Findings of the current investigation are consistent with previous reports of lower mI/Cr in the basal ganglia [24] and ACC [26] of young MRJ-users compared to NU participants, substantiating previous global findings [27] to now include lower mI/Cr levels in the lThal. No other significant group differences in metabolite levels were observed in any of the other regions examined. Lower levels of lThal mI/Cr in MRJ-dependent participants were associated with greater cognitive impulsivity on the BIS-11, but no other measures. The cognitive impulsivity subscale of the BIS-11 includes items such as “I like to think about complex problems” and “I am a careful thinker” [28]. A number of studies have utilized the cognitive impulsivity subscale as a reliable construct of rapid decision-making and ability to focus attention (e.g. [28,34]). Overall, the present lThal mI/Cr findings are consistent with the interpretation that mI/Cr levels and cognitive focus are potentially altered by chronic MRJ use [27].

Anandamide (arachidonylethanolamide), an endogenous functional agonist at the endocannabinoid CB1 receptor, is widely distributed in the thalamus [35], suggesting that CB1 receptor-mediated changes in synaptic activity within the thalamus may have direct consequences for the complex transmission of information to the cortex [36]. As THC also acts as an agonist at CB1 receptors, it is possible that chronic MRJ consumption may alter the proper funneling of information through thalamo-cortical networks. In the present study, MRJ-related metabolite differences were regionally specific to the lThal. Asymmetry in thalamic biochemical [37] and functional processing [38] has been previously established. Chronic MRJ use has been shown to interfere with verbal working memory [39], which is a cognitive process hypothesized to be lateralized to left hemispheric regions that includes the left thalamus [40]. Moreover, the left thalamus is also hypothesized to be a critical component of left-brain lateralized cognitive inhibition circuitry [41]. The thalamo-cortical circuit that includes the thalamus and frontal regions such as the DLPFC and ACC is involved in inhibitory processing [14] and is thought to suppress or diminish inappropriate behavioral responses. Left hemisphere dysfunction in regions like the thalamus can interfere with cognitive inhibition circuitry, thereby potentially elevating attention and behavioral problems, such as impulsivity [41]. Thus, future studies should investigate the possibility that left-hemisphere-mediated cognitive functions, such as verbal working memory and behavioral inhibition, and associated left-hemisphere regional metabolite levels are uniquely vulnerable to MRJ-induced dysfunction in male and female cohorts.

Present findings are consistent with previous reports documenting significant relationships between greater levels of impulsivity and MRJ use [6,27,42]. The observed relationships also extend previous findings to suggest that MRJ-related reductions are specific to lThal mI/Cr levels, which were significantly associated with greater cognitive impulsivity. mI is a glial marker that is actively transported into astroglia that are involved in neuronal metabolism, neurotransmitter synthesis, and regulating cellular volume and energy consumption [23]. Receptors for neurotransmitters such as glutamate and dopamine are present on the surface of astroglia, and are therefore directly involved in neural signaling [43,44] and neurotransmission [45]. Reductions in mI levels are suggestive of regionally altered glial metabolism, glial dysfunction, or glial loss [22] in the lThal in MRJ users, and could also reflect altered regional neurotransmission, potentially affecting thalamo-cortical signaling. Disrupted thalamo-cortical signaling may interfere with left-brain lateralized cognitive inhibitory processes by reducing astroglia-mediated energy metabolism [46]. One outcome of diminished inhibitory processing in MRJ users may be elevated cognitive impulsivity. Decreased lThal mI levels in MRJ-dependent participants may also reflect decreased intracellular mI concentration, which could interfere with glial function in neurotrophic signaling and neuroplasticity in that region [47]. The possibility, however, those differences in lThal mI levels and the relationship between lThal mI and cognitive impulsivity may have preceded the onset of MRJ use and/or served catalytically to initiate and maintain MRJ use cannot be ruled out.

There are strengths and limitations of the study that warrant attention when evaluating these preliminary findings. Notable strengths of the study included collection of multiple high field (4T) spectra using 2D J-resolved MRSI that were combined with the use of GAMMA-simulated basis sets for LCModel fitting to generate a precise, theoretically-correct template for each spectral extraction. Moreover, the MRSI technique permitted voxels to be accurately positioned in the Thal, TC, and POC and minimized the number of spectra collected with poor signal to noise quality due to interference by subcutaneous lipids. Partial volume effects were also determined from the voxels of interest using tissue segmentation. Lastly, to eliminate any potential limitations associated with self-reported drug use, same day urine samples were collected and tested to determine urinary THC/Cr levels, which serve as a significant objective measure to confirm recent MRJ use.

