| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Center for Pulmonary and Infectious Disease Control, University of Texas Health Center at Tyler, Tyler, TX 75708, USA
2 C4 101 Veterinary Medical Center, Cornell University, Ithaca, NY 14850, USA
3 Department of Microbiology and Molecular Genetics, Harvard Medical School, Boston MA 02115, USA
4 The University of New Mexico School of Medicine, Albuquerque, NM87108, USA
5 Department of Medicine, Albuquerque Veterans Affairs Medical Center, 1501 San Pedro, SE, Albuquerque, NM 87108, USA
Correspondence
Thomas F. Byrd
tbyrd{at}salud.unm.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Present address: Department of Oncological Sciences, Mount Sinai School of Medicine, NY 10029, USA.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). M. abscessus has also been identified as an emerging pulmonary pathogen in patients with cystic fibrosis (Cullen et al., 2000; Fauroux et al., 1997; Olivier et al., 2003; Sermet-Gaudelus et al., 2003), including those who have undergone lung transplantation (Sanguinetti et al., 2001). Two important clinical questions are whether isolation of M. abscessus, or other non-tuberculous mycobacteria (NTM), from the sputum of a patient represents colonization or invasive pulmonary infection, and whether lung colonization inevitably leads to the development of invasive pulmonary infection (Brown-Elliott & Wallace, 2005; Cullen et al., 2000).
NTM have long been recognized as having rough and smooth colony phenotypes (Fregnan & Smith, 1962). Studies have shown that colony morphology in NTM is influenced by cell wall glycopeptidolipid (GPL) (Barrow & Brennan, 1982; Eckstein et al., 2000). In Mycobacterium smegmatis and Mycobacterium avium GPLs are associated with sliding motility and the ability to form biofilms (Carter et al., 2003; Recht et al., 2000). Both of these characteristics are thought to play a role in bacterial colonization (Recht et al., 2000). Several reports have found a correlation between colony morphology and virulence, with rough variants generally being more virulent than smooth variants (Belisle & Brennan, 1989; Collins & Cunningham, 1981; Schaefer et al., 1970). These reports are consistent with results from our previous studies in which we reported that a rough M. abscessus clinical isolate became attenuated after it spontaneously dissociated into a smooth variant (Byrd & Lyons, 1999). During serial passage of the smooth variant, one colony spontaneously reverted back to the rough phenotype affording a unique opportunity to examine the role of GPL in colony phenotype, and the influence of colony phenotype on biofilm formation and the ability of M. abscessus to cause persistent, invasive infection.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Howard et al., 2002). After several passages of 390S on 7H10-OADC plates, a rough colony revertant was isolated and inoculated into broth. Aliquots of this new rough variant, which was named 390V, were stored at –70 °C. Additional M. abscessus strains were obtained from P. Conville, Public Health Service, NIH, Bethesda, MD, USA, and had the following morphologies: strains 6639 and 8988, smooth; strains 1056 and 1475, rough; strain 8243, intermediate. M. smegmatis strain mc2155 was used as a control for GPL expression.
Colony morphology and growth pattern.
Colony morphology was assessed on 7H11 plates. Examination for cording was performed by inoculating bacteria into 7H9 broth at a concentration of 1x104 bacteria ml–1 and culturing them at 37 °C under stationary phase conditions for 7 days. Culture suspensions were spun onto glass slides using a cytocentrifuge (Shandon), and the bacteria were heat fixed and stained with crystal violet dye. Bacteria were examined at x400 magnification using a light microscope (Nikon) and photographed.
Genetic analysis of bacterial strains.
M. abscessus 390S is lacking 14.2 kbp of DNA that is present in the original 390R strain (Howard et al., 2002). We used this genomic deletion to verify the identity of our 390V strain. For genomic DNA preparation and Southern blotting, bacteria were scraped from plate cultures and lysed by bead beating using 0.1 mm silica/zirconium beads in a BioSpec Mini-BeadBeater. DNA was extracted using the PUREGENE DNA isolation kit for Gram-positive bacteria (Gentra Systems), and then recovered following phenol/chloroform extraction and 2-propanol precipitation. Southern blots were prepared and probed (Sambrook et al., 1989) following the electrophoresis of 2 µg genomic DNA digested with the indicated restriction enzymes.
To further confirm the identity of the 390V strain, PCR products from the deletion region were compared among the 390 strains. For analysis of the deletion region, five primers were designed, based on the published sequence (Howard et al., 2002). R5del5 (5'-GTCCTCGTAGAAGTACCGATC-3') is located within ORF5, 220 bp upstream of the start of the deletion. R18del3 (5'-GTTCGTCGGAATGGCAC-3') is located within ORF18, approximately 240 bp downstream of the deletion. R6M3 (5'-GTTCGAGCAACGTGCTG-3') is located within the deletion, in ORF6. MabR3int (5'-CCAGCGACGGAGTATCTC-3') and MabR3-5end (5'-GCCAGACTGTAGGACATGA-3') are located in ORF3, outside of the deletion. Reactions were performed with 20 ng M. abscessus DNA, 1 pmol µl–1 appropriate primers and Promega PCR Master Mix, with an initial denaturation followed by 30 cycles of 95 °C/30 s, 55 °C/30 s, 72 °C/30 s, and a final 7 min extension at 72 °C.
