Vol. 59, No. 12
INFECTION AND IMMUNITY, Dec. 1991, p. 4621-4627 0019-9567/91/124621-07$02.00/0 Copyright © 1991, American Society for Microbiology
Transformation of Actinobacillus actinomycetemcomitans by Electroporation, Utilizing Constructed Shuttle Plasmids PREM K. SREENIVASAN,1 DONALD J. LEBLANC,2 LINDA N. LEE,2 AND PAULA FIVES TAYLOR1* Department of Microbiology and Molecular Genetics, University of Vermont, Burlington, Vermont 05405,1 and Department of Microbiology, University of Texas Health Science Center at San Antonio, San Antonio, Texas 782842 Received 18 June 1991/Accepted 26 August 1991
Actinobacillus actinomycetemcomitans, a periodontal pathogen, has been strongly implicated in human periodontal disease. Advances in the molecular analysis of A. actinomycetemcomitans virulence factors have been limited due to the unavailability of systems for genetic transfer, transposon mutagenesis, and gene complementation. Slow progress can be traced almost exclusively to the lack of gene vector systems and methods for the introduction of DNA into A. actinomycetemcomitans. An electrotransformation system that allowed at least five strains of A. actinomycetemcomitans to be transformed with stable shuttle plasmids which efficiently replicated in both Escherichia coli and A. actinomycetemcomitans was developed. One plasmid, a potential shuttle vector designated pDL282, is 5.7 kb in size, has several unique restriction enzyme sites, and codes for resistance to spectinomycin and ampicillin. E. coli and A. actinomycetemcomitans were transformed with equal efficiencies of approximately 105 transformants per ,ug of DNA. Similar transformation efficiencies were obtained whether the plasmid DNA was isolated from A. actinomycetemcomitans or E. coli. In addition, frozen competent cells of A. actinomycetemcomitans yielded comparable efficiencies of transformation. Restriction enzyme analysis of pDL282 isolated after transformation confirmed the presence of intact donor plasniids. A plasmid isolated from A. pleuropneumoniae was also capable of transforming some isolates of A. actinomycetemcomitans, although generally at a lower frequency. The availability of these shuttle plasmids and an efficient transformation procedure should significantly facilitate the molecular analysis of virulence factors of A. actinomycetemcomitans. actinomycetemcomitans and this strain are in the same genus, pYGlO and the electroporation conditions established for A. pleuropneumoniae were used as starting points for this study. In this report, we describe the development of a highly efficient transformation system for A. actinomycetemcomitans that is useful for a variety of A. actinomycetemcomitans strains. A potential shuttle vector containing a small, naturally occurring, cryptic plasmid isolated from an A. actinomycetemcomitans strain was constructed. All plasmids utilized were stable in both Escherichia coli and A. actinomycetemcomitans. The availability of these plasmids and an efficient transformation system should facilitate the molecular analyses of A. actinomycetemcomitans virulence factors.
Microbial, immunologic, and clinical studies have implicated Actinobacillus actinomycetemcomitans, a gram-negative bacterium, in the pathogenesis of juvenile and adult periodontitis (22). A. actinomycetemcomitans elaborates several soluble extracellular virulence factors which include a collagenase (16), a fibroblast inhibition factor (17), epitheliotoxin (2), bone resorbing factors (15), leukocyte migration inhibition factor (20), and leukotoxin specific for human and primate polymorphonuclear neutrophils (19). Several lines of evidence suggest that A. actinomycetemcomitans is a significant periodontal pathogen. Patients infected with A. actinomycetemcomitans elaborate.a significant humoral immune response to the bacterium (7, 22). A. actinomycetemcomitans has been recovered from within the diseased gingival tissue (4), and recently, this laboratory has demonstrated that A. actinomycetemcomitans can invade epithelial cells in culture (12). Substantial effort is currently directed toward understanding the precise role each of these virulence factors plays in the pathogenesis of A. actinomycetemcomitans-mediated periodontal disease (18). The molecular analysis of virulence factors of A. actinomycetemcomitans has been hampered by the lack of efficient systems for genetic manipulation. Protocols for DNA transformation, generalized transduction, transposon mutagenesis, and allelic exchange are all unavailable. An electrotransformation system for A. pleuropneumoniae which uses pYG10, a 5.0-kb naturally occurring plasmid isolated from a multiply antibiotic-resistant strain of A. pleuropneumoniae 80-8141, does exist (8). Since both A. *
MATERIALS AND METHODS Bacterial strains and culture conditions. The bacterial strains used in these studies and their relevant characteristics are listed in Table 1. A. actinomycetemcomitans strains were routinely grown in Trypticase soy broth (Difco Laboratories, Detroit, Mich.) containing 0.6% yeast extract (Difco) and 0.04% sodium bicarbonate (Sigma Chemical Co., St. Louis, Mo.) (TSBYE broth) at 37°C in a GasPak containing a generator of 10% CO2 (BBL, Cockeysville, Md.). The antibiotics vancomycin and bacitracin (Sigma) were added to final concentrations of 5 and 75 jig/ml, respectively, unless otherwise indicated. Solid media were prepared by adding 1.5% Bacto-agar (Difco) to TSBYE. For long-term maintenance, A. actinomycetemcomitans strains were frozen at -70°C in TSBYE containing 100 ,ul of dimethyl sulfoxide per ml. E. coli strains were grown aerobically in YT broth (14) at 370C.
Corresponding author. 4621
4622
SREENIVASAN ET AL.
INFECT. IMMUN.
TABLE 1. Relevant characteristics of bacterial strains used in this study Bacterium
Strain
designation
A. actinomycetemcomitans
VT674 VT675 VT679 VT746 VT849
SUNY75S SUNY75R SUNY465 Y4 ATCC 29522
Zambon, SUNY, Buffalo, N.Y. Zambon Zambon Holt, UTHSC, San Antonio, Tex. ATCC, Md.
None None None None None
Smooth Rough Smooth Smooth Smooth
A. pleuropneumoniae
VT897
80-8141
Lalonde, Stanford, Calif.
pYGlO
Cmr
E. coli
DH5a
DH5a
Life Technologies, Gaithersburg, Md.
None
ApS Cms
a
Sourcea
Plasmid
Colony
SUNY, State University of New York; UTHSC, University of Texas Health Science Center.
