Anatabine Ameliorates Experimental Autoimmune Thyroiditisinsanemedicine.com/wp-content/...Experimental-Autoimmune-thyroid… · developed thyroid infiltrates when drinking anatabine

Anatabine Ameliorates Experimental AutoimmuneThyroiditis

Patrizio Caturegli, Alessandra De Remigis, Marcella Ferlito,Melissa A. Landek-Salgado, Shintaro Iwama, Shey-Cherng Tzou,and Paul W. Ladenson

Department of Pathology (P.C., A.D.R., M.A.L.-S., S.I., S.-C.T.), and Divisions of Endocrinology andMetabolism (P.C., P.W.L.) and Cardiology (M.F.), Department of Medicine, Johns Hopkins UniversitySchool of Medicine, and Feinstone Department of Molecular Microbiology and Immunology (P.C.), JohnsHopkins Bloomberg School of Public Health, Baltimore, Maryland 21205

Tobacco smoking favorably influences the course of Hashimoto thyroiditis, possibly through theantiinflammatory proprieties of nicotine. In this study we tested anatabine, another tobacco al-kaloid, in a model of experimental autoimmune thyroiditis. Experimental autoimmune thyroiditiswas induced by different doses of thyroglobulin, to produce a disease of low, moderate, or highseverity, in 88 CBA/J female mice: 43 drank anatabine supplemented water and 45 regular water.Mice were bled after immunization and killed to assess thyroid histopathology, thyroglobulinantibodies, T4, and thyroid RNA expression of 84 inflammatory genes. We also stimulated in vitroa macrophage cell line with interferon-� or lipopolysaccharide plus or minus anatabine to quan-titate inducible nitric oxide synthase and cyclooxygenase 2 protein expression. Anatabine reducedthe incidence and severity of thyroiditis in the moderate disease category: only 13 of 21 mice (62%)developed thyroid infiltrates when drinking anatabine as compared with 22 of 23 (96%) controls(relative risk 0.59, P � 0.0174). The median thyroiditis severity was 0.5 and 2.0 in anatabine andcontrols, respectively (P � 0.0007 by Wilcoxon rank sum test). Anatabine also reduced the antibodyresponse to thyroglobulin on d 14 (P � 0.029) and d 21 (P � 0.045) after immunization and improvedthe recovery of thyroid function on d 21 (P � 0.049). In the thyroid transcriptome, anatabinerestored expression of IL-18 and IL-1 receptor type 2 to preimmunization levels. Finally, anatabinesuppressed in a dose-dependent manner macrophage production of inducible nitric oxide synthaseand cyclooxygenase 2. Anatabine ameliorates disease in a model of autoimmune thyroiditis, mak-ing the delineation of its mechanisms of action and potential clinical utility worthwhile.(Endocrinology 153: 4580–4587, 2012)

Tobacco smoking has numerous detrimental effects onhuman health, but it has also been associated with a

few apparent salutary actions, including the ameliorationof autoimmune (Hashimoto) thyroiditis and ulcerativecolitis. Smokers in the Third National Health and Nutri-tion Examination Survey were found to have lower prev-alence of thyroperoxidase and/or thyroglobulin antibod-ies than nonsmokers (1). This protective effect of smokingwas confirmed in two additional cross-sectional studies,one from the Amsterdam autoimmune thyroid disease co-hort (2) and the other from the Danish population (3), as

well as in a 5-yr prospective study also based on the Am-sterdam autoimmune thyroid disease cohort (4). In theprospective study, cigarette smoking women who had oneor more relative with documented thyroid autoimmunitybut no thyroid dysfunction or autoantibodies at study en-try showed lower odds of developing thyroperoxidaseand/or thyroglobulin antibodies (4). Similarly in ulcerativecolitis, smoking has been shown to decrease flares (5), hos-pitalizations (6), and a need for oral glucocorticoids (7) sothat low-dose smoking resumption has been successfullyused in ex-smokers with refractory disease (8).

ISSN Print 0013-7227 ISSN Online 1945-7170Printed in U.S.A.Copyright © 2012 by The Endocrine Societydoi: 10.1210/en.2012-1452 Received April 23, 2012. Accepted June 27, 2012.First Published Online July 17, 2012

Abbreviations: COX2, Cyclooxygenase 2; EAT, experimental autoimmune thyroiditis; IL-1R2, IL-1 receptor type 2; iNOS, inducible nitric oxide synthase; NF-�B, nuclear factor-�B.

T H Y R O I D - T R H - T S H

4580 endo.endojournals.org Endocrinology, September 2012, 153(9):4580–4587

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 13 April 2016. at 21:43 For personal use only. No other uses without permission. . All rights reserved.

