2025 年 48 巻 11 号 p. 1687-1693
Multiple sclerosis (MS) develops due to an abnormal T-cell immune response to autoantigens and control of T-cell activation is a mainstream approach for its treatment. In the present study, neogenin, a key molecule for T-cell activation, was used as a targeted molecular gene therapy for MS. Lipid nanoparticles (LNPs) loaded with small interfering RNA (siRNA) targeting neogenin (LNPsiNeo) were prepared, and their therapeutic effect on experimental autoimmune encephalomyelitis (EAE) induced by myelin oligodendrocyte glycoprotein, a model of MS was evaluated. Neogenin gene expression was reduced by LNPsiNeo in mouse EL-4 cells and splenocytes of LNPsiNeo-treated EAE mouse. Additionally, fluorescence-activated cell sorting (FACS) revealed that the number of CD4+ T cells in the splenocytes of EAE mouse decreased after intravenous injection of LNPsiNeo. Furthermore, the progression of encephalomyelitis symptoms was significantly suppressed by LNPsiNeo, whereas the lipid nanoparticle with control siRNA failed to show any effect. The present study suggests that neogenin is a target molecule for EAE gene therapy and LNPsiNeo may be suitable for the MS treatment.
Multiple sclerosis (MS) is an autoimmune disease of the central nervous system (CNS); it is an inflammatory demyelinating disease caused by an abnormal immune response to CNS antigens, such as myelin components. The use of immunosuppressive drugs, such as steroids and disease-modifying drugs (DMDs), to control the immune response to autoantigens is the mainstream approach for the clinical treatment of MS. However, because of their non-specific immunosuppressive effect and cytotoxicity, patients with MS frequently suffer side effects due to infections, such as progressive multifocal leukoencephalopathy (PML), which often limits the treatment of the disease.1,2) Therefore, the development of molecular-targeted therapies for MS is urgently needed.
Neogenin belongs to the membrane-bound immunoglobulin superfamily, which is expressed in various organs such as the brain, large intestine, and skin, and functions as a receptor for ligands, such as bone morphogenetic protein, netrin, and repulsive guidance molecule-a (RGMa).3) Neogenin has broad functions and is involved in nerve cell differentiation, axon elongation, placental formation, and angiogenesis during embryonic development.4,5) Boneschansker et al. reported that neogenin promoted T cell invasion into the inflammatory sites in the presence of netrin-1.6) Muramatsu et al. demonstrated that T cell activation was induced by the binding of RGMa expressed on antigen-presenting dendritic cells to neogenin expressed on the surface of the T cell membrane, and MS was subsequently developed.7) Therefore, neogenin is a key molecule in MS pathogenesis and could be a targetable molecule for its gene therapy. In this study, a lipid nanoparticle (LNP)-based small interfering RNA (siRNA) medicine was prepared to suppress the gene expression of neogenin, the gene-silencing potency of LNPs loaded with siRNA targeting neogenin (LNPsiNeo) was investigated, and the therapeutic effect of LNPsiNeo on experimental autoimmune encephalomyelitis (EAE), a representative model of MS, was demonstrated.
Synthesis of the autoantigen peptide myelin oligodendrocyte glycoprotein (MOG35-55: MEVGWYRSPFSRVVHLYRNGK) was outsourced to GenScript Japan Inc. (Tokyo, Japan). Dilinoleylmethyl-4-dimethylaminobutyrate (DLin-MC3-DMA) was purchased from MedChemExpress (Monmouth Junction, NJ, U.S.A.). Distearoylphosphatidylcholine (DSPC), cholesterol (CHOL), and dimyristoylmethoxypolyethylene glycol-2000 (DMG-PEG2000) were supplied by Nippon Fine Chemical Co., Ltd. (Hyogo, Japan). The siRNA was purchased from Hokkaido System Science Co., Ltd. (Sapporo, Japan) and nucleotide sequences were as previously reported: 5ʹ-CAAUUCCAUGGAUAGCAAU-3ʹ and 5ʹ-AUUGCUAUCCAUGGAAUUG-3ʹ for neogenin (siNeo); and 5ʹ-GGCUACGUCCAGGAGCGCACC-3ʹ and 5ʹ-UGCGCUCCUGGACGUAGCCUU-3ʹ for green fluorescent protein (GFP) (siGFP) as a control.8,9)
Cell CultureThe mouse T-lymphoma cell line of EL-4 was obtained from Japanese Collection of Research Bioresources Cell Bank and cultured in RPMI medium (FUJIFILM Wako Pure Chemical Corporation, Osaka, Japan) supplemented with 10% fetal bovine serum (FBS, Hyclone) and penicillin/streptomycin (FUJIFILM Wako) in a CO2 incubator.