Plasmids. The detailed restriction maps and construction of plasmids described in the text are described in detail elsewhere (9). In brief, plasmid pVT736-1, a naturally occurring 1.9-kb A. actinomycetemcomitans plasmid isolated from A. actinomycetemcomitans VT736, did not code for any selectable antibiotic phenotype tested. Therefore, pVT736-1 was unsuitable for developing a transformation system in A. actinomycetemcomitans without further manipulation. To obtain a selectable marker, attempts were made to clone the plasmid on an E. coli vector. Hence, pVT736-1 and pUC19 were digested with KpnI, mixed, and treated with T4 DNA ligase. The ligation mixture was used to transform E. coli DH5a, and transformants were selected on YT agar containing ampicillin and 5-bromo-4-chloro-3-indolyl-f-D-galactopyranoside (X-Gal). Recombinant colonies were screened for the presence of pUC19 containing a 1.9-kb KpnI fragment insert. Two populations of recombinant plasmids represented by pDL279 and pDL280 were obtained; they contain pVT736-1 in opposite orientations. Plasmid pDL281 was constructed by digesting plasmids pVT736-1 and pGEM7Zf(-) (obtained from Promega, Madison, Wis.) with XhoI and transforming the resulting ligation reaction into E. coli DH5a. Transformants were selected on YT agar containing ampicillin and X-Gal, and recombinant colonies were screened for the presence of pUC19 containing a 1.9-kb XhoI fragment insert. A single population of clones carrying pVT736-1 in one orientation was obtained. Plasmid pYG10, a naturally occurring plasmid from A. pleuropneumoniae, coding for chloramphenicol resistance (Cm%), was obtained from Guy Lalonde (8). Plasmid isolation. Five-milliliter cultures of A. actinomycetemcomitans transformants were screened for the presence of plasmids by a modification of the procedure of Anderson and McKay (1). Briefly, bacterial pellets were washed and resuspended in 6.7% sucrose-50 mM Tris HCl-10 mM EDTA, pH 8.0. Bacteria were incubated in the presence of lysozyme at 37°C for 20 min and lysed by the addition of sodium dodecyl sulfate. Nucleic acids precipitated by 2-propanol were washed twice with 70% ethanol and once in 99% ethanol and allowed to air dry. Plasmid DNA preparations were digested with restriction enzymes and analyzed on 0.8% Tris-borate-EDTA gels by using standard protocols (11). Large-scale plasmid preparations were obtained from 500-ml cultures of A. actinomycetemcomitans transformants, using a modification of the Anderson and McKay procedure, as described above. Plasmids from E. coli transformants were isolated by the alkaline lysis method of Birnboim and Doly (3) and analyzed on agarose gels following restriction enzyme digestion. Large-scale plasmid preparations for use in transformation were isolated
from 1-liter cultures and purified by cesium chloride density gradient centrifugation (11). Restriction enzyme analysis. All restriction enzymes were purchased from Life Technologies, Inc., Gaithersburg, Md., or Boehringer Mannheim Biochemicals, Indianapolis, Ind., and were used in accordance with the manufacturer's instructions. Restriction mapping of plasmids was carried out by using single and double enzyme digestions, and analysis was by electrophoresis on 0.8% Tris-borate-EDTA agarose gels. Plasmid transformation. (i). A. actinomycetemcomitans transformation. On day 1, 0.1 ml of frozen A. actinomycetemcomitans stock was inoculated into 1.0 ml of TSBYE broth and grown overnight as described above. The next day, the culture was diluted 1:20 and grown overnight again. On day 3, the A. actinomycetemcomitans cultures were diluted with an equal volume of fresh broth and allowed to grow to early logarithmic phase (3 h). The density of the cultures was read at 600 nm, and the bacteria were harvested by centrifugation in a Sorvall RT 6000B refrigerated centrifuge for 15 min at 3,000 rpm. The pellets were washed twice in half-volumes of ice-cold electroporation buffer (EPB; 15% glycerol, 272 mM sucrose, 2.43 mM K2HPO4, 0.57 mM KH2PO4, pH 7.4) and resuspended in fresh EPB to obtain a final optical density (OD) of 6.0 at 600 nm. This concentrated bacterial suspension was incubated on ice for 10 min prior to high-voltage pulse. Plasmid DNA was ethanol precipitated and resuspended in water at a concentration of 10 ,ug/ml (unless otherwise indicated) for use in transformation experiments. Fifty microliters of the concentrated bacterial suspension was mixed with the plasmid DNA in sterile cuvettes (Bio-Rad, Richmond, Calif.) with an interelectrode distance of 0.2 cm. High-voltage pulses were delivered with a GenePulser (Bio-Rad) set at 2,500 V, 200 Ql, and 25 ,uF. Following electroporation, bacteria were immediately transferred into 1.0 ml of warm TSBYE broth without antibiotics, and the broth was incubated for a period of 5 h (two doublings), unless otherwise indicated, to allow for expression of the transformed phenotype. Dilutions of transformed cultures were plated on TSBYE plates containing chloramphenicol (5 ,ug/ml), ampicillin (100 ,ug/ml), or spectinomycin (50 ,ug/ml), and cultures were incubated for 48 h as described above. P-Lactamase activity was assayed with Beta Lactam Reagent Disks (Marion Scientific, Kansas City, Mo.) in accordance with the manufacturer's instructions. Transformation efficiencies were expressed as CFU per milliliter per microgram of DNA from at least duplicate experiments. Frozen competent cells were prepared by rapidly freezing 250-,ul aliquots of A. actinomycetemcomitans cells resuspended in EPB to a density of 6.0 at 600 nm in a dry ice-ethanol bath for
VOL. 59, 1991
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
4623
TABLE 2. Transformation of A. actinomycetemcomitans strains TransformantsJml/,ug of DNA
Plasmid VT674
pYGlO pDL279 pDL280 pDL281
0 6 0.3 5
pUC19 -DNA
0 0
VT675
VT679
0 8 8 x 102 4.0 x 102 0.5 1.8 x 102 8.3 x 102 42
0 0
10 0
VT746
6.5 1.8 5.4 4.8
x x x x
50 0
ATCC 2952
102 5.14 x 104 104 1.8 x 1033 103 1.3 x 103 103 30
40 0
5 min and then storing at -70°C. For transformation, the vials were thawed on ice and allowed to stabilize on ice for an additional 10 min prior to electroporation. (ii) E. coli transformation. Frozen competent E. coli DH5a cells were purchased from Life Technologies, Inc., and transformed with gradient-purified plasmid DNA in accordance with the manufacturer's instructions. E. coli transformants were selected on YT agar supplemented with chloramphenicol (10 ,ug/ml), ampicillin (100 ug/lml), or spectinomycin (50 ,ug/ml). P-Lactamase activity was assayed as described above. Results are an average of duplicate experiments and are expressed as described above.
