Biosynthetic specificity of the rhamnosyltransferase gene of Mycobacterium avium serovar 2 as determined by allelic exchange mutagenesis

Joel N. Maslow1,2, Vida R. Irani2, Sun-Hwa Lee3, Torsten M. Eckstein4, Julia M. Inamine4 and John T. Belisle4

1 Section of Infectious Diseases, VA Medical Center (151), University and Woodland Aves, Philadelphia, PA 19104, USA
2 Division of Infectious Diseases, University of Pennsylvania, Philadelphia, PA, USA
3 New England Regional Primate Center, Southborough, MA, USA
4 Mycobacterial Research Laboratories, Department of Microbiology, Immunology and Pathology, Colorado State University, Fort Collins, CO, USA

Correspondence
Joel N. Maslow
joel.maslow{at}med.va.gov


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In prior studies, through recombinant expression in Mycobacterium smegmatis, the rtfA gene of Mycobacterium avium was shown to encode a rhamnosyltransferase that catalyses the addition of rhamnose (Rha) to the 6-deoxytalose of serovar 2-specific glycopeptidolipid (GPL). Whether RtfA also catalyses the transfer of Rha to the alaninol of the lipopeptide core is unknown. An isogenic rtfA mutant of M. avium serovar 2 strain TMC724 was derived using a novel allelic exchange mutagenesis system utilizing a multicopy plasmid that contained the katG gene of Mycobacterium bovis and the gene encoding green fluorescent protein (gfp). Overexpression of KatG in M. avium resulted in increased susceptibility to isoniazid, thus providing counter-selection by enriching for clones that had lost plasmid DNA. Plasmid loss was confirmed by screening for gfp-negative clones to select putative allelic exchange mutants. Two exchange mutants were created, confirmed by Southern hybridization, and demonstrated loss of serovar 2-specific GPL by thin-layer chromatography (TLC). Gas chromatography of alditol acetate derivatives revealed the loss of Rha and the terminal 2,3-O-Me-fucose and preservation of 3-O-Me-Rha and 3,4-O-Me-Rha substituents at the terminal alaninol of the lipopeptide core. Complementation of rtfA in trans through an integrative plasmid restored serovar 2-specific GPL expression identical to wild-type TMC724. This result shows that rtfA encodes an enzyme responsible only for the transfer of Rha to the serovar 2-specific oligosaccharide and provides a system of allelic exchange for M. avium as a tool for future genetic studies involving this species.


Abbreviations: 6-dTal, 6-deoxytalose; GPL, glycopeptidolipid; INH, isoniazid; LP, lipopeptide; nsGPL, non-specific glycopeptidolipid; Rha, rhamnose; ssGPL, serovar-specific glycopeptidolipid


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Members of the Mycobacterium avium complex (MAC) are ubiquitous organisms found in soil and natural bodies of water and are common colonizers of potable hot water systems (Brooks et al., 1984; George et al., 1980; von Reyn et al., 1993, 1994). Although initially considered as an opportunistic pulmonary pathogen, disseminated MAC infection, especially among patients with end-stage AIDS, was described in many reports in the 1980s and 1990s (Hawkins et al., 1986; Horsburgh, 1991; Nightingale et al., 1992). In the last 10 years, there have been significant advances in antiretroviral treatment of HIV infection as well as the introduction of prophylactic regimens to prevent the onset of disseminated MAC (DMAC) infection (Havlir et al., 1996; Pierce et al., 1996; Shafran et al., 1996). Although the incidence of DMAC infection has been significantly reduced, this disease continues to cause significant morbidity among patients with AIDS (Burman et al., 1998).

The pathogenesis of DMAC is largely unresolved. Numerous bacterial molecules have been postulated as candidate virulence factors or factors that modulate the host cellular response. The glycopeptidolipids (GPLs), the dominant MAC glycolipid (Brennan, 1988; McNeil et al., 1987), have long been considered to have a role in pathogenesis (Draper & Rees, 1970, 1973); however, there are conflicting data as to the relative effect of GPL from different serovars and the relative contribution of each GPL subunit. Heat-killed staphylococci coated with serovar 4, 12 and 17 GPL strongly promoted macrophage uptake and markedly inhibited phagosome–lysosome (PL) fusion, GPL of serovars 5, 7, 8 and 10 stimulated both phagocytosis and PL fusion, and GPL of serovars 9, 13, 16 and 19 neither promoted phagocytosis nor inhibited PL fusion (Minami, 1998). Using whole bacteria, a virulent serovar 4 strain of M. avium induced protein kinase C-dependent TNF{alpha}, IL-1{beta} and IL-6 (Gan et al., 1993; Newman et al., 1991), effects that appear to be related to GPL alone (Barrow et al., 1995). In contrast, others have reported that serovar 8-specific GPL induced significant levels of TNF{alpha}, IL-1{beta} and IL-6 from human peripheral blood mononuclear cells (PBMCs) and serovar 4 GPL did not (Barrow et al., 1995; Horgan et al., 2000). While two groups of investigators demonstrated that, following {beta}-elimination of GPL, the modified lipopeptide (LP) core of serovar 4-specific GPL, as opposed to the reduced oligosaccharide, inhibited mitogen-induced blastogenesis (Tassell et al., 1992) and suppressed phytohaemagglutinin-induced lymphoproliferation of PBMCs (Barrow et al., 1995), others have reported that the LP core of serovar 4-specific GPL had only a minimal ability to induce TNF{alpha} relative to unmodified GPL (Barrow et al., 1995). Significantly limiting the development of a consistent framework of the role of GPL in mycobacterial pathogenesis has been the inability to construct isogenic strains differing in GPL structure, necessitating the use of genetically distinct strains of differing serotypes.

