Correlation of the rate of protein synthesis and the third power of the RNA : protein ratio in Escherichia coli and Mycobacterium tuberculosis

Robert A. Cox

Division of Mycobacterial Research, National Institute for Medical Research, London NW7 1AA, UK

Correspondence
Robert A. Cox
rcox{at}nimr.mrc.ac.uk


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
DISCUSSION
REFERENCES
 
In order to further understand the different physiological states of the tubercle bacillus, a frame of reference was sought by first correlating the macromolecular compositions of Escherichia coli with specific growth rates and also with the rates of protein synthesis. Data for DNA : protein : RNA were converted to the average amounts of DNA [mDNA(av)], protein [mp(av)] and RNA [mRNA(av)] per cell. The specific growth rate µ was found to be directly proportional to mRNA(av)/mp(av). The specific protein synthesis rate per average cell [{omega}p(av)] was shown to be directly proportional to the third power of the ratio mRNA(av)/mp(av) which reflects the ribosome concentration. The equations derived were shown apply to both E. coli (µ=1·73 h-1) and Mycobacterium bovis BCG (µ=0·029 h-1).


Abbreviations: BCG, bacillus Calmette–Guérin


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
DISCUSSION
REFERENCES
 
Why does Mycobacterium tuberculosis require 24 h to divide when Mycobacterium smegmatis only takes 2 h? (Jacobs, 2000)

Mycobacteria are aerobic, non-motile, rod-shaped bacteria that are Gram-positive and acid-fast. Mycobacterium leprae and M. tuberculosis are human pathogens; their slow growth [generation times of 12 days (Shepard, 1960) and approaching 1 day (Wayne, 1994), respectively] is a notable property. The emergence of drug-resistant strains of M. tuberculosis has led to tuberculosis again becoming a threat to world health. Pathogenic mycobacteria have challenging properties such as an ability to survive within host cells. They are engulfed by host macrophages but survive and grow within phagosomes (Armstrong & D'Arcy-Hart, 1971; Ferrari et al., 1999). Treatment of tuberculosis is lengthy because bacilli persist despite chemotherapy, allowing the illness to resume its course if drug treatment is stopped prematurely. In the persistent state of the bacillus, drug resistance appears physiological rather than genetic in origin. It is known that, under hypoxic conditions, M. tuberculosis can exist in a dormant state that is resistant to standard antimycobacterial drugs. Tubercle bacilli can encounter hypoxic conditions in vivo (Weber et al., 2000) and oxygen starvation is thought to halt growth and lead to dormancy (Wayne & Hayes, 1996). Reactivation of dormant cells is thought to be the cause of the disease appearing many years after the exposure to the tubercle bacillus.

A knowledge of mycobacterial physiology during exponential growth in vitro and in vivo, within macrophages, adaptation to oxygen starvation and of the dormant state, would further our understanding of the course of tuberculosis. Current knowledge has been limited by technical problems such as slow growth and cell aggregation (for review see Ratledge, 1982; Wheeler & Ratledge, 1994). For example, the macromolecular compositions of M. tuberculosis grown under different conditions have not been reported.

Proteins comprise approximately one half of the dry mass of a bacterial cell (Bremer & Dennis, 1996). Ninety-five per cent of the energy used by the cell to synthesize macromolecules is devoted to the synthesis of proteins and 5 % is used to synthesize DNA, RNA, peptidoglycan, phospholipid, lipopolysaccharides and polysaccharides (Ingraham et al., 1983). The RNA content of a cell, of which approximately 83 % is rRNA (Bremer & Dennis, 1996; Butcher et al., 1999), reflects the number of ribosomes per cell. The central role played by ribosomes in protein biosynthesis suggests that there is a relationship between protein content, the rate of protein synthesis and RNA content.

The significance of the relationship between the specific growth rates and the protein and RNA contents of Mycobacterium bovis bacillus Calmette–Guérin (BCG), a close relative of M. tuberculosis (Brosch et al., 2000), is the subject of this study. Two questions were formulated. First, how is the average ribosome concentration of a bacterial culture related to µ, the specific growth rate? Second, how is the average ribosome concentration related to the average rate of protein synthesis?