RESULTS Transformation of A. actinomycetemcomitans strains with several plasmids. All five A. actinomycetemcomitans strains examined were transformed with at least one of the plasmids in Table 2. The A. pleuropneumoniae plasmid pYGlO was the least efficient at transforming the A. actinomycetemcomitans strains. Even though pYGlO can replicate in E. coli (6), several A. actinomycetemcomitans isolates could not be transformed at detectable efficiencies. The shuttle plasmid pDL279 transformed with highest efficiency and transformed all five strains of A. actinomycetemcomitans. VT674, a spontaneous smooth-colony derivative, was transformed at a significantly lower efficiency than its rough-colony parent, VT675. The E. coli vector pUC19 transformed three of the A. actinomycetemcomitans strains at low efficiencies. Transforming plasmid could be isolated from all transformants except those transformed with pUC19. In addition, the amount of ,-lactamase produced by the pUC19 transformants was minimal compared with that of the other Apr transformants, suggesting that the bla gene was present in a low copy number. These data are consistent with the plasmid integrating into the A. actinomycetemcomitans chromosome; however, Southern hybridization was not performed to test this hypothesis. VT746 and ATCC 29522 were transformed with the highest efficiencies and with all of the plasmids used. A. actinomycetemcomitans ATCC 29522 had the shortest generation time (120 min) and grew to the highest OD; it is easily available to other investigators. Hence, it was chosen to optimize the transformation conditions. Construction of the shuttle vector pDL282. Plasmid pDL279 transformed all isolates of A. actinomycetemcomitans with high efficiency, making it the ideal plasmid for the construction of a shuttle vector. A spectinomycin resistance gene (isolated from Enterococcus faecalis [10]) was introduced into the unique SmaI site of pDL279 to obtain a recombinant plasmid with two antibiotic resistance markers. This new plasmid, designated pDL282, was used to transform E. coli DH5a. E. coli transformants were selected on
Hindill
--
Smal/Ndel
/
_
'Smal/Clal Xba
Kpni
Xba I
FIG. 1. Restriction endonuclease map of the E. coli-A. actinomycetemcomitans shuttle vector pDL282. spc, spectinomycin resistance gene; bla, ampicillin resistance gene; pVT 736-1, cryptic plasmid of A. actinomycetemcomitans VT736; E. coli ori, replication origin for pUC19.
YT agar containing spectinomycin. Ten transformants were arbitrarily picked and transferred onto YT agar plates containing both ampicillin and spectinomycin. Doubly resistant transformants were screened for the presence of a 5.7-kb recombinant plasmid, and construction was confirmed by restriction enzyme analysis. pDL282 was stably maintained in E. coli DH5a. A restriction map of pDL282 is shown in Fig. 1. The bla gene has a unique Scal site which permits cloning of insert DNA into bla with selection for Spr and Aps.
Effect on transformation of host modification of plasmid DNA. The source of plasmid DNA played a role in the transformation efficiency of A. actinomycetemcomitans. Plasmids pYGlO and pDL282 were regularly recovered from both E. coli and A. actinomycetemcomitans transformants regardless of the source of DNA (Table 3). pYGlO isolated from A. actinomycetemcomitans transformed both A. actinomycetemcomitans and E. coli more efficiently than DNA isolated from E. coli. On the other hand, pDL282 transformed E. coli more efficiently than it transformed A. actinomycetemcomitans regardless of the source of donor DNA. Interestingly, pDL282 isolated from A. actinomycetemcomitans transformed E. coli better than A. actinomycetemcomitans. Perhaps, the fact that DH5a is restriction minus plays a role in the higher transformation efficiency. Effect of stage of growth. Bacteria were grown to different stages in their growth cycle before electroporation to deterTABLE 3. Effect on transformation efficiency of modification of plasmid DNA by A. actinomycetemcomitans (Aa) ATCC 29522 and E. coli DH5a Plasmid
Source of plasmid DNA
Recipient
pYGlO
E. coli E. coli Aa Aa
E. coli Aa E. coli Aa
2.06 5.14 1.22 1.34
E. coli E. coli Aa Aa
E. coli Aa E. coli Aa
1.7 1.6 1.78 1.92
pDL282
Transformation efficiency/p.g of DNA x 106 x 105
x 107 x 106
x 106 x x x
105 105 104
4624
INFECT. IMMUN.
SREENIVASAN ET AL.
z
0.5
0
E c 0
105. 0.4
0 _
n
0.3
-.
0
0.2
-o
I-0 E
*
E IL
103
0
Transformants -0-- Optical density
I0
2
4
6
8
-0.1
'a0 C
0
10
Hour FIG. 2. Effect of stage of bacterial growth on transformation. Closed symbols indicate number of transformants per milliliter per microgram of DNA. Open symbols indicate OD at 600 nm at the time of harvesting. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
mine the optimal stage of growth for transformation. The highest transformation efficiencies were achieved when the bacteria were in the early logarithmic phase of growth (4 h) (Fig. 2). The number of transformants fell rapidly after this period of time. Effect of plasmid DNA concentration. The effect of varying plasmid DNA concentration on total number of transformants was assessed by electroporating aliquots of prepared bacterial samples with concentrations of DNA ranging from 0.1 to 10 ,ug/ml. A dose-response curve using different DNA concentrations is depicted in Fig. 3. The number of transformants increased with increasing DNA concentration and attained saturation at approximately 2.5 ,ug of DNA per ml. Since transformation efficiency is dependent on the concentration of DNA, the transformation efficiency remained fairly constant at 2.027 0.57 x 105 transformants per ,ug of DNA with DNA concentrations ranging from 0.5 to 3.5 ,ug/ml. Effect of bacterial cell density. Different cell densities of early-logarithmic-phase cultures of ATCC 29522 were assayed for ability to be transformed. A linear increase in the number of transformants with increased bacterial density occurred over an OD range of 1 to 6 at 600 nm (Fig. 4). The system appears to saturate at an OD of 6.0. The results show approximately a 10-fold increase in the number of transformants with increasing bacterial density from an OD of 1 to 6. Effect of pH of EPB. The pH of the EPB has been demonstrated to affect the transformation efficiency in several systems (6). Since the pH of our standard EPB could not be adjusted without affecting the ionic strength, we tested EPB at pH 7.4 with and without the addition of phosphates. Transformation efficiencies were comparable with both buffers (data not shown), suggesting that phosphates were not required for transformation. A series of EPB without phosphates with pHs of 4 to 8 were prepared, and bacterial cultures were resuspended in these buffers and transformed under identical conditions. An acidic pH was more favorable for transformation than a neutral pH (Fig. 5). At pH 5.5, the number of transformants recovered was twofold higher than
that at pH 7.4. Thus, the transformation system appears quite sensitive over a significant pH range. Effect of length of recovery time. Five hours was initially chosen for the recovery time as this time represents approximately two generations for A. actinomycetemcomitans and should allow for maximal expression of the antibiotic resistance marker on the plasmid. The optimal recovery period following electroporation was determined by varying the recovery time from 1 to 5 h (Fig. 6). The number of
±
E IL U. -
c 0
E 0
ca
DNA concentration (ug/ml) FIG. 3. Effect of DNA concentration on transformation. DNA concentrations of 0.1 to 10 ,ug/ml were added to the bacterial samples. Concentrations of DNA were added in a constant volume of 1 ,ul. Symbols indicate number of transformants per milliliter recovered at each DNA concentration tested. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
VOL. 59, 1991
106 z a L-
E
10-
C) 0 0
10 4-
0
a-
10 3,3
.
