Abstract
Mycobacteria population can undergo mutations in their DNA sequence during replication, which if not repaired, would be transferred to future generations. In this study, in vitro spontaneous mutations in Mycobacterium smegmatis mc2155 (Msm) conferring resistance to isoniazid (INHr), rifampicin (RIFr), kanamycin (KANr) and streptomycin (STRr) were determined at several concentrations in a fluctuation assay. Mutation rate was estimated using the P₀ method, and estimates were then compared with the Lea-Coulson method of the median and Ma-Sandri-Sarkar Maximum Likelihood Estimator (MSS-MLE) method available on the Fluctuation analysis calculator (FALCOR). The mutation rates of RIFr ranged from 9.24 × 10-8 - 2.18 × 10-10, INHr 1.2 × 10-7 - 1.20×10-9, STRr 2.77 × 10-8 - 5.31 × 10-8 and KANr 1.7 × 10-8 mutations per cell division. This study provides mutation rate estimates to key antitubercular drugs at a range of concentrations.
Introduction
Antibiotic resistance allows bacteria to withstand the inhibitory effects of high concentrations of antibiotics, typically beyond the minimum concentration required to inhibit the growth of the bacterium. In mycobacteria, drug resistance occurs exclusively as a result of spontaneous mutations of genes encoding drug target (1), through a natural process of DNA replication. As such, resistant mutants can then be selected for in the presence of an antibiotic (2). Mycobacterium tuberculosis (Mtb) immune to several antitubercular drugs is referred to as multi-drug resistant and accounted for 490,000 cases of MDR-TB infections in 2017 and caused 190,000 deaths globally (3). The antibiotics used in this study are shown (Table 1, Figure 1).
Kanamycin (KAN) is an aminoglycoside antibiotic (Fig. 1a), belonging to the same oligosaccharide group of water-soluble antibiotic, as streptomycin (STR). Administered as a second-line anti TB drug, especially against STR-resistant Mtb (4), KAN is bactericidal and acts, by binding irreversibly to the 16s rRNA of the 30S ribosomal subunit thereby inhibiting the synthesis of protein (6). In vitro KAN-resistant Mycobacterium smegmatis (Msm) have been shown to possess altered ribosomes (9), which appear to be conferred by mutation in the rrs gene (6). STR (Fig. 1b) is an aminocyclitol aminoglycoside that inhibits translation during protein synthesis (6). Mutations in the rpsL gene and rrs genes result to STR-resistance (2, 10). RIF (Fig. 1c) binds to the β subunit of RNA polymerase, the enzyme responsible for transcription of RNA inhibiting transcription (11). RIF-resistance is due to altered β sub-unit of the polymerase at one of the three loci in rpoB gene (7). Together with RIF, INH (Fig. 1d) forms the core of TB therapy. Catalase-peroxidase enzyme (katG) activates INH into its active form and exerts its effect by inhibiting mycolic acid synthesis, which makes up a large amount of mycobacterial cell envelope (4). Generally, resistance to INH is often linked to a range of mutations affecting one or more genes e.g. mutations in the catalase-peroxidase gene, KatG, inhA-encoded long chain enoyl-ACP reductase, which is involved with the biosynthesis of mycolic acid (12). Recently also alkyl-hydroperoxide reductase (aphC) that is involved in the cellular response of oxidative stress (13, 5) has been reported to cause resistance in INH.
In this study, we isolated Msm mutants at a variety of concentrations and determined mutation rates using the P₀ method (15), and estimates were then compared with the Lea-Coulson method of the median (LC-MM) and the Ma-Sandri-Sarkar Maximum Likelihood Estimator method (MSS-MLE), available on the fluctuation analysis calculator (16).
Results
Results for Minimum Inhibitory Concentration
Mutants are easier to detect at concentrations greater than the MIC, hence to find out what antibiotic concentration will result in the proliferation of resistant mutants, the minimum concentration, required to inhibit the visible growth of Msm was assayed. The results are shown (Table 2).
Results for fluctuation assay
Table 3 shows the results for mutation rate calculated using the P₀ method.
Results for mutation rate using Fluctuation Analysis Calculator (FALCOR)
Due to the fact that the P0 method is not applicable across all values of m i.e. the number of mutations per culture (15), the LC-MM and MSS-MLE, available on the Fluctuation analysis calculator (16) were also employed in estimating mutation rate of Msm to the test antibiotics (Table 4) and then and compared with the P0 method.
Discussion
We have isolated mutants to Msm at a range of antitubercular drug concentrations. RIF was tested at 100, 200 and 500 µg ml-1, INH at 500 and 1000 µg ml-1, STR at 20 and 100 µg ml-1 and KAN at 100 µg ml-1 using the fluctuation assay. We next estimated mutation rates using the P0 method, and estimates were then compared with LC-MM and MSS-MLE methods obtainable on the fluctuation analsysis calculator. In our study, the mutation rates estimated using the three methods were comparable. For instance at 100 µg ml-1 KAN, the mutation rate was 1.70 × 10-8 mutations per cell division with the P0 method and 1.54 10-8 with the LC-MM.
The hypothesis surrounding spontaneous mutation predicts a large fluctuation around the average for the count taken from the individual cultures. As such, a mutation occurring earlier in the growth of the culture results in a higher number of mutated cells (17). This was observed for 500 µg ml-1 INH and mutation rate in our study was significantly raised (10-7) for post log phase growth. Complex networks of factors influence the rate and type of mutants that can be selected with a given antibiotic. One of such factors that play a significant part in the mutation rate is the concentration of the antibiotic (18, 19). Thus it could be observed that, when the concentration of RIF increased from 100 to 200 and 500 µg ml-1 on agar plates, the number of mutants selected reduced and the rate of mutations ranged from 9.24 × 10-8, 4.47 × 10-10, and 2.18 ×10-10. Respectively for estimation using the P0 method, LC-MM and MSS-MLE.
