Cortical GABA in Subjects at Ultra-High Risk of Psychosis: Relationship to Negative Prodromal Symptoms

Cortical GABA in Subjects at Ultra-High Risk of Psychosis: Relationship to Negative Prodromal... Background: Whilst robust preclinical and postmortem evidence suggests that altered GABAergic function is central to the development of psychosis, little is known about whether it is altered in subjects at ultra-high risk of psychosis, or its relationship to prodromal symptoms. Methods: Twenty-one antipsychotic naïve ultra-high risk individuals and 20 healthy volunteers underwent proton magnetic resonance imaging at 3T. Gamma-aminobutyric acid levels were obtained from the medial prefrontal cortex using MEGA- PRESS and expressed as peak-area ratios relative to the synchronously acquired creatine signal. Gamma-aminobutyric acid levels were then related to severity of positive and negative symptoms as measured with the Community Assessment of At-Risk Mental States. Results: Whilst we found no significant difference in gamma-aminobutyric acid levels between ultra-high risk subjects and healthy controls P ( = .130), in ultra-high risk individuals, medial prefrontal cortex GABA levels were negatively correlated with the severity of negative symptoms ( P = .013). Received: May 24, 2017; Revised: August 5, 2017; Accepted: August 18, 2017 © The Author(s) 2017. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons. org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is 114 properly cited. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 115 Conclusion: These findings suggest that gamma-aminobutyric acidergic neurotransmission may be involved in the neurobiology of negative symptoms in the ultra-high risk state. Keywords: psychosis, magnetic resonance spectroscopy, GABA, ultra-high risk of psychosis, negative symptoms Introduction Methods Converging evidence from postmortem and preclinical studies Procedures were approved by the Research Ethics Committee indicates that dysfunction of the gamma-aminobutyric acider - of King’s College London and South London and Maudsley NHS gic (GABAergic) neurotransmitter system plays a major role in Trust. All participants provided informed consent. the pathophysiology of schizophrenia (Marin, 2012). Postmortem Twenty-one UHR individuals and 20 healthy volunteers, all research has demonstrated decreased mRNA expression of glu - males aged 18to 30 years, were included. UHR psychopathology tamic acid decarboxylase and reduced density of fast-spiking was assessed using the Community Assessment of At-Risk parvalbumin-positive interneurons in a corticolimbic circuitry Mental States (CAARMS) ( Yung et  al., 2005). UHR inclusion cri- involving the prefrontal cortex and the amygdala in schizo - teria required the presence of one or more of the following: (1) phrenia (Lewis et  al., 2005; Akbarian and Huang, 2006; Benes, attenuated psychotic syndrome, (2) a brief psychotic episode 2010). Furthermore, animal models of psychosis suggest a link of <1 week duration that spontaneously remits without anti - between disrupted cortical GABAergic function and dysregula - psychotic medication/hospitalization, and (3) trait vulnerability tion of subcortical dopaminergic signaling characteristic of the (schizotypal personality disorder or a first-degree relative with disorder (Grace, 2010). Such models propose that inhibitory dis - psychosis) plus a marked decline in psychosocial functioning. ruption would underlie not only dopamine-dependent positive Healthy control subjects were recruited from the local com - symptoms of psychosis, but would also influence other neural munity. They were excluded if they had a personal or familial pathways (e.g., including the basolateral nucleus of the amyg - history of psychiatric disorder, neurological illness, or drug/ dala, or the medial prefrontal cortex [MPFC]) putatively involved alcohol dependence based on the DSM-V ( American Psychiatric in the development of the negative symptoms of psychosis Association, 1994). Current/past medication use and current/ (Grace, 2016). Moreover, the role of GABA in the development past use of tobacco and cannabis was assessed using a semis- of psychosis is further supported by preclinical evidence that tructured interview adapted from the Early Psychosis Prevention peripubertal (i.e., premorbid) pharmacological intervention on and Intervention Centre Drug and Alcohol Assessment Schedule GABA-Aα5 receptors prevents schizophrenia-like GABA cell loss (http://www.eppic.org.au). All subjects were safe for MRI, had an and blocks the development of psychosis-like features in adult IQ in the normal range as assessed using the Wechsler Adult rats (Du and Grace, 2013, 2016). Intelligence Scale-III (Velthorst et  al., 2013 ), and were antipsy- From animal and human postmortem studies, it may thus chotic-naïve, and none were taking benzodiazepines. be hypothesized that cortical GABAergic function is reduced in Subjects underwent H-MRS on a General Electric Signa schizophrenia and that this abnormality can be detected in the HDx TwinSpeed 3T scanner at the Centre for Neuroimaging premorbid stages of the disorder ( Modinos et al., 2015). However, Sciences, Institute of Psychiatry, Psychology & Neuroscience recent meta-analytical evidence from human imaging studies (King’s College London). GABA levels were obtained from the using proton magnetic resonance spectroscopy (H-MRS) did not MPFC using MEGA-PRESS, which incorporates a standardized show a significant difference in regional GABA levels between chemically selective suppression water suppression routine patients with schizophrenia and healthy volunteers ( Egerton (TE = 68 milliseconds, TR = 2000 milliseconds). For each acquisi- et al., 2017). Research in patients with schizophrenia is compli- tion, unsuppressed water reference spectra (16 averages) were cated by previous antipsychotic exposure and heterogeneity of also acquired. Shimming was optimized, with auto-prescan clinical subgroups (Kegeles et  al., 2012). In this context, stud - performed twice before each scan. The region of interest in the ies in subjects at ultra-high risk (UHR) of developing psychosis MPFC was prescribed from the midline sagittal localizer, and are a useful resource to investigate neurobiological correlates the center of the 40- × 35- × 20-mm region of interest was placed of psychosis-like characteristics without confounds associ - above the middle section of corpus callosum ( Figure 1A). Spectra ated with the use of antipsychotics or illness chronicity on the were analyzed using LCModel 6.3-1L with the basis set provided imaging data. The 3 available MPFC GABA studies in UHR indi- by its author Pr ( ovencher, 2016), which contained the metabo - viduals have