The gain of hydrogen peroxide resistance benefits growth fitness in mycobacteria under stress

Graphical abstract

Dear Editor,

Tuberculosis (TB) is a formidable challenge to global health and is caused by an ancient pathogen, Mycobacterium tuberculosis (WHO, 2010). In the pre-antibiotic era, TB killed 20% of adults in Europe and North America in the period from the seventeenth to the nineteenth centuries (Wilson, 2005). Since antibiotics have been included in TB treatment regimens, millions of human lives have been saved by this chemotherapy. However, the standard anti-tuberculosis therapy is only effective for drug-susceptible TB but not for multi-drug resistant and extensively drug-resistant TB. Unfortunately, our knowledge of drug resistance in mycobacteria remains unclear and thus hampers the development of new anti-tubercular drugs. The basic mechanisms of drug resistance in mycobacteria are complicated and the outcome of drug resistance stems from evolution in the final analysis. Understanding the evolutionary strategies used by M. tuberculosis might give new insights into designing control strategies against M. tuberculosis (Little et al,. 2012).

As the most successful bacterial pathogen, M. tuberculosis has evolved elegant strategies to evade a host’s immune defensive system. For example, D’Arcy Hart and colleagues observed that M. tuberculosis-containing phagosomes do not fuse with the lysosome inside murine macrophages (Armstrong and Hart, 1971). Several studies indicated that M. tuberculosis succeeds as an intracellular pathogen is partially due to its ability to survive and persist inside macrophages, which provide hostile environments with starvation and other stresses (Podinovskaia et al., 2013). When M. tuberculosis infects macrophages, the mycobacteria must overcome exogenous reactive oxygen species (ROS), one of the classical innate defense mechanisms against infection. Additionally, during latency, M. tuberculosis continues to be under oxidative stress; thus, the accumulation of mutations caused by oxidative DNA damage was predicted to be a potential risk for drug resistance (Ford et al., 2011). Recently, studies have shown a correlation of ROS damage of DNA with antibiotic resistance (Kohanski et al., 2010), although this notion has been disputed by other recent studies that argued that antibiotic exposure did not induce the formation of ROS (Keren et al., 2013). The role of ROS in the evolution of drug resistance is still an open question.

Currently, the view of evolution is mostly derived from genomic sequencing and lacks direct experimental evidence. In this study, to explore the correlation between hydrogen peroxide (H₂O₂) resistance and growth benefits in mycobacteria under stress, we developed a model for evolutionary screening of H₂O₂-resistant mutations in a pathogenic mycobacteria model, M. smegmatis, using a clinically important stressor, H₂O₂ (Fig. 1). This strategy of screening for H₂O₂-resistant mutations under increased H₂O₂ conditions in M. smegmatis was chosen to allow genome-wide identification of essential genes involved in oxidative stress because setting up a library of transposon mutants would fail, by default, to identify candidates that are essential genes. To demonstrate that resistance phenotypes resulted from chromosomal mutations, we continuously sub-cultured resistant strains for 10 generations without H₂O₂ treatment. The actual minimum inhibitory concentrations (MICs) of H₂O₂ in different mutants selected under different H₂O₂ concentrations were different, indicating that gradually increasing H₂O₂ resistance was due to the accumulation of mutated genes rather than a key mutated gene that caused the resistance to H₂O₂ (Fig. S1). After 54 days of selection, the MIC of the mutant strains with the maximum resistance to H₂O₂, selected at an H₂O₂ concentration of 1.5 mmol/L, was 3.125 mmol/L, while that of the wild type strain mc²155 under the same conditions (7H9 medium with 10% ADS) was 0.039 mmol/L. We named the mutant strain mc²51. As evolutionary cross-stress protection was observed under various stresses, we compared the growth advantage between mc²51 and mc²155 in corresponding stresses. We first measured the doubling time of both wild type mc²155 and mutant strain mc²51. A significant difference in doubling time was detected: the doubling time of wild type mc²155 was 2.734 ± 0.051 (n = 30), while the doubling time of mc²51 was 3.448 ± 0.087 (n = 30) (Fig. 2B), and the generation time increased up to 126%, compared to wild type mc²155. A growth advantage was observed for mutant strain mc²51 in M9 minimal medium compared to wild type mc²155, but this did not occur in M9 minimal medium with 10% ADS, suggesting that the mutant strain has a growth advantage under starvation conditions. Additionally, to investigate whether mc²51 has a survival advantage in host cells, cell line THP1 was infected with either mutant strain mc²51 or wild type strain mc²155 at an MOI of 1:10. The mutant strain mc²51 had a statistically significant growth advantage in macrophages compared to wild-type mc²155 (Fig. 2D). As a correlation of ROS damage and antibiotic resistance was indicated, we also detected the MICs of various antibiotics in both mc²155 and mc²51. Previous studies have suggested that H₂O₂ affected KatG mutation frequency and caused changes in resistance to isoniazid (INH) (Timmins and Deretic, 2006). A 100-fold difference in the MIC of INH was detected between mc²155 and mc²51 (Table S1). To provide insight into the mechanisms involved, we sequenced the whole genomes of both mutant strain mc²51 and wild type mc²155. Compared with mc²155, there are 22 single nucleotide polymorphisms (SNPs) in the mc²51 genome, of which 19 are located in coding regions, and 12 of these 19 SNPs are nonsynonymous mutations (Table S2). The data showed that G:C > T:A transversion and G:C > A:T transition were the major mutation types (Fig. 2E). The ratio of G:C > T:A transversion to G:C > A:T transition is an index to measure the contribution of oxidative DNA damage to the development of antibiotic resistance; it was 0.7 in mc²51 but is 0.25 under normal growth conditions. This increased ratio indicated that oxidative DNA damage caused the DNA mutations of mc²51. It should be noted that the ratio for mutations of M. tuberculosis isolated from macaques with latent infection increased drastically to 0.67 (Ford et al., 2011), indicating that oxidative DNA damage is a dominant source of mutation during both active and latent phases of M. tuberculosis. To further determine the possible roles of those genes in H₂O₂ resistance, the sequences of the corresponding genes amplified from wild type mc²155 genomic DNA were cloned into the integrating vector pMV361 (Table S4) and introduced into mc²51 via electroporation. Surprisingly, there are no changes in the MIC of H₂O₂ for most detected transformed strains, except for one strain of mc²51 with pMV361-fur (msmeg_3460) (Table S3 and data not shown). Furthermore, because fur is located upstream of msmeg_3461, which encodes KatG, the mutated Fur (Ferric uptake regulation protein) with substituted A28 V would influence KatG expression. Increased peroxidase and catalase activities were found in mc²51 (Fig. 2F), and KatG was confirmed by mass spectrometry analysis (Fig. 2F).

