Nicotinic Attenuation of Central Nervous System Inflammation and Autoimmunity¹

A multitude of studies suggest that nicotine, a psychoactive component of tobacco products that acts as does the natural neurotransmitter, acetylcholine, on nicotinic acetylcholine receptors (nAChRs)⁷ found in many organ systems, has profound immunological effects (1, 2). During ontogeny, nicotine exposure can modulate T (3) and B cell development and activation (4). Nicotine exposure also suppresses the T cell response and alters the differentiation, phenotype, and functions of APCs including dendritic cells (5, 6) and macrophages (7). Findings have been mixed about effects of nicotine exposure on immune responses in vivo. Exposure to nicotine and/or related agents appears to dampen inflammatory responses and reduce mortality in a mouse model of sepsis (8) and to protect against induction of type 1 diabetes in mice (9). In addition, cigarette smoke inhalation produces sustained suppression of humoral autoimmunity in a murine model of systemic lupus erythematosus (10). Moreover, several epidemiological studies reveal a strong inverse correlation between smoking and human autoimmune responses manifest as systemic lupus erythematosus and ulcerative colitis (10, 11). By contrast, other studies suggest that smoking behavior in humans might exacerbate multiple sclerosis (MS) and Crohn’s disease (12, 13, 14). These apparently conflicting observations suggest that the impact of nicotine on immune responses in vivo is complex, being potentially influenced by drug dosage and duration of exposure, by the specific organ systems involved in the immune response, by the stage and type of disease, and by the level of involvement of autoimmune and inflammatory mechanisms.

Inflammatory and immune responses within the CNS are capable of shaping the clinical outcome of brain diseases including stroke, trauma, Alzheimer’s disease, Parkinson’s disease, epilepsy, encephalomyelitis, and MS (15). Compared with other organ systems, the CNS has several unique properties with respect to immune responses. First, the spectrum of APCs differs from that in the periphery; in the CNS, resident microglia and astrocytes are active participants (16, 17). Second, peripheral immune cells migrating into the CNS are reactivated upon encountering myelin and other Ags, enhancing their capacity to recognize a wide spectrum of ambient Ags through “determinant spreading” (18). Third, the nature and magnitude of immune responses within the CNS are likely influenced by signals intrinsic to this unique microenvironment, including local cytokine signaling, but also including presumably higher levels and more concentrated signaling by chemical messengers such as acetylcholine. More work is needed to elucidate roles of nAChRs and nicotinic cholinergic signaling on immune responses in the periphery, but this need for further investigation may be even more acute for studies of immune and inflammatory responses in the CNS where additional cell types express nAChRs, especially given implication of nAChRs and nicotinic signaling in stroke, Alzheimer’s disease, Parkinson’s disease, epilepsy, and perhaps other brain diseases (19, 20).

To ascertain influences of nicotine exposure on a CNS autoimmune response, we used the murine experimental autoimmune encephalomyelitis (EAE) model of MS. Immunization of C57BL/6 (B6) mice with myelin oligodendrocyte glycoprotein (MOG) peptide activates T cells in the periphery and subsequently generates pronounced cellular infiltration and demyelination in the CNS and a monophasic neurological deficiency that resembles a form of MS in humans, acute disseminated encephalomyelitis (21). We found that nicotine exposure delays and dramatically attenuates CNS inflammation and autoimmune responsiveness to myelin Ags, suggesting several novel mechanisms of neuroimmune interaction.

B6 (H-2^b) mice purchased from Taconic Farms were housed in pathogen-free animal facilities. Female mice used were 7–8 wk of age at the experiment’s inception. Experiments were conducted in accordance with institutional guidelines.

The murine MOG_35–55 peptide (M-E-V-G-W-Y-R-S-P-F-S-R-V-V-H-l-Y-R-N-G-K) and proteolipid protein (PLP)_139–151 peptide (H-S-l-G-K-W-l-G-H-P-D-K-F) were synthesized by BioSynthesis.

To induce acute EAE, B6 mice were injected s.c. in the hind flank with 200 μg of MOG_35–55 peptide in CFA (Difco) containing 500 μg of nonviable, desiccated Mycobacterium tuberculosis. On the day of and 2 days after immunization, the mice also were inoculated with 200 ng of pertussis toxin (List Biologic) i.p.