Two forms of GPL are expressed on smooth morphotypes of M. avium: serovar-specific glycopeptidolipid (ssGPL) and non-specific glycopeptidolipid (nsGPL). Both nsGPLs and ssGPLs possess a LP core composed of D-phenylalanine-D-allo threonine-D-alanine-L-alaninol (Phe-Thr-Ala-alaninol) with a fatty acid N-linked to Phe, a 6-deoxytalose (6-dTal) substitution on the allo Thr, and mono- or dimethylated rhamnose (Rha) linked to the C terminus of alaninol. The ssGPLs are further glycosylated with a variable haptenic oligosaccharide (Aspinall et al., 1995). For all serovars of the M. avium complex, except serovars 5, 10 and 11, Rha appears as the first sugar moiety attached to the 6-dTal residue (Aspinall et al., 1995) and was therefore chosen as the initial target for study within the serovar gene cluster.

Genes encoding the glycosylation of the serovar 2-specific GPL are localized to a contiguous 35 kb region on the M. avium chromosome termed the ser2 gene cluster (Belisle et al., 1991). Four loci, serAD, were identified and demonstrated to encode putative glycosyltransferases, methyltransferases and genes for the de novo synthesis of L-fucose (Belisle et al., 1993a, b; Mills et al., 1994). Biosynthesis of the LP core was shown to reside downstream of the ser2 gene cluster (Belisle et al., 1993b). Within the ser2A locus, three open reading frames (ORFs 1–3) were identified, including two putative glycosyltransferases and a methyltransferase (Eckstein et al., 1998). Recombinant expression in Mycobacterium smegmatis demonstrated that ORF2 encodes a glycosyltransferase that links Rha to 6-dTal, and was termed rtfA (Eckstein et al., 1998). However, these experimental data could not rule out that RtfA was also responsible for the transfer of Rha to the L-alaninol of the GPL LP core.

To demonstrate clearly the role of rtfA in the biosynthesis of ssGPL, an rtfA deletion mutant was created in serovar 2 strain TMC724 and represents the first successful creation of allelic exchange mutants of M. avium. This was accomplished using an episomal vector carrying the katG gene of M. bovis BCG to confer isoniazid (INH) susceptibility and provided the ability to counter-select for clones that had lost plasmid DNA. Positive co-selection was accomplished with the hygromycin-resistance gene (hyg) and the green fluorescent protein (gfp) gene. Complementation of the rtfA knockout mutant with wild-type rtfA restored GPL expression. Thus, we were able to demonstrate that disruption of rtfA resulted in the inability to glycosylate 6-dTal but did not disrupt the transfer of Rha to the L-alaninol of the LP core.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and chemicals.
The mycobacterial strains and plasmids used in this study are listed in Tables 1 and 2, respectively. Escherichia coli DH5{alpha} was used as the host strain for plasmid construction and propagation. Growth and selection of recombinant E. coli clones were carried out using Luria–Bertani (LB) medium (American Bioanalytical) with 50 µg ampicillin ml-1 or 25 µg kanamycin ml-1. Mycobacterium spp. were grown in Middlebrook 7H9 broth or 7H11 agar (Difco) supplemented with 10 % OADC (oleic acid, albumin, dextrose complex). M. smegmatis mc2155 is a transformation competent strain that is widely used to express or carry recombinant mycobacterial genes (Snapper et al., 1990). M. avium TMC724 is a serovar 2 strain for which the region encoding GPL biosynthesis has been isolated and sequenced (GenBank accession no. AF125999) (Belisle et al., 1991; Eckstein et al., 1998; Mills et al., 1994). M. avium 920A-6 is an electrocompetent, serovar 8 bloodstream isolate cultured from a patient with AIDS (Arbeit et al., 1993; Lee et al., 2002). For M. smegmatis and M. avium, selection of recombinant clones was accomplished using hygromycin at 100 µg ml-1 or kanamycin at 50 µg ml-1. All chemicals were purchased from Sigma.


View this table:
[in this window]
[in a new window]
 
Table 1. Bacterial strains used in this study

 
Bacterial transformation.
Transformation of E. coli was performed using calcium-chloride-treated cells (Sambrook et al., 1989). Transformation of M. smegmatis was performed by electroporation as described by Snapper et al. (1990). Transformation of M. avium was performed by electroporation as described by Lee et al. (2002).

Isolation and identification of the BCG katG gene.
The katG gene was isolated from an M. bovis BCG genomic library contained in the mycobacterium–E. coli shuttle vector pYUB415 (kindly provided by Stoyan Bardarov). This genomic library was used to transform M. smegmatis mc2155 and five putative katG-positive transformants were identified as susceptible to INH at 1 µg ml-1. The presence of katG in each of the five transformants was confirmed by PCR. Bacterial lysates for PCR analyses were derived from late-exponential-phase cultures grown on selective agar. Bacteria were suspended in PBS (40 mM KH2PO4, 60 mM K2HPO4, pH 7·0) and heated to 95 °C for 10 min. After centrifugation at 15 000 g, the supernatant was recovered as a DNA lysate. The forward primer 5'-GAGATTGCCAGCCTTAAGAGC-3' corresponded to base positions 1360–1380 of the Mycobacterium tuberculosis katG gene (GenBank accession no. X68081; Zhang et al., 1992) and the reverse primer 5'-CAGGTGATACCCATGTCGAG-3' corresponded to base positions 1984–2003 to yield a 644 bp PCR product. Amplification was performed in a programmable thermal controller (MJ Research) with Taq polymerase (Perkin-Elmer) using a programme of 35 cycles of 92 °C for 2 min, 56 °C for 2 min and 72 °C for 1 min. Cosmid pVAP1 was isolated from one of the five transformants, propagated in E. coli DH5{alpha} and retransformed into M. smegmatis to confirm INH susceptibility.

Integrating cosmid pVAP9.2 was constructed by ligating the 40 kb PacI insert of pVAP1 to PacI/XbaI arms of pYUB412. The ligation product was packaged into E. coli phage {lambda} (Gigapack III Gold; Stratagene), transduced into DH5{alpha}, and selected on LB agar with ampicillin. A functional katG gene was isolated from pVAP1 as a 6 kb EcoRV fragment and cloned into the E. coli–mycobacterium shuttle vector pNBV1 (Howard et al., 1995) to create pVAP25 (11·8 kb).