4625
Perhaps the plasmid was lost in the absence of selective pressure. Effect of freezing on competence. The long generation time of A. actinomycetemcomitans strains and the length of time required for recovery made for a long experimental day. Hence, the effect on transformation efficiency of freezing competent cells was determined. Since EPB contains 15% glycerol, it was reasoned that cells could be prepared to this stage in the procedure and then quickly frozen in a dry ice-ethanol bath prior to storage at -70°C. The transformation efficiencies of freshly prepared ceUs and frozen competent cells were comparable (data not shown). We have tested cells that have been frozen for up to 2 months and have not seen a loss in transformation efficiency. However, once the cells have been thawed, they cannot be refrozen without a , _._,_._,_._,_._,_.significant loss in competence. 8Statistical analysis. The mean and standard deviation used I. 2 . to describe the results were calculated from at least duplicate experiments, using the software package Lotus 1-2-3. Bacterial density at 600 nm
FIG. 4. Effect of bacterial density on transformation. Logarithmic-phase cultures were resuspended in EPB to densities corresponding to the OD indicated. Symbols indicate number of transformants per milliliter per microgram of DNA recovered at each density tested. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
transformants recovered was optimal at 3 h and gradually declined thereafter. CFU of the same samples plated on nonselective media were determined to ascertain whether this decline was due to a decrease in total bacterial numbers. The results shown in Fig. 5 indicate that the bacteria continued to grow for a period of 4 h prior to a decrease in CFU. The number of transformants decreased after 3 h, suggesting that cell death was not responsible for the decrease in the number of transformants observed at 4 h.
z 0
E 0
0 0 I-
4
5
6
7
8
pH FIG. 5. Effect of pH of the EPB on transformation. A. actinomycetemcomitans cells were resuspended in EPB without phosphates at each of the pHs indicated. Symbols indicate the number of transformants per milliliter per microgram of DNA recovered at each pH tested. Standard deviation bars are not visible for certain data points because the limits of the SD bar do not extend beyond the limits of the square used to depict the data point.
DISCUSSION In this report, we describe an efficient method for DNA transformation of A. actinomycetemcomitans and the construction of a stable chimeric plasmid that can be used as an E. coli-A. actinomycetemcomitans cloning vector. The plasmids utilized in these studies are the first plasmids shown to transform several A. actinomycetemcomitans strains. An earlier study reported the transformation of an A. actinomycetemcomitans strain with a broad-host-range plasmid (21). To our knowledge, this work has not yet been published in full and the transformation efficiency is unclear. The shuttle vector pDL282 was capable of transforming to a high efficiency all A. actinomycetemcomitans strains tested. pDL282 has several useful features as a potential cloning vector, such as the following: a unique restriction site within the bla gene for marker inactivation (ScaI), a relatively small size (5.7 kb), and a gene for the phenotype Spr for selection. Since pDL282 replicates equally well in E. coli, this vector will prove to be useful in the study of A. actinomycetemcomitans genes. The transformation of A. actinomycetemcomitans was accomplished by electroporation, and the protocol was standardized by using A. actinomycetemcomitans ATCC 29522 and the shuttle vector pDL282. In general, the electroporation protocols were based on methods available for A. pleuropneumoniae (8), Campylobacterjejuni (13), and E. coli (5). Observations on the growth and metabolic requirements for A. actinomycetemcomitans made in this laboratory significantly influenced the optimization of the electroporation protocol. Several factors were investigated to optimize the transformation protocol. The stage of growth of A. actinomycetemcomitans influenced electroporation efficiencies. A. actinomycetemcomitans cultures in the early logarithmic phase of growth transformed to the highest efficiency. Bacterial cells harvested in the stationary phase of the growth cycle remained transformable, although at much lower efficiencies, suggesting that A. actinomycetemcomitans was transformable at every stage of growth tested. These results indicate that ATCC 29522, with a generation time of 120 min, needs to undergo at least two cycles of cell division before it can attain its highest transformation efficiency. Varying the concentration of DNA added to an electroporation mixture yielded a typical saturation curve. We routinely obtained 105 transformants by using 2.5 jig of pDL282 DNA per ml. These data will be important for the
4626
SREENIVASAN ET AL.
INFECT. IMMUN. 10 11
lo'U
z
0
E L0
10 10 0 6I
*10 9
.108
-.
0 c 0
10 7
co
E
0 C
LL
105
10 3
* Transformants -0- Viable bactena
-06
I-
cc
0 CD 0 cc .
.0 .0 CI
-1n 5
0
1
2
3
4
5
6
Hour FIG. 6. Effect of length of recovery time following electroporation on transformation. Immediately following electroporation, A. actinomycetemcomitans cells were placed in prewarmed TSBYE without antibiotics and incubated for the times indicated. The closed symbols indicate the number of Spr transformants per milliliter per microgram of DNA, while the viable count on nonselective media is indicated by the open symbols.