Three concepts are implicit in the questions posed. First, the principal features of protein synthesis are thought to be common to all bacteria irrespective of growth rate (Maaløe & Kjelgaard, 1966). Second, the concentrations of reactants strongly influence the rate of a chemical reaction. Third, the ratio of RNA : protein is thought to be directly proportional to the concentration of ribosomes (Bremer & Dennis, 1996).

Progress was made by first analysing definitive data for the macromolecular compositions of cells of Escherichia coli grown optimally in five media (doubling time, tD, 0·4–1·67 h). The macromolecular composition of M. bovis BCG was calculated from its chemical composition (Winder & Rooney, 1970). The relationships established for E. coli were then shown to apply to M. bovis BCG.

Theoretical section
Definition of slow growth.
Traditionally (see, for example, Wayne & Kubica, 1986) mycobacteria are classified as either fast-growing or slow-growing according to whether colonies appear on a solid medium within 5 days (fast-growers) or longer than 5 days (slow-growers). Bacteria growing optimally are defined according to µ, their specific growth rate (Fig. 1); slow growth spans the range µ<=0·14 h-1 (tD>=5 h), fast growth span the range µ>0·14 h-1<0·7 h-1 (tD 1–5 h). Faster-growing bacteria (µ>0·7 h-1 tD<1 h) are classified as ultra-fast growers.



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 1. Definition of slow, fast and ultra-fast growth. Each category is defined by the range of maximum growth rates shown by the labelled, double-headed arrows. All bacteria are capable of growing at less than maximum rates when conditions are less favourable; that is, all bacteria are capable of slow growth. When the generation time, tD, is more than 1 h the genome is usually replicated within the cell division cycle (see text); in other words, tD exceeds the period, C, required for DNA replication (tD>C) and each newborn cell has a single genome. Usually, when tD is less than 1 h (hatched bars) the replication of the genome takes place over more than one cell division cycle (C>tD) so that each newborn cell has more than one genome equivalent, this property defines ultra-fast growth. Mycobacterium chelonae (M. C. Nuñez & M. J. Garcia, Univerisidad Autonoma, Madrid, Spain. unpublished work); M. tuberculosis (Wayne, 1994); M. smegmatis (Gonzalez-y-Merchand et al., 1999); Pseudomonas aeruginosa (Yoshimura & Nikaido, 1982); Proteus vulgaris (Schaechter et al., 1962); Salmonella typhimurium (Schaechter et al., 1962); Escherichia coli (Koch, 1979).

 
A feature of ultra-fast growth is that genome replication spans more than one cell division cycle (Cooper & Helmstetter, 1968; Helmstetter & Cooper, 1968) giving rise to newborn cells that have more than one genome equivalent per cell. In contrast, newborn cells of fast- and slow-growing bacteria, including mycobacteria (Hiriyanna & Ramakrishnan, 1986), are thought to have a single genome that is replicated within the cell division cycle. When growth conditions are less favourable (µ<0·7 h-1), ultra-fast-growers resemble fast-growers and slow-growers, with genome replication being started and completed within the cell division cycle (Fig. 1).

Definitions, axioms and assumptions.
The analysis presented below concerns exponentially growing cell cultures of E. coli for which it is known that the conditions of growth govern both the specific growth rate µ and macromolecular composition (DNA : protein : RNA). Specifically, different growth rates correlate with different macromolecular compositions of cell cultures (see, for example, Bremer & Dennis, 1996). In contrast, when the specific growth rate was 0·2 h-1 or less, it was noted that the macromolecular content changed very little. Cultures had minimal contents of protein and RNA as judged by the DNA : protein and DNA : RNA ratios (Jacobsen, 1974 cited by Ingraham et al., 1983). For cells of minimal protein and RNA contents, the growth rates are designated µmin; the subscript min denotes that protein and RNA contents are minimal. These cells are thought to have an excess of ribosomes over the apparent demand for protein synthesis, as was shown for a Vibrio sp. by Flärdh et al. (1992). Cultures characterized by µmin are not included in the analysis. However, the concept of minimal cells could be relevant to dormant mycobacterial cells. (Symbols are defined in Table 1).


