Working memory load-dependent changes in cortical network connectivity estimated by machine learning.

Working memory engages multiple distributed brain networks to support goal-directed behavior and higher order cognition. Dysfunction in working memory has been associated with cognitive impairment in neuropsychiatric disorders. It is important to characterize the interactions among cortical networks that are sensitive to working memory load since such interactions can also hint at the impaired dynamics in patients with poor working memory performance. Functional connectivity is a powerful tool used to investigate coordinated activity among local and distant brain regions. Here, we identified connectivity footprints that differentiate task states representing distinct working memory load levels. We employed linear support vector machines to decode working memory load from task-based functional connectivity matrices in 177 healthy adults. Using neighborhood component analysis, we also identified the most important connectivity pairs in classifying high and low working memory loads. We found that between-network coupling among frontoparietal, ventral attention and default mode networks, and within-network connectivity in ventral attention network are the most important factors in classifying low vs. high working memory load. Task-based within-network connectivity profiles at high working memory load in ventral attention and default mode networks were the most predictive of load-related increases in response times. Our findings reveal the large-scale impact of working memory load on the cerebral cortex and highlight the complex dynamics of intrinsic brain networks during active task states.

Localization and Diagnostic Specificity of Glutamic Acid Decarboxylase Transcript Alterations in the Dorsolateral Prefrontal Cortex in Schizophrenia.

BACKGROUND: Working memory (WM) deficits in schizophrenia are thought to reflect altered inhibition in the dorsolateral prefrontal cortex (DLPFC). This interpretation is supported by findings of lower transcript levels of the 2 enzymes, GAD67 and GAD65, which mediate basal and activity-dependent GABA (gamma-aminobutyric acid) synthesis, respectively. However, the relative magnitude, location within the depth of the DLPFC, and specificity to the disease process of schizophrenia of alterations in GAD67 and/or GAD65 remain unclear. METHODS: Levels of GAD67 and GAD65 messenger RNAs (mRNAs) in superficial (layers 2/superficial 3) and deep (deep layer 6/white matter) zones of the DLPFC were quantified by quantitative polymerase chain reaction in subjects with schizophrenia (n = 41), major depression (n = 42), or bipolar disorder (n = 39) and unaffected comparison (n = 43) subjects. RESULTS: Relative to the unaffected comparison group, GAD67 and GAD65 mRNA levels in the schizophrenia group were lower (p = .039, effect size = -0.69 and p = .027, effect size = -0.72, respectively) in the superficial zone but were unaltered in the deep zone. In the major depression group, only GAD67 mRNA levels were lower and only in the superficial zone (p = .089, effect size = 0.70). No differences were detected in the bipolar disorder group. Neither GAD67 nor GAD65 mRNA alterations were explained by psychosis, mood disturbance, or common comorbid factors. CONCLUSIONS: Alterations in markers of GABA synthesis demonstrated transcript, DLPFC zone, and diagnostic specificity. Given the dependence of WM on GABA neurotransmission in the superficial DLPFC, our findings suggest that limitations to GABA synthesis in this location contribute to WM impairments in schizophrenia, especially during demanding WM tasks, when GABA synthesis requires the activity of both GAD67 and GAD65.

Diagnostic Specificity and Association With Cognition of Molecular Alterations in Prefrontal Somatostatin Neurons in Schizophrenia.