also presented mixed results, including increases lites GABA, glutamine, glutamate, Glx (glutamate + glutamine), (de la Fuente-Sandoval et al., 2016), decreases (Menschikov et al., and N-acetyl-aspartate (NAA). We used Cramer-Rao minimum 2016), and no differences (Wang et  al., 2016) when compared variance bounds (CRLB) >20% as reported by LCModel, which with healthy controls. Nevertheless, heterogeneity of clinical are estimates of fit of the metabolite peaks, and signal-to-noise subgroups is also a potential confounder in UHR studies ( Fusar- ratio <8 to exclude poorly fitted metabolite peaks from statis - Poli et al., 2016), and only 1 study to date investigated associa - tical analysis (Mouchlianitis et al., 2016 ; Provencher, 2016). Data tions between GABA levels and severity of positive and negative from all 41 participants in the present study met these criteria. symptoms in this group (de la Fuente-Sandoval et al., 2016 ). The Metabolites were expressed as ratios relative to the synchron - present study sought to address these issues by using H-MRS ously acquired creatine signal from the unedited MEGA-PRESS in a homogenous sample of antipsychotic-naïve subjects at spectra. This is a well-established normalization procedure in UHR of psychosis to test the hypotheses that: (1) GABA levels clinical H-MRS studies that has been extensively used in pre - in the MPFC would be reduced in UHR subjects compared with vious studies of MPFC GABA levels in patients with schizophre - healthy controls ( Marin, 2012), and that (2) GABA levels would nia (Goto et  al., 2009; Ongur et  al., 2010; Kegeles et  al., 2012; be inversely related to the severity of positive and negative pr -o Marsman et  al., 2014)  and UHR subjects (Marenco et  al., 2016; dromal symptoms (Grace, 2016). Menschikov et al., 2016). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup 116 | International Journal of Neuropsychopharmacology, 2018 Figure 1. (A) Voxel placement on medial prefrontal cortex (MPFC) and representative sample H-MRS spectra. (B) MPFC metabolite levels by group. (C) Scatterplot of the significant association between CAARMS negative symptom severity and GABA/Cr levels (ß = -.556, t = -2.761, R  = .310, P = .013). (D) Scatterplot of the nonsignificant associations between CAARMS positive symptom severity and GABA/Cr levels ( P  =  .298). Cr, creatine; Glu, glutamate; Glx, glutamate + glutamine; NAA, N-Acetyl- aspartate; UHR, ultra-high risk of psychosis. Analysis of demographic and metabolite data was per - analysis of this metabolite was not performed ( Snyder and formed with SPSS 24. Group differences were tested using Wilman, 2010). Finally, Pearson’s product-moment correlation independent-sample t tests, and significant effects are reported was used to examine potential associations between GABA lev - at P < .05. Associations between GABA/Cr levels and severity of els and age as well as cigarette use in UHR subjects, and Mann CAARMS positive and negative symptoms were assessed with Whitney-U test was used to examine potential group effects linear regression, and results were Bonferroni-corrected at between UHR with and without current or past cannabis use. P < .025. In line with previous studies (e.g., Kegeles et  al., 2012; de la Fuente-Sandoval et al., 2016), exploratory analyses on the Results other metabolites in the spectra (Glu/Cr, Glx/Cr, and NAA/Cr) were conducted for completion but will not be discussed. These Table 1 summarizes participant characteristics and metabolite v- al analyses explored: (1) group differences using t tests, (2) asso - ues. All UHR subjects met Attenuated Psychosis Syndrome criteria. ciations with CAARMS positive and negative symptoms using linear regression, and (3) correlations between GABA/Cr and the GABA Levels other metabolites using Pearson’s product-moment correlation (Bonferroni-corrected at P < .05/3). As field strengths of 4T or There were no significant differences in the creatine-scaled GABA levels between UHR subjects and healthy controls ( P = .130). more are needed to measure glutamine accurately, exploratory Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 117 Table 1. Participants’ Characteristics and Spectral Data HC (n = 20) UHR (n = 21) HC vs UHR Mean (SD, range) Mean (SD, range) Statistic P Age (y) 23.7 (2.7, 20–28) 22.2 (3.0, 18–29) t = 1.613 .111 Estimated IQ 117.1 (10.4, 95–132) 110.0 (12.5, 75–128) t = 1.736 .092 CAARMS positive - 11.9 (6.7, 1–34) - - CAARMS negative - 8.4 (6.4, 0–29) - - Tobacco (cigarettes/d) - 4.75 (6.7) - - Cannabis now (yes/no) - 11/12 - - Cannabis ever (yes/no) - 16/4 - - SNR 21.6 (2.8) 21.7 (2.80) -.076 .939 Line width 6.2 (1.8) 6.4 (1.28) -.284 .778 GABA/creatine .4 (.1) .3 (.05) 1.546 .130 GABA % CRLB 5.5 (1.0) 5.9 (1.4) -1.052 .300 Glutamate/creatine .5 (.1) 5.5 (.1) -.769 .446 Glutamate % CRLB 7.6 (1.5) 6.6 (1.5) 2.001 .053 Glx/creatine .8 (.1) .8 (.1) -.255 .800 Glx % CRLB 5.6 (1.4) 5.2 (1.5) .835 .409 N-acetyl-aspartate/creatine 1.1 (.1) 1.1 (1.1) .823 .415 N-acetyl-aspartate % CRLB 1.15 (.37) 1.63 (1.34) -1.512 .146 Abbreviations: CAARMS, Clinical Assessment for At-Risk Mental States; CRLB, Cramer Rao Lower Bounds; GABA, gamma-aminobutyric acid; Glx, glutamate + glutam- ine ratio; HC, healthy control subjects; SNR, signal to noise ratio; UHR, ultra-high risk subjects. Exploratory analysis of the other metabolites in the spectra also significantly differed between the groups (GABA and NAA: showed no significant differences between groups corrected for z = 1.7, P = .089; GABA and Glu: z = -.69, P = .490). multiple comparisons (all P > .017) (Table 1; Figure 1B). Within the UHR group, MPFC GABA/Cr levels were inversely Discussion associated with the severity of negative symptoms ( ß = -.556, t = -2.761, P = .013, significant after Bonferroni correction at We did not find evidence that cortical GABA levels (creatine- P < .025), but there was no relationship with positive symptoms scaled) in subjects at UHR of psychosis differed from those in (ß = -.245, t = -1.071, P = .298) (Figure 1C–D). healthy controls. The 3 previous GABA MRS studies in UHR sub - MPFC GABA/Cr levels were not significantly associated with jects reported either increased ( de la Fuente-Sandoval et  al., age (r = -.027, P = .908), cigarette use ( r = -.195, P = .410), or differed 2016), decreased (Menschikov et al., 2016), or no difference from in UHR subjects with current or past cannabis use compared healthy controls (Wang et  al., 2016). Although the location of with those without (current use: U = 38.0, Z = -1.723, P = .085; past the H-MRS voxel in those previous UHR studies was more ven- use: U = 22.0, Z = -.945, P = .345). Groups did not differ in spectral trally placed within the MPFC than in our study, another recent quality (CRLB, P = .484; SNR, P = .939; linewidth, P = .778) (Table 1). study in unaffected relatives using an overlapping voxel to ours Follow-up clinical data revealed that 3 of the 20 UHR subjects did find a significant decrease in the relatives ( Marenco et  al., developed psychosis at a mean follow-up time o 18  months. 