A purF mutant of Mycobacterium smegmatis has impaired survival during oxygen-starved stationary phase

Jacquie Keera,1, Marjan J. Smeulders1 and Huw D. Williams1

Department of Biology, Imperial College of Science, Technology and Medicine, Imperial College Road, London SW7 2AZ, UK1

Author for correspondence: Huw D. Williams. Tel: +44 20 75945383. Fax: +44 20 75842056. e-mail: h.d.williams{at}ic.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In this study it was demonstrated that a range of transposon mutants of Mycobacterium smegmatis, previously described as having impaired survival in carbon-starved stationary phase, were not markedly affected in O2-starved stationary-phase survival. One exception was 329B, a purine auxotroph, which showed a precipitous reduction in viability from 108 to 103 c.f.u. ml-1 during the first 5–10 d in O2-starved stationary phase. This was followed by an equally rapid recovery in culturability to a level within 10–100-fold of wild-type levels by 10–20 d into stationary phase. Transduction of the mutation into a clean genetic background demonstrated that the phenotype was due to the transposon insertion, which was shown to be in the purF gene. purF encodes phosphoribosylpyrophosphate amidotransferase, which catalyses the first committed step in purine biosynthesis. The M. smegmatis purF gene, which encodes a protein with a very high degree of similarity to the PurF homologues of Mycobacterium tuberculosis and Mycobacterium leprae, was cloned and shown to substantially complement the O2-starvation phenotype. The recovery in culturabilty of the purF mutant in O2-starved stationary phase did not involve movement of the transposon. In addition, when cells that had recovered culturability were retested, their survival kinetics in stationary phase were identical to the original culture, indicating that their recovery was not explained by the accumulation of suppressor mutations. It is concluded that the survival curve in O2-starved stationary phase for the purF mutant represents its true phenotype and is not a result of subsequent genetic changes in the culture. It is argued that the purF cells lose culturability for a finite period of time in stationary phase. Whether this is due to a fraction of the population dying and then regrowing using a previously undiscovered fermentation pathway, or becoming transiently dormant, or entering an active nonculturable state and subsequently undergoing resuscitation cannot be distinguished at this stage.

Keywords: dormancy, Mycobacterium tuberculosis, purine, VBNC, oxygen

The GenBank accession number for the sequence reported in this paper is AJ278609.

a Present address: LGC, Queens Road, Middlesex TW11 0LY, UK.


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
It is estimated that a third of the world’s population is infected with Mycobacterium tuberculosis, although most are latently infected, carrying an asymptomatic infection in which the bacilli persist for long periods without causing disease (Parrish et al., 1998 ; Wayne, 1994 ). How the bacteria survive during clinical latency is not known.

We are studying the mechanisms by which mycobacteria survive nutrient deprivation as a model for mycobacterial persistence. We have previously described the physiological changes accompanying the adaptation of Mycobacterium smegmatis to carbon-starved stationary phase (Smeulders et al., 1999 ). Carbon-starved cultures maintain viability for at least 2 years. M. smegmatis may sense when the carbon source is approaching limiting concentrations and then initiate an adaptive response to stationary-phase survival. During early stationary phase cells undergo reductive cell division and become more resistant to environmental stress and stabilize their mRNA. However, it is apparent that continuous cell growth and cell division occurs in stationary phase. Strains with altered colony morphology appear and take over stationary-phase cultures, and competition experiments between stationary-phase-adapted and exponential-phase-adapted strains show that variants appear that have growth and survival advantages in stationary phase (Smeulders et al., 1999 ).

We have recently described the isolation of mutants that are defective in stationary-phase survival in both rich medium and during carbon starvation (Keer et al., 2000 ). These were identified from a Tn611 (Guilhot et al., 1994 ) mutant library by screening for mutants that had lost viability in carbon-starved stationary phase. These mutants were also defective in stationary-phase survival in rich medium. Interestingly, all the mutants were disrupted in genes with homologues in the M. tuberculosis genome (Keer et al., 2000 ). The identification of one of these as a gene (ponA) encoding a penicillin-binding protein is consistent with the need for cell division in stationary phase (Smeulders et al., 1999 ; Keer et al., 2000 ).

