MYMD-1 is a synthetic derivative of tobacco alkaloids, compounds that possess immunoregulatory properties and have been linked to the epidemiological observation that smoking reduces the odds of developing thyroid Abs and hypothyroidism. To assess the effect and mechanism(s) of the action of MYMD-1, we chose the NOD.H-2h4 mouse model of spontaneous thyroiditis. We began in vitro using T cells isolated from NOD.H-2h4 spleens and found that MYMD-1 suppressed TNF-α production by CD4+ T cells in a dose-dependent manner. We then treated 58 NOD.H-2h4 mice for 12 wk with either unsupplemented water that contained (10 mice) or did not contain (16 mice) MYMD-1 (185 mg/l) or water supplemented with sodium iodide (500 mg/l) that contained (16 mice) or did not contain (16 mice) MYMD-1. Mice were bled at baseline and then every 2 wk until sacrifice. MYMD-1 decreased the incidence and severity (p < 0.001) of thyroiditis, as assessed by histopathology. Similarly, the number of CD3+ T cells and CD19+ B cells infiltrating the thyroid was dampened by MYMD-1, as assessed by flow cytometry. Interestingly, the subset of thyroidal CD3+CD4+Tbet+RORγT− effector Th1 cells and the systemic levels of TNF-α were decreased by MYMD-1. Serum thyroglobulin Abs decreased in the MYMD-1 group. Thyroid hormones did not differ among the four groups, whereas thyroid-stimulating hormone increased upon iodine supplementation but remained normal in MYMD-1–treated mice. Overall, the study suggests that MYMD-1 ameliorates thyroiditis acting on specific lymphoid subsets. Further studies, including other models of autoimmunity, will confirm the potential clinical use of MYMD-1 as a novel immunometabolic regulator.
Autoimmune thyroiditis, also known as Hashimoto thyroiditis or chronic lymphocytic thyroiditis, is characterized by lymphocytic infiltration of the thyroid gland, the presence of circulating Abs against thyroid-specific Ags (thyroperoxidase and thyroglobulin), and varying degrees of thyroid dysfunction (1). Risk factors for thyroiditis are white race, female (F) sex, older age, and increased consumption of iodine (2). Tobacco smoking, on the contrary, seems to have a protective effect, as suggested by numerous epidemiological studies showing that smokers have a lower prevalence of thyroid autoantibodies than nonsmokers (reviewed in Supplemental Table IV in Ref. 3). A similar protective effect of smoking is reported for ulcerative colitis, in which smokers are less likely to require colectomy (4). Overall, the mechanism(s) through which smoking confers apparent protection against some autoimmune diseases remains unclear.
Autoimmune thyroiditis has been modeled in animals since the mid-1950s by injection of whole thyroid extracts (5), purified thyroglobulin (6), or purified thyroperoxidase (7). More recently, thyroiditis has been described as developing spontaneously in a congenic mouse model, the NOD.H-2h4 strain, which was derived from the diabetes-prone NOD mice. NOD.H-2h4 mice carry the thyroiditis-susceptible K allele at the MHC class II A locus (Kk, Ak, E0, and Db) and develop thyroiditis spontaneously while losing the type 1 diabetes typical of the parental strain (8). The NOD.H-2h4 thyroiditis incidence is greatly accelerated by the addition of iodine to the drinking water (9) and characterized by infiltration of the thyroid gland with CD4+, CD8+, and B lymphocytes, disruption of the normal thyroid architecture, and development of thyroglobulin Abs followed by thyroperoxidase Abs (reviewed in Ref. 10). In contrast to the human counterpart, the NOD.H-2h4 thyroiditis does not differ between males (M) and F and seems not to cause thyroid dysfunction, although this aspect has been predominantly assessed by measuring the thyroid hormone thyroxine (T4) rather than the pituitary hormone thyroid-stimulating hormone (TSH).
Lymphoid subsets infiltrating the thyroid gland, cytokines, and environmental factors such as exposure to smoking have all been investigated in the pathogenesis of thyroiditis. CD4+ T cells expressing the α/β TCR are considered the main disease mediators because crossing NOD.H-2h4 to mice lacking the α/β TCR or administering CD4+-depleting Abs completely inhibit thyroiditis development (9, 11). Thyroid-infiltrating CD4+ T cells recognize specific iodinated thyroglobulin peptides and release IFN-γ upon activation (12), suggesting that Th1 subsets mediate important effector functions in disease pathogenesis. Furthermore, the increase in thyroidal IL-17 transcripts following iodine administration and the reduction in disease incidence and severity upon crossing to IL-17–deficient mice suggest a pathogenic contribution of the Th17 subset (13). Regulatory T cells have also been reported in the thyroid lymphocytic infiltrate, but their role remains to be defined (10). B cells are important for their ability to produce autoantibodies, present self-antigens to T cells, and regulate other lymphoid populations. Thyroglobulin Abs correlate significantly with the severity of thyroid lesions and are therefore used as disease markers, although there are NOD.H-2h4 mice that develop thyroiditis in the absence of detectable serum Abs (9). B cells, in particular follicular B cells that infiltrate the thyroid gland, are also fundamental for their role as professional APCs. When depleted by anti-CD20 treatment, in fact, the absence of B cells prevents the development or attenuates the progression of thyroiditis (14). Lastly, B cells of the regulatory phenotype (CD19+CD5+CD1dhi) decrease upon iodine administration, correlating negatively with the proportion of effector Th17 cells, thus suggesting an important pathogenic role in thyroiditis (15).