For adoptive transfer of EAE (passively induced EAE), lymph node cells were extracted from primary inoculants described above on day 8 after immunization, and cells were cultured at a density of 2 × 10⁶/ml in Click’s Eagle’s Ham’s amino acids medium supplemented with 15% FCS, 20 ng/ml rIL-12, and 50 μg/ml MOG_35–55 peptide. After 4 days of culture, cells were harvested, and 3 × 10⁷ viable cells were injected i.p. into each recipient mouse irradiated at 350 rad 1 h earlier (22). For both actively and passively induced EAE, the mice were monitored daily for symptoms scored on an arbitrary scale of 0–5 with 0.5 increments: 0, no symptoms; 1, flaccid tail; 2, hind limb weakness or abnormal gait; 3, complete hind limb paralysis; 4, complete hind limb paralysis with forelimb weakness or paralysis; 5, moribund or deceased. Values presented indicate averages of disease symptom scores on a given day, the absolute maximum score seen for all animals tested, the mean score for all animals on each day and from the time of initial presentation of symptoms until termination of the experiment, or the terminal (day of the end of the experiment) disease symptom score.

Nicotine bitartrate was purchased from Sigma-Aldrich. A 100 mg/ml solution in PBS or a solution of PBS alone was freshly prepared 24 h before pump implantation and loaded into Alzet osmotic minipumps (model 1007D, Durect Corporation). The pumps were implanted s.c. on the right side of the back of the mouse and continuously delivered either PBS or nicotine salt at 12 μl/day for 7 days, and then the pumps were removed. This equated to delivery of 0.39 mg of nicotine-free base per mouse per day. For an ∼30 g mouse, which is at the upper end of weight for animals used in the study, this equates to ∼13 mg of nicotine-free base/kg/day or ∼0.54 mg of nicotine free base/kg/h. Plasma nicotine levels in mice are ∼100–200 ng/ml (∼0.6–1.2 mM) after infusion of ∼2–4 mg/kg/h of drug and ∼45 ng/ml (∼280 nM) after infusion at ∼0.5 mg/kg/h (23). For comparison, human smokers have peak plasma nicotine levels of 10–50 ng/ml (∼60–310 nM; Ref. 23). Thus, nicotine levels in plasma (extrapolated to be ∼49 ng/ml or ∼300 nM) of mice used in the studies are comparable to those in the plasma of human smokers. Some control mice received PBS via direct injections rather than through minipumps, but either delivery method produced similar results.

To evaluate the effects of nicotine pretreatment on EAE-associated autoimmune responses, mice received nicotine or PBS daily for 7 days starting on the day of or 7 days before MOG immunization. To analyze effects of nicotine exposure on an activated autoimmune response, mice received nicotine or PBS daily for 7 days starting on day 7, which is the first day of manifestation of disease signs, after EAE induction. For the adoptive EAE transfer study, lymphocytes used for secondary injection were isolated from mice treated for 7 days with nicotine or PBS initiated at the time of MOG immunization.

Mice were anesthetized with pentobarbital on the 25th day after immunization and perfused by intracardiac puncture with 50 ml of cold PBS. Spinal cords were removed and fixed in 10% formalin/PBS. Paraffin-embedded, longitudinal sections running from the cervical enlargement of the cord were prepared and stained for H&E, myelin (luxol fast blue), and axons (Biechowsky silver). Manual tracing was used to define the degree of inflammation, demyelination, and axonal damage across the entire spinal cord section for each mouse. Pathological changes in each spinal cord were scored as follows: 0, no changes; 1, focal area involvement; 2, <5% of total myelin area involved; 3, 5–10% of total myelin area involved; 4, 10–20% involved area; 5, >20% of total myelin area involved (22).

Mononuclear cells were isolated from the spleens of EAE mice at day 11 post immunization that were treated with nicotine or PBS. Cells were suspended in culture medium containing DMEM (Life Technologies) supplemented with 1% (v/v) MEM (Life Technologies), 2 mM glutamine (Flow Laboratory), 50 IU/ml penicillin, 50 mg/ml streptomycin, and with 10% (v/v) FCS (all from Life Technologies). Four × 10⁵ cells in 200 μl of culture medium were placed in each well of 96-well, round-bottom microtiter plates (Nunc). Ten microliters of immunizing Ag MOG_35–55 peptide (10 μg/ml), control myelin Ag PLP₁₃₉-₁₅₁ peptide (10 μg/ml), or Con A (5 μg/ml) (Sigma-Aldrich) were then added (triplicates per condition). After 3 days of incubation, the cells were pulsed for 18 h with 10-μl aliquots containing 1 μCi of [methyl-³H]thymidine (specific activity of 42 Ci/mmol; MP Biomedicals) per well. Cells were harvested onto glass fiber filters and thymidine incorporation proportional to the degree of cell proliferation was then measured. The results are expressed as cpm.

