Toward a better understanding of inflammatory microvesicles in alcohol use disorder

We have read with great interest the study by Crews et al. (2021) who investigated the effects of ethanol on pro-inflammatory and microglial functional gene expression using primary organotypic brain slice cultures (OBSCs). The authors reported that ethanol caused the release of inflammation-promoting microvesicles (MVs), and that blockade of MV release by imipramine and other methods prevented ethanol-induced pro-inflammatory signaling (Crews et al., 2021). These findings are important given that limited information is known about the effects of ethanol on CNS-derived extracellular vesicles (EVs), such as MVs, and that the molecular cargo of central nervous system (CNS)-derived EVs, as well as the quantity, have been used as biomarkers for traumatic brain injury, Alzheimer’s disease, Parkinson’s disease, HIV-associated neurocognitive disorders, and methamphetamine use disorder (Badhwar & Haqqani, 2020; Dagur et al., 2020; Ohmichi et al., 2019; Pulliam et al., 2019; Sandau et al., 2020). It is now well established that the composition of a cell’s EVs can report on its underlying physiological state and gene/protein expression pattern. To standardize and facilitate this work, The International Society for Extracellular Vesicles published an updated position paper in 2018 (Minimal Information for Studies of Extracellular Vesicles: MISEV2018), which recommends experimental and reporting requirements pertinent to EV research that are intended to standardize the field and improve reproducibility through added rigor (Lötvall et al., 2014; Théry et al., 2018; Witwer et al., 2017). The guidelines are grouped into six subcategories and include (a) nomenclature of the EVs utilized in the study, (b) collection and preprocessing of tissue culture medium that EVs are isolated from, (c) experimental details related to EV separation and concentration, (d) EV characterization, (e) functional studies, and (f) reporting (Théry et al., 2018).

In accordance with MISEV2018 (Théry et al., 2018), Crews et al. include experiments that are recommended for functional studies, including (a) secreted MVs isolated from OBSCs treated with nonconditioned ‘Control’ medium (CON-MV), (b) MV-depleted supernatant, and (c) activity of isolated MVs versus exosomes. More specifically, in CON-MVs and MV-depleted supernatant, there was attenuated expression of the pro-inflammatory genes TNF-α and Il-1β, compared to MVs isolated from the medium of ethanol-treated OBSCs. They also demonstrate that the cargo of pelleted exosomes (100,000× g) isolated from the medium of ethanol-treated microglial had reduced levels of high mobility group box 1 (HMGB1), a pro-inflammatory danger-associated molecular patterns protein, compared to pelleted MVs (21,000× g; Crews et al., 2021). Together, these controls demonstrate that the functional activity of their EV preparations in part resides in their 21,000× g MV pellet. However, the functional activity experiments would benefit from a dose–response assessment and methodology used to control for the amount of EVs transferred to OBSCs. Standardizing the amount of EVs transferred, either by EV particle number or protein concentration, negates confounds associated with treatments that effect EV secretion (e.g., ethanol) or potential inconstancies in the efficiency of EV isolation protocols.

Crews et al. define their EV preparations as MVs or exosomes, based on pellets obtained from 21,000× g and 100,000× g ultracentrifugation spins, respectively (Crews et al., 2021). MISEV2018 recommends that MV and exosome nomenclature is reserved for preparations where the route of biogenesis is definitively associated with plasma membrane budding or the late endosome, respectively (Théry et al., 2018). In regard to the EVs isolated in the 100,000× g pellet, evidence that these vesicles originate from the late endosome is lacking (Crews et al., 2021). However, evidence that the 21,000× g pellet contains MVs is supported by reduced EV numbers in imipramine-treated OBSCs (Crews et al., 2021), which is reported to block MV biogenesis through inhibition of acid sphingomyelinase (Arenz, 2010). As acknowledged by Crews et al., imipramine has a number of other effects including, but not limited to, blocking dopamine D2 receptors and inhibiting serotonin reuptake, which have both been associated with modulating neuroinflammation (Dionisie et al., 2020; Śmiałowski, 1991; Tatsumi et al., 1997; Xia et al., 2019). Furthermore, imipramine treatment may indirectly effect exosome biogenesis considering that serotonin has also been shown to stimulate the secretion of exosomes from cultured BV-2 microglia, via cAMP and Ca(2+)-dependent signaling pathways (Glebov et al., 2015). To better understand the role of inflammatory EVs in alcohol use disorder, future experiments that track alterations in MV and exosome biogenesis following either ethanol or imipramine treatment are needed to demonstrate mechanism(s) of action.

For future studies aimed to investigate the role of extracellular MVs in pro-inflammatory responses to ethanol, we propose that experiments include a more rigorous characterization of EV preparations. Limiting characterization to nanoparticle tracking analysis provides size and concentration measurements for all isolated particles including MVs, exosomes, and apoptotic bodies, but lacks the capability to definitively differentiate EV subtypes. Further characterization of EV preparations by electron microscopy and immunoblots would provide evidence that the functional activity is attributed to MVs and not contaminants that are inherent to heterogeneous culture assays and crude EV preparation methods (ultracentrifugation). More specifically, electron microscopy would confirm the presence of intact isolated vesicles that are consistent with the size of MVs and exosomes and devoid of contaminating proteins, lipoproteins, and larger apoptotic bodies (Nielsen et al., 2018). Also, immunoblots using protein extracts from the resuspended pellets would confirm enrichment for markers of MVs (Annexin A1) versus classical exosomes (CD9, CD63, and CD81) in the 21,000× g and 100,000× g pellets, respectively (Jeppesen et al., 2019). Immunoblots are also used to demonstrate a lack of contaminants associated with other intracellular compartments (e.g., Golgi apparatus), secretory pathways not associated with MV or exosome biogenesis, or the culture conditions (Théry et al., 2018). In regard to the culture conditions, the use of EV-depleted serum eliminates confounds associated with contaminating serum EVs (Beninson & Fleshner, 2015; Shelke et al., 2014). Alternatively, Crews et al. control for the potential effect of contaminating serum EVs by demonstrating that MV-depleted supernatant does not have a pro-inflammatory effect (Crews et al., 2021).

Finally, Crews et al. aptly highlight that “pro-inflammatory signaling in brain is a key component of neuropathology in alcohol use disorder,” but missed an opportunity to better understand sex differences in ethanol-induced inflammation, as slices from both sexes were pooled together for the experiments. It was recently reported that female and male mice differ in regard to ethanol’s effects on inflammatory-related plasma miRNAs EV profiles, with evidence suggesting that females are more vulnerable than males to the inflammatory effects of ethanol exposure (Ibáñez et al., 2020). Other studies, not specific to EVs or their cargo, similarly find differences between females and males in the inflammatory effects of ethanol on brain and behavior (e.g., Chang et al., 2018; Lowe et al., 2020; Pascual et al., 2017; Wilhelm et al., 2017). In particular, one study in humans found that alcohol (ethanol)-induced inflammatory changes were more prominent in females who had increased levels of danger-associated molecules, such as HMGB1 (Orio et al., 2018), providing not only evidence that there are sex-related differences in the response to ethanol’s inflammatory effects, but also suggesting that some of the conclusions made by Crews et al. (e.g., regarding HMGB1) may be confounded by the use of pooled slices from female and male mice.

Despite these methodological limitations, the research by Crews et al. provides valuable new insights regarding the signaling pathways associated with ethanol-induced neuroinflammation. This contribution helps move the field forward and closer to understanding mechanisms of immune activation in alcohol use disorder.