risk populations and helping reduce the devastating toll this condition takes on families worldwide.
Methods
Participants come from the Prenatal Exposures and Preeclampsia Prevention Project (PEPP3): Mechanisms of Preeclampsia and Impact of Obesity (P01 HD30367), which was conducted at the Magee-Women’s Research Institute of the University of Pittsburgh Medical Center from 2008-2014. Details of study procedures have been previously described (Bell et al., 2013). Participants provided informed consent for all study procedures, and provided consent for samples to be used in future studies. For the current study, we used a case control study of 111 participants, 37 with preeclampsia and 74 normotensive controls with available biobanked samples. Participants were matched 2:1 based on pre-pregnancy BMI, race, smoking status, and gestational age at sample collection at the three time points of sample collection. For the current study, participants had a BMI ≥25 and had available biobanked samples. Participants with preexisting diabetes and hypertension were excluded. As this was an analysis of existing data an a priori power analysis was not completed. However, a post-hoc power analysis for a 3 x 3 Mixed MANOVA testing for an interaction effect at an alpha level = .10 (for exploratory purposes), a power level of .80, and of an expected medium effect size indicated sample size of 80 is sufficient to statistically significant findings.
As part of the parent study, participants completed 3 study visits across pregnancy. At each study visit, participants provided blood samples, which were aliquoted and stored as whole blood, plasma, and serum stored at −80°C. DNA was as extracted from whole blood samples for genotype analysis. Plasma samples were used for cytokine quantification.
Genotype Analysis
We used standard TaqMan allelic discrimination to determine participant genotype for rs1800795 (IL6 (−174 C/G)) and rs1800896 (IL10 (−1082 A/G)). Commercially available TaqMan Allele Discrimination Assays came from ThermoFisher. PCR reactions were run on an ABI QuantStudio™ under standard cycling conditions. Genotypes were assigned automatically using the instrument software. Data quality was ensured by manually reviewing the allelic discrimination plots, genotypes double-blind called and compare, and deviation from Hardy-Weinberg equilibrium assessed.
Inflammatory Maker Analysis
The methods for inflammatory marker quantification have been previously described (Jancsura et al., 2023, 2025). Briefly, for each trimester, inflammatory marker concentrations were measured in duplicate from plasma using the ThermoFisher Inflammation 20 plex Human ProcartaPlex™ Panel. For the current analysis we included inflammatory markers (IL-1α, IL-1β, IL-4, IL-6, IL-8, IL-10, IL-12p70, IL-13, IL-17A, GM-CSF, IFN-α, IFN-γ, and TNF-α) that were significantly associated with PE status in our previous work in this sample (Jancsura et al., 2023).
Statistical Analysis
Interaction Effects Between Genotype and Pregnancy Trimester on Cytokines
To evaluate the potential interaction effect between IL6 and IL10 genotype groups and pregnancy trimester (1st, 2nd, and 3rd) on cytokine levels, we conducted four 3 × 3 mixed multivariate analyses of variance (MANOVAs). IL6 and IL10 genotypes were assessed in separate models each with one model examining pro-inflammatory cytokines (IL-1α, IL-1β, IL-12p70, IL-17A, IL-6, IL-8, GM-CSF, IFN-γ, and TNF-α) and one model examining anti-inflammatory (IL-10, IL-13, IL-4, and IFN-α) cytokines. In all four analyses, the between-subjects factor was the IL6 or IL10 genotype, while the within-subjects factor was pregnancy trimester. The interaction effect between IL6 or IL10 genotypes and pregnancy trimester was assessed in each MANOVA model. Due to unequal sample sizes across genotype groups, Pillai’s Trace statistic (V) was used for the omnibus test, as it is more robust to unbalanced sample sizes compared to other statistics, such as Wilks’ Lambda (Ateş et al., 2019). Significant omnibus MANOVA models were examined for post-hoc analyses with a discriminate function analysis, as recommended by Barton et al. (2016); Barton et al. (2016). Additionally, to examine if genotypes were predictive of having preeclampsia, logistic regression models were conducted on genotypes that were found to have significant cytokines over pregnancy trimesters. Missing data for the MANOVA were handled with listwise deletion and therefore excluded in the logistic regression model. No missing values existed in the logistic regression model.