Given the modest sample size in this study and the inclusion of only male MRJ users, these findings require replication with a larger sample size that includes female MRJ users. Additionally, given that this study is a retrospective analysis of data from an earlier preliminary investigation [27], significance levels for ANCOVAs and correlations were not corrected for multiple comparisons. Notably, a large effect size was achieved for the lThal difference. In addition, post hoc power analyses revealed that only an additional five subjects, yielding a total sample size of 25 subjects, would be needed to survive a conservative Bonferroni correction. MRJ users in the present study were well characterized as young chronic users and clinically diagnosed with marijuana dependence without meeting criteria for any other substance abuse disorders; however, additional studies that include female MRJ users will help to clarify the generalizability of the relationship between lThal mI levels and cognitive impulsivity. Furthermore, future studies should also explore if concurrent alcohol and marijuana dependence produce measurable changes in metabolite levels relative to marijuana dependence alone. In the present study, while the MRJ and NU groups significantly differed on the number of alcoholic drinks consumed per week, including drinks per week in metabolite analyses as a covariate did not alter the lThal mI/Cr ratio findings. Finally, studies with larger sample sizes may improve the ability to detect additional metabolite alterations in brain regions that are known to be affected by long-term MRJ use.

Total Cr was used to determine metabolite ratios in the current study, which has its own inherent limitations. There is some controversy over the optimal internal reference sample for quantitation of spectral signal intensities. Though Cr levels, a common internal reference, generally reflect little inter-individual variation [33], it is possible that absolute concentrations of Cr may differ under pathological conditions or possibly due to overlap between the Cr signal and other compounds [48]. Thus, Cr was used as an internal reference only after analyses indicated that ratios of Cr/total signal did not differ between groups in any region except in the rTC. Metabolite data from the rTC were generally of insufficient quality, due to signal loss, which has been reported as a potential outcome in imaging medial temporal lobe regions that are particularly subject to differences in magnetic susceptibility [49]. It is possible in the present study that total signal intensity in the rTC was inconsistent across participants, which may reflect individual variability in temporal lobe signal distortion and loss that is in line with previous reports [49]. Subsequent analyses revealed that there were no significant differences in any ratios of metabolites divided by the Cr integral in the rTC. No group differences were observed in the number of voxels with high spectral quality or in the overall tissue percentages across all voxels, suggesting that it is improbable that the lThal-specific mI/Cr reduction in MRJ users was moderated by these measures. Lastly, the self-reported measures of impulsivity and mood state, in addition to drug use, collected in the present study have their own limitations. Self-report can include imprecise or under-reporting of drug use as well as individual differences in response style or interpretation of behavioral or mood questions that are asked in the assessment [50]. Nonetheless, urinary THC/Cr levels provided objective corroboration of self-reported MRJ use in the present study.

In conclusion, this exploratory investigation takes advantage of the ability of MRSI at high field to examine regional specificity of MRJ effects on brain chemistry, and as a result, revealed that young MRJ-dependent men exhibited lower mI/Cr levels in the lThal. Furthermore, these lower lThal mI/Cr levels were a significant neurobiological correlate of greater cognitive impulsivity in MRJ users, reflecting associations that were distinct from those measured in non-MRJ-using age-matched study participants. The present regional findings support earlier work in the same study cohort, which demonstrated significantly greater global mI/Cr levels and less distribution of this glial metabolite in WM in young MRJ-dependent men relative to non-MRJ-using participants. Overall, findings suggest that the lThal may be uniquely vulnerable to long-term MRJ use, and that alterations in lThal mI/Cr levels may influence thalamo-cortical signaling, and in turn, left-lateralized cognitive inhibition circuitry. Furthermore, the present finding of localized metabolite alterations to the lThal indicates that this region may be distinctly impacted by long-term MRJ use and is associated with cognitive impulsivity. Thus, future functional and neurobiological imaging studies, as well as neuropsychological assessments conducted in healthy and MRJ-using populations will be imperative for elucidating the physiological significance of the current exploratory findings.