construction of gene libraries and in isolating specific transformants occurring at low frequencies. Additional factors that influenced electroporation efficiency included the density of the bacterial suspension, the pH of the EPB, and the duration of the recovery period following electroporation. As expected, more transformants were recovered by increasing the bacterial density used for electroporation. Saturation occurred at a bacterial density of 6.0 at 600 nm. The pH of the EPB was also a factor affecting electroporation; twice the number of transformants were recovered with EPB at pH 5.5 versus pH 7.4. The recovery period following the delivery of the electrical pulse also plays a role in transformation efficiency. The maximum number of transformants was recovered following 3 h of recovery. In contrast, the total number of viable bacteria in the culture increased for 4 h after electroporation. These data suggest that the transformants may lose their plasmid in the absence of selective pressure. Transformation efficiency with logarithmic-phase bacterial cultures suspended in EPB and frozen to -70°C was comparable to that obtained with fresh bacterial cultures. The A. pleuropneumoniae plasmid pYGlO was able to transform several A. actinomycetemcomitans strains. This plasmid has been reported to transform A. pleuropneumoniae and E. coli (8), suggesting that pYGlO is a broad-hostrange plasmid. The role of E. coli and A. actinomycetemcomitans modification systems on transformation was examined by using the shuttle vectors pDL282 and pYG10. We routinely obtained equal numbers of A. actinomycetemcomitans or E. coli transformants (105 per ,ug of DNA) with pDL282 isolated from the heterologous strain. In the case of pYG10, plasmid DNA isolated from A. actinomycetemcomitans was more efficient at transforming both E. coli and A. actinomycetemcomitans, suggesting a role for an A. actinomycetemcomitans modification system. However, the modification system of A. actinomycetemcomitans does not limit transformation of E. coli with either pDL282 or pYG10. The structures of pDL282 and pYGlO isolated from A. actinomycetemcomitans and E. coli transformants were found to be intact by
single and double restriction enzyme digestion, suggesting that the plasmids did not undergo any deletions or rearrangements following transformation. The structural stability of these plasmids is of significance in applying these plasmids for routine DNA manipulations in either A. actinomycetemcomitans or E. coli. Surprisingly, pDL282 isolated from A. actinomycetemcomitans transformed A. actinomycetemcomitans 10-fold less than pDL282 isolated from E. coli and transformed into E. coli. The reason for this result is unclear. The convenience of using DNA obtained from an E. coli donor for high-efficiency transformation of A. actinomycetemcomitans will prove invaluable in utilizing the extensive tools of E. coli genetics for the analysis of A. actinomycetemcomitans genes. Our laboratories are investigating the amount of passenger DNA that can be cloned and stably maintained in pDL282. We are also adapting pDL282 as a potential transposon delivery vehicle. The availability of these genetic tools will accelerate a molecular understanding of A. actinomycetemcomitans virulence factors. REFERENCES 1. Anderson, D., and L. L. McKay. 1983. Simple and rapid method for isolating large plasmid DNA from lactic streptococci. Appl. Environ. Microbiol. 46:549-552. 2. Birkedal-Hansen, H., P. W. Claufield, Y. M. Wannemuehier, and R. Pierce. 1982. A sensitive screening assay for epitheliotoxins produced by oral microorganisms, abstr. 125. Abstr. Annu. Meet. J. Dent. Res. 61:192. 3. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1521. 4. Christersson, L. A., B. Albini, J. J. Zambon, U. M. E. Wikesjo, and R. J. Genco. 1987. Tissue localization of Actinobacillus actinomycetemcomitans in human periodontitis. i. Light, immunofluorescence and electron microscopic studies. J. Periodontol. 58:529-539. 5. Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145. 6. Dunny, G. M., L. L. Lee, and D. J. LeBlanc. 1991. Improved electroporation and cloning vector system for gram-positive bacteria. Appl. Environ. Microbiol. 57:1194-1201.
VOL. 59, 1991
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
7. Genco, R. J., and J. Slots. 1984. Host responses in periodontal diseases. J. Dent. Res. 63:441-451. 8. Lalonde, G., J. F. Miller, and L. S. Tompkins. 1989. Transformation of Actinobacillus pleuropneumoniae and analysis of R factors by electroporation. Am. J. Vet. Res. 50:1957-1960. 9. LeBlanc, D. J., L. N. Lee, A. R. Abu-AM-Jaibat, P. K. Sreenivasan, and P. M. Fives-Taylor. Submitted for publication. 10. LeBlanc, D. J., L. N. Lee, and J. M. Inamine. 1991. Cloning and nucleotide base sequence analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob. Agents Chemother. 35:1804-1810. 11. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 12. Meyer, D., P. Sreenivasan, and P. Fives-Taylor. 1991. Evidence for invasion of a human oral cell line by Actinobacillus actinomycetemcomitans. Infect. Immun. 59:2719-2726. 13. Miller, J. F., W. J. Dower, and L. S. Tompkins. 1988. Highvoltage electroporation of bacteria: genetic transformation of Campylobacter jejuni with plasmid DNA. Proc. Natl. Acad. Sci. USA 85:856-860. 14. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 15. Nowotny, A., U. H. Behling, B. Hammond, C.-H. Lai, M. Listgarten, P. H. Pham, and F. Sanavi. 1982. Release of toxic microvesicles by Actinobacillus actinomycetemcomitans. Infect. Immun. 37:151-154.
4627
16. Robertson, P. B., M. Lantz, P. T. Marucha, K. S. Kornman, C. L. Trummel, and S. C. Holt. 1982. Collagenolytic activity associated with Bacteroides species and Actinobacillus actinomycetemcomitans. J. Periodontal. Res. 17:275-283. 17. Schenker, B. J., M. E. Kushner, and C.-C. Tsai. 1982. Inhibition of fibroblast proliferation by Actinobacillus actinomycetemcomitans. Infect. Immun. 38:986-992. 18. Slots, J., and R. J. Genco. 1984. Black-pigmented Bacteroides species, Capnocytophaga species, and Actinobacillus actinomycetemcomitans in human periodontal disease: virulence factors in colonization, survival, and tissue destruction. J. Dent. Res. 63:412-421. 19. Tsai, C.-C., B. J. Shenker, J. M. DiRienzo, D. Malamund, and N. S. Taichman. 1984. Extraction and isolation of a leukotoxin from Actinobacillus actinomycetemcomitans with polymyxin B. Infect. Immun. 43:700-705. 20. Van Dyke, T. E., E. Bartholomew, R. J. Genco, J. Slots, and M. J. Levine. 1982. Inhibition of neutrophil chemotaxis by soluble bacterial products. J. Periodontol. 53:502-508. 21. Yip, J., D. Furgang, E. Weidman, P. Goncharoff, D. Figurski, and D. H. Fine. 1990. Transfer of the promiscuous plasmid RK2 from E. coli to Actinobacillus actinomycetemcomitans (Aa). Program Abstr. 68th Int. Assoc. Dent. Res., abstr. 1959. J. Dent. Res. 69:353. 22. Zambon, J. J. 1985. Actinobacillus actinomycetemcomitans in human periodontal disease. J. Clin. Periodontol. 12:1-20.