In addition to IFN-γ and IL-17 discussed above, TNF-α is known to be produced by the thyroid-infiltrating CD4+ T cells and B cells and contributes to the formation of germinal centers within the thyroid gland (14). TNF-α and TGF-β promoted hyperplasia and the proliferation of thyrocytes in IFN-γ −/− NOD.H-2h4, resulting in a chronic state of thyroid dysfunction and fibrosis (9).
Tobacco smoking has numerous effects on thyroid physiology and disease (16). It reduces the odds of developing autoimmune hypothyroidism, thyroperoxidase Abs, and thyroglobulin Abs, an effect that is dose dependent and transient (it disappears within 3 y of smoking cessation) (16). Given the complexity of the tobacco smoking mixture, it has been difficult to identify the compounds endowed with beneficial immunoregulatory properties, although some studies have focused on the alkaloids, small molecules produced by plants of the Solanaceae family in response to attacks from insect herbivores. For example, the alkaloid harmine specifically enhances the differentiation of regulatory T cells and inhibits that of Th17 cells, at least in part by dampening the activity of the kinase DYRK1A (17). The alkaloid nicotine possesses immunosuppressive properties mainly mediated by activation of the cholinergic pathway (18) but cannot be used therapeutically because of its side effects: short t1/2, lack of receptor specificity, and addictive properties (19). The alkaloid anatabine ameliorated disease in a mouse model of autoimmune thyroiditis induced by thyroglobulin (20). Nicotine and anatabine, however, similarly nornicotine and anabasine, undergo nitrosylation during the curing process, leading to the formation of tobacco-specific N′-nitrosamines that are carcinogenic (21). On the contrary, myosmine and isonicoteine do not form tobacco-specific N′-nitrosamines in vivo (22) and are thus more amenable to clinical applications. We designed the current study to assess the effect of MYMD-1, a synthetic tobacco alkaloid derivative, in the NOD.H-2h4 mouse model of autoimmune thyroiditis.
Materials and Methods
MYMD-1 was provided by MyMD Pharmaceuticals (Tampa, FL) and dissolved in 95% ethanol to make a stock solution of 1.2 M (corresponding to 185 mg/ml).
In vitro studies of MYMD-1 to assess its effect on cell proliferation and cytotoxicity
Cell proliferation was measured using the BrdU ELISA kit (catalog number ab126556; Abcam, Cambridge, MA) in a mouse pancreatic β cell line (MIN6, a kind gift of Dr. M. Hussain) and macrophage cell line (RAW 264.7 purchased from American Type Culture Collection). Cells were plated in 96-well plates and treated with different concentrations of MYMD-1 (0, 0.05, 0.5, 1, 2, and 4 mM) for 24 h. Cells were then treated with BrdU, according to the manufacturer’s recommendations. Plates were read using the SpectraMax iD3 Microplate Reader (Molecular Devices, Sunnyvale, CA) at a dual wavelength of 450/550 nm. Cytotoxicity was assessed using T cells isolated from normal NOD.H-2h4 splenocytes and the LIVE/DEAD Aqua viability dye (Life Technologies), following 24-h incubation with MYMD-1 at concentrations of 0 (medium alone), 0.05, 0.1, 0.5, 1, and 2 mM.
In vitro studies of MYMD-1 to assess its effect on the cytokines produced by T cells and the differentiation of naive CD4+ T cells into T regulatory cells
Total CD4+ T cells and naive CD4+CD25− T cells were isolated as described (23) and then cultured with different concentrations of MYMD-1 (2, 1, 0.5, 0.1, and 0.05 mM) or medium alone to assess their ability to produce cytokines or differentiate into a regulatory phenotype, respectively. For cell isolation and in vitro cultures, RPMI 1640 medium supplemented with penicillin/streptomycin (final concentration 100 U/ml), FBS (10%), and l-glutamine (2 mM) was used. Briefly, splenocytes (from three to seven NOD.H-2h4 mice in six independent experiments, n = 31 mice) were isolated by dissociation between frosted glass slides, treated with ammonium chloride potassium buffer to lyse the erythrocytes, and washed with PBS. Total CD4+ T cells were isolated using the CD4+ T Cell Isolation Kit (Miltenyi Biotec), according to the manufacturer’s instructions. CD4+ T cells (1 × 106 in 1 ml in 24-well plate) were stimulated with PMA (final concentration 50 mg/ml) and ionomycin (final concentration 1 μg/ml) in the presence of brefeldin A and monensin for 6 h at 37°C. After washing with ice-cold PBS, lymphocytes were stained with LIVE/DEAD Aqua viability dye (Life Technologies) according to the manufacturer’s instructions and then stained with Abs against CD3 (17A2) and CD4+ (GK1.5, both from BioLegend) surface markers for 30 min at 4°C. Cells were then fixed and permeabilized with eBioscience Foxp3/Transcription Factor Staining Buffer Set and stained for intracellular cytokines using Abs against TNF-α (MP6-XT22), IFN-γ (XMG1.2), IL-17A (TC11-18H10.1, all from BioLegend), and IL-4 (11B11, from BD Biosciences) for 30 min at 4°C.
