METTL3/ALKBH5-Mediated N6-Methyladenosine Modification Drives Macrophage M1 Polarization via the SLC15A3-TASL-IRF5 Signaling Axis in Psoriasis

Abstract

Impaired N6-methyladenosine (m⁶A) modification has been implicated in regulating various inflammatory diseases, but its role in psoriasis remains unclear. Here, m⁶A modification and its methyltransferase METTL3 are revealed to be upregulated in psoriatic macrophages, while the demethylase ALKBH5 is downregulated. Conditional knockout of Mettl3 in macrophages alleviated psoriasis-like symptoms in mice, whereas knockout of Alkbh5 exacerbated them. Both in vivo and in vitro, Mettl3 deficiency inhibited IMQ-induced M1 macrophage polarization, while Alkbh5 deficiency promoted M1 polarization. The regulation of macrophage polarization by m⁶A is likely mediated by targeting Slc15a3. SLC15A3 enhances the recruitment of TASL, a recently identified endolysosomal IRF5 adaptor, which functions similarly to the IRF3 adaptors STING and MAVS at the endoplasmic reticulum (ER) and mitochondria, respectively, to augment IRF5 signaling via SLC15A4. The findings underscore the critical role of m⁶A RNA modification in psoriasis pathogenesis and unveil a novel regulatory mechanism of TASL-IRF5 signaling through m⁶A modification, suggesting potential new therapeutic targets for psoriasis treatment.

Keywords: N6‐methyladenosine; SLC15A3; TASL; macrophage polarization; psoriasis.

Figure 1

Mettl3 deficiency in macrophages alleviates psoriasis‐like phenotype in mice. A) Dot blot of m⁶A (250, 125, or 62.5 ng total RNA) in the BMDMs from Mettl3^f/f;Lyz2‐Cre^±   and Mettl3^f/f mice. B) m⁶A (red) and F4/80 (green) co‐immunofluorescence staining in IMQ‐induced lesions of WT and Mettl3 cKO mice (left panel), and statistical analysis of m⁶A fluorescence intensity in F4/80⁺ cells (right panel). Arrows indicate F4/80⁺ macrophages. Scale bar, 50 µm. Each dot represents one cell, 3 mice for each group. C) Psoriasis area and severity index (PASI score) in IMQ‐induced lesions of WT and Mettl3 cKO mice (n = 3). D) H&E staining of non‐lesions and IMQ‐induced psoriasis‐like lesions of WT and Mettl3 cKO mice (n = 5). Left, representative picture; right, statistics of epidermal thickness. Scale bar, 100 µm. E) EdU (red) staining of IMQ‐induced psoriasis‐like lesions (n = 4) of WT and Mettl3 cKO mice. Statistical analysis of the numbers of EdU⁺ cells in epidermis. Scale bar, 100 µm. F) Flow cytometry of immune cell infiltration, including CD45⁺ immune cells, γδT cells, and Langerhans cells (non‐lesion n = 3, lesion n = 4). The statistical data are shown in the right panels. G) The RNA expression of cytokines was quantified by qPCR in non‐lesions (n = 3) and IMQ‐induced psoriasis‐like lesions (n = 4) of WT and Mettl3 cKO mice. Each dot represents one mouse. An unpaired t test was used for statistical analysis. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 2

Alkbh5 deficiency in macrophages exacerbates psoriasis‐like phenotype in mice. A) Dot blot of m⁶A (250, 125, or 62.5 ng total RNA) in the BMDMs from Alkbh5^f/f;Lyz2‐Cre^±   and Alkbh5^f/f mice. B) m⁶A (red) and F4/80 (green) co‐immunofluorescence staining in IMQ‐induced lesions of WT and Alkbh5 cKO mice (left panel), and statistical analysis of m⁶A fluorescence intensity in F4/80⁺ cells (right panel). Arrows indicate F4/80⁺ macrophages. Scale bar, 50 µm. Each dot represents one cell, 3 mice for each group. C) Psoriasis area and severity index (PASI score) in IMQ‐induced lesions of WT and Alkbh5 cKO mice (n = 3). D) H&E staining of non‐lesions (n = 3) and IMQ‐induced psoriasis‐like lesions (n = 5) of WT and Alkbh5 cKO mice. Left, representative picture; right, statistics of epidermal thickness. Scale bar, 100 µm. E) EdU (red) staining of IMQ‐induced psoriasis‐like lesions of WT and Alkbh5 cKO mice (n = 5). Statistical analysis of the numbers of EdU⁺ cells in epidermis. Scale bar, 100 µm. (F) Flow cytometry of immune cell infiltration, including CD45⁺ immune cells, γδT cells, and Langerhans cells (non‐lesion n = 3, lesion n = 4). The statistical data are shown in the right panels. G) The RNA expression of cytokines was quantified by qPCR in non‐lesions (n = 3) and IMQ‐induced psoriasis‐like lesions (n = 5) of WT and Alkbh5 cKO mice. Each dot represents one mouse. An unpaired t test was used for statistical analysis. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 3

