Nicotine self-administration suppresses both pro- and anti-inflammatory accumbens cytokines but does not induce apoptosis in female rats

Abstract

Nicotine can alter neuroimmune signaling within the brain reward pathway, and specifically, within the nucleus accumbens (NAc). Despite clinically meaningful sex differences in nicotine use as well as known sex differences in immune system function, no studies have evaluated NAc neuroimmune consequences following nicotine use in females. Thus, the present study evaluated cytokine expression and neuronal apoptosis in the ventral striatum (which encompasses the NAc) in ovary-intact female Long-Evans rats following nicotine or saline self-administration. Circulating cytokine levels were also evaluated in serum. Overall, pro-inflammatory cytokine levels were suppressed within the ventral striatum, with tumor necrosis factor-alpha (TNFα) being the most significantly suppressed. Interestingly, there was an increase in the chemokine fractalkine (FKN), a neuronal “help me” signal during distress. These results raise the possibility that chronic volitional nicotine use results in cleavage of FKN in response to neuronal distress, and glial cells may be unresponsive to these signals. In contrast, circulating serum cytokine levels remained mostly unchanged except for an increase in TNFα, suggesting that the immunosuppression observed following nicotine self-administration was largely localized to the brain. No measurable NAc apoptotic debris was identified, in line with our findings that interleukin-1β (IL-1β) was suppressed and raising the possibility that nicotine use suppresses NAc inflammasome activation. Thus, chronic nicotine use in female rats may disrupt ventral striatal neuroimmune homeostasis. Together, targeting nicotine-induced disruptions in neuroimmune homeostasis may represent novel therapeutic opportunities to mitigate the harmful effects of nicotine in females.

1. Introduction

Cigarette smoking remains the leading cause of preventable death and disease in the United States, causing more than 480,000 deaths annually and reducing life expectancy by at least 10 years [1–5]. Nicotine, the primary constituent of both combustible cigarettes and electronic nicotine delivery systems (ENDS), is widely recognized as the addictive component responsible for long-term use of these products [6, 7]. Despite the availability of Food and Drug Administration (FDA)-approved pharmacotherapies for smoking cessation including nicotine replacement therapies and varenicline (Chantix), these interventions have limited success in maintaining long-term abstinence [8]. Given the low cessation success rates as well as the finding that smoking cessation assists in the recovery from other substances [9], there is critical need for discovery of novel pharmacotherapies that better promote long-term abstinence [10–13]. Recently, studies have revealed a potentially important role of the neuroimmune system in nicotine use [14] and withdrawal [15,16], highlighting the possibility that druggable targets in this system may prove beneficial towards this end.

While studies have found that nicotine exposures can suppress the immune system [17,18], and conversely, nicotine withdrawal can lead to activation of pro-inflammatory signals [16,19], there are critical sex differences in immune system function and nicotine use patterns that require consideration. Indeed, the immune system is sexually dimorphic and thus sex differences in immune function are governed by genetics, environment, and adaptive immunity [20–22], with females being less susceptible to infectious diseases but more prone to develop long-term autoimmune or inflammatory conditions as compared to males. Additionally, nicotine use produces sex-specific consequences, with women exhibiting distinct physiological and behavioral responses during nicotine dependence and withdrawal as compared to men [23]. Although men and women are equally likely to maintain short-term abstinence from nicotine, women are less likely than men to sustain long-term abstinence following an unaided quit attempt, and this disparity cannot be fully explained by psychosocial variables such as perceived stress or smoking history [24], supporting a need to explore biological factors that may contribute to enhanced nicotine use vulnerability in women. Moreover, women may face increased vulnerability to nicotine-related comorbidities, including a higher incidence of cancer, an increased risk of coronary heart disease, and earlier onset of menopause [25–28]. Given that a large number of preclinical studies evaluating nicotine neurobiology (e.g., [29–33]) and the few studies evaluating nicotine use and withdrawal-associated immune consequences have focused on males (e.g., [16,19], although see [15] which conducted sex comparisons and is described below), there is a need to characterize female immune profiles that may contribute to nicotine use.

We and others have found that nicotine self-administration (SA) induces aberrant glutamatergic plasticity within the nucleus accumbens (NAc) in male rats [30,34,35], and evidence indicates that glutamate signaling in the brain is heavily regulated by the neuroimmune system [36–38]. In addition to dysregulated NAc synaptic plasticity following extinction from nicotine SA, we found increased glial fibrillary acidic protein (GFAP) and tumor necrosis factor alpha (TNFα) as well as increased levels of the microglial reactivity marker CD40 following extinction and cue-induced reinstatement to nicotine seeking [19]. In line with our prior findings, others have shown that NAc microglia are reactive and cytokine levels are increased following withdrawal from nicotine exposure [16]. There are also established sex differences in nicotine-induced immune system consequences, as increased blood brain barrier (BBB) permeability in the prefrontal cortex (PFC) was found in female as compared to male mice following withdrawal from non-volitional nicotine exposure [15]. These results indicate that females may be more vulnerable to nicotine-induced suppression of the immune system as compared to males, which supports our prior report that females demonstrate greater immune suppression by nicotine and vulnerability to comorbid disease states [39].

