Population heterogeneity in Mycobacterium smegmatis and Mycobacterium abscessus

Abstract

Bacteria use population heterogeneity, the presence of more than one phenotypic variant in a clonal population, to endure diverse environmental challenges - a 'bet-hedging' strategy. Phenotypic variants have been described in many bacteria, but the phenomenon is not well-understood in mycobacteria, including the environmental factors that influence heterogeneity. Here, we describe three reproducible morphological variants in M. smegmatis - smooth, rough, and an intermediate morphotype that predominated under typical laboratory conditions. M. abscessus has two recognized morphotypes, smooth and rough. Interestingly, M. tuberculosis exists in only a rough form. The shift from smooth to rough in both M. smegmatis and M. abscessus was observed over time in extended static culture, however the frequency of the rough morphotype was high in pellicle preparations compared to planktonic culture, suggesting a role for an aggregated microenvironment in the shift to the rough form. Differences in growth rate, biofilm formation, cell wall composition, and drug tolerance were noted among M. smegmatis and M. abscessus variants. Deletion of the global regulator lsr2 shifted the M. smegmatis intermediate morphotype to a smooth form but did not fully phenocopy the naturally generated smooth morphotype, indicating Lsr2 is likely downstream of the initiating regulatory cascade that controls these morphotypes. Rough forms typically correlate with higher invasiveness and worse outcomes during infection and our findings indicate the shift to this rough form is promoted by aggregation. Our findings suggest that mycobacterial population heterogeneity, reflected in colony morphotypes, is a reproducible, programmed phenomenon that plays a role in adaptation to unique environments and this heterogeneity may influence infection progression and response to treatment.

Keywords: Lsr2; Mycobacterium abscessus; Mycobacterium smegmatis; heterogeneity; morphology.

Fig. 1.

Mycobacterium smegmatis exists in three semi-stable phenotypic variants. Morphotype A colonies were round, smooth, and hydrophilic. Morphotype B colonies were irregular, wrinkly, and moderately hydrophilic. Morphotype C colonies were irregular, hydrophobic, and raised. (a-c) Morphotype A after 3 days incubation at 37 °C, after seven additional days incubation at 23 °C, and under light microscopy after 2 days incubation at 37 °C using 10× objective. (d) Confluent morphotype A with 5 µl water in centre. (e-g) Morphotype B after 3 days incubation at 37 °C, after seven additional days incubation at 23 °C, and using light microscopy after 2 days incubation at 37 °C using 10× objective. (h) Confluent morphotype B with 5 µl water in centre. (i-k) Morphotype C after 3 days incubation at 37 °C, after seven additional days incubation at 23 °C, and using light microscopy after 2 days incubation at 37 °C using 10× objective. (l) Confluent morphotype C with 5 µl water in centre. (m-o) M. smegmatis Δlsr2 colonies after 3 days incubation at 37 °C, after seven additional days incubation at 23 °C, and using light microscopy after 2 days incubation at 37 °C using 10× objective. (p) Confluent M. smegmatis Δlsr2 with 5 µl water in centre.

Fig. 2.

Mycobacterium abscessus exists in two semi-stable phenotypic variants. Smooth colonies were round, domed, and wet. Rough colonies were irregular, wrinkly, and dry. Rough colonies appeared to exhibit a cording-like phenomenon when examined using light microscopy. (a) Smooth morphotype after 3 days incubation on LB agar at 37 °C. (b) Smooth morphotype under light microscopy after 2 days incubation at 37 °C using 10× objective. (c) Rough morphotype after 3 days incubation on LB agar at 37 °C. (d) Rough morphotype under light microscopy after 2 days incubation at 37 °C using 10× objective.

Fig. 3.

M. tuberculosis was observed in only one form, similar to the rough forms of M. smegmatis and M. abscessus . (a) Colonies by gross observance were dry and rough, while the (b) colonies under light microscopy exhibited cording-like structures.

Fig. 4.

M. smegmatis B morphotype switched to A (a) and C (b) morphotypes over time. The A morphotype was able to shift to the B morphotype in extended settling culture (c). M. abscessus smooth morphotype had increased number of rough colonies when settled over time (d). Each coloured circle represents an individual biological replicate. Both M. smegmatis and M. abscessus had increased number of rough forms in pellicle preparation after 3 weeks across four biological replicates (e, f). Error bars represent standard error of the mean.

Fig. 5.

M. smegmatis A morphotype grew more slowly compared to other morphotypes. OD_600nm was measured over 24 h of stirring aerobic growth in DTA in three biological replicates. Error bars represent standard error of the mean. Doubling times of B and C morphotypes as well as M. smegmatis Δlsr2 were the same, A morphotype doubling time was significantly longer (P<0.001).

Fig. 6.

The A morphotype was deficient in biofilm formation while the C morphotype produced most robust biofilm. Dried biofilms of M. smegmatis A (a), B (b), C (c) morphotypes, and M. smegmatis Δlsr2 (d). M. abscessus dried biofilms of smooth (e) and rough (f) morphotypes. Quantification of M. smegmatis biofilm (g) and M. abscessus biofilm (h) using Gram safranin (*** P≤0.001). Images are representative of three biological replicates.

Fig. 7.

M. smegmatis C form was devoid of sliding motility and glycopeptidolipid production. M abscessus rough form had no sliding motility. Motility agar 2 weeks after inoculation with M. smegmatis (a) A, (b) B, and (c) C morphotypes, (d) M. smegmatis Δlsr2, (e) M. abscessus smooth, and (f) M. abscessus rough morphotypes. Images are representative of three biological replicates. (g) Thin layer chromatography showing acylated glycopeptidolipids.

Fig. 8.

M. smegmatis morphotypes responded differently to extended drug pressure. Drug concentrations were as follows: (a) rifampicin 12.5 µg ml⁻¹, (b) isoniazid 125 µg ml⁻¹, (c) norfloxacin 2.5 µg ml⁻¹, (d) kanamycin 2.5 µg ml⁻¹, (e) chlorpromazine 60 µg ml⁻¹, (f) streptomycin 0.5 µg ml⁻¹, and (g) ethambutol 2 µg ml⁻¹. Gold lines represent the A morphotype, black lines represent the B morphotype, purple lines represent the C morphotype, and teal lines represent the M. smegmatis lsr2 deletion mutant. Symbols represent the mean of three biological replicates with error bars representing standard error of the mean (* P≤0.05, ** P≤0.01, *** P≤0.001, **** P≤0.0001).

Fig. 9.

M. abscessus morphotypes responded differently to extended drug pressure. The rough form was less affected by drug than the smooth form. Drug concentrations were as follows: (A) moxifloxacin 4 µg ml⁻¹, (B) amikacin 6 µg ml⁻¹, (C) imipenem 2 µg ml⁻¹, (D) tigecycline 4 µg ml⁻¹, (E) bedaquiline 4 µg ml⁻¹, and (F) clofazimine 5 µg ml⁻¹. Black lines represent the smooth morphotype and pink lines represent the rough morphotype. Symbols represent the mean of three biological replicates with error bars representing standard error of the mean (* P≤0.05, ** P≤0.01, *** P≤0.001, **** P≤0.0001).