Single cell suspensions (4 × 10⁷ cells) were prepared and labeled with 0.5 μM CFSE at 37°C for 10 min. Subsequently, when cells were cultured, levels of CFSE staining declined with each cell division, allowing for cell proliferation to be monitored. Cells with or without CFSE were incubated at 37°C for 3 days in round-bottom plates (2 × 10⁶ cells/well) with or without Ags (MOG 10 μg/ml). After harvesting, cells were stained for surface markers with fluorochrome-conjugated mAbs including anti-CD3-PE/Cy5 (17A2), anti-CD4-allophycocyanin/Cy7 (GK1.5), and anti-CD8α-PE/Cy7 (53-6.7) (BD Biosciences). Isotype-matched negative mAbs were used as controls.

Cell viability was assessed by trypan blue dye exclusion. For detection of cell apoptosis, spleen mononuclear cell suspensions were collected from PBS- or nicotine-treated mice on day 11 (peak stage). Single cell suspensions were washed in PBS and resuspended in binding buffer containing annexin V-FITC to monitor apoptosis- or necrosis-associated plasma membrane alterations and propidium iodide to monitor cell death-associated DNA exposure (both from BD Biosciences) for 20 min at room temperature. The samples were analyzed on a FACSAria using Diva software (BD Biosciences).

Spleen mononuclear cell suspensions were collected from PBS- or nicotine-treated mice on day 11 (i.e., at the peak of the EAE response). Single cell suspensions were prepared and stained for one or more of the following Ags (targeted by the indicated Ab fluorescently tagged with either FITC, PE, allophycocyanin, PE-Cy5, or PE-Cy7): CD25 (PC61.5), CD3 (17A2), CD4 (GK1.5), CD8 (53–6.7), NK1.1 (PK136), CD11b (M1/70), CD11c (HL3), CD19 (1D3), CD80 (16-10A1), CD86 (GL1), and MHC class II (M5/114.15). Intracellular Foxp3 (FJK-16s) staining was performed according to the manufacturer’s protocol (eBioscience). Appropriate isotype controls were always included. All samples were analyzed on a FACSAria using Diva. The absolute number of a particular cell subset was calculated based on the percentage of cells stained for the appropriate markers determined by FACSAria flow cytometry and the number of mononuclear cells per mouse spleen defined based on hemocytometer counts.

For CNS cell isolates, at day 11 after EAE induction, mice were sacrificed and perfused with PBS delivered transcardially to eliminate contaminating blood cells in the CNS. CNS mononuclear cells were then isolated from the brains and spinal cords of five to six mice based on their characteristic sedimentation features on Percoll density gradients (30∼70%) and stained for cell surface markers as for splenocytes. Abs were directly labeled and analyzed as done for splenocytes. Absolute numbers and percentages of particular CNS mononuclear cell subsets were determined as described above and in Ref. 22 .

Single cell suspensions were incubated at 37°C for 3 days in round-bottom plates (2 × 10⁶ cells/well) with or without Ags (MOG 10 μg/ml, PLP 10 μg/ml, or Con A 2.5 μg/ml) and then stimulated with PMA (20 ng/ml)/ionomycin (1 μg/ml)/brefeldin A (5 μg/ml) for 5 h at 37°C. After harvesting, cells were stained for surface markers with fluorochrome-conjugated mAbs targeting CD3, CD4, and/or CD8 as described above and/or for intracellular cytokines using anti-IFN-γ, anti-IL-4, anti-IL-10, or anti-IL-17 mAbs conjugated with Alexa 647 after fixation and permeabilization using Cytofix/Cytoperm kit (BD Biosciences). All samples were analyzed on a FACSAria using Diva. For assessment of effects of treatments on cytokine induction, supernatants were collected 3 days after in vitro boosting. IFN-γ, IL-10, IL-2, and TGF-β were measured by optEIA kits (BD Pharmingen and eBioscience).

