Antibody responses to Mycobacterial glycans and glycolipids as potential correlates of protection from mycobacterium tuberculosis infection in BCG vaccinated children

Abstract

Background: Globally, tuberculosis (TB) has remained the main cause of death from a single infectious pathogen. The only licensed TB vaccine to date, BCG, has variable efficacy. We need to understand the mechanism underlying the protection of BCG mediated immunity against TB to develop more efficacious vaccines. This study aimed to evaluate antibody responses to mycobacterial glycoconjugate antigens in BCG vaccinated children as potential correlates of protection from M.tb infection. Methodology: A retrospective case-control study design was used to determine the relationship between antibody responses to mycobacterial glycoconjugate antigens and M.tb infection in BCG vaccinated children. Cases were defined by positive T-SPOT.TB test while controls tested negative on this test. The cases and controls were matched based on cofounders; age, sex, BCG scar and area of residence. Enzyme-linked Immunosorbent Assays (ELISA) were used to measure antibody responses in archived children’s samples. Initial investigations were conducted to determine optimal arabinogalactan (AG), lipoarabinomannan (LAM) and Ag85 complex antigen concentration and plasma dilutions for IgG antibody ELISA assays. Anti-AG, anti-LAM and anti Ag85 complex IgG antibody responses were then measured in 88 cases and 176 matched controls from birth to 5 years of age. Results: Optimisation assays indicated that the optimal antigen concentration was 2.5µg/ml for Anti-AG, anti-LAM and anti-Ag85 complex IgG antibody ELISA assays. On the other hand, 1 in 25 was the optimal sample dilution for detection of anti-Ag85 Complex and anti-AG antibodies whereas 1 in 50 was optimal for detection of anti-LAM antibodies. Following ELISA testing of the samples it was found that anti-LAM antibodies were strongly associated with an increased likelihood of M.tb infection at the 1-year time point (adjusted odds ratio=6.302, 95% confidence interval (CI): 1.970, 20.165 p=0.002) whereas anti-Ag85 complex antibodies were weakly associated with a reduced likelihood of M.tb infection at only the 4-year (adjusted odds ratio=0.35, 95% confidence interval (CI): 0.101, 1.206, p=0.096) and 5-year time points (adjusted odds ratio=0.285, 95% confidence interval (CI): 0.069, 1.185, p=0.084). However, we found no evidence of association between anti-AG antibodies and M.tb infection at all the time points investigated. Conclusion: Our findings suggest that anti-LAM antibodies at the 1st year of childhood are associated with increased risk of M.tb infection rather than protection whereas anti-Ag85 complex antibodies at the 4th and 5th years of childhood have are weakly associated with protection. On the other hand, anti-AG antibodies are not associated to M.tb infection in children. These associations should be further investigated using in vitro assays to determine if these antibodies enhance or inhibit M.tb cellular infection.