Our previous work used O2-sufficient, stationary-phase cultures. However, one model for mycobacterial persistence that has been used in a number of laboratories is O2 starvation (Wayne, 1994 ; Wayne & Hayes, 1996 ; Lim et al., 1999 ). In this model, cultures of the bacteria are subjected to a temporal gradient of O2 depletion by incubating in sealed tubes. This model is based on the idea that bacteria may be deprived of O2 in the caseous, necrotic centres of tuberculous lesions. Two genes encoding sigma factors are upregulated during O2-limited stationary phase (Hu & Coates, 1999a ), as is the 16 kDa {alpha}-crystallin heat-shock protein (Hu & Coates, 1999b ; Yuan et al., 1996 ) and the M. tuberculosis hmp gene product, which is homologous to the Escherichia coli flavohaemoglobin (Hu et al., 1999 ), as well as a histone-like protein in M. smegmatis (Lee et al., 1998 ).

An important question is whether the adaptation mechanisms for survival are the same under carbon- (O2-sufficient) and O2-starvation conditions. One way to address this question is to look for mutants that are defective in carbon- but not O2-starvation survival and vice versa. In this study we have looked at the O2-starvation survival of mutants previously identified as being impaired in stationary-phase survival under O2-sufficient, carbon-starved conditions, and we describe the characterization of a mutant that is defective in O2-but not carbon-starvation survival.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains and plasmids.
The strains and plasmids used in this study are shown in Table 1. 272A, 272E, 317C, 3910D, 492A and 412A are stationary-phase survival mutants described previously (Keer et al., 2000 ). Each, except 412A, has been shown to be mutated in a gene with a homologue in M. tuberculosis and the identities of these are indicated in Table 1 and described by Keer et al. (2000) .


View this table:
[in this window]
[in a new window]
 
Table 1. Strains and plasmids

 
Media and growth conditions.
E. coli cultures were grown in Luria broth at 37 °C. M. smegmatis was grown as previously described (Smeulders et al., 1999 ) at 37 °C with shaking at 200 r.p.m. in Lab-lemco medium or Hartmans–de Bont minimal medium (HdB; Hartmans & De Bont, 1992 ). Tween 80 was added to 0·05% (v/v). When required, adenosine and guanosine were added to media at a final concentration of 0·15 µg ml-1. For carbon-limitation-induced stationary-phase experiments, M. smegmatis was grown either as 5 ml cultures in 25 ml Universal tubes, or as 2 ml cultures in 18 ml test tubes in HdB medium containing 0·02% glycerol (Smeulders et al., 1999 ). For O2-starvation experiments, cultures were grown in sealed (Suba seals; SLS) 18 ml test tubes containing 12 ml HdB medium plus 0·08% glycerol, with a resultant head-space ratio (HSR) of 1·5 (Wayne & Hayes, 1996 ). Preliminary experiments in which the HSR was varied inferred that with a HSR of 1·5 cultures entered stationary phase due to O2-starvation. For example, with 4 ml medium, giving a HSR of 4·5, the culture entered stationary phase with an OD600 of ~1·1, whereas with a HSR of 1·5, stationary phase was entered at an OD600 of ~0·25. Therefore, the most plausible explanation for the growth arrest at a lower OD600 with a HSR of 1·5 is O2-starvation. For viability testing of O2-starved cultures, 100 µl samples were taken through the Suba seals. We did not equalize the pressure following sampling so as not to compromise the head-space gas composition. Therefore, following sampling the cultures were under slight negative pressure. It was obvious if the pressure did equalize with the atmosphere due to failure of the Suba seal, and if this occurred the culture was not used further. Viability was assayed by plating samples of cultures, diluted appropriately in PBS (8 g NaCl l-1, 0·2 g KCl l-1, 1·44 g Na2HPO4 l-1, 0·24 g KHPO4 l-1; pH 7·4) plus 0·05% (v/v) Tween 80, onto selective Lab-lemco plates at 37 °C for 5 d. All viability experiments were performed at least three times and representative data are shown. When required, kanamycin and streptomycin were added to final concentrations of 50 and 20 µg ml-1, respectively.

Isolation of transposon mutants impaired in stationary-phase survival.
M. smegmatis mutants impaired in stationary-phase survival were isolated from a mutant library constructed by transposon mutagenesis using Tn611 (Guilhot et al., 1994 ), as described by Keer et al. (2000) .