Given our prior findings of aberrant NAc glutamate plasticity in female rats that is specific to nicotine and modulated by ovary-derived steroid hormones [40], as well as prior findings indicating that nicotine exposure suppresses pro-inflammatory cytokine production from glial cells [41], it is possible that nicotine may induce neuronal distress and impair neuron-glia communication [42,43]. Indeed, glia (including microglia and astrocytes) secrete cytokines which are peptide signaling molecules that bind to receptors located on other cells including neurons [44]. Cytokines, specifically TNFα, increase glutamate release from astrocytes [45] and microglia [46], but inhibit neuronal glutamate plasticity [47]. Chemokines are messengers of communication between neurons and glia, exerting potent chemotactic and immunoinflammatory activities [48]. Cytokines such as TNFα and chemokines such as fractalkine (FKN; aka CX3C chemokine ligand 1, or CX3CL1 [42, 43,49]; predominantly expressed on neurons in the NAc [50]) are critical messengers for immune cell-neuron communication [43]. Importantly, neurons secrete FKN in response to stress as a “help me” signal to microglia [42,50,51], as its receptor, CX3CR1, is predominantly expressed on microglia [50] and binds FKN. Thus, it is possible that nicotine SA-induced glutamate dyshomeostasis in female rats [40] is accompanied by neuronal release of FKN.

The current study evaluated cytokine and chemokine expression in both the NAc and in circulating serum to determine the consequences of volitional nicotine use on brain and circulating levels of these immune signaling molecules in females. We hypothesize that nicotine SA will suppress pro-inflammatory cytokines but will enhance FKN levels in the NAc. Given that neuronal release of FKN occurs during apoptosis and functions as a “find me” signal during this process [52], we further tested if nicotine SA increases NAc neuronal apoptotic debris.

2. Methods

Cytokine and apoptosis studies were conducted separately at different times; thus behavioral data are presented separately in Figs. 1 (for cytokines) and 4 (for apoptosis). All surgical and behavioral procedures were identical between the two studies; however, rats in the cytokine study were sacrificed via rapid decapitation, whereas rats in the apoptosis study were perfused and tissue fixed for immunostaining. Experimental timelines for the two studies are shown in Figs. 1A (cytokine study) and 4A (apoptosis study).

Fig. 1. Female Rats Acquire Nicotine SA.

(A) Experimental timeline. (B) Food pellet responding depicted as percentage of active to inactive lever presses in overnight food training experiment. Response allocation to the active lever was not significantly different between future SA groups (p > 0.05). (C₁) Nicotine and saline infusions did not change across 10 sessions (p > 0.05). (C₂) Active lever discrimination for both nicotine and saline SA rats was above 66.67% (the 2:1 active to inactive press ratio criterion, indicated here by the dotted line). Lever discrimination did not differ as a function of SA group (p > 0.05). (C₃) Nicotine consumption did not change across 10 sessions (p > 0.05). (C₄) Infusions from the saline and nicotine SA groups summed across 10 sessions were not significantly different (p > 0.05).

2.1. Subjects

In both cytokine and apoptosis studies, female rats weighed 200–225 g at the initiation of experimentation and were individually housed on a 12-h reverse light cycle with ad libitum access to food and water before experimental procedures. Animals were handled daily upon arrival and every day during experimental procedures, and all animal-use protocols were approved by the Institutional Animal Care and Use Committee (IACUC) of the University of Kentucky (UK; Protocol #2020–3438). Specific rat numbers are outlined below.

Cytokine Study:

Thirty ovary-intact female Long-Evans (Nicotine SA=14, Saline SA=16) rats were purchased from Charles River. No rats were removed from the study.

Apoptosis Study:

Eighteen ovary-intact female Long-Evans rats (Nicotine SA=9, Saline SA=9) were purchased from Charles River. Three rats lost patency and were removed from the study.

2.2. Study drugs

Nicotine tartrate (MP Biomedicals, LLC, Solon, OH, USA) was dissolved in 0.9% saline and adjusted to pH 7.4. The dose of nicotine for this study was 0.06 mg/kg/infusion and was based on free base weight [53]. This nicotine dose was selected because rats readily acquire nicotine self-administration on this dose. Ketamine (Akorn Animal Health, Lake Forest, IL), xylazine (Akorn Animal Health, Lake Forest, IL), cefazolin (Qilu Pharmaceutical, Shandong, China), meloxicam (Norbrook Inc., Overbrook, KS), and heparin (Sagent Pharmaceuticals, Schaumburg, IL) were administered as described in Section 2.3.

2.3. Surgical procedures

Rats were implanted with an indwelling jugular vein catheter (JVC). Rats were anesthetized with intramuscular (IM) ketamine (80–100 mg/kg) and xylazine (8 mg/kg), and aseptic surgical techniques were utilized. Rats received a 7-day recovery period following surgery before initiating SA procedures. Rats received cefazolin (100 mg/kg, i.v.), heparin (20 usp, i.v.), and meloxicam (1 mg/kg, i.m.) on the day of surgery and once daily for three days during the post-surgical recovery period. Heparin was administered throughout the post-surgical recovery and following the SA session to maintain catheter patency. Catheter patency was checked throughout SA experiments using Brevital (3 mg/kg i.v.), which induces and maintains anesthesia for 2–5 min. Please see our prior studies for additional surgical details [40,54].

2.4. Operant conditioning chambers

Food training and SA sessions were conducted in 28 sound-attenuated Med Associates operant rat chambers with ventilation fans. At the start of each session, one active and one inactive lever were presented. Above each lever was a stimulus light, and each chamber contained a house light. Infusion pumps were located outside each chamber. Experimental events were recorded by MED-PC software on a computer in the same experimental room outside of the sound attenuating chambers.