The mechanisms underlying this influence of tobaccosmoking on some autoimmune diseases have been relatedto the effects of tobacco components on the immune sys-tem (9). There are numerous (�4000) components in to-bacco, including alkaloids (such as nicotine and anat-abine), gases (e.g. carbon monoxide), and carcinogens(e.g. polycyclic aromatic hydrocarbons, aldehydes, freeradicals, and solvents), and of them nicotine is known topossess antiinflammatory properties (10). Nicotine actsvia binding to the nicotinic receptor, a pentameric ionchannel (mainly for sodium and calcium) formed by thearrangement of 16 different subunits in hetero- or homo-meric conformations (11). The receptor is classically ex-pressed in the peripheral (all preganglionic fibers and neu-romuscular synapses) and central nervous system, butmore recently it has been described in cells of the immunesystem, including CD4 T lymphocytes, dendritic cells, andmacrophages (12). Indeed, the �7-homopentameric nico-tinic receptor has emerged as a novel therapeutic target fordiseases with an inflammatory pathogenesis (13).

Nicotine has been used successfully in mice with ex-perimental autoimmune encephalomyelitis in which it re-duced disease severity, shifting the autoimmune profilefrom pathogenic Th1 and Th17 responses to protectiveTh2 responses (14). Nicotine, however, cannot be used inhumans because it is addictive and toxic and has a short3-hplasmahalf-life.Consequently,we reasoned thatotheralkaloids of tobacco could share similar antiinflammatoryproperties but have a more favorable pharmacologicalprofile. The minor tobacco alkaloid anatabine is nonad-dictive and nontoxic at therapeutic doses and has a longer8-hr half-life. Furthermore, anatabine has been recentlyshown to inhibit nuclear factor-�B (NF-�B) activation andreduce neuroinflammation in a mouse model of Alzheimerdisease (15). In the present study, we therefore tested theantiinflammatory properties of anatabine in a mousemodel of experimental autoimmune thyroiditis.

Materials and Methods

Mice and induction of experimental autoimmunethyroiditis (EAT).

EAT was induced in 8-wk-old female CBA/J mice (The Jack-son Laboratory, Bar Harbor, ME) as described (16). Briefly,mouse thyroglobulin was purified by S-300 gel chromatographyand emulsified 1:1 in complete Freund’s adjuvant (which con-tains 5 mg/ml of Mycobacterium tuberculosis strain H37 Ra;Difco Laboratories, Detroit, MI). Mice were injected sc with theemulsion on d 0 and 7 and then killed on d 21 [see a videodemonstration of the immunization procedure published else-where (17)]. Three different induction protocols were used toobtain a thyroiditis of low, moderate, or high severity, based on

the amount of mouse thyroglobulin in the immunogen (50, 75,or 100 �g, respectively). Each experiment included two groups,both immunized in the same manner and coded for blinding. Thecontrol group drank regular water; the anatabine group drankwater supplemented with anatabine. Eighty-eight mice werestudied: 43 anatabine treated and 45 controls, immunized ineight separate experiments by three investigators (A.D.R.,M.A.L.-S., and S.-C.T.). Experiments were conducted in accor-dance with the standards established by the U.S. Animal WelfareActs, set forth in the National Institutes of Health guidelines andapproved by the Johns Hopkins University Animal Care and UseCommittee.

Anatabine source and administrationAnatabine was provided by Rock Creek Pharmaceuticals

(Washington, DC), its commercial producer. The racemic mix-ture was synthesized by Anthem Biosciences Pvt. Ltd. (Banga-lore, India) and then purified and formulated by Emerson Re-sources Inc. (Norristown, PA). Anatabine was added to the waterbottle at the start of the experiment (d 0) at a concentration of0.05 mg/ml, a concentration derived from the first, and thus faronly, study that used anatabine in mice (15). Based on a youngadult CBA/J female mouse drinking approximately 5 ml of waterper day and weighing approximately 20 g, we administered anat-abine at a dose of 0.25 mg/d per mouse to approximate a dose of12.5 mg/kg�d, which corresponds to a human adult dose of 1mg/kg�d, based on the formula reported by Reagan-Shaw et al.(18). The same bottle was used for the 3-wk duration of theexperiment, and the water was changed twice a week. The con-sumption of water in each cage was estimated by measuring theweight of the water bottle. There was no difference in waterconsumption between cages drinking anatabine water and cagesdrinking regular water.

