We used male and female C57BL/6 mice (10 weeks old, 20~25 g; Central Lab. Animal Inc., Seoul, Korea). Experimental animals were maintained under controlled temperature (24±2℃) and humidity (55±10%) conditions, with a 12/12 h light/dark cycle, and given free access to water and standard food. The animal protocols conformed to current international laws, and to the policies of the National Institutes of Health (NIH) Guide for the Care and Use of Laboratory Animals (NIH Publication No. 85-23, 1985, revised 1996). All experimental procedures were conducted in accordance with the Guidelines for the Care and Use of Laboratory Animals of Jeju National University (Permission No. 2017-0019). Every effort was made to minimize the number of animals used and their suffering.

Mice in the EAE group were treated with 1 mg/ml of the MOG_35-55 peptide (purity >96.44%; GL Biochem Ltd., Shanghai, China) emulsified in complete Freund's adjuvant (Difco Laboratories Inc., Detroit, MI, USA) and supplemented with 5 mg/ml Mycobacterium tuberculosis H37Ra (Difco Laboratories Inc.) by subcutaneous injection into the hind flank. Pertussis toxin (500 ng; List Biological Laboratories, Inc., Campbell, CA, USA) was injected intraperitoneally at days 0 and 2 post-immunization (PI) [27]. Following immunization, the experimental animals were weighed daily and their clinical signs were monitored and scored as follows: grade 0 (G.0), no signs; G.1, floppy tail; G.2, mild paraparesis; G.3, severe paraparesis; G.4, tetraparesis; G.5, moribund condition or death; R.0, recovery; and R.1, recovery with floppy tail.

Mice were fixed by transcardial perfusion with 4% paraformaldehyde in phosphate buffer (0.1 M, pH 7.4) under deep ether anesthesia at each sampling time point, including the paralytic peak (G.3, day 23 PI) and chronic paralytic (R.1, day 83 PI) stages. The skull and spinal cord were collected at each stage. For histological examination, the skull was decalcified for 10 days using decalcifying solution (formic acid-sodium citrate method) [28]. Following decalcification, the skull, including the nasal cavity, olfactory bulb, and spinal cord, was embedded in paraffin wax, and 5-µm-thick sections were prepared using a rotary microtome (RM 2135; Leica, Nussloch, Germany). Paraffin sections were deparaffinized in xylene, dehydrated in a serial ethanol solution, and then stained with hematoxylin and eosin to evaluate inflammation. For western blot analysis, the olfactory bulb and olfactory mucosa were removed and frozen at −80℃ until use.

Total RNAs of the olfactory bulb, which were collected from different individuals in control and EAE-affected mice at the paralytic peak stage (n=3 per group), were extracted using TRIzol RNA Isolation Reagent (Life Technologies, Thermo Fisher Scientific, Waltham, MA, USA). The olfactory mucosa in control and EAE-affected mice (n=3 per group) was used to further investigate olfactory transduction signaling at the paralytic stage. The quantity and quality of total RNA were evaluated using the Agilent 2100 Bioanalyzer RNA kit (Agilent, Santa Clara, CA, USA). The isolated RNA was processed to prepare an mRNA sequencing library using the Illumina TruSeq Stranded mRNA Sample Preparation kit (Illumina, San Diego, CA, USA) according to the manufacturer's protocol. All libraries were quantified by quantitative polymerase chain reaction using the CFX96 Real Time System (Bio-Rad, Hercules, CA, USA) and sequenced using NextSeq500 sequencers (Illumina).

Potential sequencing adapters and raw quality bases in the low reads were trimmed using Skewer [29]. After trimming the low quality bases and sequencing adapters, the cleaned high quality reads were mapped to the reference genome using STAR software [30].

Gene annotation of the reference genome from the UCSC genome browser database (https://genome.ucsc.edu) in GTF format was used as a gene model, and the expression values were calculated as fragments per kilobase of transcript per million fragments mapped (FPKM). The DEGs between the two selected biological conditions were analyzed by Cuffdiff software in the Cufflinks package [31].

Total RNA of the olfactory bulb in the control, and in paralytic and recovery stages (n=5 per group), was isolated with TRIzol RNA Isolation Reagent (Life Technologies, Thermo Fisher Scientific) and cDNA was prepared using ReverTra Ace-α (TOYOBO, Osaka, Japan). The primer sequences were as follows: Omp (203 bp), forward 5′-CTCATTGTCCCTCGCCTTG-3′ and reverse 5′-GCCTATCTCCTGCCTCCA-3′; and Stoml3 (193 bp), forward 5′-ATGGAGTCGTCTATTACAGAATCTACAG-3′ and reverse 5′-TCGGTGGCATCATCAAGCAAG-3′. PCR reactions were performed using the CFX96 Real Time System (Bio-Rad) using SYBR Green Supermix (2×) (TOYOBO) as follows: 40 cycles of denaturation (10 s, 95℃), annealing (10 s, 58℃), and extension (10 s, 72℃).

Following deparaffinization, sections were prepared as previously described [32]. Briefly, following incubation with matched blocking serum (10% [v/v] normal goat serum in phosphate-buffered saline [PBS]; Vectastain Elite ABC kit; Vector Laboratories, Burlingame, CA, USA), the samples were incubated with rabbit anti-ionized calcium binding protein-1 (Iba-1) (1:1,000, 019-19741, Lot. LKH4161; Wako Pure Chemical Industries, Ltd., Osaka, Japan) and mouse monoclonal anti-glial fibrillary acidic protein (GFAP) (1:400, MAB360, Lot. 2041104; EMD Millipore Corp., Temecula, CA, USA) for 1 h at room temperature. After three washes in PBS, the peroxidase reaction was developed using a diaminobenzidine substrate kit (Vector Laboratories).

