Color profile: Disabled Composite Default screen 351 Bacterial fitness and plasmid loss: the importance of culture conditions and plasmid size M. Alex Smith and Michael J. Bidochka Abstract: Several pBluescript-derived plasmids of various sizes were constructed to study the effects of multicopy plasmid size on bacterial fitness and plasmid loss. Transformed and untransformed bacterial clones were grown in media with or without ampicillin. Bacterial fitness (measured by growth rate), plasmid presence or absence, and plasmid copy number were assessed during successive subculturings. In selective media (minimal medium or Luria Broth plus ampicillin), the clone transformed with the largest plasmid (pBluescript with a 9000-bp insert) had a significantly longer lag phase than all other clones. In nonselective media the rate of plasmid loss during successive subculturings was greatest in the clone with the largest insert. The clone with the largest insert displayed a lower plasmid copy number than clones with a small insert or no insert at all. Plasmid loss in the form of segregational instability and plasmid copy number reduction in nonselective environments are important to the understanding of the evolution of the bacteria–plasmid associations and the appreciation of the potential for altering the genetic properties of a clone maintained or subcultured on a standard medium. Key words: pBluescript, plasmid, stress, fitness, starvation. Résumé : Nous avons construit quelques plasmides de diverses tailles dérivés de pBluescript en vue d’étudier les effets de la taille d’un plasmide en plusieurs copies sur la compétence de la bactérie et la perte de plasmides. Des clones de bactéries transformées et non-transformées ont été cultivés dans des milieux avec ou sans ampicilline. La compétence des bactéries (évaluée par le taux de croissance), la présence ou l’absence de plasmides et le nombre de copies d’un plasmide ont été mesurés lors de repiquages successifs. Dans les milieux sélectifs (milieux minimaux de culture ou bouillon de culture Luria avec ampicilline), le clone transformé avec le plus gros plasmide (pBluescript plus une insertion de 9000 pb) affichait une phase de latence significativement plus longue que tous les autres clones. Dans des milieux non-sélectifs, le taux le plus élevé de perte plasmidique lors des repiquages successifs a été observé avec le clone portant la plus lourde séquence d’insertion. Le clone contenant la plus lourde séquence d’insertion présentait un nombre de copies de plasmides inférieur à celui de clones ayant une petite séquence d’insertion ou aucune insertion du tout. La perte de plasmides à la période d’instabilité de division et la diminution du nombre de copies plasmidiques dans des environnements non-sélectifs sont des facteurs importants pour comprendre l’évolution des associations bactérie–plasmide et pour évaluer la capacité d’altérer les propriétés génétiques d’un clone maintenu ou sous-cultivé sur un milieu standard. Mots clefs : pBluescript, plasmide, stress, compétence, jeune. Introduction Experimental studies in molecular biology with transformed Escherichia coli assume that the recombinant plasmids are stable. However, differing growth and storage conditions can evoke changes in the transformed population. Environmental conditions can influence segregational instability or plasmid deamplification (Gerdes et al. 1986; Rohde 1995; GuerrierTakada et al. 1997). Furthermore, bacteria containing plasmids of different sizes or copy number may have differing fitnesses in different environments (Kues and Stahl 1989; Valenzuela et al. 1996). Effects of plasmid maintenance on the fitness of Received June 26, 1997. Revision received January 21, 1998. Accepted January 27, 1998. M.A. Smith1 and M.J. Bidochka. Biology Department, Trent University, Peterborough, ON K9J 7B8, Canada. 1 Author to whom all correspondence should be addressed ([email protected]). Can. J. Microbiol. 