ETV5 is a Putative Differentially Expressed Transcription Factor of Interleukin 1β Under Neuroinflammatory Conditions in Murine Cerebellum

The inflammation of brain and spinal cord, together constituting our central nervous system (CNS), is known as neuroinflammation. Being implicated in a variety of somatic and neurological illnesses, the most common cues of its initiation known till date are toxicity to various metabolites, traumatic brain and spinal cord injuries, viral and/or bacterial toxicity, and autoimmune mediated neuroinflammation (Lyman et al., 2014; Feinstein et al., 2016). Apart from being central to the mutual pathology of a number of acute and chronic neurological disorders, neuroinflammation has also been found to be a major cause and aggravating factor for the development of neurodegenerative diseases. Moreover, its role in the early development of some chronic conditions, e.g., dementia, has also been reported (Lyman et al., 2014; Kwon and Koh, 2020; DiSabato et al., 2016).

Being pleiotropic mediators and effectors of cells’ and tissues’ inflammatory processes and of various autoimmune disorders, interleukin-1 α (IL-1α) and interleukin-1 β (IL-1β) are archetypal proinflammatory cytokines. Of these, IL-1β also performs crucial homeostatic functions in a healthy organism. These include the regulation of body temperature, sleep, and feeding (Dinarello, 1996, 2004). Consequently, its overproduction is associated with various pathophysiological anomalies such as inflammatory bowel disease, rheumatoid arthritis, osteoarthritis, vascular diseases, neuropathy, Alzheimer’s disease and multiple sclerosis (Braddock and Quinn, 2004).

Some of the common documented events during the process of neuroinflammation include: Increased activation of brain’s resident immune cells (glial cells), altered permeability of blood brain barrier (BBB), leukocyte invasion, and increased concentration of the pro-inflammatory cytokines (Rejo et al., 2008; Guzyman-Martinez, 2019). IL-1 being upregulated in several of the neurodegenerative disorders such as Parkinson’s disease (PD) (Leal et al., 2013; Stojakovic et al., 2017), Alzheimer’s disease (Cacabelos et al., 1991, 1994), multiple sclerosis (Lin and Edelson, 2017) and amyotrophic lateral sclerosis (Meissner et al., 2010), also makes IL-1β a key player in the process of neuroinflammation too, along with other inflammatory processes of the body. The pathways activated by overproduction of IL-1β include NFkB-dependent transcription of some other pro-inflammatory cytokines such as TNF-α, interferons and IL-6; activation of neutrophil recruiting chemokines; and stimulation of synaptophysin and Tau proteins’ production in neurons (Acuner-Ozbabacan et al., 2014; Fan et al., 2007; Pyrillou et al., 2020).

The regulatory elements of IL-1B include a TATA box within the promoter, a CRE (cAMP responsive element), an AP-1 (activator protein 1) promoter site, an SP-1/PU.1 (the ETS family of transcription factor) binding site, and an NF-κB (nuclear factor kappa B) binding site (Tjitro et al., 2018; Perez et al., 1999; Lahiri et al., 2002; Shirakawa et al., 1993). The induction of IL-1B gene expression can be brought about via various stimuli including both microbial and non-microbial types. The initial translation product is a propeptide (31 kD), the pro-IL-1β, which is cleaved to the mature and biologically active form of IL-1β by the action of cysteine protease caspase-1 (Zepter et al., 1997; Herzog et al., 2005; Eder, 2009).

The presence of both neuroinflammation and tissue damage in post mortem brain tissues makes it difficult to determine whether upregulation of IL-1B is caused by, or a consequence of neuroinflammation (Hewett et al., 2012). This also urged the need to explore IL-1B along with its potential mediators to get a better understanding of its precise role in the process of neuroinflammation. This study was therefore aimed at identification of the transcription factors (TF) specific to IL-1B promoter/enhancer region and then study their differential expression under neuroinflammatory conditions using in vivo LPS mouse models.