AnimalsC57BL/6 female mice were purchased from Japan SLC (Shizuoka, Japan). All animal experiments were performed at the Hamamatsu University School of Medicine and approved by the Animal and Ethics Committee of the university. The animals were cared for according to Animal Facility Guidelines. EAE was induced by MOG immunization. MOG (200 µg) mixed with complete Freund’s adjuvant containing Mycobacterium butyricum (500 µg, Chondrex, Inc., Woodinville, WA, U.S.A.) was subcutaneously injected into the back of C57BL/6 mice; and then pertussis toxin solution (FUJIFILM Wako) was intravenously injected into the mice via a tail vein at days 0, 2, and 4 (200 ng/d).
Preparation of siRNA-Loading LNPEthanol solutions containing DLin-MC3-DMA, DSPC, CHOL, and DMG-PEG2000 at molar ratio of 49.3/10.2/38.9/1.6 was prepared. The lipid solution was mixed with three times the volume of 25 mM sodium acetate solution (pH 4.5) containing siGFP or siNeo (flow rate: 1.5 mL/min) to give a negative/positive (ratio of 6 in a microfluidic mixer (Y-type, YMC Co., Ltd., Kyoto, Japan) using two syringe pumps (YMC Co., Ltd.). Then, the mixed solution was dialyzed with a Spectra/Por® 4 Dialysis Membrane Standard RC Tubing MWCO:12–14 kD (Spectrum Laboratories, Piscataway, NJ, U.S.A.) in an adequate volume of 50 mM MES/citrate buffer (pH 6.7) for 4 h and subsequently in phosphate-buffered saline (PBS) (pH 7.4) overnight. The resulting LNP solution was ultrafiltered using an Amicon Ultra-4 100 K filter (Millipore, Bedford, MA, U.S.A.) and diluted with 66 mM phosphate buffer (pH 7.4) to obtain LNP loaded with siGFP (LNPsiGFP) or siNeo (LNPsiNeo). For fluorescence imaging assay, HiLyte Fluor™ 488-modified positive control siRNA (NIPPON GENE CO., LTD., Tokyo, Japan) was loaded into LNP to obtain fluorescently-labeled LNPsiRNA (LNPsiRNA-FL). The particle size and ζ-potential of the LNPs were measured using a zeta-potential and particle size analyzer, ELSZ-1000ZS (Otsuka Electronics Co., Ltd., Osaka, Japan). The amounts of cholesterol and siRNA in siRNA-loading LNP solution were measured using a LabAssay Cholesterol (FUJIFILM Wako) and Quant-iT RiboGreen RNA Assay Kit (Thermo Fisher Scientific, Waltham, MA, U.S.A.), respectively. Following negative staining with uranyl acetate, transmission electron microscopy (TEM; JEM1400, JEOL, Tokyo, Japan) was performed at the Advanced Research Facilities & Services, Hamamatsu University School of Medicine.