Thyroid histopathology and macrophage stainingThyroid glands were collected 21 d after the first immuniza-

tion. For most mice (72 of 88), the entire gland still attached tothe trachea was removed and fixed overnight in 10% bufferedformalin. For 16 mice, eight anatabine treated and eight controls,one lobe was removed for gene expression analysis (see below),and the other lobe attached to the trachea was used for histo-pathology. After processing and embedding in paraffin, thyroidspecimens were cut completely to obtain numerous nonsequen-tial 5-�m sections, which were then stained with hematoxylinand eosin. Sections were scored blindly (by P.C.) to obtain a scoreexpressed on a discrete scale from 0 (normal) to 5 (completethyroid infiltration and destruction) that summarizes the severityof autoimmune thyroiditis. The presence of macrophages inthe thyroid glands with ongoing EAT was assessed by immu-nohistochemistry using a rat antimouse antibody (F4/80; Se-rotec, Raleigh, NC; diluted 1:16) using the procedure previ-ously described (19).

Thyroglobulin and purified protein derivativeantibodies

Mice were bled on d 0, 10, 14, and 21 after the first immu-nization to collect serum for the measurement of thyroglobulinantibodies, as previously described (16). Briefly, Immunlon2ELISA plates (Dynex Technologies, Chantilly, VA) were coatedwith 100 ng/well of mouse thyroglobulin, blocked, and then

Endocrinology, September 2012, 153(9):4580–4587 endo.endojournals.org 4581


incubated overnight with mouse sera diluted according to theday after immunization (1:100 for d 0, 1:400 for d 10 and 14, and1:1600 for d 21). After washing, incubation with a secondaryantibody recognizing mouse IgG and IgM conjugated with al-kaline phosphatase, and color development by addition of thep-nitrophenylphosphate substrate, plates were read at 405 nmusing Emax microplate reader (Molecular Devices, Sunnyvale,CA). Results were expressed as arbitrary units per milliliter usingin each plate a homemade standard curve derived from serialdilutions of a serum sample with known thyroglobulin antibod-ies. To control for thyroid specificity, we also assessed the anti-body response against purified protein derivative, a componentof the M. tuberculosis contained in the complete Freund’s adju-vant. Purified protein derivative (Mycos Research, LLC, Love-land, CO) was used to coat (1 �g/well) ELISA plates (Immulon2;ISC BioExpress, Kaysville, UT), which were then incubated over-night with mouse sera (in triplicates) diluted 1:100 in Tris-buff-ered saline. After extensive washing, a secondary antibody con-jugated to alkaline phosphatase recognizing mouse IgG and IgMwas used to detect specific binding. After addition of the p-ni-

trophenylphosphate substrate, color development was assessedand expressed as OD.

Assessment of thyroid function by serum total T4

Total T4 was measured on d 0, 10, 14, and 21 using a com-mercial RIA (GammaCoat 125T4; Diasorin, Stillwater, MN) ac-cording to the manufacturer’s protocol.

Analysis of thyroidal gene expression by real-timeRT-PCR

Thyroid lobes from d 21 mice of the moderate severity group,eight from anatabine treated and eight from controls, werepooled into separate microcentrifuge tubes and processed byRNeasy minikit (QIAGEN, Valencia, CA) to extract total RNA.We also used 10 lobes from unimmunized (d 0) wild-type CBA/Jfemale mice as baseline control. Yield and quality of RNA weredetermined by spectrophotometry and denaturing agarose gelelectrophoresis. Total RNA was reverse transcribed using theRT2 first-strand kit (QIAGEN) following the manufacturer’s in-

FIG. 1. Anatabine suppresses the incidence and severity of EAT. A, Thyroid histopathology score in the two groups of mice according to theseverity of EAT: low (50 �g of thyroglobulin), moderate (75 �g), and high (100 �g) severity. B, Thyroid histopathology in the moderate (75 �g)EAT group. Lines represent the median score.

4582 Caturegli et al. Anatabine Ameliorates Thyroiditis Endocrinology, September 2012, 153(9):4580–4587


structions. cDNA products (175 ng per reaction) were mixed tothe supplied quantitative PCR master mixture, added to a 96-well mouse inflammatory marker plate (PAMM-011; QIA-GEN), and amplified using the Bio-Rad CFX96 modular thermalcycler platform (Bio-Rad Laboratories, Hercules, CA). Thisplate contains primers for the detection of 40 chemokines (27ligands and 13 receptors), 37 cytokines (25 ligands and 12 re-ceptors), and seven additional inflammatory molecules (the com-plete list of these 84 genes can be seen at the QIAGEN web site:http://www.sabiosciences.com/rt_pcr_product/HTML/PAMM-011A.html). Results normalized for the expression ofhousekeeping genes were analyzed using the QIAGEN web sitesoftware appropriate for the product. We considered significantthose changes in gene expression that in pairwise comparisonsyielded a 4-fold difference (i.e. values � 4 for increases and �0.25 for decreases).