Various online available databases were utilized for the identification of the promoter region of Il-1B gene. These include GeneCards (https://www.genecards.org/), the eukaryotic promoter database (https://epd.epfl.ch//index.php), and super-enhancer database (http://www.licpathway.net/sedb/). These databases also identified the TFs having the binding sites for the promoter of Il-1B and among them, only those TFs were preferred having the highest scores from the online databases. The resulting TFs were further screened on the basis of their tissue specific high or low expression profiles so as to select only those relevant to the tissue being used in the study i.e., brain.

Male and female BALB/c mice, approximately 6 weeks of age, were housed in groups of 4 in cages containing rodent bedding. Food and water were available ad libitum and temperature was maintained at 25°C. All experiments were carried out on mice divided into two groups, experimental drug group and vehicle control group receiving physiological saline injections only.

BALB/c mice 6-8 weeks of age were administered with a single dose of 5 mg/kg of lipopolysaccharide (LPS) solution (Sigma Cat # L2880-25MG) dissolved in 0.9% saline solution via intraperitoneal route. The controls received injections of saline solution only. The sacrifice of animals was carried out after 24 h of LPS treatment.

The mice were sacrificed by cervical dislocation. Briefly, the backside of a knife was placed on the dorsal side of the neck at the junction between skull and the vertebrae. Then a little pressure was applied through the knife while pulling the animal from its tail so as to dislocate the head from the rest of the body in a single jerk to minimize animal discomfort. The bioethical clearance certificate was obtained from the Bioethics Committee of the School of Biological Sciences, University of the Punjab.

The brains were further dissected to get two regions i.e., hippocampus and cerebellum. These two regions are shown in the Supplementary Figure S1 which represents the regional map of murine brain. The protocol for dissection of these regions was followed from Li (2011). The cerebellum is located on the dorsal side of the brain and on the lower side as can be seen in the Figure 1. First, the portion of the brain containing the cerebellum was separated from the cerebral cortices by making a vertical cut. Then the tissue underneath the cerebellum consisting of midbrain, pons and medulla was separated from it very carefully with the help of fine curved forceps.

The dissection of hippocampi is a bit difficult than the cerebellum as they are located on the inside of the cerebral cortices and are covered with tissue. Therefore, the dissection procedure of hippocampus is explained a little more as in the Figure 1B, C since the crescent shaped hippocampi are placed on the dorsal side of the brain covered with a fatty tissue the removal of which makes the hippocampus visible.

For total cell RNA extraction, the tissues were homogenized in TRIzol reagent via trituration and total RNA was extracted according to the manufacturer’s protocol. The RNA was used for complementary cDNA synthesis after DNase treatment using Revert Aid First Strand cDNA synthesis kit (Thermofisher Scientific, Cat # K1622).

Extracted RNA was quantified using pico200 spectrophotometer at A260/280. For DNase treatment of RNA sample, Invitrogen kit (Cat # 18068015) was used for this procedure as per instructions of the kit. The cDNA was synthesized using Maxima H Minus First Strand cDNA Synthesis Kit (Cat # K1681).

The quantitative expression analysis of genes was done by real time PCR analysis. Maxima SYBR Green/ROX qPCR master mix (2X; Thermofisher Scientific, Cat # K0222), cDNA, and 5 µM forward and reverse primers’ mix were used for RT-PCR (the sequences of the primers are provided in Table I. Two-step thermal cycle protocol with optimized annealing temperatures (Table I) of each primer pair was used for amplification and quantification of amplicons on PikoReal Real-Time PCR system (Thermofischer Scientific). Melting curve data with every 0.2°C increment was taken to determine the uniformity of PCR product. HMBS was used as a reference gene in the study. Significance analysis for fold differences of biological replicates was done using t-test for means.

The genes, for which mRNA expression was significantly differential, were further cross examined with respect to their protein expression at basal level using Expression Atlas database (Expression Atlas, 2020). This database is maintained by the European Bioinformatics Institute that is a hub of information for RNA and protein expression for different studies, organisms, and experimental conditions. Since mRNA of Etv5 was being differentially expressed in the cerebellum of both the genders, its basal level protein was checked using this database.