Cloning the M. smegmatis purF gene.
A M. smegmatis gene library in the cosmid vector LAWRIST4 was kindly provided by Neil Stoker, London School of Hygiene and Tropical Medicine. Five hundred and seventy-six cosmid clones in LAWRIST4 were screened in hybridization experiments with the 640 bp PstI/HindIII fragment isolated from the Tn611 mutant 329B to identify cosmids containing putative purF genes. Two hybridizing cosmids were isolated and 1·5 kb PstI, 3·8 kb BamH1 and 4·5 kb SacI fragments were subcloned from these cosmids into pBluescript. The 1·5 kb fragment was completely sequenced and contained the majority of the M. smegmatis purF gene. Sequencing fragments from the other clones allowed us to obtain the complete purF sequence. To complement 329B, the 4·5 kb SacI fragment from pPUR3 was subcloned into the mycobacterial shuttle cosmid pNBV1 (Howard et al., 1995 ) to form pPUR10 (Table 1).

Transduction.
Mycobacteriophage I3 lysates from Tn611 mutant strains were prepared according to Sundar Raj & Ramakrishnan (1970) . Mutations were transduced by infecting 5x109 c.f.u. exponentially growing M. smegmatis mc2155 with mutant I3 lysates at a m.o.i. of 2, and transductants were selected on Lab-lemco plates containing kanamycin.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Purine auxotrophy leads to impaired O2-starvation survival of M. smegmatis
One proposed model of clinical latency in M. tuberculosis utilizes a slow shiftdown into a non-growing state, induced by exposure to a temporal gradient of O2 depletion (Wayne & Hayes, 1996 ). To determine if a reduced ability to survive carbon-starved stationary phase or stationary phase in rich medium correlated with reduced survival in the O2 depletion model, the viability of a number of starvation-survival mutants previously reported (Keer et al., 2000 ), including a known auxotroph, 329B, described here, was tested under conditions of O2 starvation. The survival of wild-type M. smegmatis was different during O2-starved stationary phase compared to carbon-starved stationary phase (Figs 1, 2b), with viability dropping by 100-fold during the first 80 d of O2 starvation compared to <10-fold during carbon starvation. The survival of M. smegmatis during O2 starvation (Fig. 1) was similar to that described for M. tuberculosis during the first 10 d of O2-limited stationary phase (Wayne & Hayes, 1996 ). However, after this initial 10 d period there is no comparable data for M. tuberculosis in the literature. All the previously described stationary-phase survival mutants (Keer et al., 2000 ) were tested under O2-starvation conditions. The results shown in Fig. 1(a) for two mutants, 317C and 492A, are typical. Both 317C and 492A show an initial loss of viability that is a little faster than observed in the wild-type, but by day 25 viabilities had stabilized at wild-type levels. Similar survival curves under O2-starvation conditions were obtained for 272A, 272E, 3910D, 412A and all the other mutants tested (Keer et al., 2000 , data not shown) with one exception. This exception was 329B, which showed a precipitous, almost 105-fold reduction in viability during the first 10 d of O2 starvation (Fig. 1b). This was followed by a rapid increase in viability to within ~10-fold of the wild-type level. The pattern of survival for 329B was reproducible, although the actual viability levels varied between experiments. The loss of viability during the first 10 d of O2 starvation was between 103 and 106-fold c.f.u. ml-1 lower than the wild-type. In all experiments viability recovered to within 10–100-fold c.f.u. ml-1 of the wild-type viability by about day 15–20 and then remained constant at this level until the end of the experiment at about day 70. 329B is an auxotroph and requires exogenously added purines, adenosine and guanosine to grow in minimal medium, and it should be noted that these were present in the experiments described in Figs 1 and 2(b). The survival of 329B in rich Lab-lemco medium and under carbon-starved stationary-phase conditions is shown in Fig. 2. The mutant showed a marked loss of viability compared to the wild-type during the first 40 d in stationary phase in Lab-lemco, but following prolonged incubation viability levels recovered to close to those of the wild-type (Fig. 2a). Mutant 329B was not defective in stationary-phase survival under carbon starvation conditions in HdB minimal medium supplemented with purines (Fig. 2b).



View larger version (13K):
[in this window]
[in a new window]
 
Fig. 1. Survival of M. smegmatis strains in O2-starved stationary phase. Strains were grown in 0·02% glycerol-containing Hartmans–de Bont minimal medium with kanamycin, adenosine and guanosine, under conditions which result in stationary phase entry due to O2 starvation (see Methods). Viability was followed by plating onto Lab-lemco medium plus kanamycin. (a) M. smegmatis mc2155 ({blacksquare}), 317C ({circ}) and 492A ({triangleup}). (b) M. smegmatis mc2155 ({blacksquare}), purF mutant 329B ({bullet}), and phage I3 transductants of Tn611::purF from 329B: 329B N1 ({triangleup}), 329B N2 ({bigtriangledown}), 329B N3 ({star}) and 329B N4 ({circ}).