In vitro effect of anatabine on inflammatorymarkers using a macrophage cell line

Given the presence of macrophages infiltrating the thyroidgland in the EAT model and the availability of well-establishedmouse macrophage cell lines, we used the RAW 264.7 line (fromAmerican Type Culture Collection, Manassas, VA) to assess theeffects of anatabine on key inducible inflammatory markers likecyclooxygenase 2 (COX2) and inducible nitric oxide synthase(iNOS)-2. Macrophages were grown at 37 C in a CO2 incubatorusing RPMI 1640 medium (Mediatech, Manassas, VA) supple-mented with 10% heat-inactivated fetal bovine serum and 1%penicillin and streptomycin. When confluent, cells were stimu-lated overnight with interferon-� (10 ng/ml), a well-character-ized activator of macrophages. Thirty minutes later, anatabinewas added to the overnight culture at concentrations of 250, 350,500, 1000, or 2000 �M. Protein extracts were then prepared bycell lysis, separated by SDS-PAGE on gradient gels (NuPAGENovex 4–12%; Novex, San Diego, CA), and transferred ontopolyvinyl difluoride membranes (Bio-Rad Laboratories). Afterblocking with 5% nonfat dry milk, membranes were incubatedovernight with COX2 antibody (Cell Signaling Technology, Inc.,Danvers, MA; diluted 1:5,000) or iNOS antibody (Santa CruzBiotechnology, Inc., Santa Cruz, CA; also diluted 1:1,000). Toadjust for loading in each well, membranes were probed with atubulin antibody (Abcam, Cambridge, MA; diluted 1:5000). Sig-nal was detected by the addition of a secondary anti-IgG anti-body conjugated to horseradish peroxidase and quantified usingImageJ (National Institutes of Health, Bethesda, MD), adjustingfor the levels of �-tubulin in each lane. Parallel experiments werealso performed with lipopolysaccharide as a stimulant, ratherthan interferon-�, as indicated in the Supplementary Materials,published on The Endocrine Society’s Journals Online web siteat http://endo.endojournals.org.

Statistical analysisThe primary study outcome was the autoimmune response as

assessed by thyroid histopathology. Its severity was comparedbetween the two experimental groups (anatabine treated andcontrols) by the Mann-Whitney U rank-sum test and its inci-dence by risk ratio. Secondary outcomes of the study were serumthyroglobulin antibodies and total T4. They were analyzed usingmultiple linear regression with generalized estimating equations,which takes into account multiple values (at the various immu-

nization days) from the same mouse, correcting for the possibil-ity that characteristics of a single mouse over time are likely to becorrelated with one another. Analyses were performed usingStata 12 (Stata Corp., College Station, TX).

Results

Anatabine decreases the incidence and severity of EATAnatabine significantly decreased the severity of EAT

induced with our standard induction protocol. The me-dian thyroiditis score decreased from 2.0 in the controlgroup to 0.5 in the anatabine group (Fig. 1A, middle pair

FIG. 2. Antibodies to thyroglobulin (A), antibodies purified proteinderivative (PPD) (B), and total T4 (C) during EAT in mice receivinganatabine or control water. Analytes were assessed before (d 0) and10, 14, or 21 d after the first thyroglobulin immunization. The dashedlines (C) represent the mean (4.78 �g/dl) � 1 SD (1. 49 �g/dl) of totalT4 in 70 unimmunized female CBA/J mice studied by our laboratory inprevious publications (16, 30).



http://www.sabiosciences.com/rt_pcr_product/HTML/PAMM-011A.html

http://www.sabiosciences.com/rt_pcr_product/HTML/PAMM-011A.html

of box plots, and Fig. 1B, P � 0.0007). The incidence ofEAT was also significantly decreased by anatabine (Fig.1B): eight of 21 mice did not develop EAT when drinkinganatabine compared with one of 23 of those drinking reg-ular water (relative risk 0.59, 95% confidence interval0.39–0.9, P � 0.0174). The microscopic appearance ofthe hematopoietic infiltrate was similar in both experi-mental groups, composed mainly of lymphocytes andmacrophages (Fig. 1C, inset), but the infiltration was lessextensive and destructive of the thyroid architecture in theanatabine-treated mice (Fig. 1D) than in controls (Fig.1C). When EAT was induced to achieve complete diseasepenetrance and greatest severity, anatabine was not ob-served to ameliorate thyroiditis severity (Fig. 1A, rightpair). Conversely, when EAT was induced with low thy-roglobulin doses, thyroiditis was mild in both experimen-tal groups (Fig. 1A, left pair), implying that anatabine perse is not harmful to the thyroid. Overall these results dem-onstrate that anatabine reduces the incidence and amelio-rates the severity of this experimental form of autoimmunethyroiditis.

