Nicotinamide N-Methyltransferase in the Inflammatory Pathogenesis of Graves’ Orbitopathy

Nicotinamide N-methyltransferase (NNMT) has been implicated in inflammatory autoimmune disease pathogenesis, although its pro-inflammatory role in Graves’ orbitopathy (GO) is unclear. Therefore, we investigated the influence and mechanisms of NNMT in GO inflammation.

NNMT mRNA expression levels were higher in GO orbital tissues than in healthy orbital tissues. Tissues from patients with type Ⅱ GO showed higher NNMT expression than those with type Ⅰ GO. Pro-inflammatory stimulation induced NNMT expression in dose- and time-dependent manners. NNMT siRNA and antagonists attenuated the expression of pro-inflammatory cytokines (IL-6, IL-8, and monocyte chemotactic protein-1), cyclooxygenase-2, and prostaglandin E2 in orbital fibroblasts. NNMT silencing downregulated the active forms of intracellular signaling molecules (extracellular signal-regulated kinase, c-Jun-terminal kinase, and p38).

Reagents and Chemicals Dulbecco's modified Eagle's medium (DMEM), fetal bovine serum (FBS), penicillin, and gentamicin were purchased from Hyclone Laboratories, Inc. (Logan, UT, USA). An NNMT small-interfering RNA (siRNA; #111655) and a negative-control siRNA (siCon; #4390843) were obtained from Ambion/Applied Biosystems (Austin, TX, USA). Recombinant human interleukin (IL)-1β was purchased from R&D Systems (Minneapolis, MN, USA). We purchased 3-(4,5-dimethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt (MTS) assay solution from Promega Corporation (Madison, WI, USA). Antibodies against IL-6, IL-8, monocyte chemotactic protein (MCP)-1, and β-actin were obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Antibodies against cyclooxygenase (COX)-2, phosphorylated signal transducer and activator of transcription (STAT)-3 (p-STAT3), total STAT3 (t-STAT3), p-nuclear factor kappa-light-chain-enhancer of activated B cells (p-NF-κB), t-NF-κB, p-protein kinase B (p-Akt), t-Akt, p-extracellular signal-regulated kinase (p-ERK), t-ERK, p-c-Jun-terminal kinase (p-JNK), t-JNK, p-p38, and t-p38 were purchased from Cell Signaling Technology (Beverly, MA, USA). Two small molecule NNMT inhibitors, namely 5-amino-1-methylquinolinium iodide (5-amino-1MQ, SML2832) and 1-methylnicotinamide chloride (1-MNA, SML0704), were purchased from Sigma-Aldrich, Inc. (St. Louis, MO, USA).

NNMT expression in orbital tissues from individuals with or without GO were assessed via reverse transcription-quantitative PCR (RT-qPCR), following a previously described method. 28 RNA was isolated from orbital tissues (type Ⅰ GO, n = 10; type Ⅱ GO, n = 9; and non-GO, n = 19) using the TRIzol reagent (Invitrogen, Carlsbad, CA, USA). The sequences of primers targeting the NNMT and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) genes are shown in Supplementary Table S2 . Complementary DNA was synthesized using the RNeasy Lipid Tissue Mini Kit (Qiagen, Valencia, CA, USA). PCR was performed using TaqMan Universal PCR Master Mix and a QuantStudio 3 Real-Time PCR thermocycler (Applied Biosystems, Carlsbad, CA, USA). The PCR results were evaluated using the 2 −ΔΔCt method to determine relative exponential changes by comparing threshold cycle (Ct) values, normalized to the Ct value for GAPDH. Each RT-qPCR experiment was conducted in duplicate.

Western blotting was performed on orbital fibroblasts to evaluate changes in NNMT expression after inflammatory stimulation and to assess the inhibition of pathogenic GO-inflammation mechanisms following NNMT silencing. Only orbital fibroblasts from patients with type Ⅱ GO were used as GO cells in all Western blot experiments. Western blotting was performed as previously described. 28 , 31 Orbital fibroblasts treated with different reagents were rinsed with phosphate-buffered saline and then lysed in cell-lysis buffer containing 1 mM phenylmethylsulfonyl fluoride (PMSF), 50 mM NaF, 20 mM HEPES (pH 7.2), 10% glycerol, 0.1 mM dithiothreitol, 1 mg/mL leupeptin, 1 mg/mL pepstatin, 10 mM Na 3 VO 4 , and 1% Triton X-100 for 30 minutes on ice. Cell lysates were centrifuged at 13,000 g to obtain homogeneous cell fractions, which were subsequently boiled in lysis buffer. Proteins were resolved via 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis, transferred to polyvinylidene fluoride membranes (Immobilon; Millipore Corp., Billerica, MA, USA), and incubated overnight with primary antibodies in Tris-buffered saline with Tween-20. Immunoreactive bands were detected using a horseradish peroxidase-conjugated secondary antibody. The bound peroxidase was visualized via chemiluminescence (Amersham Pharmacia Biotech, Inc., Piscataway, NJ, USA) and exposed to X-ray film (Amersham Pharmacia Biotech, Inc.). The relative amounts of protein in each band were quantified via densitometry after normalization to the β-actin levels in the same sample.

