Tuesday, January 11, 2022

New genetics study of 5.4 million individuals on height: The top ~12k genetic variants explain ~40% of individual differences in Europeans, and 10-20% in other ancestries

A Saturated Map of Common Genetic Variants Associated with Human Height from 5.4 Million Individuals of Diverse Ancestries. Loic Yengo et al. bioRxiv Jan 10 20222. https://doi.org/10.1101/2022.01.07.475305

Abstract: Common SNPs are predicted to collectively explain 40-50% of phenotypic variation in human height, but identifying the specific variants and associated regions requires huge sample sizes. Here we show, using GWAS data from 5.4 million individuals of diverse ancestries, that 12,111 independent SNPs that are significantly associated with height account for nearly all of the common SNP-based heritability. These SNPs are clustered within 7,209 non-overlapping genomic segments with a median size of ~90 kb, covering ~21% of the genome. The density of independent associations varies across the genome and the regions of elevated density are enriched for biologically relevant genes. In out-of-sample estimation and prediction, the 12,111 SNPs account for 40% of phenotypic variance in European ancestry populations but only ~10%-20% in other ancestries. Effect sizes, associated regions, and gene prioritization are similar across ancestries, indicating that reduced prediction accuracy is likely explained by linkage disequilibrium and allele frequency differences within associated regions. Finally, we show that the relevant biological pathways are detectable with smaller sample sizes than needed to implicate causal genes and variants. Overall, this study, the largest GWAS to date, provides an unprecedented saturated map of specific genomic regions containing the vast majority of common height-associated variants.

Rates of Forced Sexual Experiences Among High School Students From 2001 to 2019: Rates of forced sex maintained for girls, there was a decrease over time for boys; as girls and boys aged, the risk of forced sex increased

Rates of Forced Sexual Experiences Among High School Students From 2001 to 2019. Tiffany L. Marcantonio, James Weese, Malachi Willis. Journal of Interpersonal Violence, January 6, 2022. https://doi.org/10.1177/08862605211055155

Abstract: Public awareness of sexual assault and initiatives aimed at preventing sexual assault continue to increase over the years. However, whether rates of sexual assault have diminished because of such cultural shifts remains unclear. The purpose of this study was to assess if rates of sexual assault (i.e., forced sex) have changed over the past 18 years for adolescent girls and boys as well as potential differences across racial/ethnic identities. Using nationally representative data from the Youth Risk Behavioral Surveillance Survey from 2001 to 2019, we conducted logistic regressions to assess rates of experiences of forced sex by sex and by sex and racial/ethnic identity, while accounting for grade level. Participants included 135,837 high school students. From 2001 to 2019, rates of forced sex maintained for girls; however, there was a decrease over time for boys. For girls, there were inconsistent differences in rates of forced sex by racial/ethnic identities. However, boys who identified as Black, Hispanic, Multi-Racial, and Other Race/Ethnicity were at higher risk to report forced sex than their White peers, until 2015; only Other Race/Ethnicity was at higher risk in 2019. As girls and boys aged, the risk of forced sex increased. Despite prevention efforts, rates of forced sex did not decrease from 2001 to 2019 for adolescent girls disregarding race/ethnicity, and for racial/ethnic minority boys. That rates of forced sex continue to be high is problematic as experiencing sexual assault at an earlier age is associated with myriad consequences. Further, results suggest current prevention initiatives may be inadequate at addressing risk factors for forced sex, and more broadly, sexual assault. Moving forward, researchers and educators may want to re-evaluate the strategies used to address and measure sexual assault experiences.

Keywords: sexual assault, forced sex, adolescent, YRBS, race/ethnicity minority, gender

Psychopathy factor 2 (erratic lifestyle & anti-sociality) weakly linked to lower intelligence; otherwise Dark Triad traits are relatively independent of IQ

Michels, M. (2022). General intelligence and the dark triad: A meta-analysis. Journal of Individual Differences, 43(1), 35-46. Jan 2022. http://dx.doi.org/10.1027/1614-0001/a000352

Abstract: The dark triad of personality (D3) – consisting of psychopathy, Machiavellianism, and narcissism – is a set of socially aversive personality traits. All three traits encompass disagreeable behavior and a particular disregard for the well-being of others, but also a tendency to strategic and deceptive manipulation of social environments in order to attain one′s goals. To exercise these complex manipulations effectively it seems beneficial to have high cognitive abilities. Therefore, a meta-analysis was conducted to examine possible relationships between intelligence and the dark triad. A total of 143 studies were identified to estimate the strength of relationships between the D3 and general, verbal, and nonverbal intelligence. The results indicate that none of the constructs of the dark triad are meaningfully related to intelligence. However, there was a small negative correlation between intelligence and Factor 2 psychopathy. The substantial heterogeneity regarding the observed effect sizes could not be explained with meta-regression for the most part. There was no evidence for a publication bias. In total, the results challenge the notion that the dark triad is an adaptive set of personality traits that enables individuals to effectively manipulate their social surroundings.