Anatabine decreases specifically the antibodyresponse to thyroglobulin and improves therecovery of thyroid function

Serum thyroglobulin antibodies in EAT typically be-gin to increase on d 10 after the first immunization, peakat approximately d 60, and then gradually decrease byd 100 without disappearing entirely (16). Anatabinetreated mice developed lower levels of thyroglobulinantibodies than controls on d 14 (P � 0.029) and d 21(P � 0.045) (Fig. 2A), suggesting that anatabine atten-uates the thyroid-specific autoimmune response in-duced by thyroglobulin immunization. This attenuationwas specific to thyroglobulin antibodies because it wasnot observed for the antibody response against purifiedprotein derivative (Fig. 2B), suggesting that anatabineadministration is not associated with generalizedimmunosuppression.

Total T4 is known to decrease markedly and rapidlyafter the injection of complete Freund’s adjuvant due to anonthyroidal illness effect (20) and then to decrease atlater time points due to the autoimmune destruction of thethyroid gland (16). Anatabine-treated mice showed higherserum T4 concentrations than controls on d 21 after im-munization (Fig. 2C, P � 0.049), suggesting a prompterrecovery of euthyroidism. This effect tended to be presentalso on d 10 and 14 but did not reach statistical signifi-cance (Fig. 2C).

Anatabine in vivo decreases specificproinflammatory signals in the thyroid gland withongoing EAT

To begin addressing the mechanism of action of anat-abine, we compared the thyroidal expression of a panel ofcytokines and chemokines between anatabine-treated andcontrol EAT mice, each adjusted for the expression level ofunimmunized thyroid genes. Of the total 84 genes presenton the array, eight increased significantly (�4-fold) duringEAT (Table 1). Anatabine suppressed the EAT-mediatedincrease of IL-1 receptor type 2 (IL-1R2, Figure 3A) andIL-18 (Fig. 3B), restoring their expression to levels similarto those seen in normal unimmunized thyroids. These re-sults reveal a specific effect of anatabine on proinflamma-tory pathways that may explain the amelioration of thy-roiditis incidence and severity observed histologically.

Anatabine in vitro decreases the specificproinflammatory signals in a macrophage cell line

We next assessed in vitro the effect of anatabine oniNOS and COX2, two enzymes that increase universallyduring inflammation. Macrophages were first activatedwith interferon-� to induce the expression of iNOS andCOX2 and then incubated with anatabine at concentra-tions of 250 or 350 �M. These concentrations were chosenbased on a dose-response curve ranging from 250 to 2000�M (Supplemental Fig. 1). Anatabine suppressed in a dose-dependent manner the increase of iNOS and COX2 in-

TABLE 1. Thyroidal expression levels of eight chemokine/cytokine genes that significantly increased (�4-fold)during EAT as compared with the expression in unimmunized thyroids

Gene symbolAnatabine treated

(fold increase over unimmunized)Control

(fold increase over unimmunized)Net effect

of anatabine

IL-1R2 3.80 15.74 Decrease 0.24IL-18 1.24 5.34 Decrease 0.23Spp1 32.65 20.85 No change 1.57CCL8 20.37 16.72 No change 1.22CCL6 6.74 7.32 No change 0.92Itgam 7.05 6.34 No change 1.11IL-1� 10.89 5.18 No change 2.10CCR2 6.25 4.82 No change 1.30

The presence of anatabine suppressed the increase of IL-1R2 and IL-18.



duced by interferon-� (Fig. 4), confirming in vitro itsantiinflammatory properties. The effect seen with inter-feron-� was also seen when macrophages were stimulatedwith lipopolysaccharide (Supplemental Fig. 1).

Discussion

Autoimmune (Hashimoto) thyroiditis, first reported in1912 and considered a rarity until the mid-1950s, has nowbecome the most common autoimmune disease. It has agenetic predisposition, such as the association with certainhuman leukocyte antigen haplotypes and polymorphismsin immunoregulatory genes (21), and a long preclinicalphase of detectable autoimmunity without hypothyroid-ism (22). Consequently, there is the theoretical opportu-nity to intervene to maintain normal thyroid structure andfunction. However, despite the disorder’s long history and

high prevalence, thyroid hormone replacement after glandfailure remains the only established medical treatment forthe condition.