Reversion rate.
To determine the reversion rate of the 390S strain from a smooth to a rough phenotype, a single 3-day-old 390S colony was picked from a 7H11 agar plate and inoculated into 5.0 ml 7H9 broth in a small shaker flask. The culture was incubated at 37 °C at 100 r.p.m. on a shaker incubator and harvested during early exponential phase growth. The entire culture volume was divided and plated on a total of 50 Petri dishes (150x15 mm) containing 7H11 agar, and cultured for 3 days at 37 °C. Plates were then examined using a dissecting microscope, and the total number of c.f.u. and rough revertants counted. Rough revertants that were discrete colonies, not bordering or touching adjacent colonies, were graded as definite revertants. Those rough revertants that were small and touching larger smooth colonies were considered possible revertants since they may have arisen secondarily out of an adjoining smooth colony. An estimate of the reversion rate was arrived at as a range with only definite revertants representing the upper limit, and both definite and possible revertants representing the lower limit of the range.
Sliding motility.
Strain motility was assessed by inoculating bacterial cells into the centre of plates prepared with M63 minimal media containing 0.3 % agarose (Recht et al., 2000). Plates were incubated in a humidified incubator at 37 °C and colony spread was assessed after 3 days.
Biofilm formation.
Bacteria in Sauton's medium were cultured in a Calgary biofilm device (Ceri et al., 1999). At the indicated time points, pegs were removed from the lid of the device and placed into microcentrifuge tubes containing sterile PBS. In addition, Sauton's medium was removed from the remaining wells and replaced with fresh medium. Bacteria were dislodged from pegs by sonication for 20 s using a cup horn sonicator, and supernatants were plated on 7H11 agar to determine c.f.u. In some experiments, bacteria on pegs attached to lids were fixed for 30 min with methanol, followed by staining with 1 % crystal violet dye for 1 h. Pegs were photographed using an inverted phase-contrast microscope and then destained with 100 % ethanol. The A570 value of the crystal violet/ethanol solution was then determined (Recht et al., 2000).
Lipid extraction and analysis.
To characterize lipid profiles from the 390R, 390S and 390V M. abscessus variants, lipids from plate-grown cultures were extracted with CHCl3/CH3OH [2 : 1, v/v; 10 ml (g wet weight bacteria)–1]. Bacteria were extracted twice at 56 °C (15 min with sonication) and once at 4 °C overnight, and insoluble material was removed by centrifugation and filtration (0.20 µm PFTE filter; Millipore). Combined extracts were subjected to biphasic partitioning in CHCl3/CH3OH/H2O (4 : 2 : 1, by vol.). Lipids in the organic phase were stored at –20 °C. GPLs are resistant to mild alkaline methanolysis, which destroys nonspecific acylglycerols (McNeil et al., 1989), and alkali treatment was performed according to the method of Brennan & Goren (1979). In brief, total lipid extracts (6 mg ml–1) were treated with an equal volume NaOH (0.2 M in CH3OH, 45 min at 37 °C), neutralized with glacial acetic acid, and the alkali-stable lipids subjected to biphasic partitioning.
Total lipid extracts or alkali-stable lipids (which consist primarily of GPLs) were resuspended at 10 µg µl–1 in CHCl3/CH3OH and spotted onto aluminium-backed silica gel-60 TLC plates (EM Science). Lipids were resolved in CHCl3/CH3OH/H2O (100 : 14 : 0.8 or 65 : 25 : 4, by vol.). Plates were sprayed with 10 % CuSO4 in 8 % phosphoric acid or H2SO4 (50 % in CH3OH) to visualize all lipids. To detect glycolipids with 6-deoxyhexoses plates were sprayed with 1-naphthol (Sigma; 3 % with 10 % H2SO4 in CH3OH). Plates were charred with a heat gun until spots with hues characteristic of the different lipid classes appeared (Rhoades et al., 2003). GPLs charred to a golden-yellow colour with H2SO4 or to an orange-pink colour with 1-naphthol/H2SO4. Plates were also sprayed with orcinol (0.5 % with 15 % trichloroacetic acid in water-saturated n-butanol) to detect aldehydes, but there were no differences from the colours of the spots that had been sprayed with H2SO4 (data not shown).
Infection of human monocytes.