Keywords: intelligence, psychopathy, Machiavellianism, narcissism, meta-analysis


The meta-analysis showed that the D3 and intelligence are at most weakly related. Whereas the psychopathy-intelligence-relation is negative, for M and N there seems to be no relation at all. It should be noted that the study sample for M and N is considerably lower compared to P. Two of three expectations were corroborated. Whereas M and N were (as expected) not related to cognitive abilities, the relation between psychopathy and intelligence was significant but very small. The cause for the effect might be the overlap between P and criminality: the latter has shown to be negatively related to intelligence. This becomes particularly evident considering the small negative relation between intelligence and the P-Factor 2 (the aspect of psychopathy that comprises norm-violating behavior). Since criminality is part of many P-test-items, it would be inadequate to interpret this overlap as confounding. Furthermore, intelligence is negatively related to impulsivity (Schweizer, 2002Vigil-Coleṭ & Morales-Vives, 2005) and aggression (Ackerman & Heggestad, 1997) – two conceptual features of Factor 2 psychopathy. Alternatively, the negative P-intelligence-relation might be due to range restriction in the primary studies and might disappear in the course of a secondary analysis of all raw data – yet the analysis of raw data mentioned above suggests the opposite. Nevertheless, the results indicate that D3-individuals do not have superior cognitive abilities that might enable them to show complex manipulative behavior. On the other hand, they do not seem to have relevant cognitive deficits as well. If one assumes that D3-individuals can indeed be more successful in some contexts than others (an assumption that should be scrutinized in the first place), this analysis demonstrates that this possible success is not a consequence of high cognitive abilities.
Surprisingly, the reanalysis of the raw data showed a moderate negative relation with intelligence: it is unclear if the study sample coincidently showed a moderate effect or if the meta-analytic results might have to be reinterpreted. A reanalysis of the original data from the primary studies might have shown similar results due to an underestimation of effect sizes due to range restriction in the isolated studies. But note that an overestimation of the effect in this meta-analysis is also possible due to range restriction. On the other hand, for example, Watts et al. (2016) found similar results as in this meta-analysis regarding P-intelligence and did correct for range restriction using a formula for correcting correlation estimates by Hunter and Schmidt (1990), which did not alter their overall results. However, the results from a P-gI-meta-analysis with k > 100 might be more credible than the reanalysis of only 7 datasets. The reanalysis of raw data did not raise any reason to further inspect the D3-relations to intelligence in regards to non-linear relationships.

Limitations of the Meta-Analysis

A few limitations of this meta-analysis should be considered: First, the combined effect sizes remained heterogeneous even after moderators had been taken into account. As a result, the reported overall effects may be quite different in subpopulations not under investigation in the present study. Second, the number of studies for M and N was very small, so that the inference had to be restricted to the types of studies under investigation and cannot be further generalized due to the use of the FE model. Third, the selection of tests for M and N that were used in the primary studies was narrow – which also made possible subscale-analyses for M and N impossible. This does not apply for P and most of the studies used the PCL (which is considered the “gold standard”-measure for psychopathy). Forth, a more fine-grained analysis of intelligence subdimensions on the basis of an overarching model of intelligence – preferably the Cattell-Horn-Carroll theory (Carroll, 1993) – would have been desirable. Since the number of effect sizes per effect-size-category (e.g., P-gI) would have dropped substantially, a rather rough separation into verbal and non-verbal was the pragmatic consequence. Lastly, no gray literature was included in this analysis: Since there was no specific search for unpublished studies on the research question, a substantial body of literature might have been missed – nevertheless, the gray studies that were identified did not differ in methodology nor the reported effect size. Consequently, there was no reason to include them.