It is now widely appreciated that autoimmune diseaseshave a complex and multifactorial pathogenesis in whichseveral signaling pathways lead to overproduction of cy-tokines that damage the tissue and ultimately result inclinical disease (23). Drugs that specifically inhibit or neu-tralize these cytokines, such as the blockade of TNF� inrheumatoid arthritis or IL-1 in systemic-onset juvenile id-iopathic arthritis (23), are now commonly used to treatsevere autoimmune manifestations. For Hashimoto thy-roiditis, however, the effectiveness of T4 therapy as well asthe cost and potential for adverse events, make the cyto-kine-blocking agents inappropriate for prolonged use tomaintain thyroid gland health.

Based on epidemiological observations that smoking isassociated with lower prevalence and incidence of sero-logical autoimmune thyroiditis, we hypothesized that acomponent of tobacco must be responsible for this effect.Anatabine is an alkaloid with a structure similar to nico-tine, found in tobacco and other solanaceous plants, such

FIG. 3. Thyroid RNA expression levels of IL-1R2 (A) and IL-18 (B) on d21 after immunization in mice treated with anatabine or controls. RNAlevels are normalized for the expression of housekeeping genes andexpressed as fold change over the levels observed in unimmunized (d0) thyroids.

FIG. 4. Macrophage protein levels of iNOS and COX2 afterstimulation with interferon (IFN)-� and addition of anatabine (Anatab.)at various concentrations. In each lane, the expression level of ahousekeeping gene (�-tubulin) is used as loading control. A, Westernblot images. B, Quantification of the protein band intensity, adjustedfor the intensity of �-tubulin in each lane.



as tomatoes, potatoes, green pepper, and eggplant. Its lackof addictive potential or any demonstrated toxicity, and itslonger half-life made it a reasonable candidate to test forefficacy in preventing and/or ameliorating Hashimotothyroiditis. We thus assessed the efficacy of anatabine in awell-established mouse model of the human disease in-duced by immunization with thyroglobulin and Freund’sadjuvant. We found that the addition of anatabine to thedrinking water reduced the incidence of thyroiditis andlessened the extent of the thyroid lymphocytic infiltrate inmice that develop it. Furthermore, this attenuation of his-topathological gland involvement was accompanied bylower circulating thyroid antibody levels and prompterrestoration of euthyroidism. It is important to note thatthis effect was not observed in mice in which thyroiditiswas induced with an aggressive protocol meant to achievemaximum disease severity. This suggests that any clinicalapplication of anatabine in humans may be effective onlyin treating patients with early and/or mild autoimmunethyroid disease.

Given the structure similarity with nicotine, we postu-lated that anatabine initiates its effects by binding to thenicotine receptor and modulating the cholinergic controlof inflammation (10, 24). The nicotinic receptor that hasbeen clearly associated with antiinflammatory responsesis the �7-homopentamer, classically found on neural cellsbut also on immune cells (12). Activation of the �7-nico-tinic receptor present in lymphocytes, dendritic cells, andmacrophages has been shown to suppress nuclear trans-location of NF-�B and transcription of high mobilitygroup box 1, ultimately decreasing danger signals thatinitiate inflammation (25). Consistent with this mecha-nism, Paris and colleagues demonstrated that anatabinesuppresses in a dose-dependent manner the transcriptionof NF-�B induced by tumor necrosis factor-� (15).

This study has elucidated another mechanistic pathwaythat could explain the antiinflammatory effects of anat-abine. Using a panel of 84 cytokines and chemokines, wefound that during experimental autoimmune thyroiditis,there is an up-regulation of thyroidal genes involved inattracting immune cells (e.g. CCL6 and CCL8), augment-ing the function of lymphocytes and dendritic cells (e.g.CCR2 and integrin-�M), and tissue remodeling (e.g. theextracellular structural protein Spp1). These genes werenot affected by anatabine administration. However, anat-abine suppressed the thyroidal expression of IL-18 andIL-1R2. IL-18, a member of the IL-1 family, is producedby activated macrophages and stimulates production ofinterferon-� from T cells and natural killer cells (26), over-all acting as a proinflammatory stimulus. IL-18 has beenshown to increase during thyroid inflammation both invitro (27) and in vivo (28) and, it has been suggested, may

contribute to tissue destruction in human autoimmunethyroiditis. The decreased IL-18 level with anatabine ad-ministration likely translates into decreased generation ofpathogenic Th1-type lymphoid responses and lesser his-topathological changes. The IL-1R2 is expressed princi-pally on macrophages and B cells in which it increases instates of ongoing inflammation to dampen the inflamma-tory process. It binds IL-1� and IL-1�, like the type 1 IL-1receptor, but does not convey an intracellular signal be-cause of its very short cytoplasmic tail, serving in essenceas a decoy receptor (29). The fact that the expression ofIL-1R2 in thyroids of anatabine treated mice with exper-imental autoimmune thyroiditis is significantly lower thanthat of controls and almost identical with the levels ob-served in unimmunized thyroids suggests that anatabineadministration is associated with an overall attenuation ofthe thyroid-specific autoimmune response.