INFECTION AND IMMUNITY, Dec. 1991, p. 4621-4627 0019-9567/91/124621-07$02.00/0 Copyright © 1991, American Society for Microbiology
Transformation of Actinobacillus actinomycetemcomitans by Electroporation, Utilizing Constructed Shuttle Plasmids PREM K. SREENIVASAN,1 DONALD J. LEBLANC,2 LINDA N. LEE,2 AND PAULA FIVES TAYLOR1* Department of Microbiology and Molecular Genetics, University of Vermont, Burlington, Vermont 05405,1 and Department of Microbiology, University of Texas Health Science Center at San Antonio, San Antonio, Texas 782842 Received 18 June 1991/Accepted 26 August 1991
Actinobacillus actinomycetemcomitans, a periodontal pathogen, has been strongly implicated in human periodontal disease. Advances in the molecular analysis of A. actinomycetemcomitans virulence factors have been limited due to the unavailability of systems for genetic transfer, transposon mutagenesis, and gene complementation. Slow progress can be traced almost exclusively to the lack of gene vector systems and methods for the introduction of DNA into A. actinomycetemcomitans. An electrotransformation system that allowed at least five strains of A. actinomycetemcomitans to be transformed with stable shuttle plasmids which efficiently replicated in both Escherichia coli and A. actinomycetemcomitans was developed. One plasmid, a potential shuttle vector designated pDL282, is 5.7 kb in size, has several unique restriction enzyme sites, and codes for resistance to spectinomycin and ampicillin. E. coli and A. actinomycetemcomitans were transformed with equal efficiencies of approximately 105 transformants per ,ug of DNA. Similar transformation efficiencies were obtained whether the plasmid DNA was isolated from A. actinomycetemcomitans or E. coli. In addition, frozen competent cells of A. actinomycetemcomitans yielded comparable efficiencies of transformation. Restriction enzyme analysis of pDL282 isolated after transformation confirmed the presence of intact donor plasniids. A plasmid isolated from A. pleuropneumoniae was also capable of transforming some isolates of A. actinomycetemcomitans, although generally at a lower frequency. The availability of these shuttle plasmids and an efficient transformation procedure should significantly facilitate the molecular analysis of virulence factors of A. actinomycetemcomitans. actinomycetemcomitans and this strain are in the same genus, pYGlO and the electroporation conditions established for A. pleuropneumoniae were used as starting points for this study. In this report, we describe the development of a highly efficient transformation system for A. actinomycetemcomitans that is useful for a variety of A. actinomycetemcomitans strains. A potential shuttle vector containing a small, naturally occurring, cryptic plasmid isolated from an A. actinomycetemcomitans strain was constructed. All plasmids utilized were stable in both Escherichia coli and A. actinomycetemcomitans. The availability of these plasmids and an efficient transformation system should facilitate the molecular analyses of A. actinomycetemcomitans virulence factors.
Microbial, immunologic, and clinical studies have implicated Actinobacillus actinomycetemcomitans, a gram-negative bacterium, in the pathogenesis of juvenile and adult periodontitis (22). A. actinomycetemcomitans elaborates several soluble extracellular virulence factors which include a collagenase (16), a fibroblast inhibition factor (17), epitheliotoxin (2), bone resorbing factors (15), leukocyte migration inhibition factor (20), and leukotoxin specific for human and primate polymorphonuclear neutrophils (19). Several lines of evidence suggest that A. actinomycetemcomitans is a significant periodontal pathogen. Patients infected with A. actinomycetemcomitans elaborate.a significant humoral immune response to the bacterium (7, 22). A. actinomycetemcomitans has been recovered from within the diseased gingival tissue (4), and recently, this laboratory has demonstrated that A. actinomycetemcomitans can invade epithelial cells in culture (12). Substantial effort is currently directed toward understanding the precise role each of these virulence factors plays in the pathogenesis of A. actinomycetemcomitans-mediated periodontal disease (18). The molecular analysis of virulence factors of A. actinomycetemcomitans has been hampered by the lack of efficient systems for genetic manipulation. Protocols for DNA transformation, generalized transduction, transposon mutagenesis, and allelic exchange are all unavailable. An electrotransformation system for A. pleuropneumoniae which uses pYG10, a 5.0-kb naturally occurring plasmid isolated from a multiply antibiotic-resistant strain of A. pleuropneumoniae 80-8141, does exist (8). Since both A. *
MATERIALS AND METHODS Bacterial strains and culture conditions. The bacterial strains used in these studies and their relevant characteristics are listed in Table 1. A. actinomycetemcomitans strains were routinely grown in Trypticase soy broth (Difco Laboratories, Detroit, Mich.) containing 0.6% yeast extract (Difco) and 0.04% sodium bicarbonate (Sigma Chemical Co., St. Louis, Mo.) (TSBYE broth) at 37°C in a GasPak containing a generator of 10% CO2 (BBL, Cockeysville, Md.). The antibiotics vancomycin and bacitracin (Sigma) were added to final concentrations of 5 and 75 jig/ml, respectively, unless otherwise indicated. Solid media were prepared by adding 1.5% Bacto-agar (Difco) to TSBYE. For long-term maintenance, A. actinomycetemcomitans strains were frozen at -70°C in TSBYE containing 100 ,ul of dimethyl sulfoxide per ml. E. coli strains were grown aerobically in YT broth (14) at 370C.
Corresponding author. 4621
4622
SREENIVASAN ET AL.
INFECT. IMMUN.
TABLE 1. Relevant characteristics of bacterial strains used in this study Bacterium
Strain
designation
A. actinomycetemcomitans
VT674 VT675 VT679 VT746 VT849
SUNY75S SUNY75R SUNY465 Y4 ATCC 29522
Zambon, SUNY, Buffalo, N.Y. Zambon Zambon Holt, UTHSC, San Antonio, Tex. ATCC, Md.
None None None None None
Smooth Rough Smooth Smooth Smooth
A. pleuropneumoniae
VT897
80-8141
Lalonde, Stanford, Calif.
pYGlO
Cmr
E. coli
DH5a
DH5a
Life Technologies, Gaithersburg, Md.
None
ApS Cms
a
Sourcea
Plasmid
Colony
SUNY, State University of New York; UTHSC, University of Texas Health Science Center.