A previously developed assay for growth of M. abscessus variants in human monocytes was used (Byrd & Lyons, 1999) with slight modification. Human monocyte monolayers in Iscove's medium/1 % normal human serum were infected with M. abscessus strains for 90 min, followed by three washes with Iscove's medium/1 % normal human serum. A time zero count of monocyte-associated c.f.u. was plated and then 60 µg amikacin ml–1 was added to the tissue culture media to kill any remaining extracellular bacteria. After 48 h M. abscessus c.f.u. from cell lysates were plated and the remaining wells washed twice with Iscove's medium/1 % normal human serum to remove amikacin, followed by readdition of medium to the wells. At 72 h c.f.u. from cell lysates were plated again. Middlebrook 7H9 broth (Difco) was used for dilution of cell lysates prior to plating for c.f.u. Middlebrook 7H11 agar (Difco) plates (100x15 mm bacteriological Petri dishes) were used for plating c.f.u. from infected monolayers. In addition, nuclear counts in replicate wells were determined, and c.f.u. standardized to 105 nuclei to account for any differences in monocyte number in the different monolayers, as has been described previously (Byrd, 1997). Viability was analysed in replicate infected wells by Trypan blue exclusion at each time point to ensure that the variants did not have a differential effect on monocyte viability independent of bacterial growth, as has been described previously (Byrd, 1997).
Murine pulmonary infection assay.
M. abscessus variants were used to infect SCID mice as previously described (Byrd & Lyons, 1999), with the exception that 50 µl fluid drops containing a bacterial suspension (105 c.f.u.) of M. abscessus 390R, 390S or 390V were inoculated intranasally rather than intratracheally. Immediately after infection, and at various time points thereafter, mice were sacrificed, and the lungs and spleens removed and placed in 3.0 ml PBS. The organs were homogenized and dilutions plated in triplicate on Middlebrook 7H11 agar.
The experimental protocol used in these studies was approved by Institutional Animal Care and Use Committee of the University of New Mexico Health Sciences Center.
Statistics.
Data were compared using Student's t-test. Data were considered significant with a P<0.05.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). Although the phenotypes of rough and smooth variants were stable over multiple passages, we obtained a single rough colony following routine plating of the smooth variant, and cultured it for further analysis. The rough revertant, named 390V, has a colony morphology similar to that of the original 390R isolate, but it has a drier, more wrinkled appearance (Fig. 1a–c). All 390 strains had identical growth rates in liquid media (data not shown).
|
). Although this region is intact in the parental 390R strain (Fig. 2a) (Howard et al., 2002), our previous study showed that some rough clinical isolates have extensive deletions within this region, indicating that the deletion in 390S was unlikely to be responsible for the change to a smooth colony morphology. However, the R/S-D1 deletion did provide a suitable target for confirming that the 390V rough revertant was derived from 390S. Therefore, we hybridized a Southern blot with a DNA probe (Fig. 2a) that overlaps one end of the deletion, and found that 390V and 390S shared the same band pattern (Fig. 2b, lanes 2, 3, 5, 6), which differed from the 390R band pattern (Fig. 2b, lanes 1, 4).
|
). 390V and 390S showed the same PCR amplification pattern with primer sets. In particular, amplification using primers 1 and 3, which anneal to regions flanking opposite sides of the deletion, yielded a 460 bp product with both 390V and 390S, but not with 390R because the 14 kbp region between the two primers would not be amplified under the conditions used (Fig. 2c, lanes 5–8). These data show that 390V is likely to be a rough revertant from 390S. Examining the spontaneous reversion of 390S to a rough phenotype under controlled conditions indicated a reversion rate in the range of 1 in 105 to 1 in 106. This lends further support to the evidence that our rough variant 390V arose from 390S.
M. abscessus 390R and the 390V revertant exhibit cord formation while the 390S strain does not
Mycobacterium tuberculosis and several NTM grow in what have been described as ropes, bundles or serpentine cords of acid-fast bacilli in liquid media or on agar plates (Middlebrook et al., 1947; Lorian, 1966). A particular rough colony phenotype is associated with cord formation in NTM. This type of colony is characterized microscopically by numerous irregular parallel filaments that form ridges and grooves throughout the colony (Fregnan & Smith, 1962). A comparison of the colony morphology of our M. abscessus strains to that described in these studies indicates that both 390R and 390V exhibit the rough colony phenotype associated with cording (Fig. 1d–f). Importantly, both M. abscessus 390R and M. abscessus 390V also form cords when grown in broth, but M. abscessus 390S lacks this ability (Fig. 1g–i).
GPL is abundantly expressed in the M. abscessus 390S strain but is minimal in both the 390R and 390V revertants
Since smooth and rough colony morphologies are associated with the presence and absence, respectively, of GPLs in M. avium and M. smegmatis (Barrow & Brennan, 1982; Billman-Jacobe et al., 1999; Eckstein et al., 2000), we examined whether GPL was expressed by our variants using TLC.