Gene-Silencing Assay in VitroEL-4 cells were seeded in a 35-mm dish (1 × 105 cells/dish) and precultured overnight. The cells were incubated with LNPsiGFP or LNPsiNeo (1, 10, or 100 nM siRNA). As a positive control, recombinant transforming growth factor-β1 (TGF-β1, 5 nM, R&D Systems, Minneapolis, MN, U.S.A.), an immunosuppressive cytokine, was also incubated with the cells. After 48 h of incubation, the cells were lysed and gene expression of neogenin was confirmed using a RT-PCR assay: Total RNA was extracted with a NucleoSpin® RNA kit (TaKaRa Bio Inc., Shiga, Japan); reverse transcription reaction was carried out with a PrimeScript RT Master Mix (TaKaRa Bio Inc.); and real-time PCR was performed using TB Green Premix Ex Taq II (TaKaRa Bio Inc.) containing neogenin or β-actin primers (TaKaRa Bio Inc.) with a Thermal Cycler Dice Real Time System III (TaKaRa Bio Inc.). For fluorescence observation of neogenin expression, EL-4 cells attached to MAS-coated glass slides (Matsunami Glass Ind., Ltd., Osaka, Japan) were stained with anti-mouse neogenin goat immunoglobulin G (IgG) antibody (R&D Systems) as the primary antibody followed by Alexa594-conjugated donkey anti-goat IgG antibody as the secondary antibody (Thermo Fisher Scientific). Fluorescence was observed using a fluorescence microscope (IX73; Olympus, Tokyo, Japan) and scanned using a DP80 digital camera (Olympus).
In Vivo Fluorescence Imaging AssayEAE mice (n = 2) were intravenously injected with LNPsiRNA-FL via a tail vein at 10 µg/mouse on day 8 after MOG immunization, and the spleens were dissected from the mice after 24 h. Fluorescence of LNPsiRNA-FL accumulated in the spleens was analyzed by using a in vivo imaging system (IVIS Lumina III (PerkinElmer, Inc., Waltham, MA, U.S.A.)).
In Vivo Gene Silence Assay and Fluorescence-Activated Cell Sorting (FACS)EAE mouse was intravenously injected with LNPsiGFP or LNPsiNeo via a tail vein at 10 µg/mouse on days 11 and 14 after MOG immunization and the spleen was dissected from the mouse on day 15. The splenocytes were separated from the spleen in RPMI-1640 medium, passed through a 40 µm-mesh-sized BD cell strainer (Corning, Corning, NY, U.S.A.) to obtain single cells, and hemolyzed with ACK lysing buffer. To evaluate the gene-silencing effect, real-time RT-PCR analysis was performed as previously described. To determine the number of CD4+ T cells in splenocytes, FACS analysis was as follows: splenocytes were suspended in FACS washing buffer (2% FBS +0.05% NaN3-PBS), blocked with Clear Back (MBL), and labeled with Anti-mouse CD4 monoclonal antibody (mAb)-fluorescein isothiocyanate (FITC) (MBL). The FACS scan was carried out with a Gallios flow cytometer (Beckman Coulter, Brea, CA, U.S.A.) and data was analyzed with FlowJo v10 software (BD, Franklin Lakes, NJ, U.S.A.).
Therapeutic ExperimentMOG-induced EAE mice were intravenously injected with phosphate buffer (Control), LNPsiGFP, or LNPsiNeo (10 µg/mouse/d as siRNA dosage) via a tail vein on days 11, 14, and 17. Clinical EAE signs of the mice were monitored and scored according to the following 13 grades (0: no clinical signs; 0.5: slightly limp tail; 1: tail tip paralyzed; 1.5: partially limp tail; 2: completely limp tail; 2.5: completely limp tail and weakness of hind legs; 3: uncoordinated movement; 3.5: one hind limb paralyzed; 4: both hind limbs paralyzed; 4.5: both hind limbs paralyzed and one forelimb paralyzed; 5: both hind limbs paralyzed and both forelimbs paralyzed; 5.5: Moribund state; and 6: death). Changes in body weight of the mice were also monitored.
LNPsiNeo were prepared using an ethanol dilution and self-assembly preparation method with microfluidic mixing, which was originally reported by Cullis and colleagues,10) enabling the loading of a large amount of siRNA into LNP-containing ionizing lipids. Dynamic light scattering (DLS) measurements and TEM showed that LNPsiNeo was stably dispersed in aqueous solution, with mean particle sizes of 160 and 151 nm for LNPsiGFP and LNPsiNeo, respectively (Table 1, Fig. 1). The polydispersity index (PDI) values were less than 0.12, indicating that the siRNA-loaded LNPs were nearly monodisperse in aqueous solutions. In addition, ζ-potentials of LNPsiNeo and LNPsiGFP were −4.2, and −2.1, respectively, therefore, these siRNA-loading LNPs are slightly negative-charged. Furthermore, the percentage of siRNA loaded onto the inner side of the LNP exceeded 70%.
