Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Abstract

Continued human and animal research has strengthened evidence for aberrant excitatory–inhibitory neural processes underlying autism and schizophrenia spectrum disorder psychopathology, particularly psychosocial functioning, in clinical and nonclinical populations. We investigated the extent to which autistic traits and schizotypal dimensions were modulated by the interactive relationship between excitatory glutamate and inhibitory GABA neurotransmitter concentrations in the social processing area of the superior temporal cortex using proton magnetic resonance spectroscopy. In total, 38 non-clinical participants (20 females; age range = 18–35 years, mean (standard deviation) = 23.22 (5.52)) completed the autism spectrum quotient and schizotypal personality questionnaire, and underwent proton magnetic resonance spectroscopy to quantify glutamate and GABA concentrations in the right and left superior temporal cortex. Regression analyses revealed that glutamate and GABA interactively modulated autistic social skills and schizotypal interpersonal features (p_corr < 0.05), such that those with high right superior temporal cortex glutamate but low GABA concentrations exhibited poorer social and interpersonal skills. These findings evidence an excitation–inhibition imbalance that is specific to psychosocial features across the autism and schizophrenia spectra.

Keywords

autism GABA glutamate magnetic resonance spectroscopy schizophrenia social behavior social cognition

The hyper-glutamatergic and hypo-GABAergic hypotheses of autism have gained significant attention in recent years (Cellot & Cherubini, 2014; Deidda, Bozarth, & Cancedda, 2014; Fatemi, 2008; Fatemi, Reutiman, Folsom, & Thuras, 2009; Rojas, 2014a). Normal glutamatergic and GABAergic neurotransmission is essential for the modulation of respective cortical excitatory and inhibitory processes, with abnormalities in one or both neurotransmitters leading to an excitation/inhibition (EI) imbalance. Such an imbalance affects neural signaling and transmission, and the subsequent behavioral and cognitive functions that they support (for a recent review, see Foss-Feig et al., 2017).

Aberrant cortical EI imbalance in autism spectrum disorder (ASD) has been associated with several underlying mechanisms, from reduced GABA_A (Fatemi et al., 2009), NMDA, and AMPA receptor densities (Purcell, Jeon, Zimmerman, Blue, & Pevsner, 2001) to abnormalities in the synthesis of glutamate to glutamine (glutaminase; Shimmura et al., 2013), and glutamate to GABA (GAD65/67; Fatemi et al., 2002). Mouse models of autism provide evidence for the EI imbalance hypothesis of ASD, with optogenetically controlled increase in the ratio of excitation/inhibition inducing social interaction deficits (Yizhar et al., 2011), and pharmacologically increased inhibitory neurotransmission reducing such social deficits in mouse models (Han, Tai, Jones, Scheuer, & Catterall, 2014). In fact, pharmaceutical interventions targeting glutamatergic and GABAergic receptors have demonstrated very early promise in alleviating social difficulties in ASD (Deidda et al., 2014; Rojas, Singel, Steinmetz, Hepburn, & Brown, 2014). More recently, studies in humans have shown that a dysregulation of excitatory and inhibitory processes is central to ASD and may be particularly relevant to core psychosocial dysfunction symptomology (Cochran et al., 2015; Ford, Nibbs, & Crewther, 2017b; Hegarty, Weber, Cirstea, & Beversdorf, 2018). EI has also been implicated in aberrant functional connectivity in patients with ASD (Ajram et al., 2017; Hegarty et al., 2018; von dem Hagen et al., 2011). Moreover, pharmaceutical intervention with riluzole, which blocks pre-synaptic release of glutamate and facilitates GABA_A receptor function, has been shown to increase prefrontal inhibitory flux (quantified as the ratio of GABA/GABA+ glutamate) in adult males with ASD, while reducing inhibitory flux in controls (Ajram et al., 2017). There were, however, no differences in prefrontal inhibition between clinical and non-clinical groups at baseline (Ajram et al., 2017). Furthermore, riluzole was shown to “normalize” prefrontal functional connectivity for patients, suggesting a strengthening of network pathways through pharmacological glutamate-GABA modulation (Ajram et al., 2017).

Aberrant cortical EI has also been used to explain dysfunction in schizophrenia spectrum disorders, and the EI imbalance hypothesis has been suggested to provide a possible mechanism to understanding behavioral dysfunction across the autism and schizophrenia spectra (e.g. Ajram et al., 2017; Ajram et al., 2019; Deidda et al., 2014; Ford, Abu-Akel, & Crewther, 2019; Foss-Feig et al., 2017; Gao & Penzes, 2015), particularly in light of accumulating evidence showing that autism and schizophrenia spectrum disorders share multiple phenotypes, including psychosocial symptomatology (Chisholm, Lin, Abu-Akel, & Wood, 2015; Gao & Penzes, 2015). In support of this account, research has shown, for example, that negative symptoms have been associated with glutamatergic dysregulation (Gruber, Santuccione Chadha, & Aach, 2014), that EI modulation via TMS, which specifically reduced intracortical GABA inhibition, has been associated with greater autistic trait severity in schizophrenia patients (Oliveira, Mitjans, Nitsche, Kuo, & Ehrenreich, 2018), and that EI abnormalities across both spectra is underlined by common genetic risk factors (Gao & Penzes, 2015). Furthermore, cortical gamma frequency oscillations, governed by GABAergic inhibition of parvalbumin cells, are disrupted in schizophrenia (Uhlhaas & Singer, 2010) and ASD (Rojas, 2014b). Taken together, these studies suggest that abnormal EI processes might be associated specifically with the dysfunction central to both conditions. However, since both ASD and schizophrenia spectrum disorder are heterogeneous and exhibit different clinical phenotypes (i.e. positive symptoms), investigating the specificity of EI processes to individual symptoms and symptom domains, such as psychosocial dysfunction, is consistent with the need for cross-diagnostic multidimensional explorations (Ajram et al., 2019; Foss-Feig et al., 2017; Hegarty et al., 2018).