For studying the functional protein association networks of Etv5, STRINGS (Search Tool for Retrieval of Interacting Genes and protein, Version 11.0, https://string-db.org/), BioGRID (https://thebiogrid.org/), Database of Interacting Proteins (DIP, https://dip.doe-mbi.ucla.edu/dip/Main.cgi) and IntAct (https://www.ebi.ac.uk/intact/home) databases were used (Szklarczyk et al., 2018) to find not only the established biological interacting partners from experiments and curated databases but also the predicted interactions from gene neighborhood, fusions, co-occurrence, co-expression, homology and text-mining techniques.

The re-analysis of RNA-seq datasets taken from Gene Expression Omnibus (GEO) database was done (Khan et al., 2022). In short, fastq files of LPS versus control treated mouse brain (accession number GSE153369), neurons, astrocytes and microglia (accession number GSE75246) were analysed using different packages from R Bioconductor such as limma, edgeR, gplots, org.Hs.eg.db, RcolorBrewer and Glimma (Warnes et al., 2005; McCarthy et al., 2012; Ritchie et al., 2015; Su et al., 2017). and the average expression along their p values were noted for significance analyses.

The online search resulted in the identification of a number of TFs interacting with IL-1B gene regulatory elements. Further sorting of these TFs on the basis of their differential expression in different tissues resulted in the identification of Klf16 and Etv5 exhibiting the highest expression in brain as compared to other tissues (Supplementary Table SI).

mRNA expression analysis revealed that Klf16 did not show any significant difference in expression in both the brain regions of male and female mice (Fig. 2). However, in case of Etv5, significant (p < 0.005) upregulation of 1.25 folds was observed only in male hippocampus (Fig. 3A). Quite interestingly, its expression was found to be downregulated in male cerebellum (Fig. 3B) as well as in female hippocampus (Fig. 3C) and cerebellum (Fig. 3D), although these fold differences were found to be non-significant.

After getting significant expression difference in Etv5 transcript, we further searched for the interacting protein partners of Etv5 of both human and mouse origin of which the mRNA expression of Cop1, GDNF, Det1, Bcl6b and Dusp6 was further analyzed using RT-PCR. The protein partners for mouse and human Etv5 protein from STRING database are given in Figure 4 in the form of node diagrams. Supplementary Table SII shows the interacting protein partners of Etv5 from all the databases for mouse and human, respectively.

In line with the significant upregulation of Etv5 gene, three of its interacting partners, i.e., Cop1, Det1 and Bcl6b showed significant downregulation, while non-significant upregulation was observed for GDNF and Dusp6 in male hippocampus. In male cerebellum, significant downregulation of Bcl6b and Dusp6 was seen. In female mice, only Bcl6b was found to be showing significant downregulation in both the hippocampus and cerebellum.

When the protein expression of Etv5 was checked as a validation to see its baseline conditions under normal conditions via Human Protein Atlas project (Expression Atlas, 2020), it was found that in the cerebellum of both male and female subjects, its expression was moderate (Fig. 5).

When the expression of Etv5 and its selected interacting partners was checked from RNA-seq data sets for brain tissue, neurons, microglia and astrocytes, only Det1 significant difference in average expression with a value of -1.729 (p < 0.005) (Table II).

Table II. Average expression of Etv5 from LPS treated neuroinflammation RNA-seq datasets. * (p < 0.05), ** (p < 0.005), NS (p > 0.05 = non-significant).

This study was focused on the identification of IL-1B specific transcription factors of brain as IL-1β serves a pivotal role in the mediation of inflammatory responses. Consistent with the significant differential expression of Etv5 mRNA only in male hippocampus, most of its interacting partners (Cop1, Det1 and Bcl6b) also showed significant differential expression (downregulation) only in male hippocampus. An interesting difference to note was that while Etv5 was being upregulated, the interacting partners were being downregulated in male hippocampi. In the male cerebellum, the only genes significantly differentially expressed were Bcl6b and Dusp6 (downregulation), and in female hippocampi and cerebellum only Bcl6b showed significant downregulation. Further exploration of the genes under study from the RNA-seq data sets of LPS treated brain tissue as well as from neurons, microglia and astrocytes revealed significant downregulation of Det1 in astrocytes (Table II). Among all the genes, Bcl6b was found to be common among the two brain regions in both the genders, all exhibiting downregulation of it. Also, the downregulation of Det1 in male hippocampus was corroborated with its down regulation in astrocytic RNA-seq data.