Importance: Individuals with schizophrenia (SZ) exhibit pronounced deficits in somatostatin (SST) messenger RNA (mRNA) levels in the dorsolateral prefrontal cortex (DLPFC). Molecularly distinct subtypes of SST neurons, located in the superficial and deep zones of the DLPFC, are thought to contribute to different functional processes of this region; understanding the specificity of SST alterations in SZ across these zones could inform the functional consequences of those alterations, including cognitive impairments characteristic of SZ. Objective: To quantify mRNA levels of SST and related neuropeptides in the DLPFC in individuals with SZ, bipolar disorder (BPD), or major depressive disorder (MDD) and unaffected comparison individuals. Design, Setting, and Participants: This case-control study, conducted from January 20, 2020, to March 30, 2022, used postmortem brain tissue specimens previously obtained from individuals with SZ, MDD, or BPD and unaffected individuals from a community population through 2 medical examiners' offices. Demographic, clinical, and educational information was ascertained through psychological autopsies. Exposures: Diagnosis of SZ, BPD, or MDD. Main Outcome and Measures: The main outcome was levels of SST and related neuropeptide mRNA in 2 DLPFC zones, examined using laser microdissection and quantitative polymerase chain reaction or fluorescent in situ hybridization (FISH). Findings were compared using educational attainment as a proxy measure of premorbid cognition. Results: A total of 200 postmortem brain specimens were studied, including 65 from unaffected comparison individuals (42 [65%] male; mean [SD] age, 49.2 [14.1] years); 54 from individuals with SZ (37 [69%] male; mean [SD] age, 47.5 [13.3] years); 42 from individuals with MDD (24 [57%] male; mean [SD] age, 45.6 [12.1] years); and 39 from individuals with BPD (23 [59%] male; mean (SD) age, 46.2 [12.5] years). Compared with unaffected individuals, levels of SST mRNA were lower in both superficial (Cohen d, 0.68; 95% CI, 0.23-1.13; P = .004) and deep (Cohen d, 0.60; 95% CI, 0.16-1.04; P = .02) DLPFC zones in individuals with SZ; findings were confirmed using FISH. Levels of SST were lower only in the superficial zone in the group with MDD (Cohen d, 0.58; 95% CI, 0.14-1.02; P = .12), but the difference was not significant; SST levels were not lower in either zone in the BPD group. Levels of neuropeptide Y and tachykinin 1 showed similar patterns. Neuropeptide alterations in the superficial, but not deep, zone were associated with lower educational attainment only in the group with SZ (superficial: adjusted odds ratio, 1.71 [95% CI, 1.11-2.69]; P = .02; deep: adjusted odds ratio, 1.08 [95% CI, 0.64-1.84]; P = .77). Conclusions and Relevance: The findings revealed diagnosis-specific patterns of molecular alterations in SST neurons in the DLPFC, suggesting that distinct disease processes are reflected in the differential vulnerability of SST neurons in individuals with SZ, MDD, and BPD. In SZ, alterations specifically in the superficial zone may be associated with cognitive dysfunction.

Uncovering and mitigating bias in large, automated MRI analyses of brain development.

Large, population-based MRI studies of adolescents promise transformational insights into neurodevelopment and mental illness risk 1,2. However, MRI studies of youth are especially susceptible to motion and other artifacts 3,4. These artifacts may go undetected by automated quality control (QC) methods that are preferred in high-throughput imaging studies, 5 and can potentially introduce non-random noise into clinical association analyses. Here we demonstrate bias in structural MRI analyses of children due to inclusion of lower quality images, as identified through rigorous visual quality control of 11,263 T1 MRI scans obtained at age 9-10 through the Adolescent Brain Cognitive Development (ABCD) Study6. Compared to the best-rated images (44.9% of the sample), lower-quality images generally associated with decreased cortical thickness and increased cortical surface area measures (Cohen's d 0.14-2.84). Variable image quality led to counterintuitive patterns in analyses that associated structural MRI and clinical measures, as inclusion of lower-quality scans altered apparent effect sizes in ways that increased risk for both false positives and negatives. Quality-related biases were partially mitigated by controlling for surface hole number, an automated index of topological complexity that differentiated lower-quality scans with good specificity at Baseline (0.81-0.93) and in 1,000 Year 2 scans (0.88-1.00). However, even among the highest-rated images, subtle topological errors occurred during image preprocessing, and their correction through manual edits significantly and reproducibly changed thickness measurements across much of the cortex (d 0.15-0.92). These findings demonstrate that inadequate QC of youth structural MRI scans can undermine advantages of large sample size to detect meaningful associations.

