INSIGHT of Medicine(phase 26,2024)


1.Native-state proteomics of Parvalbumin interneurons identifies unique molecular signatures and vulnerabilities to early Alzheimer's pathology

Dysfunction in fast-spiking parvalbumin interneurons (PV-INs) may represent an early aspect of Alzheimer's Disease (AD) pathology. This study utilized cell-type-specific biotinylation coupled with mass spectrometry to characterize native-state PV-IN proteomes. PV-INs showed heightened metabolic and translational activity, with enrichment of AD-risk and cognitive resilience-related proteins. In human and mouse models, PV-IN proteins correlated with cognitive decline and neuropathology. Early Aβ pathology revealed PV-IN proteomic signatures indicating increased mitochondria, synaptic disruption, and decreased mTOR signaling, not evident in whole-brain proteomes. Pre-synaptic defects in PV-to-excitatory neurotransmission validated these findings. The study sheds light on PV-IN roles in cognitive resilience and AD pathogenesis, offering valuable molecular insights for translational research.

2.Tomosyns attenuate SNARE assembly and synaptic depression by binding to VAMP2-containing template complexes

Tomosyn proteins were previously believed to inhibit membrane fusion by competing with synaptobrevin-2/VAMP2 for SNARE-complex assembly. However, new evidence contradicts this theory. Tomosyn-1/2 deficiency in mice lowered the fusion barrier, resulting in stronger synapses with faster depression and slower recovery. Wild-type tomosyn-1m rescued these effects, while a SNARE motif substitution did not. Single-molecule force measurements revealed that tomosyn's SNARE motif cannot substitute synaptobrevin-2/VAMP2. Instead, tomosyns bind to synaptobrevin-2/VAMP2, preventing SNAP-25 association. Structure-function analyses suggest the C-terminal polybasic region contributes to tomosyn's inhibitory function. These findings unveil a new mechanism where tomosyns regulate synaptic transmission by cooperating with synaptobrevin-2/VAMP2, limiting initial synaptic strength and equalizing it during repetitive stimulation.

3.Self-help mobile messaging intervention for depression among older adults in resource-limited settings: a randomized controlled trial

DOI: 10.1038/s41591-024-02864-4

The PRODIGITAL-D trial evaluated a mobile messaging psychosocial intervention, "Viva Vida," for treating depression in older adults in socioeconomically deprived areas of Guarulhos, Brazil. Participants aged 60+ with depressive symptoms received either the 6-week intervention or a single message without professional support. At 3 months, 42.4% in the intervention group showed improved depressive symptoms, compared to 32.2% in the control group. No severe adverse events were reported. These findings highlight the effectiveness of digital messaging interventions for short-term improvement in depressive symptoms among older adults, suggesting their potential integration into primary care programs for depression treatment.

4.Whole-exome sequencing in UK Biobank reveals rare genetic architecture for depression

DOI: 10.1038/s41467-024-45774-2

This study explored the impact of rare coding variants on depression using whole-exome sequencing in 320,356 UK Biobank participants. Seven depression definitions were examined based on surveys, questionnaires, and health records. Rare damaging coding variants in loss-of-function intolerant genes were associated with depression risk across various definitions. Genetic correlation analyses revealed distinct relationships between common and rare variants across depression definitions. Additionally, the effects of rare variants and polygenic risk scores on depression risk were additive. Gene set burden analyses indicated shared rare genetic components with developmental disorders, autism, and schizophrenia. These findings shed light on the role of rare coding variants, alone and combined with common variants, in depression risk and their genetic links with neurodevelopmental disorders.

5.Clinical decision support for bipolar depression using large language models

DOI: 10.1038/s41386-024-01841-2

This study investigated the use of large language models (LLMs) in guiding pharmacotherapy decisions for bipolar depression. Clinical vignettes were presented to bipolar disorder experts and LLMs, with or without augmentation by treatment guidelines. The augmented LLM identified optimal treatments for 50.8% of cases, outperforming the un-augmented LLM (23.0%) and community clinicians (23.1%). The augmented LLM's performance was consistent across demographic permutations, suggesting reduced bias risk. This approach presents a scalable strategy for clinical decision support. However, strategies to mitigate clinician overreliance on LLMs and address potential biases are warranted alongside prospective efficacy studies.