Re-arrangements of gene transcripts at glutamatergic synapses after prolonged treatments with antipsychotics: A putative link with synaptic remodeling

Highlights

• Chronic antipsychotics at different doses may impact synaptic plasticity genes.

• Molecular imaging is used to test haloperidol, asenapine and olanzapine effects.

• Homers, Shank1, PSD-95 and Arc expression levels are differentially modulated.

• Antipsychotics trigger molecular changes in brain regions linked to schizophrenia.

Abstract

Objectives

The postsynaptic density (PSD) represents a site of dopamine-glutamate integration. Despite multiple evidence of PSD involvement in antipsychotic-induced synaptic changes, there are no direct head-to-head comparisons of the effects at the PSD of antipsychotics with different receptor profile and at different doses after chronic administration.

Methods

Molecular imaging of gene expression was used to investigate whether chronic treatment with first and second generation antipsychotics (haloperidol, asenapine and olanzapine) may induce changes in the expression levels of PSD transcripts involved in schizophrenia pathophysiology, i.e. Homers, Shank1, PSD-95 and Arc.

Results

Genes' expression patterns were differentially modulated after chronic administration of typical and atypical antipsychotics as well as by the same compound administered at different doses. Antipsychotic treatment reduced gene expression in cortical regions, while Homer1a was still induced in striatum by haloperidol even after prolonged treatment. Moreover, chronic treatments appeared to cause a “de-recruitment” of brain regions demonstrated to be activated in acute treatments, with a prominent effect in the cortex rather than in striatum.

Conclusions

These results let hypothesize that prolonged antipsychotic treatment may trigger a set of plastic changes involving scaffolding and effector molecules causing a possible re-arrangement of PSD transcripts in brain regions relevant to schizophrenia pathophysiology.

Introduction

There is increasing interest for searching putative brain structural changes after chronic antipsychotic treatment, and to understand how antipsychotics with different receptor profiles and at different doses may impact brain plasticity (Ho et al., 2011, Vita et al., 2015). The postsynaptic density (PSD) is an attractive target for studying antipsychotic-induced changes after acute and chronic treatments. The PSD is an electron-dense thickness beneath post-glutamatergic synapses, where receptors, adaptors, and scaffolding proteins interact at the crossroads of dopamine-glutamate interaction (Iasevoli et al., 2013, Verpelli et al., 2012), and participate in synaptic plasticity processes (de Bartolomeis et al., 2014, Gao et al., 2013, Iasevoli et al., 2014).

PSD proteins have been involved in psychosis pathophysiology (de Bartolomeis et al., 2014, Dean et al., 2015, Hall et al., 2015), which is in agreement with the hypothesis that schizophrenia may be a disease of aberrant synaptic plasticity. Antipsychotic agents, the mainstay of schizophrenia pharmacological treatment, significantly modulate the expression and topography of PSD transcripts after acute administration in animal studies (de Bartolomeis et al., 2015, de Bartolomeis et al., 2013a, Fumagalli et al., 2008). However, a direct head-to-head comparison of the effects at the PSD of first and second-generation antipsychotics at different doses in a chronic paradigm is still lacking. To fill this gap, we evaluated the expression of key PSD transcripts, i.e. Homer1a, Arc, Homer1b, PSD-95, and Shank, after chronic administration of: i) the prototype first generation antipsychotic haloperidol, which has a relatively selective D2 receptor profile (Correll, 2010); ii) the prototype second generation antipsychotic olanzapine, which has a broad multi-receptor profile (Correll, 2010); and iii) the novel multi-receptor targeting antipsychotic asenapine, which holds a dopamine D1 and D2 receptor blockade ratio of approximately 1 (Shahid et al., 2009). In an acute paradigm of administration, increasing doses of these same antipsychotics have been reported to induce a progressive recruitment of cortical/sub-cortical regions (de Bartolomeis et al., 2015). However, molecular effects of antipsychotics may dramatically change according to the timing of administration (de Bartolomeis et al., 2016). Acute paradigms of antipsychotic administration may shed lights on adaptive and reactive synaptic mechanisms that follow an intense and punctual receptor perturbation. On the other hand, chronic antipsychotic paradigms may best recapitulate the synaptic changes occurring during or after prolonged treatments (Kontkanen et al., 2002), more closely mimicking real-world therapeutic approaches. Therefore, in the present study we tried to address the following questions:

• Does chronic administration of antipsychotics with different receptor profiles induce changes in the expression of PSD transcripts differentially with respect to cortical/sub-cortical regions?

• Do different doses of the same antipsychotic impact the topography of gene expression and therefore the recruitment of different cortical/sub-cortical circuitries?

• Do antipsychotics with different receptor profile or different doses of the same antipsychotic affect the relative ratio of transcript expression of the Homer1 isomers (i.e. Homer1a and Homer1b), which have been described to exert opposite molecular effects (Kammermeier, 2008)?

The PSD molecules studied herein were chosen based on the fact that they have all been implicated in schizophrenia pathophysiology (de Bartolomeis et al., 2014, Fromer et al., 2014, Purcell et al., 2014), as well as demonstrated to be responsive to antipsychotic treatment (de Bartolomeis et al., 2013a, Iasevoli et al., 2011, Iasevoli et al., 2010, Iasevoli et al., 2009). Moreover, all transcripts studied herein are reported to directly or indirectly interact with each other and mainly mediate postsynaptic type 5 metabotropic glutamate receptor (mGluR5)-dependent signaling (Bertaso et al., 2010, Sala et al., 2005, Tu et al., 1999, Zhang and Lisman, 2012).

The Homers1 are scaffolding and adaptor proteins associated with schizophrenia and depression-like behaviors and including constitutive forms (i.e. Homer1b/c) and an inducible early gene form (Homer1a) (de Bartolomeis and Iasevoli, 2003, Newell and Matosin, 2014, Serchov et al., 2015, Shiraishi-Yamaguchi and Furuichi, 2007, Wagner et al., 2015). PSD-95 is a membrane-associated guanylate kinase (MAGUK) scaffolding protein interacting with N-Methyl-d-Aspartate (NMDA) receptor NR2 subunit and shaker-type potassium channels (Chen et al., 2015). Shank is a PSD protein that has been associated to schizophrenia and autism spectrum disorders (Sala et al., 2015), and demonstrated to be instrumental in physically and functionally coupling of NMDA receptors and mGluR5 at the PSD (Hwang et al., 2005). Finally, Arc (activity-regulated cytoskeletal protein, also known as Arg-1) is a master regulator of synaptic plasticity, involved in long-term potentiation and both mGluR and NMDA receptor-dependent forms of long-term depression (Guzowski et al., 2000, Park et al., 2008).