The remaining CD4+ T cells not used for the analysis of the intracellular cytokine expression were processed for naive CD4+ CD25− T cell isolation using the CD25 MicroBead Isolation Kit (Miltenyi Biotec), as described (23). CD4+ CD25− T cells (1 × 106 in 1 ml) were plated into 24-well plates coated with anti-mouse CD3 Ab (10 μg/ml; clone 145-2C11) and then incubated for 5 d at 37°C with recombinant mouse TGF-β1 (5 ng/ml; R&D Systems), an Ab against CD28 (2 μg/ml; clone 37.51; both from BD Biosciences), and the same concentrations of MYMD-1 used above. T cells were then collected, washed with ice-cold PBS, stained with LIVE/DEAD Aqua viability dye, and surface stained for 30 min at 4°C with Abs against CD3 (17A2), CD4 (GK1.5, both from BioLegend), and CD25 (eBio7D4 [7D4], from Thermo Fisher Scientific). Following fixation and permeabilization with eBioscience Foxp3/Transcription Factor Staining Buffer Set, T cells were stained for 30 min at 4°C with an Ab against Foxp3 (FJK-16s, from eBioscience). Cells were acquired on BD LSR II flow cytometer (BD) and data analyzed using FlowJo software (Tree Star).
In vivo studies to assess MYMD-1 toxicity
We first assessed in vivo MYMD-1 toxicity using a cohort of 58 adult (21 ± 10 wk of age) mice of mixed strain and sex (29 M and 29 F). Mice were injected i.p. with MYMD-1 doses of 300 (n = 5 mice), 190 (n = 24), 95 (n = 10), 47.5 (n = 5), or 0 (n = 14) mg/kg, monitored for a minimum of 24 h postinjection, and then sacrificed.
Mouse cohort, study design, and outcomes
We next performed in vivo studies using a cohort of 58 NOD.H-2h4 mice established at the Johns Hopkins University pathogen-free facility from a breeding pair kindly gifted by Dr. G. Carayanniotis (Memorial University of Newfoundland, St. Johns, Canada). When 8 wk old, mice were divided into two treatment groups, each containing two subgroups. In the first treatment group, 26 mice (17 M, 9 F) drank unsupplemented water that contained (10 mice: 7 M, 3 F) or did not contain (16 mice: 10 M, 6 F) MYMD-1 (185 mg/l). In the second treatment group, 32 mice (20 M, 12 F) drank water supplemented with 500 mg/l of sodium iodide (Sigma-Aldrich, St. Louis, MO) that contained (16 mice: 10 M, 6 F) or did not contain (16 mice: 10 M, 6 F) MYMD-1 (185 mg/l). Mice were staggered into small groups of four to six animals each and sacrificed at different days but all after 12 wk of treatment for a total of 10 different sacrifice days. At sacrifice, we collected thyroid, lymph nodes, and spleen in all mice to assess thyroiditis by H&E histology and alterations of immune responses by flow cytometric analysis using single-cell suspensions. The pancreas was also collected from 38 mice as a specificity control for the thyroid effect. Mice were bled from the submandibular plexus at baseline (8 wk of age) 2, 4, 6, 8, 10, and 12 wk after treatment to measure the study outcomes indicated below. The study was approved by The Johns Hopkins Animal Care and Use Committee.
Choice of iodine and MYMD-1 administration
The normal iodine intake of an adult mouse is estimated to be 2 μg/d (24), assuming a daily consumption of 5 g of feed and 5 ml of water. Mice in our study were fed with the Teklad 2818SX 18% protein rodent feed (Envigo, Madison, WI), which contains 6 μg/g of iodine (thus yielding a theoretical iodine intake of 30 μg/d). This diet, thus, already contains 15 times more iodine than the normal daily intake. And in fact, NOD.H-2h4 mice housed in our animal facility do spontaneously develop thyroiditis, albeit with varying penetrance. To ensure the development of thyroiditis and thus have a positive control for the experiments, we added an additional 2500 μg/d of iodine, corresponding to a dose of iodine 1000 times higher than the normal daily intake (9).
MYMD-1 stock solution was dissolved in autoclaved distilled water to a final concentration of 0.185 mg/ml (= 0.925 mg/d), thus yielding an estimated daily dose of ∼1 mg per mouse.