Construction of allelic exchange vector pVAP41 and rtfA-containing vectors used for complementation.
The construction of the allelic exchange vector pVAP41 is presented schematically in Fig. 1. The rtfA gene was amplified from pJTB21 (Belisle et al., 1991). The forward primer 5'-CGCCGATGTCGTGGAGGACTT-3' (SHL01) and reverse primer 5'-GCTGCCATACCTGGTCAGATC-3' (SHL02) corresponded to positions 4030–4051 and 5924–5944, respectively, of the ser2 gene cluster of TMC724 (GenBank accession no. AF125999). PCR was performed using a programme of 35 cycles of 94 °C for 1 min, 62 °C for 2·5 min and 72 °C for 1 min. The 1914 bp PCR product was ligated into the EcoRV site of pBluescript SK (+/-) to create pVAP36 (4·8 kb). The 1·3 kb BamHI hyg cassette from pBSHGH (Lee et al., 2002) was inserted via blunt-end ligation into the unique NcoI site within rtfA to create pVAP37 (6·2 kb). The 3·2 kb HindIII–PstI fragment of pVAP37 containing rtfA : : hyg was ligated into the respective sites in the multiple cloning region of pMVGFP2 (Lee et al., 2002) to create pVAP40 (8·3 kb). Finally, katG, contained within the 6 kb EcoRV fragment of pVAP1, was ligated into the NheI site of pVAP40 to create pVAP41 (14·1 kb).



View larger version (18K):
[in this window]
[in a new window]
 
Fig. 1. Construction of allelic exchange vector pVAP41. The rtfA gene was amplified as a 1914 bp product from pJTB21 (Belisle et al., 1991) using forward primer SHL01 and reverse primer SHL02 and ligated into the EcoRV site of pBluescript SK (+/-) to create pVAP36 (4·8 kb). The 1·3 kb hyg BamHI cassette of pBSHGH (Lee et al., 2002) was inserted via blunt-end ligation into the unique NcoI site within rtfA to create pVAP37 (6·2 kb). The 3·2 kb HindIII–PstI fragment of pVAP37 containing rtfA : : hyg was ligated into the respective sites in the multiple cloning region of pMVGFP2 (Lee et al., 2002) to create pVAP40 (8·3 kb). Finally, the 6 kb EcoRV fragment of pVAP1 containing katG was ligated into the NheI site of pVAP40 to create pVAP41 (14·1 kb).

 
The stability of the pAL5000-based vector pVAP41 was determined for M. avium strain 920A-6. Following transformation with pVAP41, a gfp-positive, hygromycin-resistant clone was propagated non-selectively in 7H9 broth. Serial dilutions were inoculated onto selective 7H11 agar containing 100 µg hygromycin ml-1. After 13 days of non-selective incubation, 24·3 % of colonies remained hygromycin-resistant versus only 5·8 % of colonies after 20 days incubation. All colonies were green fluorescent protein (GFP)-positive, demonstrating a very low rate of spontaneous hygromycin resistance. Based on these data, the half-life of pVAP41 for M. avium appears to range between 5 and 6·5 days. These data were consistent for two separate cultures.

Plasmid pVAP52 was constructed as an integrative vector to complement the interrupted rtfA gene. To construct plasmid pVAP52, rtfA was amplified using primers SHL01and SHL02, except that SHL01 was extended upstream by 12 bp (5'-ACCATAAAGCTT-3') to include a HindIII restriction site (underlined) and SHL02 was extended downstream by 11 bp (5'-ACTCCAAGCTT-3') to include a HindIII restriction site. After restriction digestion with HindIII, this fragment was ligated into the HindIII site of pIGFP2 (Lee et al., 2002).

Isolation of genomic DNA.
Genomic DNA for Southern blot analysis was isolated from 30 ml late-exponential-phase cultures. Bacteria were washed with 30 ml TES (100 mM Tris/HCl, pH 8·0, 100 mM EDTA, pH 8·0, 100 mM NaCl), suspended in 2·5 ml of TES with 300 µl lysozyme (10 mg ml-1), and incubated at 37 °C for 1·5 h. Proteinase K and SDS were added to final concentrations of 1 mg ml-1 and 1 % (w/v), respectively, followed by incubation at 55 °C for 5 h. Phenol/chloroform/isoamyl alcohol (25 : 24 : 1 by vol.) was added at 1·5 vol., the suspension was vortexed for 2 min and then centrifuged at 15 000 g for 15 min at room temperature. The aqueous layer was transferred into a new microcentrifuge tube and 0·7 vol. of 2-propanol was added. After incubation at room temperature for 5 min, genomic DNA was collected by centrifugation at 15 000 g for 10 min at room temperature and washed with 70 % ETOH. The DNA pellet was air-dried and suspended in H2O.

Southern blot analysis was performed for genomic DNA subjected to overnight restriction digestion with HindIII (Promega) and resolved in a 0·8 % agarose gel at 35 V for 18 h. The restriction fragments were transferred to a non-charged nylon membrane (Duralon-UV; Schleicher and Schuell) by capillary action (Sambrook et al., 1989). Southern blots were probed for rtfA using a 1·9 kb PCR amplicon generated as above using the North2South Direct HRP labelling and detection kit (Pierce), following the manufacturer's directions.

Isolation and analysis of GPLs.
GPLs were isolated from lyophilized 100 ml bacterial cultures. Total lipids were extracted with CHCl3/CH3OH (2 : 1, v/v) and enrichment for GPLs was achieved by mild alkaline hydrolysis as described previously (Eckstein et al., 1998). The GPLs (100 µg) were resolved by TLC on 20 cmx20 cm silica gel 60 plates (Merck) using CHCl3/CH3OH/H2O (90 : 10 : 1 by vol.) as the mobile phase. Lipids were visualized by spraying with a solution of 10 % H2SO4 in ethanol and heating.

