Profiling the microRNA signature of the peripheral sensory ganglia in experimental autoimmune encephalomyelitis (EAE)
Multiple sclerosis is an autoimmune disease with a distinct female bias, as well as a high prevalence of neuropathic pain in both sexes. The dorsal root ganglia (DRG) contain the primary sensory neurons that give rise to pain, and damage to these neurons may lead to neuropathic pain. Here, we investigate the sex differences of the DRG transcriptome in a mouse model of MS.
Next-generation sequencing was used to establish RNA and microRNA profiles from the DRG of mice with MOG35–55-induced EAE, a model of CNS inflammation that mimics aspects of MS. Differential expression and multiple meta-analytic approaches were used to compare expression profiles in immunized female and male mice. Differential expression of relevant genes and microRNAs were confirmed by qPCR.
Three thousand five hundred twenty genes and 29 microRNAs were differentially expressed in the DRG of female mice with MOG35–55-EAE, while only 189 genes and 3 microRNAs were differentially expressed in males with MOG35–55-EAE. Genes related to the immune system were uniquely regulated in immunized female mice. Direct comparison of sex within disease indicates significant differences in interferon and phagosomal pathways between the sexes. miR-21a-5p is the primary dysregulated microRNA in both sexes, with females having additional dysregulated microRNAs, including miR-122-5p.
This study provides evidence that females are uniquely affected by MOG35–55-EAE and that this difference may result from additional signaling not present in the male. The altered transcriptome of females correlates with other studies finding hyperactivity of pain-sensing neurons and suggests underlying sex-specific pathways for neuropathic pain.
KeywordsMicro-RNA Pain Inflammation Multiple sclerosis DRG
Complete Freund’s adjuvant
Central nervous system
Differential expression of genes
Dorsal root ganglia
Experimental autoimmune encephalomyelitis
Myelin oligodendrocyte glyocoproteinn
Peripheral nervous system
Multiple sclerosis (MS) is an autoimmune disease of the central nervous system, characterized by demyelination, disturbances in neuronal function, and progressive neurodegeneration . Experimental autoimmune encephalomyelitis (EAE) is a mouse model used to study the pathophysiology of MS, where an autoimmune response is artificially induced against the myelin of the central nervous system (CNS). The model produces immune-related demyelination of the CNS [2, 3, 4]. People with MS experience altered motor function but recent estimates report that 63% of people living with MS are additionally negatively impacted by disturbances to sensation such as chronic, neuropathic pain . Neuropathic pain arises from damage to the neurons involved in transmitting the “pain” signal to the brain. This damage can be localized to any part comprising the pain system, from the pain-sensing periphery to the pain-perceiving central nervous system .
The dorsal root ganglia (DRG) residing in the peripheral nervous system (PNS) contain primary sensory neurons including pain-sensing neurons. The precise contribution of DRG neurons and the mechanisms mediating neuropathic pain in MS and EAE remain elusive. While transcriptome dysregulation has been studied in the CNS and immune cell populations in the context of EAE [7, 8, 9], less research has been done on the PNS, a potential source of neuropathic pain in the disease. Recently, our laboratory has described differences in the development of pain states between female and male mice in the EAE model. These differences include differences in immune and neurodegenerative histochemical and oxidative stress profiles [2, 4]. These sex differences extend beyond the CNS, affecting the DRG . Knowing that sex differences exist in EAE and that there are functional alterations correlated to pain, we sought to describe the transcriptional profile of the DRG in EAE from female and male mice with the disease.
MicroRNAs (miRNAs) are ~ 22 nucleotide long, single-stranded RNA molecules that repress translation of RNA species with complementary sequences . miRNAs have risen to the forefront of research into the regulation of the transcriptome in pathophysiological states. It is estimated that the majority of all transcriptionally related pathways are under miRNA regulation . Increasing amounts of evidence suggest that “miRNA-signatures” in various tissues can be correlated with specific disease states. As miRNAs can function as one-to-many vectors by regulating the expression of multiple genes, we were also interested to see if a “miRNA-signature” existed at the level of the DRG in addition to a functional analysis of the “transcriptome signature.” This information could yield insights into the mechanisms of pain in EAE and MS more generally.
Here, we identify EAE-related miRNAs that are differentially expressed in both male and female DRGs. We compare the two sex-specific signatures at the transcriptional level for mice with EAE. We describe the analysis of RNASeq and miRSeq datasets, and functional clustering of the differentially expressed genes and miRNAs into disease-relevant signatures. Our findings highlight extensive sex differences that must be considered with animal models of disease and identify potential targets for pain-modifying therapies in the disease.