To confirm the GFAP-positive cells in the olfactory bulb, we performed double immunofluorescence staining as in our previous study [33]. For double staining, the sections were incubated with mouse anti-GFAP antibody, which were visualized with antimouse IgG conjugated to tetramethyl rhodamine isothiocyanate (TRITC) (1:50, T5393, lot 103K9165; Sigma-Aldrich, St, Louis, MO, USA). The sections were washed three times with PBS and then incubated with rabbit anti-glutamic acid decarboxylase (GAD) 65/67 (1:1,000, AB1511, lot 2324512; Millipore, Temecula, CA, USA) and visualized using anti-rabbit IgG conjugated fluorescein isothiocyanate (FITC) (1:50, F0382, lot 018K60572; Sigma-Aldrich). Images were captured using a BX-51 fluorescence microscope (Olympus, Tokyo, Japan) and merged using Adobe Photoshop (Adobe Systems, San Jose, CA, USA).

All measurements represent averages of three independent experiments. All values are presented as means±standard error of the mean. The results were analyzed using one-way analysis of variance followed by Student-Newman-Keuls post hoc test for multiple comparisons. A p-value <0.05 was considered to reflect statistical significance.

Following immunization with the MOG_35-55 peptide in mice, tail atony was observed at approximately day 12 PI. Hind limb paralysis followed at days 18~24 PI. The tail atony lasted for a considerable period. By day 30 PI, EAE-affected mice had recovered from hind limb paralysis and showed chronic paralysis (Fig. 1).

Greater inflammatory cell infiltration was detected in the spinal cord (Fig. 2B) and olfactory bulb (Fig. 2D) of EAE-affected mice compared with controls (Fig. 2A and 2C). Furthermore, microgliosis and astrogliosis in the olfactory bulb of EAE-induced mice were remarkably high (Fig. 2E and 2G) compared with control mice (Fig. 2F and 2H). Some Iba-1-positive cells were observed along the olfactory nerve (Fig. 2F), while the decalcified tissue did not show Iba-1-positive cells in the normal olfactory epithelium (Fig. 2E). Thus, the olfactory bulb of EAE-affected mice is significantly injured following immunization, and may be influenced by inflammatory mediators in the cerebrospinal fluid, as well as by inflammatory cells.

We screened for DEGs in the olfactory bulb of control and EAE-affected mice at the paralytic stage. According to the cut-off value for DEG screening (log2 [fold change] >1 and p-value <0.05), the expression levels of 563 genes were significantly changed (44 upregulated and 519 downregulated). These altered genes may be involved in the Gene Ontology terms biological process (588), cellular component (69), or molecular function (51); CORUM protein complexes (4); human phenotype ontology (97); Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways (50); miRBase microRNAs (14); or TRANSFAC TFBS (143) (Supplementary Data 1). Hierarchical clustering heat maps of the DEGs are shown in Fig. 3. We further analyzed inflammation and olfactory transduction signaling in the olfactory bulb.

Gene expression of Il1b, Ccl2, Cd86, Stat1, and Irf5, markers of the M1 macrophage phenotype, was significantly increased (Table 1). M2 macrophage marker genes, including Trem2 and Cd163 were also upregulated. We further examined the expression of interleukins, which have been associated with EAE pathogenesis [34]. Some interleukins (although not the majority), including Il1a, Il1b, and Il18bp, were significantly upregulated (data not shown).

We further analyzed DEG for markers of olfactory ensheathing cells (OECs) [35,36]. The p75Ngfr was significantly downregulated, while Gfap was significantly upregulated (Table 2). With double immunofluorescence staining, most of the increased GFAP was seen in astrocytes and the olfactory nerve layer, although it was also expressed in few interneurons (Fig. 4).

Since EAE-affected mice showed olfactory dysfunction in a previous study [27], we analyzed olfactory transduction signaling (KEGG pathway: mmu04740). Olfactory receptor genes were not detected in the olfactory bulbs (data not shown). Among 15 upregulated genes, Rtp4 was significantly increased. Among 29 downregulated genes, Omp was significantly decreased (Fig. 5A). In addition, the Stoml3 gene, which is involved in olfactory transduction [3], was also significantly decreased in the olfactory bulb of EAE-affected mice. Downregulation of Omp and Stoml3 was confirmed using quantitative RT-PCR (Fig. 5B and 5C).

We focused on the evaluation of olfactory transduction signaling-related genes (KEGG pathway: mmu04740) in the olfactory mucosa and, in particular, whether similar changes in olfactory transduction signaling-related genes occurred in the olfactory mucosa. In the olfactory mucosa of EAE-affected mice, inflammation-related genes, including Il1b, Ccl2, and Cd86, were upregulated (data not shown), as in the olfactory bulb. In total, 120 genes, including 119 olfactory receptors and Slc8a1, were significantly increased in the olfactory mucosa of EAE-affected mice, while Gng13 and Calml3 were significantly decreased in EAE-affected mice. In addition, we also found that odorant-binding proteins, including Obp1a, Obp1b, and Obp2a, which are important in the odorant perception process, were significantly decreased in the olfactory mucosa of EAE-affected mice (Fig. 6A and 6B).