MOG-reactive Abs were quantified using ELISA. In brief, microtiter plates (Corning Glass Works) were coated with 100 μl/well of murine MOG_35–55 (10 μg/ml) overnight and blocked with 10% FBS (4°C). Mouse serum samples obtained on day 11 (peak responses after EAE induction) were then added to the plates and incubated overnight at 4°C. Plates were then incubated for 2 h with biotinylated rabbit anti-mouse IgG, IgG1, IgG2a, IgA, IgG3, or IgG2b (Invitrogen) followed by addition of alkaline phosphatase-conjugated ABC reagent (Dakopatts; R&D Systems). Immune complexes were visualized colorimetrically after exposure to p-nitrophenyl-phosphate and quantified based on assessments of OD at 450 nm.

Differences between groups were evaluated by performing ANOVA. Fisher’s exact test and Mann-Whitney’s U test were applied to analyze disease incidence and severity, respectively, of symptomatic manifestations.

To determine how nicotine might influence the course of EAE, B6 mice were infused using an implantable minipump with PBS or with nicotine at a dose of ∼13 mg/kg/day for 7 days. The choice on dosage and route of administration was based on recently published guidelines for testing the effects of nicotine in vivo (23) and to create in mice plasma nicotine levels comparable to those found in human cigarette smokers (∼300 nM). Nicotine delivery under these conditions had no observable effects on animal behavior.

A single injection of MOG_35–55 peptide in CFA augmented with pertussis toxin induced moderate to severe EAE in untreated animals or in animals receiving control infusions of PBS only (maximum disease symptom score 4.25 ± 0.80, mean disease symptom score over days 7–26 of 2.99 ± 0.67; Fig. 1). The average time after immunization to initial presentation of disease symptoms was 7.6 ± 0.53 days in these control animals (Fig. 1,A). Disease symptoms subsequently increased, with paralysis being evident at days 9 to 14 after immunization, before control animals showed a slow and incomplete recovery stabilizing ∼25–26 days after immunization (i.e., at termination of the experiment; terminal disease symptom score of ∼2.75 ± 0.25; Fig. 1). By contrast, mice receiving nicotine 7 days before or on the day of immunization had a delayed onset of EAE, with disease symptoms first becoming evident 11 days after immunization and peaking 16–18 days after immunization (Fig. 1,A). Moreover, severity of EAE was attenuated in these nicotine-treated animals, reaching mean maximum disease symptom scores of 3.17 ± 0.78 for the pretreatment group (Fig. 1,A) and 3.13 ± 0.25 for the nicotine cotreatment group (Fig. 1,A). Differences in times to initial presentation, times to peak symptoms, and maximum or mean disease severities between controls receiving PBS and nicotine pre- or cotreated animals all were statistically significant (p < 0.01). Also striking was the ability of nicotine pre- or cotreatment to accelerate the rate of recovery and to increase the extent of recovery from EAE (Fig. 1 A). Disease symptom scores fell to <1.0 by the experiments’ termination (∼25–26 days after immunization; disease symptom scores were 0.85 ± 0.25 for nicotine cotreatment or 0.83 ± 0.26 for nicotine pretreatment; p < 0.01 compared with PBS controls). Thus, nicotine exposure preceding or simultaneous with disease induction significantly delayed EAE onset, attenuated EAE severity, and promoted disease recovery.

If nicotine treatment began 7 days after immunization, i.e., when disease symptoms (flaccid tail, hind limb weakness, etc.) typically are first evident and when myelin-reactive T cells have already become activated, there was no striking difference in the time to peak effect (Fig. 1,A). However, the maximum and mean severities of EAE were attenuated (3.97 ± 0.26 and 2.99 ± 0.67, respectively, for controls; 3.17 ± 0.24 and 1.89 ± 0.29, respectively, for delayed nicotine treatment; p < 0.05), and disease severity at termination of the experiment was even more reduced (2.75 ± 0.25 for controls and 0.49 ± 0.01 for delayed nicotine treatment; p < 0.01; Fig. 1 A). Therefore, attenuation of disease severity and acceleration of recovery also were evident even if nicotine exposure was delayed until the day when disease symptoms first became evident and even when myelin-reactive T cells were activated.