Thyroid histopathology and scoring
Thyroid glands were collected at sacrifice and used either entirely for H&E histopathology in 12 of 58 mice (21%) or split through the isthmus in the remaining 46 mice (79%) to assign one lobe to histopathology and the other lobe to flow cytometry. For histopathology, entire glands or lobes were fixed overnight in 10% buffered formalin and then processed and embedded in paraffin. Thyroid blocks were cut to obtain nonsequential 5-μm sections that were then stained with H&E. Sections were scored from 0 to 4 to quantify the severity of the mononuclear cell infiltration into the thyroid gland, as originally described in (25). Briefly, 0 indicates a normal thyroid gland with none or scanty mononuclear cells visible; 1 indicates a hematopoietic mononuclear cell infiltration involving <20% of the thyroid area; 2 indicates an involvement between 20 and 30%; 3 indicates an involvement between 31 and 50%; and 4 indicates an involvement between 51 and 100%, a marked infiltration that almost completely effaces the normal follicular architecture.
Mouse thyroglobulin Ag was purified by gel filtration chromatography, diluted to 2 ng/μl in carbonate-bicarbonate buffer, and used to coat Immulon 2 High Binding microtiter ELISA plates (supplier number 3455; Thermo Fisher Scientific, Waltham, MA) at an amount of 100 ng/well. Mouse sera were diluted 1:100 in TBS, added to the plates, and incubated for 2 h at room temperature. After washing, binding of the serum Abs to thyroglobulin was detected using an affinity-purified whole IgG Ab directed against mouse IgG and IgM (H chain + L chain) conjugated to alkaline phosphatase (catalog number 115-055-068; Jackson ImmunoResearch Laboratories, West Grove, PA). Following washing and addition of the p-nitrophenol phosphate substrate, enzymatic reaction was measured at 405 nm using the SpectraMax iD3 Microplate Reader (Molecular Devices). Each plate included in-house standards derived from serial dilutions of pooled sera derived from mice immunized with mouse thyroglobulin and thus containing high titers of thyroglobulin Abs.
Thyroid function as assessed by serum T4, triiodothyronine, and TSH
Serum levels of thyroid hormones (T4 and triiodothyronine [T3]) and TSH were measured using Luminex technology (catalog number RTHYMAG-30K; MilliporeSigma, Billerica, MA), following the manufacturer’s recommendations. Although the kit is designed to measure rat TSH, it has been used to measure mouse TSH (26), given the high degree of identity (90%) between the two species. Thyroid hormones, which are made of two iodine-modified tyrosines, are identical in all species (27).
Serum levels of cytokines (TNF-α, IFN-γ, IL-1β, IL-6, IL-10, IL-17A) were measured using Luminex technology (catalog number M60-0007NY; Bio-Rad Laboratories, Hercules, CA), following the manufacturer’s recommendations.
Ex vivo characterization of immune cell subsets by flow cytometry
One thyroid lobe, pooled lymph nodes (mainly from the draining cervical network), and spleen were collected into RPMI 1640 medium from each mouse (n = 46) at the time of sacrifice. Single-cell suspension from the individual thyroid lobes was prepared by mechanical disaggregation, followed by incubation with dispase I (final concentration of 0.04 mg/ml) and collagenase II (final concentration of 1 mg/ml) for 2 h at 4°C and then 15 min at 37°C. Cell suspension was passed through a 70-μm cell strainer, washed twice with PBS, and used for flow cytometric analysis. Lymph nodes and spleen were mechanically disrupted using frosted glass slides, strained through a 70-μm cell strainer, and washed with PBS. Spleen cells were then fractionated by Ficoll-Paque density gradient centrifugation to isolate the mononuclear cells. Thyroid (on average 27,390 ± 12,300 cells acquired per lobe), lymph node (1 × 106 cells), and spleen (1 × 106 cells) single-cell suspensions were stained with LIVE/DEAD Aqua viability dye, according to the manufacturer’s instructions. Cells were first incubated for 10 min on ice with an Ab to mouse CD16/32 to block Fc receptors, washed with FACS staining buffer (PBS with 1% BSA and 2 mM EDTA), and then stained for 30 min at 4°C with fluorochrome-labeled Abs directed against the following lymphoid surface markers: CD3 (17A2), CD4 (GK1.5), CD25 (eBio7D4 [7D4]), CD24 (M1/69), CD5 (53-7.3; all from Thermo Fisher Scientific), CD8 (53-6.7), CD19 (6D5), CD23 (B3B4; all from BioLegend), TCRγδ (GL3), CD1d (1B1), and CD21 (7G6; all from BD Biosciences). After fixation and permeabilization with eBioscience Foxp3/Transcription Factor Staining Buffer Set, cells were stained for 30 min at 4°C for intracellular transcription factors using Abs against Foxp3 (FJK-16s; from Thermo Fisher Scientific), RORγT (Q31-378; from BD), GATA-3 (16E10A23), and T-bet (4B10; both from BioLegend). Cells were acquired on a BD LSR II flow cytometer and analyzed using FlowJo software to assess the following lymphoid populations: CD3+CD4+ T cells, CD3+CD8+ T cells, CD3+ TCRγδ+ cells, regulatory T cells (CD3+CD4+Foxp3+CD25+), Th1 T cells (CD3+CD4+T-bet+RORγT−), Th2 T cells (CD3+CD4+GATA3+), Th17 T cells (CD3+CD4+T-bet−RORγT+), Th1/17 double positive (CD3+CD4+T-bet+RORγT+), total B cells (CD3−CD19+), follicular B cells (CD19+CD21intCD24int), and transitional type 1 B cells (CD19+CD21loCD23hi).