Being a member of the PEA3 family of TFs, Etv5 binds to a conserved DNA motif having sequence GGAA/T (Oh et al., 2010). Some of the important known roles of Etv5 include: Coordination of limb development; regulation of epithelial-to-mesenchymal transition in some cancer types; control of gene expression in spermatogonial stem cells; inhibition of sonic hedgehog gene expression; and positive regulation of some miRNAs important in spermatogonial stem cells renewal (Zhang et al., 2009; 2010; Wu et al., 2011; Niu et al., 2011). Besides these, although still poorly understood, the roles of Etv5 in immune responses include an increased interferon gamma (IFN-γ) production from TH1 cells (Thieu et al., 2008; Ouyang et al., 1999), increased IL-17 production in TH17 cells (thus playing a role in allergic inflammation development) (Pham et al., 2014), and suppression of the activation of adipose tissue resident macrophages (Hu et al., 2021). Since activation of these macrophages contributes in the development of chronic inflammation, and Etv5 being identified as a major suppressor of the process, the upregulation of Etv5 transcript in male hippocampus in the current study presents an interesting target for future studies of Etv5 on brain resident macrophages i.e., microglia. Corroborated with the previous studies indicating a regulatory role of Etv5 on transcriptional activation of IL-10, IL-17 and IL-9 expression (Pham et al., 2014; Koh et al., 2016, 2017), the precise mechanism of its role in macrophage activation regulation is yet to be unveiled.

We observed a downregulation of Cop1 gene in hippocampi of both the genders and in male cerebellum although significant results were those of male hippocampus only. Previous studies have found a neuroprotective role of Cop1 in neurodegenerative disorders (Gorantla et al., 2007, 2008). Corroborating the neuroprotective role, recent knockout studies of Cop1 have shown an increased pro-inflammatory response in brain resident macrophages, i.e., microglial cells (Ndojaa et al., 2020). Since Det1 is a partner of Cop1 E3 ubiquitin ligase complex, its downregulation is in concert with Cop1 downregulation (which has been found to be almost double in magnitude to that of Det1 (Fig. 3B). Previous roles of Det1 have been found in the mediation of cellular differentiation and proliferation in multicellular organisms (Nassrallah et al., 2018), however, no known roles could be found of Det1 in the process of neuroinflammation. Similarly, although previous roles of Bcl6b and Dusp6 have been found in the regulation of inflammatory responses (Bunting and Melnick, 2013) and endothelial inflammation (Hsu et al., 2018), respectively, their precise roles in the process of neuroinflammation are yet to be studied. These findings present important genes for further detailed studies with respect to the modulation of neuroinflammatory process in hippocampus tissue. Also, the brain region along with gender differences are also evident from the current study highlighting the need to focus on individual brain regions of each gender.

In the current study, we have presented the expression profiles IL-1B associated transcription factor Etv5 along with some of its interacting partners as molecular targets of neuroinflammatory process. Also, the current study presents the dissection of these TFs with respect to not only the two important brain regions hippocampus and cerebellum due to their functional implications in memory and motor function, respectively, but also with respect to gender differences. Moreover, the identification of significantly differentially expressed genes present interesting target genes for further studies on neuroinflammation. Furthermore, future research on the extensive molecular characterization of all brain regions as well as individual cell types could provide a complete landscape not only for the recognition of each of these brain regions and cells in the process of neuroinflammation, but also would guide the development of more targeted therapeutic interventions.

Funds for this study were provided by the University of the Punjab.

The study was approved by the Institutional Review Board (IRB) and Institutional Animal Care and Use Committee (IACUC) of University of the Punjab

All the data has been included in this article.

There is supplementary material associated with this article. Access the material online at: https://dx.doi.org/10.17582/journal.pjz/20220612100622

The authors have declared no conflict of interest.