EAE is a T cell mediated disease that can be produced in recipients of encephalitogenic T cells (24). Thus, we performed adoptive transfer experiments in which T cells from immunized animals that were exposed to PBS only or nicotine starting at the time of immunization was introduced into naive animals, and evolution of EAE in recipients was monitored. Robust development of EAE was evident in control animals. The maximum and mean disease severity scores in these animals were 5.0 ± 0.25 and 4.22 ± 0.36, respectively (Fig. 1,B). By contrast, disease severity scores were 2.5 ± 0.82, p < 0.001 (maximum) and 1.0 ± 0.15, p < 0.001 (mean) for recipients from nicotine-treated animals (Fig. 1,B). Moreover, whereas there was no recovery at all over the course of the study in recipients of encephalitogenic T cells, recipients of cells from animals immunized and treated with nicotine showed nearly complete resolution of symptoms by the end of the experiment (Fig. 1 B). These results showing protection against adoptive transfer of EAE suggest that nicotine exposure may modulate (directly or indirectly) T cell properties/functions.

Pronounced cellular infiltration, demyelination, and axonal damage are pathological hallmarks of EAE and MS. To evaluate whether nicotine can alter these pathological changes, we conducted histological examinations of spinal cords isolated 25 days after immunization from control mice immunized to produce EAE and from mice immunized but also exposed to nicotine for 7 days starting on the day of immunization.

Examination of white matter in longitudinal sections of these tissues from control mice stained with H&E revealed marked multifocal and lymphohistiocytic inflammation that was both perivascular and diffuse (Fig. 1,C). Myelin loss as revealed by luxol fast blue staining was widespread, especially around inflamed areas (Fig. 1,D). In sharp contrast, most sections from nicotine-treated mice had few infiltrating cells, and myelin sheets were largely preserved (Fig. 1, F and G). Furthermore, axonal damage revealed by silver staining was clearly present in the submeningeal areas of PBS control mice, whereas axons from nicotine-treated mice were scarcely affected and then only in regions immediately surrounding foci of inflammation (Fig. 1, E and H). Quantitative analysis of histological indices showed that inflammation, demyelination, and axonal degeneration were significantly greater in PBS-treated compared with nicotine-treated mice (Fig. 1 I; p < 0.01). Therefore, nicotine exposure protected the CNS against inflammation, demyelination, and axonal damage that are pathological hallmarks of EAE.

Nicotine exposure can modify T and B cell development and survival, although a consensus has yet to develop regarding mechanisms involved and how differences observed may relate to the degree of cell maturity during drug exposure (25). To define effects of nicotine exposure on lymphocyte status during an autoimmune response to MOG, we quantified various lymphocyte subpopulations based on cellular phenotype from mice immunized with MOG under control conditions or in concert with nicotine treatment commenced on the day of MOG immunization (day 0). Results obtained for animals treated with nicotine 7 days before of after immunization were similar and so are not detailed in this study.

Compared with control EAE mice, nicotine-treated mice had various degrees of reductions in the percentages and numbers of CD3⁺, CD4⁺, or CD8⁺ T cells or CD19⁺CD3⁻ B cells among splenocytes sampled on day 11 post immunization (Fig. 2, A–D; p < 0.05). Concurrently, there were no significant effects of nicotine treatment on numbers or proportions of NK cells (CD3⁻NK1.1⁺), NKT cells (CD3⁺NK1.1⁻), or CD4⁺CD25⁺ regulatory T cells (Fig. 2,B and 3A). Interestingly, the expression of Foxp3 was clearly augmented in the nicotine-treated mice (Fig. 3, B and C).

Although the numbers and percentages of peripheral CD11b⁺ mononuclear/microglial (Fig. 4,A) and CD11c⁺ dendritic (Fig. 4,B) cells were not dramatically altered by nicotine treatment, reductions in the expression of MHC class II, CD80, and CD86 were notable on these cells (Fig. 4). Similar data were also obtained from assessments of cells derived from lymph nodes or peripheral blood and at several other time points during the course of EAE (data not shown).

To address how nicotine affected expansion of MOG_35–55-specific T cell responses in nicotine-treated EAE mice, splenic mononuclear cells were isolated at peak of the automimmune response (11 days post immunization) from animals immunized on the same day as initiation of control or nicotine treatment. T cell proliferation in response to myelin Ags was quantified by measuring [³H]lthymidine incorporation and based on CFSE assays. Compared with baseline proliferation responses in tissue culture medium, there were slightly but not significantly higher responses in both control and nicotine-treated groups to PLP, but responses to MOG in nicotine-treated animals were no higher and were significantly (p < 0.05) lower than responses of cells from control animals to MOG challenge (Fig. 5,A). Similarly, there were fewer CD3⁺, CD4⁺, and CD8⁺ T cells with lower levels of CFSE staining, indicative of a reduced T cell proliferative response to MOG in nicotine-treated mice (Fig. 5,B). By contrast, annexin V and propidium iodide double staining for cells undergoing apoptosis or necrosis revealed no increase in cell death upon nicotine exposure, at least at the current dose (Fig. 5,C). Thus, the reduced numbers of several lymphocyte subpopulations recorded in Fig. 3 were likely the result of a selective decline in proliferation related to nicotine exposure rather than an increase in apoptosis.