Since this study was exploratory in nature, examining clinical relevance, and power was limited by uneven sample sizes across groups, results were considered statistically significant at a p < .10. Statistical significance at p < .10 has been justified in clinical context when the risk of type two errors could be more problematic (AbdulRaheem, 2024; Barkan, 2015; Gaus et al., 2015).
Post-hoc Analysis
A total of four 3 × 3 mixed MANOVAs were conducted in this study. For models that had statistically significant p < .05 or approached significance interaction (p < .10; due to their clinical relevance and exploratory purposes), a post-hoc discriminant function analyses were performed. These analyses aimed to identify the specific dependent variable(s) most strongly contributing to the interaction effects observed between genotype and pregnancy trimester. The within subject-effects, in this case, the effect over trimesters has been reported in Jancsura et al. (2023) and thus was not assessed with a post-hoc analysis.
Covariates were not included in the models since this was a retrospective case-controlled study, participants were already matched 2:1 based on demographic factors mentioned in the methods, reducing the variance being accounted for by such factors. Controlling for a covariate can alter what the variable is representing (Darlington & Hayes, 2016).
Data Transformations and Preparation
Prior to conducting any analyses, the data distributions of cytokine levels in the first trimester were examined. All cytokine variables violated normal distribution assumptions and were positively skewed, necessitating log transformations. Log transformations successfully adjusted skewness and kurtosis to within acceptable ranges for IL-1α, IL-1β, IL-12p70, GM-CSF, TNF-α, and IL-4. However, for IL-17A, IFN-γ, IL-8, IL-6, IL-10, IL-13, and IFN-α, a more robust Box-Cox transformation was applied, which effectively brought their skewness and kurtosis values into the appropriate ranges. Given the repeated measures design of this study, any transformation applied to cytokine measurements in the first trimester was also applied to the corresponding measurements in the second and third trimesters to maintain consistency. Transformed variables were utilized for all inferential statistical analyses. However, descriptive statistics were calculated using the original, untransformed variables to facilitate interpretability.
Results
Sample Descriptives
This study included a sample of N = 111 pregnant women. Participants were predominantly young (M = 24.1 years), had a mean pre-pregnancy body mass index (BMI) in the obese range (M = 35.0), and were primarily Black (72%). The majority of participants were single/not living with a partner (66.7%) and 15.3% completed an associate’s degree or higher level education. The average infant birth weight was 3,181.44 grams (7.0 lbs.), corresponding to the 52nd percentile for birth weight.
Sample Characteristics of IL-6 Genotype
Sample Characteristics of IL-10 Genotype
Among IL6 genotype, the CC group exhibited a notably lower average infant birth weight of 2,937.1 grams (6.5 lbs.), placing this group in the 44th percentile for birth weight. In contrast, IL10 genotypes showed minimal differences in descriptive measures across groups.
Interaction Effects: Genotype and Pregnancy Trimester on Pro-inflammatory Cytokines
Pro-inflammatory cytokine levels significantly changed across pregnancy (V = 0.33, F (18, 67) = 1.84, p = .038, η p 2 = 0.33). However, the between-subjects effect of IL6 genotype on inflammatory cytokines was not significant (V = 0.15, F (18, 154) = 0.71, p = .800), nor was the interaction between IL6 genotypes and pregnancy trimester on pro-inflammatory cytokines significant (V = 0.47, F (36, 136) = 1.17, p = .257). Similarly, the between-subjects effect of IL10 genotype was not significant (V = 0.24, F (18, 154) = 1.15, p = .310),nor was the interaction between IL10 genotype and pregnant trimester on pro-inflammatory cytokines (V = 0.32, F (36, 136) = 0.72, p = .873).