View this table:
[in this window]
[in a new window]
 
Table 1. Definitions of symbols

 
At any time, t, during the exponential growth of a culture the mass of protein, pt, is given by equation (1) where p0 is the mass of protein at t=0.

The mass of protein, pt, is the product of the number of cells, nt, present at time t and the average mass of protein mp(av) per cell. Although the number of cells per culture increases exponentially their age distribution remains unchanged (Powell, 1956). A cell's age, a, ranges from a=0 (a newborn cell) to a=1 (a cell about to divide), where a=t/tD; t=0 for newborn cells and t=tD for cells about to divide. Hence, the average amount of protein per cell is given by equation (2).

where na is the number of cells aged a; na=0 is the number of cells aged a=0; mp(a) is the amount of protein per cell aged a. Evaluation of equation (2) reveals (Ingraham et al., 1983) that mp(av) is equal to mp(a=0)/ln2 [1·4 mp(a=0)] where mp(a=0) is the mass of protein per newborn cell. Other average cell properties are similarly defined; for example, v(av) (fl or µm3), average dry cell mass mdc(av), average dry cell mass RNA mRNA(av) and average dry cell mass DNA mDNA(av). In each case, mass is measured in femtograms (fg). Each of these properties, which serve to define an ‘average’ cell, has a constant value for a particular set of growth conditions because the age distribution of cells remains unchanged throughout exponential growth.

A culture, which at time t comprises nt cells, accumulates protein at an instantaneous rate specified by equation (3), which is the differential of equation (1).

Substitution for pt=nt{bullet}mp(av) on the right-hand side of equation (3) leads to equation (4)

Rearrangement of equation (4) leads to the identities shown in equation (5), which also defines {omega}p(av), the specific protein synthesis rate (fg protein synthesized per cell per hour).

In other words, dpt/dt is equal to the product of nt and {omega}p(av).

It is thought that the amount of protein per cell increases exponentially over the range t=0 for a newborn cell to t=tD when the cells divide (Cooper, 1988). Hence, mp(t), the amount of protein per cell at time t, is given by equation (6) where mp(t=0) is the amount of protein per newborn cell.

Over the same period (t=0 to t=tD) the specific protein synthesis rate [{omega}p(t)] at time t is given by equation (7).

When mp(t)=mp(av), equation (7) may be written as equation (8).

Thus, equations (3) and (8) are related through nt [equation (5)], which further illustrates the notion that the properties of a cell culture are determined by the properties of average cells. Thus, both mp(av) and {omega}p(av) are parameters that usefully describe the exponential growth of a cell culture.

Macromolecular composition of E. coli as a function of the specific growth rate.
For cells growing at a rate in the range 0·42–1·73 doublings h-1 (Table 2), the mass of protein [mp(av)] and dry cell mass [mdc(av)] increase concomitantly with growth rate; that is, mp(av) appears to be in constant proportion, {beta}m, of mdc(av). The average cell volume [v(av)] will be directly proportional to mdc(av) and hence, to mp(av) if the fraction of cell mass due to water, {beta}aq, and the buoyant density, {rho}, are independent of growth rate. A value for {beta}aq of 0·7 was established for E. coli growing optimally in glucose minimal medium (Ingraham et al., 1983); this value appears to be accepted for E. coli cells in general. The buoyant density, {rho}, of E. coli was found to lie within the range 1·09±0·015 g cell mass (ml cell volume)-1 [or pg fl-1 (or µm3)] for cells with growth rates of 0·4–3·0 h-1 and to remain constant during the growth cycle (Woldringh et al., 1981); and was found to be independent of the composition of the growth medium below 1000 mosM (Baldwin et al., 1994). Thus, as a first approximation, {rho} may be regarded as a constant. The parameters {rho}, mdc(av) (or mp(av)/{beta}m) and v(av) are related by equations (9a) and (9b).