Carbohydrate analysis of the GPLs from the wild-type and recombinant strains of M. avium was performed by gas chromatography-mass spectroscopy (GC-MS) as described by Eckstein et al. (1998). Briefly, lipids were hydrolysed with 2 M CF3COOH and the free sugars were reduced with NaBD4 and acetylated with acetic anhydride. The resulting alditol acetates were resolved on a DB-5 column (10 mx0·18 mm; J&W Scientific) using a TraceTM GC 2000 (Thermo Quest) coupled to a Finnigin Polariz mass spectrometer (Thermo Quest). The programme used consisted of 1 min at 50 °C followed by an increasing temperature gradient of 30 °C min-1 to 150 °C and a final temperature gradient of 5 °C min-1 to 260 °C.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
M. avium cells transformed with katG display increased susceptibility to INH
The rtfA gene of M. avium strain TMC724 was chosen as our initial target to examine the biosynthesis of GPL. Preliminary experiments first investigated whether an allelic exchange system based on pPR27 (Pelicic et al., 1996a), containing both sacB and the temperature-sensitive origin of replication for Mycobacterium fortuitum plasmid pAL5000, could be used for M. avium. Electrocompetent M. avium strain 920A-6 was transformed with a pPR27-based plasmid, pVAP39 (data not shown). Both sucrose and temperature selection provided only minimal decreases in cell counts (<50 %; data not shown). Moreover, transformation of TMC724, a strain with a baseline low efficiency of transformation (Lee et al., 2002), could not be successfully electroporated with pVAP39. Thus, to generate such mutants in M. avium a more amenable counter-selection strategy was required.

KatG-mediated INH selection has been used successfully as a counter-selection marker for allelic exchange of M. bovis BCG (Norman et al., 1995). Given that M. avium is naturally resistant to relatively high concentrations of INH, we sought to determine whether an episomal copy of the M. bovis BCG katG gene could confer INH susceptibility to M. avium and be used as a marker to select mutants derived by homologous recombination.

Initially, M. smegmatis was transformed with the katG-containing plasmids pVAP1, pVAP9.2 and pVAP25 to create strains 127R, 146R.1 and 173R.1, respectively. INH susceptibility of each M. smegmatis transformant was compared to wild-type M. smegmatis. All transformants demonstrated increased susceptibility to INH compared with wild-type M. smegmatis. Strains 127R and 146R.1, carrying large episomal and integrative cosmids, demonstrated a 4–5 log decrease in cell numbers at an INH concentration of 0·25 µg ml-1 and were fully susceptible at 0·5 µg ml-1, whereas strain 173R.1 was fully susceptible to INH at a concentration of 0·05 µg ml-1. KatG-mediated INH susceptibility was further tested for recombinant clones 174R, 185R and 186R of M. avium 920A-6 SmO (Lee et al., 2002) carrying plasmids pVAP1, pVAP9.2 and pVAP25, respectively. Clones 174R and 185R demonstrated a pattern of INH susceptibility similar to wild-type 920A-6 (Fig. 2a) whereas clone 186R was highly susceptible to INH and demonstrated a 3·5 log decrease in cell numbers at an INH concentration of 0·5 µg ml-1 (Fig. 2a).



View larger version (26K):
[in this window]
[in a new window]
 
Fig. 2.
 
Finally, KatG-mediated INH susceptibility was assessed for the episomal delivery vector pVAP41. Strain 920A-6 SmO transformed with pVAP41 demonstrated bacterial killing intermediate between pVAP25 and pVAP1 with a 6 log decline in bacterial numbers at 0·5 µg ml-1 but no further increases in INH susceptibility (data not shown). Wild-type M. avium strain TMC724 was fully resistant to INH <=10 µg ml-1 and demonstrated an ~4·5 log decline at INH concentrations >=25 µg ml-1, whereas TMC724 transformed with pVAP41 demonstrated a 6·2 log decline at 10 µg ml-1 (Fig. 2b). These data further demonstrated that KatG-mediated INH susceptibility could be used to select mutants that had undergone homologous recombination, but that the level of KatG-mediated INH susceptibility and optimal concentration was highly strain-dependent.

Allelic exchange mutagenesis of rtfA of serovar 2 strain TMC724
The episomal delivery vector (pVAP41) was constructed to contain rtfA interrupted with the hygromycin-resistance gene (rtfA : : hyg), katG and gfp. TMC724(pVAP41) transformants were grown non-selectively to allow for homologous recombination and then inoculated onto INH- and hygromycin-containing medium. INH-susceptible clones that retained episomal copies of plasmid DNA should be eliminated, whereas clones that had lost plasmid DNA would be retained. Conversely, hygromycin selection would select for INH-resistant clones that had undergone either homologous recombination with rtfA : : hyg for rtfA or chromosomal recombination of plasmid DNA via single crossover mutagenesis or by illegitimate recombination. The loss of gfp-mediated fluorescence (Lee et al., 2002) could then be used to detect clones that had undergone homologous recombination.

M. avium strain TMC724 electroporated with 1 µg of pVAP41 yielded four GFP-positive colonies. Each of the four clones yielded both a 1·9 kb PCR product corresponding to wild-type rtfA and a 3·2 kb product corresponding to rtfA interrupted with hyg (rtfA : : hyg). One transformant was grown non-selectively in 7H9 medium for 24 days to provide sufficient time for homologous recombination and to allow for spontaneous plasmid loss. The cell suspension was then inoculated onto 7H11 medium supplemented with INH and hygromycin to select for allelic exchange mutants. Screening by fluorescence microscopy detected six GFP-negative clones, 218R.1 to 218R.6, of approximately 12 000 colonies examined.

PCR for rtfA was then performed for the six clones. Three clones (218R.3, 218R.4 and 218R.5) yielded a single 3·2 kb PCR product consistent with rtfA : : hyg. Clone 218R.6 yielded both a 1·9 kb PCR product and a 3·2 kb band, signifying the presence of wild-type rtfA and rtfA : : hyg. Clones 218R.1 and 218R.2 yielded no detectable PCR products, representing probable spontaneous deletion mutants in the genes encoding GPL biosynthesis (Belisle et al., 1993b).

Southern blot analysis was performed to confirm the genotype of the allelic exchange mutants (Fig. 3). Genomic DNA from strain 218R.4 was digested with HindIII (there are no HindIII sites in either rtfA or hyg) and hybridized for rtfA. Wild-type TMC724 yielded a 17·4 kb restriction fragment that hybridized to rtfA. In comparison, clone 218R.4 yielded an 18·7 kb hybridizing fragment corresponding to a 1·3 kb increase from the hyg insert.