Preliminary TLC results obtained using the same methods used to study M. smegmatis GPLs (Recht & Kolter, 2001) indicated that 390R, 390V and other rough M. abscessus variants (strains 1056 and 1475) lacked lipid bands that were present in 390S and other smooth variants (strains 6639 and 8988), as well as a variant with an intermediate phenotype (strain 8243). These bands aligned with GPLs from M. smegmatis (data not shown), suggesting that only the smooth M. abscessus strains expressed GPLs. To further define the cell wall lipids of our 390R, 390V and 390S M. abscessus variants, lipid extracts were compared by TLC after spraying with sulfuric acid. The total lipid profiles of the 390R and 390V strains were very similar, containing trehalose mycolates, phosphatidylinositol mannosides and other phospholipids; these lipids were classified based on their mobility and characteristic hues, purple, brown-orange and yellow, respectively, as compared to extracts from Mycobacterium bovis BCG and M. tuberculosis (data not shown) (Rhoades et al., 2003). The proportions of the lipids in the 390R and 390V extracts were also similar except for the most polar of the trehalose dimycolates (Fig. 3a, b, asterisk), which was more abundant in the 390R extract. These classes of lipids were also present in the 390S extract except for one of the trehalose monomycolates (Fig. 3b, small arrow) and an unidentified wax (Fig. 3b, large arrowhead). Notably, the 390S extract contained numerous additional lipids that charred to a gold-yellow hue with sulfuric acid (Fig. 3b). These lipids exhibited RF values similar to or greater than those of the trehalose mycolates (Fig. 3b, square bracket). The colour and mobility of the extra lipids are similar to those of apolar GPLs of M. avium (McNeil et al., 1989), and it has been reported that M. abscessus expresses GPLs (Lopez-Marin et al., 1994). To provide further evidence that these lipids are M. abscessus GPLs, the plates were sprayed with 1-naphthol to detect deoxyhexoses found in GPLs, and extracts were also subjected to mild alkaline methanolysis to which GPLs are resistant (McNeil et al., 1989; Torrelles et al., 2002). 1-naphthol turned the extra lipids of the 390S extract to a characteristic orange-pink (Fig. 3c), and the only other lipids detected were the trehalose mycolates, which have a different deoxyhexose that charred to a lavender hue. Upon mild alkaline methanolysis most lipids, including the phospholipids and trehalose mycolates, were degraded; however, the extra lipids in the 390S extract remained (Fig. 3c). The shift in mobility of the saponified lipids was likely due to partial deacetylation caused by the methanolysis treatment. Interestingly, lipids that aligned with these were also present at very low proportions in alkali-treated 390R and 390V extracts, indicating that GPLs are made by all three strains. These results show that, in addition to trehalose glycolipids and phospholipids expressed by all three strains, the 390S strain expresses a significantly higher proportion of GPLs. As 390R was the parental strain to 390S, and 390V was derived from 390S, our results indicate that high level GPL expression is a reversible phenotype.
|
) and form biofilms is due to the presence of GPL (Recht et al., 2000; Recht & Kolter, 2001). Smooth and intermediate strains of M. abscessus, including 390S, were motile and formed a thin film of growth spreading 1–2 cm from the centre of the plate, while the rough strains were non-motile (data not shown). In addition, the 390S variant had the capability to adhere to and grow on the surface of polystyrene pegs in the Calgary biofilm device, while both 390R and 390V lacked this ability (Fig. 4). These results indicate that 390S is able to form biofilms and is consistent with our finding of GPL expression by this variant.
|
), 390S declined in number over 3 days, whereas 390R caused persistent, progressive infection. Importantly, 390V showed a similar pattern as 390R, with overlapping numbers of c.f.u. in monocytes at day 3 (Fig. 5). These results strongly suggest that the ability of 390R and 390V to persistently infect human monocytes is a consequence of their rough surface phenotype.
|
) was regained by the 390V revertant we infected SCID mice and examined the growth of M. abscessus strains in the lungs over a 4-week period. Comparison of the 390S strain to the 390V strain shows that the revertant causes persistent, invasive infection (Fig. 6). At 28 days, no c.f.u. were recovered from the lungs of any of the four mice infected with the 390S strain. In contrast, at this time point, c.f.u. were recovered from the lungs of each of the four mice infected with either the 390R or the 390V strain. Consistent with the ability of 390V to cause invasive disease was the finding that both 390R (2 out of 4 mice, mean=310 c.f.u.) and 390V (3 out of 4 mice, mean=140 c.f.u.) were found in the spleens of mice at 21 days. In contrast, no c.f.u. from the 390S strain were found in the spleens of any mice throughout the course of infection.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
The results of our study are consistent with studies of NTM that have compared rough and smooth colony variants, and found that rough variants tend to be more virulent in experimental infection models. It has been reported that rough forms of Mycobacterium kansasii persist longer than smooth variants in experimentally infected mice (Belisle & Brennan, 1989; Collins & Cunningham, 1981). The relationship between virulence and colony phenotype is more complicated in M. avium strains. Rough variants of M. avium are generally pathogenic for mice and chickens, whereas smooth opaque variants are generally non-pathogenic (Schaefer et al., 1970). However, a smooth transparent colony phenotype is associated with increased virulence (Reddy et al., 1994), and a rough transparent colony phenotype with even greater virulence (Kansal et al., 1998). Although our M. abscessus 390S variant lacks the ability to cause invasive infection it exhibits both sliding motility and biofilm formation, which are postulated to play a role in surface colonization (Martinez et al., 1999; Recht et al., 2000). Progressive infection by NTM is often preceded by colonization of various anatomic sites. In addition, colonization of medical devices by NTM can lead to subsequent serious infection, for example catheter-related bacteraemia (Brown-Elliott & Wallace, 2002). Both sliding motility and biofilm formation are dependent on GPL expression (Recht et al., 2000; Recht & Kolter, 2001). Consistent with these observations, our results demonstrate abundant expression of GPL by the 390S variant but minimal expression by the original 390R isolate and the 390V revertant. It is thus likely that GPL is responsible for the sliding motility and biofilm formation that we have observed in the 390S variant.