2.5. Food training

Food restriction was maintained for at least 6 h before the food training session was initiated and maintained throughout the experiment, including during nicotine SA. Rats underwent a 15-hour food training session, during which one active lever press (fixed-ratio 1 [FR1] schedule of reinforcement) delivered one food pellet (45 mg/pellet; Bio-Serv). Food pellets were also delivered non-contingently every 9 min. Operant chambers were illuminated with a house light during the overnight session to simulate the reverse light cycle, and water was provided ad libitum during the session. Both active and inactive lever presses were recorded, and acquisition criteria were set at a 2:1 active-to-inactive lever-press ratio during the session. If this was not achieved, additional overnight sessions were conducted until the criteria were met. The largest number of food training sessions needed to meet food training acquisition criterion was two. Twelve rats needed 2 sessions (N = 9 for the cytokine study; N = 3 for the apoptosis study).

2.6. SA procedures

Seven days following JVC surgery, rats were food-restricted to 85% of their free-feed weight (as established before the initiation of experimental procedures) and remained food-restricted throughout the experiment [40,55]. During SA sessions, upon an active lever press (FR-1), lights above both levers were illuminated and a tone (2900 Hz) was presented simultaneously with drug infusion. Infusions were followed by a 20-s dark timeout period, during which active lever responses were recorded but produced no programmed consequences. An inactive lever was always present but produced no consequences when pressed. Acquisition criteria was set to responding a minimum of 10 times on the active lever. Lever discrimination was monitored by using a 2:1 active:inactive lever press ratio (or 66.67% allocation of responding to the active lever, which is denoted in Figure C₂ and D₂ by a dotted line) or higher on the active lever for two consecutive sessions. Rats underwent a minimum 10 2-h sessions of nicotine SA during which infusions of saline or 0.06 mg/kg/infusion (0.1 mL/infusion) of nicotine were delivered for 5.9 s following one response on the active lever. Our prior work has demonstrated that ovary-intact female rats readily acquire nicotine SA with these methodological parameters [40,55,56].

2.7. Vaginal cytology

We selected the estrus phase to be the terminal estrous cycle phase following completion of the minimum number of SA sessions as outlined above. We determined estrous cycle phase via vaginal cytology, with the identification of cornified cells as in our prior studies (See Supplemental Figure 1 for vaginal cytology demonstrating estrus) [40,57,58]. If the vaginal cytology indicated that the female rat was not in estrus, extra SA sessions were completed until the first positive estrus swab was reached to reduce potential hormonal variability contributions to neuroimmune outcomes. The average number of sessions completed was 13.5 with the minimum number of completed sessions being 10 and the maximum number of sessions needed to reach estrus was 17. To confirm that rats were in the estrus phase of the estrous cycle, vaginal cytology was performed daily from the first day of SA until sacrifice. Sterile cotton swabs were dipped into sterile 0.9% saline solution then inserted into the vaginal opening for cell collection. Samples were then wet mounted and then viewed with a Olympus BX40 upright microscope for manual identification of cornified cells. Images were taken using AmScope microscope digital camera (MU1803) and software.

2.8. Cytokine analysis

Following a minimum of 10 SA sessions and confirmed estrus vaginal cytology, rats were sacrificed via live decapitation and trunk blood and whole brain tissue were immediately collected. Fresh tissue punches of the ventral striatum were taken, and the tissue was immediately flash-frozen. At least 2 ml of trunk blood was collected from each rat. In separate 1.5 ml microcentrifuge tubes, 1 ml of trunk blood was spun at 18,000 x g for 15 min at 4 °C within 1 h of collection. The serum was removed, placed in a new microcentrifuge tube and stored at − 20 °C. Serum samples and ventral striatum punches were then sent to Eve Technologies (Calgary, Alberta, Canada) for cytokine analysis. Eve Technology uses Luminex^® xMAP^® technology to detect and quantify twenty-seven rat cytokines, chemokines, and growth factors simultaneously. The multiplexing analysis was performed by Eve Technology Corporation using the Luminex^® 200^™ system (Luminex Corporation/DisSorin, Saluggia, Italy) with Bio-Plex Manager^™ software (Bio-Rad Laboratories Inc, Hercules, California, USA). Twenty-seven markers were measured in the samples using the Eve Technologies' Rat Cytokine/Chemokine 27-Plex Discovery Assay^® Array (RD27) as per the manufacturer’s instructions for use (MILLIPLEX^® Rat Cytokine/Chemokine Magnetic Bead Panel Cat. # RECYTMAG-65 K, MilliporeSigma, Burlington, Massachusetts, USA). The 27-plex consisted of: EGF, Eotaxin/CCL11, Fractalkine/CX3CL1, G-CSF/CSF-3, GM-CSF, GROα/CXCL1/KC/CINC-1, GROβ/CXCL2/MIP-2/CINC-3, IFN-γ, IL-1α, IL-1β, IL-2, IL-4, IL-5, IL-6, IL-10, IL-12(p70), IL-13, IL-17A/CTLA-8, IL-18, IP-10/CXCL10, Leptin, LIX, MCP-1/CCL2, MIP-1α/CCL3, RANTES/CCL5, TNF-α and VEGF-A with assay sensitivities of these markers range from 0.3 – 30.7 pg/mL.