View this table:
[in this window]
[in a new window]
 
Table 2. Properties of E. coli B/r growing exponentially at different growth rates at 37 °C

 
The concentration of protein [mp(av)/v(av)], which was estimated to be 0·16 pg fl-1 (or µm3), is thought to be not only constant during the growth cycle but also to be independent of growth rate (see also Bremer & Dennis, 1996); that is, both {beta}m and {beta}aq are also thought to be independent of growth rate.

Whereas the concentration of protein appears to be independent of the specific growth rate, the mass of protein per cell is not. The relationship between mp(av) and µ [equation (10)] is illustrated in Fig. 2.

where {alpha}=4·1x10-3 (fg protein)-1 h-1.



View larger version (12K):
[in this window]
[in a new window]
 
Fig. 2. Relationship between specific growth rate, µ, average protein content, trun -1mp(av) and average cell volume, v(av). The average cell volume was calculated by means of equation (9b), Table 2. {circ}, Data for E. coli (Table 2); {blacksquare}, datum for M. bovis BCG (Table 3); solid line, empirical plot; broken line, linear plot for E. coli (curve I) and M. bovis BCG (curve II).

 
The plot of RNA : protein ratios [mRNAav/mp(av)] against µ (Fig. 3) has features in common with the plot of mp(av) against µ (Fig. 2). A linear correlation [equation (11)] of µ with mRNA(av)/mp(av) was found (see data of Table 2).

where b=4·37 h-1. The intercept on the abscissa raises the possibility that there is a limiting value of mRNAav/mp(av){approx}0·08 for viable cells.



View larger version (14K):
[in this window]
[in a new window]
 
Fig. 3. Specific growth rate, µ, is directly proportional to the RNA : protein ratio. For average ribosome concentrations see Tables 2 and 3. {circ}, Data for E. coli; {blacksquare}, datum for M. bovis BCG; broken line, extrapolation of curve for µ against mRNA(av)/mp(av).

 
Whereas the RNA : protein ratio was found to be directly proportional to the specific growth rate when cells were grown optimally, the specific protein synthesis rate {omega}p(av) [equation (5)] was not (Fig. 4). Both the specific protein synthesis rate and hence, the associated rate of consumption of energy were found to increase as the specific growth rate increased. The dependence of the specific protein synthesis rate on the RNA : protein ratio (Fig. 5a) was similar to the plot shown in Fig. 4; the similarities in the two graphs are to be expected since µ and mRNAav/mp(av) are linearly related [equation (11)]. The specific protein synthesis rate was found to increase approximately eight-fold for a doubling of the RNA : protein ratio (Fig. 5a), which suggests that {omega}p(av) is proportional to the third power of the ratio mRNA(av)/mp(av). This conclusion is supported by the plot of log [{omega}p(av)] against log [mRNA(av)/mp(av)], which was found to have a slope of 3·2 (Fig. 5b), in accord with a third power relationship. The plot of the specific protein synthesis rate against the third power of the RNA : protein ratio (Fig. 5c) was found to be linear [equation (12)] with a gradient {theta}=7·5x103 fg protein synthesized h-1.



View larger version (16K):
[in this window]
[in a new window]
 
Fig. 4. Correlations between {omega}p(av), the concomitant rate of energy consumption and specific growth rate. The rate of energy consumption needed to maintain the rate of protein biosynthesis was calculated on the basis of the assumption that 4·2 high energy phosphate bonds (~P) are required for the formation of one peptide bond (Ingraham et al., 1983). {circ}, Data for E. coli calculated by means of equation (4) using the information given in Table 2; {blacksquare}, datum for M. bovis (Table 3).

 


View larger version (13K):
[in this window]
[in a new window]
 
Fig. 5. {omega}p(av) Is a function of the RNA : protein ratio.{circ}, Data for E. coli, Table 2; {blacksquare}, datum for M. bovis BCG, Table 3. (a) {omega}p(av)=mp(av){bullet}µ as a function of mRNA(av)/mp(av); (b) log [{omega}p(av)] as a function of log [mRNA(av)/mp(av)]; (c) {omega}p(av) is directly proportional to the third power of the RNA : protein ratio.

