Increased vitamin D receptor expression from macrophages after stimulation with M. tuberculosis among persons who have recovered from extrapulmonary tuberculosis

We performed a case-control study. Cases were defined as persons with previously treated extrapulmonary TB. We defined three sets of controls: 1) persons with previously treated pulmonary TB, 2) persons with LTBI, and 3) persons who had been exposed to culture-positive pulmonary TB but were not infected (i.e., tuberculin skin test (TST) < 5 mm or negative interferon gamma release assay (IGRA)). Persons with both pulmonary and extrapulmonary TB were counted as extrapulmonary for the purposes of analysis. Inclusion criteria consisted of: age ≥ 18 years at time of diagnosis of TB disease or infection; HIV-seronegative; culture-confirmed disease and either near completion (within one month) or after completion of therapy (for extrapulmonary TB cases and pulmonary TB controls), and TST induration ≥5 mm or positive IGRA assay (for LTBI controls). We did not require persons to complete therapy for LTBI to be enrolled. Only contacts of culture-positive pulmonary TB cases were included as controls—both those with and without evidence of M. tuberculosis infection. Contacts of culture-positive pulmonary TB cases were tested for LTBI at the beginning of the contact investigation and 8–12 weeks later if the initial test was negative [16]. Exclusion criteria consisted of: serum creatinine > 2 mg/dL; use of corticosteroids or other immunosuppressive agents at the time of diagnosis or study entry; malignancy; and diabetes mellitus. HIV-positive persons were excluded because of the known increased risk of extrapulmonary TB associated with HIV/AIDS [17‐19]. Although it is unclear if local cell-mediated immune responses affect the systemic immune response, we excluded individuals with pleural TB from our study because of their exaggerated immune response at the site of disease, which may differ from other forms of extrapulmonary TB [20].

All participants were enrolled from Tennessee. Extrapulmonary TB cases and pulmonary TB controls were identified from the Tennessee Department of Health TB registry. Ongoing contact investigations at local and regional TB clinics were reviewed to identify patients in the remaining control groups. Demographic and clinical characteristics were collected from the patient or the Tennessee TB registry. The institutional review boards of Vanderbilt University Medical Center, Nashville Davidson Metro Public Health Department, and the Tennessee Department of Health approved the study. Study participants provided written informed consent.

Each subject had HIV serology, complete blood count, and vitamin D measurement performed. We measured 25-hydroxyvitamin D, the major circulating form of vitamin D. Peripheral blood mononuclear cells (PBMCs) were isolated within 24 h under sterile conditions by Ficoll-Paque (GE Healthcare Bio-Science) density centrifugation. Viability was estimated by trypan blue dye exclusion. PBMCs were divided and two aliquots of 2 × 10⁶ cells/mL were plated in a 12-well flat bottom cell culture plate in antibiotic free RPMI 1640 containing 10% heat-inactivated human serum. Gamma-irradiated H37Rv (10 μl of 1 mg/mL, γ-H37Rv) was added to one well. The plate was incubated for three days in 5% CO₂ at 37 °C.

Monocytes were isolated from PBMCs using the autoMACS system (Miltenyi, San Diego, CA). CD14⁺ cells were isolated using positive selection with CD14 microbeads (Miltenyi Biotech) and incubated in RPMI 1640 for two hours. After two hours, the cells were washed and re-plated in RPMI 1640 containing 10% heat-inactivated human serum and M-CSF (4 ng/μL; Sigma Aldrich, St. Louis, MO) and incubated for three days.

M. tuberculosis strain H37Rv (ATCC 27294) was grown to log-phase in an upright, vented tissue culture flask in 7H9 broth supplemented with OADC and 0.05% Tween-80. On day four, macrophages from each experiment were stimulated with the following: a specific TLR2 agonist, 19 kDa LpqH (EMC Micro-collections, 10 μg/mL) [21], γ-H37Rv (10 μg/mL), or live H37Rv (multiplicity of infection of 5–10). One well of macrophages was not stimulated. After a two-hour stimulation and incubation in 5% CO₂ at 37 °C, media was removed and replaced with media supplemented with 1,25α vitamin D [22]. After overnight incubation, the supernatant was removed and frozen at − 80 °C for ELISA. 500 μl of Trizol was added to each well and macrophages were lysed by pipetting. RNA was extracted using the Qiagen RNeasy Kit (Qiagen, Valencia, CA). Similar procedures were used to isolate RNA after 24 and 48 h.

