- Email: [email protected]
Plasmid curing from an acidophilic bacterium of the genus Acidocella
FEMS Microbiology Letters 183 (2000) 271^274
www.fems-microbiology.org
Plasmid curing from an acidophilic bacterium of the genus Acidocella S. Ghosh a , N.R. Mahapatra b
a;1
, T. Ramamurthy b , P.C. Banerjee
a;
*
a Indian Institute of Chemical Biology, 4 Raja S.C. Mullick Road, Calcutta 700032, India National Institute of Cholera and Enteric Diseases, P-33 CIT Road, Scheme XM, Calcutta 700010, India
Received 2 July 1999; received in revised form 2 December 1999 ; accepted 3 December 1999
Abstract Preservation of the acidophilic heterotroph, Acidocella sp. strain GS19h, at 4³C in stab culture eliminated all indigenous plasmids from this bacterium. Growth at 42³C initially caused changes in the plasmid profile followed by total elimination of plasmids after 10 cycles of growth. Concomitant to this loss of all plasmids, the cured derivatives became sensitive to CdSO4 and ZnSO4 , and the MIC value of the salts dropped from 1 M for each in the case of parental strain to 2 mM and 5 mM, respectively, suggesting plasmid-mediated inheritance of metal resistance in this bacterium. The cured derivatives could not utilise lactose, indicating this metabolic activity to be plasmid-associated in this strain. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Acidophilic bacterium ; Acridine orange; Acri£avine ; High-temperature curing ; Low-temperature curing ; Metal resistance ; Plasmid curing
1. Introduction Bacterial plasmids are known to harbour genes for: (i) resistances to antibiotics and metals [1,2], (ii) catabolic pathways such as lactose utilisation and degradation of hydrocarbons [3,4], (iii) biosynthesis of certain antibiotics, etc. [5]. In many cases, however, characteristics of the host organism conferred by the plasmids remain elusive, and such cryptic plasmids are abundant in nature [6]. Curing of a cryptic plasmid from a bacterial strain is a method to substantiate the relationship between a genetic trait and carriage of that speci¢c trait in the plasmid. Various methods involving chemical and physical agents have been developed to eliminate plasmids [7]. Depending on the nature of the bacterial host and/or the plasmid, some methods work better in a system than others do. As an example, the Inc-8 plasmid FP2 was cured from Pseudomonas aeruginosa only when the host cells had a dht mutation ; the curing procedure consisting of freezing
* Corresponding author. Tel. : +91 (33) 4733491; Fax: +91 (33) 4735197; E-mail: [email protected] 1 Present address: Department of Medicine and Center for Molecular Genetics, University of California at San Diego, and Veterans A¡airs Medical Center, San Diego, CA 91261-9111, USA.
cells in 15% glycerol at 370³C for at least 48 h was a novel one [8]. Protocols for curing plasmids consist frequently of exposure of a culture to sub-inhibitory concentrations of some chemical agents, e.g. acridine orange, acri£avine, sodium dodecyl sulfate or to a super-optimal temperature followed by selection of cured derivatives [7]. Although such methods are widely practised for elimination of plasmids from host bacterial cells, not a single report on plasmid curing from acidophilic bacteria of acidic mine environment is available. This work was initiated from the observation that cold preservation of the Acidocella strain GS19h in stab culture resulted in the elimination of plasmids, suggesting that curing may be achieved in the case of acidophilic bacteria applying physical or chemical methods. 2. Materials and methods 2.1. Bacterial strains and growth conditions From an Indian copper mine, a mesophilic, acidophilic, heterotrophic culture was isolated in MGY medium of the following composition in g l31 : glucose, 1; yeast extract, 0.1; (NH4 )2 SO4 , 2.0; K2 HPO4 , 0.25; MgSO4 W7H2 O, 0.25; and KCl, 0.1; pH of the medium was adjusted to 3 with 1 N H2 SO4 . The culture was puri¢ed by repeated single
0378-1097 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 9 9 ) 0 0 6 4 0 - 0
FEMSLE 9212 3-2-00
272
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