44: 351–355 (1998) W98-020.CHP Tue Jun 02 09:10:28 1998 the bacterial host have been documented in several clones of E. coli (Godwin and Slater 1979; Seo and Bailey 1985; Boe et al. 1987; de Taxis du Poet et al. 1987; Lenski and Bouma 1987; Caldwell et al. 1989; Modi and Adams 1991; Rhee et al. 1994) and Bacillus subtilis (Bron et al. 1988; Leonhardt and Alonso 1991). pBluescript (Stratagene Cloning Systems, La Jolla, Calif.) is a commercially available multicopy plasmid used in many experimental studies as a cloning vector. This plasmid has, among other features, a gene encoding ampicillin resistance and a multiple cloning site (MCS). Escherichia coli transformed with pBluescript are conventionally grown in a selective media containing ampicillin to maintain the plasmid. In this investigation we asked (i) whether plasmid size (pBluescript plus an insert in the MCS) affected bacterial fitness in a selective environment and (ii) whether larger plasmids are lost and (or) deamplified (copy number reduced) at a greater rate in nonselective environments compared with pBluescript itself. Previous work has indicated that plasmid copy number and © 1998 NRC Canada Color profile: Disabled Composite Default screen 352 Can. J. Microbiol. Vol. 44, 1998 Table 1. Clones used in this study: designated nomenclature, plasmids contained, and overall plasmid size. culated with the following equation: %CFU with plasmid = CFU with ampicillin / CFU without ampicillin × 100. Clones used Plasmid contained XL1-Blue XL1-BB MB-700 MB-9000 None pBluescript pBluescript and 700 bp pBluescript and 9000 bp Effects of starvation on plasmid copy number Three clones that resumed growth in starvation situations were checked for copy number differences. Copy number was determined using two methods and three calculations. In the first method, cultures from each of the clones of MB-9000, which resumed growth in the second experiment (MB-9000, MB-9000-2, MB-9000-4), pBluescript, MB-700, and MB-9000 were standardized using a haemocytometer with respect to the number of bacteria per millilitre of culture. Plasmids were isolated using methods described in Berghammer and Auer (1993). The 1-mL samples (equal number of bacteria, balanced to 1 mL total volume with Tris–EDTA buffer) were centrifuged at 13 200 rpm for 5 min. A Gene-Quant II spectrophotometer (Pharmacia, Biotech) was used to quantify DNA in solution. Ten readings were taken for each sample. The second method was the “Saran Wrap” method for DNA estimation (Sambrook et al. 1989). In this method, plasmid DNA was added to Tris–EDTA buffer and ethidium bromide (0.2 mg/mL) and the suspensions were serially diluted. The dilutions were compared with dilutions of a known standard, in this case, pBluescript (1 mg/mL). A photo was taken of these 10-mL dilution drops under shortwave UV light and the irradiance of each drop was compared with the most similar droplet from the known standard solution. The estimations of plasmid DNA (ng/mL) from either method were then employed to determine plasmid copy number as follows: Plasmid size (kb) 0.0 2.961 3.661 11.961 segregation can influence bacterial host fitness (Godwin and Slater 1979; Moser and Campbell 1983; Bron and Luxen 1985; Hakkaart et al. 1985; Griffiths et al. 1990). Much of that work was conducted in rich (Luria–Bertani) media with antibiotics present to select for the phenotypic marker of each individual plasmid. Here we report significant impacts of plasmid size and copy number on bacterial host fitness in starvation media with, and without, the presence of antibiotics. Bacterial growth rate is reduced when maintaining larger plasmids. The metabolic stress of maintaining plasmids in nonselective environments may be alleviated through plasmid segregational instability and (or) plasmid deamplification. Methods and materials Plasmid constructs and clones Genomic DNA from the fungus Metarhizium anisopliae was digested with EcoRV and ligated with the plasmid pBluescript and transformed into E. coli XL1Blue using standard techniques (Sambrook et al. 1989). Several clones were isolated using blue–white selection and standard techniques. Transformants had fungal inserts of 700 and 9000 bp. The DNA was sequenced and contained no open reading frames. Four clones (Table 1) were used in experimental studies: a control with no plasmid (XL1Blue), only the plasmid pBluescript (XL1-BB, 2.961 kb), pBluescript with a 700-bp insert (MB-700, 3.661 kb), and pBluescript with a 9000-bp insert (MB-9000, 11.961 kb). Growth rates Bacterial growth rates were monitored by measuring the turbidity in culture at 610 nm or by monitoring the colony-forming units (CFU) after plating culture aliquots. The optical densities at 610 nm were