In a study by (20), mutation frequencies of Msm mc26, rather than mutation rate was estimated, to 100 µg ml-1 STR and 500 µg ml-1 RIF and were >2 × 10-4 and >2.4 ×10-5 respectively which were higher than the results obtained in the current study. However, the Msm strains employed in both studies were different (mc26 versus mc2155). It was difficult to obtain STRr and KANr mutants. Causes of resistance in streptomycin have been extensively investigated in many bacteria and require a very specific base substitution in ribosomal genes for one of the ribosomal proteins. Mutations in the 16srRNA gene rrs have been found to confer mutation in streptomycin and kanamycin (6). Hence the mutation rates calculated using the Po method for this study were low i.e. 2.77 × 10-8 for streptomycin at 20 μg ml1 and 5.31 × 10-8 for streptomycin at 100 μg ml-1. These results are comparable to streptomycin mutation rate of 2.29 × 10-8 in Mtb (2). Mutation rate of Msm to 100 μg ml-1 KAN was 1.70 ×10-8 in this study. Just as it occurs in Mtb, rifampicin resistance mutation rate of 3.32 × 10-9 (2) has been found to be similar to Msm mutation rate of 10-8-10-10 as we have found in our study.
Conclusions
It was not surprising that M. smegmatis been naturally resistant to isoniazid had the lowest mutation rate by comparison to kanamycin, streptomycin and rifampicin. Furthermore, we found that since the L-C method estimates mutation rate based on the median value, it was not ideal for estimating mutation rate where greater than 50 per cent of cultures had no growth (15). This resulted in both an undefined m value and uncalculated mutation rate. All the three methods were ideal in estimating mutation rate provided the recommendations for P0 and m values are adhered to. These findings will promote a greater understanding of the phenomenon of mutation and how the estimation of mutation rate could be of importance in the control of multi-drug resistant M. tuberculosis infection.
Materials and Methods Bacteria strain
Mycobacterium smegmatis mc2155 was employed in our experiments (21).
Test drugs
Isoniazid, kanamycin, Rifampicin and Streptomycin all purchased from Sigma-Aldrich Chemical Co. Ltd. (Poole, UK) were employed in this study. Standard stock solutions of isoniazid (50 mg/mL), Kanamycin (10 mg/mL) and Streptomycin (10mg/mL) were prepared by dissolving in sterile distilled water (SDW) and filtering using a 0.22 micrometre (µm) pore size cellulose membrane, while RIF (50 mg/mL) stock solution was prepared in DMSO (Fisher Scientific Ltd. Leicestershire, UK). Working solutions were prepared by diluting in SDW.
Growth medium
Nutrient broth No. 2 (NB2; Lab lemco powder 10, Oxoid Ltd, Basingstoke, England), composed of: peptone 10g, sodium chloride 5g, beef extract 5g and reverse osmosis water, to make 1000 mL. Agar Technical No. 3 (Oxoid Ltd, Basingstoke, England) at 1.5 % was used. Sterilisation was by autoclaving at 121°C, 15 pound per square inch (psi) for 20 minutes. Antibiotics were added to media after cooling to 55°C.
Supplements
Glycerol 0.5 % volume by volume (v/v), 0.1 % (v/v) Tween 80 was used to supplement the broth. DMSO was also used in rifampicin stock solution preparation (Fisher Scientific Ltd, Leicestershire, UK).
Determination of Minimum Inhibitory Concentration (MIC)
MIC assays were determined based on log2 serial dilution of broth using NB2 containing 0.1 % Tween 80 in 5 mL tubes using the procedure by (22). This was incubated for 48 h at 37°C and tubes were then observed for visible growth.
Fluctuation assay
The distribution of mutant numbers in parallel cultures was determined using Fluctuation analysis method. A small number of cells (OD, 0.002) were grown under non-selective conditions in a 15 mL centrifuge tube. After about 36 h, a 1:1 serial dilution was done appropriately by introducing 1 mL of inoculum into 1 mL sterile NB2 medium. This was to ensure the numbers of cells in all tubes were the same. About 10 to 24 parallel independent bijou bottles containing 2 mL were used in the fluctuation assay. Thereafter, the cells were grown to saturation (after 4 to 7 days), resulting in 106 to 109 cells, and selected for mutant growth. Microfuge tubes (1.5 mL) were then used to pellet the culture (10 000 g, 5 minutes) and made into 300 L volume. 200 µL volume was plated on antibiotic selective plates at the indicated concentrations. In order to estimate the number of viable cells, the remaining 100 µL was serially diluted and plated on non-selective NB2 agar plates. The average cell number was then calculated. This was the final number of cells, Nt that was used for mutation rate calculation.
Isolation of antibiotic-resistant mutants
Msm mc2155 cultures were spread out on NBA plates at the desired antibiotic concentrations and incubated for 4-6 days at 37 °C. Mutant colonies were confirmed by re-streaking on antibiotic selective plates containing the antibiotics at their respective concentrations.
Data analysis
The numbers of mutant cells were analysed using P0 method and fluctuation analysis calculator (FALCOR) available online.
Funding information
The authors acknowledge TETFUND Nigeria for funding Nyinoh, I. W’s doctoral studies.
Conflicts of interest
None declared.