 
Growth and macromolecular composition of M. tuberculosis.
The era of the study of the macromolecular composition of bacteria (see, for example, Maaløe & Kjeldgaard, 1966) was centred around 1960–1970 and such studies are now unfashionable. The only study of the chemical composition of a mycobacterium was carried out by Winder & Rooney (1970), who reported a careful and comprehensive study of Mycobacterium bovis BCG, a member of the M. tuberculosis complex that is very closely related to M. tuberculosis (Brosch et al., 2000). The mycobacterium was grown in a standard medium and the generation time was approximately 24 h. The results presented in Table 3 were obtained from the data of Winder & Rooney (1970) by making a number of simple calculations based on the definitions of the units used, and current data for the size of the genome, the average number of genomes per cell (footnote * of Table 3) and the size of the rRNA moiety of ribosomes (Cole et al., 1998). The numbers of cells per millilitre, mDNA(av), mRNA(av), mp(av) and v(av) were calculated. Thus, Table 3 summarizes all the information that is available for the macromolecular composition of a member of the M. tuberculosis complex in particular and mycobacteria in general.


View this table:
[in this window]
[in a new window]
 
Table 3. Properties of M. bovis BCG (Glaxo) during growth (at 37 °C) in batch cultures

 
The limited data for M. bovis BCG are presented in Figs 2–5. In Fig. 2, a plot of µ against mp(av), the datum for M. bovis BCG (curve II) does not conform with the data for E. coli (curve I). More data were needed for M. bovis in order to explain why this should be the case. In contrast, in Figs 3–5 data for the mycobacterium were found to form a smooth curve with data for E. coli, suggesting that the associated equations (11) and (12) may also apply to M. bovis BCG growing optimally. This was confirmed, as may be seen from the following example. Very similar protein contents [mp(av){approx}156 fg] were reported for E. coli (tD=1 h) grown in a glycerol medium (Table 2) and for M. bovis BCG (tD=24 h) growing exponentially (Table 3). In contrast, the corresponding RNA : protein ratios of E. coli (Table 2) and M. bovis BCG (Table 3) were found to be 0·25 and 0·085, respectively, almost a three-fold difference. Substitution of the RNA : protein ratios in equation (11) yielded values of µ of 0·73 and 0·025 h-1, respectively, for E. coli and M. bovis BCG; the observed values were 0·69 and 0·029 h-1, respectively. The RNA : protein ratios were also substituted in equation (12) in order to calculate the specific protein synthesis rate [{omega}p(av)]. The calculated specific protein synthesis rates were 117 and 4·6 fg protein synthesized per cell per hour, respectively, for E. coli and M. bovis BCG; the observed values, calculated by means of equation (5), were 108 and 4·5 fg protein synthesized per cell per hour, respectively. Thus, equations (11) and (12) which were derived for E. coli apply also to M. bovis BCG. The ability of these equations to describe both E. coli and M. bovis BCG emphasizes that the concentration of RNA which is measured by RNA : protein ratio is an informative cell parameter.

The macromolecular composition of stationary phase cells (day 8 of Table 3) provides a guide to the properties of minimal cells of the mycobacterium. When appropriate conversion factors (Table 3) were applied to data for M. bovis BCG as to data for E. coli, the following picture emerged. Minimal cells of E. coli were estimated to have an average of 6200 ribosomes, a concentration of 8200 ribosomes per femtolitre of cell volume. In contrast, minimal cells of M. bovis BCG were estimated to have an average of 2200ribosomes per cell, a concentration of approx. 4500 ribosomes per femtolitre of cell volume, which is close to the limit inferred for viable cells from the intercept on the RNA : protein axis in Fig. 3.