After 24 and 48 h of incubation, supernatant from separate wells was filtered by centrifugation and frozen at − 80 °C. Supernatants from the 12, 24, and 48 h time points were thawed, diluted 1:10 in assay buffer, and evaluated for IL-1β using the IL-1β Human Antibody Pair kit (Life Technologies, Grand Island, NY).

On day four, PBMCs previously incubated with or without γ-H37Rv were washed in 15 mL 1X HBSS and re-suspended in 1 mL of complete RPMI 1640 and macrophages infected with H37Rv were added. Additionally, there were two wells that contained only infected macrophages. Plates were incubated for three days. On the third day, media was removed from the wells and cells were lysed with 50 μl of a 1:5 dilution of 25% sodium dodecyl sulfate and incubated 10 min at 4 °C. Immediately after incubation, 50 μl 10% bovine serum albumin was added to each tube and mixed. Ten-fold serial dilutions were prepared in 7H9 broth to 10⁶ and spread on pre-warmed 7H10 agar plates, and incubated at 37 °C for three weeks. Colonies were counted and CFU calculated at 14 and 21 days.

Total RNA was extracted from macrophages using TRIzol and subsequently purified using the Qiagen RNeasy Mini Kit. cDNA was synthesized using the iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA) according to manufacturer’s instructions. For qRT-PCR, we used the relative gene expression method (Relative Unites = 2^-ΔΔCt). Macrophage expression levels of VDR, cathelicidin (CATH), toll-like receptor 2 (TLR2), and tumor necrosis factor-alpha (TNF-α) were measured by quantitative PCR (qPCR) using gene-specific TaqMan Gene Expression Assays (Applied Biosystems, Foster City, CA) on the StepOne Real-Time PCR instrument (Applied Biosystems). GAPDH was used as a normalizer. Specific primers’ GenBank accession number and assay ID are provided in Additional file 1: Table S1).

Based on log-transformed data from a study that evaluated TLR2 mRNA expression after stimulation with M. tuberculosis H37rV in active pulmonary TB patients, [23] we estimated a sample size of 12 cases and 24 in each control group would provide 80% power at 5% two sided-significance level to detect a 1.0 log difference. Based on natural log transformed data from a study evaluating TLR2 surface expression by flow cytometry in patients with active TB, [24] we estimated a sample size of 9 cases and 18 persons in each control group would provide 80% power at 5% two sided-significance level to detect a 2.0 log difference. A sample size of 15 extrapulmonary TB cases and 30 persons in each of the 3 control groups was estimated to provide 80% power to detect a 2.5 fold difference in VDR mRNA and a 2-fold difference in cathelicidin mRNA between cases and latent infection controls.

Differences in demographic and clinical factors across exposure categories were assessed using non-parametric tests: Fisher’s exact test for differences in proportions for categorical variables, and the Kruskall-Wallis test for differences in distributions of continuous variables. Experiments were performed in duplicate. Gene expression outcome was a continuous measure (in relative value units; RVU), as was IL-1β production (in μg/mL) and accordingly, the non-parametric distribution of each outcome was described using medians and inter-quartile ranges. Unadjusted and adjusted differences in median values of gene expression and IL-1β by exposure group were determined by quantile regression, and standard errors were calculated via bootstrapping with 20 replicates [25‐27]. Because three markers were compared simultaneously in each regression model, the Holm-Bonferroni correction was applied to control the family-wise error rate across multiple tests [28, 29]. Due to a limited sample size for analysis, we could accommodate only a small number of covariates in our models. Because foreign-born status differed by patient group (Table 1) and outcome, we controlled for foreign-born status in the adjusted model of the outcome biomarkers. We did not control for sex or race because the outcome biomarkers did not vary by sex or race. All tests were two-tailed, and estimates were statistically significant if the corrected p < 0.00555. All statistical analyses were performed using Stata 12.1 (Stata Corporation, College Station, TX).