TEXT

Top Abstract Text References

The RecA protein is a central component of the SOS response, which involves expression of a regulon comprising more than 20 unlinked genes and operons (17). The SOS response is characterized by an increased capacity for DNA repair and mutagenesis (9). In the pathogenic mycobacteria Mycobacterium tuberculosis and Mycobacterium leprae, the recA gene is comprised of a single open reading frame, which is interrupted by an intein coding sequence (2, 4). These two protein introns are different in size, sequence, and location of insertion of their coding sequences into the respective recA genes (4), suggesting that they have been acquired independently. In general, inteins are excised precisely from the precursor protein, and the flanking exteins are ligated to form the mature protein (3, 7, 8, 12). A wide variety of heterologous systems have been used to demonstrate splicing, and most of the available evidence points to the fact that protein splicing is an autocatalytic reaction, which does not require the involvement of accessory molecules (14). In contrast to the M. tuberculosis RecA, the 79-kDa precursor protein of the M. leprae RecA is spliced only in the native organism and not in Escherichia coli (4).

Thus, it was inferred that the M. leprae RecA intein may provide an example of conditional protein splicing, with the splicing reaction requiring some accessory mycobacterial protein or with the splicing activity controlled by a regulatory protein, although the possibility remained that the failure to splice in E. coli was a result of incorrect protein folding (1, 4).

Mycobacterium smegmatis recA mutant strain KS (5) was transformed with the E. coli-mycobacterium shuttle vector pRML-58a, a derivative of pHint (6) carrying a 4.1-kbp SphI/ClaI M. leprae recA fragment; this vector integrates at the mycobacterial attB site. Successful integration was demonstrated by PCR and Southern blot analysis (data not shown). For further investigations, M. smegmatis recA mutant strains complemented with M. tuberculosis recA (KS recA::aph/recA⁺ M. tuberculosis) or M. smegmatis recA (KS recA::aph/recA⁺ M. smegmatis) were used as controls.

Expression of RecA was investigated after ofloxacin induction (1 µg/ml for 5 h) (5, 11), by Western blot analysis (Fig. 1). A single band of approximately 38 kDa is found in protein extracts from the recA⁺ strain M. smegmatis mc² 155 SMR5 (lane 1) but is absent from M. smegmatis KS recA::aph (lane 2). Upon complementation of the recA mutant with recA genes from M. smegmatis (lane 3), M. tuberculosis (lane 4), and M. leprae (lane 5), RecA expression is found. Significantly, M. smegmatis recA mutant cells complemented with pRML-58a show a protein with a molecular mass of 38 kDa, indicating expression of the mature form of the introduced M. leprae recA gene. Although the antiserum used is able to recognize the unspliced precursor form of the M. leprae RecA protein (79 kDa), a protein with a corresponding molecular mass is not observed.

View larger version (62K): [in this window] [in a new window]   FIG. 1.   Detection of mycobacterial RecA protein in extracts from M. smegmatis. Lane 1, mc2 155 SMR5; lane 2, KS recA::aph; lane 3, KS recA::aph/recA+ (M. smegmatis); lane 4, KS recA::aph/recA+ (M. tuberculosis); lane 5, KS recA::aph/recA+ (M. leprae L58a). Approximately 10 µg of protein was loaded on a 12.5% sodium dodecyl sulfate gel, electroblotted, and developed following incubation with antiserum raised against M. tuberculosis RecA. The antiserum cross-reacts with the RecA proteins from M. leprae, M. tuberculosis, and M. smegmatis.

To functionally characterize the M. leprae RecA, complemented M. smegmatis recA mutant cells were investigated with respect to the ability to perform DNA repair following treatment with DNA-alkylating agents (ethyl methanesulfonate-methyl methanesulfonate) or UV irradiation (5). Transformation of M. smegmatis KS recA::aph with pRML-58a fully restored the wild-type phenotype, i.e., growth in the presence of ethyl methanesulfonate-methyl methanesulfonate (data not shown), and resistance to UV irradiation (Fig. 2) of the complemented strain was indistinguishable from that of recA⁺ mc² 155 SMR5 and from that of the KS recA::aph strains complemented with the recA gene from M. tuberculosis or M. smegmatis, respectively.

View larger version (14K): [in this window] [in a new window]   FIG. 2.   UV resistance of M. smegmatis recA mutant complemented with recA from M. smegmatis, M. tuberculosis, and M. leprae. Symbols: triangles, M. smegmatis mc2 155 SMR5; stars, KS recA::aph; diamonds, KS recA::aph/recA+ (M. smegmatis); inverted triangles, KS recA::aph/recA+ (M. tuberculosis); rectangles, KS recA::aph/recA+ (M. leprae L58a). The ability to form colonies after UV irradiation was determined by plating dilutions on 7H10 agar plates. The number of CFU was calculated after 3 days of incubation at 37°C.