| LNP | N/P ratio | Particle size (d.nm) | Polydispersity index | ζ-Potential (mV) | siRNA loading (%) |
|---|---|---|---|---|---|
| LNPsiGFP | 6 | 160 ± 15 | 0.11 ± 0.06 | −4.2 ± 2.9 | 70.3 ± 8.04 |
| LNPsiNeo | 6 | 151 ± 14 | 0.12 ± 0.05 | −2.1 ± 1.3 | 70.0 ± 9.90 |
Average values were calculated from four different preparations. N/P, negative/positive.
Negative staining with uranyl acetate was performed, and LNPsiGFP and LNPsiNeo were observed using TEM. TEM, transmission electron microscopy.
The gene-silencing efficacy of LNPsiNeo was determined using EL-4 cells as a model of T lymphocytes. As shown in Fig. 2A, neogenin expression was strongly suppressed by incubation with LNPsiNeo at a concentration of 1 nM. In contrast, the LNPsiGFP-bearing control siRNA showed very little knockdown efficiency at a high concentration of 100 nM. TGF-β, as an immunosuppressive cytokine used as positive control also showed an obvious gene-silencing effect. Fluorescence observation of neogenin expression demonstrated that neogenin protein expression decreased (Fig. 2B).
EL-4 cells (1 × 105 cells) were incubated with LNPsiGFP or LNPsiNeo for 48 h. Real-time RT-PCR analysis was performed to evaluate changes in neogenin gene expression. Data represent the average and upper limit values of four different measurements. TGF-β, an immunosuppressive cytokine, was used as a positive control (A). After incubation with LNPs, immunofluorescence staining of neogenin was performed, and the cells were observed using a fluorescence microscope (B). DIC: differential interference contrast.
To demonstrate the accumulation of siRNA at spleen after intravenous injection into EAE mouse, fluorescently-labeled LNPsiRNA (LNPsiRNA-FL) was intravenously injected into EAE or normal mouse and the fluorescence in the spleen was scanned. The result indicated that the fluorescence was observed in the spleens of both mice. Then, to evaluate in vivo gene-silencing effect of LNPsiNeo, neogenin gene expression in the spleen of mouse after intravenous injection of LNPsiNeo was investigated. The mRNA level of neogenin was decreased in the splenocytes of LNPsiNeo-injected EAE mouse compared with that in non-treated EAE mouse and this effect was not observed in LNPsiGFP-treated mouse (Fig. 3A).
(A) Accumulation of LNPsiRNA in spleen. Fluorescently-labeled LNPsiRNA (LNPsiRNA-FL) was intravenously injected into MOG-EAE (n = 1) or normal mouse (n = 1) via a tail vein and allow to circulate for 24 h. Then, the spleen of each mouse was dissected and the fluorescence was scanned by an in vivo imaging system. NT: non-treatment. (B) In vivo gene-silencing effect mediated by LNPsiNeo injection. LNPsiGFP or LNPsiNeo was intravenously injected into EAE mouse (n = 1) via a tail vein at 10 µg/mouse/d as the siRNA dosages on days 11 and 14 after MOG immunization. After the dissection of the spleen, real-time RT-PCR analysis was carried out to determine the expression of the neogenin gene. Data represent the average and upper limit values of four different measurements. (C) Change of splenic CD4+ T cells after the treatment with LNPsiNeo. Dissected splenocytes were probed with Anti-mouse CD4 mAb-FITC to identify the CD4+ T cells and FACS was performed. Lymphocyte fraction was gated from side scatter (SSC) versus forward scatter (FSC) plot (black circle). Similar results were obtained in all separate experiments.