GPLs are found in the outermost portion of the mycobacterial cell envelope of NTM (Barrow et al., 1980; Furuchi & Tokunaga, 1972; Goren et al., 1972; Ortalo-Magne et al., 1996) and contain antigenic determinants for a number of mycobacterial species (Brennan & Goren, 1979; Camphausen et al., 1985; Lopez-Marin et al., 1994). The GPL molecule typically consists of a tripeptide-amino alcohol core with an amide linked long chain fatty acid. This lipopeptide core is substituted with 6-deoxytalose and O-methylated rhamnose to generate the non-specific core GPLs found in many species of NTM (Brennan & Goren, 1979). These non-specific GPLs are further modified by the addition of oligosaccharides to produce antigenic serovar specific GPLs in some mycobacterial species (Brennan & Nikaido, 1995; Lopez-Marin et al., 1994).
M. abscessus has been found to possess five major GPLs, which are also found in Mycobacterium chelonae and exhibit cross-reactivity with M. chelonae antiserum (Lopez-Marin et al., 1994). The differences between these five groups are due to differences in the location and/or number of the acetyl and sugar moieties. Only two M. abscessus isolates were analysed in the report by Lopez-Marin et al. (1994) and it was not noted whether these isolates had a rough or a smooth phenotype. Based on our results it is likely that these isolates expressed the smooth phenotype.
Phenotypic change associated with GPL expression has been described previously. Spontaneously occurring M. avium mutants lacking GPL were identified by a change from a smooth to a rough colony phenotype. Loss of M. avium GPL can occur as a result of deletion of large genomic regions encoding GPL synthesis proteins (Eckstein et al., 2000). A gene encoding a mycobacterial peptide synthetase designated mps is required for assembly of the lipopeptide core of GPL in M. smegmatis (Billman-Jacobe et al., 1999). We have identified an mps gene homologue in our three M. abscessus variants that has strong similarity to the mps gene of M. smegmatis and M. avium (unpublished data). This suggests that all three mycobacterial species utilize similar mechanisms for GPL synthesis. However, whereas deletions were responsible for observed differences in M. avium colony morphotypes (Eckstein et al., 2000), the ability of the M. abscessus 390 isolate to bidirectionally change colony phenotype suggests a reversible mechanism, rather than a deletion of genes involved in GPL synthesis. In addition, the expression of small amounts of GPL by our rough variants also argues against deletion.
In contrast to the M. abscessus 390S variant, both 390R and 390V variants form cords in broth medium that are morphologically identical to those of M. tuberculosis (Attorri et al., 2000). The mechanism for M. abscessus cording likely involves trehalose 6,6' dimycolate (TDM). Our TLC results suggest that TDM is present in each of our three variants; however, further compositional analysis of these bands is under way to determine their chemical structure. Recent studies indicate that the fine chemical structure of TDM determines the cording phenotype, and that TDM molecules with cording capability enhance the virulence of M. tuberculosis (Glickman et al., 2000; Rao et al., 2005). Thus, cord formation by M. abscessus may in part be responsible for the ability of the 390R and 390V variants to persist and cause invasive infection. The lack of cording by the 390S variant may be due to localization of GPL to the outermost portion of the M. abscessus cell wall preventing the interaction of TDM molecules from contiguous bacteria necessary for cording to occur. A similar hypothesis has been proposed to account for the difference in virulence between smooth and rough strains of M. kansasii (Belisle & Brennan, 1989). Alternatively the attenuation of the 390S variant could in some way be related to our finding that one of the trehalose monomycolates and an unidentified wax are absent from the 390S variant, but present in both the 390R and 390V variants.
