S-allylmercaptocysteine inhibits TLR4-mediated inflammation through enhanced formation of inhibitory MyD88 splice variant in mammary epithelial cells

Abstract

Mastitis is an inflammatory disease affecting mammary tissues caused by bacterial infection that negatively affects milk quality and quantity. S-Allylmercaptocysteine (SAMC), a sulfur compound in aged garlic extract (AGE), suppresses lipopolysaccharide (LPS)-induced inflammation in mouse models and cell cultures. However, the mechanisms underlying this anti-inflammatory effect remain unclear. In this study, we demonstrated that oral administration of AGE suppressed the LPS-induced immune response in a mastitis mouse model and that SAMC inhibited LPS-induced interleukin-6 production and nuclear factor κB p65 subunit activation in HC11 mammary epithelial cells. Global phosphoproteomic analysis revealed that SAMC treatment downregulated 910 of the 1,304 phosphorylation sites upregulated by LPS stimulation in mammary cells, including those associated with toll-like receptor 4 (TLR4) signaling. Additionally, SAMC decreased the phosphorylation of 26 proteins involved in pre-mRNA splicing, particularly the U2 small nuclear ribonucleoprotein complex. Furthermore, we found that SAMC increased the production of the myeloid differentiation factor 88 short form (MyD88-S), an alternatively spliced form of MyD88 that negatively regulates TLR4 signaling. These findings suggest that SAMC inhibits TLR4-mediated inflammation via alternative pre-mRNA splicing, thus promoting MyD88-S production in mammary epithelial cells. Therefore, SAMC may alleviate various inflammatory diseases, such as mastitis, by modulating immune responses.

Fig. 1

Effect of S-allylmercaptocysteine (SAMC) on the induction of pro-inflammatory cytokines and chemokines and activation of NF-κB by lipopolysaccharide (LPS). (a) The chemical structure of SAMC. (b) IL-6 secretions in cultured HC11 cells treated with SAMC (10-300 μM) and LPS (100 ng/mL) for 24 h were measured by ELISA. n = 4. (c) mRNA expression of pro-inflammatory cytokines (Il6 and Tnfa) and chemokines (Cxcl1 and Ccl2) in HC11 cells treated with LPS and SAMC (300 μM) for 1 h was examined by real-time quantitative PCR. n = 4-5. (d) Phosphorylation level of NF-κB p65 were examined by immunoblotting in HC11 cells upon LPS stimulation (1 μg/mL) for 10-90 min and SAMC treatment for 10-90 min. Data are shown as mean±SD, n = 3. ** denotes significant difference (p < 0.01). N.D. means not detected.

Fig. 2

Effects of S-allylmercaptocysteine (SAMC) on phosphoproteome in HC11 cells treated with lipopolysaccharide (LPS). (a) Workflow of phosphoproteomic analysis. (b) The proportion of phosphorylated amino acid residues in detected phosphopeptides (pSer = phosphorylated serine, pThr = phosphorylated threonine, pTyr = phosphorylated tyrosine). (c) The proportion of phosphopeptides increased by LPS stimulation (≥ 1.5-fold, 10.7%) and the proportions of these phosphopeptides increased (≥ 1.5-fold, 5.2%) or decreased (≤ 0.75-fold, 69.8%) by SAMC treatment. (d, e) The heat map of the phosphopeptides affected by LPS and SAMC in Fig. 2c (d) and the phosphorylation level of protein kinases in the TLR signaling pathway (e). LPS and SAMC columns show the log2 fold changes of the phosphopeptide abundance in LPS-treated cells vs. control and in LPS- and SAMC- treated cells vs. LPS alone-treated cells, respectively.

Fig. 3

S-Allylmercaptocysteine (SAMC) down-regulates the phosphorylation of mRNA splicing-related proteins. (a-c) Gene ontology term annotation for cellular component (a) and biological process (b) and Wikipathway analysis (c) of the phosphoproteins increased by lipopolysaccharide (LPS) treatment (≥ 1.5-fold vs. control) and decreased by SAMC treatment (≤ 0.75-fold vs. LPS group). (d) MA plot of the phosphoproteins downregulated by SAMC. (e) Protein-protein interactions among the spliceosome-related proteins in (d) were represented by Markov Cluster Algorithm clustering performed by STRING database. The proteins associated with U2 snRNP and other spliceosomes excepting U2 snRNP were shown as blue and red spheres, respectively.

Fig. 4

S-Allylmercaptocysteine (SAMC) down-regulates the phosphorylation of various kinases in HC11 cells. (a) The heatmap of phosphorylation of kinases increased by lipopolysaccharide (LPS) (≥ 1.5-fold). LPS and SAMC columns showed the log₂ fold changes of the phosphopeptide abundance in LPS vs. control and SAMC vs. LPS, respectively. (b) Protein-protein interactions among the kinases increased by LPS (≥ 1.5-fold) and decreased by SAMC (≤ 0.75-fold) represented by Markov Cluster Algorithm (MCL) clustering. TLR-related kinases are shown as red spheres, and CDK family is shown as yellow spheres. (c) Protein-protein interactions among the spliceosome-related proteins in Fig. 3f and CDKs family in Fig. 4b represented by MCL clustering. The proteins associated with U2 snRNP, other spliceosome (except U2 snRNP), and CDKs family are shown as green, red, and blue spheres, respectively.

Fig. 5

S-Allylmercaptocysteine (SAMC) induces the gene expression of MyD88-S. (a) Real-time qPCR analysis of the effect of SAMC (300 µM) on mRNA expression of Myd88s in HC11 cells upon lipopolysaccharide (LPS) stimulation (100 ng/mL) for 30–120 min. (b, c) Real-time qPCR analysis of the concentration-dependent effect of SAMC (10–300 µM) on mRNA expression of Myd88s (b) and canonical Myd88l (c) in HC11 cells treated with LPS for 2 h. Data are shown as mean ± SD, n = 4–5. ** denotes significant difference (p < 0.01) and * denotes significant difference (p < 0.05).