Reduced GABAergic neuropil and interneuron profiles in schizophrenia: Complementary analysis of disease course-related differences
Introduction
Although the dopamine and glutamate hypotheses are leading theories regarding the pathophysiology of schizophrenia (Howes et al., 2015; Iasevoli et al., 2014; Olney, 1995), numerous studies have also suggested abnormal GABAergic neurotransmission as a factor (Akbarian et al., 1995; Bird et al., 1977; Blum and Mann, 2002; Glausier et al., 2015; Guidotti et al., 2000). Glutamic acid decarboxylase (GAD), a key enzyme in the synthesis of the neurotransmitter GABA (Lee et al., 2019; Steiner et al., 2016), is found in two isoforms, GAD65 and GAD67, which are present in most GABA-containing neurons in the brain. However, transiently-activated GAD65 appears to be restricted to membranes and nerve endings, whereas constitutively active GAD67 is more widely-distributed in cells (Bu et al., 1992; Kaufman et al., 1991). Expression of GAD65 appears to be regulated more at the epigenetic level and thus may play a more important role in the abnormal plasticity of central GABA synapses in chronic disease (Pan, 2012). GAD67-mediated GABA synthesis is specifically involved in regulation of GABAergic axonal and synaptic morphogenesis (Chattopadhyaya et al., 2007). Moreover, activity-driven expression of GAD67 controls the synthesis of GABA for synaptic release (Lau and Murthy, 2012). Reports on GAD67 are more prominent in schizophrenia research with most studies suggesting decreased expression of the GAD67 gene (for a review see: Akbarian and Huang, 2006).
We recently observed a significant decrease in GAD65/67 immunoreactive (GAD-ir) profiles in the hippocampal CA1 region and superior temporal gyrus (STG) layer V of paranoid compared to residual schizophrenia patients, using manual software-aided quantification of GAD-ir profiles and manual adjustment of the staining intensity threshold (Steiner et al., 2016). A previous study of a similar cohort revealed increased GAD-ir cell profiles in the subiculum and parahippocampal gyrus (Schreiber et al., 2011). However, the latter study did not provide a comparative analysis of residual versus paranoid schizophrenia, and GAD-immunostained tissue sections were evaluated via manual GAD-ir cell counting (Schreiber et al., 2011). Therefore, user-dependent subjective influences cannot be ruled out.
Here, we aimed to re-assess the previous findings using a comprehensive histological investigation with fully automated quantitative analysis. We also attempted to account for diagnostic issues and potential impact of illness duration. The automated approach enabled an objective, simultaneous evaluation of GAD-ir neuropil and somata in six different brain regions thought to be involved in schizophrenia pathophysiology.
The STG is active during auditory hallucinations and thus relevant for studies of paranoid schizophrenia (Nenadic et al., 2010). The STG layers III and V were analysed separately as our previous study indicated possible differences in GAD distribution, potentially corresponding to their different functions in auditory processing (Steiner et al., 2016). The limbic thalamic mediodorsal (MD) and laterodorsal (LD) nuclei represent integrative hubs in higher cognitive and emotional processes, which are disturbed and show morphological abnormalities in schizophrenia patients (for reviews see: Alelú-Paz and Giménez-Amaya, 2008; Dorph-Petersen and Lewis, 2017; Pergola et al., 2015). The hippocampal regions CA1 and dentate gyrus (DG) were selected because a role of GABA in hippocampal hyperactivity is suspected in schizophrenia (Heckers et al., 2002; Heckers and Konradi, 2015). Dysfunction of the CA1 region has been associated with positive symptoms and found to predict illness progression (Schobel et al., 2013; Zierhut et al., 2013). Consistent with previous results of GAD-ir neuropil evaluation (Steiner et al., 2016), we found abnormalities in glial markers in this region, which differed between paranoid and residual schizophrenia (Busse et al., 2012; Gos et al., 2014; Steiner et al., 2008).
