Comorbidity and Coaggregation of Major Depressive Disorder and Bipolar Disorder and Cannabis Use Disorder in a Controlled Family Study.

Importance: Cannabis use disorder (CUD) is increasing in the US. Clarification of the potential mechanisms underlying the comorbidity between mood disorders and CUD may help prevent CUD. Objective: To examine co-occurrence and familial aggregation of CUD and mood disorder subtypes. Design, Setting, and Participants: In this cross-sectional, community-based study in the Washington, DC, metropolitan area, semistructured diagnostic interviews and family history reports assessed lifetime DSM-IV disorders in probands and relatives. Familial aggregation and coaggregation of CUD with mood disorders were estimated via mixed-effects models, adjusting for age, sex, recruitment source, and comorbid mood, anxiety, and other substance use disorders. A total of 586 adult probands (186 with bipolar disorder; 55 with CUD) and 698 first-degree relatives (91 with bipolar disorder; 68 with CUD) were recruited from a community screening of the greater Washington, DC, metropolitan area from May 2004 to August 2020. Inclusion criteria were ability to speak English, and availability and consent to contact at least 2 living first-degree relatives. Main Outcomes and Measures: Lifetime CUD in first-degree relatives. Results: Of 586 probands, 395 (67.4%) were female; among 698 relatives, 437 (62.6%) were female. The mean (SD) age was 47.5 (15.2) years for probands and 49.6 (18.0) years for relatives. In the proband group, 82 participants (14.0%) self-identified as African American or Black, 467 (79.7%) as White, and 37 (6.3%) as American Indian or Alaska Native, Asian, more than one race, or another race or ethnicity or declined to respond. In the relative group, 53 participants (7.6%) self-identified as African American or Black, 594 (85.1%) as White, and 51 (7.3%) as American Indian or Alaska Native, Asian, more than one race, or another race or ethnicity or declined to respond. These groups were combined to protect privacy owing to small numbers. CUD in probands (55 [9.4%]) was associated with an increase in CUD in relatives (adjusted odds ratio [aOR], 2.64; 95% CI, 1.20-5.79; P = .02). Bipolar disorder II (BP-II) in probands (72 [12.3%]) was also associated with increased risk of CUD in relatives (aOR, 2.57; 95% CI, 1.06-6.23; P = .04). However, bipolar disorder I (114 [19.5%]) and major depressive disorder (192 [32.8%]) in probands were not significantly associated with CUD in relatives. Among relatives, CUD was associated with BP-II (aOR, 4.50; 95% CI, 1.72-11.77; P = .002), major depressive disorder (aOR, 3.64; 95% CI, 1.78-7.45; P < .001), and mean (SD) age (42.7 [12.8] years with CUD vs 50.3 [18.3] years without CUD; aOR, 0.98; 95% CI, 0.96-1.00; P = .02). Familial coaggregation of BP-II with CUD was attenuated by the inclusion of comorbid anxiety disorders. Further, rates of CUD were highest in relatives with both a familial and individual history of BP-II (no familial or individual history of BP-II: 41 [7.2%]; familial history but no individual history of BP-II: 13 [19.1%]; individual history but no familial history of BP-II: 10 [22.2%]; familial and individual history of BP-II: 4 [28.6%]; Fisher exact test, P < .001). The onset of mood disorder subtypes preceded CUD in probands and relatives in most cases. Conclusions and Relevance: The findings confirmed a familial aggregation of CUD. The increase in risk of CUD among relatives of probands with BP-II suggests that CUD may share a common underlying diathesis with BP-II. Taken together with the temporal precedence of depression and mania with respect to CUD onset, these findings highlight a potential role for BP-II intervention as CUD prevention.

Divergent C-terminal motifs in Gα12 and Gα13 provide distinct mechanisms of effector binding and SRF activation.

The G12/13 subfamily of heterotrimeric guanine nucleotide binding proteins comprises the α subunits Gα12 and Gα13, which transduce signals for cell growth, cytoskeletal rearrangements, and oncogenic transformation. In an increasing range of cancers, overexpressed Gα12 or Gα13 are implicated in aberrant cell proliferation and/or metastatic invasion. Although Gα12 and Gα13 bind non-redundant sets of effector proteins and participate in unique signalling pathways, the structural features responsible for functional differences between these α subunits are largely unknown. Invertebrates encode a single G12/13 homolog that participates in cytoskeletal changes yet appears to lack signalling to SRF (serum response factor), a transcriptional activator stimulated by mammalian Gα12 and Gα13 to promote growth and tumorigenesis. Our previous studies identified an evolutionarily divergent region in Gα12 for which replacement by homologous sequence from Drosophila melanogaster abolished SRF signalling, whereas the same invertebrate substitution was fully tolerated in Gα13 [Montgomery et al. (2014) Mol. Pharmacol. 85: 586]. These findings prompted our current approach of evolution-guided mutagenesis to identify fine structural features of Gα12 and Gα13 that underlie their respective SRF activation mechanisms. Our results identified two motifs flanking the α4 helix that play a key role in Gα12 signalling to SRF. We found the region encompassing these motifs to provide an interacting surface for multiple Gα12-specific target proteins that fail to bind Gα13. Adjacent to this divergent region, a highly-conserved domain was vital for SRF activation by both Gα12 and Gα13. However, dissection of this domain using invertebrate substitutions revealed different signalling mechanisms in these α subunits and identified Gα13-specific determinants of binding Rho-specific guanine nucleotide exchange factors. Furthermore, invertebrate substitutions in the C-terminal, α5 helical region were selectively disruptive to Gα12 signalling. Taken together, our results identify key structural features near the C-terminus that evolved after the divergence of Gα12 and Gα13, and should aid the development of agents to selectively manipulate signalling by individual α subunits of the G12/13 subfamily.