Machine learning: Men's sexual satisfaction was overall more predictable than women's

Identifying the strongest self-report predictors of sexual satisfaction using machine learning. Laura M. Vowels, Matthew J. Vowels, Kristen P. Mark. Journal of Social and Personal Relationships, January 11, 2022. https://doi.org/10.1177/02654075211047004

Abstract: Sexual satisfaction has been robustly associated with relationship and individual well-being. Previous studies have found several individual (e.g., gender, self-esteem, and attachment) and relational (e.g., relationship satisfaction, relationship length, and sexual desire) factors that predict sexual satisfaction. The aim of the present study was to identify which variables are the strongest, and the least strong, predictors of sexual satisfaction using modern machine learning. Previous research has relied primarily on traditional statistical models which are limited in their ability to estimate a large number of predictors, non-linear associations, and complex interactions. Through a machine learning algorithm, random forest (a potentially more flexible extension of decision trees), we predicted sexual satisfaction across two samples (total N = 1846; includes 754 individuals forming 377 couples). We also used a game theoretic interpretation technique, Shapley values, which allowed us to estimate the size and direction of the effect of each predictor variable on the model outcome. Findings showed that sexual satisfaction is highly predictable (48–62% of variance explained) with relationship variables (relationship satisfaction, importance of sex in relationship, romantic love, and dyadic desire) explaining the most variance in sexual satisfaction. The study highlighted important factors to focus on in future research and interventions.

Keywords: Sexual satisfaction, machine learning, random forests, Shapley values

Our results showed that we could predict between 48 and 62% of the variance in sexual satisfaction using a random forest algorithm, up to two to three times more than previous studies even after deleting relationship satisfaction from the model (Byers & Macneil, 2006Laumann et al., 2006). The algorithm is also explainable because it does not suffer from suppression and cancellation effects or multicollinearity. The results show that using machine learning can help move psychological research into a new era of highly predictive and accurate models that generalize better to the population and have a higher utility in practice (Yarkoni & Westfall, 2017).

The strongest predictors

Because of the importance of sexual satisfaction on relationship quality (Joel et al., 2020McNulty et al., 2016L. M. Vowels & K. P. Mark, 2020b) and overall well-being (Davison et al., 2009Del Mar Sánchez-Fuentes et al., 2014), understanding factors that are the most, and the least, strongly associated with sexual satisfaction is important. This can enable researchers and practitioners to target individuals who may be at a particular risk of poor sexual satisfaction and helps to address factors that are the most likely to induce changes in sexual satisfaction while ignoring those that are the least likely to produce change. Thus, we added to the literature by examining which factors were the most, and least, predictive of sexual satisfaction in two samples.

Several variables that have previously been identified as important predictors of sexual satisfaction were included in the top-10 predictors: relationship satisfaction (Joel et al., 2020McNulty et al., 2016L. M. Vowels & K. P. Mark, 2020b), dyadic desire (Kim et al., 2020Mark, 20122014), romantic love (L. M. Vowels & K. P. Mark, 2020a), sexual communication (Impett et al., 2019), and perception of love and sex (Hendrick & Hendrick, 2002). Importantly, when relationship satisfaction was low, it had up to three times higher impact on the model outcome compared to when relationship satisfaction was high. Furthermore, participants in Sample 1 who viewed sex as an important part of their relationship and those who had sex regularly also had higher sexual satisfaction compared to participants who placed less importance on sex and more on love and had sex less frequently. Similarly, participants who reported a higher frequency of more varied sexual behaviors such as giving and receiving oral sex and mutual masturbation in Sample 2 reported higher levels of sexual satisfaction. These results suggest that frequency and value of sex as well as a more varied sexual repertoire in relationships are important predictors of sexual satisfaction. More varied sexual repertoire is also likely to lead to more satisfying sexual experiences, especially for women given that women have a higher likelihood of orgasm from clitoral stimulation than from intercourse. These results confirm earlier findings using traditional statistical models (Haavio-Mannila & Kontula, 1997Laumann et al., 2006).

Gender was not an important predictor of sexual satisfaction suggesting that men and women overall had similar levels of sexual satisfaction in both samples which is consistent with some studies (Mark et al., 2018McClelland, 2011) and inconsistent with others (Laumann et al., 2006). Men’s sexual satisfaction was overall more predictable than women’s. This may be because women’s sexuality is thought to be more complex than men’s (Basson, 2001). There were also some notable differences in the top-10 predictors for men and women. Attachment avoidance was only in the top-10 predictors for women’s sexual satisfaction (18th for men changing the outcome very little). Women who were higher in attachment avoidance reported lower sexual satisfaction compared to women lower in attachment avoidance. Attachment avoidance is associated with fear of closeness and intimacy, which tend to be more strongly tied to sexuality for women than men (Péloquin et al., 2014), which may explain why attachment avoidance was particularly important for women.