colony isolation on MGY-agarose (0.6%, w/v) medium, which was prepared by adding an equal volume of sterile 2UMGY medium (pH 3) to sterile 1.2% (w/v) agarose at 50^55³C just before plating. The strain GS19h was placed in the genus Acidocella after 16S rDNA sequencing [9]. It was routinely cultured at 30³C on a rotary shaker in MGY medium of pH 3. Growth of the strain and its plasmidcured derivatives in the presence of metal salts was tested in the same medium as reported previously [10]. In replica plating experiments, MGY-agarose (0.6%, w/v) medium with or without a metal salt was used. 2.2. Plasmid preparation Plasmid was prepared following an alkaline lysis method as described before [11]. Normal gel electrophoresis of plasmids was carried out with 0.7-0.8% (w/v) agarose. 2.3. Determination of metal resistance spectrum The MICs were determined by allowing the wild-type Acidocella strain and its cured derivatives to grow up to 7 days in metal salt containing MGY medium (pH 3) at 30³C on a rotary shaker in triplicate sets [10]. Growth of the cultures was monitored by measuring optical density at 540 nm. 2.4. Biochemical characteristics of the wild-type strain and cured derivatives Biochemical characterisation of the Acidocella strain and its plasmid-cured derivatives was carried out using API 20 E kit of BioMerieux, France. The instructions supplied by the company were followed except that the time of incubation was 3 days instead of 24 h.
3. Results It was a chance observation that the Acidocella strain GS19h lost all three plasmids when stored at 4³C in stab cultures for more than 2 years. While testing viability of the stab cultures in MGY-agarose, it was detected that the majority of the cultures were alive but lost the plasmids. Plasmid-less derivatives of the strain were sensitive to 2 mM CdSO4 and 5 mM ZnSO4 while the parental strain exhibited extreme resistance being sensitive to only 1 M of these salts, which indicated involvement of plasmids in imparting metal resistance. Attempts were, therefore, ¢rst made to cure the plasmids by the DNA intercalating agents acridine orange and acri£avine. The level of these chemicals tolerated by this strain in MGY medium was extremely high ( s 400 Wg ml31 and s 700 Wg ml31 for acridine orange and acri£avine, respectively). There was, however, no change in the plasmid pro¢le of this strain even after repeated sub-culturing in the presence of such high concentrations of these curing agents. Since cryo-preservation resulted in the loss of plasmids from the Acidocella strain, it was intriguing to study the thermal e¡ect on plasmid stability in this bacterium. Since the strain GS19h could not usually grow at 37³C [9], an initial attempt was made to let it grow at this temperature. Normal growth of the strain at 37³C was observed after prolonged incubation for many days; but the duration of growth period was reduced to almost normal range after a few transfers at this temperature. Although the culture or isolated colonies from the same showed some deviation in respect of mobility of plasmids in agarose gel from the typical plasmid pro¢le of the strain grown at 30³C, metal resistance characteristics of the strain remained unchanged. When growth of the strain at an even higher temperature of 42³C was tried, after two cycles, the small-
2.5. Curing experiments For curing with chemical agents, a freshly grown culture was used to inoculate (5%, v/v inoculum) MGY medium containing di¡erent concentrations of acridine orange (from 10 to 400 Wg ml31 ) or acri£avine (from 10 to 700 Wg ml31 ). The strain was allowed to grow at 30³C on a shaker. After su¤cient growth, a portion of each culture was withdrawn to check the plasmid pro¢le of growing cultures in agarose gel electrophoresis. The growth cycles were repeated ¢ve times to ensure the possibility of plasmid curing by these chemicals. To achieve curing by heat treatment [7], the strain was initially adapted to grow at 37³C in MGY medium for a few cycles. This culture was further adapted to grow at 42³C using 10% (v/v) inoculum. Growth at this temperature was repeated several times in fresh medium, and plasmid pro¢le was occasionally checked in agarose gel. Single colonies from such cultures were isolated and subjected to plasmid pro¢le analysis.