recorded at 3-h intervals. At 6-h intervals, a 1:10 000 dilution was made (in phosphate-buffered saline and plated onto Luria–Bertani (LB; 10 g tryptone, 5 g yeast extract, 0.5 g NaCl, and 2 mL of 1 M NaOH per litre of distilled water) agar plates containing 75 µg ampicillin/mL. Plates were incubated for approximately 18 h at 37°C and the CFU on each plate for each clone were counted. The experiments were repeated five times. Effects of plasmid sizes on bacteria in a starvation situation Three clones were used: control (no insert; XL1Blue), 700-bp insert (MB-700), and 9000-bp insert (MB-9000). All clones were grown in minimal media (M9) broth (6.0 g Na2HPO4, 3.0 g KH2PO4, 5 g NaCl, 1.0 g NH4Cl plus 10 mL of 0.01 M CaCl2 after autoclaving.) This broth constitutes a starvation situation as only a limited supply of glucose and salts are available. Supplemental amino acids were not added to further enhance the creation of a starvation environment. Cultures were grown until (and if) an optical density (OD) of 0.6 at 610 nm was reached. This value represents a mid-log phase of growth. Once an OD610 of 0.6 had been reached that culture was subcultured again into M9 broth. At every third subculturing a 1:10 000 dilution was plated onto LB agar and grown at 37°C for 18 h. The resultant bacterial colonies were replica plated onto LB agar with ampicillin (Lederberg and Lederberg 1952). The CFU from each plate were compared and the percentage plasmid retention was cal- [1] 1 mole plasmids 6.02×1023 plasmids = × bacteria single plasmid mass (ng) 1 mole amount of plasmid DNA (ng) no. of bacteria × × 1 mL of solution 1 mL of culture Statistical comparisons of bacterial growth profiles The statistical significance of the OD of the growth curves for each clone was determined using a Tukey multiple comparison test performed on the length of time each clone took to exceed an OD610 of 0.1. (An OD610 of 0.1 was chosen to represent the end of the lag phase of the bacterial growth profile.) The Tukey test was also employed to determine interclone differences in exponential growth rates. Results Effects of plasmid sizes on bacterial growth profiles In M9 broth, the following lag phases and exponential growth rates (µ) were observed for the various clones: lag phase = 12.6 h and µ = 0.21 for XL1Blue, lag phase = 12.6 h and µ = 0.20 for XL1-BB, lag phase = 12.9 h and µ = 0.22 for MB-700, and lag phase = 19.8 h and µ = 0.2 for MB-9000. The lag phases and the exponential growth rates for XL1Blue, XL1BB, and MB-700 were statistically similar (Fig. 1). However, the lag phase of MB-9000 was statistically longer than the lag phase of XL1Blue, XL1-BB, and MB-700 (F = 34.87, n = 10, df = 3, p < 0.005) (Fig. 1). The exponential growth rate of MB-9000 was similar to the other three clones (F = 2.00, n = 8, df = 3, p > 0.10) (Fig. 1). In LB broth the following lag phases and expontential growth rates were observed for the various clones: lag phase = 1.5 h and µ = 0.31 for XL1Blue, lag phase = 2.1 h and µ = 0.35 for XL1Blue BB, lag phase = 2.5 h and µ = 0.35 for MB-700, and lag phase = 5.1 h and µ = 0.32 MB-9000. The lag phases and exponential growth rates of XL1Blue, XL1BB, and MB-700 were statistically similar (p < 0.005) (Fig. 1). However, as in M9 media, the lag phase of MB-9000 was © 1998 NRC Canada W98-020.CHP Tue Jun 02 09:10:30 1998 Color profile: Disabled Composite Default screen 353 Smith and Bidochka Fig. 1. Growth of four clones of E. coli as measured by the change in OD610 in M9 containing 75 µg ampicillin/mL (closed symbols) or LB containing 75 µg ampicillin/mL (open symbols). An OD610 of 0.1 represented the end of the lag phase. d and s, XL1-Blue; r and e, XL1-BB; j and h, MB-700; m and n, MB-9000. Absorbance at 610 nm 2 1.5 1 0.5 0 0 3 6 9 12 15 Culture A ge (h) significantly longer (p < 0.005) than the lag phase of the other clones (Fig. 1). Similar between clone patterns were exhibited in M9 and LB broth. A 10-mL sample of each culture was diluted 10 000 fold, and plated on LB agar every 6 h to determine the number of CFU that corresponded with the OD readings. Patterns similar to those of the OD readings were observed in the CFU comparison (data not shown). Plasmid size and segregation (continuous subculture on M9 with ampicillin) The OD did not change for all clones (XL1Blue, XL1-BB, MB-700, and