Consistent with previous studies using both traditional analyses (Rubin et al., 2012L. M. Vowels & K. P. Mark, 2020a) and machine learning (Joel et al., 2020L. M. Vowels et al., 2021), including partner effects added little additional variance. However, both actor and partner variables were among the top-10 most important predictors. Partner effects alone could also explain around half as much variance as only actor effects. Important partner variables included partner’s sexual satisfaction, romantic love, relationship satisfaction, and dyadic desire. Interestingly, for women, their male partner’s sexual satisfaction was just as important a predictor for their own sexual satisfaction than their relationship satisfaction. This is consistent with several studies finding that women partnered with men tend to answer questions of sexual satisfaction relative to their partner’s satisfaction as much as their own (McClelland, 20112014Pascoal et al., 2014) and may be due to there being a societal expectation on women to prioritize men’s pleasure. For men, their female partner’s sexual satisfaction only accounted for about third as much change in sexual satisfaction compared to their own relationship satisfaction. These findings suggest that while we may be able to predict actor’s sexual satisfaction relatively well using only their own variables, accounting for both partners’ variables can provide important additional insights.

The present study also provided an important addition to the literature by evaluating which factors were unimportant for sexual satisfaction. Many of the variables that have previously been associated with sexual satisfaction in traditional analyses were less important compared to other predictors. These included variables such as gender, sexual orientation, children, religiosity, attitudes toward sexuality, and mental health (Del Mar Sánchez-Fuentes et al., 2014Laumann et al., 2006). This suggests that even though differences in demographic variables may be statistically significant in some studies especially when sample sizes are large (e.g., Laumann et al., 2006), this does not mean that the differences are meaningful. In fact, the present study suggests the opposite; couple’s overall relationship and sexual behaviors are more proximal to sexual satisfaction and appear more important than who the person is. Understanding which variables are less related to the outcome is important, so that researchers and practitioners do not waste their time and resources on factors that are less likely to change the outcome.

Implications for research, theory, and practice

The study has several strengths as well as important implications for research, theory, and practice. We used explainable machine learning and cross-validation in which the model performance is tested on unseen data to avoid overfitting and thus improve the generalizability of the results. The code used in the study is readily available and provides a pipeline to relationship researchers to conduct more robust and predictable science. The results showed that dyadic level variables are the most likely to contribute to sexual satisfaction while individual predictors are less important. Furthermore, examining individuals’ perceptions of love and sex (Hendrick & Hendrick, 2002), keeping sex as a central element of relationships, and broadening couple’s sexual repertoire may enhance their sexual satisfaction. Finally, we expect many of these variables to have a bidirectional association with sexual satisfaction meaning that improving one (e.g., introducing more varied sexual behaviors) may produce a positive change in the other (e.g., enhanced sexual satisfaction) which will in turn improve the first variable (e.g., increased desire to try new things).

Limitations and future directions

The study also has several limitations that should be considered when interpreting the results. While the study included many predictors that have been associated with sexual satisfaction in previous research, there are other variables that we did not account for, that predict sexual satisfaction (e.g., responsiveness, self-esteem, personality, sociocultural variables). We also only had access to self- and partner-report measures. Thus, the algorithm could only make the predictions based on the variables that were available in the dataset. Therefore, future research should consider a greater number of individual, relational, and societal factors and include behavioral measures to predict sexual satisfaction. We also used data from two relatively large samples including a large subset of couples, the data were convenience samples and limited in their generalizability; most of the participants were white and well-educated and all participants in Sample 2 were in mixed-sex relationships, albeit nearly half the participants were bisexual. We also did not ask participants about any disabilities which may have contributed to their sexual satisfaction. Therefore, future research is needed to examine predictors of sexual satisfaction in a more representative sample. Random forests are a powerful tool that will take advantage of any correlations and interactions in the data, no matter how non-linear, it cannot be used to estimate causality. However, in the absence of a means to reliably estimate causality when examining factors relating to sexual satisfaction, we believe that using a predictive model is perhaps the best option. There are limitations to the Shapley method which have been discussed elsewhere (Kumar et al., 2020), and the notion that the human-interpretable Shapley model sufficiently explains our model suggests that a simpler model may be adequate to begin with, even if the simpler model is harder to identify (Rudin, 2019).