Plasmids. The detailed restriction maps and construction of plasmids described in the text are described in detail elsewhere (9). In brief, plasmid pVT736-1, a naturally occurring 1.9-kb A. actinomycetemcomitans plasmid isolated from A. actinomycetemcomitans VT736, did not code for any selectable antibiotic phenotype tested. Therefore, pVT736-1 was unsuitable for developing a transformation system in A. actinomycetemcomitans without further manipulation. To obtain a selectable marker, attempts were made to clone the plasmid on an E. coli vector. Hence, pVT736-1 and pUC19 were digested with KpnI, mixed, and treated with T4 DNA ligase. The ligation mixture was used to transform E. coli DH5a, and transformants were selected on YT agar containing ampicillin and 5-bromo-4-chloro-3-indolyl-f-D-galactopyranoside (X-Gal). Recombinant colonies were screened for the presence of pUC19 containing a 1.9-kb KpnI fragment insert. Two populations of recombinant plasmids represented by pDL279 and pDL280 were obtained; they contain pVT736-1 in opposite orientations. Plasmid pDL281 was constructed by digesting plasmids pVT736-1 and pGEM7Zf(-) (obtained from Promega, Madison, Wis.) with XhoI and transforming the resulting ligation reaction into E. coli DH5a. Transformants were selected on YT agar containing ampicillin and X-Gal, and recombinant colonies were screened for the presence of pUC19 containing a 1.9-kb XhoI fragment insert. A single population of clones carrying pVT736-1 in one orientation was obtained. Plasmid pYG10, a naturally occurring plasmid from A. pleuropneumoniae, coding for chloramphenicol resistance (Cm%), was obtained from Guy Lalonde (8). Plasmid isolation. Five-milliliter cultures of A. actinomycetemcomitans transformants were screened for the presence of plasmids by a modification of the procedure of Anderson and McKay (1). Briefly, bacterial pellets were washed and resuspended in 6.7% sucrose-50 mM Tris HCl-10 mM EDTA, pH 8.0. Bacteria were incubated in the presence of lysozyme at 37°C for 20 min and lysed by the addition of sodium dodecyl sulfate. Nucleic acids precipitated by 2-propanol were washed twice with 70% ethanol and once in 99% ethanol and allowed to air dry. Plasmid DNA preparations were digested with restriction enzymes and analyzed on 0.8% Tris-borate-EDTA gels by using standard protocols (11). Large-scale plasmid preparations were obtained from 500-ml cultures of A. actinomycetemcomitans transformants, using a modification of the Anderson and McKay procedure, as described above. Plasmids from E. coli transformants were isolated by the alkaline lysis method of Birnboim and Doly (3) and analyzed on agarose gels following restriction enzyme digestion. Large-scale plasmid preparations for use in transformation were isolated
from 1-liter cultures and purified by cesium chloride density gradient centrifugation (11). Restriction enzyme analysis. All restriction enzymes were purchased from Life Technologies, Inc., Gaithersburg, Md., or Boehringer Mannheim Biochemicals, Indianapolis, Ind., and were used in accordance with the manufacturer's instructions. Restriction mapping of plasmids was carried out by using single and double enzyme digestions, and analysis was by electrophoresis on 0.8% Tris-borate-EDTA agarose gels. Plasmid transformation. (i). A. actinomycetemcomitans transformation. On day 1, 0.1 ml of frozen A. actinomycetemcomitans stock was inoculated into 1.0 ml of TSBYE broth and grown overnight as described above. The next day, the culture was diluted 1:20 and grown overnight again. On day 3, the A. actinomycetemcomitans cultures were diluted with an equal volume of fresh broth and allowed to grow to early logarithmic phase (3 h). The density of the cultures was read at 600 nm, and the bacteria were harvested by centrifugation in a Sorvall RT 6000B refrigerated centrifuge for 15 min at 3,000 rpm. The pellets were washed twice in half-volumes of ice-cold electroporation buffer (EPB; 15% glycerol, 272 mM sucrose, 2.43 mM K2HPO4, 0.57 mM KH2PO4, pH 7.4) and resuspended in fresh EPB to obtain a final optical density (OD) of 6.0 at 600 nm. This concentrated bacterial suspension was incubated on ice for 10 min prior to high-voltage pulse. Plasmid DNA was ethanol precipitated and resuspended in water at a concentration of 10 ,ug/ml (unless otherwise indicated) for use in transformation experiments. Fifty microliters of the concentrated bacterial suspension was mixed with the plasmid DNA in sterile cuvettes (Bio-Rad, Richmond, Calif.) with an interelectrode distance of 0.2 cm. High-voltage pulses were delivered with a GenePulser (Bio-Rad) set at 2,500 V, 200 Ql, and 25 ,uF. Following electroporation, bacteria were immediately transferred into 1.0 ml of warm TSBYE broth without antibiotics, and the broth was incubated for a period of 5 h (two doublings), unless otherwise indicated, to allow for expression of the transformed phenotype. Dilutions of transformed cultures were plated on TSBYE plates containing chloramphenicol (5 ,ug/ml), ampicillin (100 ,ug/ml), or spectinomycin (50 ,ug/ml), and cultures were incubated for 48 h as described above. P-Lactamase activity was assayed with Beta Lactam Reagent Disks (Marion Scientific, Kansas City, Mo.) in accordance with the manufacturer's instructions. Transformation efficiencies were expressed as CFU per milliliter per microgram of DNA from at least duplicate experiments. Frozen competent cells were prepared by rapidly freezing 250-,ul aliquots of A. actinomycetemcomitans cells resuspended in EPB to a density of 6.0 at 600 nm in a dry ice-ethanol bath for
VOL. 59, 1991
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
4623
TABLE 2. Transformation of A. actinomycetemcomitans strains TransformantsJml/,ug of DNA
Plasmid VT674
pYGlO pDL279 pDL280 pDL281
0 6 0.3 5
pUC19 -DNA
0 0
VT675
VT679
0 8 8 x 102 4.0 x 102 0.5 1.8 x 102 8.3 x 102 42
0 0
10 0
VT746
6.5 1.8 5.4 4.8
x x x x
50 0
ATCC 2952
102 5.14 x 104 104 1.8 x 1033 103 1.3 x 103 103 30
40 0
5 min and then storing at -70°C. For transformation, the vials were thawed on ice and allowed to stabilize on ice for an additional 10 min prior to electroporation. (ii) E. coli transformation. Frozen competent E. coli DH5a cells were purchased from Life Technologies, Inc., and transformed with gradient-purified plasmid DNA in accordance with the manufacturer's instructions. E. coli transformants were selected on YT agar supplemented with chloramphenicol (10 ,ug/ml), ampicillin (100 ug/lml), or spectinomycin (50 ,ug/ml). P-Lactamase activity was assayed as described above. Results are an average of duplicate experiments and are expressed as described above.