2.9. Immunostaining and quantification of apoptotic debris

Following at least 10 SA sessions, rats were euthanized during estrus by perfusion with phosphate-buffered saline (PBS; pH 7.4), then fixed in 4% paraformaldehyde. The fixed brains were removed and stored overnight in 4% paraformaldehyde. The following day, the whole brains were removed from 4% paraformaldehyde and placed overnight at 4 °C in PBS. For long-term storage, whole brains were transferred to 0.1% sodium azide (Sigma-Aldrich, S2002) until immunostaining. For immunostaining, brains were sliced into 60 μm sections using a vibratome. The Anti-fractin (Millipore, AB3150) antibody, which recognizes caspase-cleaved β-actin and is a sensitive and specific marker of capase-3-dependent apoptotic neuronal debris, was used as the primary antibody [52,59,60]. For fractin staining, immunoperoxidase detection was performed using the ImmPRESS^® HRP Polymer Kit (Vector Laboratories, MP-7801) according to the manufacturer's instructions with minor modifications. In brief, brain slices were incubated in sterile distilled water for 5 min, the water was removed, and the process was repeated for a second wash. The brain slices were then incubated in BLOXALL Blocking Solution for 10 min, followed by wash buffer (PBS + 0.1% Tween 80) for 5 min, and incubated for 20 min in 2.5% horse serum. Brain slices were then incubated at 37 °C for 1 hour with the primary anti-fractin antibody diluted 1:1000 (PBS + 1% BSA), followed by a 5-minute incubation in wash buffer. Finally, brain slices were incubated for 30 min with ImmPRESS Polymer Reagent, washed twice in wash buffer, and incubated for 4 min with IMMPACT DAB reagent. Following a final wash in wash buffer, the brain slices were mounted on a cover slide with ProLongTM Gold antifade Reagent (Invitrogen P36930). Apoptosis was quantified by counting apoptotic bodies. First, the NAc was located using topography and landmarks such as the anterior commissure per the Rat Brain Atlas [61]. We averaged the number of stained cells per NAc section across two coronal sections containing the NAc per rat and quantified the results for 9 Nicotine SA and 9 Saline SA rats. We averaged the number of stained precipitate per NAc section across two coronal sections containing the NAc per rat and quantified the results for 9 Nicotine SA and 9 Saline SA rats. For each slice containing the NAc, stained precipitate was manually counted from surrounding brain regions as a proxy for whole brain apoptosis (see Fig. 4D₁ for an example slice). Brightfield images were acquired with an Olympus BX40 upright microscope and AmScope microscope digital camera (MU1803) and software.

Fig. 4. Nicotine SA Does Not Initiate Canonical Apoptosis Pathways.

(A) Experimental timeline for the apoptosis study. (B) No statistically significant differences were found in food pellet responding (depicted here as percentage of active to inactive lever presses in the overnight food training session; p > 0.05). (C₁) Nicotine and saline infusions across 10 SA sessions. There was no statistically significant difference in intake between SA groups. (C₂) Active lever discrimination for both nicotine and saline SA rats were above 66.67% (the 2:1 active to inactive press ratio criterion, indicated here by the dotted line). (C₃) Nicotine consumption across 10 sessions. (D₁) A representative schematic of brain slice demonstrating fractin immunoassay staining to quantify apoptotic bodies with highlighted regions. Scale bar = 500 pixels; white arrows=example apoptotic bodies. (D₂) No statistically significant difference was found in NAc apoptotic bodies between nicotine and saline SA (p > 0.05; Mean ± SEM). (D₃) There were no statistically significant differences in apoptotic bodies (p > 0.05; Mean ± SEM) within randomly sampled brain slices from whole brain regions.

2.10. Statistical analysis

2.10.1. Food training

Data for food training were processed for statistical analysis using the tidyverse suite of packages [62] in R Version 4.5.1 [63]. Lever-press discrimination (% active lever presses) as a function of predetermined SA drug (saline or nicotine) on the rat’s terminal food training session was analyzed using independent samples t-tests. Food training data were analyzed and reported separately for each experiment to determine whether there were baseline differences in food training before undergoing SA.

2.10.2. SA procedures

SA infusions and lever-press discrimination (% active lever presses) across the terminal 10 sessions before rats were sacrificed were analyzed using linear mixed-effects (LME) modeling in R Version 4.5.1 [63] reported separately for each experiment. LMEs were conducted using lme4 and lmertest packages [64,65]. Post-hoc testing was conducted in the emmeans package [66]. Models were specified with random effects of subject and session to account for the interdependence of repeated measurements in line with best practices for repeated-measures data [67,68], a fixed effect of drug (saline or nicotine), and the session by drug interaction. Post-hoc parameter inspection, simple slope estimation, and pairwise comparisons were conducted on the highest order effect involving each variable as appropriate and corrected with Tukey’s post-hoc correction (significance set at p < 0.05). Lever discrimination was assessed as the ratio of active/active + inactive lever presses (including all infusion-earning and timeout lever presses). Three rats that lost catheter patency during nicotine acquisition were omitted from the data analysis. Given that rats received different numbers of SA sessions dependent upon estrous cycle phase and the identification of estrus following successful completion of SA acquisition criteria, the first 10 SA sessions are presented (as all rats finished at least 10 sessions), as well as the summed total nicotine consumption for all sessions as in our prior study [55]. Rats in saline SA underwent at least 10 SA sessions (without application of acquisition criteria) until estrus phase was identified.