The primary outcome measure of the study was the severity of thyroiditis, as assessed by H&E histopathology (a categorical scale ranging from zero to four) or flow cytometry (a continuous scale representing the absolute number of live CD3 T cells in one thyroid lobe). Differences in the median severity among the four experimental groups were analyzed first by Kruskal–Wallis test and then by Wilcoxon rank-sum test for pairwise comparisons.
Secondary outcomes included thyroglobulin Ab incidence and change in the Ab titer over time, median percentages of the various lymphoid subsets, serum cytokine levels, serum thyroid hormones (T4 and T3), and TSH. Median percentages were compared by Kruskal–Wallis test. Longitudinal analysis of serum thyroglobulin Abs, cytokines, or thyroid function was performed with multiple linear regression with generalized estimating equations, according to (28). Briefly, longitudinal data analysis examines in each animal multiple values of a given variable, called clusters, and corrects for the fact that measures repeated over time on the same animal are more likely to be correlated with one another. Multiple logistic regression was used to assess the relationship between development of thyroiditis (as determined by the presence of CD3 T cell or CD19 B cell infiltration into the thyroid gland) and MYMD-1 treatment, with adjustment for water type (regular versus iodinated), sex of the mice, and date of the experiment. The adjusted odds ratios produced by the logistic model were back transformed to adjusted probabilities for easiness of interpretation, as described in (29).
Statistical analyses were performed using Stata statistical software, release 15.1 (Stata, College Station, TX).
MYMD-1 does not impair cell proliferation or viability in vitro
To assess the effect of MYMD-1 on cell proliferation, MIN6 and RAW 264.7 cells were treated with MYMD-1 for 24 h followed by a 2 h incubation with BrdU. Cell proliferation was not reduced by treatment with MYMD-1 at concentrations of up to 2 mM (Fig. 1A). To investigate the effect of MYMD-1 on cell viability, CD4+ T cells were isolated from NOD.H-2h4 splenocytes and cultured with MYMD-1 for 24 h, followed by stimulation with PMA and ionomycin for 6 h and stained with a viability dye and a CD3 Ab. The percentage of live CD3+ T cells was at least 70% in all culture conditions and did not significantly differ among cultures with various MYMD-1 concentrations (p = 0.796, Fig. 1B). Similar results were obtained by manually counting the number of viable RAW 264.7 cells using microscopy (Supplemental Fig. 1).
MYMD-1 suppresses the production of TNF-α by CD4+ T cells in vitro
To determine whether MYMD-1 shared similar anti-inflammatory properties with other tobacco alkaloids, CD4+ T cells were isolated from NOD.H-2h4 splenocytes, cultured for 24 h with MYMD-1 (0, 0.05, 0.1, 0.5, 1, and 2 mM), and then stimulated for 6 h with PMA and ionomycin in the presence of brefeldin A and monensin. Lymphocytes were stained for intracellular cytokines using Abs against TNF-α, IFN-γ, IL-17A, and IL-4 and then analyzed by flow cytometry. MYMD-1 significantly decreased TNF-α production by CD4+ T cells in a dose-dependent fashion (Fig. 2A). No differences were instead observed in IFN-γ, IL-17A, or IL-4 expression by CD4+ T cells (Fig. 2B–D).
MYMD-1 shows no toxicity at doses of up to 95 mg/kg in vivo
To determine MYMD-1 toxicity in vivo, 58 mice of mixed strains and sexes were injected i.p. with MYMD-1 doses of 300 (n = 5 mice), 190 (n = 24), 95 (n = 10), 47.5 (n = 5), or 0 (n = 14) mg/kg. MYMD-1 showed no toxicity when used at doses up to 95 mg/kg (Supplemental Fig. 2). Significant toxicity (tremors, convulsions, and breathing difficulties) was observed at doses of 190 mg/kg and 100% mortality at doses of 300 mg/kg (Supplemental Fig. 2).
MYMD-1 ameliorates the severity and/or incidence of autoimmune thyroiditis
A cohort of 58 NOD.H-2h4 mice drank either regular water or iodine-supplemented water, which contained or did not contain MYMD-1. The treatment began at 8 wk of age and continued for 12 wk when the study was censored. No toxicity was observed in any of the mice during the entire study period. At sacrifice, thyroid glands were assessed for lymphocytic infiltration by histopathology (one lobe) or flow cytometry (the other lobe). MYMD-1 ameliorated the severity and/or incidence of autoimmune thyroiditis (Fig. 3). In the regular water group, thyroiditis severity decreased 4-fold, from a mean value of 0.89 to 0.2 (Fig. 3A, p = 0.047) and thyroiditis incidence from 70% to 22% (Fig. 3B, p = 0.025) upon MYMD-1 administration. Thyroid glands were enlarged and showed clear mononuclear cell infiltration in most of the mice in the regular water subgroup (Fig. 3C), whereas they showed a near normal histopathology upon MYMD-1 administration (Fig. 3D). Similar findings were observed in the iodinated water group, although in this group, all of the disease features were exacerbated. Thyroiditis severity decreased from a mean value of 2.19 to 1.38 (Fig. 3E, p = 0.068) and incidence decreased from 100 to ∼60% (Fig. 3F, p = 0.018) upon MYMD-1 administration. Thyroid glands were enlarged and prominently infiltrated by hematopoietic mononuclear cells that often effaced the normal follicular architecture in the iodine-only subgroup (Fig. 3G), features that were still present but greatly attenuated in the iodine plus MYMD-1 subgroup (Fig. 3H).