Fig. 1. Plasmid DNA pro¢le on 0.7% (w/v) agarose gel. Lane 1, wildtype strain GS19h; lane 2, derivative of the strain after two cycles of growth at 42³C; lane 3, derivative of the strain after four cycles of growth at 42³C; lane 4, completely cured derivative. `chr' denotes chromosomal DNA.
FEMSLE 9212 3-2-00
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
est plasmid (26.7 kb) and the largest plasmid (77 kb), and after four cycles, the 36 kb plasmid also could not be detected by gel electrophoresis. Instead, three new plasmid bands of molecular sizes of 72 (band a), 75 (band b) and 89.3 (band c) kb were detected (Fig. 1, lanes 1^3). That the change in plasmid pro¢le was permanent was ascertained when any of the original plasmids could not be visualised by gel electrophoresis of the plasmid preparation from the same culture after its repeated growth at 30³C. The culture treated at elevated temperature for four cycles (partially cured) could not grow with 80 mM ZnSO4 or 10 mM CdSO4 in MGY medium unlike the original strain. On further sub-culturing of this strain at 42³C for 10 cycles or more, no plasmid band could be detected and the metal resistance levels were found even lower, 2 mM for CdSO4 and 5 mM for ZnSO4 . The cured derivatives, however, could grow in the presence of high concentrations of acridine orange and acri£avine like the parental strain. That no plasmid band remained hidden in the chromosomal region of the cured derivative (Fig. 1, lane 4) was checked by restriction enzyme digestion of the DNA preparation followed by gel electrophoresis. Probability of chromosomal integration of the plasmid at temperature stress was ruled out after Southern hybridisation using total plasmids of the strain GS19h as probe (data not shown). In addition to the metal resistance determinants, the plasmids of strain GS19h may contain genes of catabolic pathways. Therefore, biochemical characteristics (done by API 20 E kit) of two cured derivatives, one obtained through cold preservation and the other by growing at 42³C, were compared with those of the parent strain to test this possibility. Interestingly, the cured derivatives possessed all the typical characteristics of the strain GS19h except lactose utilisation. 4. Discussion Improvement of bio-leaching operations through the application of genetic engineering techniques in acidophilic mining bacteria is currently envisaged. Towards this end, indigenous plasmids encoding speci¢c phenotypes such as metal resistance may have immense practical implications [12]. But presence and identi¢cation of such plasmids in acidophilic bacteria could not be achieved for a long period. In an attempt to search for such plasmids in acidophilic bacteria for development of genetic technology for mining microbes, the plasmid-bearing highly metal resistant Acidocella sp. strain GS19h was chosen. Since one or more plasmids of this strain harbour metal resistance conferring gene(s) [11], it is obvious that its plasmid-less derivatives will be more sensitive to metals. This report af¢rms that curing of plasmids may be achieved arti¢cially by applying physical methods even in the case of acidophilic bacteria of mine environment. Curing and eventual drastic decrease of metal resistance
273
characteristics were observed through sub-culturing of the cells at super-optimal growth temperature for more than 10 cycles but not by acridine orange and acri£avine, suggesting the speci¢city of these curing agents to some group of plasmids as well as to some bacteria [5]. It is of great interest to note that there was su¤cient change in plasmid pro¢le of the strain when it was grown at super-optimal temperature. Structural instability of the plasmids at higher temperature [13] might be the reason behind plasmid curing in this strain. When grown in MGY medium, the plasmid-less derivatives of Acidocella strain GS19h were observed to grow with a faster rate as compared to the plasmid-bearing wild-type strain. This was probably due to the following reasons: (i) the plasmids do not contain genes essential for carrying out normal physiological processes; (ii) the plasmid-related growth inhibition of the host strain was released in the absence of plasmids after curing [14]. It would be more convincing, speci¢cally for the latter proposition, if some properties of the strain could be studied after reintroduction of the plasmids in the plasmid-cured derivative. Attempts to transform plasmid-less derivatives of strain GS19h applying standard methods of CaCl2 treatment and electroporation were, however, not successful. Interestingly, besides the loss of metal resistance characteristics, the cured derivatives of the strain were unable to use lactose as carbon source for growth. It may therefore be suggested that this metabolic activity is also plasmid-mediated in this strain similar to the previous report [3]. In conclusion, this work reports for the ¢rst time curing of plasmids from an acidophilic bacterial strain. It was observed that propagation of the strain for several cycles at higher temperature but not in the presence of DNA intercalating agents caused elimination of plasmids from the cells. Thus this report may be useful in resolving the phenotypic characteristics imparted by the cryptic plasmids of other acidophilic bacteria. Acknowledgements S. Ghosh and N.R. Mahapatra thank the Labonya Prova Bose Trust, Calcutta, India, and Council of Scienti¢c and Industrial Research (CSIR), New Delhi, India, for fellowships. The authors express thanks to G.B. Nair and his group, National Institute of Cholera and Enteric Diseases, Calcutta, India, for help in this work.
