Cannabis Smoke, Aging, and Brain Inflammation: What Mouse Data Show

Abstract

Study at a Glance

Study Snapshot

Study Facts Table

What Researchers Actually Did

Gazarov and colleagues enrolled 40 C57Bl/6J mice (half female, 20 per age group) and assigned them to one of four conditions: younger-placebo, younger-cannabis, older-placebo, or older-cannabis (n=10 per group after pooling sexes). Cannabis exposure was delivered via an automated TE-10 cigarette smoking machine: mice remained in their home cages inside the exposure chamber while three cannabis cigarettes (5.5–6.2% THC) or three placebo cigarettes (≤0.1% THC) were sequentially burned and pumped into the chamber over approximately 40 minutes per session, seven days per week for 30 consecutive days. This exposure protocol was validated in prior work from the same group, with documented THC and THC metabolite levels in plasma and brain. On day 31, mice were sacrificed via rapid decapitation, and trunk blood, prefrontal cortex tissue, and hippocampal tissue were collected for inflammatory marker quantification.

Serum samples were analyzed for 40 cytokines using a custom quantitative sandwich-based antibody array (Raybiotech). Brain tissue was homogenized, protein-standardized to 1.5 mg/mL, and analyzed for 120 inflammatory markers using a separate commercial array (QAM-CAA-2000-1, Raybiotech). Statistical analysis proceeded in two stages: individual cytokine two-way ANOVAs (Age x Drug) with false discovery rate correction (FDR, q=0.05) across all markers within each tissue type, followed by principal component analysis (PCA) of z-scored cytokine profiles to capture coordinated, multi-cytokine shifts. The top five cytokine contributors to each principal component were examined with exploratory two-way ANOVAs and post-hoc Bonferroni-corrected pairwise comparisons. Multivariate PCA outlier detection led to exclusion of two serum samples (older cannabis mice) and one HPC sample (younger placebo mouse).

Key Findings: Primary Outcomes

• Serum, age effects: Older mice had significantly higher serum IL-12p40 (F(1,34)=14.4, FDR-adjusted p=0.011) and RANTES (F(1,34)=12.7, FDR-adjusted p=0.010) relative to younger mice. Exploratory analyses additionally identified age-related increases in TARC (p=0.026) and IL-1Ra (p=0.017).
• Serum, cannabis effects: No significant main effect of cannabis smoke on any individual serum cytokine. No significant Age x Drug interaction reached FDR-corrected significance in serum.
• PFC, age effects: Six cytokines were significantly elevated in PFC of older mice: Galectin-3 (F(1,36)=32.3, p=0.0001), PF4 (F(1,36)=22.4, p=0.0009), KC (F(1,36)=16.0, p=0.005), OPN (F(1,36)=13.5, p=0.01), P-Selectin (F(1,36)=13.1, p=0.01), and MIP-1γ (F(1,36)=11.6, p=0.015).
• PFC, cannabis effects (global): PCA revealed a significant main effect of Drug on PC1 (F(1,36)=4.32, p=0.044), indicating cannabis exposure increased the global PFC inflammatory profile in both younger and older mice. No individual PFC cytokine reached FDR-corrected significance for drug effects.
• HPC, age effects: P-Selectin (F(1,35)=37.3, p<0.0001), PF4 (F(1,35)=34.2, p<0.0001), TNF RI (F(1,35)=16.9, p=0.006), and bFGF (F(1,35)=13.5, p=0.017) were all higher in older mice.
• HPC, Age x Drug interaction (individual cytokines): IL-13 (F(1,35)=18.7, p=0.007) and Dkk1 (F(1,35)=17.8, p=0.007) demonstrated significant interactions. Post-hoc testing showed: older placebo mice had higher IL-13 (p=0.025) and Dkk1 (p=0.008) than younger placebo mice; older cannabis mice had lower IL-13 (p=0.020) and Dkk1 (p=0.0498) than older placebo mice; younger cannabis mice had higher IL-13 (p=0.008) and Dkk1 (p=0.005) than younger placebo mice.
• HPC, Age x Drug interaction (global profile): PCA PC1 showed a significant Age x Drug interaction (F(1,35)=8.96, p=0.005), accounting for 25.5% of total HPC cytokine variance. PC2 was significantly affected by Drug alone (F(1,35)=4.99, p=0.030).

Key Findings: Secondary Outcomes and Subgroup Analyses

• HPC PC1 top contributors: Exploratory ANOVAs on ALK1 (F(1,35)=4.65, p=0.038) and Gas6 (F(1,35)=9.56, p=0.004) revealed significant Age x Drug interactions. Post-hoc testing showed older placebo mice had higher Gas6 than younger placebo mice (p=0.004).
• HPC PC2 top contributors: Significant Age x Drug interactions were identified for IL-2Ra (F(1,35)=5.66, p=0.023), Amphiregulin (F(1,35)=5.66, p=0.023), EGF (F(1,35)=5.44, p=0.026), and IL20 (F(1,35)=5.05, p=0.031). Younger cannabis mice had higher IL-2Ra (p=0.004) and Amphiregulin (p=0.026) compared to younger placebo mice.
• PFC PC2: Significant main effect of Age (F(1,36)=7.78, p=0.008), driven by PF4, MIP-1γ, and KC, consistent with individual cytokine findings.
• Regional specificity: IL-13 did not show a significant Age x Drug interaction in PFC (adjusted p=0.989), and Dkk1 was below the detection limit in most PFC samples, underscoring that the bidirectional cannabis effects were specific to the hippocampus.
• Sex as a variable: Male and female data were pooled for all analyses due to insufficient statistical power for sex-stratified comparisons.