2.10.3. Cytokine statistical analysis

For each cytokine, the number of missing values in each condition (nicotine and saline) was evaluated, and the cytokine was retained for analysis if there were three or more non-missing values in both the saline and nicotine sample groups, ensuring that statistics could be calculated using the non-missing values only. Cytokine abundances were normalized to each raťs total nicotine consumption standardized for body weight and summed across sessions (mg/kg). A Welch two-sample t-test was calculated between nicotine and saline non-missing log2-transformed cytokine abundances. P-values were adjusted using the Benjamini-Hochberg false-discovery rate (FDR) correction. Calculations used R version 4.3.0 [63].

Associations between ventral striatum cytokine concentrations and measures of drug intake were examined using Kendall’s tau rank correlation in R version 4.3.0 [63]. Specifically, we correlated cytokine abundance with total infusions earned across all sessions (Supplementary Tables 1 & 2), total infusions earned across the terminal 10 sessions before rats reached their terminal estrus phase (Supplementary Tables 3 & 4), and infusions earned on their day of sacrifice (Supplementary Tables 5 & 6). Infusions earned rather than measures of consumption were used to include saline rats. Kendall’s tau rank correlations were computed to account for potential deviations from normality and extreme values typical of cytokine abundance data, in addition to potentially monotonic rather than strictly linear relationships between consumption and cytokines [69]. Correlations were calculated separately for each cytokine within each tissue type and treatment group (nicotine and saline), and p‑values were adjusted for multiple comparisons using the same FDR procedure described earlier.

2.10.4. Statistical analysis of apoptotic debris quantification

Apoptotic debris was manually counted in each stained section and summed for each rat. Group differences in apoptotic counts were analyzed using two-tailed unpaired t-tests for both the NAc and whole-brain assays. When tests violated the assumption of equal variances, Welch’s correction was applied.

3. Results

3.1. Food training

Cytokine Study.

There was no statistically significant effect of predetermined drug group assignment on active lever discrimination during food training [t(28)=−1.957, p = 0.060] indicating no baseline differences in lever press response acquisition before SA (Fig. 1B).

Apoptosis Study.

There was no statistically significant effect of predetermined drug group assignment on active lever discrimination during food training [t(16)=0.716, p = 0.484], indicating no baseline differences in lever press response acquisition before SA (Fig. 4B).

3.2. SA

Cytokine Study.

LMEs were conducted to examine infusions earned and discrimination (% active lever presses) as a function of drug group and session. There were no statistically significant effects or interactions for either infusions earned (Figure 1C₁; [Fs≤2.318, ps≥.139]) or lever discrimination (Figure 1C₂; [Fs≤3.903, ps≥.058]). These results indicate that intake during SA did not increase over time (Fig. 1C₃) and infusions earned were similar between nicotine and saline SA rats (Fig. 1C₄).

Apoptosis Study.

LMEs were conducted to examine infusions earned and discrimination (% active lever presses) as a function of drug group and session. There were no statistically significant effects or interactions for either infusions earned (Figure 4C₁; [Fs≤1.982, ps≥.178]) or lever discrimination (Figure 4C₂; [Fs≤1.08, ps≥.315]). These results indicate that intake during SA did not increase over time (Fig. 4C₃).

3.2. Nicotine SA suppresses ventral striatum cytokines

Of the 21 proinflammatory cytokines examined, nicotine SA produced a statistically significant reduction (adjusted p < 0.05) in cytokine levels for 8 cytokines. Specifically, nicotine reduced levels of IL-5 (adjusted p = 4.8 × 10^–5), G-CSF (adjusted p = 0.046), IL-1β (adjusted p = 0.046), Leptin (adjusted p = 1.6 × 10^–8), VEGF (adjusted p = 1.6 × 10^–8), IL-17α (adjusted p = 6.0 × 10^–7), MIP-2 (adjusted p = 4.8 × 10^–5), and TNF-α (adjusted p = 1.1 × 10^–3), compared to saline SA controls (Fig. 2A and B). Similarly, of the 6 evaluated anti-inflammatory cytokines, IL-6 (adjusted p = 0.046) and IL-13 (adjusted p = 6.0 × 10^–3) were significantly reduced during nicotine SA (Fig. 2A and B). Notably, the only elevated cytokine was FKN (adjusted p = 2.1 × 10^–5; see Fig. 2A and B). As shown in Fig. 2A, TNFα had the largest ratio of suppression with log2(nicotine/saline) = − 1.34. We found no significant associations between any striatum cytokine level and either total infusions earned across all sessions, total infusions earned across the terminal 10 sessions before rats reached their terminal estrus phase, or infusions earned on their day of sacrifice (ps≥0.100). The full table of correlations can be found in Supplementary Table 1.

Fig. 2. Nicotine SA Suppresses Pro-Inflammatory Cytokine Expression in the Ventral Striatum of Female Rats.

(A) Representative bar graph relative to 100% of saline control group demonstrating level of all cytokines measured in ventral striatum tissue from nicotine SA rats. Cytokines are grouped and color-coded according to their immunomodulatory properties (i.e., whether they serve neuroprotective (blue), neuroinflammatory (red), or neuromodulatory (yellow) roles). The cytokines that are grouped as neuromodulatory are defined as those that can modify the activity of other immune signals. (B) Ventral striatum cytokine abundance differential results: Black lines indicate the 95% confidence interval from the t-test around the log2 fold changes. Points are sized according to the adjusted p-value. One cytokine, FKN, was significantly upregulated. Cytokines that were significantly suppressed were IL-1α, IL-1β, IL-5, IL-6, IL-13, IL-17α G-CSF, Leptin, VEGF, MIP-2, and TNFα.