As an intracellular pathogen, the success of M. tuberculosis is partially due to its ability to survive and persist in macrophages. For example, ROS produced by the host is one of the most important classical innate defense mechanisms against infection. To defend against ROS stress, mycobacteria have evolved many detoxification strategies to scavenge H₂O₂ (Kumar et al., 2011). Several studies have suggested that accumulation of mutations might link to the emerging drug resistance. In this study, we first selected an H₂O₂-resistant strain, mc²51, which displayed an 80-fold higher MIC of H₂O₂ than that in wild type mc²155. The mutant strain mc²51 has a growth advantage under starvation conditions and within host macrophage cells, but no resistance or susceptibility to the many investigated antibiotics except INH. To characterize the correlation of SNPs in mc²51 with the adaptive evolution of M. smegmatis resistance to H₂O₂, we sequenced the genome of mc²51 and showed that the SNP in fur contributed the observed resistance to H₂O₂ (Table S3). Furthermore, the mass spectrometry results indicated that the mutation of fur changed the expression of katG, causing high resistance to H₂O₂. Our results showed that the development of H₂O₂ resistance in mycobacteria led to the gain of growth benefits in starvation conditions and within macrophages.

The ratio of G:C > T:A transversion to G:C > A:T transition mutation is used as an index to measure the contribution of oxidative DNA damage to the development of antibiotic resistance (Hershberg et al., 2008). If the oxidative DNA damage contributes to DNA mutation, this ratio will increase significantly. In this study, we calculated the ratio as 0.78, which indicated that oxidative DNA damage is a dominant source of mutation for mycobacteria (Fig. 2E). A similar ratio (0.67) was also reported in mutations of M. tuberculosis isolated from macaques with latent infection (Ford et al., 2011). Additionally, previous studies have shown that the ratio of G:C > T:A transversion to G:C > A:T transition that confers resistance to anti-TB drugs is 0.71 (Wang et al., 2013), which is similar to our results (0.78). However, in our study, the MICs of various antibiotics did not differ between wild type mc²155 and mc²51 except that of INH, a prodrug that needs bacterial ROS activation. The results suggested that the mutation ratio is an effect rather than a cause for the acquisition of antibiotic resistance.