The N6‐methyladenosine in macrophages promotes M1 polarization. In vivo, IMQ‐induced psoriatic lesions were enzymatically digested into single‐cell suspensions, followed by flow cytometric analysis. In vitro, bone marrow‐derived macrophages (BMDMs) were isolated and subsequently stimulated with IMQ (2ug ml⁻¹). A) The proportion of macrophages in IMQ‐induced psoriasis‐like lesions in mice (Up n = 4, down n = 6). The statistical data are shown in the right panels. (B) Proportion of M1 macrophages in IMQ‐induced psoriasis‐like lesions in mice (n = 3). The statistical data are shown in the right panels. C) Flow cytometry examines the percentages of M‐CSF induced macrophages from WT, Alkbh5 cKO and Mettl3 cKO bone marrow‐derived monocytes (n = 5). The statistical data are shown in the right panels. D) Flow cytometry examines the percentages of M1 macrophages from WT, Alkbh5 cKO, and Mettl3 cKO mice (n = 3). The statistical data are shown in the right panels. E) Flow cytometric examines reactive oxygen species (ROS) in macrophages from psoriatic skin lesions and in IMQ‐stimulated bone marrow‐derived macrophages (BMDMs) of WT, Alkbh5 cKO, and Mettl3 cKO mice (n = 3). The statistical analysis was shown in the right panels. Each dot represents one sample. F) The RNA expression of cytokines Il23p19, Il6 and Tnfα are quantified by qPCR in WT, Alkbh5 cKO, and Mettl3 cKO BMDMs (n_IMQ‐ = 3, n_IMQ+ = 5). G) Mettl3/Alkbh5 rescues the expression of Il23p19, Il6, and Tnfa. The wild‐types and catalytic‐activity‐dead mutants (Mettl3 ^D395A&W398A, Alkbh5 ^H205A) of Mettl3/Alkbh5 were delivered with lentiviral vectors into Mettl3‐KO and Alkbh5‐KO BMDMs, respectively (n = 3). The mRNA of Mettl3, Alkbh5, Il23p19, Il6, and Tnfa was examined by qRT‐PCR. Each dot represents one mouse. An unpaired t test was used for statistical analysis. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 4

Slc15a3 is modified by m⁶A in macrophages. A) Principal component analysis of the transcriptomes of BMDMs isolated from WT group (gray), Alkbh5 cKO group (orange), and Mettl3 cKO group (green) with 2ug ml⁻¹ IMQ treatment 4 h (n = 2). B) Heatmap of genes with differential m⁶A peaks based on MeRIP‐seq data. C) Read density of m⁶A in Slc15a3 transcript in WT, Alkbh5 cKO, and Mettl3 cKO BMDMs. D,E) qPCR and MeRIP‐qPCR analysis showed Slc15a3 RNA level and m⁶A enrichment in WT, Alkbh5 cKO and Mettl3 cKO macrophages in psoriatic lesions (D) and BMDMs (E), respectively. F) Insertion of the identified Slc15a3 m⁶A sites, but not its mutated versions, reduces the activity of a luciferase reporter in Raw264.7 and THP‐1 cells (n = 3). G) Co‐Immunoflurosecence staining of Slc15a3 (red) and F4/80⁺ (green). Arrows indicate F4/80⁺ macrophages. Each dot represents one cell, 3 mice for each group. Scale bar, 50 µm. The statistical data are shown in the right panel. H) Decay examination of Slc15a3 mRNA (n = 3). Three group BMDMs were treated with Actinomycin D. I,J) Ythdf1 knockdown reduces the expression of Slc15a3, Il23p19, Il6, and Tnfa in BMDMs with IMQ treatment. Each dot represents one mouse. K) (Upper panel) RNA immunoprecipitation (RIP)‐qPCR analysis of Slc15a3 mRNA enrichment in YTHDF1‐IP fractions from Raw264.7 macrophages transfected with wild‐type Slc15a3 (WT) or its m⁶A site mutant (A‐to‐C). (Lower panel) RIP‐qPCR analysis of Slc15a3 mRNA in YTHDF1‐IP fractions from WT, Mettl3‐KO, and Alkbh5 KO bone BMDMs treated with IMQ (2 µg/mL, 4 h). Data are normalized to input RNA levels and presented as fold change relative to WT (mean ± SEM, n = 3). An unpaired t test was used for statistical analysis. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 5