3.3. Nicotine SA increases circulating TNFα

In contrast to cytokine levels within the ventral striatum, blood serum cytokine levels were not suppressed. Interesting, we found an increase in TNFα by nicotine SA as compared to saline SA (adjusted p = 0.012; see Fig. 3A and B). Cytokines MIP-2, GM-CSF, and GRO/KC were removed from analysis due to below threshold levels of detectability. We found no significant associations between any striatum cytokine level and total infusions earned across all sessions, total infusions earned across the terminal 10 sessions before rats reached their terminal estrus phase, and infusions earned on their day of sacrifice (ps≥0.560). The full table of correlations can be found in Supplementary Table 2.

Fig. 3. TNFα Is Upregulated in the Serum of Female Rats Following Nicotine SA.

(A) Representative bar graph relative to 100% of saline control group demonstrating level of all cytokines measured in serum from nicotine SA rats. Cytokines are grouped and color-coded according to their immunomodulatory properties (i.e., whether they serve neuroprotective (blue), neuroinflammatory (red), or neuromodulatory (yellow) roles. The cytokines that are grouped as neuromodulatory are defined as those that can modify the activity of other immune signals). (B) Serum cytokine abundance differential results. Black lines indicate the 95% confidence interval around the log2 fold change. Points are sized according to the adjusted p-value. Only one cytokine, TNFα, was significantly upregulated following nicotine SA.

3.4. Nicotine SA does not induce NAc apoptotic debris

A Welch-corrected unpaired t-test indicated no statistically significant difference in NAc apoptotic bodies between nicotine- and saline-self-administering rats (t(8.60)=1.33, p = 0.22; Fig. 4D₂). Nicotine produced a higher mean (1.33) than saline (0.17), but this was not statistically significant. Variances differed significantly between groups (F(8, 8)=26.75, p < 0.001), supporting the use of Welch’s correction.

An unpaired t-test found no statistically significant difference in whole-brain apoptotic bodies between nicotine- and saline- self-administering rats (t(16)=1.61, p = 0.13; Fig. 4D₃). Nicotine produced a higher mean count (12.67) than saline (3.78), but this was not statistically significant. Variances did not differ significantly between groups (F(8,8)=3.57, p = 0.09), supporting the use of the standard unpaired t-test.

4. Discussion

Here we provide the first investigation of ventral striatum and circulating cytokines in females following volitional nicotine SA. These results are important because they provide a new perspective regarding how the glutamate system may be dysregulated by circulating cytokines, and how cytokine profiles in females may underlie women-specific clinical experiences and trajectories in nicotine use. In line with our hypothesis, we found that FKN (the neuronal help me [42,50,51] and find me signal during apoptosis [52]) was significantly upregulated in the ventral striatum. Importantly, this upregulation occurred against a backdrop of general suppression of pro-inflammatory cytokines by nicotine SA. Contrary to our hypothesis, the degree of cytokine and chemokine changes were not significantly correlated with nicotine consumption. Finally, we found no evidence of apoptotic debris in the NAc by nicotine, findings which align with nicotine-induced suppression of IL-1β and thus a potential inhibition of NAc inflammasome signaling [70,71].

4.1. Nicotine suppresses TNFα but increases FKN in the ventral striatum: therapeutic implications

Glial cells such as microglia and astrocytes are key regulators of synaptic glutamate [72,73]. Importantly, FKN is a soluble chemokine that functions as a chemoattractant [74] that is cleaved by TNFα converting enzyme (TACE, which is also known as disintegrin and metalloproteinase 17 or ADAM17 [75]). Given that we show suppressed TNFα, we hypothesize that there is a mechanism by which TACE is selectively activated to cleave FKN from neurons rather than TNFα from surrounding microglia. Importantly, our cytokine assay does not account for differences in membrane bound or soluble forms of cytokine protein, meaning that low TNFα abundance could also be a consequence of decreased protein formation rather than selective enzyme activity of TACE. FKN and/or TACE have become promising targets for therapeutic applications, and has been suggested that inhibition of the FKN (CXCL1)–CX3CR1 axis and specifically TACE inhibitors may prove beneficial in autoimmune disorders [76], cancers [77,78], as well as other pathologies [79]. While this may be a viable therapeutic path for nicotine use, it is important to note that both FKN and TNFα are cleaved by TACE, and thus given our findings, drug development should identify compounds that allow for inhibition of its enzymatic actions at specific sites.

In contrast to the suppression of TNFα found in the ventral striatum, here we show an increase in circulating TNFα levels. These findings indicate that nicotine use-induced suppression of TNFα is likely localized to the brain and has important implications for strategies targeting this cytokine as a smoking cessation therapeutic. It is possible that the increase in peripheral circulating TNFα could be a consequence of suppressed brain TNFα levels, although it is also possible that the differential nicotine-induced changes in this cytokine are due to tissue or cell-type [80] specificity of its secretion and function. For example, TNFα can promote cell survival but it can also trigger apoptosis and necroptosis, and these differential actions are governed by its activation of either pro-survival or pro-apoptotic intracellular signaling pathways [81,82]. Indeed, blood serum does not allow for determination of the tissue type (s) from which this cytokine originated or what functions it may serve in the peripheral nervous system. Additionally, TNFα is pleiotropic and thus may serve specific functions depending upon where is it localized and its concentration [83].