Genetic patterning for child psychopathology is distinct from that for adults and implicates fetal cerebellar development.

Childhood psychiatric symptoms are often diffuse but can coalesce into discrete mental illnesses during late adolescence. We leveraged polygenic scores (PGSs) to parse genomic risk for childhood symptoms and to uncover related neurodevelopmental mechanisms with transcriptomic and neuroimaging data. In independent samples (Adolescent Brain Cognitive Development, Generation R) a narrow cross-disorder neurodevelopmental PGS, reflecting risk for attention deficit hyperactivity disorder, autism, depression and Tourette syndrome, predicted psychiatric symptoms through early adolescence with greater sensitivity than broad cross-disorder PGSs reflecting shared risk across eight psychiatric disorders, the disorder-specific PGS individually or two other narrow cross-disorder (Compulsive, Mood-Psychotic) scores. Neurodevelopmental PGS-associated genes were preferentially expressed in the cerebellum, where their expression peaked prenatally. Further, lower gray matter volumes in cerebellum and functionally coupled cortical regions associated with psychiatric symptoms in mid-childhood. These findings demonstrate that the genetic underpinnings of pediatric psychiatric symptoms differ from those of adult illness, and implicate fetal cerebellar developmental processes that endure through childhood.

Association of adverse prenatal exposure burden with child psychopathology in the Adolescent Brain Cognitive Development (ABCD) Study.

OBJECTIVE: Numerous adverse prenatal exposures have been individually associated with risk for psychiatric illness in the offspring. However, such exposures frequently co-occur, raising questions about their cumulative impact. We evaluated effects of cumulative adverse prenatal exposure burden on psychopathology risk in school-aged children. METHODS: Using baseline surveys from the U.S.-based Adolescent Brain and Cognitive Development (ABCD) Study (7,898 non-adopted, unrelated children from 21 sites, age 9-10, and their primary caregivers), we examined 8 retrospectively-reported adverse prenatal exposures in relation to caregiver-reported total and subscale Child Behavior Checklist (CBCL) scores. We also assessed cumulative effects of these factors on CBCL total as a continuous measure, as well as on odds of clinically significant psychopathology (CBCL total ≥60), in both the initial set and a separate ABCD sample comprising an additional 696 sibling pairs. Analyses were conducted before and after adjustment for 14 demographic and environmental covariates. RESULTS: In minimally and fully adjusted models, 6 exposures (unplanned pregnancy; maternal alcohol, marijuana, and tobacco use early in pregnancy; pregnancy complications; and birth complications) independently associated with significant but small increases in CBCL total score. Among these 6, none increased the odds of crossing the threshold for clinically significant symptoms by itself. However, odds of exceeding this threshold became significant with 2 exposures (OR = 1.86, 95% CI 1.47-2.36), and increased linearly with each level of exposure (OR = 1.39, 95% CI 1.31-1.47), up to 3.53-fold for ≥4 exposures versus none. Similar effects were observed in confirmatory analysis among siblings. Within sibling pairs, greater discordance for exposure load associated with greater CBCL total differences, suggesting that results were not confounded by unmeasured family-level effects. CONCLUSION: Children exposed to multiple common, adverse prenatal events showed dose-dependent increases in broad, clinically significant psychopathology at age 9-10. Fully prospective studies are needed to confirm and elaborate upon this pattern.

Association of Prenatal Exposure to Population-Wide Folic Acid Fortification With Altered Cerebral Cortex Maturation in Youths.