 


View larger version (10K):
[in this window]
[in a new window]
 
Fig. 2. Stationary-phase survival of M. smegmatis 329B. M. smegmatis mc2155 ({blacksquare}) and 329B ({square}) were grown into stationary phase in Lab-lemco medium plus kanamycin (a) or in 0·02% glycerol-containing Hartmans–de Bont minimal medium with kanamycin, adenosine and guanosine (b), conditions under which the cultures enter stationary phase due to glycerol exhaustion (Smeulders et al., 1999 ). Viability was followed by plating onto Lab-lemco medium plus kanamycin.

 
Transduction of the mutant phenotypes
To demonstrate unequivocally that the phenotype of 329B resulted from insertional mutagenesis by Tn611, the mutation was transduced into a clean genetic background using the mycobacterial transducing phage I3, and the resultant transductants were analysed. A 640 bp HindIII/PstI fragment from the transposon delivery vector pCG79 (Guilhot et al., 1994 ) was used as a hybridization probe on PstI digests of genomic DNA isolated from the original mutant and a number of transductants. This confirmed that the transposon was inserted into the same site in all strains (data not shown). 329B transductants showed similar patterns of O2-starved stationary-phase survival to the original mutant strains (Fig. 1b), confirming that the presence of the transposon insertion is responsible for the mutant phenotype.

Identification of purF as the disrupted gene in mutant 329B: cloning and sequencing of purF
To identify the gene disrupted in the purine auxotroph 329B, chromosomal DNA isolated from the mutant strain 329B was digested with NotI, which does not cut within the transposon vector itself, producing large chromosomal fragments. The fragments were recircularized and transformed into E. coli. Selection for kanamycin- and streptomycin-resistant colonies yielded plasmids containing the transposon insertion in its entirety plus additional flanking genomic sequences. PstI digestion of the clones produced the parental pCG79 fragments, plus an extra PstI fragment containing a portion of flanking chromosomal DNA. This fragment was sequenced and a BLAST search indicated it to be highly homologous to purF genes from a range of bacteria (data not shown). The purF gene encodes phosphoribosylpyrophosphate amidotransferease, which catalyses the first committed step in purine biosynthesis. This PstI fragment was used to screen a M. smegmatis genomic cosmid library in LAWRIST4. Two cosmid clones were isolated and 1·5 kb PstI, 3·8 kb PstI and 4·5 kb SacI fragments were subcloned from these into pBluescript (Table 1). The 1·5 kb PstI fragment containing the majority of the gene was used for sequencing. The GenBank accession number for this sequence is AJ278609. Alignment with M. tuberculosis and Mycobacterium leprae sequences shows very clearly the strong homology between the PurF proteins from these three mycobacterial species (Fig. 3).



View larger version (66K):
[in this window]
[in a new window]
 
Fig. 3. Alignment of the M. smegmatis PurF protein (Msm.pro) and the corresponding proteins in M. tuberculosis (Mtb.pro) and M. leprae (Mlep.pro). Identical residues are shown on a black background while boxed residues all match the residue group in the M. smegmatis sequence. Residue groupings are: (DE), (HKR), (AFILMPVW) and (CGNQSTY). Gaps are indicated by dashes.

 
Complementation of the purF::Tn611 mutant 329B
A 4·5 kb SacI fragment containing the entire purF gene was subcloned into the mycobacterial shuttle vector pNBV1 (Howard et al., 1995 ) to form pPUR10 (Table 1). pPUR10 was transformed into 329B and complementation of its purine auxotrophy was confirmed by assaying its ability to grow on unsupplemented minimal medium. Full complementation was apparent during growth and quantitative plating onto Lab-lemco medium and HdB medium, with and without added purines. During O2-starvation survival the viability of the complemented mutant was returned to near wild-type levels (Fig. 4).



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 4. Complementation of the M. smegmatis 329B purF mutant. Survival in O2-starved stationary phase of M. smegnatis mc2155 ({blacksquare}), 329B purF/pNBV1 ({circ}) and 329B/pPUR10 (purF+) ({bigtriangledown}) was determined as described in the legend to Fig. 1.