   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
DISCUSSION
REFERENCES
 
This study contributes towards establishing a theoretical framework for further understanding the growth of M. tuberculosis in two ways. First, an upper limit was placed on the protein and RNA contents for minimal cells of M. bovis BCG. Second, the correlations observed between specific growth rates and macromolecular compositions of an ultra-fast-growing organism (E. coli) can be extrapolated to a very-slow-growing organism (M. tuberculosis). The RNA : protein ratio was found to be a key property of the bacterial cells studied. For cells growing optimally, the specific growth rate was found to be linearly related to the RNA : protein ratio [equation (11)] and the specific protein synthesis rate was found to be directly proportional to the third power of the RNA : protein ratio [equation (12)].

The significance of the correlations specified in equations (11) and (12) lies in the relationship between the RNA : protein ratio and ribosome concentration. A simple proportionality between the RNA : protein ratio and ribosome concentration requires mRNA(av) to be directly proportional to the average number of ribosomes, which seems likely because rRNA and tRNA are thought to account for 98 % or so of the RNA fraction. Furthermore, mp(av) is required to be directly proportional to cell volume, or, more specifically, to the volume in which protein synthesis takes place. In other words, the concentration of protein is required to have a constant value that is independent of growth rate. The available data for E. coli (Table 2) indicate that protein constitutes a fixed proportion, {beta}m, of dry cell mass that is independent of the specific growth rate; there are no data reported for M. tuberculosis.

On the basis of the assumption that the RNA : protein ratio is proportional to ribosome concentration, equation (12) may be revised to relate the specific protein synthesis rate directly to the third power of the ribosome concentration. This proposed relationship can be shown to be in accord with our knowledge of mRNA-directed protein synthesis. The exponential increases in mass, volume, RNA content and protein content during growth ensure that the concentrations of RNA and protein are kept constant and require that the amounts of substrates, enzymes and products increase concomitantly to maintain constant concentrations. The balance between components of the machinery for protein synthesis is maintained if the proportions of individual components such as tRNA and protein factors such as elongation factor EF-Tu, etc., are kept in balance with the number of ribosomes. For example, 9·26–9·29 tRNAmolecules per ribosome and five to seven copies of EF-Tu were reported for E. coli independently of growth rate (Table 2).

Protein synthesis involves many steps and many components (for review see Al-Karadagh et al., 2000). The rate-limiting step in peptide bond formation is the interaction of a ternary complex (tc) composed of aminoacyl tRNA (aatRNA), elongation factor EF-Tu and GTP with the A-site of the ribosome (Pape et al., 1998). The overall reaction is given by equation (13).

The reaction involves ribosomes, aatRNA and EF-Tu. The concentrations of the latter two components may be directly related to ribosome concentration (Table 2), suggesting that the rate of peptide bond formation is a function of the third power of the ribosome concentration. Two further factors apply to protein synthesis during cell growth; namely, the concentration of protein remains constant, whereas the cell volume increases exponentially.

The correlation observed between specific growth rates and macromolecular compositions of both an ultra-fast and a very-slow grower reflects the importance of both protein concentration and the rate of protein biosynthesis in cell growth. For example, the protein content reflects cell volume, the RNA : DNA ratio reflects the number of ribosomes per cell and the RNA : protein ratio reflects the concentration of ribosomes and hence, the specific protein synthesis rate. Finally, the ability to measure the ratios DNA : protein : RNA simply and accurately, using a minimal number of cells, would be advantageous; flow cytometry may offer this potential (Diaper & Edwards, 1994: Turner et al., 2000).


   ACKNOWLEDGEMENTS
 
I thank Simon A. Cox for his help in the preparation of this manuscript, my colleague Dr I. D. J. Burdett for his expert advice and constant encouragement and my colleague Dr J. Ecclestone for advice on nomenclature. I thank Dr R. Rosenberger of the National Institute for Biological Standards and Control, South Mimms, UK, for his interest and Professor C. Ratledge of the Department of Biological Sciences, The University of Hull, UK, for his invaluable advice on mycobacterial metabolism.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
DISCUSSION
REFERENCES
 
Al-Karadagh, S., Kristensen, O. & Liljas, A. (2000). A decade of progress in understanding the structural basis of protein synthesis. Prog Biophys Mol Biol 73, 167–193.[CrossRef][Medline]

Armstrong, J. A. & D'Arcy-Hart, P. (1971). Response of cultured macrophages to Mycobacterium tuberculosis, with observations on fusion of lysosomes with phagosomes. J Exp Med 134, 713–740.