Female and male 6–8-week-old C57BL/6 mice were acquired from Charles River Laboratories, Canada, and habituated for 2 weeks in the housing facility with baseline handling and behavioral testing. After this period, MOG35–55-EAE, (hereon referred to as “EAE”) was induced by 50 μg subcutaneous injections of myelin oligodendrocyte glycoprotein (sp) (MOG35–55; Peptide Synthesis Facility, University of Calgary) emulsified in complete Freund’s adjuvant (CFA) at a concentration of 1.5 mg/ml. An additional intraperitoneal injection of 300 ng of pertussis toxin in 0.2 ml saline was given on the day of induction as well as 48 h later. Control (termed CFA hereon after) mice received identical injection protocols except for the lack of emulsified MOG. CFA is known to produce transient alterations in immune activity  but is still an effective tool for EAE induction and an appropriate control. Mice were assessed daily for clinical progression of EAE following a four-point scale: grade 0, normal mouse; grade 1, flaccid tail (disease onset); grade 2, mild hindlimb weakness with quick righting reflex; grade 3, severe hindlimb weakness with slow righting reflex; grade 4, hindlimb paralysis in one hindlimb or both. Tissue was extracted on the first day a mouse presented with clinical signs (see below) that is denoted as “onset.”
Tissue collection and total RNA extraction
Animals were euthanized by Euthanyl® (sodium pentobarbital) overdose injected intrapertioneally. After injection, animals were monitored for level of consciousness and dissections did not proceed until no response to toe pinch or corneal contact was observed. Cardiac punctures were performed to confirm euthanization and animals were perfused with 10 mL of saline. Whole DRGs were immediately extracted and suspended in Qiazol (Qiagen Biosystems) and stored at − 80 °C for later processing. Upon collection of the full tissue set, total RNA was extracted using miRNeasy® Mini kits (Qiagen Biosystems).
Total RNA was supplied to ArrayStar Inc. for sequencing. Total RNA was enriched by oligoDT magnetic beads, and library preparation was completed using KAPA Stranded RNA-Seq Library Prep Kit. Transcriptome data was acquired by sequencing on an Illumina HiSeq 4000 machine according to manufacturer’s instructions. MicroRNA data was acquired by 3′-adapter ligation with T4 RNA ligase 2 (truncated), 5′-adapter ligation with T4 RNA ligase, cDNA synthesis with RT primer, PCR amplification, and extraction and purification of ~ 130–150 bp PCR amplified fragments and sequencing on an Illumina HiSeq 2000 machine according to manufacturer’s instructions.
Analysis of differential expression
Both RNASeq and miRSeq differential expression was investigated through the R  package DESeq2. Briefly, raw transcript sequences from NGS were aligned to the UCSC “mm10” database of known genes using the Genomic Features package function “exonsBy.” miR read counts were directly uploaded into R. The resulting data structures were processed through the DESeq2 package following the author’s instructions found in their published vignette [14, 15, 16]. The results of each contrasting analysis (e.g., M EAE vs. M CFA) were published in a tabular format for usage in subsequent analyses. Genes were considered significantly differentially expressed using the threshold of q < 0.1 after Benjamini-Hochberg correction. Clustergram and PCA analyses were performed using MATLAB (MATLAB and Statistics Toolbox Release 2017b, The MathWorks, Inc., Natick, MA, USA).
Ontological analysis with GOrilla/REViGo
Single gene lists ranked by p-adjusted values were analyzed with the GOrilla  web software, using a p value threshold of 10E−9. Significant GO terms from each category (process, function, component) were then processed through REViGO  to screen for redundant terminology.
Functional analysis with ingenuity pathway analysis
Activation patterns of canonical pathways were analyzed through the use of IPA (QIAGEN Inc., https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis) . Briefly, gene lists ranked by p-adjusted values were analyzed using the “Core Analysis” feature using default parameters and a p-adjusted cutoff of 0.1.
For validation of DEmiRs, 50 ng of total RNA was reverse transcribed using the miScript II RT Kit (Qiagen Biosystems). miRs were detected with the Mm_miR-21_2 and Mm_miR-122a_1 miScript Primer Assays (Qiagen Biosystems) using the miScript SYBR Green PCR Kit (Qiagen Biosystems). Values detected were standardized to sno72. For validation of DEGs, 100 ng of total RNA was processed with DNase I (Invitrogen) followed by reverse transcription using Superscript III (Invitrogen). Transcripts were detected using the following primers: Ppia: RT2 qPCR Primer Assay for Mouse Ppia (Qiagen Biosystems), C3: Forward, GAGGCACATTGTCGGTGGTG, Reverse, CCAGGATGGACATAGTGGCG, C5ar1: Forward, ATGGACCCCATAGATAACAGCA, Reverse, GAGTAGATGATAAGGGCTGCAAC, Stat1: RT2 qPCR Primer Assay for Mouse Stat1 (Qiagen Biosystems) RT2 qPCR Primer Assay for Mouse Tlr7 (Qiagen Biosystems), RT2 qPCR Primer Assay for Mouse Tlr8 (Qiagen Biosystems), and RT2 qPCR Primer Assay for Mouse Trem2 (Qiagen Biosystems). Transcripts detected were standardized to Ppia values. All qPCR reactions were 20 uL volumes utilizing a StepOnePlus machine (Applied Biosystems).
qPCR statistical analysis
Statistical analysis was performed using Graphpad Prism 6.01. qPCR analysis utilized two-way ANOVA testing with Sidak’s correction for multiple testing. qPCR statistics were performed on untransformed values but plotted as linearized values for ease of visualization. Alpha was set to 0.05 for all testing.