As ROS is correlated with M. tuberculosis pathogenesis, we explored whether mycobacteria obtain any benefits when evolving to adjust their resistance to H₂O₂. Our results showed that endogenous H₂O₂ resistance confers bacterial growth advantages. Newly identified genes, such as carD, moaD1, and pp54, involved in oxidative stress, have been shown to be important for bacterial survival inside host cells (Mestre et al., 2013). A correlation has also been shown between resistance to oxidative stress and survival in macrophages. In line with this hypothesis, our results showed that H₂O₂-resistant strain mc²51 had a statistically significant growth advantage in macrophages compared with wild-type mc²155 (Fig. 2D). Consistent with a previous study on the relationship of Fur to INH and KatG, the resistance of H₂O₂ was partially abrogated when wild type Fur was expressed in mc²155 (Table S3). In addition, MS data showed that mc²51 had increased enzyme activity of KatG (Fig. 2F). Moreover, the SNPs in Fur were also identified in INH-resistant isolates (Ramaswamy et al., 2003). This finding implies that mutant Fur could be produced in humans. Thus, these findings support our hypothesis that the evolutional adaptation of mycobacteria to H₂O₂, accompanied by slow growth, confers growth benefits within macrophages on the mycobacteria.

References

[1] Armstrong JA, Hart PD (1971) Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J Exp Med 134:713–740

[2] Balbi KJ, Rocha EP, Feil EJ (2009) The temporal dynamics of slightly deleterious mutations in Escherichia coli and Shigella spp. Mol Biol Evol 26:345–355

[3] Ford CB, Lin PL, Chase MR, Shah RR, Iartchouk O, Galagan J, Mohaideen N, Ioerger TR, Sacchettini JC, Lipsitch M et al (2011) Use of whole genome sequencing to estimate the mutation rate of Mycobacterium tuberculosis during latent infection. Nat Genet 43:482–486

[4] Hershberg R, Lipatov M, Small PM, Sheffer H, Niemann S, Homolka S, Roach JC, Kremer K, Petrov DA, Feldman MW et al (2008) High functional diversity in Mycobacterium tuberculosis driven by genetic drift and human demography. PLoS Biol 6:e311

[5] Keren I, Wu Y, Inocencio J, Mulcahy LR, Lewis K (2013) Killing by bactericidal antibiotics does not depend on reactive oxygen species. Science 339:1213–1216

[6] Kohanski MA, DePristo MA, Collins JJ (2010) Sublethal antibiotic treatment leads to multidrug resistance via radical-induced mutagenesis. Mol Cell 37:311–320

[7] Kumar A, Farhana A, Guidry L, Saini V, Hondalus M, Steyn AJ (2011) Redox homeostasis in mycobacteria: the key to tuberculosis control? Expert Rev Mol Med 13:e39

[8] Little TJ, Allen JE, Babayan SA, Matthews KR, Colegrave N (2012) Harnessing evolutionary biology to combat infectious disease. Nat Med 18:217–220

[9] Mestre O, Hurtado-Ortiz R, Dos Vultos T, Namouchi A, Cimino M, Pimentel M, Neyrolles O, Gicquel B (2013) High throughput phenotypic selection of Mycobacterium tuberculosis mutants with impaired resistance to reactive oxygen species identifies genes important for intracellular growth. PLoS One 8:e53486

[10] Podinovskaia M, Lee W, Caldwell S, Russell DG (2013) Infection of macrophages with Mycobacterium tuberculosis induces global modifications to phagosomal function. Cell Microbiol 15:843–859

[11] Ramaswamy SV, Reich R, Dou S-J, Jasperse L, Pan X, Wanger A, Quitugua T, Graviss EA (2003) Single nucleotide polymorphisms in genes associated with isoniazid resistance in Mycobacterium tuberculosis. Antimicrob Agents Chemother 47:1241–1250

[12] Timmins GS, Deretic V (2006) Mechanisms of action of isoniazid. Mol Microbiol 62:1220–1227

[13] Wang ZY, Xiong M, Fu LY, Zhang HY (2013) Oxidative DNA damage is important to the evolution of antibiotic resistance: evidence of mutation bias and its medicinal implications. J Biomol Struct Dyn 31:729–733

[14] WHO (2010). Multidrug and extensively drug-resistant TB (M/XDR-TB): 2010 global report on surveillance and response. WHO/HTM/TB/20103

[15] Wilson LG (2005) Commentary: medicine, population, and tuberculosis. Int J Epidemiol 34:521–524

RIGHTS & PERMISSIONS

The Author(s) 2014.