To further characterize effects of nicotine exposure on autoreactive T cells, splenocytes isolated at the peak of the EAE response (11 days after immunization) from animals subjected to control PBS or nicotine treatment starting on the day of immunization were then subjected to 3 days of cell culture alone or in the presence of Ags (MOG 10 μg/ml, PLP 10 μg/ml, or Con A 2.5 μg/ml) and then stimulated with PMA (20 ng/ml)/ionomycin (1 μg/ml)/brefeldin A (5 μg/ml) for 5 h. After harvesting, cells were stained for cell surface markers and/or fixed and permeabilized to allow staining for intracellular cytokines, all of which were quantified using flow cytometry, and cell culture supernatants were collected and assayed for cytokine release by enzyme immunoassays.

Secreted IFN-γ by splenocytes isolated from nicotine-treated animals was reduced (∼64 ± 3.0 ng/ml compared with ∼166 ± 32 ng/ml in controls; p < 0.01 at day 11; ∼96 ± 35 ng/ml compared with ∼34 ± 12 ng/ml in controls; p < 0.05 at day 26) (Table I). The reduction in secreted IFN-γ by cells isolated from nicotine-exposed animals was most marked in CD8⁺ compared with CD4⁺ cells (Table II). There was no significant effect of nicotine exposure in IL-17 secretion (Table Ι and Table ΙΙ); secreted IL-2 was reduced (Table I). However, secreted IL-10 or TGF-β1 levels were increased from cells derived from nicotine-treated animals (Table I and Table II). Production of IL-4 was undetectable. The results imply nicotine treatment induced a shift from Th1 toward Th2 or Th3 responses.

Autoantigen responses are influenced by helper T cell levels and actions. For example, in mice, IFN-γ secreted from Th1 cells is believed to drive the IgG2b response, whereas the IgG1 response is driven by Th2 cells principally involved in Ab-mediated immunity and help direct toward B cells. We used ELISAs to measure MOG-specific Abs of specific isotypes present in the sera of animals 11 days after immunization and initiation of control or nicotine exposure. IgG2b production was halved in nicotine-treated mice, whereas levels of IgG1 and IgA were significantly increased >3-fold and >25%, respectively (Fig. 6). The decrease in IgG2b is consistent with the decrease in IFN-γ with nicotine exposure and a blunted Th1 cell response.

As described earlier, histopathological studies revealed that there are less inflammatory infiltration in CNS tissues from nicotine-treated animals than in spinal cord tissues from PBS-treated control animals. To assess the impact of nicotine exposure on the CNS immune cell profile, animals at day 11 after EAE induction and initiation of control or nicotine treatment were cleared of blood cells in the CNS and used to isolate CNS mononuclear cells (i.e., resident microglial cells and monocytes that had migrated into the CNS from the blood). Cells were then stained for surface markers to define absolute numbers of and percentages of specific, CNS mononuclear cell subsets using flow cytometry.

CD3⁺, CD4⁺, and CD8⁺ T cells and CD3⁻CD19⁺ B cells were abundant in the CNS from PBS-treated EAE mice, but numbers of all of these cells were drastically (≥4-fold) reduced in central tissues from nicotine-treated animals (Fig. 7).

We also examined APCs in the CNS during EAE, including CD11c⁺ dendritic cell and CD11b⁺ macrophage/microglial cell (CD11b⁺CD45⁺) populations. CD45 was initially used to distinguish microglia from infiltrating macrophages. However, upon activation, expression of CD45 no long clearly distinguishes the two cell types (26). CD11b⁺CD45⁺ and CD11c⁺ cells were dramatically (∼10-fold) reduced in CNS tissues from nicotine-exposed animals (Fig. 8, A and B). There also were reductions (∼2–4-fold in proportions and ≥10-fold in absolute numbers) of MHC class II⁺, CD80⁺, and CD86⁺ levels on CD11b⁺ cells and >10-fold reductions in levels of those markers on CD11c⁺ cells with nicotine treatment (Fig. 8, C–F). Importantly, the magnitude of reduction in nicotine-exposed animals of CD80 and CD86 levels on CD11b⁺ or CD11c⁺ cells was much greater for CNS than for peripheral monocytes (see Fig. 4). This suggests that nicotine exposure may have distinct immunomodulatory effects in different anatomic compartments, perhaps reflecting different mechanisms beyond different magnitudes of effect.