Slc15a3 is a key mediator for M1 polarization in Mettl3 and Alkbh5 deficient macrophages. A) Mettl3/Alkbh5 rescues the expression of Slc15a3. The wild‐types and catalytic‐activity‐dead mutants (Mettl3 ^D395A&W398A, Alkbh5 ^H205A) of Mettl3/Alkbh5 were delivered with lentivirus vectors in Mettl3‐KO and Alkbh5‐KO BMDMs, respectively. Each dot represents one mouse (n = 3). The mRNA of Slc15a3 was examined by qRT‐PCR. B,C) Slc15a3 mediates IMQ‐induced expression of Il23p19, Il6, and CD86. Two independent siRNAs targeting Slc15a3 were transfected into Raw264.7 macrophages, and the mRNA expression levels of Il23p19 and Il6 were examined by qRT‐PCR (B), and the cell membrane expression of CD86 was examined by FACS (C). n = 3. Slc15a3 mediates M1 polarization. D,E) Slc15a3 functions downstream of Mettl3 and Alkbh5 to mediate IMQ‐induced M1 polarization. Mettl3 KO BMDMs were infected with lentiviral vector encoding Slc15a3 and Alkbh5 KO BMDMs were transfected with Slc15a3 specific siRNA, The proportion of M1 macrophages, expressing the CD86 marker on the cell surface (D) and the mRNA expression of Slc15a3, Il23p19 and Il1b (E) were detected by flow cytometry and qRT‐PCR in WT, WT+IMQ, Mettl3 KO‐Slc15a3 OE +IMQ, Alkbh5 KO‐Slc15a3 KD+IMQ groups (n = 3). An unpaired t test was used for statistical analysis. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 6

Slc15a3 promotes endolysosomal localization of TASL to activate TASL‐IRF5 signal. A–C) Protein‐protein interactions between Slc15a3, Slc15a4, and TASL. Co‐Immunoprecipitation assays were performed with cell extracts from the HEK239T cells transfected with the plasmids encoding the indicated genes. D) Slc15a3 promotes endolysosomal localization of TASL. The plasmids encoding Slc15a3‐BFP (blue) and Tasl‐Aausfp1 (green) were transfected into Raw264.7 macrophages. Lysotracker (red) was used for endolysosome labeling. Each dot represents one lysosome. Scale bar, 10 µm. E) m⁶A modification mediates IMQ‐ or R848‐stimulated IFR5 activation. Western blot showed the expression of phosphorylated IRF5 (p‐IRF5), total IRF5, and GAPDH in WT, Alkbh5 KO, and Mettl3 KO BMDMs treated with TLR7 ligands IMQ or R848. F) Slc15a3 functions downstream of m⁶A modification to mediate IMQ‐stimulated IRF5 activation. Western blot showed the expression of p‐IRF5, IRF5, and GAPDH in WT, Alkbh5 KO, and Mettl3 KO BMDMs transfected with the siRNA targeting Slc15a3 or the plasmid encoding Slc15a3. G,H) Flow cytometry of IMQ induced M1 polarization and the mRNA expression of Il23p19 and Tnfa in WT, Mettl3‐KO, and Alkbh5‐KO BMDMs. C5 was employed to inhibit IRF5 signaling activated by TASL. Flow cytometry was used to assess the proportion of CD86⁺ macrophages in IMQ‐stimulated bone marrow‐derived macrophages (BMDMs) with C5 treated 24 h (G). In parallel, quantitative PCR (qPCR) was performed to evaluate the expression levels of Il23 and Tnfa following IMQ stimulation (H). The data represent one out of three independent biological replicates. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.

Figure 7

The METTL3/ALKBH5‐m⁶A‐SLC15A3 axis in macrophages is associated with the severity of psoriasis. A) Immunofluorescence staining shows the expression levels of m⁶A/ALKBH5/METTL3 (red) in CD68⁺ macrophages (green) of healthy skin and psoriatic lesions. Each dot represents one cell, 3 mice for each group. Scale bar, 50 µm. The corresponding fluorescence intensity statistical analysis is shown in the right panels. Each dot represents one microscope field of view from 6 healthy people and 8 psoriasis patients. An unpaired t test was used for statistical analysis. B) IGV visualization shows the m⁶A peaks located on SLC15A3 transcripts in healthy skin and psoriatic lesions (data analysis based on the public database, GSE155702). (C) MeRIP‐qPCR and qRT‐PCR reveal the m⁶A modification at the 3’‐UTR and the mRNA expression levels of SLC15A3 in the skin and peripheral monocytes from healthy people and psoriasis patients. D) Immunofluorescence staining shows the expression levels of SLC15A3 (red) in CD68⁺ macrophages (green) of healthy skin and psoriatic lesions. The experiments were performed as the same as those in A. E) Pearson correlation analysis shows the positive correlation between the PASI scores and m⁶A Mean Fluorescence Intensity (MFI), m⁶A modification of SLC15A3, and the relative expression of SLC15A3 in CD68⁺ macrophages. Each dot represents one clinical sample. F) The proportion of M1‐polarized macrophages. Peripheral blood mononuclear cells (PBMCs) were isolated from healthy donors (n = 3) and psoriasis patients (n = 8), followed by monocyte separation and differentiation into macrophages. During IMQ‐induced M1 polarization, macrophages were treated with the METTL3 inhibitor STM2457. The statistical analysis was shown in the right panels. Each dot represents one sample. ^* P < 0.01, ^** P < 0.01, ^*** P < 0.001.