To address the thyroid specificity of MYMD-1, we analyzed the exocrine and endocrine pancreas, anatomic locations where occasionally NOD.H-2h4 mice housed in our facility feature minor mononuclear cell infiltration. Only minimal, scattered infiltration clustered around veins was found in 6 (2 F and 4 M) of 38 mice (16%), with no difference among the four experimental groups (data not shown).
Characterization of the thyroidal lymphoid subsets by flow cytometry confirmed and expanded the histopathological results. In the regular water group, MYMD-1 significantly decreased the number of T cells infiltrating the thyroid gland, from a mean value of 461 cells per thyroid lobe to a mean value of 120 (Fig. 4A, left panel, p = 0.042). A decrease was also observed in the iodinated water group, although it did not reach statistical significance (Fig. 4A, right panel). CD3+ T cells were the most numerous infiltrating hematopoietic cells, outnumbering and preceding the infiltrating B cells (Fig. 4B). Multiple logistic regression showed that MYMD-1 significantly decreased the probability of having CD3 T cells infiltrating the thyroid gland after adjusting for the type of water the mice drank (regular versus iodinated water), sex, and time when the experiment was performed. The adjusted probability of CD3 T cell infiltration in the thyroid gland, in fact, decreased upon MYMD-1 treatment from 0.64 to 0.14 in the regular water group and from 1 to 0.71 in the iodinated water group (Fig. 4C). Similarly, the adjusted probability of having CD19+ B cell infiltration in the thyroid gland decreased upon MYMD-1 treatment from 0.36 to 0.18 in the regular water group and from 1 to 0.51 in the iodinated water group (Fig. 4D).
MYMD-1 decreases the Ab response to thyroglobulin
MYMD-1 significantly decreased the percentage of mice developing thyroglobulin Abs in the iodinated water group from 75 to 19% (Fig. 5A, right panel, p = 0.005). There was a decrease also in the regular water group from 25 to 10%, although it did not reach statistical significance (Fig. 5A, left panel). Thyroglobulin Abs at sacrifice correlated positively with the severity of thyroiditis (Fig. 5B, p = 0.001) with a concordance of 62% (Fig. 5B, upper right and lower left quadrants). The remaining 38% of mice developed thyroiditis but no thyroglobulin Abs (Fig. 5B, upper left quadrant). Thyroglobulin Abs were not detectable at baseline (8 wk of age) in any of the four experimental subgroups. They had a mean of 15 AU/ml, a median of 16, and an SD of 9. We chose a value of 40 AU/ml, corresponding to the 99th percentile (Fig. 5C, 5D, dashed line), as the threshold for seroconversion. In the iodinated water group, thyroglobulin Abs began to increase after 4 wk, continuing their upward trend up to 12 wk (Fig. 5C, filled circles). Treatment with MYMD-1 significantly retarded this humoral response, with thyroglobulin Abs remaining in the normal range at week 4 (p = 0.012) and 6 (p = 0.024) of treatment (Fig. 5C, open circles). A similar retardation was seen in the regular water group, although the Ab titer was overall lower in the absence of iodine. In the absence of MYMD-1, thyroglobulin Abs became detectable at week 6 and then remained around 100 AU/ml up to the end of the study (Fig. 5D, closed squares). Treatment with MYMD-1 maintained thyroglobulin Abs in the normal range up to 10 wk of treatment, with significantly lower levels seen at week 6 (p = 0.012) and 8 (p = 0.025) (Fig. 5D, open squares).
MYMD-1 specifically affects the Th1 (CD3+CD4+Tbet+RORγT−) cell subset in thyroid, lymph nodes and spleen
In the regular water group, MYMD-1 significantly decreased the percentage of Th1 cells (defined as CD3+CD4+Tbet+RORγT−) in the thyroid gland (Fig. 6A, left panel, p = 0.037), spleen (Fig. 6B, left panel, p = 0.001), and lymph nodes (Fig. 6C, left panel, p = 0.02). In the iodinated water group, thyroidal Th1 cells similarly tended to decrease upon MYMD-1 administration, although this decrease did not reach statistical significance (Fig. 6A, right panel, p = 0.058). No effect was seen for the Th1 T cell populations residing in spleen (Fig. 6B, right panel) and lymph nodes (Fig. 6C, right panel). Th17 cells (defined as CD3+CD4+Tbet−RORγT+) were not affected by MYMD-1 in the thyroid gland (Fig. 6D) and lymph nodes (Fig. 6F), although they tended to decrease in the spleen of mice drinking regular water (p = 0.074, Fig. 6E, left panel). Examples of the Th1/Th17 gate in thyroid, spleen, and lymph nodes are shown in Fig. 6G–I. No marked effect was seen for Th2 (CD3+CD4+GATA-3+), T regulatory (CD3+CD4+CD25+Foxp3+), or CD3+TCRγδ+ T cell subsets (data not shown).