References [1] Foster, T.J. (1983) Plasmid determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiol. Rev. 47, 361^409. [2] Silver, S. (1996) Bacterial resistances to toxic metal ions- a review. Gene 179, 9^19.
FEMSLE 9212 3-2-00
274
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
[3] Chassy, B.M., Gibson, E.M. and Giu¡rida, A. (1978) Evidence for plasmid associated lactose metabolism in Lactobacillus casei subsp. Casei. Curr. Microbiol. 1, 141^144. [4] Ghosal, D., You, I.S., Chatterjeee, D.K. and Chakrabarty, A.M. (1985) Plasmids in the degradation of chlorinated aromatic compounds. In: Plasmids in Bacteria (Helniski, D.R., Cohen, S.N., Clewell, D.B., Jackson, D.A. and Hollander, A., Eds.), pp. 667^687. Plenum Press, New York. [5] Stanisich, V.A. (1984) Identi¢cation and analysis of plasmids at the genetic level. In: Methods in Microbiology, Vol. 17 (Bennett, P.M. and Grinsted, J., Eds.), pp. 5^32. Academic Press, London. [6] Couturier, M., Bex, F., Bergquist, P.L. and Maas, W.K. (1988) Identi¢cation and classi¢cation of bacterial plasmids. Microbiol. Rev. 52, 375^395. [7] Crosa, J.H., Tolmasky, M.E., Actis, L.A. and Falkow, S. (1994) Plasmids. In: Methods for General and Molecular Bacteriology (Gerhardt, P., Ed.), pp. 365^386. American Society for Microbiology, Washington, DC. [8] Potter, A.A. and Loutit, J.S. (1983) FP2 plasmid curing in Pseudomonas aeruginosa. Can. J. Microbiol. 29, 732^733.
[9] Banerjee, P.C., Ray, M.K., Koch, C., Bhattacharyya, S., Shivaji, S. and Stackebrandt, E. (1996) Molecular characterization of two acidophilic heterotrophic bacteria isolated from a copper mine of India. Syst. Appl. Microbiol. 19, 78^82. [10] Mahapatra, N.R. and Banerjee, P.C. (1996) Extreme tolerance to cadmium and high resistance to copper, nickel and zinc in di¡erent Acidiphilum strains. Lett. Appl. Microbiol. 23, 393^397. [11] Ghosh, S., Mahapatra, N.R. and Banerjee, P.C. (1997) Metal resistance in Acidocella strains and plasmid-mediated transfer of this characteristic to Acidiphilium multivorum and Escherichia coli. Appl. Environ. Microbiol. 63, 4523^4527. [12] Rawlings, D.E. and Silver, S. (1995) Mining with microbes. Biotechnology 13, 773^778. [13] Craynest, M., Barbotin, J.N., Tru¡aut, N. and Thomas, D. (1996) Stability of plasmid pHV1431 in free and immobilized cell cultures. Ann. N.Y. Acad. Sci. 782, 311^322. [14] Eberhard, G.W. (1989) Why do bacterial plasmids carry some genes and not others? Plasmid 21, 167^174.