We conducted 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assays to evaluate the effects of siNNMT and 2 different NNMT inhibitors on the viability of orbital fibroblasts. For siNNMT, siNNMT- and siCon-transfected orbital fibroblasts from individuals with or without GO were incubated for 24, 48, or 72 hours in medium supplemented with 10% FBS and antibiotics. Thereafter, MTT solution was added and each plate was incubated again for 4 hours under the same conditions. For NNMT inhibitors, primary cultured orbital fibroblasts were seeded on 24-well culture plates and treated with various concentrations of each NNMT inhibitor for 24 hours. Subsequently, MTS solution was added, and the plate was incubated for an additional 4 hours under identical conditions. Dye absorbance was measured at 490 nm using an ELISA plate reader (EL 340 Biokinetics Reader; Bio-Tek Instruments, Winooski, VT, USA). Cell viabilities were calculated as a percentage relative to untreated control cells.

GO Tissues Showed Increased NNMT Expression To evaluate NNMT expression, whole orbital tissue explants were obtained from patients with type Ⅰ GO, type Ⅱ GO, and from individuals without GO, and the relative NNMT mRNA expression levels were analyzed via RT-qPCR ( Fig. 1 ). NNMT transcript levels in type Ⅱ GO tissues were significantly higher than those in non-GO control tissues ( P < 0.001) and type Ⅰ GO tissues ( P = 0.021). Figure 1. NNMT mRNA expression in GO orbital tissues. Orbital tissues were obtained from patients with GO ( n = 19) and healthy subjects without GO ( n = 19). Patients with GO from whom orbital tissues were obtained were divided into two groups based on CT scans: those with type Ⅰ GO (non-infiltrative type and lipogenic type, n = 9) and those with type Ⅱ GO (infiltrative type and myogenic type, n = 10). The mRNA-expression levels of NNMT were compared via RT-qPCR. Differences among the three groups were assessed using Kruskal-Wallis test (χ 2 [2, n = 38] = 17.512, P < 0.001), with post hoc analyses using Dunn's method (* P < 0.05, ** P < 0.01). The results are presented as mean ± standard error (SE).

Before evaluating the effect of NNMT on the pathogenesis of GO through NNMT silencing, we analyzed the effects of siNNMT on the viabilities of GO and non-GO orbital fibroblasts. MTT assay showed that siNNMT transfection did not decrease the cell viability of orbital fibroblasts to below 95% ( Supplementary Fig. S1 ). We investigated whether NNMT contributed to inflammatory response in GO and non-GO orbital fibroblasts via Western blotting ( Fig. 3 ). The protein-expression levels of NNMT in GO and non-GO fibroblasts demonstrated that siNNMT successfully silenced NNMT expression relative to siCon. The protein levels of pro-inflammatory cytokines (IL-6, IL-8, and MCP-1), and COX-2 were examined via Western blotting. We found that siNNMT transfection substantially decreased IL-1β-induced IL-6, IL-8, MCP-1, and COX-2 protein expression in GO and non-GO orbital fibroblasts. Cells that were not treated with IL-1β showed no noticeable differences pre- and post-siNNMT treatment. Figure 3. Inhibitory effect of NNMT silencing on the expression of pro-inflammatory cytokines in GO and non-GO orbital fibroblasts. Orbital fibroblasts from ( A ) healthy control and ( B ) patients with GO ( n = 3, each) were transfected with 10 nM siNNMT or siCon, cultured for 48 hours, and then either treated with IL-1β (10 ng/mL) for 72 hours or left untreated. NNMT and pro-inflammatory cytokines such as IL-6, IL-8, MCP-1, and COX-2 were analyzed via Western blotting. The assays were performed in triplicate with cells from three different individuals. The results showed the mean and SD of the relative density ratios for NNMT and pro-inflammatory cytokines (** P < 0.01 versus siCon transfectants).

MTT assays were performed to identify non-toxic doses of NNMT inhibitors (5-amino-1MQ and 1-MNA) for orbital fibroblasts. Orbital fibroblasts from patients with or without GO were treated with small-molecule inhibitors for 24 hours. The 0 to 100 µM range of 5-amino-1MQ and the 0 to 5 mM range of 1-MNA did not affect decrease cell viability below 95% in normal and GO orbital fibroblasts ( Supplementary Fig. S2 ). Pro-inflammatory cytokine expression was evaluated in GO orbital fibroblasts treated with an NNMT inhibitor post-IL-1β stimulation. Western blot analysis revealed that treatment with the NNMT inhibitors, 1-MNA and 5-amino-1MQ, significantly reduced the IL-1β-induced expression of IL-6, IL-8, MCP-1, and COX-2, when compared to its levels in untreated cells ( Fig. 5 ). Figure 5. Inhibitory effect of small-molecule NNMT inhibitors on the expression of pro-inflammatory cytokines in GO orbital fibroblasts. Orbital fibroblast from individuals with GO ( n = 3) were not treated or were pretreated with either 100 µM 5-amino-1MQ or 5 mM 1-MNA for 30 minutes before exposure to 10 ng/mL IL-1β for 24 hours. The expression levels of NNMT and pro-inflammatory cytokines (IL-6, IL-8, MCP-1, and COX-2) were analyzed via Western blotting. The assays were performed in triplicate with cells from three different individuals. The mean and SD of the relative density ratios are shown (** P < 0.01 versus IL-1β treated cultures).

