Ongoing and past projects
Genetic Mechanisms of Intergenerational Depression
Using novel dentate gyrus-specific gene expression-based polygenic risk scores based on gene expression data obtained from rodent early life stress experiments, we show that expression-based scores predict individual differences in hippocampal structure and interacts with maternal distress to predict children most at risk for hippocampal changes and KSADS depression diagnosis. Our converging results across clinical and population samples suggest that individual differences in genetics predict children’s sensitivity to the environment (both positive and negative), which in turn modulates hippocampal mechanisms of susceptibility to intergenerational depression.
Can Dentate Gyrus-specific Expression-based Polygenic Scores predict Dentate Gyrus Structure and Function and Depression Onset?
Polygenic Scores for Psychiatric Traits mediate the Impact of Multigenerational Family History for Depression on Offspring Psychopathology
Do polygenic scores for psychiatric and cognitive traits mediate the impact of multigenerational family history of depression on offspring’s psychopathology?
In this cohort study of 8,111 preadolescents in the US, depression and bipolar disorder polygenic scores were significantly associated with multigenerational family history of depression. Depression polygenic score significantly mediated the relationship between family history of depression and childhood psychopathology.
The results suggest that the intergenerational transmission of depression is partially mediated by common genetic variants associated with depression. (Lee* & van Dijk* et al., in submission)
Hippocampal Mechanisms of Intergenerational Depression
Altered dentate gyrus microstructure in individuals at high familial risk for depression predicts future symptoms
Background: Offspring of individuals with major depressive disorder (MDD) are at increased risk for developing MDD themselves. Altered hippocampal and specifically dentate gyrus (DG) structure and function may be involved in depression development. However, hippocampal abnormalities could also be a consequence of the disease. For the first time, we tested whether abnormal dentate gyrus micro and macrostructure was present in offspring of individuals with MDD and whether these abnormalities predicted future symptomatology.
Methods: We measured DTI grey matter mean diffusivity (MD), a measure of microstructure, and structural MRI volume of the DG in 102 children (Generation 2) and grandchildren (Generation 3) at high and low risk for depression, with risk defined as the presence of moderate or severe depression (high risk) or absence (low risk) in generation 1. Prior and current (HAM-D) and future depressive symptoms (PHQ-9, IDAS, WHODAS) were tested for association with hippocampal structure.
Results: DG MD was increased in individuals at high risk for depression, regardless of a personal history of depression. While DG mean diffusivity was not associated with past or current depressive symptoms, higher MD predicted higher symptom scores eight years later. DG microstructure partially mediated the association between risk and future symptoms. DG volume was reduced in the high-risk children but not in high-risk grandchildren.
Conclusion: Together, these findings suggest a role for the DG in the development of depression. Furthermore, DG microstructure, more than macrostructure, is a sensitive risk marker for depression and partially mediates future depressive symptoms.
Background: Maternal stress (MS) is a well-documented risk factor for impaired emotional development in offspring. Rodent models implicate the dentate gyrus (DG) of the hippocampus in the effects of MS on offspring depressive-like behaviors, but mechanisms in humans remain unclear. Here, we test across two independent cohorts whether MS is associated with depressive symptoms and with DG micro- and macro-structural alterations in offspring.
Methods: We analyzed DG DTI mean diffusivity (DG-MD) and volume in a 3-Generation Family Risk for Depression study (TGS; n=69, mean age 35.0) and the Adolescent Brain Cognitive Development Study (ABCD; n=5196, mean age 9.9) using generalized estimating equation models and mediation analysis. MS was assessed by the Parenting Stress Index (in TGS) and a measure compiled from the Adult Response Survey (ABCD). PHQ-9 and rumination scales (TGS) and Child Behavior Checklist (ABCD) measured offspring depressive symptoms at follow-up. Schedule for Affective Disorders and Schizophrenia–Lifetime interview measured depression diagnoses.
Results: Across cohorts, MS was associated with future symptoms and higher DG-MD (indicating disrupted microstructure) in offspring. Higher DG-MD was associated with higher symptom scores measured 5 years (TGS) and 1-year (ABCD) after MRI. In ABCD, DG-MD is increased in high-MS offspring who have depressive symptoms at follow-up, but not in offspring who remain resilient or whose mother had low MS.
Conclusions: Converging results across two independent samples extend previous rodent studies and suggest a role for the DG in exposure to MS and offspring depression.
In utero mechanisms of intergenerational depression: SSRIs
Serotonin shapes brain structure and function during early development across phylogenetically diverse species. In mice and humans, perinatal SSRI exposure produces brain alterations and increases anxiety/depression-related behaviors in the offspring. It remains unclear whether shared brain circuit changes underlie the behavioral impact of perinatal SSRIs across species. We examine how developmental SSRI-exposure in mice and humans changes fear-related brain activation and behavior. SSRI-administered mice showed increased defense responses to a predator odor that were associated with stronger fMRI-based fear circuit activation when compared to saline controls. Similarly, human adolescents exposed to SSRIs in utero showed greater activation of fear brain structures and exhibited higher anxiety and depressive symptoms than unexposed adolescents. Perinatal SSRI enhances innate fear-related responses and fear brain circuit activation that are conserved across species. (Zanni* & van Dijk* et al., 2023)
Intergenerational Depression
Background: Three-generation family studies of depression have established added risk of psychopathology for offspring with 2 previous generations affected with depression compared with 1 or none. Because of their rigorous methodology, there are few of these studies, and existing studies are limited by sample sizes. Consequently, the 3-generation family risk paradigm established in family studies can be a critical neuropsychiatric tool if similar transmission patterns are reliably demonstrated with the family history method. We examined the association of multigenerational family history of depression with lifetime depressive disorders and other psychopathology in children.