FEMSLE 9212 3-2-00
www.fems-microbiology.org
Plasmid curing from an acidophilic bacterium of the genus Acidocella S. Ghosh a , N.R. Mahapatra b
a;1
, T. Ramamurthy b , P.C. Banerjee
a;
*
a Indian Institute of Chemical Biology, 4 Raja S.C. Mullick Road, Calcutta 700032, India National Institute of Cholera and Enteric Diseases, P-33 CIT Road, Scheme XM, Calcutta 700010, India
Received 2 July 1999; received in revised form 2 December 1999 ; accepted 3 December 1999
Abstract Preservation of the acidophilic heterotroph, Acidocella sp. strain GS19h, at 4³C in stab culture eliminated all indigenous plasmids from this bacterium. Growth at 42³C initially caused changes in the plasmid profile followed by total elimination of plasmids after 10 cycles of growth. Concomitant to this loss of all plasmids, the cured derivatives became sensitive to CdSO4 and ZnSO4 , and the MIC value of the salts dropped from 1 M for each in the case of parental strain to 2 mM and 5 mM, respectively, suggesting plasmid-mediated inheritance of metal resistance in this bacterium. The cured derivatives could not utilise lactose, indicating this metabolic activity to be plasmid-associated in this strain. ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. Keywords : Acidophilic bacterium ; Acridine orange; Acri£avine ; High-temperature curing ; Low-temperature curing ; Metal resistance ; Plasmid curing
1. Introduction Bacterial plasmids are known to harbour genes for: (i) resistances to antibiotics and metals [1,2], (ii) catabolic pathways such as lactose utilisation and degradation of hydrocarbons [3,4], (iii) biosynthesis of certain antibiotics, etc. [5]. In many cases, however, characteristics of the host organism conferred by the plasmids remain elusive, and such cryptic plasmids are abundant in nature [6]. Curing of a cryptic plasmid from a bacterial strain is a method to substantiate the relationship between a genetic trait and carriage of that speci¢c trait in the plasmid. Various methods involving chemical and physical agents have been developed to eliminate plasmids [7]. Depending on the nature of the bacterial host and/or the plasmid, some methods work better in a system than others do. As an example, the Inc-8 plasmid FP2 was cured from Pseudomonas aeruginosa only when the host cells had a dht mutation ; the curing procedure consisting of freezing
* Corresponding author. Tel. : +91 (33) 4733491; Fax: +91 (33) 4735197; E-mail: [email protected] 1 Present address: Department of Medicine and Center for Molecular Genetics, University of California at San Diego, and Veterans A¡airs Medical Center, San Diego, CA 91261-9111, USA.
cells in 15% glycerol at 370³C for at least 48 h was a novel one [8]. Protocols for curing plasmids consist frequently of exposure of a culture to sub-inhibitory concentrations of some chemical agents, e.g. acridine orange, acri£avine, sodium dodecyl sulfate or to a super-optimal temperature followed by selection of cured derivatives [7]. Although such methods are widely practised for elimination of plasmids from host bacterial cells, not a single report on plasmid curing from acidophilic bacteria of acidic mine environment is available. This work was initiated from the observation that cold preservation of the Acidocella strain GS19h in stab culture resulted in the elimination of plasmids, suggesting that curing may be achieved in the case of acidophilic bacteria applying physical or chemical methods. 2. Materials and methods 2.1. Bacterial strains and growth conditions From an Indian copper mine, a mesophilic, acidophilic, heterotrophic culture was isolated in MGY medium of the following composition in g l31 : glucose, 1; yeast extract, 0.1; (NH4 )2 SO4 , 2.0; K2 HPO4 , 0.25; MgSO4 W7H2 O, 0.25; and KCl, 0.1; pH of the medium was adjusted to 3 with 1 N H2 SO4 . The culture was puri¢ed by repeated single
0378-1097 / 00 / $20.00 ß 2000 Federation of European Microbiological Societies. Published by Elsevier Science B.V. All rights reserved. PII: S 0 3 7 8 - 1 0 9 7 ( 9 9 ) 0 0 6 4 0 - 0
FEMSLE 9212 3-2-00
272
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