View larger version (53K):
[in this window]
[in a new window]
 
Fig. 3. Southern blot analysis of allelic exchange mutants. Genomic DNA from wild-type TMC724 and clone 218R.4 was digested with HindIII and probed for rtfA. TMC724 (lane 1) yielded a 17·4 kb band, whereas clone 218R.4 (lane 2) yielded a 18·7 kb band, corresponding to the incorporation of the 1·3 kb hyg gene. Plasmid pVAP41 yielded a fragment of 16·4 kb, somewhat larger than the predicted 14·1 kb (data not shown).

 
RtfA encodes addition of Rha to 6-dTal and not to the alaninol of the LP core as confirmed by TLC and GC-MS
It was hypothesized that the creation of an insertional deletion mutant in rtfA of TMC724 via allelic exchange would prevent the addition of Rha as the proximal serovar-specific carbohydrate to 6-dTal within the GPL core (Eckstein et al., 1998). The rtfA knockout mutants should thus express only the nsGPL and be devoid of Rha and 2,3-O-Me-fucose (2,3-O-Fuc). As expected, TLC of alkaline stable lipids (Fig. 4) demonstrated that TMC724 expressed both ssGPL and nsGPL, whereas clones 218R.4 and 218R.5 expressed only nsGPL. For both mutant clones, the pattern of nsGPL expression was unchanged compared to the wild-type serovar 2 strain (Fig. 4, and data not shown).



View larger version (92K):
[in this window]
[in a new window]
 
Fig. 4. TLC. Alkaline-stable lipids were isolated from wild-type and mutant bacteria and resolved on silica plates in CHCl3/CH3OH/H2O (90 : 10 : 1 by vol.). Lane 1 contains M. avium TMC724 (serovar 2), lanes 2 and 3 contain {Delta}rtfA mutant 218R.4 and lane 4 contains strain 235R.1 (clone 218R.4 complemented with wild-type rtfA contained on integrative plasmid pVAP52). The arrow denotes serovar 2-specific GPL.

 
GC analysis of the alditol acetate derivatives yielded five peaks for TMC724 corresponding to 6-dTal, Rha, 2,3-O-Me-Fuc, 3-O-Me-Rha and 3,4-O-Me-Rha (Fig. 5). The spectrum of sugars from clone 218R.5 showed loss of the Rha peak and the distal 2,3-O-Me-Fuc and preservation of 3-O-Me-Rha and 3,4-O-Me-Rha. Relative to wild-type bacteria, however, the rtfA mutant displayed an increase in 3-O-Me-Rha. This confirms the conclusion of Eckstein et al. (1998) that rtfA is responsible for the addition of Rha to 6-dTal as the proximal sugar of serovar 2-specific GPL, but does not catalyse the addition of Rha to alaninol of the LP core.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 5. GC of alditol derivatives of GPLs of TMC724 and {Delta}rtfA mutant 218R.4. (a) Shows a Rha standard used to differentiate the closely migrating Rha and 6-dTal peaks. (b) Shows sugars associated with GPL of M. avium TMC724. (c) Shows sugars associated with GPL of 218R.4. The asterisks indicate peaks that were not derived from sugars as determined by their MS profile.

 
Finally, proof that disruption of rtfA resulted in loss of serovar 2-specific GPL expression was confirmed by complementation of rtfA in trans. Clone 218R.4 was transformed with the integrative plasmid pVAP52 containing wild-type rtfA. TLC of this clone (235R.1) showed recovery of the baseline pattern for GPL expression (Fig. 4).


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Site-directed mutagenesis via allelic replacement had been described for Mycobacterium intracellulare (Marklund et al., 1995), M. bovis BCG (Aldovini et al., 1993; Norman et al., 1995; Pelicic et al., 1996a; Wards et al., 2000), M. tuberculosis (Pelicic et al., 1997) and M. smegmatis (Husson et al., 1990), and has been extended to include M. avium in this report. By in situ replacement of rtfA with rtfA : : hyg, we were able to confirm that RtfA catalyses the addition of Rha only to 6-dTal as the first serovar-specific sugar but does not catalyse the addition of Rha to alaninol of the LP core of GPL. Thus, through the construction of mutants in GPL biosynthesis, to include rtfA, the role of GPL in the pathogenesis of M. avium can be studied in a consistent manner.

Homologous recombination techniques for mycobacteria have typically utilized replicative plasmids (Baulard et al., 1996) to allow for greater time for single crossover versus use of linear strands of non-replicative DNA (Balasubramanian et al., 1996). In fact, for M. intracellulare, homologous recombination has been shown to require an intermediate state created by a single crossover with subsequent elimination of plasmid DNA sequences by a double crossover state. Enrichment of allelic exchange mutants that have lost plasmid DNA is facilitated by the inclusion of selection and counter-selection markers allowed as part of the vectors (Pelicic et al., 1997).

Counter-selection markers used for the successful creation of allelic exchange mutants of M. tuberculosis and M. smegmatis employed a plasmid vector that incorporated a temperature-sensitive origin of replication (ts-OriM) and contained the sacB gene (Pelicic et al., 1996a, b, 1997). Unfortunately, we found that neither sacB nor the ts-OriM provided minimal if any selection for M. avium. Moreover, the lowered incubation temperatures required for the ts-OriM (Lee et al., 2002) precluded successful transformation strains with lower transformation efficiencies such as TMC724. In addition, while other genes such as rpsL have been used for counter-selection for mycobacteria (Sander et al., 1995), their use for M. avium also appears limited (D. Carroll, personal communication). Thus, other selection and counter-selection markers were investigated.