Proton magnetic resonance spectroscopy (¹H-MRS) has proven effective in quantifying regional glutamate and GABA abnormalities, with the glutamate/GABA ratio providing an index for altered excitation–inhibition in ASD (Ajram et al., 2019). Several studies report increased regional glutamate (Bejjani et al., 2012; Doyle-Thomas et al., 2014; Hassan et al., 2013; Hegarty et al., 2018; Joshi et al., 2013; Naaijen, Lythgoe, Amiri, Buitelaar, & Glennon, 2015) and reduced regional GABA (Gaetz et al., 2014; Harada, Kubo, Nose, Nishitani, & Matsuda, 2011; Ito et al., 2017; Kubas et al., 2012; Port et al., 2017; Puts et al., 2017; Rojas et al., 2014), as well as an increased glutamate/GABA ratios (Hegarty et al., 2018) in ASD samples. Similarly, schizophrenia samples have been characterized by an increased glutamate and reduced GABA (Chiu et al., 2017; Guidotti et al., 2005; Marsman et al., 2014; Moghaddam & Javitt, 2012). These findings are inconsistent, however, with the variability largely due to differences in brain region, clinical sample characteristics (age, diagnosis, and symptoms), and data acquisition techniques (Ajram et al., 2019).

We have previously demonstrated that autistic traits (Ford et al., 2019), and psychosocial functioning specifically (Ford, Nibbs, & Crewther, 2017a; Ford et al., 2017b), are associated with increased glutamate and reduced GABA concentrations—a proxy for increased cortical excitation to inhibition—in the socially relevant right superior temporal cortex (STC) region. In contrast, psychosis-proneness exhibited the opposite interaction (Ford et al., 2019). Reduced GABA in ASD has also been reported in temporal brain regions (Gaetz et al., 2014; Port et al., 2017; Rojas et al., 2014). Furthermore, reduced respective neuronal and BOLD signal in the right temporoparietal junction, which overlaps the STC, during a social game has been observed for those with autism (Yuk et al., 2018) and higher ASD traits (Abu-Akel, Apperly, Wood, & Hansen, 2017), as well as reduced white matter in the right STC (von dem Hagen et al., 2011).

Together, scientific research clearly points toward alterations in EI processes in ASD and schizophrenia, which might be specifically associated with the social dysfunction characteristic of both conditions; however, the extent to which increased glutamate and/or reduced GABA processes contribute to such alterations remains unknown. Understanding the respective contributions of glutamate and GABA to such social dysfunction could prove valuable in the search for more targeted pharmacological interventions. This study is the first to directly probe the extent to which excitatory glutamate and inhibitory GABA interactively predict autistic and schizotypal symptom severity. This is specifically predicated on evidence for reduced GABA (Gaetz et al., 2014; Rojas et al., 2014) and increased glutamate (Brown, Singel, Hepburn, & Rojas, 2013) concentration in the auditory cortex of individuals with ASD. In utilizing a high functioning subclinical population, the potential effects of intelligence, medication use, excess motion during scanning, and acute symptomatology are strongly mitigated. It was hypothesized that glutamate and GABA would interactively modulate autistic and schizotypal symptom expressions associated with social functioning, specifically in the direction of increased glutamate and reduced GABA concentrations.

Methods and materials

Ethical approval was granted by Swinburne University’s Human Research Ethics Committee, and all participants provided written informed consent before commencing the study.

Participants

Participant recruitment and experimental procedures have been reported previously (Ford & Crewther, 2014; Ford et al., 2017b). Briefly, 38 non-clinical participants (18 males, 20 females; mean age (SD) = 23.22 (5.52)) were recruited for the ¹H-MRS component of a much larger study of 1678 young adults aged 18–40 years (Ford et al., 2017b) based on their scores on a questionnaire of pseudo-randomized Autism-Spectrum Quotient (AQ) and Schizotypal Personality Questionnaire (SPQ) items (Ford & Crewther, 2014). Participants were recruited for the larger study via social media and on-campus advertising, and were invited to complete the online questionnaire. Factor analysis revealed three dimensions of autism–schizotypy traits, with the first factor specifically relating to social difficulties. Participants in the upper and lower 20% of scores on this social difficulties dimension were invited to participate in the ¹H-MRS sub-study.

Participants were excluded if they had a current psychiatric condition. No participants reported a history of autism or schizophrenia, or current psychiatric condition; six participants disclosed a previous psychiatric condition (two depression and anxiety, one depression, one anxiety, one bipolar II, and one anorexia). One female participant was taking psychiatric medications, and was thus excluded (n = 37; 18 males, 19 females; mean age (SD) = 23.11 (5.27)); all remaining participants were free from medications, illicit substances, and nicotine at the time of the scan as determined by structured interview. This degree of prevalence of these conditions is to be expected given the multidimensionality of spectrum conditions, as well as the prevalence of mood disorders in the non-clinical population, and is unlikely to have affected the results of this study as there were no cases of active symptomatology.

Due to the potential modulating effect of the ovulation phase of the female menstrual cycle on GABA concentrations (De Bondt, De Belder, Vanhevel, Jacquemyn, & Parizel, 2015), the estimated menstrual cycle day for female participants was established based on number of days post-onset of menstrual bleeding. Linear and quadratic regression models demonstrated that menstrual cycle day was not associated with left or right STC GABA levels (p > 0.1), see Supplementary Information for more details.

Psychometric scales

The AQ is a 50-item scale that quantifies autistic tendency across five subscales: social skill, communication, attention switching, attention to detail, and imagination (Baron-Cohen, Wheelwright, Skinner, Martin, & Clubley, 2001). Responses to AQ items were given on a 4-point Likert-type scale from 1 (strongly disagree) to 4 (strongly agree), with original scores preserved—rather than scaling as per the original scoring system—to generate a total score out of 200 and to improve response reliability (Cronbach’s α = 0.82) (Austin, 2005; Ford & Crewther, 2014).