Methods: In this analysis of the Adolescent Brain Cognitive Development (ABCD) study data, retrospective, cross-sectional reports on psychiatric functioning among 11 200 children (generation 3 [G3]) and parent reports on parents' (G2) and grandparents' (G1) depression histories were analyzed. Four risk categories were created, reflecting how many prior generations had history of depression: (1) neither G1 nor G2 (G1-/G2-), (2) only G1 (G1+/G2-), (3) only G2 (G1-/G2+), and (4) both G1 and G2 (G1+/G2+). Child lifetime prevalence and relative risks of psychiatric disorders were based on child and caregiver reports and grouped according to familial risk category derived from G1 and G2 depression history.
Results: Among 11 200 included children, 5355 (47.8%) were female, and the mean (SD) age was 9.9 (0.6) years. By parent reports, the weighted prevalence of depressive disorder among children was 3.8% (95% CI, 3.2-4.3) for G1-/G2- children, 5.5% (95% CI, 4.3-7.1) for G1+/G2- children, 10.4% (95% CI, 8.6-12.6) for G1-/G2+ children, and 13.3% (95% CI, 11.6-15.2) for G1+/G2+ children (Cochran-Armitage trend = 243.77; P < .001). The weighted suicidal behavior prevalence among children was 5.0% (95% CI, 4.5-5.6) for G1-/G2- children, 7.2% (95% CI, 5.8-8.9) for G1+/G2- children, 12.1% (95% CI, 10.1-14.4) for G1-/G2+ children, and 15.0% (95% CI, 13.2-17.0) for G1+/G2+ children (Cochran-Armitage trend = 188.66; P < .001). By child reports, the weighted prevalence of depressive disorder was 4.8% (95% CI, 4.3-5.5) for G1-/G2- children, 4.3% (95% CI, 3.2-5.7) for G1+/G2- children, 6.3% (95% CI, 4.9-8.1) for G1-/G2+ children, and 7.0% (95% CI, 5.8-8.5) for G1+/G2+ children (Cochran-Armitage trend = 9.01; P = .002), and the weighted prevalence of suicidal behaviors was 7.4% (95% CI, 6.7-8.2) for G1-/G2- children, 7.0% (95% CI, 5.6-8.6) for G1+/G2- children, 9.8% (95% CI, 8.1-12.0) for G1-/G2+ children, and 13.8% (95% CI, 12.1-15.8) for G1+/G2+ children (Cochran-Armitage trend = 46.69; P < .001). Similar patterns were observed for other disorders for both parent and child reports and across sex, socioeconomic status, and race/ethnicity.
Conclusions: Having multiple prior affected generations was associated with increased risk of childhood psychopathology. Furthermore, these findings were detectable even at prepubertal ages and existed in diverse racial/ethnic and socioeconomic groups. Clinically, they underscore the need for screening for family history in pediatric settings and highlight implications for biological research with homogenous subgroups using magnetic resonance imaging or genetic analyses.
Early Life and Neural Mechanisms of Family Risk for Suicide
Why do some individuals die by suicide while others who have a similar background and experiences do not? Our goal is to identify neurobiological and environmental risk factors (and their interplay) that can help determine who goes on to develop suicidal attempts and ideation (susceptibility) versus who does not (resilience).
The study will test these questions in a unique three-generation longitudinal family study of depression that contains a battery of neuroimaging and early childhood data. The clustering of suicidal attempts and ideation within families in the study sample allows us to study the role of hippocampus and other frontal-limbic biomarkers as well as early life adversity across groups of (1) individuals at high familial risk with a personal history of suicide attempts and ideations (susceptible); (2) individuals at high familial risk without a personal history of suicide attempts and ideations (resilience); (3) individuals without a personal or family history of suicide attempts and ideations (controls).
The broad aims of the study are to (1) identify frontal-limbic markers of susceptibility and resilience in individuals at high family risk for suicide; (2) identify early life adversity markers that predispose to subsequent suicide attempts and ideations and to test whether these are mediated by the frontal-limbic circuitry.
This study is funded by a Young Investigator Award from the American Foundation for Suicide Prevention.
DG mechanisms of cognition
The dentate gyrus (DG) is crucial for behaviorally discriminating similar spatial memories, predicting that DG place cells change ("remap") their relative spatial tuning ("place fields") for memory discrimination. This prediction was never tested, although DG place cells remap across similar environments without memory tasks. We confirm this prior finding but find that DG place fields do not remap across spatial tasks that require DG-dependent memory discrimination. Instead of remapping, place-discriminating discharge is observed transiently among DG place cells, particularly when memory discrimination is most necessary. The DG network may signal memory discrimination by expressing distinctive sub-second network patterns of co-firing at memory discrimination sites. This involves increased coupling of discharge from place cells and interneurons, as was observed during successful, but not failed, behavioral expression of memory discrimination. Instead of remapping, these findings indicate that memory discrimination is signaled by sub-second patterns of correlated discharge within the dentate network. (van Dijk and Fenton, 2018)