Expression of nicotinic acetylcholine receptors on nonneuronal cells such as APCs and other immune system cells underscores the idea that nAChRs have functions well beyond mediation of nicotinic cholinergic neurotransmission. Although considerable evidence indicates that nicotine exerts anti-inflammatory and immune modulatory effects in vivo (2), its specific roles in modulating CNS inflammation and autoimmune responses are not well defined. Using the EAE model, we have now demonstrated that nicotine exposure can dramatically delay disease onset, attenuate disease severity, and facilitate recovery. This occurs whether nicotine treatment begins before, at the time of, or after immunization with myelin Ags to induce EAE. Moreover, nicotine exposure also suppresses disease development on adoptive transfer of autoimmune T cells. At the histopathological level, nicotine exposure curtails the infiltration of inflammatory cells into the CNS and the destruction of myelin and axons. These results thus reveal new aspects of nicotine’s functions as a potent immunomodulator, suggest novel roles of nAChRs in those processes, and highlight the importance of understanding interactions between nervous and immune systems, especially in the context of development and treatment of CNS inflammatory disorders.

The brain and spinal cord are considered immunologically privileged organs in that, under normal physiological circumstances, entry of lymphocytes and inflammatory mediators into the CNS is blocked by the blood/brain barrier (27). Nevertheless, the initiation of EAE, and perhaps MS, requires activation of T cells in peripheral lymphoid organs and homing of myelin-reactive T cells as well as APCs to the CNS. Aided by local APCs, such as microglia and astrocytes, infiltrating T cells undergo reactivation and, in concert with other cellular and soluble components of the immune system, orchestrate the induction of CNS pathology (16, 17). Mechanisms underlying the dramatic effects of nicotine in our model can be multifactorial. First, we demonstrated that the expansion of MOG-reactive T cells from the spleen in nicotine-treated mice was significantly dampened. In these animals, MOG-reactive Th cells produced less IFN-γ and IL-2 than cells from PBS-treated controls, whereas production of IL-10, and particularly TGF-β, was augmented. A marginal, but not significant reduction of IL-17 was observed in mice that received nicotine. Our observation is somewhat surprising given the augmentation of TGF-β in nicotine-exposed mice. We reasoned that the timing of TGF-β production might contribute to lack of effect of Th-17 development. Nicotine also did not appear to induce apoptosis in autoreactive T cells. This outcome invites the prediction that the immunological effects induced by nicotine may have contributed to the decreased T cell proliferation and altered cytokine profile. Increased production of TGF-β, by itself, can suppress MOG-reactive T cells. In contrast, TGF-β may promote the generation of CD4⁺CD25⁺ regulatory T cells (28, 29). Indeed, we observed that, although the absolute numbers of CD4⁺CD25⁺ regulatory T cells were not dramatically altered by nicotine exposure, expression of FoxP3 was significantly up-regulated. These regulatory T cells with enhanced FoxP3 expression may contribute to the suppression of T effector/autoreative cells.

The other factor that contributes to the decreased expansion and a shift in cytokine profile of MOG-reactive T cells is an altered APC phenotype in nicotine-treated animals. Consistent with previous studies (30), we found that nicotine significantly reduced levels of MHC class II, CD80, and CD86 expression on peripheral CD11c⁺ and CD11b⁺ cells. Notably, these changes were more dramatic for CD11b⁺ cells. In an experimental sepsis model, Wang and colleagues reported that nicotine prevented activation of the NF-κB pathway and inhibited HMGB1 secretion from macrophages, which appear to be responsible for improved survival of the animals (8). It has been well established that a specific type of APC can direct differentiation of Th cells to produce regulatory cytokines and moderate immune responses (31, 32, 33, 34). It is highly likely that the altered APC phenotype in nicotine-treated animals may, at least in part, reduce the encephalitogenic capacity of MOG-reactive T cells in the EAE model.