The SPQ is a 74-item scale, with three superordinate dimensions encapsulating the nine dimensions of schizotypal personality disorder: psychosis-proneness (ideas of reference, odd beliefs, unusual perceptual experiences, and suspiciousness); interpersonal features (social anxiety, no close friends, and constricted affect); and disorganized features (odd behavior and odd speech). Responses to SPQ items were given on a 4-point Likert-type scale from 1 (strongly disagree) to 4 (strongly agree), with original scores preserved to generate a score out of 296; Cronbach’s α SPQ = 0.86, psychosis-proneness = 0.77, interpersonal features = 0.77, disorganized features = 0.69 (Ford & Crewther, 2014).

¹H-MRS procedure

The ¹H-MRS protocol has been reported in detail in Ford et al. (2017b). To summarize, the T1-weighted image was acquired by a 3T Siemens TIM Trio whole-body magnetic resonance imaging system (Erlangen, Germany) using an MPRage pulse sequence with an inversion recovery (176 slices, slice thickness = 1.0 mm, voxel resolution = 1.0 mm³, TR = 1900 ms, TE = 2.52 ms, TI = 900 ms, bandwidth = 170 Hz/Px, flip angle = 9^o, field of view 350 mm x 263 mm × 350 mm, orientation sagittal, acquisition time = ~5 min). The image was used to position separate left and right STC voxels (20 mm × 30 mm × 20 mm) for ¹H-MRS acquisition. STC voxel location is presented in Figure 1, and includes parts of the insular and inferior prefrontal cortex. Glutamate concentration was quantified using an 80-ms TE PRESS sequence (Schubert, Gallinat, Seifert, & Rinneberg, 2004) with chemical shift selective (CHESS; Haas, 1986) water suppression (TR = 2000 ms, bandwidth = 1200 Hz, 80 averages, acquisition time = 2 min 48 s). Eight spectral water averages were acquired with identical PRESS parameters and shim. GABA+ macromolecules (GABA+) concentration was quantified using a MEGA-PRESS (Mescher, Merkle, Kirsch, Garwood, & Gruetter, 1998) editing sequence with CHESS water suppression (TE = 68 ms, TR = 1500 ms, bandwidth = 1000 Hz, edit-on pulse frequency = 1.9 ppm, edit-off pulse frequency = 7.5 ppm, edit pulse bandwidth = 44 Hz, 120 averages acquired, duration = 6 min 6 s). Twelve spectral unsuppressed water averages were acquired with identical MEGA-PRESS parameters. For all ¹H-MRS scans, an automatic shim was complemented with manual shimming until the line-width was less than 20 Hz. The right hemisphere 80-ms TE PRESS data for one participant were not recorded.

Figure 1.

Example voxel placement and average 80-ms TE PRESS and MEGA-PRESS spectra fits. Example left (a) and right (d) 2 × 2 × 3 voxel placement; average 80-ms TE PRESS spectra with isolated glutamate spectrum (inset) for left (b) and right (e) voxels; and average MEGA-PRESS edited spectra with isolated GABA+ peak at 3.0 ppm (inset) for left (c) and right (f) voxels. The shading represents ±1 standard error from the mean.

¹H-MRS analysis

To obtain glutamate concentrations, the 80-ms TE PRESS data were processed using TARQUIN version 4.3.7 (Wilson, Reynolds, Kauppinen, Arvanitis, & Peet, 2011), with eddy current correction. All participants’ data had adequate fit (see Figure 1 and Table 1), with a signal-to-noise ratio greater than 20 and water linewidth less than 12 Hz; however, left hemisphere data were excluded for one participant due to large standard deviation of water frequency (SD = ‒6.88).

Table 1.

Means and standard deviations (SD) for voxel tissue composition and fit statistics for GABA and glutamate spectra.

N = 37	Left hemisphere	Right hemisphere
Gray matter (%)	66.20 (5.41)	66.89 (5.70)
White matter (%)	25.86 (6.79)	24.26 (7.42)
Cerebrospinal fluid (%)	7.94 (2.61)	8.85 (2.82)
Glutamate water FWHM (Hz)	7.57 (1.01)	8.41 (1.77)
Glutamate water frequency, SD (Hz)	0.36 (1.54)	–0.24 (2.46)
Glutamate SNR	46.19 (9.97)	48.46 (11.94)
Glutamate CRLB^a	1.24 (0.50)	0.99 (0.62)
Glutamine CRLB^a	0.61 (0.23)	0.48 (0.26)
GABA+ creatine FWHM (Hz)	9.08 (0.87)	9.02 (0.81)
GABA+ water frequency, SD (Hz)	0.81 (0.66)	1.08 (1.11)
GABA+ fit error (%)	10.57 (2.66)	8.72 (2.39)
GABA+ SNR	10.15 (3.05)	12.29 (3.21)

FWHM: full width-half maximum; SNR: signal-to-noise ratio; CRLB: Cramer–Rao Lower Bound.

The CRLB reported here is the predicted metabolite fit error given in standard deviation.

GABA+ was quantified from MEGA-PRESS data using Gannet’s GABA analysis toolkit for MATLAB (version 2.0) (Edden, Puts, Harris, Barker, & Evans, 2014). The edit-on spectrum was subtracted from the edit-off spectrum to expose GABA+ concentration (institutional units (IU)) at 3.0 ppm relative to the water spectra (Mullins et al., 2014). All GABA+ spectra were adequate, see Ford et al. (2017b) for fit details. Glutamate and GABA+ concentrations were corrected for individual variation in gray matter, white matter, and cerebrospinal fluid using the formula provided in Harris, Puts, and Edden (2015).

