Prefrontal GABA levels, hippocampal resting perfusion and the risk of psychosis

Gemma Modinos,Fatma Şimşek,Ilaria Bonoldi,Oliver D Howes,Anthony A Grace,Philip Mcguire,David J Lythgoe,Jesus Perez,Matthew R Broome,Matilda Azis,James M Stone,Fernando Zelaya,Matthijs Bossong,Beverly Quinn,Paul Allen,Carly Samson

doi:10.1038/s41386-017-0004-6

Abstract

Preclinical models propose that the onset of psychosis is associated with hippocampal hyperactivity, thought to be driven by cortical GABAergic interneuron dysfunction and disinhibition of pyramidal neurons. Recent neuroimaging studies suggest that resting hippocampal perfusion is increased in subjects at ultra-high risk (UHR) for psychosis, but how this may be related to GABA concentrations is unknown. The present study used a multimodal neuroimaging approach to address this issue in UHR subjects. Proton magnetic resonance spectroscopy and pulsed-continuous arterial spin labeling imaging were acquired to investigate the relationship between medial prefrontal (MPFC) GABA+ levels (including some contribution from macromolecules) and hippocampal regional cerebral blood flow (rCBF) in 36 individuals at UHR of psychosis, based on preclinical evidence that MPFC dysfunction is involved in hippocampal hyperactivity. The subjects were then clinically monitored for 2 years: during this period, 7 developed a psychotic disorder and 29 did not. At baseline, MPFC GABA+ levels were positively correlated with rCBF in the left hippocampus (region of interest analysis, p = 0.044 family-wise error corrected, FWE). This correlation in the left hippocampus was significantly different in UHR subjects who went on to develop psychosis relative to those who did not (p = 0.022 FWE), suggesting the absence of a correlation in the latter subgroup. These findings provide the first human evidence that MPFC GABA+ concentrations are related to resting hippocampal perfusion in the UHR state, and offer some support for a link between GABA levels and hippocampal function in the development of psychosis.

Highlights

Post-mortem and preclinical studies have provided consistent evidence that the pathophysiology of psychotic disorders involves an alteration in GABAergic neurotransmission [1, 2]
While we found no group differences in the standalone GABA 1H-MRS or hippocampal regional cerebral blood flow (rCBF) measures, the GABA+ correlation with hippocampal perfusion appeared to be driven by the ultra-high risk (UHR) subjects who subsequently developed a psychotic disorder: there was a strong correlation in this subgroup, but no correlation in the UHR subjects who did not develop psychosis
Independent work using positron emission tomography indicates that subcortical dopamine function is increased in psychosis [50] and in UHR subjects [51,52,53,54], and that the level of increase in UHR subjects is linked to the later onset of psychosis [55]

Summary

Introduction

Post-mortem and preclinical studies have provided consistent evidence that the pathophysiology of psychotic disorders involves an alteration in GABAergic neurotransmission [1, 2]. One study reported higher levels in subjects at UHR compared with controls [21]; one described lower GABA levels in UHR subjects than in controls [22], and two studies including a recent one from our group found no differences between UHR subjects and healthy controls [23, 24].

Objectives

Results

Conclusion