The suppressive effect of LNPsiNeo on T cell differentiation was confirmed by measuring the number of CD4+ T cells in the splenocytes of EAE mouse. The lymphocyte population was gated from the splenocytes of the EAE mouse and FACS analysis was performed following CD4+ cell selection. The number of CD4+ T cells decreased in the splenocytes of LNPsiNeo-treated EAE mouse in comparison with that of LNPsiGFP-treated or control mouse (Fig. 3B), suggesting that treatment with LNPsiNeo suppressed the differentiation of naive T cells to CD4+ T cells in the splenocytes of EAE mouse.
Therapeutic Effect of LNPsiNeo on EAETo evaluate the therapeutic effect, MOG-induced EAE mice were used as a model of MS, and a therapeutic experiment with LNPsiNeo was performed. The injection of the samples commenced when symptoms of encephalomyelitis, such as motor neuron disorders, were observed in the mice (first injection: day 11), and changes in the clinical score of EAE symptoms were monitored. The treatment with LNPsiNeo significantly suppressed the EAE symptoms during therapy (Fig. 4A). In contrast, LNPsiGFP showed slight effect during the injection period, but failed to keep for a long term and the clinical symptom grade of these groups became to be similar to that of untreated EAE mice. In addition, when body weight was monitored as an indicator of the side effects of LNPsiNeo injection, no change in body weight was observed in the LNPsiNeo-treated mice (Fig. 4B). Furthermore, nor is any liver injury caused by the injection detected, which histology was similar to that seen for the non-treated EAE mouse (Supplementary Fig. S1). These results suggest that LNPsiNeo has therapeutic potential for EAE.
(A) Suppression of clinical EAE symptoms using LNPsiNeo injection. EAE mice (n = 4) were intravenously injected with phosphate buffer (Control), LNPsiGFP or LNPsiNeo (10 µg/mouse/d as the siRNA dosages) via a tail vein on days 11, 14, and 17 (black arrows) after immunization with MOG. Clinical symptoms of EAE were monitored and scored based on 13 grades. Data represent the mean ± standard error of the mean (S.E.M.). Significant differences were shown only at day 36 (*, p < 0.05; **, p < 0.01, Tukey HSD). Similar result was obtained in a separate experiment. (B) Body-weight change in EAE mice. Data represent the mean ± standard deviation (S.D.) and black arrows indicate the days of drug injection.
Abnormalities in the immune reaction generally cause immune disease onset and progression, therefore, chemotherapy to regulate the immune system has become the gold standard strategy for the treatment of autoimmune diseases, including MS, which is caused by the overreaction of autoimmune T-cell responses against myelin components in the CNS.11) The first line clinical treatments of MS are to use DMDs to prevent the T cells from autoimmune activation. Interferon (INF)-β formulations have a protective effect on relapsing-remitting MS by suppressing T cell activation12) and some steroid drugs have an immunosuppressive effect on T cell-mediated immune reaction. However, targeted molecular therapies are also of interest, since the mechanisms of MS pathology have been elucidated. Fingolimod is a functional antagonist against the sphingosine-1-phosphate (S1P) receptor that prevents the T cells from being transferred from secondary lymphatic organs to the blood circulation.13) Natalizumab is a genomically-modified antibody drug against α4β1 integrin, which is highly expressed in CD4+ T cells infiltrating into brain parenchymal tissue during MS development and has a suppressive effect on the immune cell invasion into CNS.14,15) The anti-CD20 antibody drug, ofatumumab, has been recently developed, because the CD20 molecule is expressed on the surface of B cells and a certain type of T cells which play an important role in production of autoantibodies. However, the side effects of these DMDs remain, and patients with MS are more susceptible to infectious diseases, such as PML, caused by systemic immunosuppression.16) Therefore, molecular screening of a targeted molecule for MS treatment is now ongoing and RNA interference (RNAi)-based therapy using siRNA or microRNA is a potential targeted therapy for MS17–22): Doi et al. demonstrated that the treatment of anti-CD3 antibody-stimulated CD4+ cells derived from PBMC with siRNA targeting NR4A2, an orphan nuclear receptor, reduced the secretion of inflammatory cytokines of both interleukin (IL)-17 and IFN-γ, and its adoptive immunotherapy showed the improvement of clinical encephalomyelitis symptoms.23) Yang et al. reported similar