Statistical analyses

Using standardized predictor values, regression analyses with Wald chi-square statistics tested the association of GABA+ and glutamate concentrations, and their interaction, with total autism trait expression, and each of its subscales, as well as the three SPQ domains: psychosis-proneness, interpersonal features and disorganized features. Regression analyses were performed using Generalized Linear Models (GLMs) with robust estimators, which allow for the treatment of non-normally distributed data and guards against the unduly influence of outliers (Rousseeuw & Leroy, 2005). To estimate the amount of variance explained by each model, we calculated Pseudo R², using the following formula

Pseudo R^{2} = 1 - \frac{D e v i a n c e}{N u l l D e v i a n c e}

Significant interactions were probed by adapting the method of Hayes and Matthes (2009) to accommodate regression parameters produced by the GLMs with robust estimators. Specifically, the effect of glutamate concentration on AQ subscales and SPQ dimension scores were examined, by convention, at the mean, at 1SD below the mean, and at 1SD above the mean of GABA+ concentration. The change in Pseudo R², which indicates the additional amount of variance explained by the inclusion of the interaction term vis-à-vis the main effect-only model, was also calculated. It is noteworthy that this interaction probe procedure does not involve splitting the sample into smaller groups using these cutoff points. Rather, it estimates the effect of a predictor on the dependent variable, while holding constant the other predictor at a discrete point. Accordingly, this approach allows us to infer from the model the effect of discrete levels of excitation and inhibition on autistic traits and schizotypal dimensions within the entire sample. Finally, to adjust for false discovery rate (FDR), we applied the Benjamini and Hochberg (1995) procedure where applicable (q-value cutoff 0.05).

Results

Demographics

There was no difference between male or female participants in age or on any of the SPQ or AQ subscales (all t < 0.98, all p > 0.22; see Table 2).

Table 2.

Sample characteristics presented in means with standard deviations in parentheses.

	Overall (N = 37)	Males (N = 18)	Females (N = 19)	T (df = 35)	p-value^a
Age	23.11 (5.27)	23.78 (5.96)	22.47 (4.61)	0.75	0.46
AQ subscales
Social skill	3.70 (2.70)	3.56 (2.33)	3.84 (3.06)	–0.32	0.75
Communication	3.59 (2.37)	3.39 (2.61)	3.79 (2.18)	–0.51	0.62
Attention switching	3.73 (2.82)	3.67 (2.87)	3.79 (2.86)	–0.13	0.90
Attention to detail	5.22 (1.62)	5.22 (1.96)	5.21 (1.27)	0.02	0.98
Imagination	3.46 (1.52)	3.72 (1.71)	3.21 (1.32)	1.02	0.31
SPQ subscales
Interpersonal features	9.84 (7.01)	10.22 (7.32)	9.47 (6.89)	0.32	0.75
Psychosis-proneness	8.43 (5.54)	8.61 (5.42)	8.26 (5.79)	0.19	0.85
Disorganized features	6.27 (4.27)	7.17 (4.68)	5.42 (3.78)	1.25	0.22

AQ: Autism Spectrum Quotient; SPQ: Schizotypal Personality Questionnaire.

p-values are of two-tailed independent t-tests.

Left superior temporal voxel

The Wald chi-square omnibus tests were non-significant for the AQ subscales or the SPQ dimensions (1.73 < χ²_{(df = 3)} < 4.96, ps_corr > 0.44).

Right superior temporal voxel

The Wald chi-square omnibus tests revealed a significant model for the AQ social skills subscale (χ²_{(df = 3)} = 12.76, p_corr = 0.021, Pseudo R² = 0.298) and the SPQ interpersonal dimension (χ²_{(df = 3)} = 13.04, p_corr = 0.021, Pseudo R² = 0.304), respectively, explaining 29.8% and 30.4% of the variance. The omnibus tests were non-significant for the other AQ subscales of communication, attention switching, attention to details, and imagination and for SPQ psychosis-proneness and disorganized dimensions (3.97 < χ²_(df ₌ ₃₎ < 8.73, ps_corr > 0.088).

With respect to the total AQ social skills model, parameter estimates revealed a significant and negative interaction of glutamate and GABA+ concentrations (β (SE) = ‒0.547 (0.193), Wald_χ² = 8.04, p = 0.005), contributing 24.2% to the overall variance explained by the model (Pseudo R² change = 0.072). Probing the interaction term revealed that glutamate concentration was significantly and positively associated with the social skills scores when GABA+ concentration was low (GABA+ = ‒1SD; β (SE) = 1.66 (0.49), t = 3.39, p = 0.002) and at the mean (GABA+ = mean; β (SE) = 1.11 (0.44), t = 2.54, p = 0.016). This effect attenuated and became non-significant at high GABA+ concentration (GABA+ = +1SD; β (SE) = 0.56 (0.49), t = 1.15, p = 0.258) (see Figure 2(a)).

Figure 2.

Visualization of the interactive effect of right hemisphere (RH) glutamate and GABA+ concentrations on social skills scores of the AQ and the interpersonal skills scores of the SPQ. Regression lines represent the effect of glutamate concentration (standardized z-score) on AQ social skills (panel a) and SPQ interpersonal skills (panel b) scores when GABA+ concentrations are at 1 standard deviation (SD) from the mean, at the mean (M), and at 1SD above the mean. Corresponding 3D scatter plots represent the raw participant data. Higher scores indicate greater difficulties in social/interpersonal skills.

With respect to the SPQ interpersonal dimension, parameter estimates revealed a significant and negative interaction of glutamate and GABA+ concentrations (β (SE) = ‒1.56 (0.477), Wald_χ² = 10.66, p = 0.001), contributing 29.4% to the overall variance explained by the model (Pseudo R² change = 0.089). Probing the interaction term revealed that glutamate concentration was significantly and positively associated with the SPQ interpersonal scores when GABA+ concentration was low (GABA+ = ‒1SD; β (SE) = 4.46 (1.26), t = 3.53, p = 0.001) and at the mean (GABA+ = mean; β (SE) = 2.91 (1.12), t = 2.59, p = 0.014). This effect attenuated and became non-significant at high GABA+ concentration (GABA+ = +1SD; β (SE) = 1.35 (0.72), t = 1.26, p = 0.291; see Figure 2(b)).