Detection of EAE-related sex-specific signatures
GOrilla/REViGO analysis of differentially expressed genes in female EAE. Top 10 non-dispensable terms are reported
Extracellular region part
Side of membrane
External side of plasma membrane
GOrilla/REViGO analysis of differentially expressed genes in male EAE. Top 10 non-dispensable terms are reported
NADH dehydrogenase complex
Extracellular region part
Mitochondrial membrane part
Validation of RNASeq
qPCR statistics for DEGs and DEmiRs
Adjusted p value
Adjusted p value
Differences in transcriptional signatures by sex within EAE
Canonical pathways differentially expressed between female vs. male EAE
Ingenuity canonical pathways
Acute phase response signaling
Neuroinflammation signaling pathway
Cytotoxic T lymphocyte-mediated apoptosis of target cells
Hepatic fibrosis/hepatic stellate cell activation
GP6 signaling pathway
Validation of miRSeq
EAE is a complex disease that primarily targets the CNS. Less is known about how the PNS is affected in EAE. We have shown that a disease-related transcriptional signature is detectable in the DRG of EAE mice and that sex is an important factor in the characterization of this signature. To accomplish this, we utilized next-generation sequencing for mRNA and miRNAs from the DRG at the clinical onset of EAE and profiled the differentially expressed genes and miRNAs to acquire a functional signature for each sex. The identification of functional signatures for complex disease states or comorbidities such as pain is relevant for prompting strategic experimental design and stratifying future therapies.
Detecting the signature
We first wanted to describe the transcriptome of our animals, considering sex as a factor. Sex differences have been previously described in immunity [26, 27, 28] and inflammatory responses [29, 30], and both of these processes play a key role in our disease model. Indeed, we found extensive differential expression of the transcriptome in female EAE mice but only small amounts of differentially expressed genes (DEGs) in male EAE mice. Large changes in gene expression have been reported in the DRG after peripheral nerve injury, and these changes have been associated with neuropathic pain behaviors [31, 32]. This supports previous evidence from our laboratory, indicating that female but not male DRG neurons are hyper-responsive at the onset of EAE . In addition to these differences in the quantity of DEGs, we also found major differences in the functional signature of this altered transcriptional profile. Our data indicates that there are extensive immune and cell-signaling-related processes disturbed in the female DRGs, a potential by-product of immune cells transiently localized there .
Sex-specific disease effects
Comparing the sexes directly, we noted the appearance of phagosomal activation pathways, complement, and interferon signaling as being differentially affected in female EAE versus their CFA controls (Table 3). The phagosome-related terms contained references to DEGs such as Trem2, Csf1r, Tlr7, and most of the complement system. Taken together, this data indicates sex-specific dysfunction of inflammatory activity in phagocytic cells. Laffont et al. have reported increased interferon responses of estrogen activated human plasma dendritic cells to TLR7 ligands, suggesting that this mechanism may be sex-hormone linked . Tlr7 is an X-linked gene and this may account for the sex-bias we observed, as X chromosome genes may escape silencing.
Interferon signaling is an approved target for MS therapy, as beta-interferons have been utilized in relapsing-remitting MS since the late 1990s . Although the effects of interferon inhibitors are modest, all clinical trials conducted to date have been with mixed female and male populations. The possibility of female specific interferon dysregulation raises the possibility of higher efficacy in a subset of the clinical population. We validated the female-EAE-specific increased expression of Stat1—a downstream effector of types I and II interferons—suggesting that interferon signaling is activated in these animals. In male EAE mice however, Stat1 levels were significantly reduced suggestive of a silencing of interferon signaling. As beta-interferon therapy would activate these pathways, it follows that their effect may have differential effects based on sex.
Activation of interferon signaling through Stat1 typically results in increased inflammation and enhanced immune activity . While these pathways do seem to be activated in the diseased group of females, it is apparent that it is not the case for the males. The high activation of interferon-signaling and Stat1 in females is complemented by a similar increase in transcripts associated with inflammatory macrophages, including most members of the complement system. Complement proteins are involved with the recruitment and phagocytic capacity of innate immune cells [37, 38]. Based on our transcriptomic data, females may be poised to specifically respond to interferon stimulation, resulting in heightened phagocytic activity in the DRG in response to EAE.