MB-9000) after reinoculating the culture grown in M9 onto M9. XL1Blue (with no plasmid) subcultures did reach the an OD610 of 0.6 but only after 5 days of growth. This was then subcultured five times during the 500-h duration of this experiment. Two replicates of clone MB-9000 did eventually reach an OD610 of 0.6 after approximately 3 days of no apparent growth (i.e., OD did not change). Observations of colony morphology and colour (on LB and MacConkey’s agar) and a microscopic examination of these MB-9000 colonies indicated that they were not extraneous contaminants. No replicate of the clone MB-700 grew to reach an OD610 of 0.6 and therefore, no between clone statistical comparisons were possible, as all clones did not reach an OD610 of 0.6 (when subculturing would have occurred). For each of the two MB9000 replicates that did grow, three additional replica plate repetitions were made. The proportion of the colonies that contained the plasmid decreased drastically after the initial growth to an OD610 of 0.6 (Fig. 2). 18 21 24 Plasmid size and amplification (continuous subculture on M9 with ampicillin) Two different methods and two different equations for copy number determination produced different values for the plasmid copy number (Fig. 3). Amounts of plasmid DNA did not differ because of a differential ability to isolate DNA, since the amount of chromosomal DNA extracted from each sample was constant as observed by agarose gel electrophoresis. Two equations with the Gene Quant method gave similar results. Two patterns are evident in Fig. 3. There is a negative correlation between the insert size in a multicopy plasmid such as pBluescript and the number of plasmids per bacterium after successive subculture on nonselective media. There were no significant differences between the original MB-9000 clone and the MB-9000 clones that grew following multiple subculturings (500 h) in M9 starvation media (F = 0.02, p = 0.890). Discussion The principal findings of this study were (i) plasmid size was negatively correlated with plasmid copy number (MB-9000 copy number was less than either XL1-BB or MB-700, (ii) continual subculture in minimal media did not affect copy number in plasmids of any size tested, and (iii) any growth observed during continual subculture in minimal media without ampicillin was due to segregational loss of the plasmid (as demonstrated through replica-plating). It has been hypothesized that when bacterial cells are starved they retain any plasmids present but lose the enzymes required to detoxify the antibiotic (Griffiths et al. 1990). The work reported here indi© 1998 NRC Canada W98-020.CHP Tue Jun 02 09:10:32 1998 Color profile: Disabled Composite Default screen 354 Can. J. Microbiol. Vol. 44, 1998 Fig. 2. The proportion of colony MB-9000 that retained the plasmid through subculturing in M9 without ampicillin. Subculture intervals occurred over 500 h of growth in a starvation environment with no antibiotic present. Results are means of four replicates with standard error bars. 100 Fig. 3. Different methods of copy number determination using the “Saran Wrap” method (Sambrook et al. 1989) (h), a calculation utilized in Modi and Adams (1991) with data collected using Gene Quant ( ), and an equation determined here with data collected using Gene Quant (j). 600 u Copy Number % Colonies Retaining Plasmid 500 80 60 u 40 400 300 200 100 u 20 u u u u 0 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 0 2.96 3.66 11.96 Plasmid Size (kb) 11.96 (Post 500 h starvation) Subculture No. cates that when E. coli XL1-Blue cells are starved, the plasmid is lost and the rate of loss is positively correlated with the size of the plasmid. Maintenance of large plasmids causes a fitness reduction in the bacterial host in nutrient-limiting culture conditions. The maintenance of plasmids during subculture is important to studies in molecular biology where transformants are used and stored under different conditions. Plasmid maintenance is also important in describing the evolution of the bacteria–plasmid association. The segregational stability of the bacteria–plasmid association and potential changes in plasmid copy number must be considered when culturing or storing bacteria containing multicopy plasmids. We monitored bacterial fitness when E. coli was grown under the stress of subculture