Furthermore, the data were cross-sectional and therefore we could not examine which predictors may account for the most change in sexual satisfaction over time, or indeed whether sexual satisfaction is predictable over time. Joel et al. (2020) found that they could predict little relationship satisfaction longitudinally. Cross-sectional self-report measures are also prone to shared method variance which results in higher correlation among variables collected at the same point in time. We attempted to overcome some of these issues by testing the models without relationship satisfaction given its high correlation with sexual satisfaction and only using partner effects to predict actor’s sexual satisfaction. The models with relationship satisfaction excluded were still predictive but predicted less variance. The models with partner effects alone could predict nearly 30% of the variance in actor’s sexual satisfaction which is higher than most other previous studies using actor or actor and partner effects. Future longitudinal and behavioral research is needed to understand whether the self-report variables measured in this study are predictive over time or whether behavioral measures could also be predictive. Finally, we examined whether men and women differed in the predictors that were important for their sexual satisfaction and future research could also examine whether the predictors of sexual satisfaction differ by sexual orientation.

Judgments About Male Victims of Sexual Assault: 2019 female cohort more likely to attribute encouragement to victim (27% in 2019 vs 4% in 1984) & to believe that the victim derived pleasure or had less stress (25% vs 5%) than female victims

Judgments About Male Victims of Sexual Assault by Women: A 35-Year Replication Study. Emma K. PeConga, Jacqueline E. Spector, Ronald E. Smith. Journal of Interpersonal Violence, January 3, 2022. https://doi.org/10.1177/08862605211062990

Abstract: Sexual assault of men by women has received increasing attention in recent years, as has research on rape myths about male victims. This study is a cross-generational replication of a 1984 study of college students’ judgments about male and female victims in a scenario involving a sexual assault carried out by male or female assailants. The 1984 data (n = 172) were compared with those of a 2019 cohort (n = 372) in a 2 (participant gender) x 2 (assailant gender) x 2 (victim gender) x 2 (cohort) factorial design to assess potential generational changes in perceptions of victims. Judgments by male participants of male victims of assaults carried out by women changed notably over time. The 2019 male cohort was less likely to judge that the victim initiated or encouraged the incident (40% in 1984 compared with 15% in 2019) and derived pleasure from it (47.4% in 1984 compared with 5.8% in 2019). In contrast, the 2019 female cohort was more likely to attribute victim encouragement (26.9% compared with 4.3% in 1984) and pleasure to the male victim (25% in 2019 compared with 5% in 1984). A similar gender pattern occurred in judgments of how stressful the event was for the male victim. Analysis of the 2019 data revealed that overall, despite scientific and cultural shifts that have occurred over the past three decades, participants continued to judge the male victim of assault by a female to have been more encouraging and to have experienced more pleasure and less stress than in any other assailant/victim gender combination. Results are discussed in relation to gendered stereotypical beliefs and male rape myths, as well as possible sensitization to power differentials inspired by the #MeToo movement. We emphasize the need for greater awareness and empirical attention to abuse that runs counter to preconceived notions about sexual victimization.

Keywords: rape, sexual assault, rape myths, male victimization, gender

Vascular and blood-brain barrier-related changes underlie stress responses and resilience in female mice and depression in human tissue

Vascular and blood-brain barrier-related changes underlie stress responses and resilience in female mice and depression in human tissue. Laurence Dion-Albert, Alice Cadoret, Ellen Doney, Fernanda Neutzling Kaufmann, Katarzyna A. Dudek, Beatrice Daigle, Lyonna F. Parise, Flurin Cathomas, Nalia Samba, Natalie Hudson, Manon Lebel, Signature Consortium, Matthew Campbell, Gustavo Turecki, Naguib Mechawar & Caroline Menard. Nature Communications volume 13, Article number: 164. Jan 10 2022. https://www.nature.com/articles/s41467-021-27604-x