In sharp contrast with control EAE mice, nicotine-treated animals had relatively few cellular infiltrates in CNS. Further, flow cytometry analysis of the cellular infiltrates showed a significant reduction of CD4⁺, CD8⁺, CD19⁺, CD11c⁺, CD11b⁺, and CD11b⁺CD45⁺cell populations, and the reductions in CD19⁺ B cells and CD11c⁺ dendritic cells seem to reflect diminished migration into the CNS from the periphery. It is presently unclear whether the reduction of CD4⁺ and CD8⁺ cells in the CNS stems from reduced influx from the periphery, impaired expansion in the CNS after migration, or both. Whatever the mechanism, it is clear that there is significantly reduced expression of Ag presentation machinery by resident or infiltrating CD11c⁺ and CD11b⁺ cells.

There are several possible ways for nicotine exposure to affect disease symptoms, even when applied after EAE has been initiated and initial presentation of CNS symptoms recover after the peak stage. First, nicotine may inhibit myelin-reactive T cell determinant spreading when T cells migrating from the periphery encounter CNS Ags. Second, a large percentage of C57BL/6 mice will spontaneously recover after peak stage of neurological deficit via unknown mechanisms that may be particularly sensitive to nicotinic action. Third, nicotine may directly or indirectly facilitate function of oligodendrocytes, myelin-forming cells, to stimulate regeneration after inflammatory insults. These possibilities are currently under investigation in our laboratory.

In agreement with previous studies that nicotine could suppress the migration of leukocytes to the inflammation/infection site (35), we also demonstrated that there was a dramatic reduction in overall inflammatory cell numbers, including monocytes that migrated from the periphery and CNS microglial cells. Considerable evidence suggests that microglia have multiple and sometimes contrasting functions in inflammatory and degenerative disorders of the CNS (36, 37, 38, 39, 40); not only can microglia only serve as APCs, but they also possess neurotoxic or neuroprotective activities (17, 41, 42, 43). The factors dictating which of these two effects are exerted by microglia are currently unknown. However, the preservation of myelin and axons during EAE in nicotine-treated animal could be in part due to protection against microglial neurotoxicity in addition to attenuation of the inflammatory response in the CNS.

Nicotine dosing in the current study was designed to mimic in mouse plasma levels of nicotine achieved in the typical human cigarette smoker, and the EAE model is a surrogate for at least some forms of MS. More specifically, EAE most closely resembles the human demyelinating disorder, acute disseminated encephalomyelitis, one of the subforms of MS. Further studies are required to verify whether nicotine’s effects on EAE can be generalized to other forms CNS disorders, including MS. The literature on relationships between tobacco product use and autoimmune MS is mixed. Some epidemiological studies suggested that smoking was associated with aggravation of MS symptoms (44, 45, 46, 47), whereas other studies documented no influence of smoking on disease progression and MS severity (48). It is only recently that a consensus developed that the risk for developing Parkinson’s disease is about one half in smokers of the risk in nonsmokers (19, 20), and perhaps similar studies should be initiated with regard to risks for developing MS. It is important to note that nicotine exposure is not the only factor influencing health in smokers; many other elements in tobacco could alter nervous and immune system function (2, 49, 50, 51). For example, acrolein affects neutrophil function and decreases the resistance of the lungs to infections (52, 53). Hydroquinone has been shown to inhibit the activation and proliferation of T cell (54, 55). Chronic exposure to benzo[a]pyrene induces dose-related decreases in the mass and cellularity of lymphoid tissues, and maternal exposure to benzo[a]pyrene alters the development of T cells and immune responses in the offspring (55, 56). Many compounds are associated with brain toxicity including vinyl chloride, arsenic, and hydrogen cyanide (57, 58). Also, the timing of nicotine exposure or smoking behavior with regard to disease stage has not been considered. However, it does make sense that if pure nicotine acts as an immunosuppressive agent, under the right circumstances, that activity could be leveraged.

Our study provides evidence that nicotine exposure can prevent the loss of tolerance to myelin Ags. Nicotine may have these effects by acting on multiple steps in the autoimmune response. Expression of α7 nicotinic acetylcholine receptors by CNS microglia and/or astrocytes invites the prediction that this receptor subtype may mediate the effects of nicotine observed in the EAE model. Further elucidation of roles played by α7 nicotinic receptors and other potential entities involved in mediating beneficial effects of nicotine on EAE promises to illuminate novel and potentially superior strategies for treatment of human neuroinflammatory and neurodegenerative diseases.

We thank X. Bai for histological analysis; J. Wu, A. Simard, and P. Whiteaker for fruitful discussions and advice; S. Miller, L. Lucero, and S. Rhodes for technical assistances; and P. Minick for editing the manuscript.

The authors have no financial conflict of interest.