Persistence of Burkholderia thailandensis E264 in lung tissue after a single binge alcohol episode

Abstract

Background: Binge drinking, an increasingly common form of alcohol use disorder, is associated with substantial morbidity and mortality; yet, its effects on the immune system's ability to defend against infectious agents are poorly understood. Burkholderia pseudomallei, the causative agent of melioidosis can occur in healthy humans, yet binge alcohol intoxication is increasingly being recognized as a major risk factor. Although our previous studies demonstrated that binge alcohol exposure increased B. pseudomallei near-neighbor virulence in vivo and increased paracellular diffusion and intracellular invasion, no experimental studies have examined the extent to which bacterial and alcohol dosage play a role in disease progression. In addition, the temporal effects of a single binge alcohol dose prior to infection has not been examined in vivo.

Principal findings: In this study, we used B. thailandensis E264 a close genetic relative of B. pseudomallei, as useful BSL-2 model system. Eight-week-old female C57BL/6 mice were utilized in three distinct animal models to address the effects of 1) bacterial dosage, 2) alcohol dosage, and 3) the temporal effects, of a single binge alcohol episode. Alcohol was administered comparable to human binge drinking (≤ 4.4 g/kg) or PBS intraperitoneally before a non-lethal intranasal infection. Bacterial colonization of lung and spleen was increased in mice administered alcohol even after bacterial dose was decreased 10-fold. Lung and not spleen tissue were colonized even after alcohol dosage was decreased 20 times below the U.S legal limit. Temporally, a single binge alcohol episode affected lung bacterial colonization for more than 24 h after alcohol was no longer detected in the blood. Pulmonary and splenic cytokine expression (TNF-α, GM-CSF) remained suppressed, while IL-12/p40 increased in mice administered alcohol 6 or 24 h prior to infection. Increased lung and not intestinal bacterial invasion was observed in human and murine non-phagocytic epithelial cells exposed to 0.2% v/v alcohol in vitro.

Conclusions: Our results indicate that the effects of a single binge alcohol episode are tissue specific. A single binge alcohol intoxication event increases bacterial colonization in mouse lung tissue even after very low BACs and decreases the dose required to colonize the lungs with less virulent B. thailandensis. Additionally, the temporal effects of binge alcohol alters lung and spleen cytokine expression for at least 24 h after alcohol is detected in the blood. Delayed recovery in lung and not spleen tissue may provide a means for B. pseudomallei and near-neighbors to successfully colonize lung tissue through increased intracellular invasion of non-phagocytic cells in patients with hazardous alcohol intake.

Fig 1. Bacterial load in lung and spleen of binge-drinking mice intranasally infected with decreasing doses of B. thailandensis.

Colony forming units (CFUs) per lung (A). Colony forming units (CFUs) per spleen (B). C57BL/6 mice were administered alcohol (4.4 g/kg) or PBS intraperitoneally (i.p.) and 0.5 h later mice were inoculated intranasally with B. thailandensis at doses of (3 x 10⁵, 5 x 10⁴, 8 x 10³, or 500). Tissues were collected 24 h post infection and bacterial tissue burden was determined (CFU/Tissue). Each bar represents the mean of each group inoculated with a respective dose with SD, N = 6 per group. Average fold increase comparison of lung and spleen tissue (C). Fold increase is based on initial bacterial dose and mean final bacterial burden with SD. ND = Not Detected; no bacteria was cultured on any LB media plate. (Control) indicates infected mice (3 x 10⁵ CFU) not administered alcohol. Horizontal lines and asterisks (*) represent statistical comparison of (10⁵ dose) and subsequent lower doses by Student’s t-test with Welch’s correction. ****, p ≤ 0.0001.

Fig 2. Bacterial load in lung and spleen of mice intranasally infected with B. thailandensis and administered different alcohol doses.

Colony forming units (CFUs) per lung (A). Colony forming units (CFUs) per spleen (B). C57BL/6 mice were administered alcohol at doses of (4.4, 3, 2, 1 g/kg) or PBS intraperitoneally (i.p.) and 0.5 h later mice were inoculated intranasally with B. thailandensis. Dashed line represents the inoculating bacterial dose of (3 x 10⁵) CFUs. Tissues were collected 24 h post infection and bacterial tissue burden was determined (CFU/Tissue). Each bar represents the mean of each group administered a respective alcohol dose with SD, N = 6 per group. Measured BAC (%) indicated with () and line represents average BAC measured from blood collected 0.5 h prior to infection from each group (0.254, 0.152, 0.0265, 0.00397%) or PBS respectively. ND = Not Detected; no bacteria was cultured on any LB media plate. Horizontal lines and asterisks (*) represent statistical comparison of (4.4 g/kg dose) and subsequent lower alcohol doses by Unpaired Student’s t-test. **, p ≤ 0.01, ****, p ≤ 0.0001.