Interaction Effects: Genotypes and Pregnancy Trimester on Anti-inflammatory Cytokines
Anti-inflammatory cytokines significantly changed throughout pregnancy trimesters, (V = 0.24, F (8, 67) = 3.06, p = .005, η p 2 = 0.24). However, the between-subjects effect of IL6 genotype on anti-inflammatory cytokines was not statistically significant (V = 0.24, F (18, 154) = 1.15, p = .310), nor was the interaction effect between IL6 genotype and pregnancy trimester on anti-inflammatory cytokines (V = 0.32, F (36, 136) = 0.72, p = .873). The between-subjects effect of IL10 genotype on anti-inflammatory cytokines was statistically significant, with a moderate-to-large effect size (V = 0.22, F (8, 162) = 2.48, p = .014, η p 2 = 0.11) and the interaction effect between IL10 genotypes and pregnancy trimester was significant, with a large effect size (V = 0.28, F (16, 154) = 1.56, p = .087, η p 2 = 0.14).
Follow-Up Discriminant Function Analysis for IL-10 Genotype and Trimester Interaction on Anti-inflammatory Cytokines
Structure Matrix of Discriminant Functions
Note. An “*” indicates significant loading on the function at the ≤.05.
Anti-inflammatory Cytokines Descriptive by Genotype and Trimester
IL-10 Genotype Predicting Preeclampsia
To evaluate whether a participant’s IL10 genotype could predict their preeclampsia status (normative or preeclampsia), a binary logistic regression analysis was performed using the TT genotype as the reference group. The results revealed that the model was significant (χ2 (2) = 4.66, p = .097, Nagelkerke R 2 = .06). It was observed that individuals with the TC genotype exhibited a 25% higher likelihood of developing preeclampsia compared to those with the TT genotype (β = 0.90, SE = 0.46, Wald = 3.87, p = .049). No statistically significant difference was found in the likelihood of preeclampsia between the CC and TT genotypes (p = .925). However, it is important to note that the results may have been affected by the small sample size in the CC genotype group (n = 13) relative to the TC (n = 39) and TT (n = 34) groups.
Discussion
This study looked at how genes and cytokines interact in a group that faces higher risks during pregnancy, predominantly young, Black, pregnant women with obesity. The IL6 genotype did not appear to affect circulating cytokine levels, but IL10 genotype significantly influenced anti-inflammatory cytokine levels and may contribute to changes in cytokine levels across pregnancy. Furthermore, IL10 genotype, specifically TC was associated with risk for preeclampsia. These findings suggest that IL10 genotype may contribute to cytokine regulation in pregnancy and represent a risk factor for preeclampsia.
The IL6 genotype demonstrated no significant effect on either cytokine levels or trimesters, which is surprising given the overwhelming support from the literature linking IL6 SNPs to circulating pro-inflammatory levels (Hu et al., 2018; Koc et al., 2023; Damavandi et al., 2022). This could be in part due to the sample consisting of only women with obesity as obesity is known to contribute to a chronic inflammatory environment (Rees et al., 2022; Spradley et al., 2015) and the IL6 polymorphisms have been associated with obesity (Chmurzynska et al., 2019; Todendi et al., 2016). Thus, in the context of obesity, IL6 genotype may not have meaningful impact on circulation of inflammatory maker levels in pregnancy. As our study did not include women with normal BMIs, we cannot determine if the associations would differ in women without obesity. Alternatively, it is possible that other contextual factors obscured genotype-protein associations. Our sample consisted of predominantly single mothers not living with a partner, with low levels of education. Such socioeconomic characteristics have been associated with alterations in circulating cytokines through stress pathways (Gillespie et al., 2021, 2022).