In this study, we examined the roles of NNMT in GO pathogenesis, especially in inflammation. Our results demonstrated that inflammation in orbital fibroblasts led to increased NNMT expression. Silencing NNMT through siNNMT transfection or inhibiting NNMT activity with small molecule inhibitors significantly reduces pro-inflammatory signals, such as the production and secretion of pro-inflammatory cytokines and phosphorylation of signaling molecules. NNMT silencing also inhibited pro-inflammatory intracellular signaling pathways. These data indicate that NNMT inhibition may have therapeutic potential against GO. The main chemical reaction mediated by NNMT is nicotinamide methylation to form methylnicotinamide, with SAM acting as the methyl donor. Therefore, NNMT activity is directly related to nicotinamide levels because NNMT facilitates the removal of excess nicotinamide. Nicotinamide has therapeutic potential for GO, as evidenced by both in vitro and clinical studies. Previous in vitro data showed that nicotinamide treatment inhibited superoxide-induced fibroblast proliferation. 22 Superoxide radicals are known to promote orbital fibroblast proliferation and differentiation. 32 – 34 The results of another study further revealed that nicotinamide inhibited the induction of various cell-surface proteins, such as intercellular adhesion molecule-1 and human leukocyte antigen-DR, which are typically induced by cytokines interferon gamma and tumor necrosis factor alpha in GO fibroblasts. Nicotinamide delayed GO progression in a clinical study. 35 Furthermore, data from clinical trials showed that nicotinamide treatment relieved the clinical symptoms of GO, as assessed using the NOSPECS total eye-score system. 24 Collectively, the results of these studies have indicated that nicotinamide, an NNMT substrate, can potentially be used as an antioxidant to treat GO. NNMT has been extensively studied as a regulator of NAD + levels, especially in the context of autoimmune diseases. In the case of IBD, treatment-related changes in NAD + levels correlated with shifts in NNMT expression. 19 Several previous reports showed that NAD + levels were diminished in patients with MS, and that the extent of this decrease correlated with disease severity. 36 , 37 In vivo data derived using mice with experimental autoimmune encephalomyelitis (EAE), a well-established animal model of MS, showed that maintaining higher NAD + levels through NNMT inhibition is a beneficial mechanism for current MS treatment. 20 The results of another study of mice with EAE showed that NAD + treatment inhibited inflammasome formation, modulated T cell differentiation, reduced the levels of pro-inflammatory factors, and increased expression of anti-inflammatory cytokine, IL-10. 21 Although the pathways linking NNMT to inflammation may differ slightly across diseases, findings with IBD and MS suggested that inhibiting NNMT to regulate NAD + levels is a promising therapeutic approach. To the best of our knowledge, this study is the first to provide insights on the role of NNMT in GO pathogenesis. Previously, we performed RNA sequencing to compare GO and non-GO orbital tissues, revealing a 2.55-fold increase in NNMT gene expression and 1.32-fold decrease in nicotinamide nucleotide adenylyltransferase 3 expression in the GO orbital tissues. 38 Therefore, we compared NNMT-expression levels between GO and non-GO tissues and found that NNMT expression was significantly higher in type Ⅱ GO orbital tissues than in non-GO or type I GO orbital tissues. Owing to type Ⅱ GO's association with greater inflammation and more severe outcomes, such as restrictive myopathy and diplopia, NNMT may play an important role in the changes such as inflammation, primary symptoms, and more severe sequelae of type Ⅱ GO. Furthermore, this insight into NNMT is an important clue for identifying unknown factors that might precede changes to NNMT in distinguishing between type Ⅰ and type Ⅱ GO. Orbital fibroblasts form patients with GO are continuously exposed to pro-inflammatory stimuli from immune cells infiltrating into orbital tissues. Therefore, treating orbital fibroblast cultures with pro-inflammatory stimulants such as IL-1β is useful for observing their response in an inflammatory condition, similar to in vivo event. IL-1β induces stronger production of pro-inflammatory cytokines and consequent characteristic remodeling in orbital fibroblasts from patients with GO, 39 and IL-1β has been used as an effective pro-inflammatory stimulant in many in vitro GO studies. 29 – 31 Our data showed that the inflammatory condition induced by IL-1β increased NNMT levels in orbital fibroblasts in time- and dose-dependent manners. Thus, we can assume that NNMT contributes to the inflammatory mechanism of GO in orbital tissues. Cytokines are essential for GO studies, as various types of T cells and the cytokines they release drive the primary immune response in the local orbital environment. 1 , 2 ,