Discussion

Based on the hyper-glutamatergic and hypo-GABAergic hypotheses of autism (Cellot & Cherubini, 2014; Fatemi, 2008; Fatemi et al., 2002), we hypothesized that the expression of autistic social skills and schizotypal interpersonal skills would be interactively modulated by cortical excitation and inhibition in the STC. In support of this hypothesis, we found that the positive association of right STC glutamate concentration with deficits in the autism social skills (Figure 2(a)) and the schizotypal interpersonal domains decreased with increase in the GABA+ concentration (Figure 2(b)). Importantly, this effect was not observed for psychosis-proneness or disorganized features of schizotypy; hence, an EI imbalance, in the direction of increased glutamate with concurrently reduced GABA+, appears specific to psychosocial difficulties across the autistic and schizotypal spectra.

Our findings are consistent with evidence from animal studies showing that the social deficits caused by increasing the cellular EI balance can be ameliorated following an increase in the inhibitory neurotransmission of GABAergic neurons (Han et al., 2014; Yizhar et al., 2011), as well as with evidence linking higher glutamine and lower GABA/Cr levels to impairments in social cognition and functioning in adolescents with ASD (Cochran et al., 2015). Together, these findings add to the growing body of literature suggesting that aberrant glutamatergic and GABAergic modulation might be central neural marker and drug target for ASD (Ajram et al., 2019), and schizophrenia (Stone, 2011), and perhaps specifically, psychosocial difficulties (Deidda et al., 2014; Rojas et al., 2014).

There were no significant sex differences in AQ and SPQ dimension scores, which is somewhat unexpected given several studies show sex differences in AQ and SPQ scores, and ASD is more prevalent in males than females (Baron-Cohen et al., 2001; Raine, 1993). The absence of any sex differences could be in part due to the nature of participant recruitment, given we invited those in the upper and lower ends of a social difficulties spectrum scores. This sample characteristic may in fact be a strength, given there is an even distribution of males and females at the lower and upper ends of the subscales relating to social functioning in this sample.

Of note is the laterality of the interaction effect to the right STC. While the STC in general is a putative region for social processes (von dem Hagen et al., 2011), the right hemisphere is more specifically involved in the processing of prosody and the paralinguistic aspects of language (Lindell, 2006). The right hemisphere specificity observed herein is consistent with previous reports of an association between autistic traits and reduced BOLD signal in the right temporoparietal junction, which overlaps the STC, during a theory of mind task (Abu-Akel et al., 2017). Autistic traits have also been associated with reduced right posterior STC white matter (von dem Hagen et al., 2011).

It is important to note that, due to the interrelationship between glutamine, glutamate and GABA metabolism, specialized imaging and analysis techniques are required to distinguish between them. We utilized gold-standard acquisition techniques to quantify glutamate (80-ms TE PRESS) and GABA+ (MEGA-PRESS) in this study, and can thus be confident of the accuracy of the concentrations used in subsequent statistical analyses.

Investigating these associations in non-pathological individuals with subclinical symptom expressions may limit, some may argue, relevance for understanding clinical psychosocial dysfunction; however, the association of trait characteristics of ASD and schizotypy with inhibitory and excitatory signaling in healthy individuals, and the concordance of these findings with previous studies in animals and humans, confirm the importance of dimension-based approaches to advancing psychopharmacological research into the psychosocial dysfunction central to multidimensional spectrum conditions such as ASD and schizophrenia spectrum disorder.

Our findings inform this active area of research by showing that specific phenotypes might be affected by an EI imbalance in the STC. Future research should attempt to replicate our findings in the STC and to test our hypothesis in other regions implicated in the expressions of social processing difficulties in ASD and schizophrenia spectrum disorder. Finally, examining the interactive effect of excitatory and inhibitory neurotransmission on outcomes provides a framework to understanding inter-individual differences, and possibly how shared mechanisms across conditions can result in heterogeneous outcomes.

Supplemental Material

AUT866030_Lay_Abstract – Supplemental material for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Supplemental material, AUT866030_Lay_Abstract for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission by Talitha C Ford, David P Crewther and Ahmad Abu-Akel in Autism

Supplemental Material

AUT866030_Supplemental_material – Supplemental material for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Supplemental material, AUT866030_Supplemental_material for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission by Talitha C Ford, David P Crewther and Ahmad Abu-Akel in Autism

Footnotes

Acknowledgements

The authors thank Swinburne Neuroimaging for their support through data collection and analysis. The authors also acknowledge the facilities of Swinburne Neuroimaging (SNI) and its flagship funding from the National Imaging Facility (NIF) under the National Collaborative Researcher Infrastructure Strategy (NCRIS) implemented by the Australian Government.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by a Swinburne University Neuroimaging Grant awarded to Prof. D.P.C. and Dr T.C.F., and a National Health and Medical Research Council of Australia grant awarded to Prof. D.P.C. (APP1004740). Dr T.C.F. was supported by a Swinburne University Postgraduate Research Award.

ORCID iDs

Talitha C Ford

Ahmad Abu-Akel

Supplemental material

Supplemental material for this article is available online.

References

Abu-Akel

Apperly

I. A.

Wood

S. J.

Hansen

P. C.

(2017). Autism and psychosis expressions diametrically modulate the right temporoparietal junction. Social Neuroscience, 12, 506–518. doi:10.1080/17470919.2016.1190786

Ajram

L. A.

Horder

Mendez

M. A.

Galanopoulos

Brennan

L. P.

Wichers

R. H.

… McAlonan

G. M.

(2017). Shifting brain inhibitory balance and connectivity of the prefrontal cortex of adults with autism spectrum disorder. Translational Psychiatry, 7, e1137. doi:10.1038/tp.2017.104

Ajram

L. A.

Pereira

A. C.

Durieux

A. M. S.

Velthius

H. E.

Petrinovic

M. M.

McAlonan

G. M.

(2019). The contribution of [1H] magnetic resonance spectroscopy to the study of excitation-inhibition in autism. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 89, 236–244. doi:10.1016/j.pnpbp.2018.09.010