Abstract: Prevalence, symptoms, and treatment of depression suggest that major depressive disorders (MDD) present sex differences. Social stress-induced neurovascular pathology is associated with depressive symptoms in male mice; however, this association is unclear in females. Here, we report that chronic social and subchronic variable stress promotes blood-brain barrier (BBB) alterations in mood-related brain regions of female mice. Targeted disruption of the BBB in the female prefrontal cortex (PFC) induces anxiety- and depression-like behaviours. By comparing the endothelium cell-specific transcriptomic profiling of the mouse male and female PFC, we identify several pathways and genes involved in maladaptive stress responses and resilience to stress. Furthermore, we confirm that the BBB in the PFC of stressed female mice is leaky. Then, we identify circulating vascular biomarkers of chronic stress, such as soluble E-selectin. Similar changes in circulating soluble E-selectin, BBB gene expression and morphology can be found in blood serum and postmortem brain samples from women diagnosed with MDD. Altogether, we propose that BBB dysfunction plays an important role in modulating stress responses in female mice and possibly MDD.


Only a handful of studies have explored BBB sex differences, most indirectly and in vitro37 but, to our knowledge, none did so in the context of chronic stress in mice or MDD. Overall, our findings indicate that chronic social and subchronic variable stressors alter BBB integrity in the mouse female brain through loss of the tight junction protein Cldn5 in the PFC and, to some extent, other mood-related brain regions such as the NAc. Importantly, these vascular alterations are also present in postmortem human brain samples from women with MDD. In mice, viral-mediated downregulation of Cldn5 in the PFC is sufficient to promote anxiety- and depression-like behaviours including social avoidance, anhedonia, and helplessness supporting a causal role in the establishment of maladaptive stress responses and possibly, mood disorders. We did not observe stress-induced BBB dysfunction in the male PFC7 indicating that chronic stress and depression affect the neurovasculature in a sex-specific manner. Different stress paradigms elicit specific anxiety- and depression-like behaviours according to sex, each recapitulating certain aspects of the symptoms and molecular features of MDD38.

A potential limitation of the female CSDS model we have used here is exposure to a male stressor (i.e., antagonistic social confrontations by a larger male aggressor). In male C57BL/6, the CSDS paradigm has been shown to have relevant etiological, predictive and face validity14. However, additional studies are warranted to confirm that this model is also etiologically relevant for female mice. Furthermore, SI tests were conducted using a male social target and we cannot exclude a possible confounding effect of intersex social interaction on the SI ratio values. Performing these tests with a female target may result in a different stratification of SS and RES mice, as we report for our virus-injected cohorts. We chose to use male mice as social targets in other contexts because social stress was performed by male aggressors19. Physical injuries are always a concern when running the CSDS paradigm either in males or females. Indeed, sickness behaviours could account for behavioral changes through stimulation of the immune system and entry of inflammatory mediators into the brain via altered BBB permeability. Therefore, physical examinations of animals were performed and no difference in the number of wounds was observed (Supplementary Figs. 1a and 8b) suggesting that susceptible vs resilient behavioral phenotypes were not due to physical injury. Finally, stress resilience is a fluid concept, and it is still debated whether or not it can be defined as a trait39. The complex neurobiological interactions underlying SS and RES phenotypes are not fully understood and are highly context-dependent. Coping strategies (i.e., social interactions) may be considered adaptative in some circumstances and maladaptive in others, and such mechanisms have been poorly investigated in females40. Thus, further studies are needed to better understand individual differences in stress-induced resilience vs vulnerability and how these behavioural responses and underlying mechanisms relate to human mood disorders.

Only females are susceptible to 6-d SCVS; however, both sexes display depression-like behaviours after weeks of stress exposure915 or if behavioural testing is performed 30 days after the 6-d SCVS paradigm41. Discrepancies in these behaviours could explain the sex-specific regional vascular effects observed and, possibly, sex differences associated with MDD. Magnetic resonance imaging (MRI) studies also support this idea, with unmedicated women with MDD showing decreased grey matter volume in limbic regions, including the left ventral PFC, while this reduction is observed in striatal regions for men with MDD42. Our study is also in line with recent clinical observations reporting region-specific BBB disruption in psychiatric disorders4344, although sex differences were not addressed.