While the caveats listed above are important to consider in the strategic design of future studies evaluating TACE, TNFα, and FKN as potential therapeutic targets for nicotine use cessation, it is important to note that evidence from literature evaluating other neuropsychiatric diseases have identified TNFα as a potential key player. For example, circulating TNFα levels in patients have been implicated in unique cytokines signatures of as anxiety [84], depression [85], and attention deficit hyperactivity disorder [86]. Evidence indicates that supplemental intracerebroventricular administration of TNFα-like protein worsens performance on tasks designed to measure cognitive performance and depression-like behavior [87]. Conversely, inhibition of TNFα with Etanercept (human recombinant TNFα competitive fusion protein) promotes cognitive performance and reduces anxiety-like behavior as characterized by reductions in open field testing, elevated zero maze, and Y memory maze behavior in mice [88,89]. Together, therapeutic modulation of TNFα as a treatment strategy for nicotine use may be a viable pathway for drug development.

4.2. Nicotine suppresses activators of the inflammasome in the ventral striatum

Here, we also demonstrate that despite increased FKN signaling via protein cytokine analysis, there was not a subsequent increase in capase-3-dependent apoptosis via fractin immunostaining. Of note, IL-1β was downregulated in our study, and IL-1β is a key product of the microglial inflammasome [90]. Inflammasomes are dynamic, multi-protein complexes that are responsible for initiating neuroinflammation to respond to pathogenic insults [91]. Inflammasomes promote the pyroptosis and subsequent release of IL-1β, and it is thought that nicotine may prevent priming and activation of microglial inflammasomes and downstream nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) signaling through α7 nicotinic acetylcholine receptor activation. Taken together, we hypothesize that non-inflammatory pathways of apoptosis may be upregulated rather than inflammatory-dependent pathways via the inflammasome; however, we instead found no differences in apoptosis within the NAc between drug naive and drug exposed conditions. This offers an important insight that nicotine may be neuroprotective of synapses experiencing glutamate excitotoxicity which further promotes stronger synaptic function to strengthened synapses in the drug reward pathway. Importantly, other drugs of dependence such as ethanol, opioids, and stimulants may also interact with the inflammasome, but these other drugs have been shown to activate immune responses [92–94] rather than inhibit like nicotine, making nicotine unique in this way.

4.3. Female-specific implications of nicotine induced ventral striatal neuroimmune suppression

Our current study is the first to report neuroimmune consequences due to volitional nicotine SA in ovary-intact female rats. The current findings build upon our prior studies whereby we have shown that nicotine SA-induced NAc glutamate plasticity of GABAergic medium spiny neurons (MSNs), which is heavily implicated in driving drug seeking behavior [34] including nicotine [30], is regulated by steroid hormones derived from ovaries as well as by exogenously delivered E2 replacement [40]. Indeed, we found that ovariectomy (surgical removal of the ovaries [57]) reduces nicotine consumption and alters NAc glutamate signaling in female rats, and supplementation with E2 partially rescues responding for nicotine and restores glutamate signaling to ovary-intact controls. Here we chose to characterize cytokine and apoptosis outcomes specifically during the estrus cycle phase, and the justification for this decision was to eliminate baseline cytokine shifts induced by the reproductive system [95]. The estrus phase of the estrous cycle is associated with increased levels of E2, and E2 can inhibit binding of NF-κB to the promoters of inflammatory genes [96]. Given that TNFα activates the NF-ĸB pathway to increase inflammatory responses [97], it is possible that our findings represent an additive or synergistic immunosuppressive effect of E2 and nicotine on the ventral striatal neuroimmune system. Further, it is possible that during low E2 phases of the estrous cycle, nicotine SA may induce neuronal apoptosis given that E2 serves a neuroprotective role through inhibition of NF-ĸB [98] and thus future studies should explore this possibility to investigate how each cycle phase may differentially impact nicotine SA behavior and cytokine abundance. It is also an important consideration that androgens serve a similar neuroprotective role in the brain [99]; however, estrogen and nicotine have a bidirectional relationship in which estrogen enhances nicotine metabolism [100] and nicotine enhances estrogen metabolism and signaling [101]. Thus, it is possible that the neuroprotective effects of estrogen have an outsized impact on nicotine neurobiology compared to androgens.

Given that female-specific factors that may contribute to nicotine use, including specific hormone contributions and the role of the hypothalamus-pituitary-gonadal axis is heavily understudied, it is not clear if steroid hormones may play an outsized role in driving nicotine use by critically regulating immune signaling. Our prior study [40] did not evaluate how use of ovariectomy as a tool to evaluate specific hormone contributions to neurobehavior may have shifted neuroimmune signaling and thus did not consider this aspect of the pentapartite synapse. Given our current findings demonstrating neuroimmune consequences of nicotine SA within the ventral striatum of female rats, future studies are needed to tease apart the relationships between the shifting ovary-derived steroid hormone milieu, the NAc pentapartite synapse, and nicotine use.

4.4. Opposing immune consequences of nicotine use and withdrawal

As noted in the introduction, literature from us and others indicates that nicotine has dichotomous effects on the immune system depending upon context (i.e., whether nicotine is on board or whether an animal is in withdrawal), it is possible that nicotine influences NAc neuroimmune system function according to an opponent process [102–104] whereby suppressed neuroimmune signals become activated when nicotine is metabolized. While studies are needed to determine if immune activation plays a causal role in withdrawal severity, the possibility remains that rebounding activation of the NAc neuroimmune system could result in enhancement of cytokines, which could lead to activation of apoptotic and necroptotic cellular pathways. While our working hypothesis is that nicotine regulates the NAc neuroimmune-glutamate circuit via an opponent process depending upon use or withdrawal state (Fig. 5), studies are needed to confirm whether these opposing neuroimmune system effects are dynamic and context-dependent within the same animal, and whether more chronic nicotine exposures and repeated withdrawal cycles may lead to enhancement of immunomodulation by nicotine use and withdrawal.