 
Stability of the purF phenotype
We were interested in establishing whether the apparent recovery in the plating efficiency of the purF mutant in O2-starved stationary phase was a result of a permanent genetic change, perhaps as a consequence of the movement of the transposon, or of the acquisition of suppressor mutations. Southern hybridization experiments indicated that the Tn611 did not move during incubation in O2-starved stationary phase (data not shown). To test for the accumulation of second-site suppressors, a sample taken from a 15-d-old, O2-starved, stationary-phase culture of 329B (the primary culture), in which plate counts had recovered to around 107 c.f.u. ml-1, was inoculated into fresh medium. The viability of the secondary culture was followed into O2-starved stationary phase (Fig. 5). Both the primary and the secondary culture showed identical survival kinetics in stationary phase. These data are consistent with the recovery of viability representing the true phenotype of the purF mutation and not being due to the accumulation of second-site suppressors.



View larger version (15K):
[in this window]
[in a new window]
 
Fig. 5. Adaptation of M. smegmatis 329B purF mutant to O2 starvation. The survival in O2-starved stationary phase of a pre-adapted (secondary) 329B culture ({triangleup}), was compared to that of a M. smegmatis wild-type culture ({blacksquare}) and a 329B (primary) culture ({circ}). The pre-adapted 329B culture was prepared by growing a 329B culture into O2-starved stationary phase for more than 15 d to allow recovery of culture viability to around 107 c.f.u. ml-1 to occur. A sample from this culture was then inoculated into fresh medium and its survival in O2-starved stationary phase followed.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
M. smegmatis is an obligate aerobe. No conditions have been described which allow its growth in the absence of O2, either using an alternative terminal electron acceptor or a fermentation mechanism for energy generation. In addition, there is no potential alternative terminal electron acceptor, such as nitrate, available in the HdB minimal medium. Therefore, restricting the amount of O2 available to that in the headspace of a sealed tube ensured that cultures entered stationary phase due to O2 starvation. Similarly, M. tuberculosis is an obligate aerobe; although the genes for dissimilatory nitrate reductase are found in its genome (Cole et al., 1998 ), to our knowledge no conditions have been described that allow its anaerobic growth. Where comparable experiments have been carried out, M. smegmatis, Mycobacterium bovis BCG and M. tuberculosis all show significant similarities in their behaviour under O2-starvation conditions, although there are also some differences (Cunningham & Spreadbury, 1998 ; Dick et al., 1998 ; Hutter & Dick, 1998 ; Lim et al., 1999 ). The paucity of knowledge of the survival of obligate aerobes in the absence of O2, together with the use of O2 starvation as a model for latent tuberculosis, led us to test the survival of stationary-phase survival mutants under O2-starved conditions. In this report we show that a purF mutant of M. smegmatis is defective in survival under O2-starvation conditions.

The purF mutant 329B was isolated using a screen to find mutants defective in carbon-starvation survival (Keer et al., 2000 ). 329B grew to give visible turbidity under the screen conditions of growth in 0·02% glycerol containing Hartmans–de Bont medium, presumably because there was sufficient carryover of purines in the inoculum to allow some growth. Its apparent loss of viability in the initial screen, compared to the wild-type, could in part reflect a lower final cell density due to purine limitation. It could also be due to the effects of O2 limitation in the microtitre plates, which were sealed with film to prevent evaporation of the medium during extended 37 °C incubation.