Exploitation of INH susceptibility, conferred by introduction of katG from a susceptible species of mycobacteria, as well as inclusion of gfp, allowed for a facile method to select for double exchange mutants that had undergone loss of plasmid DNA. The use of KatG as a counter-selection marker for M. avium was, however, shown to be strain-specific and required careful study to determine the optimal concentration of INH. We showed that, for M. avium strain TMC724, an INH concentration of 10 µg ml-1 would provide an approximate 1 log decrease in cell numbers between wild-type bacteria and strains transformed with katG contained on an episomal plasmid, whereas for M. avium strain 920A-6, INH at a concentration of 1 µg ml-1 yielded an 8 log decrease in cell numbers for transformed bacteria. Similar results were observed for the intrinsically more resistant smooth transparent morphotype of strain 920A-6. Our data would predict that once chromosomal integration of plasmid DNA occurred via an initial single crossover event [such as occurs for M. intracellulare (Marklund et al., 1997)], there would be no additional selection with KatG-mediated INH selection. Of note is that our data for serovar 8 strain 920A-6 corroborate studies that INH resistance of M. avium does not appear to result from a diffusion barrier across the cell wall (Mdluli et al., 1998). Unlike M. smegmatis, a single copy of katG did not confer INH susceptibility to M. avium, suggesting possible interference from the presence of the native M. avium katG gene.

The genetic basis of the biosynthesis of the M. avium serovar 2-specific GPL has been the subject of prior studies (Belisle et al., 1991; Eckstein et al., 1998; Mills et al., 1994). Oligosaccharide biosynthesis was initially localized to a 28 kb chromosomal region of TMC724 (Belisle et al., 1991) that was demonstrated to contain four functional domains, ser2A to ser2D (Mills et al., 1994). Transposon mutagenesis showed that the genes encoding the addition of serovar-specific oligosaccharides resided within ser2A, the region associated with rhamnosyltransferase activity being contained within a 4·5 kb region within this locus (Mills et al., 1994). Sequence analysis of ser2A demonstrated three ORFs with homology to mycobacterial glycosyltransferases (ORF1 and ORF2) and a putative methyltransferase (ORF3) (Eckstein et al., 1998). Recombinant expression of ORF1 and ORF2 in M. smegmatis demonstrated that ORF2 was able to catalyse the addition of Rha to 6-dTal of the LP core of GPL, consistent with ORF2 representing the M. avium rtfA.

While prior work (Eckstein et al., 1998) was able to demonstrate that ORF2 could catalyse the addition of Rha to the 6-dTal of the LP core, proof of this function in M. avium was yet to be proven. Through the construction of site-directed knockout in the M. avium rtfA gene and subsequent complementation with the wild-type gene, we were able to conclusively prove that RtfA catalyses the addition of Rha as the proximal sugar in the serovar-specific oligosaccharide. GC analysis of alditol acetate derivatives of GPL in the prior study (Eckstein et al., 1998) showed an increase in 2,3,4-O-Me-Rha relative for the rtfA transformant to wild-type M. smegmatis. This result suggested the unlikely possibility that rtfA could be bifunctional, catalysing both the addition of Rha to 6-dTal and the addition of Rha to alaninol of the LP core. We did not observe the same inversion of molar ratios of 6-dTal and 3,4-O-Me-Rha (2,3,4-O-Me-Rha in the prior study) showing that, as expected, RtfA catalyses only the addition of Rha to 6-dTal.

Using the techniques derived in this study, it is possible to consistently study the genetics of the biosynthesis of GPL and its role in the pathogenesis of M. avium. Disruption of rtfA represents an ideal first target in studies of pathogenesis since Rha represents the proximal sugar of the oligosaccharide moiety of ssGPL, independent of serovar. In this way, a major limitation of prior studies that have relied on comparisons of strains representing different serovars (Newman et al., 1991) or have chemically modified GPL to examine GPL subfractions (Barrow, 1991; Barrow et al., 1993; Brownback & Barrow, 1988; Tassell et al., 1992) can be avoided. Further, unresolved questions as to the function of other genes within the ser2A locus can be elucidated.

In summary, by site-directed mutagenesis, allelic exchange mutants were created in rtfA of serovar 2 strain TMC724. We were able to demonstrate conclusively that RtfA only catalyses transfer of Rha to 6-dTal of the LP core as the proximal serovar-specific sugar. Through the construction of this and additional mutants in GPL synthesis, the role of GPL in the host–pathogen interaction, immunogenesis and antibiotic resistance can be determined.


View this table:
[in this window]
[in a new window]
 
Table 2. Plasmids used in this study

 

   ACKNOWLEDGEMENTS
 
Supported by a Merit Review Grant from the Veterans Affairs (J. N. M., S.-H. L.), ROI AI32783 (J. N. M.) and RO1 AI41925 (T. M. E., J. M. I., J. T. B.). William Jacobs (Department of Immunology and Microbiology, Albert Einstein School of Medicine of Yeshiva University, Bronx, NY, USA) and Stoyan Bardarov are gratefully acknowledged for the gift of strains and plasmids. Thomas Glaze, Ansel Hsiao and William R. Howe are gratefully acknowledged for technical assistance. Presented in part at the 103rd General Meeting of the American Society for Microbiology, Washington, DC, May 18–22, 2003.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Aldovini, A., Husson, R. N. & Young, R. A. (1993). The uraA locus and homologous recombination in Mycobacterium bovis BCG. J Bacteriol 175, 7282–7289.[Abstract]

Arbeit, R. D., Slutsky, A., Barber, T. W., Maslow, J. N., Niemczyk, S., Falkinham, J. O., III, O'Conner, G. T. & von Reyn, C. F. (1993). Genetic diversity among strains of Mycobacterium avium causing monoclonal and polyclonal bacteremia in patients with AIDS. J Infect Dis 167, 1384–1390.[Medline]

Aspinall, G. O., Chatterjee, D. & Brennan, P. J. (1995). The variable surface glycolipids of mycobacteria: structures, synthesis of epitopes, and biological properties. Adv Carbohydr Chem Biochem 51, 169–242.[Medline]

Balasubramanian, V., Pavelka, M. S., Jr, Bardarov, S. S., Martin, J., Weisbrod, T. R., McAdam, R. A., Bloom, B. R. & Jacobs, W. R., Jr (1996). Allelic exchange in Mycobacterium tuberculosis with long linear recombination substrates. J Bacteriol 178, 273–279.[Abstract]