MYMD-1 specifically affects follicular (CD19+CD21intCD24int) B cell subset in the thyroid gland
In the regular water group, MYMD-1 significantly decreased the follicular B cells (defined as CD19+CD21intCD24int) in the thyroid gland from a median of 63% of the total CD19+ cells to 8% (Fig. 7A, left panel, p = 0.008). A similar trend was seen in the iodinated water group, in which thyroid follicular B cells decreased from a median value of 72 to 60%, although this decrease did not reach statistical significance (Fig. 7A, right panel, p = 0.198). Interestingly, the follicular B cell effect was specific to the thyroid gland and not seen in the splenic (Fig. 7B) and nodal (Fig. 7C) compartments. Examples of the CD21/CD24 gate in thyroid, spleen, and lymph nodes are shown in Fig. 7D–F.
MYMD-1 decreases serum TNF-α levels
Considering the decreased TNF-α production by CD4+ T cells cultured in vitro in the presence of MYMD-1 and the reduction of Th1 cells in the MYMD-1–treated mice, we assessed whether MYMD-1 administration influenced the systemic levels of TNF-α, IFN-γ, IL-17A as well as those of IL-1β, IL-6, and IL-10. In the regular water group, serum levels of TNF-α increased 1.5-fold over baseline after 6 wk of treatment and remained elevated through the end of the study (Fig. 8A, filled squares). Treatment with MYMD-1 maintained serum TNF-α close to the baseline values, so that at week 12 they were significantly lower than those found in mice not drinking MYMD-1 (Fig. 8A, open squares, p = 0.042). In the iodinated water group, the trends were similar although TNF-α levels were overall higher with iodine, and the decrease induced by MYMD-1 was borderline significant at week 12 (Fig. 8B, p = 0.053). The effect of MYMD-1 on serum IFN-γ was not seen in the regular water group (Fig. 8C) and modest in the iodinated water group (Fig. 8D). No statistically significant differences were found for IL-17A, IL-1β, IL-6, and IL-10 (data not shown).
MYMD-1 maintained serum TSH in the normal reference range in NOD.H-2h4 mice
In the NOD.H-2h4 model, thyroid function has been typically assessed by measuring the serum levels of total T4 and/or T3. Only a few reports, often of small numerosity, have included the measurement of the pituitary hormone TSH (30–32). Upon iodine administration, thyroid hormones are known to remain within the normal reference range (10), although the thyroid gland has been reported to be consistently hypertrophic (24), suggesting a compensatory response of the thyroid follicular cells to increased TSH production. We, therefore, measured TSH in the four experimental groups and found that TSH increased over time in the iodine group, whereas it remained in the normal range when MYMD-1 was used (Fig. 9A, males, and Fig. 9B, females). Serum T4 and T3 did not significantly differ among the four experimental groups (Fig. 9C–F).
The study first, to our knowledge, reports the use of MYMD-1 in a mouse model of autoimmunity, revealing its beneficial effects in the absence of measurable toxicity. In particular, MYMD-1 decreased disease incidence and severity and thyroglobulin Abs, the hallmarks of autoimmune thyroiditis.
The pathogenesis of thyroiditis in the NOD.H-2h4 model largely depends on the contribution of Th1, the first lymphoid population that infiltrates the thyroid gland, Th17, and follicular B lymphocytes as well as on the proinflammatory cytokines they produce, such as TNF-α and IFN-γ (reviewed in Refs. 9, 10). MYMD-1 predominantly suppressed the number of pathogenic Th1 cells in the thyroidal, splenic, and nodal compartments. Th1 cells, a major source of IFN-γ, are critical in thyroiditis induction because NOD.H-2h4 mice lacking either IFN-γ itself (33) or the IFN-γ receptor (34) do not develop thyroid lymphocytic infiltration. Similarly, MYMD-1 markedly decreased the number of thyroid follicular B cells, a major thyroid-infiltrating population in the NOD.H-2h4 model, which is involved not only in the production of thyroid-specific Abs but also in Ag presentation (15). Overall, our findings point toward an exquisite targeting of MYMD-1 on these lymphoid subsets. The Th1 and follicular B cell decreases were not accompanied by changes in Th17 cells in lymph nodes, thyroid and spleen, reflecting perhaps a selective effect of MYMD-1 on lymphocyte trafficking or a minor role of Th17 in later stages of disease. Thyroidal Th17 cells were, in fact, shown to contribute to thyroiditis development when disease was assessed at 4 or 8 wk after iodine administration (13, 32), whereas we censored the study after 12 wk.