Intriguingly, 329B purF loses viability during O2 starvation in HdB minimal medium that has been supplemented with purines. One explanation is that purines are required for survival but are not taken up by the bacteria during the early stages of O2 starvation. Alternatively, the O2-starvation phenotype may not be a consequence of starvation for purines directly but results from a shortage of one or more intermediates in the purine biosynthesis pathway. Important in this context is the finding that a Rhizobium etli purF mutant and certain other purine biosynthesis mutants overexpress cytochrome cbb3, a cytochrome c oxidase (Soberon et al., 1997 ; García-Horsman et al., 1994 ). Analysis of the transcription of the fixNOQP operon, encoding cytochrome cbb3, in different R. etli purine biosynthetic mutants, suggests that 5 aminoimidazole-4-carboxamide ribonucleotide (AICAR) or a related metabolite is a negative regulator of this oxidase (Soberon et al., 1997 ). It would be interesting to investigate the survival phenotypes of other M. smegmatis purine biosynthesis mutants. To our knowledge, nothing is known about the respiratory chain of M. smegmatis. In preliminary studies, we have found that while overall respiratory capacity, determined as NADH oxidase activity, is unchanged in a M. smegmatis purF mutant, cytochrome c oxidase activity is two- to threefold higher in the mutant compared to the wild-type, although this activity does not reduce to wild-type levels upon complementation with pPUR10 (Keer et al., unpublished results). However, the cytochrome c oxidase activities and cytochrome c levels of M. smegmatis are low compared with, for example, Pseudomonas aeruginosa, which has a cytochrome cbb3 (Keer et al., unpublished results). This requires further investigation to show directly whether or not overexpression of a specific terminal oxidase is responsible for the O2-starvation phenotype. How could loss of control of a terminal oxidase affect survival during O2 starvation? One possibility is that inappropriate levels of oxidase expression may affect the adaptation to O2 starvation, by preventing electron flux through the most favourable respiratory pathway. Alternatively, a defect in oxidase regulation may affect exit from stationary phase, which is required for viability determination of samples taken from the O2-starved culture. There is a precedent for this in the stationary-phase exit defect found in E. coli cytochrome-bd-deficient mutants (Siegele & Kolter, 1993 ; Siegele et al; 1996 ; Goldman et al., 1996 ). However, there may be other explanations for the effect of purF mutation on O2-starvation survival. In Lactobacillus lactis there is evidence for cellular metabolic pathways being intimately related to stress responses (Duwat et al., 1999 ; Rallu et al., 2000 ). Changes in guanine nucleotide pools affect the ability to synthesize (p)ppGpp and related signal molecules, although one might expect (p)ppGpp also to have a role in carbon-starvation survival. Studies of acid stress of L. lactis suggest that modification of the flux through the purine nucleotide pathway or increased pppGpp concentrations are perceived as intracellular stress signals in this organism, leading to multistress resistance (Rallu et al., 2000 ).

M. smegmatis loses viability more rapidly in the early stages of O2-starved stationary phase compared to carbon-starved stationary phase. This may reflect the dynamic state of the culture under carbon-starvation conditions, with new variants with improved survival capabilities appearing during incubation (Smeulders et al., 1999 ), their growth being fuelled by nutrients (carbon sources) released by dying cells and the availability of O2 as a terminal electron acceptor for respiration. During O2 starvation there is no terminal electron acceptor available and so cryptic growth would only be possible if the bacteria used an undiscovered fermentation pathway. The finding that there are mutations which are important for carbon-starved stationary phase but not for O2-starved stationary phase, and that the purF mutant is affected in O2- but not carbon-starvation survival suggests that different survival mechanisms do operate under both conditions. It is interesting that mutant 317C, which is defective in a putative penicillin-binding protein (Keer et al., 2000 ), was not markedly defective in survival during O2 starvation while it is in rich medium and following carbon starvation. The need for a penicillin-binding protein in stationary-phase survival is consistent with our previous demonstration of the dynamic nature of M. smegmatis stationary-phase cultures and the appearance of new variants, which presumably requires growth and cell division (Smeulders et al., 1999 ). The absence of a survival phenotype of this mutant during O2 starvation raises the issue of whether O2-starved cultures are similarly dynamic.