Importance: Presently, 81 countries mandate the fortification of grain products with folic acid to lessen the risk of neural tube defects in the developing fetus. Epidemiologic data on severe mental illness suggest potentially broader effects of prenatal folate exposure on postnatal brain development, but this link remains unsubstantiated by biological evidence. Objective: To evaluate associations among fetal folic acid exposure, cortical maturation, and psychiatric risk in youths. Design, Setting, and Participants: A retrospective, observational clinical cohort study was conducted at Massachusetts General Hospital (MGH) among 292 youths 8 to 18 years of age born between January 1993 and December 2001 (inclusive of folic acid fortification rollout ±3.5 years) with normative results of clinical magnetic resonance imaging, divided into 3 age-matched groups based on birthdate and related level of prenatal folic acid fortification exposure (none, partial, or full). Magnetic resonance imaging was performed between January 2005 and March 2015. Two independent, observational, community-based cohorts (Philadelphia Neurodevelopmental Cohort [PNC] and National Institutes of Health Magnetic Resonance Imaging Study of Normal Brain Development [NIH]) comprising 1078 youths 8 to 18 years of age born throughout (PNC, 1992-2003) or before (NIH, 1983-1995) the rollout of folic acid fortification were studied for replication, clinical extension, and specificity. Statistical analysis was conducted from 2015 to 2018. Exposures: United States-mandated grain product fortification with folic acid, introduced in late 1996 and fully in effect by mid-1997. Main Outcomes and Measures: Differences in cortical thickness among nonexposed, partially exposed, and fully exposed youths (MGH) and underlying associations between age and cortical thickness (all cohorts). Analysis of the PNC cohort also examined the association of age-cortical thickness slopes with the odds of psychotic symptoms. Results: The MGH cohort (139 girls and 153 boys; mean [SD] age, 13.3 [2.3] years) demonstrated exposure-associated cortical thickness increases in bilateral frontal and temporal regions (9.9% to 11.6%; corrected P < .001 to P = .03) and emergence of quadratic (delayed) age-associated thinning in temporal and parietal regions (ß = -11.1 to -13.9; corrected P = .002). The contemporaneous PNC cohort (417 girls and 444 boys; mean [SD] age, 13.5 [2.7] years) also exhibited exposure-associated delays of cortical thinning (ß = -1.59 to -1.73; corrected P < .001 to P = .02), located in similar regions and with similar durations of delay as in the MGH cohort. Flatter thinning profiles in frontal, temporal, and parietal regions were associated with lower odds of psychosis spectrum symptoms in the PNC cohort (odds ratio, 0.37-0.59; corrected P < .05). All identified regions displayed earlier thinning in the nonexposed NIH cohort (118 girls and 99 boys; mean [SD] age, 13.3 [2.6] years). Conclusions and Relevance: The results of this study suggest an association between gestational exposure to fortification of grain products with folic acid and altered cortical development and, in turn, with reduction in the risk of psychosis in youths.

A Cross-Sectional Study of Dietary and Genetic Predictors of Blood Folate Levels in Healthy Young Adults.

Since 1998, the U.S. has mandated folic acid (FA) fortification of certain grain products to reduce the risk of neural tube defects. Folate intake and red blood cell (RBC) folate concentrations increased substantially post-intervention, although recent studies raise concerns about the level of ongoing benefit. This study investigated blood folate level determinants in healthy young adults, including intake of naturally occurring food folate, synthetic FA, and the interaction of naturally occurring food folate with a common missense variant in the FOLH1 gene thought to affect absorption. Participants (n = 265) completed the Diet History Questionnaire II, RBC folate testing, and were genotyped for the 484T>C FOLH1 variant. Men reported significantly greater intake of all folate sources except for supplemental FA, but RBC folate levels did not significantly differ by sex. Synthetic FA was a stronger predictor of RBC folate than naturally occurring food folate. In the largest racial group, synthetic FA and the interaction of FOLH1 genotype with naturally occurring food folate significantly predicted RBC folate, with the overall model accounting for 13.8% of the variance in RBC folate levels. Blood folate levels rely on a complex interaction of natural and synthetic folate intake as well as FOLH1 genotype.