Barrow, W. W. (1991). Contributing factors of pathogenesis in the Mycobacterium avium complex. Res Microbiol 142, 427–433.[CrossRef][Medline]

Barrow, W. W., Carvalho de Sousa, J. P., Davis, T. L., Wright, E. L., Bachelet, M. & Rastogi, N. (1993). Immunomodulation of human peripheral blood mononuclear cell functions by defined lipid fractions of Mycobacterium avium. Infect Immun 61, 5286–5293.[Abstract]

Barrow, W. W., Davis, T. L., Wright, E. L., Labrousse, V., Bachelet, M. & Rastogi, N. (1995). Immunomodulatory spectrum of lipids associated with Mycobacterium avium serovar 8. Infect Immun 63, 126–133.[Abstract]

Baulard, A., Kremer, L. & Locht, C. (1996). Efficient homologous recombination in fast-growing and slow-growing mycobacteria. J Bacteriol 178, 3091–3098.[Abstract]

Belisle, J. T., Pascopella, L., Inamine, J. M., Brennan, P. J. & Jacobs, W. R., Jr (1991). Isolation and expression of a gene cluster responsible for biosynthesis of the glycopeptidolipid antigens of Mycobacterium avium. J Bacteriol 173, 6991–6997.[Medline]

Belisle, J. T., McNeil, M. R., Chatterjee, D., Inamine, J. M. & Brennan, P. J. (1993a). Expression of the core lipopeptide of the glycopeptidolipid surface antigens in rough mutants of Mycobacterium avium. J Biol Chem 268, 10510–10516.[Abstract/Free Full Text]

Belisle, J. T., Klaczkiewicz, K., Brennan, P. J., Jacobs, W. R., Jr & Inamine, J. M. (1993b). Rough morphological variants of Mycobacterium avium. Characterization of genomic deletions resulting in the loss of glycopeptidolipid expression. J Biol Chem 268, 10517–10523.[Abstract/Free Full Text]

Brennan, P. J. (1988). Mycobacterium and other actinomycetes. In Microbial Lipids, pp. 203–298. Edited by C. Ratledge & S. G. Wilkinson. London: Academic Press.

Brooks, R. W., Parker, B. C., Gruft, H. & Falkinham, J. O., III (1984). Epidemiology of infection by nontuberculous mycobacteria. V. Numbers in eastern United States soils and correlation with soil characteristics. Am Rev Respir Dis 130, 630–633.[Medline]

Brownback, P. E. & Barrow, W. W. (1988). Modified lymphocyte response to mitogens after intraperitoneal injection of glycopeptidolipid antigens from Mycobacterium avium complex. Infect Immun 56, 1044–1050.[Medline]

Burman, W. J., Stone, B. L., Reitmeijer, C. A., Maslow, J., Cohn, D. L. & Reves, R. R. (1998). Long-term outcomes of treatment of Mycobacterium avium complex bacteremia using a clarithromycin-containing regimen. AIDS 12, 1309–1315.[CrossRef][Medline]

Draper, P. & Rees, R. J. W. (1970). Electron-transparent zone of mycobacteria may be a defence mechanism. Nature 228, 860–861.[Medline]

Draper, P. & Rees, R. J. W. (1973). The nature of the electron-transparent zone that surrounds Mycobacterium lepraemurium inside host cells. J Gen Microbiol 77, 79–87.[Medline]

Eckstein, T. M., Silbaq, F. S., Chatterjee, D., Kelly, N. J., Brennan, P. J. & Belisle, J. T. (1998). Identification and recombinant expression of a Mycobacterium avium rhamnosyltransferase gene (rtfA) involved in glycopeptidolipid biosynthesis. J Bacteriol 180, 5567–5573.[Abstract/Free Full Text]

Gan, H., Newman, G., McCarthy, P. L. & Remold, H. G. (1993). TNF-{alpha} response of human monocyte-derived macrophages to Mycobacterium avium, serovar 4, is of brief duration and protein kinase C dependent. J Immunol 150, 2892–2900.[Abstract/Free Full Text]

George, K. L., Parker, B. C., Gruft, H. & Falkinham, J. O., III (1980). Epidemiology of infection by nontuberculous mycobacteria. II. Growth and survival in natural waters. Am Rev Respir Dis 122, 89–94.[Medline]

Havlir, D. V., Dubé, M. P., Sattler, F. R. & 9 other authors (1996). Prophylaxis against disseminated Mycobacterium avium complex with weekly azithromycin, daily rifabutin, or both. California Collaborative Treatment Group. N Engl J Med 335, 392–398.[Abstract/Free Full Text]

Hawkins, C. C., Gold, J. W., Whimbey, E., Kiehn, T. E., Brannon, P., Cammarata, R., Brown, A. E. & Armstrong, D. (1986). Mycobacterium avium complex infections in patients with the acquired immunodeficiency syndrome. Ann Intern Med 105, 184–188.[Medline]

Horgan, L., Barrow, E. L. W., Barrow, W. W. & Rastogi, N. (2000). Exposure of human peripheral blood mononuclear cells to total lipids and serovar-specific glycopeptidolipids from Mycobacterium avium serovars 4 and 8 results in inhibition of TH1-type responses. Microb Pathog 29, 9–16.[CrossRef][Medline]

Horsburgh, C. R., Jr (1991). Mycobacterium avium complex infection in the acquired immunodeficiency syndrome. N Engl J Med 324, 1332–1338.[Medline]

Howard, N. S., Gomez, J. E., Ko, C. & Bishai, W. R. (1995). Color selection with a hygromycin-resistance-based Escherichia coli–mycobacterial shuttle vector. Gene 166, 181–182.[CrossRef][Medline]

Husson, R. N., James, B. E. & Young, R. A. (1990). Gene replacement and expression of foreign DNA in mycobacteria. J Bacteriol 172, 519–524.[Medline]

Lee, S. H., Cheung, M., Irani, V., Carroll, J. D., Inamine, J. M., Howe, W. R. & Maslow, J. N. (2002). Optimization of electroporation conditions for Mycobacterium avium. Tuberculosis 82, 167–174.[CrossRef][Medline]

Marklund, B.-I., Speert, D. P. & Stokes, R. W. (1995). Gene replacement through homologous recombination in Mycobacterium intracellulare. J Bacteriol 177, 6100–6105.[Abstract]

Marklund, B.-I., Mahenthiraralingam, E. & Stokes, R. W. (1997). Site-directed mutagenesis and virulence assessment of the M. intracellulare katG gene. In Abstracts of the 97th General Meeting of the American Society for Microbiology, abstract U-95. Washington, DC: American Society for Microbiology.