in minimal media. Although each culture had initially grown to stationary phase after a period of approximately 24 h, subculturing resulted in a decrease in fitness as measured by bacterial growth rate. To our knowledge, no reference of such a response to multiple subculturing in M9 minimal media by E. coli has been reported. Kolter et al. (1993) stated that when E. coli are starved, they can become less metabolically active, more resistant to stress, and smaller owing to cellular divisions with no increase in cell mass. The rapid decrease in growth rate observed when colonies were repetitively subcultured in M9 might be explained by such a reduction in size and metabolic activity. Lenski and Bouma (1987) concluded that any increase in the growth rate of bacterial transformants in selective media was due to selection against plasmid carriage. Directional selection of this type was observed in our experiments. Multicopy plasmids (such as pBluescript) are generally lost from cultures due to the imperfect segregation of plasmids into daughter cells at the time of cell division (Summers and Sherratt 1984; Lenski and Bouma 1987; Modi and Adams 1991). This is characterized as segregational instability. Structural instability is the rearrangement of plasmid DNA or changes in the number of plasmids per bacterium (copy number) (Bron et al. 1988). Therefore, any negative effect of the plasmid presence on bacterial fitness could be alleviated through the loss of plasmids at segregation (and subsequent enhanced growth of the plasmid-free cells) or a decrease in copy number. Analysis of plasmid and chromosomal DNA extracted from the various clones ruled out multimerization of the plasmids as a potential method by which plasmid copy number could be reduced. Seo and Bailey (1985) and de Taxis du Poet et al. (1987) reported that growth in a minimal media caused a sharp increase in plasmid copy number in E. coli. This contradicts the findings of Warnes and Stephenson (1986) who reported that 25 generations of culturing on selective media caused a subsequent decrease in plasmid copy number. Seo and Bailey (1985) found that maximum specific growth rates in Luria broth and in M9 media were reduced with increasing plasmid content. We found that a large plasmid size resulted in a significantly longer lag phase in either nutrient-limiting or nutrient-rich environments. Our experiments were conducted in a brief time period (24, 48, or 500 h). This may not have been sufficient time to implicate a genetic change that would reduce the negative effects of plasmid carriage in the bacterial chromosome. Hakkaart et al. (1985) reported that the effects of copy number are dominant over the effects of plasmid size relative to plasmid maintenance. In this work plasmids endured segregational loss correlated with the size of plasmid through consecutive subculturings in starvation media, while copy number remained constant. Such a finding indicates that with the E. coli XL1Blue – pBluescript association the effects of plasmid size appear to dominate the effects of copy number. Understanding plasmid maintenance in a starvation environment is important when considering the utility of E. coli and other bacteria containing plasmids in molecular biology. It is also important in understanding how the plasmid–bacteria association has endured in a natural envi© 1998 NRC Canada W98-020.CHP Tue Jun 02 09:10:34 1998 Color profile: Disabled Composite Default screen Smith and Bidochka ronment, for clearly the environment in which bacteria are cultured affects plasmid maintenance. Differing growth and storage conditions can evoke and maintain changes in subsequent generations. Environmental conditions influence segregation instability or plasmid copy number and bacteria containing plasmids of different sizes have different fitness levels in different environments. The findings reported here have important implications in the use of plasmids in molecular biology and in understanding the bacteria–plasmid association in conditions of starvation and nutrient absence. Maintenance of pBluescript is significantly improved by reductions in plasmid size. High copy number plasmids reduce the fitness of the bacterial host. Acknowledgements Thanks to Michael Berrill for continued discussions regarding the nature of these experiments and to an anonymous reviewer for comments regarding an earlier manuscript. References Berghammer, H., and Auer, B. 1993. “Easypreps”: fast and easy plasmid minipreperation for analysis of recombinant clones in E. coli. Biotechniques, 14: 527–528. Boe, L, Gerdes, K., and Molin, S. 1987. Effects of genes exerting growth inhibition and plasmid stability on plasmid maintenance. J. Bacteriol. 