colony isolation on MGY-agarose (0.6%, w/v) medium, which was prepared by adding an equal volume of sterile 2UMGY medium (pH 3) to sterile 1.2% (w/v) agarose at 50^55³C just before plating. The strain GS19h was placed in the genus Acidocella after 16S rDNA sequencing [9]. It was routinely cultured at 30³C on a rotary shaker in MGY medium of pH 3. Growth of the strain and its plasmidcured derivatives in the presence of metal salts was tested in the same medium as reported previously [10]. In replica plating experiments, MGY-agarose (0.6%, w/v) medium with or without a metal salt was used. 2.2. Plasmid preparation Plasmid was prepared following an alkaline lysis method as described before [11]. Normal gel electrophoresis of plasmids was carried out with 0.7-0.8% (w/v) agarose. 2.3. Determination of metal resistance spectrum The MICs were determined by allowing the wild-type Acidocella strain and its cured derivatives to grow up to 7 days in metal salt containing MGY medium (pH 3) at 30³C on a rotary shaker in triplicate sets [10]. Growth of the cultures was monitored by measuring optical density at 540 nm. 2.4. Biochemical characteristics of the wild-type strain and cured derivatives Biochemical characterisation of the Acidocella strain and its plasmid-cured derivatives was carried out using API 20 E kit of BioMerieux, France. The instructions supplied by the company were followed except that the time of incubation was 3 days instead of 24 h.
3. Results It was a chance observation that the Acidocella strain GS19h lost all three plasmids when stored at 4³C in stab cultures for more than 2 years. While testing viability of the stab cultures in MGY-agarose, it was detected that the majority of the cultures were alive but lost the plasmids. Plasmid-less derivatives of the strain were sensitive to 2 mM CdSO4 and 5 mM ZnSO4 while the parental strain exhibited extreme resistance being sensitive to only 1 M of these salts, which indicated involvement of plasmids in imparting metal resistance. Attempts were, therefore, ¢rst made to cure the plasmids by the DNA intercalating agents acridine orange and acri£avine. The level of these chemicals tolerated by this strain in MGY medium was extremely high ( s 400 Wg ml31 and s 700 Wg ml31 for acridine orange and acri£avine, respectively). There was, however, no change in the plasmid pro¢le of this strain even after repeated sub-culturing in the presence of such high concentrations of these curing agents. Since cryo-preservation resulted in the loss of plasmids from the Acidocella strain, it was intriguing to study the thermal e¡ect on plasmid stability in this bacterium. Since the strain GS19h could not usually grow at 37³C [9], an initial attempt was made to let it grow at this temperature. Normal growth of the strain at 37³C was observed after prolonged incubation for many days; but the duration of growth period was reduced to almost normal range after a few transfers at this temperature. Although the culture or isolated colonies from the same showed some deviation in respect of mobility of plasmids in agarose gel from the typical plasmid pro¢le of the strain grown at 30³C, metal resistance characteristics of the strain remained unchanged. When growth of the strain at an even higher temperature of 42³C was tried, after two cycles, the small-
2.5. Curing experiments For curing with chemical agents, a freshly grown culture was used to inoculate (5%, v/v inoculum) MGY medium containing di¡erent concentrations of acridine orange (from 10 to 400 Wg ml31 ) or acri£avine (from 10 to 700 Wg ml31 ). The strain was allowed to grow at 30³C on a shaker. After su¤cient growth, a portion of each culture was withdrawn to check the plasmid pro¢le of growing cultures in agarose gel electrophoresis. The growth cycles were repeated ¢ve times to ensure the possibility of plasmid curing by these chemicals. To achieve curing by heat treatment [7], the strain was initially adapted to grow at 37³C in MGY medium for a few cycles. This culture was further adapted to grow at 42³C using 10% (v/v) inoculum. Growth at this temperature was repeated several times in fresh medium, and plasmid pro¢le was occasionally checked in agarose gel. Single colonies from such cultures were isolated and subjected to plasmid pro¢le analysis.