Results: Adverse Events and Safety Profile

Adverse events were not a formal study endpoint and are not reported. The paper does not describe any animal deaths, weight loss data, or behavioral toxicity observations linked to the smoke exposure protocol over 30 days. The use of a placebo smoke condition controlled for combustion-product inflammation, isolating effects attributable to THC-containing cannabis. No safety-relevant outcomes are quantified in this report.

Statistical Approach and Rigor

The analytic sequence was appropriate for a multi-cytokine, multi-tissue study with modest group sizes. FDR correction at q=0.05 applied independently per tissue per factor is a defensible choice, though it does not control family-wise error rate across all comparisons simultaneously. The two-stage approach — individual ANOVAs followed by PCA with exploratory post-hoc tests on top contributors — is transparent, but the exploratory PCA-derived ANOVAs are not FDR-corrected, which increases the risk of false positives among the secondary findings. Multivariate PCA-based outlier exclusion (IQR threshold) is more principled than univariate exclusion alone. Log2 transformation of protein concentrations addresses right-skewed distributions typical of cytokine data. The combination of sexes, while acknowledged as a limitation, reduces the effective n per cell to five for each sex, making any sex-stratified analysis statistically underpowered. Overall, the statistical framework is scientifically sound for a discovery-oriented preclinical study with these sample sizes, but the exploratory secondary findings require independent replication before any mechanistic conclusions are drawn.

Clinical Takeaway

This preclinical study provides initial evidence that 30 days of inhaled cannabis smoke has age-dependent, brain-region-specific effects on inflammatory cytokine profiles in mice. The most clinically relevant finding is the bidirectional response in hippocampal tissue: cannabis reduced cytokine levels in older animals while increasing them in younger ones. Specific candidates, including IL-13 and Dkk1, a Wnt pathway antagonist associated with synapse loss and cognitive decline, showed the strongest statistical support for this pattern. For clinicians advising older adults who use or are considering cannabis, these data suggest the immune effects of THC in the aging brain may differ qualitatively from those observed in younger populations. That said, this is a mouse study with small group sizes, no behavioral outcomes reported alongside the immunological data, and no human translation yet established. These findings generate hypotheses; they do not support clinical recommendations.

Why This Matters Clinically

Cannabis use among adults over 65 in the United States rose from 4.8% to 7.0% between 2021 and 2023, and 12.1% of adults aged 50 to 80 reported past-year use in 2021. Most existing cannabinoid immunology research has been conducted in young adult or disease-model animals using intraperitoneal injection or oral gavage, routes that do not replicate human cannabis consumption patterns. This study advances the field by using smoke inhalation in an aged animal cohort, pairing the route of administration and the population characteristics most relevant to the growing demographic of older cannabis users. The observation that the hippocampus, a structure central to episodic memory and disproportionately vulnerable in aging and neurodegeneration, shows a distinct and age-specific inflammatory response to cannabis smoke argues for focused mechanistic investigation. If confirmed and extended, these findings could inform whether cannabis represents a legitimate anti-inflammatory adjunct for older adults at risk for or experiencing cognitive decline, a question that current evidence does not yet answer.

CED Clinical Relevance

At CED Clinic, patients frequently ask whether cannabis use might protect against, slow, or worsen age-related cognitive changes. This paper does not answer that clinical question directly, but it provides a mechanistic thread worth following: the hippocampus of older mice exposed to cannabis smoke showed lower levels of IL-13 and Dkk1, a Wnt antagonist linked to synapse loss. For older patients already using cannabis or considering it for pain, anxiety, or sleep, the inflammatory biology explored here is directly relevant. The findings also underscore that the immune effects of cannabis are not uniform across the lifespan, which means that risk-benefit assessments formulated in young adult data may not transfer to older patients. CED clinicians should note that the cannabis used here was THC-dominant with minimal CBD, relevant context when patients present with products of varying cannabinoid composition.

Fits What We Already Know

The age-related increases in serum and brain cytokines observed here are consistent with the established literature cited by the authors. Human studies document rising circulating IL-6, TNFα, IL-10, IL-12p70, and IL-1Ra with age (Álvarez-Rodríguez et al., 2012; Cruz-Almeida et al., 2015; Milan-Mattos et al., 2019). In rodent models, Jeon et al. (2012) found elevated serum Eotaxin, IL-9, and TARC in older mice, and Porcher et al. (2021) identified upregulated IL-1α, IL-1β, and IL-17 in the hippocampus of aged mice, with more pronounced increases in females. The hippocampus-specific age effects here extend those prior reports to a broader cytokine panel. The age-dependent cognitive and neurobiological effects of THC are also supported by Bilkei-Gorzo et al. (2017), who showed chronic low-dose THC improved hippocampal-dependent cognition in older mice while impairing it in younger mice, precisely the behavioral correlate that this study’s bidirectional hippocampal cytokine data could help explain. The Dkk1 findings are further supported by Nannini et al. (2023), who reported that long-term cannabis use in middle-aged humans is associated with altered expression of WNT signaling pathway genes in whole blood, lending translational plausibility to the Dkk1 observation in mice.

Physician-Led, Whole-Person Care

A doctor who takes the time to truly understand you.

Personal care that starts with listening and is guided by experience and ingenuity.

Health, Longevity, Wellness

One-on-One Cannabis Guidance

Metabolic Balance

Leave a Message
Metabolic Care
Medical Consulting
Cannabis Care