The significant IL10 genotype effect may result from the differentiated anti-inflammatory capacity of the various genotypes (Turner et al., 1997). The interaction with trimesters suggests that genotype-based differences in anti-inflammatory capacity may become more pronounced as pregnancy advances and inflammatory demands increase. Typically, the IL10 rs1800896 polymorphism establishes producer phenotypes, with TT (equivalent to AA) associated with low IL-10 production, TC (AG) with intermediate production, and CC (GG) with high production (Eskdale et al., 1998; Turner et al., 1997). However, in our cohort of women with obesity, TC carriers exhibited the highest cytokine levels of the three, deviating from their expected intermediate producer activity, and warrants further investigation. Our previous work demonstrated elevated anti-inflammatory markers in women with obesity who later developed PE (Jancsura et al., 2023), in contrast to much of the broader literature reporting reduced IL-10 levels in PE. We posited that the chronic, pro-inflammatory state of obesity may result in a compensatory increase in circulating anti-inflammatory cytokines (Jancsura et al., 2023). Taken together with the findings of the current study, we posit that IL10 genotype may cause alterations in circulating anti-inflammatory levels, further disrupting the delicate immune balance in pregnancy, thereby increasing risk for preeclampsia.
Limitations
This study was powered adequately for a medium effect size (n = 111), but the genotypic distribution was unequal. Though difficult to avoid, this imbalance could have reduced statistical power for the comparison of homozygous genotypes or could have an effect on the protective effects of a high IL-10 producer (CC) or low IL-6 producer (GG) genotype. However, statistically significant interaction between IL10 genotype and anti-inflammatory cytokines coupled with a medium to large effect size supports the presence of a meaningful interaction in pregnancy. Generalizability of this study to a more diverse population may also prove challenging, as this cohort was made up of mostly black women with obesity, a group overrepresented in healthcare disparities and increased hospital mortality rates (Ahmad et al., 2026). Additionally, it’s possible our findings would be different in women without obesity and thus is an avenue for future research. Confounders (such as diet, stress, social determinants of health and comorbidities) may also be considered for control in future studies, as they may influence both genotype-cytokine relationships and preeclampsia risk. Finally, we only examined one factor associated with cytokine production (polymorphisms). Other processes like gene expression regulation, epigenetic modification, and post-transcription regulation processes further contribute to circulating cytokine levels but could not be measured in the current study.
Conclusions
Although additional studies are needed to confirm these findings and establish clinical utility, this work adds to the existing knowledge of cytokine changes in pregnancy by demonstrating a potential role for IL10 genotype in modulating cytokine circulation in pregnancy and preeclampsia. If these results are confirmed in future studies, particular in cohorts including women without obesity, IL10 genotype may be an informative factor in conjunction with cytokines for risk stratification or screening to identify women who may benefit from increased prenatal monitoring, blood pressure surveillance, and earlier screening for preeclampsia biomarkers.
Footnotes
Ethical Considerations
The study was conducted in accordance with the ethical standards of the Institutional Review Board at the University of Pittsburgh. Ethical approval for this study was obtained from the IRB of the University of Pittsburgh (Protocol# 19110285) and The Ohio State University (Protocol #2022H0287).
Consent to Participate
Written informed consent was obtained from all participants prior to participation in the parent study and therefore this study was exempt since only de-identified data was used.
Author Contributions
Woo, Jennifer contributed to design contributed to acquisition drafted manuscript critically revised manuscript gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy Pearce, Michelle contributed to interpretation drafted manuscript critically revised manuscript gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy Hulla, Ryan contributed to analysis and interpretation drafted manuscript critically revised manuscript gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy Conley, Yvette contributed to conception and design contributed to acquisition critically revised manuscript gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy Jancsura, McKenzie contributed to conception and design contributed to acquisition, analysis, and interpretation drafted manuscript critically revised manuscript gave final approval agrees to be accountable for all aspects of work ensuring integrity and accuracy.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: K23MD016431 – PI Woo; National Institute of Nursing Research (T32NR009759) and the National Institute of Child Health and Human Development (P01HD30367) of the National Institutes of Health and the Heilbrunn Nurse Scholar’s Awards through Rockefeller University. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies. The funding agencies had no role in study design; in the collection, analysis, and interpretation of data; in the writing of the report; and in the decision to submit the report for publication.
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Data Availability Statement
Data will be made available upon reasonable request.