Relationship of systemic IL-10 levels with proinflammatory cytokine responsiveness and lung function in agriculture workers

A cross-sectional study of United States Veterans with agriculture exposures were recruited at the Omaha Veterans Affairs Medical Center and has been described previously [21]. During an outpatient clinic visit, the veteran was asked, “Have you worked on a farm for more than two years?” Those who answered “yes” and were between the ages of 40 and 80 years were eligible for this study. Furthermore, eligible subjects had to have no history of asthma, lung cancer, metastatic cancer to the lungs, or interstitial lung diseases such as pulmonary fibrosis, sarcoidosis or hypersensitivity pneumonitis, based upon self-report and medical chart confirmation. Those with a history of an infection or exacerbation of disease within the previous 3 weeks were excluded. Recruitment began March of 2008 and continued through December of 2013. The study was approved by the VA Institutional Review Board, and all subjects signed a written informed consent document before enrollment.

Demographic information, years worked on a farm, smoking status and respiratory symptoms were obtained by in-person and telephone questionnaires. Ever smokers were defined as having smoked more than 100 cigarettes in their lifetime. Chronic bronchitis was defined by the American Thoracic Society guidelines as having chronic cough and chronic phlegm for three consecutive months for at least 2 years [22]. All subjects underwent spirometry, with post-bronchodilator spirometry (0.083% albuterol) performed on individuals with a forced expiratory volume in 1 s (FEV₁₎ / forced vital capacity (FVC) ratio < 0.70. COPD was defined by the Global Initiative for Chronic Obstructive Lung Disease (GOLD) classification criteria (FEV₁/FVC < 0.7) [23]. The highest recorded FEV₁ and FVC were used to derive height-, weight-, age-, gender-, and ethnic-adjusted values based on National Health and Nutrition Survey (NHANESIII) reference equations. [24].

Organic dust was collected from a swine confined animal feeding operation and prepared as previously described [25]. Briefly, settled surface dust was extracted in phosphate buffered saline, centrifuged and filter sterilized (100% ODE). The dust extract was diluted to a final concentration of 1% (vol/vol) and used in the whole blood assay.

Blood was obtained by venipuncture and used for a cell differential and the whole blood assay [26]. Heparinized blood samples were processed within 2 h of collection. Blood was diluted at a 1:1 ratio with L-glutamine-RPMI 1640 medium (Thermo Fisher Scientific, Waltham, MA) then stimulated with ODE (1%) or sterile phosphate buffered saline. Dilute blood was incubated for 24 h at 37 °C with 5% CO₂, and then centrifuged at 500 x g for 5 min. Cell-free supernates were stored at − 80 °C.

For IL-6 and TNF-α measurement, a sandwich ELISA was employed [26]. In brief, purified (goat) anti-human IL-6 or (mouse) anti-human TNF-α antibody (both from R&D Systems, Minneapolis, MN) were used to coat a microtiter plate overnight. Whole blood assay supernates were incubated in the wells at room temperature followed by (rabbit) anti-human IL-6 antibody (Sigma-Aldrich, St. Louis, MO) or biotinylated (goat) anti-human TNF-α (R&D Systems). Human serum-absorbed peroxidase conjugated (goat) anti-rabbit IgG (Rockland Immunochemicals, Pottstown, PA) or streptavidin-HRP for TNF-α (R&D Systems) were used for detection. The colorimetric conversion of substrate (TMB, R&D systems) was quantified at 450 nm using the VERSAmax microplate reader (Molecular Devices, San Jose, CA). Cytokine concentrations were interpolated from an integrated 8-point standard curve created using purified recombinant human proteins. The limits of detectability for the human cytokine assays were: IL-6, 60 pg/mL and TNF-α, 15 pg/mL. Human IL-10 was measured using a commercially available kit with limits of detection at 31.3 pg/mL (human IL-10 Duoset, R&D Systems).

Study population characteristics were summarized using descriptive statistics such as mean ± standard deviation (SD) and n (%). Due to the skewed nature of IL-6, TNF-α and IL-10 levels, all values were transformed to the natural logarithm scale. Cytokine responsiveness to ODE was calculated as the difference between baseline and ODE-stimulated cytokine production. Associations between ΔLn IL-6 and ΔLn TNF-α levels and baseline Ln IL-10 were examined using Pearson’s correlation. Multivariable linear regression models were used to examine ΔLn IL-6 and ΔLn TNF-α as predictors of Ln IL-10 level, while adjusting for age (continuous variable), sex (male/female), body mass index (BMI kg/m², < 25, 25–29.9, ≥ 30), race (white/other), education (< high school, ≥ high school), smoking status (never, former, current) and white blood cell count. The association between chronic respiratory symptoms (cough, phlegm and bronchitis) and Ln IL-10 concentration was determined using Student’s t-test and multivariable logistic regression (adjustment for age, sex (male/female), body mass index (BMI kg/m², < 25, 25–29.9, ≥ 30), race (white/other), education (< high school, ≥ high school), smoking status (never, former, current) and white blood cell count (10³/μl, continuous variable). Associations between lung function and ln IL-10 (quartiles) concentration were determined by ANOVA. Due to the non-linear nature of the ANOVA results, lung function variables (FEV₁/FVC ratio; FEV₁, % predicted; FVC, % predicted) were classified into four levels based on quartiles and multivariable models were examined using multinominal logistic regression. Lung function quartile was treated as the nominal response variable and the model was adjusted for age (continuous variable), sex (male/female), body mass index (BMI kg/m², < 25, 25–29.9, ≥ 30), race (white/other), education (< high school, ≥ high school), smoking status (never, former, current) and white blood cell count (1 × 10³/μl, continuous variable). Corresponding OR and 95% CI of increased lung function (higher quartile relative to lower quartile) associated with Ln IL-10 concentration (continuous), were calculated.

Multivariable models were generated using backwards stepwise regression analysis to identify common sociodemographic and disease-related factors independently associated with Ln IL-10, respiratory symptoms and lung function. Given their significance in other studies [27], race, sex and education were factored into the model.