2016). However, these were asymptomatic individuals at genetic Exploratory analyses removing these subjects rendered the cor - high risk as opposed to a sample of subjects with an attenu - relation between GABA/Cr and CAARMS negative symptoms no ated psychosis syndrome. Potential sources of variability may longer significant (ß = -.377, t = -1.578, P = .135), suggesting that thus relate to voxel placement and to the nature of the high- the association was driven by those individuals who went on to risk sample under study. Future studies should consider stand - develop a psychotic disorder. ardizing voxel placement or using multiple rather than single voxels (Duyn et  al., 1993). Allowing GABA quantification from both dorsal and ventral MPFC in the same individuals would Other Metabolites help elucidate region-specific effects in people at increased risk for psychosis and clarify whether GABA function is rela - Exploratory analysis showed no significant associations between levels of the other metabolites in the voxel and posi - tively uncompromised in more dorsal MPFC areas compared with healthy individuals. Regarding the nature of the high-risk tive or negative symptom severity (Glu/Cr and CAARMS positive: ß = -.159, t = -.684, P = .503; Glu/Cr and CAARMS negative: ß = -.297, samples recruited to different studies, the UHR category is het - erogeneous with respect to both inclusion criteria and clinical t = -1.283, P = .217; Glx/Cr and CAARMS positive: ß = -.109, t = -.463, P = .649; Glx/Cr and CAARMS negative: ß = -.096, t = -.400, P = .694; outcomes, and the neuroimaging findings in a sample may vary depending on its composition (Fusar-Poli et  al., 2016). In this NAA/Cr and CAARMS positive: ß = -.173, t = -.746, P = .465; NAA/ Cr and CAARMS negative: ß = -.034, t = -.142, P = .889). None of context, our study expands the previous literature by showing results from a homogeneous sample of UHR individuals who all these metabolites were significantly correlated with age (Glu/ Cr: r = -.296, P = .192; Glx/Cr: r = -.421, P = .058; NAA/Cr: r = .342, fell under attenuated psychosis syndrome criteria. Prospective studies in similarly homogenized UHR samples are needed to P = .130). Finally, Pearson’s product-moment correlation showed that in healthy controls, GABA/Cr was positively associated with further clarify whether GABAergic dysfunction can be reliably detected in this group, whether the MPFC subregions affected NAA/Cr (r = .516, P = .020), while in UHR individuals, GABA/Cr was positively associated with Glu/Cr ( r = .460, P = .041). However, are predominantly ventral, and whether alterations in GABA levels are identifiable with H-MRS in UHR individuals who these associations did not survive Bonferroni correction, or Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup 118 | International Journal of Neuropsychopharmacology, 2018 are destined to develop a psychotic disorder. Nevertheless, it prodromal negative symptoms, which warrants replication in is worth noting that a recent meta-analysis of H-MRS studies larger samples. in schizophrenia did not find a significant effect in GABA lev - els (Egerton et al., 2017 ), suggesting a lack of convergence with Acknowledgments predictions from animal and postmortem studies. This might The authors thank the MRI radiographers for their expert assist - relate to the divergent nature of the measurements across ance and the study volunteers for their participation, and we disciplines. While preclinical and postmortem studies suggest gratefully thank members of the OASIS, CAMEO, and Warwick that the GABAergic abnormality refers to parvalbumin-positive 1 and Coventry clinical teams who were involved in the recruit - interneurons (Marin, 2012), H-MRS assesses total tissue con - ment and management of the UHR subjects in this study. centrations and as such it is likely to not be restricted to a par - ticular GABA cell type. Future translational animal and human work measuring GABA levels in homolog regions across species Funding may be able to comprehensively delineate the molecular path- This work was supported by a Wellcome Trust Programme Grant way linking GABAergic dysfunction to the expression of schizo - to P.M. (grant no. 091667, 2011) and a Wellcome Trust Programme phrenia-like characteristics, including GABA measurements in Grant to D.G.M. (grant no. 091300/Z/10/Z). D.G.M. was also sup - other anatomical regions such as the hippocampus, amygdala, ported by the Dr Mortimer D. Sackler Foundation. G.M. is funded and thalamus. by the Wellcome Trust and the Royal Society (Sir Henry Dale Although preclinical models would primarily predict that Fellowship). cortical GABA dysfunction leads to positive psychotic symp - toms through a hyper-responsive dopamine system arising Interest Statement from a glutamatergic dysregulation, our regression analysis did not reveal an association between GABA and positive Dr Grace receives consulting fees from Johnson & Johnson, symptoms. However, we observed that GABA levels were sig - Lundbeck, Pfizer, GSK, Merck, Takeda, Dainippon Sumitomo, nificantly inversely associated with the severity of negative Otsuka, Lilly, Roche, Asubio, and Abbott; and receives research symptoms. This finding is of interest, as the pathophysio - funding from Lundbeck, Lilly, Autifony, Alkermes, and Johnson & logical basis of negative symptoms is unclear ( Salamone et al., Johnson. Dr Howes has received investigator-initiated research 2015), and it merits replication in a larger sample. Previous funding from and/or participated in advisory/speaker meetings studies had not found a significant association between MPFC organized by Astra-Zeneca, Autifony, BMS, Eli Lilly, Heptares, GABA levels and severity of positive or negative symptoms in Jansenn, Lundbeck, Lyden-Delta, Otsuka, Servier, Sunovion, UHR subjects (de la Fuente-Sandoval et al., 2016) or in unmedi- Rand, and Roche. Neither Dr Howes nor his family have been cated patients with a first episode of psychosis ( Kegeles et al., employed by or have holdings/a financial stake in any biomed - 2012). Nevertheless, as mentioned above, effects of anatom - ical company. There are no other interests from any of the other ical location of the MEGA-PRESS acquisition and study sample coauthors. composition may be at play. Furthermore, although this cross- sectional study was not designed to examine longitudinal References effects, the correlation appeared