McNeil, M., Tsang, A. Y. & Brennan, P. J. (1987). Structure and antigenicity of the specific oligosaccharide hapten from the glycopeptidolipid antigen of Mycobacterium avium serotype 4, the dominant mycobacterium isolated from patients with acquired immune deficiency syndrome. J Biol Chem 262, 2630–2635.[Abstract/Free Full Text]

Mdluli, K., Swanson, J., Fischer, L., Lee, R. E. & Barry, C. E., 3rd (1998). Mechanisms involved in the intrinsic isoniazid resistance of Mycobacterium avium. Mol Microbiol 27, 1223–1233.[CrossRef][Medline]

Mills, J. A., McNeil, M. R., Belisle, J. T., Jacobs, W. R., Jr & Brennan, P. J. (1994). Loci of Mycobacterium avium ser2 gene cluster and their functions. J Bacteriol 176, 4803–4808.[Abstract]

Minami, H. (1998). Promotion of phagocytosis and prevention of phagosome–lysosome (P–L) fusion in human peripheral blood monocytes by serotype specific glycopeptidolipid (GPL) antigen of Mycobacterium avium complex (MAC). Kekkaku 73, 545–556 (in Japanese).[Medline]

Newman, G. W., Gan, H. X., McCarthy, P. L., Jr & Remold, H. G. (1991). Survival of human macrophages infected with Mycobacterium aviumintracellulare correlates with increased production of tumor necrosis factor-{alpha} and IL-6. J Immunol 147, 3942–3948.[Abstract/Free Full Text]

Nightingale, S. D., Byrd, L. T., Southern, P. M., Jockusch, J. D., Cal, S. X. & Wynne, B. A. (1992). Incidence of Mycobacterium aviumintracellulare complex bacteremia in human immunodeficiency virus-positive patients. J Infect Dis 165, 1082–1085.[Medline]

Norman, E., Dellagostin, O. A., McFadden, J. & Dale, J. W. (1995). Gene replacement by homologous recombination in Mycobacterium bovis BCG. Mol Microbiol 16, 755–760.[Medline]

Pelicic, V., Reyrat, J. M. & Gicquel, B. (1996a). Positive selection of allelic exchange mutants in Mycobacterium bovis BCG. FEMS Microbiol Lett 144, 161–166.[CrossRef][Medline]

Pelicic, V., Reyrat, J.-M. & Gicquel, B. (1996b). Expression of the Bacillus subtilis sacB gene confers sucrose sensitivity on mycobacteria. J Bacteriol 178, 1197–1199.[Abstract]

Pelicic, V., Jackson, M., Reyrat, J. M., Jacobs, W. R., Jr, Gicquel, B. & Guilhot, C. (1997). Efficient allelic exchange and transposon mutagenesis in Mycobacterium tuberculosis. Proc Natl Acad Sci U S A 94, 10955–10960.[Abstract/Free Full Text]

Pierce, M., Crampton, S., Henry, D. & 10 other authors (1996). A randomized trial of clarithromycin as prophylaxis against disseminated Mycobacterium avium complex infection in patients with advanced acquired immunodeficiency syndrome. N Engl J Med 335, 384–391.[Abstract/Free Full Text]

Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.

Sander, P., Meier, A. & Böttger, E. C. (1995). rpsL+: a dominant selectable marker for gene replacement in mycobacteria. Mol Microbiol 16, 991–1000.[Medline]

Shafran, S. D., Singer, J., Zarowny, D. P. & 9 other authors (1996). A comparison of two regimens for the treatment of Mycobacterium avium complex bacteremia in AIDS: rifabutin, ethambutol, and clarithromycin versus rifampin, ethambutol, clofazimine, and ciprofloxacin. Canadian HIV Trials Network Protocol 010 Study Group. N Engl J Med 335, 377–383.[Abstract/Free Full Text]

Snapper, S. B., Melton, R. E., Mustafa, S., Kieser, T. & Jacobs, W. R., Jr (1990). Isolation and characterization of efficient plasmid transformation mutants of Mycobacterium smegmatis. Mol Microbiol 4, 1911–1919.[Medline]

Tassell, S. K., Pourshafie, M., Wright, E. L., Richmond, M. G. & Barrow, W. W. (1992). Modified lymphocyte response to mitogens induced by the lipopeptide fragment derived from Mycobacterium avium serovar-specific glycopeptidolipids. Infect Immun 60, 706–711.[Abstract]

von Reyn, C. F., Waddell, R. D., Eaton, T. & 10 other authors (1993). Isolation of Mycobacterium avium complex from water in the United States, Finland, Zaire, and Kenya. J Clin Microbiol 31, 3227–3230.[Abstract]

von Reyn, C. F., Maslow, J. N., Barber, T. W., Falkinham, J. O., III & Arbeit, R. D. (1994). Persistent colonization of potable water as a source of Mycobacterium avium infection in AIDS. Lancet 343, 1137–1141.[Medline]

Wards, B. J., De Lisle, G. W. & Collins, D. M. (2000). An esat6 knockout mutant of Mycobacterium bovis produced by homologous recombination will contribute to the development of a live tuberculosis vaccine. Tuber Lung Dis 80, 185–189.[CrossRef][Medline]

Zhang, Y., Heym, B., Allen, B., Young, D. & Cole, S. (1992). The catalase-peroxidase gene and isoniazid resistance of Mycobacterium tuberculosis. Nature 358, 591–593.[CrossRef][Medline]

Received 12 June 2003; revised 8 August 2003; accepted 15 August 2003.