Additional experiments will be needed to elucidate fullyanatabine’s mechanism(s) of action, which could includeeffects not mediated by the nicotinic receptor and path-ways not analyzed in the present study. Even if its antiin-flammatory effects are mediated purely through the nic-otinic receptor, it remains to be understood how theactivation of a fast-switching, on/off type of ion channelleads to the prolonged effects on the control of inflamma-tion (11).

In conclusion, these studies demonstrate that anat-abine, a minor alkaloid originally derived from solana-ceous plants, attenuates in a murine model histopatholog-ical, serological, and functional manifestations of thyroidautoimmunity. Further exploration of anatabine’s antiin-flammatory mechanisms and potential clinical utility areworthwhile.

Acknowledgments

We express our gratitude to Drs. Michael Mullan and FionaCrawford (Roskamp Institute, Sarasota, FL) for their guidanceand advice.

Address all correspondence and requests for reprints to: Pa-trizio Caturegli, M.D., M.P.H., Department of Pathology, JohnsHopkins University School of Medicine, Ross 656, 720 RutlandAvenue, Baltimore, Maryland 21205. E-mail: [email protected].

This work was supported by a grant from the Walton FamilyFoundation. S.I. was supported in part by a fellowship from theUehara Memorial Foundation (Tokyo, Japan).

Disclosure Summary: P.C. and P.W.L. are consultants forRock Creek Pharmaceuticals Inc. The other authors have noth-ing to declare.



References

1. Belin RM, Astor BC, Powe NR, Ladenson PW 2004 Smoke expo-sure is associated with a lower prevalence of serum thyroid auto-antibodies and thyrotropin concentration elevation and a higherprevalence of mild thyrotropin concentration suppression in thethird National Health and Nutrition Examination Survey(NHANES III). J Clin Endocrinol Metab 89:6077–6086

2. Strieder TG, Prummel MF, Tijssen JG, Endert E, Wiersinga WM2003 Risk factors for and prevalence of thyroid disorders in a cross-sectional study among healthy female relatives of patients with au-toimmune thyroid disease. Clin Endocrinol (Oxf) 59:396–401

3. Pedersen IB, Laurberg P, Knudsen N, Jørgensen T, Perrild H, OvesenL, Rasmussen LB 2008 Smoking is negatively associated with thepresence of thyroglobulin autoantibody and to a lesser degree withthyroid peroxidase autoantibody in serum: a population study. EurJ Endocrinol 158:367–373

4. Effraimidis G, Tijssen JG, Wiersinga WM 2009 Discontinuation ofsmoking increases the risk for developing thyroid peroxidase anti-bodies and/or thyroglobulin antibodies: a prospective study. J ClinEndocrinol Metab 94:1324–1328

5. Boyko EJ, Perera DR, Koepsell TD, Keane EM, Inui TS 1988 Effectsof cigarette smoking on the clinical course of ulcerative colitis. ScandJ Gastroenterol 23:1147–1152

6. Fraga XF, Vergara M, Medina C, Casellas F, Bermejo B, MalageladaJR 1997 Effects of smoking on the presentation and clinical courseof inflammatory bowel disease. Eur J Gastroenterol Hepatol 9:683–687

7. Mokbel M, Carbonnel F, Beaugerie L, Gendre JP, Cosnes J 1998[Effect of smoking on the long-term course of ulcerative colitis].Gastroenterol Clin Biol 22:858–862

8. Calabrese E, Yanai H, Shuster D, Rubin DT, Hanauer SB 2012Low-dose smoking resumption in ex-smokers with refractory ulcer-ative colitis. J Crohns Colitis 6:756–762

9. Sopori M 2002 Effects of cigarette smoke on the immune system.Nat Rev Immunol 2:372–377

10. Tracey KJ 2007 Physiology and immunology of the cholinergic an-tiinflammatory pathway. J Clin Invest 117:289–296

11. Miwa JM, Freedman R, Lester HA 2011 Neural systems governedby nicotinic acetylcholine receptors: emerging hypotheses. Neuron70:20–33

12. Gahring LC, Rogers SW 2005 Neuronal nicotinic acetylcholine re-ceptor expression and function on nonneuronal cells. AAPS J7:E885–E894