to be driven by UHR individu - als who went on to develop a psychotic disorder. Preclinical Akbarian S, Huang HS (2006) Molecular and cellular mecha- models have proposed circuit-based approaches to the devel- nisms of altered GAD1/GAD67 expression in schizophrenia opment of psychosis-like behaviors (Lisman et al., 2008), and and related disorders. Brain Res Rev 52:293–304. it has been recently postulated that the positive, negative, American Psychiatric Association (1994) Diagnostic and and disorganized dimensions of schizophrenia may originate Statistical manual of mental disorders., 4th ed (DSM-IV) edi - from disruption of multiple, interconnected circuits involv - tion. Washington DC: American Psychiatric Association. ing GABAergic dysfunction and converging on hippocampal Benes FM (2010) Amygdalocortical circuitry in schizophrenia: from hyperactivity (Grace, 2016). Hence, larger multimodal imaging circuits to molecules. Neuropsychopharmacology 35:239–257. studies in UHR subjects investigating associations between, de la Fuente-Sandoval C, Reyes-Madrigal F, Mao X, Leon-Ortiz for example, GABA levels and hippocampal activity in relation P, Rodriguez-Mayoral O, Solis-Vivanco R, Favila R, Graff- to clinical symptomatology and outcomes are warranted to Guerrero A, Shungu DC (2016) Cortico-striatal GABAergic and expand the present findings. glutamatergic dysregulations in subjects at ultra-high risk Strengths of the current report include a homogenous UHR for psychosis investigated with proton magnetic resonance sample as determined by trained clinicians, a well-matched con - spectroscopy. Int J Neuropsychopharmacol 19: pyv105. trol group, and a validated method (MEGA-PRESS) to quantify Du Y, Grace AA (2013) Peripubertal diazepam administration pre - water-scaled GABA concentrations at 3T. A limitation intrinsic to vents the emergence of dopamine system hyperresponsivity all MEGA-PRESS studies is that the GABA signal contains some in the MAM developmental disruption model of schizophre- contribution from macromolecules, that is, diverse proteins and nia. Neuropsychopharmacology 38:1881–1888. lipids, but it is unlikely that the macromolecule contribution Du Y, Grace AA (2016) Loss of parvalbumin in the hippocampus would differ across groups. Furthermore, the results should be of MAM schizophrenia model rats is attenuated by peripuber - considered in the context of a sample of male participants and tal diazepam. Int J Neuropsychopharmacol. 19:pii: pyw065. as such may not generalize to all individuals with an attenuated doi: 10.1093/ijnp/pyw065. psychosis syndrome. Duyn JH, Gillen J, Sobering G, van Zijl PC, Moonen CT (1993) In conclusion, if there is any alteration in GABA levels meas- Multisection proton MR spectroscopic imaging of the brain. urable with in vivo H-MRS in the MPFC in people at UHR of Radiology 188:277–282. psychosis, it is small and difficult to detect even with a homoge- Egerton A, Modinos G, Ferrera D, McGuire P (2017) Neuroimaging neous sample of antipsychotic-naïve individuals. Nevertheless, studies of GABA in schizophrenia: a systematic review with our data provides a direct link between GABAergic levels and meta-analysis. Transl Psychiatry 7:e1147. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 119 Fusar-Poli P, et al. (2016) Heterogeneity of psychosis risk within Modinos G, Allen P, Grace AA, McGuire P (2015) Translating the MAM individuals at clinical high risk: a meta-analytical stratifica - model of psychosis to humans. Trends Neurosci 38:129–138. tion. JAMA Psychiatry 73:113–120. Mouchlianitis E, Bloomfield MA, Law V, Beck K, Selvaraj S, Grace AA (2010) Ventral hippocampus, interneurons, and schizo- Rasquinha N, Waldman A, Turkheimer FE, Egerton A, phrenia: a new understanding of the pathophysiology of Stone J, Howes OD (2016) Treatment-resistant schizophr -e schizophrenia and its implications for treatment and preven - nia patients show elevated anterior cingulate cortex glu - tion. Curr Directions Psychol Sci 19:232–237. tamate compared to treatment-responsive. Schizophr Bull Grace AA (2016) Dysregulation of the dopamine system in the 42:744–752. pathophysiology of schizophrenia and depression. Nat Rev Provencher SW (2016) LCModel & LCMgui User’s Manual. Neurosci 17:524–532. LCModel version 6.3-1L, <http://s-provencher.com/pub/ Kegeles LS, Mao X, Stanford AD, Girgis R, Ojeil N, Xu X, Gil R, LCModel/manual/manual.pdf> Slifstein M, Abi-Dargham A, Lisanby SH, Shungu DC (2012) Salamone JD, Koychev I, Correa M, McGuire P (2015) Elevated prefrontal cortex gamma-aminobutyric acid and Neurobiological basis of motivational deficits in psychopath- glutamate-glutamine levels in schizophrenia measured in ology. Eur Neuropsychopharmacol 25:1225–1238. vivo with proton magnetic resonance spectroscopy. Arch Gen Snyder J, Wilman A (2010) Field strength dependence of PRESS Psychiatry 69:449–459. timings for simultaneous detection of glutamate and glutam - Lewis DA, Hashimoto T, Volk DW (2005) Cortical inhibitory neu - ine from 1.5 to 7T. J Magn Reson 203:66–72. rons and schizophrenia. Nat Rev Neurosci 6:312–324. Velthorst E, Levine SZ, Henquet C, de Haan L, van Os J, Myin- Marenco S, Meyer C, Kuo S, van der Veen JW, Shen J, DeJong K, Germeys I, Reichenberg A (2013) To cut a short test even Barnett AS, Apud JA, Dickinson D, Weinberger DR, Berman KF shorter: reliability and validity of a brief assessment of intel - (2016) Prefrontal GABA levels measured with magnetic reson - lectual ability in schizophrenia--a control-case family study. ance spectroscopy in patients with psychosis and unaffected Cog Neuropsychiatry 18:574–593. siblings. Am J Psychiatry 173:527–534. Wang J, Tang Y, Zhang T, Cui H, Xu L, Zeng B, Li Y, Li G, Li C, Liu Marin O (2012) Interneuron dysfunction in psychiatric disorders. H, Lu Z, Zhang J, Wang J (2016) Reduced γ-aminobutyric acid Nat Rev Neurosci 13:107–120. and glutamate+glutamine levels in drug-naïve patients with Menschikov PE, Semenova NA, Ublinskiy MV, Akhadov TA, first-episode schizophrenia but not in those at ultrahigh risk. Keshishyan RA, Lebedeva IS, Omelchenko MA, Kaleda VG, Neural Plasticity 2016:9. Varfolomeev SD (2016) (1)H-MRS and MEGA-PRESS pulse Yung AR, Yuen HP, McGorry PD, Phillips LJ, Kelly D, Dell’Olio M, sequence in the study of balance of inhibitory and excita - Francey SM, Cosgrave EM, Killackey E, Stanford C, Godfrey tory neurotransmitters in the human brain of ultra-high K, Buckby J (2005) Mapping the onset of psychosis: the risk of schizophrenia patients. Dokl Biochem Biophys Comprehensive Assessment of At-Risk Mental States. Aust N 468:168–172. Z J Psychiatry 39:964–971. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png International Journal of Neuropsychopharmacology Oxford University Press

