A GWAS of opioid response in a Japanese sample found that the C allele of rs2952768, which is in an LD block with rs7591784 , was significantly associated with greater postoperative opioid analgesic requirements, as well as lower reward dependence in healthy volunteers, lower risk of polydrug use in volunteers with methamphetamine dependence, alcohol dependence, and eating disorders, and increased expression of CREB1 in human postmortem brains . The G allele in rs7591784 was associated in our study with lower pretreatment methamphetamine use and better treatment outcomes, both suggestive of less severe methamphetamine use disorder, and as rs7591784 and rs2952768 are strongly linked, our results provide support for the previous association between the C allele of rs2952768 and lower severity of methamphetamine use disorder observed in the Japanese GWAS. Whether rs7591784 directly effects CREB expression or function is not known, but our results and previous studies suggest that variability in CREB signaling and subsequent changes in methamphetamine-induced gene expression may influence clinical severity of methamphetamine use problems and success in quitting methamphetamine and that the CREB signaling pathway may be a target for the development of medications to treat methamphetamine use disorder. Phosphodiesterase inhibitors modulate signaling via the CREB pathway via increases in cAMP and ibudilast, a nonselective phosphodiesterase inhibitor,vertical farming company is in clinical development for methamphetamine use disorder . Previous GWAS found SNPs in CDH13 to be among the most significant SNPs associated with a diagnosis of methamphetamine dependence and with the subjective response to amphetamine among healthy volunteers . None of the SNPs related to CDH13 in our study were significantly associated with methamphetamine use frequency following Bonferroni correction. The lack of significant association in our study may be due to the different phenotypes examined in the previous GWAS compared to the current study that examined methamphetamine use frequency in a treatment-seeking sample or may be due to limited power to detect SNPs with small effect size in our small sample. Methamphetamine use frequency as well as results of our SNP analyses differed greatly between males and females.
None of the three SNPs that were nominally significant in males, including rs7591784, approached significance in females suggesting that although the female sample size was relatively small, the lack of significant associations for these SNPs in females is unlikely to be due to limited power in females alone. Previous studies in rodents have found sex differences in methamphetamine pharmacokinetics , methamphetamine-induced plasma corticosterone levels , methamphetamine-related neurotoxicity , and methamphetamine self-administration with female rats acquiring methamphetamine self-administration faster, self-administering more methamphetamine, and exhibiting higher rates of methamphetamine reinstatement than male rats . In humans, female methamphetamine users have a higher risk of Parkinson’s disease , greater reductions in hippocampal volume and higher prevalence of physiologic dependence symptoms compared to male methamphetamine users and these biological or other psycho social differences may have a greater influence on methamphetamine use frequency in females than the SNPs examined here. Interestingly, amphetamine-induced CREB-mediated transcription differs dramatically between male and female mice in the nucleus accumbens, ventral tegmental area, amygdala, and locus coeruleus with greater CREB-meditated gene transcription following amphetamine in females suggesting that the significant association between rs7591784 and methamphetamine-related phenotypes observed in our study in males but not females may be due to underlying sexual dimorphism in the CREB signaling pathway. The one SNP that was nominally associated with methamphetamine use frequency assuming an additive model in females, rs163030, was associated with caudate volume in a GWAS and rs163030 may influence methamphetamine use frequency in females by altering structure or functioning of the caudate, a brain region implicated in impulsivity and methamphetamine addiction . Additional studies investigating sex differences in the biological and social influences on methamphetamine addiction are warranted. This study has several limitations. The sample size is small and the power to detect an association between a candidate SNP and methamphetamine use frequency with a small effect size is limited.
As a result the study is subject to false negative results. Also, numerous findings from candidate gene studies have failed to replicate and results from this study are preliminary and require replication in an independent sample prior to making any conclusions. To mitigate this risk, we emphasized selection of candidate SNPs that had previously been associated with methamphetamine-relevant phenotypes in GWAS. Our study did not genotype rs2709386, which was most strongly associated with opioid sensitivity in the previous Japanese GWAS, and although rs2709386 and rs7591784 are highly linked, future studies are necessary to determine which SNP is more strongly associated with methamphetamine use and treatment outcomes. Lastly, the sample was drawn participants of several methamphetamine pharmacotherapy clinical trials and results from a treatment-seeking sample may not be generalizable to methamphetamine uses as a whole. In summary, we found an association between rs7591784 near CREB1 and pretreatment methamphetamine use, an important indicator of disease severity and predictor of subsequent treatment outcomes, as well as methamphetamine use during treatment independent of pretreatment methamphetamine use in males but not females. Replication of this result in independent samples is necessary but our results combined with previous research suggest that variability in CREB signaling may influence severity of methamphetamine use disorder as well as success in quitting methamphetamine with outpatient treatment and that medications targeting the CREB pathway such as the non selective phosphodiesterase inhibitor ibudilast may be effective treatments for methamphetamine use disorder. Future studies should examine the role of CREB-related polymorphisms and the associated epigenetic changes on response to treatment for methamphetamine use disorder and whether these biological influences on methamphetamine use differ between males and females.Worldwide, it is estimated that there are over three million youth living with HIV globally, with the majority of youth acquiring HIV perinatally . Youth with perinatally-acquired HIV may show cognitive deficits as well as developmental delay even among those with reconstituted immunologic and virologic status, making PHIV a common infectious cause of perinatally-acquired developmental disability globally . Combination antiretroviral therapy for children with PHIV has resulted in substantial improvements in health with survival beyond childhood and reductions in morbidity and mortality .