13. Bencherif M, Lippiello PM, Lucas R, Marrero MB 2011 �7 Nico-tinic receptors as novel therapeutic targets for inflammation-baseddiseases. Cell Mol Life Sci 68:931–949

14. Nizri E, Irony-Tur-Sinai M, Lory O, Orr-Urtreger A, Lavi E,Brenner T 2009 Activation of the cholinergic anti-inflammatorysystem by nicotine attenuates neuroinflammation via suppression ofTh1 and Th17 responses. J Immunol 183:6681–6688

15. Paris D, Beaulieu-Abdelahad D, Bachmeier C, Reed J, Ait-GhezalaG, Bishop A, Chao J, Mathura V, Crawford F, Mullan M 2011

Anatabine lowers Alzheimer’s A� production in vitro and in vivo.Eur J Pharmacol 670:384–391

16. Kimura H, Kimura M, Tzou SC, Chen YC, Suzuki K, Rose NR,Caturegli P 2005 Expression of class II major histocompatibilitycomplex molecules on thyrocytes does not cause spontaneous thy-roiditis but mildly increases its severity after immunization. Endo-crinology 146:1154–1162

17. Landek-Salgado MA, Tzou SC, Kimura H, Caturegli P 2010 Induc-tion of experimental autoimmune hypophysitis in SJL mice. J VisExp pii 2182

18. Reagan-Shaw S, Nihal M, Ahmad N 2008 Dose translation fromanimal to human studies revisited. FASEB J 22:659–661

19. Kimura H, Kimura M, Westra WH, Rose NR, Caturegli P 2005Increased thyroidal fat and goitrous hypothyroidism induced byinterferon-�. Int J Exp Pathol 86:97–106

20. Rocchi R, Kimura H, Tzou SC, Suzuki K, Rose NR, Pinchera A,Ladenson PW, Caturegli P 2007 Toll-like receptor-MyD88 and Fcreceptor pathways of mast cells mediate the thyroid dysfunctionsobserved during nonthyroidal illness. Proc Natl Acad Sci USA 104:6019–6024

21. Allen EM, Hsueh WC, Sabra MM, Pollin TI, Ladenson PW, SilverKD, Mitchell BD, Shuldiner AR 2003 A genome-wide scan for au-toimmune thyroiditis in the Old Order Amish: replication of geneticlinkage on chromosome 5q11.2-q14.3. J Clin Endocrinol Metab88:1292–1296

22. Hutfless S, Matos P, Talor MV, Caturegli P, Rose NR 2011 Signif-icance of prediagnostic thyroid antibodies in women with autoim-mune thyroid disease. J Clin Endocrinol Metab 96:E1466–E1471

23. Masters SL, Simon A, Aksentijevich I, Kastner DL 2009 Horrorautoinflammaticus: the molecular pathophysiology of autoinflam-matory disease (*). Annu Rev Immunol 27:621–668

24. Rosas-Ballina M, Tracey KJ 2009 Cholinergic control of inflam-mation. J Intern Med 265:663–679

25. Wang H, Liao H, Ochani M, Justiniani M, Lin X, Yang L, Al-AbedY, Wang H, Metz C, Miller EJ, Tracey KJ, Ulloa L 2004 Cholinergicagonists inhibit HMGB1 release and improve survival in experi-mental sepsis. Nat Med 10:1216–1221

26. Smith DE 2011 The biological paths of IL-1 family members IL-18and IL-33. J Leukoc Biol 89:383–392

27. Takiyama Y, Miyokawa N, Tokusashi Y, Ito K, Kato S, Kimura S,Sato K, Katagiri M 2002 Thyroid-stimulating hormone induces in-terleukin-18 gene expression in FRTL-5 cells: immunohistochemi-cal detection of interleukin-18 in autoimmune thyroid disease. Thy-roid 12:935–943

28. Liu Z, Wang H, Xiao W, Wang C, Liu G, Hong T 2010 Thyrocyteinterleukin-18 expression is up-regulated by interferon-� and maycontribute to thyroid destruction in Hashimoto’s thyroiditis. Int JExp Pathol 91:420–425

29. Gabay C, Lamacchia C, Palmer G 2010 IL-1 pathways in inflam-mation and human diseases. Nat Rev Rheumatol 6:232–241

30. Iwama S, De Remigis A, Bishop A, Kimura H, Caturegli P 2012Hürthle cells predict hypothyroidism in interferon-� transgenic miceof different genetic backgrounds. Endocrinology 153:4059–4066



Documents

Anatabine Ameliorates Experimental Autoimmune Thyroiditisinsanemedicine.com/wp-content/...Experimental-Autoimmune-thyroid… · developed thyroid infiltrates when drinking anatabine