Loading next page...
 
/lp/ou_press/cortical-gaba-in-subjects-at-ultra-high-risk-of-psychosis-relationship-39yh6ImvU4
Publisher
Collegium Internationale Neuro-Psychopharmacologicum
Copyright
© The Author(s) 2017. Published by Oxford University Press on behalf of CINP.
ISSN
1461-1457
eISSN
1469-5111
D.O.I.
10.1093/ijnp/pyx076
Publisher site
See Article on Publisher Site

Abstract

Background: Whilst robust preclinical and postmortem evidence suggests that altered GABAergic function is central to the development of psychosis, little is known about whether it is altered in subjects at ultra-high risk of psychosis, or its relationship to prodromal symptoms. Methods: Twenty-one antipsychotic naïve ultra-high risk individuals and 20 healthy volunteers underwent proton magnetic resonance imaging at 3T. Gamma-aminobutyric acid levels were obtained from the medial prefrontal cortex using MEGA- PRESS and expressed as peak-area ratios relative to the synchronously acquired creatine signal. Gamma-aminobutyric acid levels were then related to severity of positive and negative symptoms as measured with the Community Assessment of At-Risk Mental States. Results: Whilst we found no significant difference in gamma-aminobutyric acid levels between ultra-high risk subjects and healthy controls P ( = .130), in ultra-high risk individuals, medial prefrontal cortex GABA levels were negatively correlated with the severity of negative symptoms ( P = .013). Received: May 24, 2017; Revised: August 5, 2017; Accepted: August 18, 2017 © The Author(s) 2017. Published by Oxford University Press on behalf of CINP. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons. org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is 114 properly cited. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 115 Conclusion: These findings suggest that gamma-aminobutyric acidergic neurotransmission may be involved in the neurobiology of negative symptoms in the ultra-high risk state. Keywords: psychosis, magnetic resonance spectroscopy, GABA, ultra-high risk of psychosis, negative symptoms Introduction Methods Converging evidence from postmortem and preclinical studies Procedures were approved by the Research Ethics Committee indicates that dysfunction of the gamma-aminobutyric acider - of King’s College London and South London and Maudsley NHS gic (GABAergic) neurotransmitter system plays a major role in Trust. All participants provided informed consent. the pathophysiology of schizophrenia (Marin, 2012). Postmortem Twenty-one UHR individuals and 20 healthy volunteers, all research has demonstrated decreased mRNA expression of glu - males aged 18to 30 years, were included. UHR psychopathology tamic acid decarboxylase and reduced density of fast-spiking was assessed using the Community Assessment of At-Risk parvalbumin-positive interneurons in a corticolimbic circuitry Mental States (CAARMS) ( Yung et  al., 2005). UHR inclusion cri- involving the prefrontal cortex and the amygdala in schizo - teria required the presence of one or more of the following: (1) phrenia (Lewis et  al., 2005; Akbarian and Huang, 2006; Benes, attenuated psychotic syndrome, (2) a brief psychotic episode 2010). Furthermore, animal models of psychosis suggest a link of <1 week duration that spontaneously remits without anti - between disrupted cortical GABAergic function and dysregula - psychotic medication/hospitalization, and (3) trait vulnerability tion of subcortical dopaminergic signaling characteristic of the (schizotypal personality disorder or a first-degree relative with disorder (Grace, 2010). Such models propose that inhibitory dis - psychosis) plus a marked decline in psychosocial functioning. ruption would underlie not only dopamine-dependent positive Healthy control subjects were recruited from the local com - symptoms of psychosis, but would also influence other neural munity. They were excluded if they had a personal or familial pathways (e.g., including the basolateral nucleus of the amyg - history of psychiatric disorder, neurological illness, or drug/ dala, or the medial prefrontal cortex [MPFC]) putatively involved alcohol dependence based on the DSM-V ( American Psychiatric in the development of the negative symptoms of psychosis Association, 1994). Current/past medication use and current/ (Grace, 2016). Moreover, the role of GABA in the development past use of tobacco and cannabis was assessed using a semis- of psychosis is further supported by preclinical evidence that tructured interview adapted from the Early Psychosis Prevention peripubertal (i.e., premorbid) pharmacological intervention on and Intervention Centre Drug and Alcohol Assessment Schedule GABA-Aα5 receptors prevents schizophrenia-like GABA cell loss (http://www.eppic.org.au). All subjects were safe for MRI, had an and blocks the development of psychosis-like features in adult IQ in the normal range as assessed using the Wechsler Adult rats (Du and Grace, 2013, 2016). Intelligence Scale-III (Velthorst et  al., 2013 ), and were antipsy- From animal and human postmortem studies, it may thus chotic-naïve, and none were taking benzodiazepines. be hypothesized that cortical GABAergic function is reduced in Subjects underwent H-MRS on a General Electric Signa schizophrenia and that this abnormality can be detected in the HDx TwinSpeed 3T scanner at the Centre for Neuroimaging premorbid stages of the disorder ( Modinos et al., 2015). However, Sciences, Institute of Psychiatry, Psychology & Neuroscience recent meta-analytical evidence from human imaging studies (King’s College London). GABA levels were obtained from the using proton magnetic resonance spectroscopy (H-MRS) did not MPFC using MEGA-PRESS, which incorporates a standardized show a significant difference in regional GABA levels between chemically selective suppression water suppression routine patients with schizophrenia and healthy volunteers ( Egerton (TE = 68 milliseconds, TR = 2000 milliseconds). For each acquisi- et al., 2017). Research in patients with schizophrenia is compli- tion, unsuppressed water reference spectra (16 averages) were cated by previous antipsychotic exposure and heterogeneity of also acquired. Shimming was optimized, with auto-prescan clinical subgroups (Kegeles et  al., 2012). In this context, stud - performed twice before each scan. The region of interest in the ies in subjects at ultra-high risk (UHR) of developing psychosis MPFC was prescribed from the midline sagittal localizer, and are a useful resource to investigate neurobiological correlates the center of the 40- × 35- × 20-mm region of interest was placed of psychosis-like characteristics without confounds associ - above the middle section of corpus callosum ( Figure 1A). Spectra ated with the use of antipsychotics or illness chronicity on the were analyzed using LCModel 6.3-1L with the basis set provided imaging data. The 3 available MPFC GABA studies in UHR indi- by its author Pr ( ovencher, 2016), which contained the metabo - viduals have also presented mixed results, including increases lites GABA, glutamine, glutamate, Glx (glutamate + glutamine), (de la Fuente-Sandoval et