Early HIV infection, immune activation, and viral persistence during a critical period of development may be especially detrimental to developing brains in youth with PHIV . Brain development is an extended process that begins prenatally and continues throughout the first two decades of life,indoor vertical farming with increased sensitivity to experience during the first year of life in pathways responsible for sensory, language and higher order cognitive development . Adolescence is also a crucial developmental window marked by a period of rapid brain maturation via synaptic pruning and myelination. White matter volume increases while grey matter volume decreases , with parietal grey matter reduction prominent before adolescence, followed by dorsal, mesial, and orbital frontal grey matter reduction during and after adolescence . Studies including neuroimaging combined with cognitive evaluation allow for an in vivo characterization of how HIV and cART may mediate brain development . In adults, studies of post-mortem tissue and in vivo neuroimaging combined with cognitive testing have revealed atrophy in cortical and subcortical structures that is related to HIV severity and cognitive performance . Still, the effects of early HIV infection on the underlying brain in adolescents with PHIV have not been well-characterized . Adolescent brains are also subject to environmental influences, including substance use. Neuroimaging and neuropsychological studies in youth who use substances have found structural brain abnormalities, including grey matter volume reductions, as well as cognitive dysfunction .Thus, to carefully study effects of PHIV on youth treated with cART, it is important to account for substance use. We present one of the first studies to investigate the impact of HIV severity and coincident substance use on regional and total brain volumes and their association with cognition in PHIV youth. Other studies on grey matter volumes in PHIV do not focus on regional grey matter or substance use or are in PHIV populations with varying clinical characteristics from our cohort . Since PHIV youth often exhibit global cognitive functioning, working memory, and processing speed deficits , we hypothesized that frontal and parietal regions, regions important for higher-order cognitive functioning, would show volume reduction as compared to typically-developing, HIV unexposed and uninfected youth and smaller volumes would be associated with worse cognitive performance. We also hypothesized that HIV disease severity and substance use would be associated with reduced cortical grey matter volumes among adolescents with PHIV. 40 PHIV youth from a single site participating in the Adolescent Master Protocol study of the NIH Pediatric HIV/ AIDS Cohort Study network were recruited. Institutional review board approvals were obtained. Parents, legal guardians, or youth aged 18 years or older provided written informed consent; minors provided written assent. A control group of 334 typically developing, HIV-unexposed and uninfected youth was generated using frequency-matching for sex and age from the Pediatric Imaging, Neurocognition, and Genetics study Magnetic Resonance Imaging database from five sites . Of note, information regarding alcohol and drug use was not collected in the PING cohort. Total grey matter and 10 a priori cortical ROIs volumes were identified as primary outcomes. The remaining 74 ROIs were analyzed as secondary outcomes. Descriptive statistics and graphical methods were used to confirm the normality assumption for volume measures. Volumes were compared between PHIV youth and HIV-unexposed and uninfected youth using linear regression models. Results were reported for models with and without adjusting for age at scan, sex, race, caregiver education attainment, annual household income , and intracranial volume. Caregiver education was classified as high school education and below vs. greater than high school. Annual household income was classified as $30,000 and below vs. greater than $30,000. Percent change in volume as compared to HIV-unexposed and uninfected youth was calculated based on adjusted means.For evaluating associations of volume with cognitive functioning, we considered the 10 primary ROIs as well as total grey matter. Linear regression was used to evaluate associations between ROI volume as well as total grey matter and working memory, processing speed, and cognitive proficiency indices, adjusting for sex and age at scan. Three sets of sensitivity analyses were conducted: 1) including substance use in models of associations of brain volumes with HIV measures for grey matter volumes; 2) adjusting for total grey matter volume in models evaluating associations between brain volumes and substance use for secondary ROIs; 3) including substance use, sex, and age at scan in analyses evaluating associations of brain volumes with cognition. As described previously for other neuroimaging studies in this cohort , for PHIV youth, the mean interval between scanning and assessment of recent disease markers was 1.8 months with 83% of recent VL measures within three months prior to scanning. Mean interval between scanning and cognitive testing was 3.8 months . All but 2 participants completed cognitive assessments within 1 year of neuroimaging, with 31 within 3 months. One fourth of PHIV youth reported tobacco use, 35% alcohol, and 35% marijuana use. 13 youth reported both alcohol and marijuana use, and 8 youth reported use of tobacco, alcohol, and marijuana. Due to small numbers reporting illicit drug use beyond marijuana , this measure was not considered further in statistical analyses . There was no difference in substance use by race. Adjustment for substance use in models evaluating associations between primary ROI volumes and HIV disease severity as well as cognitive function did not alter findings. Similarly, adjustment for total grey matter volume in models evaluating associations of HIV severity and substance use measures with secondary ROI volumes did not alter findings, with no associations identified based on a FDR level of 0.10. Positive associations observed between primary brain volumes and cognitive function persisted after adjustment for substance use. Bilateral superior frontal volumes positively correlated with CPI, WMI, and PSI after adjustment for substance use.The current study is one of the first to examine relationships among PHIV infection, disease severity, and substance use on the brain grey matter and cognitive outcomes in PHIV youth. We found that PHIV youth had reduced total grey matter volume as well as reduced volumes in the rostral middle frontal, postcentral, precentral, and superior parietal gyri, compared to similarly-aged HIV-unexposed and uninfected youth. These patterns persisted after adjusting for sex, age at scan, race, caregiver education attainment, annual household income, and intracranial volume.