The glial perspective of autism spectrum disorder convergent evidence from postmortem brain and PET studies

Autism spectrum disorder (ASD) constitute a heterogeneous group of pervasive developmental disorders with hallmarks of persistent social interaction deficits along with repetitive stereotyped behaviors (APA DMS-V, 2013). These symptoms occur in early childhood and coincide with brain developmental period. The incidence of ASD has rapidly increased to 1 in every 50 children in the United States (Blumberg et al., 2013), with similar prevalence in countries around the world. The lack of specific biomarkers couple with the benefits of early diagnosis requires a consistent and reliable pathogenesis which may facilitate diagnostic precisions and therapeutic opinions.

The physiological mechanism of ASD cannot be characterized through a sole focus on neurons, which sparks a boom of research interest in the investigation of extra-neuronal contributors. Evidence derived from genetic, epidemiological and clinical studies converges to support the involvement of glia induced neuroinflammation in the pathogenesis of ASD (Varghese et al., 2017). However, given the extreme pathological and phenotypic heterogeneity of ASD, all individual reports should be considered provisional and need to be replicated. The characterization of glia induced neuroinflammation in the pathogenesis of ASD requires an integrative platform to combine evidence derives from various research methods (Ecker & Murphy, 2014), with postmortem brain and molecular imaging emerge as two validated and encouraged paradigms.

Despite remarkable advances in the field of ASD research, a comprehensive picture of ASD pathogenesis remains to be drawn. Evidence converges to support a critical involvement of neuroinflammation in the pathogenesis of ASD, with glial cells lie at the very core of this physiological mechanism (Siniscalco et al., 2018).

Neuroinflammation refers to an inflammatory response in the central nervous system (CNS), which emerges as a prominent hallmark of diverse pathological conditions (Glass et al., 2010). Neuroinflammation constitutes a complex process among which multiple glial cells interact with each other to form an intricate inflammatory response (Yang & Zhou, 2019). The interactions of glial cells control the promotion and inhibition of inflammatory processes through the balance of pro- and anti-inflammatory cytokines, thus shape the two-sided properties of neuroinflammatory responses (Kwon & Koh, 2020). The precise roles of different glial cells in neuroinflammation depends on the type of inflammatory stimulus and the course of pathological conditions. The crosstalk between microglia and astroglia forms a critical platform which decides the initiation, maintenance or restriction of neuroinflammatory process, and impacts the genesis, course and severity of pathological condition (Matejuk & Ransohoff, 2020). Activated microglia act in concert with reactive astrocytes to foster a long-lasting and self-sustaining neuroinflammatory response, which exacerbates the production of inflammatory cytokines and mediators, ultimately causes functional and behavioral consequences relevant to ASD (Bjorklund et al., 2016). Neuroinflammation occurs not only at the brain parenchyma but also at the brain barriers whose mechanisms should be characterized both in brain and in periphery. The consequences of neuroinflammation result from an integrated responses of resident glial cells together with infiltrated peripheral immune cells. The complexity of pathways and events in neuroinflammation requires the integration of different approaches to fully illustrate the cellular and molecular mechanisms (Han et al., 2021).

In the search for potential neuroinflammatory mechanisms, the importance of glial cells has attracted considerable attention in recent years (Di Benedetto & Rupprecht, 2013). Given the significance of glial cells in the physiology and pathology, ASD pathogenesis shifts from an exclusively neuron-centric to an additionally glia-centric paradigm.

Glial cells comprise a class of extra-neuronal cells with diverse structural, functional and molecular characteristics (Zhang, 2001), including microglia, astrocytes, oligodendrocytes, and oligodendrocyte progenitor cells (OPCs). Microglia belong to the mononuclear phagocyte family, which originate from embryonic yolk sac progenitors rather than hematopoietic stem cells. Microglia populate in the parenchyma during embryonic development and persist into the first postnatal week (Bilbo et al., 2018, Reemst et al., 2016, Voineagu and Eapen, 2013), which comprise 10%∼15% of total brain cells and 5∼10% of total glial cells. Microglia continuously monitor the physiological milieu of parenchyma and constitute the first defense mechanism of the CNS, which act as the first responder and also a robust cytokine producer under inflammatory insults. Microglia persist throughout the entire life without any substantial input from circulation as a result of their long-lived nature and self-renewal ability (Yona et al., 2013, Goldmann et al., 2013). The activation of microglia has been simply equated with the existence of neuroinflammation for a long time, thus ignores the characterization of neuroinflammation go beyond the scope of microglial activation (Setiawan et al., 2015). Astrocytes, instead, arise from neural progenitor cells within the subventricular zone and populate throughout the parenchyma later than microglia during early postnatal period (Matta et al., 2019). Astrocytes represent the most abundant cell type in the brain, which constitute approximately 30% of total brain cells and comprise 20∼40% of total glial cells (Ge et al., 2012, Molofsky et al., 2014). Although microglia and astrocytes own their individual functions, they work as a unit through persistent crosstalk to amplify or diminish their individual or shared functions. Microglia and astrocytes function synchronously and complementarily in the process of neuroinflammation to drive the fine-tuned regulation and resolution of neuroinflammatory responses (Vainchtein & Molofsky, 2020). The activated microglia induce the reactive of astrocytes and control the fate of astrocytes, the reactive astrocytes in turn amplify the activation of microglia and regulate the function of microglia. Microglia participate in the first line of defense and orchestrate the process of neuroinflammation, astrocytes receive inflammation signals from microglia and further the release of inflammatory stimuli, thereby initiate the reactive loop and exacerbate the inflammatory response (Adams and Gallo, 2018, Ransohoff and Perry, 2009). The appearance of glial cells in the postnatal brain coincides with the dynamic period of synaptic refinement. The fine processes of microglia and astroglia closely contact with synaptic boutons and spines (Bushong et al., 2002), actively participate in structurally shaping and functionally modulating synaptic connectivity (Chung et al., 2013, Stevens et al., 2007, Hong et al., 2016, Bellesi et al., 2017).