al., 2016), decreases (Menschikov et al., and N-acetyl-aspartate (NAA). We used Cramer-Rao minimum 2016), and no differences (Wang et  al., 2016) when compared variance bounds (CRLB) >20% as reported by LCModel, which with healthy controls. Nevertheless, heterogeneity of clinical are estimates of fit of the metabolite peaks, and signal-to-noise subgroups is also a potential confounder in UHR studies ( Fusar- ratio <8 to exclude poorly fitted metabolite peaks from statis - Poli et al., 2016), and only 1 study to date investigated associa - tical analysis (Mouchlianitis et al., 2016 ; Provencher, 2016). Data tions between GABA levels and severity of positive and negative from all 41 participants in the present study met these criteria. symptoms in this group (de la Fuente-Sandoval et al., 2016 ). The Metabolites were expressed as ratios relative to the synchron - present study sought to address these issues by using H-MRS ously acquired creatine signal from the unedited MEGA-PRESS in a homogenous sample of antipsychotic-naïve subjects at spectra. This is a well-established normalization procedure in UHR of psychosis to test the hypotheses that: (1) GABA levels clinical H-MRS studies that has been extensively used in pre - in the MPFC would be reduced in UHR subjects compared with vious studies of MPFC GABA levels in patients with schizophre - healthy controls ( Marin, 2012), and that (2) GABA levels would nia (Goto et  al., 2009; Ongur et  al., 2010; Kegeles et  al., 2012; be inversely related to the severity of positive and negative pr -o Marsman et  al., 2014)  and UHR subjects (Marenco et  al., 2016; dromal symptoms (Grace, 2016). Menschikov et al., 2016). Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup 116 | International Journal of Neuropsychopharmacology, 2018 Figure 1. (A) Voxel placement on medial prefrontal cortex (MPFC) and representative sample H-MRS spectra. (B) MPFC metabolite levels by group. (C) Scatterplot of the significant association between CAARMS negative symptom severity and GABA/Cr levels (ß = -.556, t = -2.761, R  = .310, P = .013). (D) Scatterplot of the nonsignificant associations between CAARMS positive symptom severity and GABA/Cr levels ( P  =  .298). Cr, creatine; Glu, glutamate; Glx, glutamate + glutamine; NAA, N-Acetyl- aspartate; UHR, ultra-high risk of psychosis. Analysis of demographic and metabolite data was per - analysis of this metabolite was not performed ( Snyder and formed with SPSS 24. Group differences were tested using Wilman, 2010). Finally, Pearson’s product-moment correlation independent-sample t tests, and significant effects are reported was used to examine potential associations between GABA lev - at P < .05. Associations between GABA/Cr levels and severity of els and age as well as cigarette use in UHR subjects, and Mann CAARMS positive and negative symptoms were assessed with Whitney-U test was used to examine potential group effects linear regression, and results were Bonferroni-corrected at between UHR with and without current or past cannabis use. P < .025. In line with previous studies (e.g., Kegeles et  al., 2012; de la Fuente-Sandoval et al., 2016), exploratory analyses on the Results other metabolites in the spectra (Glu/Cr, Glx/Cr, and NAA/Cr) were conducted for completion but will not be discussed. These Table 1 summarizes participant characteristics and metabolite v- al analyses explored: (1) group differences using t tests, (2) asso - ues. All UHR subjects met Attenuated Psychosis Syndrome criteria. ciations with CAARMS positive and negative symptoms using linear regression, and (3) correlations between GABA/Cr and the GABA Levels other metabolites using Pearson’s product-moment correlation (Bonferroni-corrected at P < .05/3). As field strengths of 4T or There were no significant differences in the creatine-scaled GABA levels between UHR subjects and healthy controls ( P = .130). more are needed to measure glutamine accurately, exploratory Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 117 Table 1. Participants’ Characteristics and Spectral Data HC (n = 20) UHR (n = 21) HC vs UHR Mean (SD, range) Mean (SD, range) Statistic P Age (y) 23.7 (2.7, 20–28) 22.2 (3.0, 18–29) t = 1.613 .111 Estimated IQ 117.1 (10.4, 95–132) 110.0 (12.5, 75–128) t = 1.736 .092 CAARMS positive - 11.9 (6.7, 1–34) - - CAARMS negative - 8.4 (6.4, 0–29) - - Tobacco (cigarettes/d) - 4.75 (6.7) - - Cannabis now (yes/no) - 11/12 - - Cannabis ever (yes/no) - 16/4 - - SNR 21.6 (2.8) 21.7 (2.80) -.076 .939 Line width 6.2 (1.8) 6.4 (1.28) -.284 .778 GABA/creatine .4 (.1) .3 (.05) 1.546 .130 GABA % CRLB 5.5 (1.0) 5.9 (1.4) -1.052 .300 Glutamate/creatine .5 (.1) 5.5 (.1) -.769 .446 Glutamate % CRLB 7.6 (1.5) 6.6 (1.5) 2.001 .053 Glx/creatine .8 (.1) .8 (.1) -.255 .800 Glx % CRLB 5.6 (1.4) 5.2 (1.5) .835 .409 N-acetyl-aspartate/creatine 1.1 (.1) 1.1 (1.1) .823 .415 N-acetyl-aspartate % CRLB 1.15 (.37) 1.63 (1.34) -1.512 .146 Abbreviations: CAARMS, Clinical Assessment for At-Risk Mental States; CRLB, Cramer Rao Lower Bounds; GABA, gamma-aminobutyric acid; Glx, glutamate + glutam- ine ratio; HC, healthy control subjects; SNR, signal to noise ratio; UHR, ultra-high risk subjects. Exploratory analysis of the other metabolites in the spectra also significantly differed between the groups (GABA and NAA: showed no significant differences between groups corrected for z = 1.7, P = .089; GABA and Glu: z = -.69, P = .490). multiple comparisons (all P > .017) (Table 1; Figure 1B). Within the UHR group, MPFC GABA/Cr levels were inversely Discussion associated with the severity of negative symptoms ( ß = -.556, t = -2.761, P = .013, significant after Bonferroni correction at We did not find evidence that cortical GABA levels (creatine- P < .025), but there was no relationship with positive symptoms scaled) in subjects at UHR of psychosis differed from those in (ß = -.245, t = -1.071, P = .298) (Figure 1C–D). healthy controls. The 3 previous GABA MRS studies in UHR sub - MPFC GABA/Cr levels were not significantly associated with jects reported either increased ( de la Fuente-Sandoval et  al., age (r = -.027, P = .908), cigarette use ( r = -.195, P = .410), or differed 2016), decreased (Menschikov et al., 2016), or no difference from in UHR subjects with current or past cannabis use compared healthy controls (Wang et  al., 2016). Although the location of with those without (current use: U = 38.0, Z = -1.723, P = .085; past the H-MRS voxel in those previous UHR studies was more ven- use: U = 22.0, Z = -.945, P = .345). Groups did not differ in spectral trally placed within the MPFC than in our study, another recent quality (CRLB, P = .484; SNR, P = .939; linewidth, P = .778) (Table 1). study in unaffected relatives using an overlapping voxel to ours Follow-up clinical data revealed that 3 of the 20 UHR subjects did find a significant decrease in the relatives ( Marenco et  al., developed psychosis at a mean follow-up time o 18  months. 