results by using siRNA targeting a transcription factor, T-bet.24) However, these approaches have not yet led to the development of systemically injected RNAi therapy. Cationic lipoplexes have been used for nucleic acid delivery, which are not applicable for systemic injection from the viewpoints of both pharmacokinetics and cytotoxicity.25) Herein, neogenin, one of the key molecules in CD4+ T cell-mediated MS development, is targeted. To produce a systemically injectable formulation of siNeo and deliver the siRNA to lymphatic immune cells, the LNP technique was applied. Cullis and Hope succeeded in producing LNP medicines loaded with nucleic acids, such as patisiran, which has been systemically injected to knockdown the protein expression of factor VII in hepatocytes,26) and an mRNA vaccine for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection.27) When LNPsiNeo were prepared by an ethanol dilution method with a microfluidic mixer, which is the same method used to prepare commercialized LNP medicines, monodisperse nanoparticles with a large amount of siNeo was loaded in the inner core phase were obtained (Table 1, Fig. 1). Its gene-silencing effect on neogenin-expressing EL-4 cells after incubation with LNPsiNeo was demonstrated and protein knockdown was observed using fluorescence imaging (Fig. 2). Next, in vivo gene-silencing and immunosuppressive effects on splenocytes following intravenous injection of LNPsiNeo into EAE mouse were evaluated, since EAE mouse was widely used as a T cell-mediated autoimmune MS model. Neogenin gene expression in splenocytes was strongly suppressed after treatment with LNPsiNeo (Fig. 3B), and FACS revealed that the population of CD4+ T cells in splenic lymphocytes was reduced (Fig. 3C). Although a slight decrease in the number of CD4+ T cells has been observed by LNPsiGFP, there was the possibility of non-specific off-target effect or immunomodulation one by injection of siRNA-loading LNPs composed of ionizable lipid. In any case, I assumed that LNPsiNeo became a systemically applicable formulation that enabled the delivery of siNeo to splenocytes, explaining the suppressive effects of LNPsiNeo on T-cell immunity in the splenocytes of EAE mouse. Kimura and Harashima have investigated the LNP biodistribution after systemic injection to mouse; and demonstrated that LNP distributed at not only liver but also spleen and showed gene expression in both organs by loaded oligonucleotides.28) They further examined the gene expression for each cell type in the spleen and have demonstrated that gene expressions were observed in some immune cell types including T cells and the expression pattern was dependent on lipid composition of LNPs. Besides, other research groups have succeeded to develop T-cell targeted LNPs which enables to effectively deliver oligonucleotides into T cells.29,30) Since lipid composition of LNPsiNeo is similar to their LNP, I predicted LNPsiNeo accumulated at spleen after intravenous injection and show the gene-silencing effect on the splenocytes. Actually, I observed the accumulation of fluorescently-labeled siRNA at the spleen after intravenous injection of its LNP formulation (Fig. 3A) and the uptake into the splenocytes including CD45+CD4+ T cells (Supplementary Fig. S2). To identify targeted cells of LNPsiNeo and reveal whole picture of the action mechanism for suppression of T-cell immunity, it is necessary to conduct further experiments. Finally, the therapeutic effect in MOG-induced EAE mice was examined. LNPsiGFP slightly showed the effect only during the injection period and it was thought that this transient suppression caused from off-target effect on T cell immunity by the LNP injection. On the other hand. LNPsiNeo injection significantly suppressed the clinical symptoms of MOG-EAE mice without any side effects and kept the effect for a long term, suggesting that the delivery of siNeo to splenocytes using LNPs can modulate the T-cell immunity via neogenin gene silencing and subsequently treat EAE (Fig. 4). To the best of our knowledge, the present study is the first to treat MS with an LNP-based siRNA to obtain a potent therapeutic effect on EAE. This LNP formulation may be a new drug modality and gene delivery system for the treatment of autoimmune diseases, including MS.
This research was supported by the AJINOMOTO Co. INNOVATION ALLIANCE PROGRAM (AIAP) 2019 and grants from the Grants-in-Aid for Scientific Research program (JSPS KAKENHI Grant Number: B; 21H02614, C; 18K06593).
The author declares no conflict of interest.
This article contains supplementary materials.
