(H-290) A rapid, high-throughput, melt-based assay for rapid optimization of oligonucleotide ligation assay to detect multi-drug resistant tuberculosis

Introduction:: Tuberculosis (TB) is the second leading cause of death by infectious diseases after COVID-19 and above HIV/AIDS [1]. In 2021, approximately 450,000 cases of TB infections showed resistance to rifampicin (RIF) or was multidrug-resistant (MDR), which is resistant to both RIF and isoniazid (INH), the two most common antimicrobial drugs for TB [1]. Thus, determining whether an individual is infected by MDR TB is essential to improve individual treatment outcomes and reduce further transmission of MDR TB. Single nucleotide polymorphisms (SNPs) have been reported to be associated with MDR TB [2]. Oligonucleotide ligation assay (OLA) can accurately detect known single-nucleotide polymorphisms (SNPs) and previously was applied to detect SNPs associated with other diseases such as HIV [3]. OLA utilizes templated ligation of DNA oligonucleotide probes, and the ligated probes are detected using an enzyme-linked immunosorbent assay (ELISA). Here, we present an improvement upon OLA. Instead of using ELISA, we have developed a simple melt-based assay that can screen and optimize ligation probes in a high-throughput fashion like ELISA but five times faster and ten times cheaper.




Materials and Methods:: DNA sequences from the rpoB, katG, and inhA genes of mycobacterium Tuberculosis were obtained from the European Nucleotide Archive. The sequences were aligned using Geneious, and three OLA probes (i.e., mutant, wild-type, and common probes) were designed for each of 10 mutations conferring resistance to RIF (rpoB gene: D435V, D435Y, H4445Y/D, H445L, S450L/W, S450F, L452P) and INH (katG gene: S315T; inhA gene: c-777t and g-154a). The ligation reaction contains ampligase thermostable ligase, NAD, DTT, salts, EvaGreen, synthetic single-stranded TB DNA targets, and a combination of ligation probes (a 5’-phosphorylated common probe and either a mutant or wild-type probe containing a single nucleotide polymorphism at its 3’ end) respective to each TB genotype. Four types of reactions (n=4) were tested to evaluate the sensitivity and specificity of the ligation reactions: MUT ligation reactions with WT or MUT templates and WT ligation reactions with WT or MUT templates. The ligation reaction was conducted using a commercial qPCR machine (Biorad CFX96). The reactions underwent an incubation at 37°C for 15 minutes for ligation and followed by denaturation at 94°C for 5 minutes, slow cooling to 25°C for 30 seconds at a rate of 1°C/s, slow heating to 50°C for 30 seconds at a rate of 1°C/s and melt curve analysis from 50°C to 95°C at a rate of 0.5°C/s with 30 seconds intervals. Ligation yields were estimated using two approaches: (1) the area under the curve (-dRFU/dT, trapezoidal approximation) and (2) the slope of -dRFU/dT near melting temperature.


Results, Conclusions, and Discussions:: We observed melt peaks of the ligated probes in all of the reactions with matched templates (i.e., WT probes with WT DNA templates and MUT probes with MUT DNA template). We did not observe melt peaks of the ligated probes in all the reactions with mismatched probe/template (i.e., WT probes with MUT templates and MUT probes with WT templates). These results indicated the ability of our assay to specifically detect MDR TB SNPs. We investigated two analysis approaches to obtain the quantitative ligation yields representing the signal from the melt peaks of ligated probes. Areas under the curves of the ligated probe melt peaks were distinguishable in all codons except for the case of S450L/W, in which we observed an unusually high baseline in WT probe/MUT template reactions. On the other hand, the baselines of melt curves from the mismatched probe/template in all cases were flatter than those from the matched probe/template. Thus, ligation yields were better differentiated using the slope. Fig.1 shows successful differentiation between the reactions with matched probe/template versus those with mismatched probe/template.


In summary, we demonstrated a newly developed assay protocol and analysis that can be used to accelerate the screening of the OLA probes and OLA assay optimization. Our new protocol requires 2 hours to complete compared to 10 hours of the original OLA protocol. Additionally, this melt analysis approach does not require labeled probes as in ELISA. It thus reduces the material costs during the research and development phase by ~10 folds (~$US 50 versus ~$US 500 per SNP by the original OLA protocol).




Acknowledgements (Optional): : We are grateful for the advice and stimulating discussion from our collaborators: Dr. Lisa Frenkel, Dr. Grace John-Stewart, and Dr. Bhavna Chohan at the University of Washington; Dr. Diana Marangu at the University of Nairobi; and Preston Omondi and Dr. Steve Wandiga at Kenya Medical Research Institute; and Ronald Odero at AMREF Africa. This work was supported by the SEATRAC New Investigator Award (Panpradist). The funder has no role in study design and result interpretation.

References (Optional): : [1] World Health Organization. (2023, April 21) Tuberculosis Fact Sheet. https://www.who.int/teams/global-tuberculosis-programme/tb-reports

[2] World Health Organization. (2021, June 25) Catalogue of mutations in Mycobacterium tuberculosis complex and their association with drug resistance. World Health Organization. https://www.who.int/publications-detail-redirect/9789240028173

[3] Beck IA, Mahalanabis M, Pepper G, Wright A, Hamilton S, Langston E, Frenkel LM. Rapid and sensitive oligonucleotide ligation assay for detection of mutations in human immunodeficiency virus type 1 associated with high-level resistance to protease inhibitors. J Clin Microbiol. 2002 Apr;40(4):1413-9. doi: 10.1128/JCM.40.4.1413-1419.2002. PMID: 11923366; PMCID: PMC140364.