M. abscessus has emerged as an important cause of infection caused by NTM (Cullen et al., 2000; Fauroux et al., 1997; Griffith et al., 1993; Howard & Byrd, 2000; Olivier et al., 2003; Sanguinetti et al., 2001; Sermet-Gaudelus et al., 2003). Since GPL expression by NTM has been postulated to play a role in environmental colonization (Recht et al., 2000), and cord formation is required for mycobacterial persistence in infected hosts (Glickman et al., 2000; Rao et al., 2005), the ability to switch phenotypes may allow M. abscessus to transition between a colonizing organism and an invasive human pathogen. There have been no studies correlating M. abscessus phenotype with clinical outcome. However, one report of a severe M. abscessus pulmonary infection in a cystic fibrosis patient who had received a lung transplant documented that the infecting organism had a rough colony phenotype (Sanguinetti et al., 2001).
The ability to transition between smooth and rough colony phenotypes could have particular relevance to M. abscessus pulmonary infection in cystic fibrosis patients. The altered pulmonary physiology of these patients makes them particularly susceptible to colonization by biofilm-forming bacteria such as Pseudomonas aeruginosa (Boucher, 2004). Chronic lung colonization by such bacteria causes a host inflammatory reaction that damages the lungs over time (Chmiel et al., 2002). Colonization of the lungs of these patients with M. abscessus may be favoured by strains with a smooth colony phenotype expressing GPL. Since M. abscessus 390S has a reversion frequency from smooth to rough in the range of 1 : 105–106, it is conceivable that a rough cord-forming variant could emerge from cystic fibrosis patients chronically colonized with a smooth variant expressing GPL. Our results raise the possibility that the emergence of such a variant expressing minimal GPL could herald the onset of more aggressive, invasive pulmonary infection. Such an emergence was recently demonstrated in a study involving Burkholderia cenocepacia. Importantly, the emergent non-biofilm-forming strain was able to persist at a much higher level in the lungs of normal mice, demonstrating that loss of biofilm-forming capability may be associated with expression of factors that increase virulence (Conway et al., 2004).
In conclusion, our study provides new insights into the pathogenesis of infection caused by M. abscessus, and establishes a framework for assessing the clinical correlation between M. abscessus infection and colony phenotype, particularly with regard to the roles of GPL expression and cord formation.
|
ACKNOWLEDGEMENTS |
|---|
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Barrow, W. W. & Brennan, P. J. (1982). Isolation in high frequency of rough variants of Mycobacterium intracellulare lacking C-mycoside glycopeptidolipids antigens. J Bacteriol 150, 381–384.
Barrow, W. W., Ullom, B. P. & Brennan, P. J. (1980). Peptidoglycolipid nature of the superficial cell wall sheath of smooth-colony-forming mycobacteria. J Bacteriol 144, 814–822.
Belisle, J. T. & Brennan, P. J. (1989). Chemical basis of rough and smooth variation in mycobacteria. J Bacteriol 171, 3465–3470.
Billman-Jacobe, H., McConville, M. J., Haites, R. E., Kovacevic, S. & Coppel, R. L. (1999). Identification of a peptide synthetase involved in the biosynthesis of glycopeptidolipids of Mycobacterium smegmatis. Mol Microbiol 33, 1244–1253.[CrossRef][Medline]
Boucher, R. C. (2004). New concepts of the pathogenesis of cystic fibrosis lung disease. Eur Respir J 23, 146–158.
Brennan, P. J. & Goren, M. B. (1979). Structural studies on the type-specific antigens and lipids of the Mycobacterium avium–Mycobacterium intracellulare–Mycobacterium scrofulaceum complex. J Biol Chem 254, 4205–4211.
Brennan, P. J. & Nikaido, H. (1995). The envelope of mycobacteria. Annu Rev Biochem 64, 29–63.[CrossRef][Medline]
Brown-Elliott, B. A. & Wallace, R. J., Jr (2002). Clinical and taxonomic status of pathogenic nonpigmented or late-pigmenting rapidly growing mycobacteria. Clin Microbiol Rev 15, 716–746.
Brown-Elliott, B. A. & Wallace, R. J. (2005). Infections caused by nontuberculous mycobacteria. In Principles and Practice of Infectious Diseases, 6th edn, vol. 2, pp. 2910–2911. Edited by G. L. Mandell, J. E. Bennett & R. Dolin. Philadelphia, PA: Elsevier.
Byrd, T. F. (1997). TNF promotes growth of virulent M. tuberculosis in human monocytes: iron-mediated growth suppression is correlated with decreased release of TNF from iron-treated, infected monocytes. J Clin Invest 99, 2518–2529.[Medline]
Byrd, T. F. & Lyons, C. R. (1999). Preliminary characterization of a Mycobacterium abscessus mutant in human and murine models of infection. Infect Immun 67, 4700–4707.
Camphausen, R. T., Jones, R. L. & Brennan, P. J. (1985). A glycolipid antigen specific to Mycobacterium paratuberculosis: structure and antigenicity. Proc Natl Acad Sci U S A 82, 3068–3072.
Carter, G., Wu, M., Drummond, D. C. & Bermudez, L. E. (2003). Characterization of biofilm formation by clinical isolates of Mycobacterium avium. J Med Microbiol 52, 747–752.