Arnstein, H. R. V. & Cox, R. A. (1992). Protein Biosynthesis: in Focus. Oxford: Oxford University Press.

Baldwin, W. W., Hirkish, M. A. & Koch, A. L. (1994). A change in a single gene of Salmonella typhimurium can dramatically change its buoyant density. J Bacteriol 176, 5001–5004.[Abstract]

Bremer, H. & Dennis, P. P. (1996). Modulation of chemical composition and other parameters of the cell growth rate. In Escherichia coli and Salmonella: Cellular and Molecular Biology, 2nd edn, pp. 1553–1568. Editor by F. C. Neidhardt and others. Washington, DC: American Society for Microbiology.

Brosch, R., Gordon, S. V., Eiglmeier, K., Garnier, T., Tekaia, F., Yeramian, E. & Cole, S. T. (2000). Genomics, biology, and evolution of the Mycobacterium tuberculosis complex. In Molecular Genetics of Mycobacteria, pp. 19–36. Edited by G. F. Hatfull & W. R. Jacobs, Jr. Washington, DC: American Society for Microbiology.

Butcher, P. O., Sole, K. M. & Mangan, J. A. (1999). RNA extraction. In Molecular Mycobacteriology: Techniques and Clinical Applications, pp. 385–350. Edited by R. A. Ollar & N. O. Connell. New York: Marcel Dekker.

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.[CrossRef][Medline]

Colston, M. J. & Cox, R. A. (1999). Mycobacterial growth and dormancy. In Mycobacteria: Molecular Biology and Virulence, pp. 198–219. Edited by C. Ratledge & J. Dale. Oxford, UK: Blackwell Science.

Cooper, S. (1988). Rate and topography of cell wall synthesis during the division cycle of Salmonella typhimurium. J Bacteriol 170, 422–430.[Medline]

Cooper, S. & Helmstetter, C. E. (1968). Chromosome replication and the division cycle of Escherichia coli B/r. J Mol Biol 31, 519–540.[Medline]

Diaper, J. P. & Edwards, C. (1994). Survival of Staphylococcus aureus in lakewater monitored by flow cytometry. Microbiology 140, 35–42.[Abstract]

Ferrari, G., Langen, H., Naito, M. & Pieters, J. (1999). A coat protein on phagosomes involved in the intracellular survival of mycobacteria. Cell 97, 435–447.[Medline]

Flärdh, K., Cohen, P. S. & Kellenberg, S. (1992). Ribosomes exist in large excess over the apparent demand for protein synthesis during carbon starvation in marine Vibrio sp. strain CCUG 15956. J Bacteriol 174, 6780–6788.[Abstract]

Gonzalez-y-Merchand, J. A., Colston, M. J. & Cox, R. A. (1999). Effects of growth conditions on expression of mycobacterial murA and tyrS genes and contributions of their transcripts to precursor rRNA synthesis. J Bacteriol 181, 4617–4627.[Abstract/Free Full Text]

Helmstetter, C. E. & Cooper, S. (1968). DNA synthesis during the division cycle of rapidly growing E. coli B/r. J Mol Biol 31, 507–518.[Medline]

Hiriyanna, K. T. & Ramakrishnan, T. (1986). Deoxyribonucleic acid replication time in Mycobacterium tuberculosis H37 Rv. Arch Microbiol 144, 105–109.[Medline]

Ingraham, J. L., Maaløe, O. & Neidhardt, F. C. (1983). Growth of the Bacterial Cell. Sunderland, MA: Sinauer Associates.

Jacobs, W. R., Jr (2000). Mycobacterium tuberculosis: a once genetically intractable organism. In Molecular Genetics of Mycobacteria, pp. 1–16. Edited by G. F. Hatfull & W. R. Jacobs, Jr. Washington, DC: American Society for Microbiology.