169: 4646–4650. Bron, S., and Luxen, E. 1985. Segregational instability of pUB110derived recombinant plasmids in Bacillus subtilis. Plasmid, 14: 235–244. Bron, S., Luxen, E., and Swart, P. 1988. Instability of recombinant pUB110 plasmids in Bacillus subtilis: plasmid-encoded stability function and effects of DNA inserts. Plasmid, 19: 231–241. Caldwell, B.A, Ye, C., Griffiths, R.P., Moyer., C.L. and Morita, R.Y. 1989. Plasmid expression and maintenance during long-term starvation-survival of bacteria in well water. J. Environ. Microbiol. 55: 1860–1864. de Taxis du Poet, P., Arcand, Y., Bernier, R., Barbotin, J.-N., and Thomas, D. 1987. Plasmid stability in immobilised and free recombinant Escherichia coli JM105(pKK223–200): importance of oxygen diffusion, growth rate, and plasmid copy number. Appl. Environ. Microbiol. 53: 1548–1555. Gerdes, K., Rasmussen, P.B., and Molin, S. 1986. Unique type of plasmid maintenance function: postsegregational killing of plasmidfree cells. Proc. Natl. Acad. Sci. U.S.A. 83: 3116–3120. Godwin, D., and Slater, J.H. 1979. The influence of the growth environment on the stability of a drug resistance plasmid in 355 Escherichia coli K12. J. Gen. Microbiol. 111: 201–210. Griffiths, R.P., Moyer, C.L., Caldwell, B.A., Ye, C., and Morita, R.Y. 1990. Long term starvation-induced loss of antibiotic resistance in bacteria. Microb. Ecol. 19: 252–257. Guerrier-Takada, C., Salavati, R., and Altman, S. 1997. Phenotypic conversion of drug-resistant bacteria to drug sensitivity. Proc. Natl. Acad. Sci. U.S.A. 94: 8468–8472. Hakkaart, H.J.J., van Gemen, B., Velkamp, E., and Nijkamp, J.J.J. 1985. Maintenance of multicopy plasmid Clo DF13 III: role of plasmid size and copy number in partitioning. Mol. Gen. Genet. 198: 364–366. Kolter, R, Siegele, D.A., and Tormo, A. 1993. The stationary phase of the bacterial life cycle. Annu. Rev. Microbiol. 47: 855–874. Kues, U., and Stahl, U. 1989. Replication of plasmids in Gram-negative bacteria. Microbiol. Rev. 53: 491–516. Lederberg, J., and Lederberg, E.M. 1952. Replica plating and indirect selection of bacterial mutants. J. Bacteriol. 63: 399–406. Lenski, R.E., and Bouma, J.E. 1987. Effects of segregation and selection on instability of plasmid pACYC184 in Escherichia coli B. J. Bacteriol. 169: 5314–5316. Leonhardt, H, and Alonso, J.C. 1991. Parameters affecting plasmid stability in Bacillus subtilis. Gene, 103: 107–111. Modi, R.I., and Adams, J. 1991. Coevolution in bacterial-plasmid populations. Evolution, 45: 656–667. Moser, D.R., and Campbell, J.L. 1983. Characterisation and complementation of pMB1 copy number mutant: effect of RNA I gene dosage on plasmid copy number and incompatibility. J. Bacteriol. 154: 809–818. Rhee, J.I., Ricci, J.C.D., Bode, J., and Schugerl, K. 1994. Metabolic enhancement due to plasmid maintenance. Biotechnol. Lett. 16: 881–884. Rohde, C. 1995. Technical information sheet No. 12: plasmid isolation from bacteria: some fast procedures. World J. Microbiol. Biotechnol. 11: 367–369. Sambrook, J., Fritshck, E.F., and Maniatis, T. 1989. Molecular cloning: a labratory manual. 2nd ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. Seo, J.H., and Bailey, J.F. 1985. Effects of recombinant plasmid content on growth properties and cloned gene product formation in Escherichia coli. Biotechnol. Bioeng. 27: 1668–1674. Summers, D.K., and Sherratt, D.J. 1984. Multimerisation of high copy number plasmids causes instability: ColE1 encodes a determinant essential for plasmid monomerisation and stability. Cell, 36: 1097–1103. Valenzuela, M.S., Ikpeazu, E.V., and Siddiqui, R.A.I. 1996. E. coli growth inhibition by a high copy number derivative of plasmid pBR322. Biochem. Biophys. Res. Commun. 219: 876–883. Warnes, A., and Stephenson, J.R. 1986. The insertion of large pieces of foreign genetic material reduces the stability of bacterial plasmids. Plasmid, 16: 116–123. © 1998 NRC Canada W98-020.CHP Tue Jun 02 09:10:35 1998
© Copyright 2024 Paperzz