Our work not only highlights fundamental sex differences in stress-induced neurovascular responses but also provides mechanistic insights by identifying key pathways and genes involved. Bulk tissue sequencing studies show a major rearrangement of transcriptional patterns in mood-related brain regions in MDD with low overlap between men vs women with MDD (~10%)911. These marked sex differences are also observed in mouse models of depression915. Neuronal contribution is undeniable, notably via changes in neurotransmitter systems9. Nevertheless, a significant enrichment for endothelium-related genes is also present9 but had never been explored. Bulk RNA sequencing revealed a similar enrichment for genes related to this cell population in schizophrenia45, reinforcing the involvement of the neurovasculature in psychiatric disorders. Although the resilience phenotype in female mice is not as clearly defined as it is for their male counterparts when analysing behaviours1924, our endothelium transcriptomic profiling revealed a distinct resilience-associated pattern when compared to animals displaying anxiety- or depression-like behaviours induced by either chronic social or subchronic variable stress. We found secreted enzyme phospholipase A2 Group VII (Pla2g7) as a gene significantly upregulated in SS vs RES mice, and Pla2g7 methylation is associated with increased risk of coronary heart disease specifically in women46. On the other hand, carnitine acyl carnine translocase (Slc25a20) is decreased in the PFC of SS females when compared to RES, which is consistent with previous studies reporting decreased acetyl-l-carnitine in MDD patients47.

Interestingly, both sexes show significant endothelium gene expression changes in the omega-3/omega-6 fatty acid synthesis pathways in the PFC following chronic social stress. However, these alterations were observed in RES vs CTRL for females and between SS vs CTRL in males. This pathway has been extensively studied in human depression as well as in animal models of mood disorders with omega-3 deficiency linked to neuronal atrophy in the medial PFC of male mice, concomitant with anxiety- and depressive-like behaviours48. Fatty acids play an important role in the regulation of systemic49 and endothelium inflammation50, thus providing an intriguing association between depression-related disruptions in the neuroimmune axis51 and BBB hyperpermeability. Similarities between the PFC female resilient and male susceptible endothelial transcriptome profiles for this pathway and the fact that both subgroups maintain BBB integrity in this brain region despite chronic stress exposure suggest that it could play a protective role in stress-induced BBB permeability loss. With previous reports linking omega-3 fatty acid consumption with resilience to stress in rodents52 and humans53, further studies are warranted to confirm their potential beneficial effect on the neurovasculature and gain mechanistic insights.

A possible mechanism for this sex- and region-specific vulnerability of the BBB is the presence of estrogen receptors on the neurovasculature. Endothelial cells express low levels of functional estrogen receptors54 and estrogen-coupled receptors can enter the cell nucleus and bind to estrogen-responsive elements (ERE) on specific DNA sequences55. Interestingly, ERE and stimulating protein 1 (Sp1) transcription factors were identified on the mouse Cldn5 gene promoter, which allows estrogen receptors to modulate Cldn5 transcription through cooperative interactions of Er/Sp1 with ERE/Sp1 elements56. While high levels of estrogen render female rats more sensitive to stress-induced PFC dysfunction57, this was found to be protective in the striatum58, providing mechanisms to explore for future studies. Our group has recently shown that permissive epigenetic regulation of Cldn5 expression paired with low endothelium Cldn5-repressive transcription factor forkhead box protein O1 is associated with stress resilience in the NAc of male mice, while increased histone deacetylase 1 level and activity is a mediator of stress susceptibility8. Thus, investigating sex-specific epigenetic vascular mechanisms underlying susceptibility vs resilience to stress and BBB hyperpermeability will be important in the future.

Additionally, estrogen-activated receptors have been shown to inhibit the proinflammatory transcription factor NfkB (nuclear factor kappa light chain enhancer of activated B cells), a known regulator of ICAM-1 and E-selectin59. Increased circulating proinflammatory cytokines and NFkB were reported in adolescent MDD and bipolar disorder, correlated with depressive symptoms severity60, and we have identified the proinflammatory NFkB pathway as a mediator of stress susceptibility in NAc endothelial cells of male8 but not female mice (Figs. 4 and 5). Accordingly, we observed a sex-specific increase in circulating sICAM-1 levels only in SS male mice vs pre-CSDS at baseline (Supplementary Fig. 10) and Icam1 levels are increased in the NAc of SS male mice8 but seem to be reduced in the PFC of SS female mice (Supplementary Fig. 10), reinforcing the idea of sex-specific regulatory mechanisms of BBB integrity, possibly through estrogen-mediated pathways. Social defeat stress increases the expression of adhesion molecules in the male mouse brain, including Icam-1 and Sele the gene encoding for E-selectin61. We did observe a trend, that did not reach significance, for higher expression of Sele in the PFC of stressed mice (Supplementary Fig. 10) suggesting that the increase in circulating sE-selectin measured in the blood serum of SS females may come, at least partly, from other brain regions or non-CNS sources. Moreover, elevated blood sE-selectin level was reported in the elderly with mild cognitive impairment and depressive mood62; however, these studies did not address sex differences. Assessment of BBB leakage using MRI scans in patients suffering from bipolar disorder allowed identification of a subpopulation of patients characterized by worse symptoms including the severity of depression, anxiety, chronicity of illness and decreased global functioning63. Despite being commonly used worldwide, it would be unrealistic to apply BBB imaging to a large population scale or in a preventive context highlighting the importance in discovering biomarkers of psychiatric diseases as we aimed to do here. It could be particularly relevant for conditions involving exacerbated inflammation and/or vascular dysfunction and for which MDD prevalence is higher than the general population; for example, stroke or Alzheimer’s disease1.

