Illegitimate recombination: an efficient method for random mutagenesis in Mycobacterium avium subsp. hominissuis

Abstract

Background: The genus Mycobacterium (M.) comprises highly pathogenic bacteria such as M. tuberculosis as well as environmental opportunistic bacteria called non-tuberculous mycobacteria (NTM). While the incidence of tuberculosis is declining in the developed world, infection rates by NTM are increasing. NTM are ubiquitous and have been isolated from soil, natural water sources, tap water, biofilms, aerosols, dust and sawdust. Lung infections as well as lymphadenitis are most often caused by M. avium subsp. hominissuis (MAH), which is considered to be among the clinically most important NTM. Only few virulence genes from M. avium have been defined among other things due to difficulties in generating M. avium mutants. More efforts in developing new methods for mutagenesis of M. avium and identification of virulence-associated genes are therefore needed.

Results: We developed a random mutagenesis method based on illegitimate recombination and integration of a Hygromycin-resistance marker. Screening for mutations possibly affecting virulence was performed by monitoring of pH resistance, colony morphology, cytokine induction in infected macrophages and intracellular persistence. Out of 50 randomly chosen Hygromycin-resistant colonies, four revealed to be affected in virulence-related traits. The mutated genes were MAV_4334 (nitroreductase family protein), MAV_5106 (phosphoenolpyruvate carboxykinase), MAV_1778 (GTP-binding protein LepA) and MAV_3128 (lysyl-tRNA synthetase LysS).

Conclusions: We established a random mutagenesis method for MAH that can be easily carried out and combined it with a set of phenotypic screening methods for the identification of virulence-associated mutants. By this method, four new MAH genes were identified that may be involved in virulence.

Figure 1

Sketch showing randomly mutated genes distributed within the M. avium genome. Genes location mapped on the genome after sequencing.

Figure 2

Sketch illustrating the genetic characterisation of the mutants MAV_1778, MAV_3128, MAV_4334, and MAV_5106. The sites of the insertion of the marker (Hyg^r gene) were identified by inverse PCR followed by sequencing of the eluted PCR products. The figure shows for four mutants the mutated gene (dark blue) with the site of insertion of the fragment (grey) carrying the Hyg^r gene (red) and the four genes located upstream and downstream of the mutated gene (light blue). Numbers in the arrows indicate the gene names. The direction of the arrows stands for gene direction. Gene sizes and distances between genes are approximations. Below the map of each mutant the size of the deletion generated as result of insertion of the marker is indicated.

Figure 3

Colony morphology upon plating on Congo Red agar plates. Well-grown broth cultures of all strains were diluted 1:10⁶ and 100 μl plated in triplicate onto Middlebrook agar with OADC containing 100 μg ml^-1 Congo Red. Plates were incubated on average for three weeks. The arrows point to smooth-domed-opaque (sdo), smooth-flat-red (sfr), smooth transparent (st), rough red (rr) and rough transparent (rt) colonies. A: WT; B: mutant MAV_2555; C: mutant MAV_1888; D: mutant MAV_4334; E: mutant MAV_5106; F: mutant MAV_1778; G: mutant MAV_3128; H: mutant MAV_3625; I: mutant MAV_2599.

Figure 4

Resistance towards pH stress. The bacteria were grown in Middlebrook 7H9 broth with OADC at pH 7 and pH 5 during 11 days; the ATP content was recorded by quantification of the amount of ATP in the cultures. The amount of ATP is represented as RLU (relative light units). A: WT and mutant MAV_1778; B: WT and mutant MAV_3128; C: WT and mutant MAV_3625; D: WT and mutant MAV_2599.

Figure 5

Induction of IL-10 cytokine secretion by infected macrophages. THP-1 cells (2.0x10⁵) were infected (MOI 50) with mutants and WT. After 24 hours cytokines from supernatants were measured by ELISA. When compared to WT a P value <0.01 (two-tailed, unpaired Mann–Whitney test) was considered very significant (**).

Figure 6

Intracellular survival of mutants compared to WT in human monocytes. Human blood monocytes (1.0x10⁶) from healthy volunteers were infected (MOI 10) with mutants and WT. Intracellular bacteria were quantified after 4 hour of infection, and after 1, 2, & 4 days. The monocytes were lysed in 1 ml of sterile water and 100 μl of 1:500 dilution in sterile water of sample were plated on Middlebrook agar plates supplemented with ADC for CFU counting. A: WT and mutant MAV_4334; B: WT and mutant MAV_5106; C: WT and mutant MAV_1778; D: WT and mutant MAV_3128. Statistical analysis was done using a two tailed, paired Student’s t test. When compared to wild-type a P < 0.05 was considered significant (*) and a P < 0.01 very significant (**).

Figure 7

Phenotype of the complemented strain MAV3128Comp compared to mutant MAV_3128 and WT. A: Colony morphology on Congo Red plates. B: Intracellular survival in human blood monocytes.