2016). However, these were asymptomatic individuals at genetic Exploratory analyses removing these subjects rendered the cor - high risk as opposed to a sample of subjects with an attenu - relation between GABA/Cr and CAARMS negative symptoms no ated psychosis syndrome. Potential sources of variability may longer significant (ß = -.377, t = -1.578, P = .135), suggesting that thus relate to voxel placement and to the nature of the high- the association was driven by those individuals who went on to risk sample under study. Future studies should consider stand - develop a psychotic disorder. ardizing voxel placement or using multiple rather than single voxels (Duyn et  al., 1993). Allowing GABA quantification from both dorsal and ventral MPFC in the same individuals would Other Metabolites help elucidate region-specific effects in people at increased risk for psychosis and clarify whether GABA function is rela - Exploratory analysis showed no significant associations between levels of the other metabolites in the voxel and posi - tively uncompromised in more dorsal MPFC areas compared with healthy individuals. Regarding the nature of the high-risk tive or negative symptom severity (Glu/Cr and CAARMS positive: ß = -.159, t = -.684, P = .503; Glu/Cr and CAARMS negative: ß = -.297, samples recruited to different studies, the UHR category is het - erogeneous with respect to both inclusion criteria and clinical t = -1.283, P = .217; Glx/Cr and CAARMS positive: ß = -.109, t = -.463, P = .649; Glx/Cr and CAARMS negative: ß = -.096, t = -.400, P = .694; outcomes, and the neuroimaging findings in a sample may vary depending on its composition (Fusar-Poli et  al., 2016). In this NAA/Cr and CAARMS positive: ß = -.173, t = -.746, P = .465; NAA/ Cr and CAARMS negative: ß = -.034, t = -.142, P = .889). None of context, our study expands the previous literature by showing results from a homogeneous sample of UHR individuals who all these metabolites were significantly correlated with age (Glu/ Cr: r = -.296, P = .192; Glx/Cr: r = -.421, P = .058; NAA/Cr: r = .342, fell under attenuated psychosis syndrome criteria. Prospective studies in similarly homogenized UHR samples are needed to P = .130). Finally, Pearson’s product-moment correlation showed that in healthy controls, GABA/Cr was positively associated with further clarify whether GABAergic dysfunction can be reliably detected in this group, whether the MPFC subregions affected NAA/Cr (r = .516, P = .020), while in UHR individuals, GABA/Cr was positively associated with Glu/Cr ( r = .460, P = .041). However, are predominantly ventral, and whether alterations in GABA levels are identifiable with H-MRS in UHR individuals who these associations did not survive Bonferroni correction, or Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup 118 | International Journal of Neuropsychopharmacology, 2018 are destined to develop a psychotic disorder. Nevertheless, it prodromal negative symptoms, which warrants replication in is worth noting that a recent meta-analysis of H-MRS studies larger samples. in schizophrenia did not find a significant effect in GABA lev - els (Egerton et al., 2017 ), suggesting a lack of convergence with Acknowledgments predictions from animal and postmortem studies. This might The authors thank the MRI radiographers for their expert assist - relate to the divergent nature of the measurements across ance and the study volunteers for their participation, and we disciplines. While preclinical and postmortem studies suggest gratefully thank members of the OASIS, CAMEO, and Warwick that the GABAergic abnormality refers to parvalbumin-positive 1 and Coventry clinical teams who were involved in the recruit - interneurons (Marin, 2012), H-MRS assesses total tissue con - ment and management of the UHR subjects in this study. centrations and as such it is likely to not be restricted to a par - ticular GABA cell type. Future translational animal and human work measuring GABA levels in homolog regions across species Funding may be able to comprehensively delineate the molecular path- This work was supported by a Wellcome Trust Programme Grant way linking GABAergic dysfunction to the expression of schizo - to P.M. (grant no. 091667, 2011) and a Wellcome Trust Programme phrenia-like characteristics, including GABA measurements in Grant to D.G.M. (grant no. 091300/Z/10/Z). D.G.M. was also sup - other anatomical regions such as the hippocampus, amygdala, ported by the Dr Mortimer D. Sackler Foundation. G.M. is funded and thalamus. by the Wellcome Trust and the Royal Society (Sir Henry Dale Although preclinical models would primarily predict that Fellowship). cortical GABA dysfunction leads to positive psychotic symp - toms through a hyper-responsive dopamine system arising Interest Statement from a glutamatergic dysregulation, our regression analysis did not reveal an association between GABA and positive Dr Grace receives consulting fees from Johnson & Johnson, symptoms. However, we observed that GABA levels were sig - Lundbeck, Pfizer, GSK, Merck, Takeda, Dainippon Sumitomo, nificantly inversely associated with the severity of negative Otsuka, Lilly, Roche, Asubio, and Abbott; and receives research symptoms. This finding is of interest, as the pathophysio - funding from Lundbeck, Lilly, Autifony, Alkermes, and Johnson & logical basis of negative symptoms is unclear ( Salamone et al., Johnson. Dr Howes has received investigator-initiated research 2015), and it merits replication in a larger sample. Previous funding from and/or participated in advisory/speaker meetings studies had not found a significant association between MPFC organized by Astra-Zeneca, Autifony, BMS, Eli Lilly, Heptares, GABA levels and severity of positive or negative symptoms in Jansenn, Lundbeck, Lyden-Delta, Otsuka, Servier, Sunovion, UHR subjects (de la Fuente-Sandoval et al., 2016) or in unmedi- Rand, and Roche. Neither Dr Howes nor his family have been cated patients with a first episode of psychosis ( Kegeles et al., employed by or have holdings/a financial stake in any biomed - 2012). Nevertheless, as mentioned above, effects of anatom - ical company. There are no other interests from any of the other ical location of the MEGA-PRESS acquisition and study sample coauthors. composition may be at play. Furthermore, although this cross- sectional study was not designed to examine longitudinal References effects, the correlation appeared to be driven by UHR individu - als who went on to develop a psychotic disorder. Preclinical Akbarian S, Huang HS (2006) Molecular and cellular mecha- models have proposed circuit-based approaches to the devel- nisms of altered GAD1/GAD67 expression in schizophrenia opment of psychosis-like behaviors (Lisman et al., 2008), and and related disorders. Brain Res Rev 52:293–304. it has been recently postulated that the positive, negative, American Psychiatric Association (1994) Diagnostic and and disorganized dimensions of schizophrenia may originate Statistical manual of mental disorders., 4th ed (DSM-IV) edi - from disruption of multiple, interconnected circuits involv - tion. Washington DC: American Psychiatric Association. ing GABAergic dysfunction and converging on hippocampal Benes FM (2010) Amygdalocortical circuitry in schizophrenia: from hyperactivity (Grace, 2016). Hence, larger