Another interesting finding that emerged from this study was the action of MYMD-1 on TNF-α, a cytokine involved in the induction and maintenance of inflammatory processes and considered “at the top of the proinflammatory cascade” (35). Blockade of TNF-α remarkably improves clinical outcomes in patients with rheumatoid arthritis and inflammatory bowel disease and is being investigated in numerous other autoinflammatory conditions. Less is known about the role of TNF-α in thyroid autoimmunity. TNF-α is produced by T cells infiltrating the thyroid of patients with Hashimoto thyroiditis (36) and follicular B cells homing to the thyroid of NOD.H-2h4 mice (14). TNF-α is also critical to disease development in the experimental models of lymphocytic thyroiditis induced by thyroglobulin immunization in the presence of LPS (37) and granulomatous thyroiditis induced by adoptive transfer of splenocytes primed with thyroglobulin and IL-12 (38). Interestingly, epithelial thyroid cells themselves possess the ability to release TNF-α when exposed to inflammatory contexts (39). We found that MYMD-1 decreased both the ability of CD4 T cells to release TNF-α in vitro and the serum levels of TNF-α, in keeping with the known ability of alkaloids to inhibit TNF-α production.
The immunoregulatory role of alkaloids, such as nicotine and MYMD-1, is a topic of great interest and potential clinical applicability. Although tobacco smoking is the classic environmental risk factor for many inflammatory and neoplastic diseases, it also contains some compounds (such as the alkaloids) endowed with anti-inflammatory properties (40). Nicotine is the best characterized tobacco alkaloid and is known to modulate the immune system mainly by binding to the homopentameric α-7 nicotinic acetylcholine receptor (41). This receptor is expressed strongly in the brain but also on a variety of nonneuronal cell types, including lymphocytes, granulocytes, dendritic cells, and macrophages (41). Nicotine binding to cells of the innate and adaptive immune system suppresses JAK2, STAT3, and NF-κB pathways, inhibiting the production of proinflammatory cytokines, effects that can be mimicked by stimulation of the vagus nerve (42). Nicotinic effects, however, are complex, multifaceted, and mechanistically traceable to changes in intracellular ion fluxes. The complexity is compounded by the existence of numerous types of acetylcholine receptors (17 types of subunits have, in fact, been identified in vertebrates), which can be differentially expressed on various immune cell subsets (reviewed in Ref. 43). Considering its immunoregulatory properties, nicotine has been tested in several rodent models of human autoimmune diseases (44). In the experimental autoimmune encephalomyelitis model of multiple sclerosis, administration of nicotine consistently results in neuroprotection acting through both the α 7 and α 9 receptors (45). This beneficial effect is associated with a suppression of TNF-α release from microglial cells (46) and the blockade of Th1 and Th17 responses (18). In the collagen-induced (47) or Mycobacterium tuberculosis H37Ra-induced (48) models of rheumatoid arthritis, nicotine ameliorates joint affliction by reducing Th1 differentiation and the release of proinflammatory cytokines such as of TNF-α from the synovium. In the streptozotocin-induced and NOD models of type 1 diabetes mellitus, the elevation of TNF-α and other inflammatory cytokines typically observed in the pancreatic islets is prevented by nicotine (49). In the coxsackievirus B3-induced model of myocarditis, i.p. injections of high-dose nicotine reduced myocarditis severity while improving left ventricular function, effects that were associated with a dose-dependent downregulation of the cardiac levels of TNF-α, IL-1β, IL-6, and IL-17A mRNAs and proteins (50). Taken together, these studies suggest a beneficial effect of nicotine on a variety of autoimmune diseases, although its toxicity and unfavorable pharmacodynamics properties prevent its clinical use. The present study of MYMD-1 showed efficacy in ameliorating thyroiditis with no evidence of toxicity throughout the entire study period, suggesting a mode of action not limited to the nicotinic “cholinergic anti-inflammatory pathway.”
In conclusion, our study supports the potential clinical use of MYMD-1, a novel and seemingly harmless compound that ameliorates thyroiditis via the reduction of Th1 responses and TNF-α release. The study paves the way to testing the potential therapeutic and/or prophylactic use of MYMD-1 in patients with Hashimoto thyroiditis as well as in other autoimmune diseases in which Th1 responses and TNF-α are critical. Assessments of MYMD-1 in rodent models of multiple sclerosis, alopecia, diabetes mellitus, and age reversal are being performed at our institution and will confirm the role of MYMD-1 as a novel immunometabolic regulator.
We thank David Zahavi and Dr. Marcella Ferlito for technical contribution to the in vitro studies of MYMD-1, Christopher Thoburn for assistance in performing the Luminex assays, and Dr. Katherine Whartenby for critical revision of the manuscript.
This work was supported by a grant from MyMD Pharmaceuticals Inc. G.D.D. was supported in part by the Division of Endocrinology of the University of Chieti-Pescara, P. Chalan by a Virginia O'Leary and John C. Wilson Autoimmune Disease Research Fellowship, and P. Caturegli in part by National Institutes of Health Grant R01 CA-194042.
The online version of this article contains supplemental material.
Abbreviations used in this article:
The authors have no financial conflicts of interest.