Many unanswered questions persist regarding BBB adaptations in both health and disease18. Our multidisciplinary approach allowed us to identify sex-specific circulating vascular potential biomarkers as well as candidate genes and pathways that could be relevant to inform on MDD diagnosis and develop treatments. Targeting and regulating tight junction protein integrity at the BBB level could represent an innovative strategy to treat mood disorders43. However, thinking beyond endothelial cells will be important to better understand the complex biology underlying BBB hyperpermeability in MDD. Single-cell sequencing of postmortem brain tissue from individuals with MDD shows important dysfunction in the PFC pyramidal neurons and oligodendrocyte-lineage cells64, but to our knowledge, this had never been investigated for endothelial cells, smooth muscle cells or pericytes. By characterizing sex- and region-specific neurovascular alterations underlying stress susceptibility in mice and human depression we provide valuable clues and highlight the need to consider sex as a biological variable while defining the role of brain barriers in psychiatric diseases. These findings are also important in the context of cardiovascular diseases (CVD), with evidence supporting a bidirectional relationship between depression and CVD, where depression is a predictor for the development of CVD, and vice versa65. Like in MDD, sex differences in age of onset, symptomatology and treatment response exist in CVD, highlighting the possibility of common etiological basis. Thus, investigating the candidate genes and circulating vascular potential biomarkers associated with depression we report here in populations of CVD patients would be highly relevant and could provide clues into shared underlying mechanisms of CVD and MDD pathologies. As an example, we observed an increase in sPAI-1, a prothrombotic plasma protein secreted by endothelial tissue, following social and subchronic variable stress in female, but not male, mice (Fig. 6c, d, Supplementary Fig. 10). Dysfunction of this pathway has been proposed as a link between MDD and CVD66 but underlying mechanisms remain to be elucidated. Genetic variants of the PAI-1 gene (SERPINE1) have been associated with MDD67. An elevated level of PAI-1 proteins is observed in the serum of individuals with major depression disorder68, but is also linked with increased risk of ischemic cardiovascular events such as thrombosis and atherosclerosis69. Antidepressant treatment decreases PAI-1 expression in the brain of male rats67. However, discrepancies exist with PAI-1 deficiency predisposing male mice to depression-like behaviours and resistance to commonly prescribed selective serotonin reuptake inhibitors70. Our findings suggest that PAI-1 may be even more relevant for female stress responses, and it was hypothesized that it could play a role in perinatal depression71, raising interest for future studies on this target to gain mechanistic insights particularly for female rodents and women with MDD.

To sum up, our study shows that chronic stress can induce BBB alterations in the female PFC promoting anxiety, depression-like behaviours, and BBB leakiness (Supplementary Fig. 11). Importantly, loss of tight junction CLDN5 expression was confirmed in postmortem brains samples from women with major depressive disorder supporting relevance for human MDD. These alterations were associated with changes in circulating vascular potential biomarkers in the mouse and human blood serum that will have to be explored in future clinical studies and larger human cohorts to confirm translational value. Conversely, despite chronic stress exposure, BBB integrity is maintained in the PFC of RES animals and this could be related to transcriptomic adaptations and activation of protective signalling pathways in the endothelium (Supplementary Fig. 11). Although this project explored neurovasculature-related changes in two brain regions, many other areas involved in emotion regulation remain unexplored and we hypothesize that stress-induced BBB changes go beyond the NAc and PFC. It will also be intriguing to investigate if age-related neurovascular changes such as BBB breakdown, which has been linked to human cognitive dysfunction72, play a causal role in late-life depression73 which is more prevalent in women74.