Fig. 1. Plasmid DNA pro¢le on 0.7% (w/v) agarose gel. Lane 1, wildtype strain GS19h; lane 2, derivative of the strain after two cycles of growth at 42³C; lane 3, derivative of the strain after four cycles of growth at 42³C; lane 4, completely cured derivative. `chr' denotes chromosomal DNA.
FEMSLE 9212 3-2-00
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
est plasmid (26.7 kb) and the largest plasmid (77 kb), and after four cycles, the 36 kb plasmid also could not be detected by gel electrophoresis. Instead, three new plasmid bands of molecular sizes of 72 (band a), 75 (band b) and 89.3 (band c) kb were detected (Fig. 1, lanes 1^3). That the change in plasmid pro¢le was permanent was ascertained when any of the original plasmids could not be visualised by gel electrophoresis of the plasmid preparation from the same culture after its repeated growth at 30³C. The culture treated at elevated temperature for four cycles (partially cured) could not grow with 80 mM ZnSO4 or 10 mM CdSO4 in MGY medium unlike the original strain. On further sub-culturing of this strain at 42³C for 10 cycles or more, no plasmid band could be detected and the metal resistance levels were found even lower, 2 mM for CdSO4 and 5 mM for ZnSO4 . The cured derivatives, however, could grow in the presence of high concentrations of acridine orange and acri£avine like the parental strain. That no plasmid band remained hidden in the chromosomal region of the cured derivative (Fig. 1, lane 4) was checked by restriction enzyme digestion of the DNA preparation followed by gel electrophoresis. Probability of chromosomal integration of the plasmid at temperature stress was ruled out after Southern hybridisation using total plasmids of the strain GS19h as probe (data not shown). In addition to the metal resistance determinants, the plasmids of strain GS19h may contain genes of catabolic pathways. Therefore, biochemical characteristics (done by API 20 E kit) of two cured derivatives, one obtained through cold preservation and the other by growing at 42³C, were compared with those of the parent strain to test this possibility. Interestingly, the cured derivatives possessed all the typical characteristics of the strain GS19h except lactose utilisation. 4. Discussion Improvement of bio-leaching operations through the application of genetic engineering techniques in acidophilic mining bacteria is currently envisaged. Towards this end, indigenous plasmids encoding speci¢c phenotypes such as metal resistance may have immense practical implications [12]. But presence and identi¢cation of such plasmids in acidophilic bacteria could not be achieved for a long period. In an attempt to search for such plasmids in acidophilic bacteria for development of genetic technology for mining microbes, the plasmid-bearing highly metal resistant Acidocella sp. strain GS19h was chosen. Since one or more plasmids of this strain harbour metal resistance conferring gene(s) [11], it is obvious that its plasmid-less derivatives will be more sensitive to metals. This report af¢rms that curing of plasmids may be achieved arti¢cially by applying physical methods even in the case of acidophilic bacteria of mine environment. Curing and eventual drastic decrease of metal resistance
273
characteristics were observed through sub-culturing of the cells at super-optimal growth temperature for more than 10 cycles but not by acridine orange and acri£avine, suggesting the speci¢city of these curing agents to some group of plasmids as well as to some bacteria [5]. It is of great interest to note that there was su¤cient change in plasmid pro¢le of the strain when it was grown at super-optimal temperature. Structural instability of the plasmids at higher temperature [13] might be the reason behind plasmid curing in this strain. When grown in MGY medium, the plasmid-less derivatives of Acidocella strain GS19h were observed to grow with a faster rate as compared to the plasmid-bearing wild-type strain. This was probably due to the following reasons: (i) the plasmids do not contain genes essential for carrying out normal physiological processes; (ii) the plasmid-related growth inhibition of the host strain was released in the absence of plasmids after curing [14]. It would be more convincing, speci¢cally for the latter proposition, if some properties of the strain could be studied after reintroduction of the plasmids in the plasmid-cured derivative. Attempts to transform plasmid-less derivatives of strain GS19h applying standard methods of CaCl2 treatment and electroporation were, however, not successful. Interestingly, besides the loss of metal resistance characteristics, the cured derivatives of the strain were unable to use lactose as carbon source for growth. It may therefore be suggested that this metabolic activity is also plasmid-mediated in this strain similar to the previous report [3]. In conclusion, this work reports for the ¢rst time curing of plasmids from an acidophilic bacterial strain. It was observed that propagation of the strain for several cycles at higher temperature but not in the presence of DNA intercalating agents caused elimination of plasmids from the cells. Thus this report may be useful in resolving the phenotypic characteristics imparted by the cryptic plasmids of other acidophilic bacteria. Acknowledgements S. Ghosh and N.R. Mahapatra thank the Labonya Prova Bose Trust, Calcutta, India, and Council of Scienti¢c and Industrial Research (CSIR), New Delhi, India, for fellowships. The authors express thanks to G.B. Nair and his group, National Institute of Cholera and Enteric Diseases, Calcutta, India, for help in this work.