Accounting for the signature
To account for the transcriptomic changes observed in EAE, we also sequenced the miR-transcriptome. We observed a small amount of overlap between the detected DEmiRs in each sex, corresponding to miRNAs previously linked to EAE and the MS literature: miRs-21 [39, 40] and − 142 . The most significant miRNA for both sexes, miR-21, was both highly expressed and confirmed to be significantly upregulated in both female and male EAE DRGs. MS brain lesions have been shown to be enriched with miR-21 in patients with MS but not neuromyelitis optica (NMO) . NMO is an autoimmune disease with similar symptomatology, potentially indicating that this signature is specific to MS/EAE pathology. Therefore, miR-21 may be suitable as a marker of EAE. miR-21 has been well studied in the context of cancer, giving rise to its label of an “oncomiR” [43, 44, 45]. Interest in miR-21 as a regenerative marker is also emerging with evidence that it is pro-regenerative  and immunomodulatory . Macrophages uptake miR-21 laden exosomes after peripheral nerve injuries  where it can act as a direct ligand for Toll-like receptors 7 and 8 [49, 50]. Interestingly, miR-21 is affected by sex hormones, containing a direct response element for androgens  and directly affected by estradiol signaling [52, 53]. As miR-21 was similarly upregulated in both sexes, we additionally screened for miRNAs unique to sex. miR-122 was only upregulated in the diseased female DRG. miR-122 is primarily described for its expression in the liver; however, it is known to increase in the circulation after injury [54, 55] and may interact with type I interferon signaling . Together with the transcriptomic signatures, our data indicates that the female DRGs may be uniquely affected by the epigenetic regulation of miRs-21 and -122. Although EAE is canonically not thought of as a PNS disease, the presence of the typical “MS-miR-signature” indicates that female DRGs may be affected concordantly with CNS lesions.
To our knowledge, this is the first presentation of transcriptome and miR-transcriptome dysregulation in DRGs using the EAE mouse model. We have shown that the miR-signature is primarily defined by miR-21, bearing similarity to previously described findings in the spinal cord and serum of mice with EAE, and MS lesions. We present evidence of a female-biased dysregulation of both “omic” analyses, the presence of distinct functional transcriptomic signatures for each sex, and that interferon signaling and phagosome function may be indicated for future analysis. Acknowledging fundamental differences between the sexes is not only important for experimental design, but for the development of potential sex-specific therapies. Our study provides a platform to study sex-specific changes in the DRG of EAE and the role that miRNAs may have on their transcriptome.
The authors wish to thank G. Tenorio for assistance in making the figures.
AC conducted the EAE experiment. TNF, MSY, and MD performed the tissue extraction and qPCR validation. TNF analyzed and interpreted the sequencing data. TNF, CAJ, AEF, and BJK were involved in experimental design and manuscript preparation. All authors read and approved the final manuscript.
Funding for this project was provided by operating grants from the MS Society of Canada (MSSC), the Alberta MS Collaboration, and the University Hospital Fund (University of Alberta). TNF and AC were supported by studentships from the MS Society of Canada. MSY was supported by the Alexander Graham Bell Canada Graduate Scholarship from NSERC.
Ethics approval and consent to participate
All animal studies complied with the Canadian Council on Animal Care Guidelines and Policies and were approved by the Animal Care and Use Committee: Health Sciences for the University of Alberta.
Consent for publication
The authors declare that they have no competing interests.
- 13.Team R. R: a language and environment for statistical computing; 2013.Google Scholar
- 15.Morgan M, Pagès H, Obenchain VHN. Rsamtools: binary alignment (BAM), FASTA, variant call (BCF), and tabix file import; 2018.Google Scholar
- 24.Mi H, Thomas P. PANTHER pathway: an ontology-based pathway database coupled with data analysis tools; 2009. p. 123–40.Google Scholar
- 34.Laffont S, Rouquié N, Azar P, Seillet C, Plumas J, Aspord C, et al. X-chromosome complement and estrogen receptor signaling independently contribute to the enhanced TLR7-mediated IFN-α production of plasmacytoid dendritic cells from women. J Immunol. 2014 Dec 1;193(11):5444–52.PubMedCrossRefGoogle Scholar
- 55.Park H-K, Jo W, Choi H-J, Jang S, Ryu J-E, Lee H-J, et al. Time-course changes in the expression levels of miR-122, −155, and −21 as markers of liver cell damage, inflammation, and regeneration in acetaminophen-induced liver injury in rats. J Vet Sci. 2016;17(1):45–51.PubMedPubMedCentralCrossRefGoogle Scholar
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.