Fig. 5.

Hypothesized Nicotine Use and Withdrawal-Induced Glutamate-Neuroimmune Dysregulations In the NAc Pentapartite Synapse. The pentapartite synapse is composed of (1) glutamatergic afferents that project from cortical regions, including the prefrontal cortex (PFC) to the NAc; (2) GABAergc medium spiny neurons (MSNs); (3) astrocytic perisynaptic processes; (4) microglia, and (5) the extracellular matrix. These 5 components communicate to maintain homeostasis within the brain reward pathway. Based on prior and current findings, we hypothesize that (A) chronic nicotine use induces specific cleavage of fractalkine (FKN) by TNFα converting enzyme (TACE) from neurons while concurrently limiting TACE-induced cleavage of TNFα from microglia, limiting microglia-neuronal communication and ability of glial cells to regulate glutamate homeostasis via their chemical messengers. We further hypothesize that nicotine inhibits efficient astrocyte-microglia communication by reducing the ability of microglia to secrete cytokines and activate inflammasome intracellular signaling pathways. Together, this nicotine-induced immunosuppression results in an inability of glia to functionally regulate neurons. (B) Conversely, during nicotine withdrawal (or removal of the immunosuppressive agent; in this case, nicotine) elicits a hyperactive inflammatory cascade accompanied by microglial inflammasome activation, pyrpotosis, and subsequent IL-1β and IL-18 release. Cytokine secretion from microglia then renders the astrocyte unable to regulate synaptically released glutamate through TNFα-induced suppression of GLT-1 transcription as well as a retraction of astrocytic perisynaptic processes from the synapse. Relapse to nicotine use then suppresses NAc neuroimmune signaling and increases GLT-1 transcription, leading to repeated nicotine use/withdrawal cycles.

4.5. Limitations of the current study

It is important to note that our current study has some limitations to consider. Here we only tested one dose of nicotine and provided 2 h SA sessions, which may limit generalizability of findings to other doses or conditions under which nicotine is self-administered across the entire day. This is important because individuals who smoke often do so at intervals throughout waking hours [105]. Thus, future studies should compare our current findings to long access nicotine SA conditions. It is also important to note that we did not incorporate a non-drug food control, which may be important because intravenous SA requires indwelling catheters and thus may result in a baseline immune-enhanced state. While this is possible, it is important to note that both the experimental (nicotine) and control (saline) group underwent jugular vein catheter surgery and thus this is not an experimental confound in the current study. Given the directionality of our effects whereby nicotine suppressed ventral striatum cytokine expression, this possibility strengthens our dataset because nicotine SA resulted in immunosuppression in an organism that may have been in a state of immune activation. Indeed, a common methodology to study immune mediators involves artificial triggering of immune activation, such as by administration of the bacteria lipopolysaccharides (LPS) [106,107]. Future studies are also needed to directly compare male and female immunomodulation by nicotine SA. Finally, while our study fills a gap regarding nicotine-specific changes in cytokine signatures as compared to a drug-naive condition, we did not observe escalation of nicotine use (i.e., increased consumption across sessions) as compared to saline groups under the selected study parameters (specifically, session duration, nicotine dose, cue delivery, and schedule of reinforcement). Here we show that saline self-administering rats respond at similar rates to rats self-administering nicotine. The dose utilized in this study (0.06 mg/kg/infusion) is on the descending limb of most nicotine dose effect curves and may not produce responding rates higher than saline under FR1 conditions due to a “ceiling effect” on nicotine intake [108–110]. Further, the response rate of the saline group here likely also reflects rats learning the response-outcome contingency during operant training that lever presses result in concurrent cue light and saline delivery as rats will respond for cue light delivery alone [111–113]. Importantly, we did observe a nicotine-specific neurobiological effect, which indicates nicotine disrupts neuroimmune physiology in the NAc. Future studies are needed to evaluate whether these cytokine changes play a causal role in driving nicotine-related behaviors.

5. Conclusions

We provide novel evidence that nicotine SA in female rats results in suppression of pro- and anti-inflammatory cytokines within the NAc, and that these suppressive effects are likely isolated to the brain. We further demonstrate converging evidence with our prior reports that along with dysregulated glutamate plasticity in NAc MSNs, neuronally secreted FKN is increased by nicotine SA. We further present supporting evidence that nicotine use does not induce NAc apoptosis during a high E2 phase of the estrous cycle, raising several possibilities including: (1) nicotine-induced immunosuppression in the NAc disrupts effective neuron-glia communication, and thus neurons are releasing distress signals that microglia cannot detect or respond to; (2) that E2 and nicotine act synergistically to suppress NAc neuroimmune signaling, and thus the NAc nicotine neuroimmune circuit is regulated by the ovarian hormone milieu; and (3) together with prior literature [17,20], nicotine use and withdrawal may confer differential immunomodulatory effects akin to an opponent process and this framework could be leveraged in drug development effects for nicotine use cessation therapeutics. Together, our findings support future sex-specific investigations of the neuroimmune system for the identification and evaluation of novel therapeutics in nicotine use cessation.

Supplementary Material

Supplementary material associated with this article can be found, in the online version, at doi: 10.1016/j.addicn.2026.100263.

Data availability

Data will be made available on request.