Section snippets
Brain characteristics
Brains of 16 schizophrenia patients and 16 controls without neuropsychiatric diseases were obtained from the Magdeburg Brain Bank in accordance with EU regulations and the local institutional review board (Steiner et al., 2016). According to DSM-IV criteria, 10 patients had suffered from paranoid schizophrenia and 6 from residual schizophrenia. DSM-IV diagnoses were obtained by experienced psychiatrists using clinical records and structured interviews with the physicians involved in the
Appearance of GAD-ir cells and neuropil
GAD-ir cells and neuropil evaluated microscopically (200× magnification) is presented in Fig. 2A/D/G. GAD-ir cells had a dark purple colour. Thus, it was possible to count these by staining intensity threshold analysis, with only GAD-ir cells surpassing the automatically-set threshold. GAD-ir cells showed different region-specific size and shape characteristics, being larger in LD and MD. GAD-ir neuropil were visible as a fine network of fibers connected to GAD-ir cells. In cross section,
Discussion
Our findings of decreased GAD-ir profiles in the investigated brain regions in schizophrenia suggest altered GABAergic neurotransmission, consistent with most previous studies, reporting decreased GAD65/67 expression in schizophrenia (Akbarian et al., 1995; Hashimoto, 2005; Hashimoto et al., 2008; Steiner et al., 2016; Thompson et al., 2009; Volk et al., 2000; Woo et al., 2004). The MD and STG layer V contributed most to the reduction in GAD-ir neuropil. An important involvement of the thalamus
Conclusion
Our results support the hypothesis of an impaired GABAergic system in schizophrenia patients. We found an overall decrease in GAD-ir neuropil and cell profiles, attributable to significant decreases in STG layer V and the MD. By using region- and subgroup-specific analyses of GAD-ir cell profiles, our study revealed that patients with paranoid schizophrenia had a more prominent loss of GAD-ir cells than the residual subgroup. Moreover, our current study suggests an influence of antipsychotic
Author contributions
JS, TG, KSf, HGB and BB conceived the study. AF and VM performed the digital image analysis of histological sections, performed the statistical analysis, created the figures and tables und supervision by HD and KSf. AF, VM, KSf and JS wrote the first version of the manuscript. TG, TF, KSz, GML reviewed the data analysis and contributed significantly to manuscript writing. CM examined all brains as an experienced neuropathologist. BB, HGB and PCG edited the final version of the manuscript and
Author statement
Antonia Förster: Methodology, Investigation, Formal analysis, Writing – Original draft, visualization. Vera Model: Methodology, Investigation, Formal analysis, Writing – Original draft, visualization. Thomas Frodl: Writing - Review & Editing, Supervision, Funding acquisition. Kolja Schiltz: Writing - Review & Editing. Henrik Dobrowolny: Software, Formal analysis, Data curation. Gabriela Meyer-Lotz: Resources, Data curation. Paul C. Guest: Writing - Review & Editing. Christian Mawrin:
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Funding
The Magdeburg Brain Bank and subsequently this investigation have been supported by the Stanley Medical Research Foundation (Grant No. 07R-1832) to BB and JS, the DFG-SFB 779/6–1 to BB, the Saxony-Anhalt Ministry of Research (XN3594O/0405 M, N2-OVGU), the German Ministry of Research (“BrainNet“, BMBF NBL-3/2 and 01ZZ0407) and the Alfried-Krupp-von-Bohlen-und-Halbach foundation.
Declaration of competing interest
The authors declare no conflict of interest.
References (80)
- et al.
Molecular and cellular mechanisms of altered GAD1/GAD67 expression in schizophrenia and related disorders
Brain Research Reviews
(2006) - et al.
Progressive brain change in schizophrenia: a prospective longitudinal study of first-episode schizophrenia
Biological Psychiatry
(2011) - et al.
Uncoupling of GABAA and benzodiazepine receptor binding activity in the hippocampal formation of schizophrenic brain
Brain Research
(1997) - et al.
Increased brain dopamine and reduced glutamic acid decarboxylase and choline acetyl transferase activity in schizophrenia and related psychoses
The Lancet
(1977) - et al.
Different distribution patterns of lymphocytes and microglia in the hippocampus of patients with residual versus paranoid schizophrenia: further evidence for disease course-related immune alterations?
Brain, Behavior, and Immunity
(2012) - et al.
GAD67-mediated GABA synthesis and signaling regulate inhibitory synaptic innervation in the visual cortex
Neuron
(2007) - et al.
Volumes of association thalamic nuclei in schizophrenia: a postmortem study
Schizophrenia Research
(2003) - et al.
Postmortem structural studies of the thalamus in schizophrenia
Schizophrenia Research
(2017) - et al.
Reduced sleep spindle activity point to a TRN-MD thalamus-PFC circuit dysfunction in schizophrenia
Schizophrenia Research
(2017) - et al.
Lower glutamic acid decarboxylase 65-kDa isoform messenger RNA and protein levels in the prefrontal cortex in schizoaffective disorder but not schizophrenia
Biological Psychiatry
(2015)