Austin

E. J.

(2005). Personality correlates of the broader autism phenotype as assessed by the autism spectrum quotient (AQ). Personality and Individual Differences, 38, 451–460. doi:10.1016/j.paid.2004.04.022

Baron-Cohen

Wheelwright

Skinner

Martin

Clubley

(2001). The autism-spectrum quotient (AQ): Evidence from Asperger Syndrome/high-functioning autism, males and females, scientists and mathematicians. Journal of Autism and Developmental Disorders, 31, 5–17. Retrieved from http://www.ncbi.nlm.nih.gov/pubmed/11439754

Bejjani

O’Neill

Kim

J. A.

Frew

A. J.

Yee

V. W.

… Levitt

J. G.

(2012). Elevated glutamatergic compounds in pregenual anterior cingulate in pediatric autism spectrum disorder demonstrated by 1H MRS and 1H MRSI. PLoS ONE, 7, e38786. doi:10.1371/journal.pone.0038786

Benjamini

Hochberg

(1995). Controlling the false discovery rate: A practical and powerful approach to multiple testing. Journal of the Royal Statistical Society. Series B (Methodological), 57, 289–300.

Brown

M. S.

Singel

Hepburn

Rojas

D. C.

(2013). Increased glutamate concentration in the auditory cortex of persons with autism and first-degree relatives: A (1)H-MRS study. Autism Research, 6, 1–10. doi:10.1002/aur.1260

Cellot

Cherubini

(2014). GABAergic signaling as therapeutic target for autism spectrum disorders. Frontiers in Pediatrics, 2, Article 70. doi:10.3389/fped.2014.00070

10.

Chisholm

Lin

Abu-Akel

Wood

S. J.

(2015). The association between autism and schizophrenia spectrum disorders: A review of eight alternate models of co-occurrence. Neuroscience & Biobehavioural Reviews, 55, 173–183. doi:10.1016/j.neubiorev.2015.04.012

11.

Chiu

P. W.

Lui

S. S. Y.

Hung

K. S. Y.

Chan

R. C. K.

Chan

Sham

P. C.

… Mak

H. K. F.

(2017). In vivo gamma-aminobutyric acid and glutamate levels in people with first-episode schizophrenia: A proton magnetic resonance spectroscopy study. Schizophrenia Research, 193, 295–303. doi:10.1016/j.schres.2017.07.021

12.

Cochran

D. M.

Sikoglu

E. M.

Hodge

S. M.

Edden

R. A.

Foley

Kennedy

D. N.

… Frazier

J. A.

(2015). Relationship among glutamine, gamma-aminobutyric acid, and social cognition in autism spectrum disorders. Journal of Child Adolescent Psychopharmacology, 25, 314–322. doi:10.1089/cap.2014.0112

13.

De Bondt

De Belder

Vanhevel

Jacquemyn

Parizel

P. M

. (2015). Prefrontal GABA concentration changes in women-influence of menstrual cycle phase, hormonal contraceptive use, and correlation with premenstrual symptoms. Brain Research, 1597, 129–138. doi:10.1016/j.brainres.2014.11.051

14.

Deidda

Bozarth

I. F.

Cancedda

(2014). Modulation of GABAergic transmission in development and neurodevelopmental disorders: Investigating physiology and pathology to gain therapeutic perspectives. Frontiers in Cellular Neuroscience, 8, Article 119. doi:10.3389/fncel.2014.00119

15.

Doyle-Thomas

K. A.

Card

Soorya

L. V.

Wang

A. T.

Fan

Anagnostou

(2014). Metabolic mapping of deep brain structures and associations with symptomatology in autism spectrum disorders. Research in Autism Spectrum Disorders, 8, 44–51. doi:10.1016/j.rasd.2013.10.003

16.

Edden

R. A.

Puts

N. A.

Harris

A. D.

Barker

P. B.

Evans

C. J.

(2014). Gannet: A batch-processing tool for the quantitative analysis of gamma-aminobutyric acid-edited MR spectroscopy spectra. Journal of Magnetic Resonance Imaging, 40, 1445–1452. doi:10.1002/jmri.24478

17.

Fatemi

S. H.

(2008). The hyperglutamatergic hypothesis of autism. Progress in Neuro-Psychopharmacology & Biological Psychiatry, 32, 911, author reply 912–913. doi:10.1016/j.pnpbp.2007.11.004

18.

Fatemi

S. H.

Halt

A. R.

Stary

J. M.

Kanodia

Schulz

S. C.

Realmuto

G. R.

(2002). Glutamic acid decarboxylase 65 and 67 kDa proteins are reduced in autistic parietal and cerebellar cortices. Biological Psychiatry, 52, 805–810. doi:10.1016/s0006-3223(02)01430-0

19.

Fatemi

S. H.

Reutiman

T. J.

Folsom

T. D.

Thuras

P. D.

(2009). GABA(A) receptor downregulation in brains of subjects with autism. Journal of Autism and Developmental Disorders, 39, 223–230. doi:10.1007/s10803-008-0646-7

20.

Ford

T. C.

Abu-Akel

Crewther

D. P.

(2019). The association of excitation and inhibition signaling with the relative symptom expression of autism and psychosis-proneness: Implications for psychopharmacology. Progress in Neuro-Psychopharmacology and Biological Psychiatry, 88, 235–242. doi:10.1016/j.pnpbp.2018.07.024

21.

Ford

T. C.

Crewther

D. P.

(2014). Factor analysis demonstrates a common schizoidal phenotype within autistic and schizotypal tendency: Implications for neuroscientific studies. Front Psychiatry, 5, Article 117. doi:10.3389/fpsyt.2014.00117

22.

Ford

T. C.

Nibbs

Crewther

D. P.

(2017a). Glutamate/GABA+ ratio is associated with the psychosocial domain of autistic and schizotypal traits. PLoS ONE, 12, e0181961. doi:10.1371/journal.pone.0181961

23.