multimodal imaging circuits to molecules. Neuropsychopharmacology 35:239–257. studies in UHR subjects investigating associations between, de la Fuente-Sandoval C, Reyes-Madrigal F, Mao X, Leon-Ortiz for example, GABA levels and hippocampal activity in relation P, Rodriguez-Mayoral O, Solis-Vivanco R, Favila R, Graff- to clinical symptomatology and outcomes are warranted to Guerrero A, Shungu DC (2016) Cortico-striatal GABAergic and expand the present findings. glutamatergic dysregulations in subjects at ultra-high risk Strengths of the current report include a homogenous UHR for psychosis investigated with proton magnetic resonance sample as determined by trained clinicians, a well-matched con - spectroscopy. Int J Neuropsychopharmacol 19: pyv105. trol group, and a validated method (MEGA-PRESS) to quantify Du Y, Grace AA (2013) Peripubertal diazepam administration pre - water-scaled GABA concentrations at 3T. A limitation intrinsic to vents the emergence of dopamine system hyperresponsivity all MEGA-PRESS studies is that the GABA signal contains some in the MAM developmental disruption model of schizophre- contribution from macromolecules, that is, diverse proteins and nia. Neuropsychopharmacology 38:1881–1888. lipids, but it is unlikely that the macromolecule contribution Du Y, Grace AA (2016) Loss of parvalbumin in the hippocampus would differ across groups. Furthermore, the results should be of MAM schizophrenia model rats is attenuated by peripuber - considered in the context of a sample of male participants and tal diazepam. Int J Neuropsychopharmacol. 19:pii: pyw065. as such may not generalize to all individuals with an attenuated doi: 10.1093/ijnp/pyw065. psychosis syndrome. Duyn JH, Gillen J, Sobering G, van Zijl PC, Moonen CT (1993) In conclusion, if there is any alteration in GABA levels meas- Multisection proton MR spectroscopic imaging of the brain. urable with in vivo H-MRS in the MPFC in people at UHR of Radiology 188:277–282. psychosis, it is small and difficult to detect even with a homoge- Egerton A, Modinos G, Ferrera D, McGuire P (2017) Neuroimaging neous sample of antipsychotic-naïve individuals. Nevertheless, studies of GABA in schizophrenia: a systematic review with our data provides a direct link between GABAergic levels and meta-analysis. Transl Psychiatry 7:e1147. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018 Copyedited by: oup Modinos et al. | 119 Fusar-Poli P, et al. (2016) Heterogeneity of psychosis risk within Modinos G, Allen P, Grace AA, McGuire P (2015) Translating the MAM individuals at clinical high risk: a meta-analytical stratifica - model of psychosis to humans. Trends Neurosci 38:129–138. tion. JAMA Psychiatry 73:113–120. Mouchlianitis E, Bloomfield MA, Law V, Beck K, Selvaraj S, Grace AA (2010) Ventral hippocampus, interneurons, and schizo- Rasquinha N, Waldman A, Turkheimer FE, Egerton A, phrenia: a new understanding of the pathophysiology of Stone J, Howes OD (2016) Treatment-resistant schizophr -e schizophrenia and its implications for treatment and preven - nia patients show elevated anterior cingulate cortex glu - tion. Curr Directions Psychol Sci 19:232–237. tamate compared to treatment-responsive. Schizophr Bull Grace AA (2016) Dysregulation of the dopamine system in the 42:744–752. pathophysiology of schizophrenia and depression. Nat Rev Provencher SW (2016) LCModel & LCMgui User’s Manual. Neurosci 17:524–532. LCModel version 6.3-1L, <http://s-provencher.com/pub/ Kegeles LS, Mao X, Stanford AD, Girgis R, Ojeil N, Xu X, Gil R, LCModel/manual/manual.pdf> Slifstein M, Abi-Dargham A, Lisanby SH, Shungu DC (2012) Salamone JD, Koychev I, Correa M, McGuire P (2015) Elevated prefrontal cortex gamma-aminobutyric acid and Neurobiological basis of motivational deficits in psychopath- glutamate-glutamine levels in schizophrenia measured in ology. Eur Neuropsychopharmacol 25:1225–1238. vivo with proton magnetic resonance spectroscopy. Arch Gen Snyder J, Wilman A (2010) Field strength dependence of PRESS Psychiatry 69:449–459. timings for simultaneous detection of glutamate and glutam - Lewis DA, Hashimoto T, Volk DW (2005) Cortical inhibitory neu - ine from 1.5 to 7T. J Magn Reson 203:66–72. rons and schizophrenia. Nat Rev Neurosci 6:312–324. Velthorst E, Levine SZ, Henquet C, de Haan L, van Os J, Myin- Marenco S, Meyer C, Kuo S, van der Veen JW, Shen J, DeJong K, Germeys I, Reichenberg A (2013) To cut a short test even Barnett AS, Apud JA, Dickinson D, Weinberger DR, Berman KF shorter: reliability and validity of a brief assessment of intel - (2016) Prefrontal GABA levels measured with magnetic reson - lectual ability in schizophrenia--a control-case family study. ance spectroscopy in patients with psychosis and unaffected Cog Neuropsychiatry 18:574–593. siblings. Am J Psychiatry 173:527–534. Wang J, Tang Y, Zhang T, Cui H, Xu L, Zeng B, Li Y, Li G, Li C, Liu Marin O (2012) Interneuron dysfunction in psychiatric disorders. H, Lu Z, Zhang J, Wang J (2016) Reduced γ-aminobutyric acid Nat Rev Neurosci 13:107–120. and glutamate+glutamine levels in drug-naïve patients with Menschikov PE, Semenova NA, Ublinskiy MV, Akhadov TA, first-episode schizophrenia but not in those at ultrahigh risk. Keshishyan RA, Lebedeva IS, Omelchenko MA, Kaleda VG, Neural Plasticity 2016:9. Varfolomeev SD (2016) (1)H-MRS and MEGA-PRESS pulse Yung AR, Yuen HP, McGorry PD, Phillips LJ, Kelly D, Dell’Olio M, sequence in the study of balance of inhibitory and excita - Francey SM, Cosgrave EM, Killackey E, Stanford C, Godfrey tory neurotransmitters in the human brain of ultra-high K, Buckby J (2005) Mapping the onset of psychosis: the risk of schizophrenia patients. Dokl Biochem Biophys Comprehensive Assessment of At-Risk Mental States. Aust N 468:168–172. Z J Psychiatry 39:964–971. Downloaded from https://academic.oup.com/ijnp/article-abstract/21/2/114/4085585 by Ed 'DeepDyve' Gillespie user on 16 March 2018

Journal

International Journal of NeuropsychopharmacologyOxford University Press

Published: Feb 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 12 million articles from more than
10,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Unlimited reading

Read as many articles as you need. Full articles with original layout, charts and figures. Read online, from anywhere.

Stay up to date

Keep up with your field with Personalized Recommendations and Follow Journals to get automatic updates.

Organize your research

It’s easy to organize your research with our built-in tools.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

Monthly Plan

  • Read unlimited articles
  • Personalized recommendations
  • No expiration
  • Print 20 pages per month
  • 20% off on PDF purchases
  • Organize your research
  • Get updates on your journals and topic searches

$49/month

Start Free Trial

14-day Free Trial

Best Deal — 39% off

Annual Plan

  • All the features of the Professional Plan, but for 39% off!
  • Billed annually
  • No expiration
  • For the normal price of 10 articles elsewhere, you get one full year of unlimited access to articles.

$588

$360/year

billed annually
Start Free Trial

14-day Free Trial