References [1] Foster, T.J. (1983) Plasmid determined resistance to antimicrobial drugs and toxic metal ions in bacteria. Microbiol. Rev. 47, 361^409. [2] Silver, S. (1996) Bacterial resistances to toxic metal ions- a review. Gene 179, 9^19.
FEMSLE 9212 3-2-00
274
S. Ghosh et al. / FEMS Microbiology Letters 183 (2000) 271^274
[3] Chassy, B.M., Gibson, E.M. and Giu¡rida, A. (1978) Evidence for plasmid associated lactose metabolism in Lactobacillus casei subsp. Casei. Curr. Microbiol. 1, 141^144. [4] Ghosal, D., You, I.S., Chatterjeee, D.K. and Chakrabarty, A.M. (1985) Plasmids in the degradation of chlorinated aromatic compounds. In: Plasmids in Bacteria (Helniski, D.R., Cohen, S.N., Clewell, D.B., Jackson, D.A. and Hollander, A., Eds.), pp. 667^687. Plenum Press, New York. [5] Stanisich, V.A. (1984) Identi¢cation and analysis of plasmids at the genetic level. In: Methods in Microbiology, Vol. 17 (Bennett, P.M. and Grinsted, J., Eds.), pp. 5^32. Academic Press, London. [6] Couturier, M., Bex, F., Bergquist, P.L. and Maas, W.K. (1988) Identi¢cation and classi¢cation of bacterial plasmids. Microbiol. Rev. 52, 375^395. [7] Crosa, J.H., Tolmasky, M.E., Actis, L.A. and Falkow, S. (1994) Plasmids. In: Methods for General and Molecular Bacteriology (Gerhardt, P., Ed.), pp. 365^386. American Society for Microbiology, Washington, DC. [8] Potter, A.A. and Loutit, J.S. (1983) FP2 plasmid curing in Pseudomonas aeruginosa. Can. J. Microbiol. 29, 732^733.
[9] Banerjee, P.C., Ray, M.K., Koch, C., Bhattacharyya, S., Shivaji, S. and Stackebrandt, E. (1996) Molecular characterization of two acidophilic heterotrophic bacteria isolated from a copper mine of India. Syst. Appl. Microbiol. 19, 78^82. [10] Mahapatra, N.R. and Banerjee, P.C. (1996) Extreme tolerance to cadmium and high resistance to copper, nickel and zinc in di¡erent Acidiphilum strains. Lett. Appl. Microbiol. 23, 393^397. [11] Ghosh, S., Mahapatra, N.R. and Banerjee, P.C. (1997) Metal resistance in Acidocella strains and plasmid-mediated transfer of this characteristic to Acidiphilium multivorum and Escherichia coli. Appl. Environ. Microbiol. 63, 4523^4527. [12] Rawlings, D.E. and Silver, S. (1995) Mining with microbes. Biotechnology 13, 773^778. [13] Craynest, M., Barbotin, J.N., Tru¡aut, N. and Thomas, D. (1996) Stability of plasmid pHV1431 in free and immobilized cell cultures. Ann. N.Y. Acad. Sci. 782, 311^322. [14] Eberhard, G.W. (1989) Why do bacterial plasmids carry some genes and not others? Plasmid 21, 167^174.
FEMSLE 9212 3-2-00