Mapping Brain Glutamate in Humans: Sex Differences in Cigarette Smokers

Glutamatergic signaling is dysregulated in addictions, including Tobacco Use Disorder (TUD) , and represents a promising target for smoking cessation therapies. In rodents, NMDA-type glutamate (Glu) receptor antagonists reduce nicotine-seeking behavior, whereas pre-treatment with NMDA-type and AMPA-type Glu receptor antagonists attenuates nicotine self-administration and nicotine-induced dopamine release. In humans, N-acetylcysteine, which regulates Glu via the cysteine-glutamate antiporter and glial Glu transporter, reduces cigarette smoking9 as well as cortical and subcortical levels of Glu itself, measured by magnetic resonance spectroscopy (MRS). These observations suggest that Glu, assayed by MRS, which is a safe and non-invasive in vivo metric, may aid in tracking regional effects of smoking and of anti-smoking interventions.

Notably, evidence has been provided that Glu in the dorsal anterior cingulate cortex (dACC) is associated with smoking-related states, that Glx (the sum of Glu and its primary metabolite, glutamine, Gln) in dACC is higher in smokers than nonsmokers, and that Glx in the insula is higher after overnight abstinence from smoking than during satiety. Further, studies in smokers combining functional magnetic resonance imaging (fMRI) with MRS have linked dACC Glu with activation of the Default Mode Network during cue-induced cigarette craving, and demonstrated reduced dACC Glx and altered dACC-to-DMN connectivity following varenicline treatment. Prior MRS studies of smoking, however, have been limited by single-voxel techniques that sample single brain regions with restricted spatial-resolution and partial volume effects (i.e., including more than one tissue type in the acquisition voxel).

We will exploit the echo-planar spectroscopic imaging (EPSI) variant of MRS in a high-density, anatomically broad assessment of how sex, acute smoking, and ovarian hormones affect brain Glu. Our pilot data indicate that dACC Glu is lower in smoking-abstinent women than men, and lower still in women with higher serum estrogen. After smoking, dACC Glu decreases in men but not in women.

We will use EPSI to determine whether our dACC Glu pilot findings: 1) replicate at high spatial resolution; 2) pertain to the anterior insula and thalamus, which are also implicated in TUD, and 3) on an exploratory basis, pertain to other regions in the rest of the brain. To accomplish this, we will use a recently developed short echo-time (TE=20 ms) variant of 3D EPSI at 3-Tesla. With this state-of-the-art technique, we typically acquire high-quality spectra simultaneously from ~80% of the brain at the very high spatial resolution of ~0.4 cc. Short-TE, moreover, improves segregation of Glu from the overlapping Gln signal.

We teste adult daily smokers (men and women) before and after their first cigarette of the day after ~12 h abstinence. Serum estrogen and progesterone in women were assayed. We addressed two specific aims: