Study setting and patients
Between June and October 2017, a cross-sectional study was conducted. Confirmed new pulmonary TB cases were recruited from Tuberculosis Clinic at Soewandhie Hospital, Surabaya, Indonesia. Bacteriological confirmation was conducted by sputum acid fast staining and RIF Xpert gene (Cepheid, Sunnyvale, CA, USA). Patients underwent fiber optic bronchoscopy to collect bronchoalveolar lavage fluid (BALF). Macrophages were collected from BALF. Patients with HIV positive, diabetes mellitus, renal abnormality, heart diseases, immune response disorders such as lupus erythematosus and rheumatoid arthritis, non-TB pulmonary diseases, and those who previously received anti-TB treatment were excluded. All samples were tested to identify MTBC strain using polymerase change reaction (PCR) targeting two specific genes: RD9 and TbD1.
Assessment of pulmonary damage
The degree of pulmonary damage was classified using the NICE Scoring System based on the total lesions in six lung areas [19]. This scoring system assessed four components: nodule (N), infiltration or consolidation (I), cavity (C) and ectasis (E) based on chest radiograph of three areas of each lung (i.e. six areas of both lungs). For each area, the score was 1 to 4 indicating lung damage area of 0–25%, >25%–≤50%, >50%–≤75% and >75%, respectively. The pulmonary damage was categorized as mild if the total score was 8 or less and severe if the total score was more than 8.
Samples collection and macrophages isolation
BAL was performed using 10 ml of saline solution as described previously [20]. The BALF was centrifuged at 2500rpm for 15 mins then supernatant was discarded and cells were resuspended to a cell count of 4x105 cells/ml with RPMI 1640 medium. The total cell count was measured using hemocytometer.
FADD and RIP3 expression by immunocytochemical staining
Pellet cells derived from the centrifugation were applied to glass slides and washed with PBS three times for 10 mins. Permeabilization was performed with a CA-630-0.5% Igepal solution (Sigma Aldrich, Saint Louis, MO, USA). H2O2 0.3% was then added and incubated for 10 mins before was washed with PBS. The slides were incubated with anti-human monoclonal antibody FADD or RIP3 followed manufacturer’s protocol (Santa Cruz, Oregon, OR, USA). The quantification of the protein expression was conducted according to previous study [21].
Apoptosis assay
The level of apoptosis in infected macrophages was determined by using the Tunel Assay apoptosis kit per manufacturer’s protocol (R&D Systems, Minneapolis, MN, USA). Tunel assay was performed with terminal deoxynucleotidyl transferase enzymes to determine the fragmentation of DNA. The level of apoptosis was measured based on previous study [22].
Strain identification and confirmation sequencing
The detection of MTBC strain was conducted from BALF. Briefly, DNA was extracted using DNeasy® Blood & Tissue kit (Ambion Inc., Austin, TX, USA). Amplification of gene-specific M. tuberculosis strain was conducted using RD9 primers (F: 5’-GTGTAGGTCAGCCCCATCC-3’, I: 5-CAATGTTTGTTGCGCTGC-3’, R: 5’-GCTACCCTCGACCAAGTGTT-3’), while M. bovis strain was identified using TbD1 primers (F: 5’-AGTGACTGGCCTGGTCAAAC-3’, R: 5’-GAGCTCTGTGCGACGTTATG-3’) [23,24]. The conditions for PCR assays were set up for 30s at 94˚C (denaturation), followed by 35 cycles of denaturation (94˚C, 30s), annealing (56˚C, 1s), and extension (72˚C, 10 min). Confirmation of the strain was conducted by sequencing nine and two of M. tuberculosis and M. bovis samples, respectively and homology analysis was conducted using Basic Local Alignment Search Tool (BLAST).
Statistical Analysis
Association between MTBC strain and degree of lung damage including each subset of NICE component were assessed using chi-squared test. To compare the level of apoptosis, FADD, and RIP3 of macrophages between M. tuberculosis and M. bovis strain groups, Man-Whitney test was employed. For all analyses, significance was assessed at α=0.05.