Ford

T. C.

Nibbs

Crewther

D. P.

(2017b). Increased glutamate/GABA+ ratio in a shared autistic and schizotypal trait phenotype termed social disorganisation. NeuroImage: Clinical, 16, 125–131. doi:10.1016/j.nicl.2017.07.009

24.

Foss-Feig

J. H.

Adkinson

B. D.

J. L.

Yang

Srihari

V. H.

McPartland

J. C.

… Anticevic

(2017). Searching for cross-diagnostic convergence: Neural mechanisms governing excitation and inhibition balance in schizophrenia and autism spectrum disorders. Biological Psychiatry, 81, 848–861. doi:10.1016/j.biopsych.2017.03.005

25.

Gaetz

Bloy

Wang

D. J.

Port

R. G.

Blaskey

Levy

S. E.

Roberts

T. P.

(2014). GABA estimation in the brains of children on the autism spectrum: Measurement precision and regional cortical variation. NeuroImage, 86, 1–9. doi:10.1016/j.neuroimage.2013.05.068

26.

Gao

Penzes

(2015). Common mechanisms of excitatory and inhibitory imbalance in schizophrenia and autism spectrum disorders. Current Molecular Medicine, 15, 146–167.

27.

Gruber

Santuccione Chadha

Aach

(2014). Magnetic resonance imaging in studying schizophrenia, negative symptoms, and the glutamate system. Front Psychiatry, 5, Article 32. doi:10.3389/fpsyt.2014.00032

28.

Guidotti

Auta

Davis

J. M.

Dong

Grayson

D. R.

Veldic

… Costa

(2005). GABAergic dysfunction in schizophrenia: New treatment strategies on the horizon. Psychopharmachology, 180, 191–205. doi:10.1007/s00213-005-2212-8

29.

Haas

(1986). Localization of unaffected spins in NMR imaging and spectroscopy (LOCUS Spectroscopy). Magnetic Resonance in Medicine, 3, 963–969.

30.

Han

Tai

Jones

C. J.

Scheuer

Catterall

W. A.

(2014). Enhancement of inhibitory neurotransmission by GABAA receptors having alpha2,3-subunits ameliorates behavioral deficits in a mouse model of autism. Neuron, 81, 1282–1289. doi:10.1016/j.neuron.2014.01.016

31.

Harada

Kubo

Nose

Nishitani

Matsuda

(2011). Measurement of variation in the human cerebral GABA level by in vivo MEGA-editing proton MR spectroscopy using a clinical 3 T instrument and its dependence on brain region and the female menstrual cycle. Human Brain Mapping, 32, 828–833. doi:10.1002/hbm.21086

32.

Harris

A. D.

Puts

N. A.

Edden

R. A.

(2015). Tissue correction for GABA-edited MRS: Considerations of voxel composition, tissue segmentation, and tissue relaxations. Journal of Magnetic Resonance Imaging, 42, 1431–1440. doi:10.1002/jmri.24903

33.

Hassan

T. H.

Abdelrahman

H. M.

Abdel Fattah

N. R.

El-Masry

N. M.

Hashim

H. M.

El-Gerby

K. M.

Abdel Fattah

N. R.

(2013). Blood and brain glutamate levels in children with autistic disorder. Research in Autism Spectrum Disorders, 7, 541–548. doi:10.1016/j.rasd.2012.12.005

34.

Hayes

A. F.

Matthes

(2009). Computational procedures for probing interactions in OLS and logistic regression: SPSS and SAS implementations. Behavior Research Methods, 41, 924–936. doi:10.3758/BRM.41.3.924

35.

Hegarty

J. P.

Weber

D. J.

Cirstea

C. M.

Beversdorf

D. Q.

(2018). Cerebro-cerebellar functional connectivity is associated with cerebellar excitation–inhibition balance in autism spectrum disorder. Journal of Autism and Developmental Disorders, 48, 3460–3473. doi:10.1007/s10803-018-3613-y

36.

Ito

Mori

Harada

Hisaoka

Toda

Mori

… Kagami

(2017). A proton magnetic resonance spectroscopic study in autism spectrum disorder using a 3-tesla clinical magnetic resonance imaging (MRI) system: The anterior cingulate cortex and the left cerebellum. Journal of Child Neurology, 32, 731–739. doi:10.1177/0883073817702981

37.

Joshi

Biederman

Wozniak

Goldin

R. L.

Crowley

Furtak

… Gonenc

(2013). Magnetic resonance spectroscopy study of the glutamatergic system in adolescent males with high-functioning autistic disorder: A pilot study at 4T. European Archives of Psychiatry and Clinical Neuroscience, 263, 379–384. doi:10.1007/s00406-012-0369-9

38.

Kubas

Kulak

Sobaniec

Tarasow

Lebkowska

Walecki

(2012). Metabolite alterations in autistic children: A 1H MR spectroscopy study. Advances in Medical Sciences, 57, 152–156. doi:10.2478/v10039-012-0014-x

39.

Lindell

A. K.

(2006). In your right mind: Right hemisphere contributions to language processing and production. Neuropsychology Review, 16, 131–148. doi:10.1007/s11065-006-9011-9

40.

Marsman

Mandl

R. C. W.

Klomp

D. W. J.

Bohlken

M. M.

Boer

V. O.

Andreychenko

… Hulshoff Pol

H. E.

(2014). GABA and glutamate in schizophrenia: A 7 T 1H-MRS study. NeuroImage: Clinical, 6, 398–407. doi:10.1016/j.nicl.2014.10.005

41.

Mescher

Merkle

Kirsch

Garwood

Gruetter

(1998). Simultaneous in vivo spectral editing and water suppression. NMR in Biomedicine, 11, 266–272.

42.

Moghaddam

Javitt

(2012). From revolution to evolution: The glutamate hypothesis of schizophrenia and its implication for treatment. Neuropsychopharmacology, 37, 4–15. doi:10.1038/npp.2011.181

43.