How do the purF mutants recover culturability from day 8–15 onwards under O2 starvation? Southern blotting showed that recovery did not involve movement of the transposon and the survival kinetics of a culture pre-adapted to O2 starvation were not consistent with the accumulation of suppressor mutations (Fig. 5). Therefore, we are left with the explanation that the dramatic loss in viability followed by a partial recovery is the phenotype of the purF mutant. This is supported by the complementation experiment in which pPUR10 complemented the ability of 329B to grow on unsupplemented minimal medium and partially, but not completely, its ability to survive O2 starvation to wild-type levels. The lack of full complementation may be due to the multicopy complementation leading to an imbalance in the levels of purine biosynthetic intermediates. Two possible mechanisms can explain the recovery in the culturability of the purF mutant in O2-starved stationary phase. Firstly, as conditions change in stationary phase, regrowth (cryptic growth) of the viable population (typically 103–104 c.f.u. ml-1 at day 8–12), occurs using nutrients released by dead cells. This seems improbable during O2 starvation for the reasons discussed above. However, the possibility cannot be ruled out that an unknown fermentation pathway is available to M. smegmatis under these conditions or that a terminal electron acceptor is made available following the breakdown of dead cells. A second explanation is that the loss of culturability during the early days of O2 starvation results from a significant fraction of the population becoming dormant or active but nonculturable (ANC). We prefer the term ANC here to viable but nonculturable (VBNC) for the reasons discussed by others (Barer, 1997 ; Barer & Harwood, 1999 ; Kell et al., 1998 ). The outcome of dormancy or the ANC state is a transient loss of culturabilty, although the physiological properties of the cells would differ markedly between these two states. If the cells become dormant they would enter a reversible state of low metabolic activity while if they became ANC they would remain metabolically active. The dormant or ANC cells then resuscitate and recover culturability as conditions change in stationary phase. Resuscitation would not necessitate regrowth. We cannot distinguish between these possibilities at present although their investigation will form the basis of future experiments. Interestingly, a recent investigation by Bogosian et al. (2000) indicates that the apparent resuscitation of Vibrio vulnificus from a VBNC (ANC) state (Whitesides & Oliver, 1997 ) may be explained by the emergence of an injured, hydrogen-peroxide-sensitive subpopulation. An intriguing possibility is that something similar may be happening in the M. smegmatis purF mutant.


   ACKNOWLEDGEMENTS
 
This work was funded by the Wellcome Trust, via a Wellcome Prize Fellowship to M.J.S and grants to H.D.W. We are very grateful to Neil Stoker for providing the cosmid library.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Barer, M. R.(1997). Viable but non-culturable and dormant bacteria: time to resolve an oxymoron and a misnomer? J Med Microbiol 46, 629-631.[Medline]

Barer, M. R. & Harwood, C. R.(1999). Bacterial viability and culturability. Adv Microb Physiol 41, 93-137.[Medline]

Bogosian, G., Noelle, D., Aardema, E. V., Bourneuf, E. V., Morris, P. J. L. & O’Neil, J. P.(2000). Recovery of hydrogen peroxide-sensitive culturable cells of Vibrio vulnificus gives the appearance of resuscitation from a viable but nonculturable state. J Bacteriol 182, 5070-5075.[Abstract/Free Full Text]

Cole, S. T., Brosch, R., Parkhill, J. & 39 other authors (1998). Deciphering the biology of Mycobacterium tuberculosis from the complete genome sequence. Nature 393, 537–544.[Medline]

Cunningham, A. F. & Spreadbury, C. L.(1998). Mycobacterial stationary phase induced by low oxygen tension: cell wall thickening and localization of the 16-kilodalton {alpha}-crystallin homolog. J Bacteriol 180, 801-808.[Abstract/Free Full Text]

Dick, T., Lee, B. H. & Murugasu-Oei, B.(1998). Oxygen depletion induced dormancy in Mycobacterium smegmatis. FEMS Microbiol Lett 163, 159-164.[Medline]

Duwat, P., Ehrlich, S. D. & Gruss, A.(1999). Effects of metabolic flux on stress response pathways in Lactococcus lactis. Mol Microbiol 31, 845-858.[Medline]

García-Horsman, J. A., Barquera, B., Rumbley, J., Ma, J. & Gennis, R. B.(1994). The superfamily of haem-copper respiratory oxidases. J Bacteriol 176, 5587-5600.[Medline]

Goldman, B. S., Gabbert, K. K. & Kranz, R. G.(1996). The temperature-sensitive growth and survival phenotypes of Escherichia coli cydDC and cydAB strains are due to deficiencies in cytochrome bd and are corrected by catalase and reducing agents. J Bacteriol 178, 6348-6351.[Abstract]

Guilhot, C., Otal, I., Van Rompaey, I., Martin, C. & Gicquel, B.(1994). Efficient transposition in mycobacteria: construction of Mycobacterium smegmatis insertional mutant libraries. J Bacteriol 176, 535-539.[Abstract]

Hartmans, S. & De Bont, J. A. M. (1992). The genus Mycobacterium – nonmedical. In The Prokaryotes,2nd edn, vol. II, pp. 1215–1237. Edited by A. Balows and others. New York: Springer.