Microglia and astrocytes are populated in parenchyma from embryonic stage onward, they are positioned to carry out critical roles in many aspects of subsequent neurodevelopment. Both microglia and astrocytes may, conceivably, act as a critical link in the characterization of disease pathogenesis itself but also to the identification of therapeutic opinion of ASD (Lai et al., 2014).

Although capture some behavioral features and pathological mechanism of ASD, animal models so far can neither reflect the phenotypic heterogeneity of ASD nor represent the core pathology of this disorder, interpretations about the neurobiological basis of ASD based on animal models should therefore be made with cautions. The investigation of ASD neuropathology mainly takes one of two parallel paths, either a microscopic approach with postmortem brain or a macroscopic approach with molecular imaging.

Postmortem brain analysis takes the advantages of characterize neuroinflammation at the molecular and cellular levels from postmortem brain tissues (Matta et al., 2019), Positron-emission tomography (PET) imaging enables the non-invasive visualization of neuroinflammation at the molecular and cellular levels from vivo subjects (Meyer et al., 2020). Postmortem brain analysis can characterize ASD pathological changes across cell classes and molecular pathways, which can serve as a complementary approach to address the limited spatial resolution of molecular imaging methods. PET imaging studies can link ASD symptom profiles to brain anatomy and function, which can act as a supplementary method to address the paucity of postmortem human tissues. The characterization of neurobiological changes and the identification of diagnostic biomarker for ASD requires the collaboration between different investigational approaches. Postmortem brain analysis and PET imaging studies can be of high relevance as they both reveal the molecular and cellular basis of ASD neuropathology, and evidence in these two fields have not been systematically integrated in the context of ASD. The presence of considerable heterogeneity between studies in terms of sample characteristics and methodological approaches limits a consistent elaboration of glia induced neuroinflammation in the pathogenesis of ASD. Although not often performed in the context of preclinical studies, the meta-analysis provides a powerful method to synthesize data on a specific topic.

The present study used the meta-analysis paradigm to appraise the nature, magnitude, and consistency of glia induced neuroinflammation in the pathogenesis of ASD. The present study systematically reviewed human data derives from both postmortem brain and PET studies, meta-analyzed the involvement of glia induced neuroinflammation in the pathogenesis of ASD, discussed the implications of these findings in relation to physiological mechanisms and therapeutic opportunities.

The present review was performed in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA). Two authors independently performed the literature search, study selection and data extraction. Discrepancies produced in these processes was resolved through robust discussions among all authors.

A systematic search for literature indexed in the entire PubMed, EMBASE, PsycINFO and Web of Science was performed with the following search terms:

For postmortem studies: (Autism Spectrum Disorder OR Autistic Disorder OR Autism OR ASD) AND (microglia OR astrocytes OR glia OR neuroglia OR Ionized calcium-binding adaptor molecule 1 OR IBA-1 OR glial fibrillary acidic protein OR GFAP) AND (post mortem OR postmortem OR post-mortem OR autopsy).

For PET studies: (Autism Spectrum Disorder OR Autistic Disorder OR Autism OR ASD) AND (Positron Emission Tomography OR PET OR Single Photon Emission Tomography OR SPET OR Single Photon Emission Computed Tomography OR SPECT).

All search terms were searched in a combination of the search fields “title”, “abstract,” and “MeSH” when available. No other limitations were applied to ensure the best possible retrieval and inclusive. Reference lists of examined papers, reviews and meta-analyses were cross-checked for additional relevant publications. No separate sources for gray materials were covered in the search because the inclusion criteria for this review was limited to peer-reviewed studies.

Only peer-reviewed primary research articles were considered as eligible studies.

Inclusion criteria were set as: included a group of ASD subjects compared with matched controls; used postmortem brain samples or adopted PET methods; reported glia related parameters, including number or density of glial cells, cell size or cell shape of glial cells, mRNA expression or protein expression of glial markers. Exclusion criteria were set as: animal studies; review articles; meta-analyzes; case reports; conference abstracts; editorials/letters.

This study did not include studies that only measured markers associated with the M1/M2 phenotype of microglia or the A1/A2 phenotype of astrocyte but did not localize these markers specifically to microglia or astrocyte. This study did not include studies that specifically used high-throughput techniques either microarray or proteomics examined the gene expression or protein expression of glial makers due to their bias towards positive results for particular markers. Due to the relatively limited number of studies yielded from the literature search, the present study did not set selection criteria for confounds in terms of sample heterogeneity, psychiatric comorbidity or medicine history. The inclusive nature of this review was necessary given the relative paucity of literature in order to decipher the complex nature of glia induced neuroinflammation in the pathogenesis of ASD.

A standardized data extraction form was used to extract data from each study.

The information extracted from these included studies can divide into three categories in terms of study information (including first author and publication year), sample information (including sample sizes, mean age, male subjects, medication use, and complication) and method information (including examined brain regions, glial maker, analysis technique, and postmortem interval (PMI) for postmortem studies as well as ligands and analysis method for PET studies).

When multiple studies used an overlap or identical sample, the study contained a larger sample size or reported the most comprehensive data was included. Where data in studies were incomplete or unclear, the authors were contacted to provide additional details. Where data in studies was presented graphically rather than numerically, data were approximated from graphs using WebPlotDigitizer. Brain areas investigated in different studies were grouped into larger categories reflected major brain structures.

留言 (0)

沒有登入
gif