Ceri, H., Olson, M. E., Stremick, C., Read, R. R., Morck, D. & Buret, A. (1999). The Calgary biofilm device: new technology for rapid determination of antibiotic susceptibilities of bacterial biofilms. J Clin Microbiol 37, 1771–1776.
Chmiel, J. F., Berger, M. & Konstan, M. W. (2002). The role of inflammation in the pathophysiology of CF lung disease. Clin Rev Allergy Immunol 23, 5–27.[CrossRef][Medline]
Collins, F. M. & Cunningham, D. S. (1981). Systemic Mycobacterium kansasii infection and regulation of the alloantigenic response. Infect Immun 32, 614–624.
Conway, B. D., Chu, K. K., Bylund, J., Altman, E. & Speert, D. P. (2004). Production of exopolysaccharide by Burkholderia cenocepacia results in altered cell-surface interactions and altered bacterial clearance in mice. J Infect Dis 190, 957–966.[CrossRef][Medline]
Cullen, A. R., Cannon, C. L., Mark, E. J. & Colin, A. A. (2000). Mycobacterium abscessus infection in cystic fibrosis. Colonization or infection? Am J Respir Crit Care Med 161, 641–645.
Eckstein, T. M., Inamine, J. M., Lambert, M. L. & Belisle, J. T. (2000). A genetic mechanism for deletion of the ser2 gene cluster and formation of rough morphological variants of Mycobacterium avium. J Bacteriol 182, 6177–6182.
Fauroux, B., Delaisi, B., Clement, A., Saizou, C. & Moissenet, D. (1997). Mycobacterial lung disease in cystic fibrosis: a prospective study. Pediatr Infect Dis J 16, 354–358.[CrossRef][Medline]
Fregnan, G. B. & Smith, D. W. (1962). Description of various colony forms of mycobacteria. J Bacteriol 83, 819–826.
Furuchi, A. & Tokunaga, T. (1972). Nature of the receptor substance of Mycobacterium smegmatis for D4 bacteriophage adsorption. J Bacteriol 111, 404–411.
Glickman, M. S., Cox, J. S. & Jacobs, W. R. (2000). A novel mycolic acid cyclopropane synthetase is required for cording, persistence, and virulence of Mycobacterium tuberculosis. Mol Cell 5, 717–727.[CrossRef][Medline]
Goren, M. B., McClatchy, J. K., Martens, B. & Brokl, O. (1972). Mycosides C: behavior as receptor site substance for mycobacteriophage D4. J Virol 9, 999–1003.
Griffith, D. E., Girard, W. & Wallace, R. J. (1993). Clinical features of pulmonary disease caused by rapidly growing mycobacteria. An analysis of 154 patients. Am Rev Respir Dis 147, 1271–1278.[Medline]
Howard, S. T. & Byrd, T. F. (2000). The rapidly growing mycobacteria: saprophytes and parasites. Microbes Infect 2, 1845–1853.[CrossRef][Medline]
Howard, S. T., Byrd, T. F. & Lyons, C. R. (2002). A polymorphic region in Mycobacterium abscessus contains a novel insertion sequence element. Microbiology 148, 2987–2996.
Kansal, R. G., Gomez-Flores, R. & Mehta, R. T. (1998). Change in colony morphology influences the virulence as well as the biochemical properties of the Mycobacterium avium complex. Microb Pathog 25, 203–214.[CrossRef][Medline]
Lopez-Marin, L. M., Gautier, N., Laneele, M., Silve, G. & Daffe, M. (1994). Structures of the glycopeptidolipid antigens of Mycobacterium abscessus and Mycobacterium chelonae and possible chemical basis of the serological cross-reactions in the Mycobacterium fortuitum complex. Microbiology 140, 1109–1118.[Abstract]
Lorian, V. (1966). Direct cord reading medium for isolation of mycobacteria. Appl Microbiol 14, 603–607.[Medline]
Martinez, A., Torella, S. & Kolter, R. (1999). Sliding motility in mycobacteria. J Bacteriol 181, 7331–7338.
McNeil, M., Chatterjee, D., Hunter, W. & Brennan, P. (1989). Mycobacterial glycolipids: isolation, structures, antigenicity, and synthesis of neoantigens. Methods Enzymol 179, 215–242.[Medline]
Middlebrook, G., Dubos, R. J. & Pierce, C. (1947). Virulence and morphological characteristics of mammalian tubercle bacilli. J Exp Med 175–189.
Olivier, K. N., Weber, D. J., Wallace, R. J. & 10 other authors (2003). Nontuberculous mycobacteria. I: Multicenter prevalence study in cystic fibrosis. Am J Respir Crit Care Med 167, 828–834.