Fig 3. Bacterial load in lung and spleen of mice intranasally infected with B. thailandensis and temporal effects of alcohol during infection.

Colony forming units (CFUs) per lung (A). Colony forming units (CFUs) per spleen (B). C57BL/6 mice were administered alcohol (4.4 g/kg) or PBS intraperitoneally (i.p.) and (0.5, 3, 6, or 24 h) later mice were inoculated intranasally with B. thailandensis. Dashed line represents the inoculating bacterial dose of (5 x 10⁵) CFUs. Tissues were collected 24 h post infection and bacterial tissue burden was determined (CFU/Tissue). Each bar represents the mean of each group administered alcohol at a respective time prior to infection with SD, N = 6 per group. Measured BAC (%) indicated with () and line represents average BAC measured from blood collected 0.5, 3, 6, or 24 h prior to infection from each group (0.254, 0.156, 0.0312, 0.0%) or PBS respectively. ND = Not Detected; no bacteria was cultured on any LB media plate. (Control) indicates infected and BAC 0.0%. Horizontal lines and asterisks (*) represent statistical comparison of (0.5 h alcohol prior to infection) and subsequent alcohol exposure prior to infection by Unpaired Student’s t-test, *, p ≤ 0.05, **, p ≤ 0.01.

Fig 4. Cytokines in lung of mice intranasally infected with B. thailandensis and temporal effects of alcohol during infection.

GM-CSF (A). TNF-α (B). IL-12/p40 (C). IL-10 per lung (D). C57BL/6 mice were administered alcohol (4.4 g/kg) or PBS intraperitoneally (i.p.) and (0.5, 3, 6, or 24 h) later mice were inoculated intranasally with B. thailandensis (5 x 10⁵) CFUs. Tissues were collected 24 h post infection and cytokine concentration determined by ELISA with corresponding protein standard. Each bar represents the mean of each group administered alcohol at a respective time prior to infection with SD, N = 6 per group. (Control) indicates infected mice not administered alcohol. Horizontal lines and asterisks (*) represent statistical comparison of (0.5 h alcohol prior to infection) and subsequent alcohol exposure prior to infection by one-way ANOVA **, p ≤ 0.01, ****, p ≤ .0001.

Fig 5. Cytokines in spleen of mice intranasally infected with B. thailandensis and temporal effects of alcohol during infection.

GM-CSF (A). TNF-α (B). IL-12/p40 (C). IL-10 (D). C57BL/6 mice were administered alcohol (4.4 g/kg) or PBS intraperitoneally (i.p.) and (0.5, 3, 6, or 24 h) later mice were inoculated intranasally with B. thailandensis (5 x 10⁵) CFUs. Tissues were collected 24 h post infection and cytokine concentration determined by ELISA with corresponding protein standard. Each bar represents the mean of each group administered alcohol at a respective time prior to infection with SD, N = 6 per group. (Control) indicates infected mice not administered alcohol. Horizontal lines and asterisks (*) represent statistical comparison of (0.5 h alcohol prior to infection) and subsequent alcohol exposure prior to infection by one-way ANOVA, *, p ≤ 0.05, ***, p ≤ 0.001, ****, p ≤ .0001.

Fig 6. B. thailandensis invasion and survival in non-phagocytic lung and intestinal cells with and without alcohol treatment.

Murine (Eph4, LET-1), intestinal (Mode K), or human (A549) lung epithelial cells were grown to confluency in cell culture media and co-cultured with B. thailandensis (MOI 1:10) with 0.0% or 0.2% v/v alcohol. ALC (Prior to infect) indicates cells treated with alcohol for 3 h prior to infection. ALC (During infect) indicates cells treated with alcohol at the time of infection. All cells were co-cultured for 3 h post alcohol treatment. Extracellular bacteria were removed by washes X4 and antibiotic treatment for 2 h. Cells were lysed, and viable bacteria recovered. Asterisks (*) represent statistical comparisons between ALC (Prior to infect) or (During infect) treatment and (Control, Non-Alcohol) determined by two-way ANOVA and Dunnett’s multiple comparisons test. Bars represent average CFU with SD. ***, p ≤ 0.001, ****, p ≤ 0.0001. Group comparison to Mode K cells determined by two-way ANOVA and Tukey’s multiple comparison test, **** p ≤ 0.0001, N = 3.