The ability of the M. leprae RecA protein to catalyze homologous recombination was investigated by transformation experiments. M. smegmatis recA⁺ and M. smegmatis recA mutant complemented with M. smegmatis or M. leprae recA were transformed with a suicide vector targeting pyrF (5). The number of clones obtained by transformation of KS recA::aph/recA⁺ (M. leprae L58a) with the suicide vector pHRM-Gm (5) and subsequent selection on gentamicin was comparable to the number of transformants obtained upon transformation of the parental recA⁺ strain M. smegmatis mc² 155 SMR5 or the recA mutant strain complemented with M. smegmatis recA. In contrast, no transformants were obtained upon transformation of the recA mutant strain (Table 1). These data suggest that the M. leprae RecA is able to mediate integration of exogenous nucleic acids by homologous recombination.

Transformants obtained by positive selection with gentamicin are expected to result from a single crossover (15). These transformants possess one functional and one inactivated copy of pyrF and thus cannot be differentiated phenotypically from the parental pyrF⁺ strain. To verify that transformants obtained with the suicide vector pHRM-Gm resulted from homologous recombination at the pyrF locus, transformants were subjected to a second counterselection (30 µg of gentamicin per ml and 100 µg of streptomycin per ml [15]) to result in intrachromosomal recombination and deletion of the wild-type pyrF gene. Transformants resistant to gentamicin and streptomycin were investigated for their PyrF phenotype. pyrF mutants are uracil auxotrophs and resistant to fluoroorotic acid (FOA), a toxic analog of orotic acid. All gentamicin- and streptomycin-resistant derivatives obtained by counterselection from the recA⁺ strain and the strains of the KS recA::aph mutant complemented with the recA fragments from M. smegmatis and M. leprae, respectively (five of five each investigated), were uracil dependent and FOA resistant, indicating allelic exchange of pyrF by two successive homologous crossover events (Table 1).

Since the discovery of protein splicing (7, 8), much effort has been devoted to the elucidation of the biochemical reactions that underlie the protein splicing process and to the identification of the catalytic groups and the structural elements of inteins that participate in protein splicing (14). The intein plus the first downstream extein residue contain sufficient information for the splicing reaction to occur. For efficient splicing of an intein, four nucleophilic attacks mediated by three of four conserved splice junction residues are required: a serine, threonine, or cysteine at the intein N terminus; an asparagine at the intein C terminus; and a serine, threonine, or cysteine at the downstream extein N terminus. A penultimate histidine at the C terminus of the intein assists the cleavage reaction (13).

The M. tuberculosis RecA intein is 440 amino acids in size and flanked by a cysteine at the intein N terminus, an asparagine at the intein C terminus, and a cysteine at the downstream extein N terminus. The intein of the M. leprae RecA is 365 amino acids in size and is also flanked by a cysteine at the intein N terminus and an asparagine at the intein C terminus but carries a serine at the downstream extein N terminus (4). In the case of the M. tuberculosis RecA, the protein was spliced appropriately when expressed in E. coli and the protein splicing reaction was found to occur in trans when using purified N- and C-terminal fragments of the intein (10). In contrast, the M. leprae 79-kDa RecA precursor protein is not spliced in E. coli (4). It was thus suggested that the M. leprae RecA intein might represent an example of conditional protein splicing (4).

M. smegmatis is a nonpathogenic mycobacterium which is frequently used in mycobacterial genetics. This species has a single contiguous, i.e., inteinless, RecA, similar to most other eubacterial species. M. smegmatis recA mutant strains are unable to promote homologous recombination (5, 11). Using complementation of the M. smegmatis recA mutant, it has been possible for the first time to perform functional investigations of the M. leprae RecA. Previously, the M. leprae RecA intein was found to splice in M. leprae itself but not when expressed in E. coli. Complementation of an M. smegmatis recA mutant strain with the M. leprae recA gene fully restored the RecA⁺ phenotype with respect to DNA repair as well as with respect to integration of exogenous nucleic acids by homologous recombination; in addition, a mature 38-kDa RecA protein was detected by immunoblotting. Presumably, if some accessory mycobacterial protein is required for functional splicing of the M. leprae RecA intein, M. smegmatis must provide some enzymatic activity allowing splicing of the M. leprae RecA precursor protein. More likely, in M. smegmatis, but not in E. coli, the RecA protein is folded properly, enhancing the self-splicing reaction. In this regard, it should be noted that the rate of synthesis of the precursor protein may be instrumental in enabling correct folding, for in M. smegmatis the M. leprae RecA was expressed from its own promoter, whereas in E. coli it was expressed from the strong T7 promoter, partly because its native promoter did not function in E. coli (4).