Ortalo-Magné, A., Lemassu, A., Lanéelle, M.-A., Bardou, F., Silve, G., Gounon, P., Marchal, G. & Daffé, M. (1996). Identification of the surface-exposed lipids on the cell envelope of Mycobacterium tuberculosis and other mycobacterial species. J Bacteriol 178, 456–461.
Rao, V., Fujiwara, N., Porcelli, S. A. & Glickman, M. S. (2005). Mycobacterium tuberculosis controls host innate immune activation through cyclopropane modification of a glycolipid effector molecule. J Exp Med 201, 535–543.
Recht, J. & Kolter, R. (2001). Glycopeptidolipid acetylation affects sliding motility and biofilm formation in Mycobacterium smegmatis. J Bacteriol 183, 5718–5724.
Recht, J., Martinez, A., Torello, S. & Kolter, R. (2000). Genetic analysis of sliding motility in Mycobacterium smegmatis. J Bacteriol 182, 4348–4351.
Reddy, V. M., Parikh, K., Luna-Herrera, J., Falkinham, J. O., Brown, S. & Gangadharam, P. R. (1994). Comparison of virulence of Mycobacterium avium complex strains (MAC) isolated from AIDS and non-AIDS patients. Microb Pathog 16, 121–130.[CrossRef][Medline]
Rhoades, E., Hsu, F., Torrelles, J. B., Turk, J., Chatterjee, D. & Russell, D. G. (2003). Identification and macrophage-activating activity of glycolipids released from intracellular Mycobacterium bovis BCG. Mol Microbiol 48, 875–888.[CrossRef][Medline]
Sambrook, J., Fritsch, E. F. & Maniatis, T. (1989). Molecular Cloning: a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory.
Sanguinetti, M., Ardito, F., Fiscarelli, E., La Sorda, M., D'Argenio, P., Ricciotti, G. & Fadda, G. (2001). Fatal pulmonary infection due to multidrug-resistant Mycobacterium abscessus in a patient with cystic fibrosis. J Clin Microbiol 39, 816–819.
Schaefer, W. R., Davis, C. L. & Cohn, M. L. (1970). Pathogenicity of transparent, opaque, and rough variants of Mycobacterium avium in chickens and mice. Am Rev Respir Dis 102, 499–506.[Medline]
Sermet-Gaudelus, I., Le Bourgeois, M., Pierre-Audigier, C. & 11 other authors (2003). Mycobacterium abscessus and children with cystic fibrosis. Emerg Infect Dis 9, 1587–1591.[Medline]
Torrelles, J. B., Ellis, D., Osborne, T., Hoefer, A., Orme, I. M., Chatterjee, D., Brennan, P. J. & Cooper, A. M. (2002). Characterization of virulence, colony morphotype and the glycopeptidolipid of Mycobacterium avium strain 104. Tuberculosis 82, 293–300.
Received 24 October 2005;
revised 7 March 2006;
accepted 7 March 2006.
This article has been cited by other articles:
|
|
 |
|
  R. Greendyke and T. F. Byrd Differential Antibiotic Susceptibility of Mycobacterium abscessus Variants in Biofilms and Macrophages Compared to That of Planktonic Bacteria Antimicrob. Agents Chemother., June 1, 2008; 52(6): 2019 - 2026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Fujiwara, N. Nakata, T. Naka, I. Yano, M. Doe, D. Chatterjee, M. McNeil, P. J. Brennan, K. Kobayashi, M. Makino, et al. Structural Analysis and Biosynthesis Gene Cluster of an Antigenic Glycopeptidolipid from Mycobacterium intracellulare J. Bacteriol., May 15, 2008; 190(10): 3613 - 3621. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. E. Jonsson, M. Gilljam, A. Lindblad, M. Ridell, A. E. Wold, and C. Welinder-Olsson Molecular Epidemiology of Mycobacterium abscessus, with Focus on Cystic Fibrosis J. Clin. Microbiol., May 1, 2007; 45(5): 1497 - 1504. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Freeman, H. Geier, K. M. Weigel, J. Do, T. E. Ford, and G. A. Cangelosi Roles for Cell Wall Glycopeptidolipid in Surface Adherence and Planktonic Dispersal of Mycobacterium avium Appl. Envir. Microbiol., December 1, 2006; 72(12): 7554 - 7558. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||
, 390R; 
,390S; 
, 390V; *, 390S versus 390R and 390V, P<0.01, t-test. (b) M. abscessus variants were inoculated as above. At 48 h the bacteria from pegs attached to lids were fixed and stained with crystal violet dye. Bacteria were then destained with ethanol and the A570 value of the ethanol solution determined. Data are the mean of quadruplicate determinations at 48 h. *, 390S versus 390R and 390V, P<0.01, t-test. (c) The tips of crystal violet-stained M. abscessus pegs from (b) were photographed using an inverted phase-contrast microscope (x100 magnification) prior to destaining. 390S demonstrates a relatively uniform layer of bacteria on the tip of the peg, whereas 390R and 390V demonstrate only occasional dark-staining bacterial aggregates.