Jacobsen, H. (1974). PhD thesis. University of Copenhagen.

Koch, A. L. (1979). Microbial growth in low concentrations of nutrients. In Strategies of Microbial Life in Extreme Environments, pp. 261–269. Edited by M. Shilo. Weinheim: Verlag Chemie.

Maaløe, O. & Kjeldgaard, N. O. (1966). Control of Macromolecular Synthesis: a Study of DNA, RNA and Protein Synthesis in Bacteria. New York: W. A. Benjamin.

Pape, T., Wintermeyer, W. & Rodnina, M. V. (1998). Complete kinetic mechanism of elongation factor Tu-dependent binding of aminoacyl-tRNA to the A-site of the E. coli ribosome. EMBO J 17, 7490–7497.[Abstract/Free Full Text]

Pedersen, S., Bloch, P. L., Reeh, S. & Neidhardt, F. C. (1978). Patterns of protein synthesis in E. coli: a catalog of the amount of 140 individual proteins at different growth rates. Cell 14, 179–190.[Medline]

Powell, E. O. (1956). Growth rate and generation time of bacteria, with special reference to continuous culture. J Gen Microbiol 15, 492–511.[Medline]

Ratledge, C. (1982). Nutrition, growth and metabolism. In Biology of the Mycobacteria, pp. 185–271. Edited by C. Ratledge & J. L. Stanford. London: Academic Press.

Schaechter, M., Williamson, J. P., Hood, J. R., Jr & Koch, A. L. (1962). Growth, cell and nuclear divisions in some bacteria. J Gen Microbiol 29, 421–434.[Medline]

Shepard, C. C. (1960). The experimental disease that follows the injection of human leprosy bacilli into the footpads of mice. J Exp Med 112, 445–454.

Turner, K., Porter, J., Pickup, R. & Edwards, C. (2000). Changes in viability and macromolecular content of long-term batch cultures of Salmonella typhimurium measured by flow cytometry. J Appl Microbiol 89, 90–99.[CrossRef][Medline]

Wayne, L. G. (1994). Cultivation of Mycobacterium tuberculosis for research purposes. In Tuberculosis: Pathogenesis, Protection and Control, pp. 73–83. Edited by B. Bloom. Washington, DC: American Society for Microbiology.

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

Wayne, L. G. & Kubica, G. P. (1986). The mycobacteria. In Bergey's Manual of Systematic Bacteriology, vol. 2, pp. 1435–1457. Edited by P. H. A., Sneath, N. S., Mair, M. E. Sharpe & J. G. Holt. Baltimore: Williams & Wilkins.

Weber, I., Fritz, C., Ruttkowski, S., Kreft, A. & Bange, F. C. (2000). Anaerobic nitrate reductase (narGHJI) activity of Mycobacterium bovis BCG in vitro and its contribution to virulence in immunodeficient mice. Mol Microbiol 35, 1017–1025.[CrossRef][Medline]

Wheeler, P. R. & Ratledge, C. (1994). Metabolism of Mycobacterium tuberculosis. In Tuberculosis; Pathogenesis, Protection and Control, pp. 353–385. Edited by B. R. Bloom. Washington, DC: American Society for Microbiology.

Winder, F. G. & Rooney, S. A. (1970). Effects of nitrogenous components of the medium on the carbohydrate and nucleic acid content of Mycobacterium tuberculosis BCG. J Gen Microbiol 63, 29–39.[Medline]

Woldringh, C. L., Binnerts, J. S. & Mans, A. (1981). Variation in Escherichia coli buoyant density measured in percoll gradients. J Bacteriol 148, 58–63.[Medline]

Yoshimura, F. & Nikaido, H. (1982). Permeability of Pseudomonas aeruginosa outer membrane to hydrophobic solutes. J Bacteriol 152, 636–642.[Medline]

Received 3 April 2002; revised 21 October 2002; accepted 12 November 2002.