Mullins

P. G.

McGonigle

D. J.

O’Gorman

R. L.

Puts

N. A.

Vidyasagar

Evans

C. J.

… Edden

R. A.

(2014). Current practice in the use of MEGA-PRESS spectroscopy for the detection of GABA. NeuroImage, 86, 43–52. doi:10.1016/j.neuroimage.2012.12.004

44.

Naaijen

Lythgoe

D. J.

Amiri

Buitelaar

J. K.

Glennon

J. C.

(2015). Fronto-striatal glutamatergic compounds in compulsive and impulsive syndromes: A review of magnetic resonance spectroscopy studies. Neuroscience & Biobehavioral Reviews, 52, 74–88. doi:10.1016/j.neubiorev.2015.02.009

45.

Oliveira

Mitjans

Nitsche

M. A.

Kuo

M. F.

Ehrenreich

(2018). Excitation-inhibition dysbalance as predictor of autistic phenotypes. Journal of Psychiatric Research, 104, 96–99. doi:10.1016/j.jpsychires.2018.06.004

46.

Port

R. G.

Gaetz

Bloy

Wang

D. J.

Blaskey

Kuschner

E. S.

… Roberts

T. P.

(2017). Exploring the relationship between cortical GABA concentrations, auditory gamma-band responses and development in ASD: Evidence for an altered maturational trajectory in ASD. Autism Research, 10, 593–607. doi:10.1002/aur.1686

47.

Purcell

Jeon

Zimmerman

Blue

Pevsner

(2001). Postmortem brain abnormalities of the glutamate neurotransmitter system in autism. Neurology, 57, 1618–1628.

48.

Puts

N. A.

Wodka

E. L.

Harris

A. D.

Crocetti

Tommerdahl

Mostofsky

S. H.

Edden

R. A.

(2017). Reduced GABA and altered somatosensory function in children with autism spectrum disorder. Autism Research, 10, 608–619. doi:10.1002/aur.1691

49.

Raine

(1993). Sex differences in schizotypal personality in a nonclinical population. Journal of Abnormal Psychology, 101, 361–364.

50.

Rojas

D. C.

(2014a). The role of glutamate and its receptors in autism and the use of glutamate receptor antagonists in treatment. Journal of Neural Transmission, 121, 891–905. doi:10.1007/s00702-014-1216-0

51.

Rojas

D. C.

(2014b). γ-Band abnormalities as markers of autism spectrum disorders. Biomarkers in Medicine, 8, 353–368.

52.

Rojas

D. C.

Singel

Steinmetz

Hepburn

Brown

M. S.

(2014). Decreased left perisylvian GABA concentration in children with autism and unaffected siblings. NeuroImage, 86, 28–34. doi:10.1016/j.neuroimage.2013.01.045

53.

Rousseeuw

P. J.

Leroy

A. M.

(2005). Robust Regression and Outlier Detection (Vol. 589). New York, NY: John Wiley.

54.

Schubert

Gallinat

Seifert

Rinneberg

(2004). Glutamate concentrations in human brain using single voxel proton magnetic resonance spectroscopy at 3 Tesla. NeuroImage, 21, 1762–1771. doi:10.1016/j.neuroimage.2003.11.014

55.

Shimmura

Suzuki

Iwata

Tsuchiya

K. J.

Ohno

Matsuzaki

Wakuda

(2013). Enzymes in the glutamate-glutamine cycle in the anterior cingulate cortex in postmortem brain of subjects with autism. Molecular Autism, 4, Article 6. doi:10.1186/2040-2392-4-6

56.

Stone

J. M.

(2011). Glutamatergic antipsychotic drugs: A new dawn in the treatment of schizophrenia? Therapeutic Advances in Psychopharmacology, 1, 5–18. doi:10.1177/2045125311400779

57.

Uhlhaas

P. J.

Singer

(2010). Abnormal neural oscillations and synchrony in schizophrenia. Nature Reviews Neuroscience, 11, 100–113. doi:10.1038/nrn2774

58.

von dem Hagen

E. A. H.

Nummenmaa

Engell

A. D.

Ewbank

M. P.

Calder

A. J

. (2011). Autism spectrum traits in the typical population predict structure and function in the posterior superior temporal sulcus. Cerebral Cortex, 21, 493–500. doi:10.1093/cercor/bhq062

59.

Wilson

Reynolds

Kauppinen

R. A.

Arvanitis

T. N.

Peet

A. C.

(2011). A constrained least-squares approach to the automated quantitation of in vivo (1)H magnetic resonance spectroscopy data. Magnetic Resonance in Medicine, 65, 1–12. doi:10.1002/mrm.22579

60.

Yizhar

Fenno

L. E.

Prigge

Schneider

Davidson

T. J.

O’Shea

D. J.

… Deisseroth

(2011). Neocortical excitation/inhibition balance in information processing and social dysfunction. Nature, 477, 171–178. doi:10.1038/nature10360

61.

Yuk

Urbain

Pang

E. W.

Anagnostou

Buchsbaum

Taylor

M. J.

(2018). Do you know what I’m thinking? Temporal and spatial brain activity during a theory-of-mind task in children with autism. Developmental Cognitive Neuroscience, 34, 139–147. doi:10.1016/j.dcn.2018.08.001

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.49 MB

0.18 MB

Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Abstract

Keywords

Methods and materials

Participants

Psychometric scales

1H-MRS procedure

1H-MRS analysis

Statistical analyses

Results

Demographics

Left superior temporal voxel

Right superior temporal voxel

Discussion

Supplemental Material

AUT866030_Lay_Abstract – Supplemental material for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Supplemental Material

AUT866030_Supplemental_material – Supplemental material for Psychosocial deficits across autism and schizotypal spectra are interactively modulated by excitatory and inhibitory neurotransmission

Footnotes

Acknowledgements

Declaration of conflicting interests

Funding

ORCID iDs

Supplemental material

References

Supplementary Material

¹H-MRS procedure

¹H-MRS analysis