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.[Medline]

Hu, Y. & Coates, A. R. M.(1999a). Transcription of two sigma 70 homologue genes, sigA and sigB, in stationary-phase Mycobacterium tuberculosis. J Bacteriol 181, 469-476.[Abstract/Free Full Text]

Hu, Y. & Coates, A. R. M.(1999b). Transcription of the stationary-phase-associated hspX gene of Mycobacterium tuberculosis is inversely related to synthesis of the 16-kilodalton protein. J Bacteriol 181, 1380-1387.[Abstract/Free Full Text]

Hu, Y., Butcher, P. D., Mangan, J. A., Rajandream, M. A. & Coates, A. R. M.(1999). Regulation of hmp gene transcription in Mycobacterium tuberculosis: effects of oxygen limitation and nitrosative and oxidative stress. J Bacteriol 181, 3486-3493.[Abstract/Free Full Text]

Hutter, B. & Dick, T.(1998). Increased alanine dehydrogenase activity during dormancy in Mycobacterium smegmatis. FEMS Microbiol Lett 167, 7-11.[Medline]

Keer, J., Smeulders, M. J., Gray, K. M. & Williams, H. D.(2000). Mutants of Mycobacterium smegmatis impaired in stationary-phase survival. Microbiology 146, 2209-2217.[Abstract/Free Full Text]

Kell, D. B., Kaprelyants, A. S., Weichart, D. H., Harwood, C. R. & Barer, M. B.(1988). Viability and activity in readily culturable bacteria: a review and discussion of the practical issues. Antonie Leeuwenhoek 73, 169-187.

Lee, B. H., Murugasu-Oei, B. & Dick, T.(1998). Upregulation of a histone-like protein in dormant Mycobacterium smegmatis. Mol Gen Genet 260, 475-479.[Medline]

Lim, A., Eleuterio, M., Hutter, B., Murugasu-Oei, B. & Dick, T.(1999). Oxygen depletion-induced dormancy in Mycobacterium bovis BCG. J Bacteriol 181, 2252-2256.[Abstract/Free Full Text]

Parrish, N. M., Dick, J. D. & Bishai, W. R.(1998). Mechanisms of latency in Mycobacterium tuberculosis. Trends Microbiol 6, 107-112.[Medline]

Rallu, F., Gruss, A., Ehrlich, S. D. & Maguin, E.(2000). Acid- and multistress-resistant mutants of Lactococcus lactis: identification of intracelluar stress signals. Mol Microbiol 35, 517-528.[Medline]

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

Siegele, D. & Kolter, R.(1993). Isolation and characterisation of an Escherichia coli mutant defective in resuming growth after starvation. Genes Dev 7, 2629-2640.[Abstract]

Siegele, D., Imlay, K. R. C. & Imlay, J. A.(1996). The stationary-phase-exit defect of cydC (surB) mutants is due to the lack of a functional terminal cytochrome oxidase. J Bacteriol 178, 6091-6096.[Abstract]

Smeulders, M. J., Keer, J., Speight, R. A. & Williams, H. D.(1999). Adaptation of Mycobacterium smegmatis to stationary phase. J Bacteriol 181, 270-283.[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]

Soberon, M., Lopez, O., Miranda, J., Tabche, M. L. & Morera, C.(1997). Genetic evidence for 5 aminoimidazole-4-carboxamide ribonucleotide (AICAR) as a negative effector of cytochrome terminal oxidase complex cbb3 production in Rhizobium etli. Mol Gen Genet 254, 665-673.[Medline]

Sundar Raj, C. V. & Ramakrishnan, T.(1970). Transduction in Mycobacterium smegmatis. Nature 228, 280-281.

Wayne, L. G.(1994). Dormancy of Mycobacterium tuberculosis and latency of disease. Eur J Clin Microbiol Infect Dis 13, 908-914.[Medline]

Wayne, L. G. & Hayes, L. G.(1996). An in vitro model for sequential study of shiftdown of Mycobacterium tuberculosis through two stages of nonreplicating persistence. Infect Immun 64, 2062-2069.[Abstract]

Whitesides, M. D. & Oliver, J. D.(1997). Resuscitation of Vibrio vulnificus from the viable but nonculturable state. Appl Environ Microbiol 63, 1002-1005.[Abstract]

Yuan, Y., Crane, D. D. & Barry, C. E.III(1996). Stationary phase-associated protein expression in Mycobacterium tuberculosis: function of the mycobacterial alpha-crystallin homolog. J Bacteriol 178, 4484-4492.[Abstract]

Received 17 July 2000; revised 2 October 2000; accepted 30 October 2000.