RESULTS Transformation of A. actinomycetemcomitans strains with several plasmids. All five A. actinomycetemcomitans strains examined were transformed with at least one of the plasmids in Table 2. The A. pleuropneumoniae plasmid pYGlO was the least efficient at transforming the A. actinomycetemcomitans strains. Even though pYGlO can replicate in E. coli (6), several A. actinomycetemcomitans isolates could not be transformed at detectable efficiencies. The shuttle plasmid pDL279 transformed with highest efficiency and transformed all five strains of A. actinomycetemcomitans. VT674, a spontaneous smooth-colony derivative, was transformed at a significantly lower efficiency than its rough-colony parent, VT675. The E. coli vector pUC19 transformed three of the A. actinomycetemcomitans strains at low efficiencies. Transforming plasmid could be isolated from all transformants except those transformed with pUC19. In addition, the amount of ,-lactamase produced by the pUC19 transformants was minimal compared with that of the other Apr transformants, suggesting that the bla gene was present in a low copy number. These data are consistent with the plasmid integrating into the A. actinomycetemcomitans chromosome; however, Southern hybridization was not performed to test this hypothesis. VT746 and ATCC 29522 were transformed with the highest efficiencies and with all of the plasmids used. A. actinomycetemcomitans ATCC 29522 had the shortest generation time (120 min) and grew to the highest OD; it is easily available to other investigators. Hence, it was chosen to optimize the transformation conditions. Construction of the shuttle vector pDL282. Plasmid pDL279 transformed all isolates of A. actinomycetemcomitans with high efficiency, making it the ideal plasmid for the construction of a shuttle vector. A spectinomycin resistance gene (isolated from Enterococcus faecalis [10]) was introduced into the unique SmaI site of pDL279 to obtain a recombinant plasmid with two antibiotic resistance markers. This new plasmid, designated pDL282, was used to transform E. coli DH5a. E. coli transformants were selected on
Hindill
--
Smal/Ndel
/
_
'Smal/Clal Xba
Kpni
Xba I
FIG. 1. Restriction endonuclease map of the E. coli-A. actinomycetemcomitans shuttle vector pDL282. spc, spectinomycin resistance gene; bla, ampicillin resistance gene; pVT 736-1, cryptic plasmid of A. actinomycetemcomitans VT736; E. coli ori, replication origin for pUC19.
YT agar containing spectinomycin. Ten transformants were arbitrarily picked and transferred onto YT agar plates containing both ampicillin and spectinomycin. Doubly resistant transformants were screened for the presence of a 5.7-kb recombinant plasmid, and construction was confirmed by restriction enzyme analysis. pDL282 was stably maintained in E. coli DH5a. A restriction map of pDL282 is shown in Fig. 1. The bla gene has a unique Scal site which permits cloning of insert DNA into bla with selection for Spr and Aps.
Effect on transformation of host modification of plasmid DNA. The source of plasmid DNA played a role in the transformation efficiency of A. actinomycetemcomitans. Plasmids pYGlO and pDL282 were regularly recovered from both E. coli and A. actinomycetemcomitans transformants regardless of the source of DNA (Table 3). pYGlO isolated from A. actinomycetemcomitans transformed both A. actinomycetemcomitans and E. coli more efficiently than DNA isolated from E. coli. On the other hand, pDL282 transformed E. coli more efficiently than it transformed A. actinomycetemcomitans regardless of the source of donor DNA. Interestingly, pDL282 isolated from A. actinomycetemcomitans transformed E. coli better than A. actinomycetemcomitans. Perhaps, the fact that DH5a is restriction minus plays a role in the higher transformation efficiency. Effect of stage of growth. Bacteria were grown to different stages in their growth cycle before electroporation to deterTABLE 3. Effect on transformation efficiency of modification of plasmid DNA by A. actinomycetemcomitans (Aa) ATCC 29522 and E. coli DH5a Plasmid
Source of plasmid DNA
Recipient
pYGlO
E. coli E. coli Aa Aa
E. coli Aa E. coli Aa
2.06 5.14 1.22 1.34
E. coli E. coli Aa Aa
E. coli Aa E. coli Aa
1.7 1.6 1.78 1.92
pDL282
Transformation efficiency/p.g of DNA x 106 x 105
x 107 x 106
x 106 x x x
105 105 104
4624
INFECT. IMMUN.
SREENIVASAN ET AL.
z
0.5
0
E c 0
105. 0.4
0 _
n
0.3
-.
0
0.2
-o
I-0 E
*
E IL
103
0
Transformants -0-- Optical density
I0
2
4
6
8
-0.1
'a0 C
0
10
Hour FIG. 2. Effect of stage of bacterial growth on transformation. Closed symbols indicate number of transformants per milliliter per microgram of DNA. Open symbols indicate OD at 600 nm at the time of harvesting. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
mine the optimal stage of growth for transformation. The highest transformation efficiencies were achieved when the bacteria were in the early logarithmic phase of growth (4 h) (Fig. 2). The number of transformants fell rapidly after this period of time. Effect of plasmid DNA concentration. The effect of varying plasmid DNA concentration on total number of transformants was assessed by electroporating aliquots of prepared bacterial samples with concentrations of DNA ranging from 0.1 to 10 ,ug/ml. A dose-response curve using different DNA concentrations is depicted in Fig. 3. The number of transformants increased with increasing DNA concentration and attained saturation at approximately 2.5 ,ug of DNA per ml. Since transformation efficiency is dependent on the concentration of DNA, the transformation efficiency remained fairly constant at 2.027 0.57 x 105 transformants per ,ug of DNA with DNA concentrations ranging from 0.5 to 3.5 ,ug/ml. Effect of bacterial cell density. Different cell densities of early-logarithmic-phase cultures of ATCC 29522 were assayed for ability to be transformed. A linear increase in the number of transformants with increased bacterial density occurred over an OD range of 1 to 6 at 600 nm (Fig. 4). The system appears to saturate at an OD of 6.0. The results show approximately a 10-fold increase in the number of transformants with increasing bacterial density from an OD of 1 to 6. Effect of pH of EPB. The pH of the EPB has been demonstrated to affect the transformation efficiency in several systems (6). Since the pH of our standard EPB could not be adjusted without affecting the ionic strength, we tested EPB at pH 7.4 with and without the addition of phosphates. Transformation efficiencies were comparable with both buffers (data not shown), suggesting that phosphates were not required for transformation. A series of EPB without phosphates with pHs of 4 to 8 were prepared, and bacterial cultures were resuspended in these buffers and transformed under identical conditions. An acidic pH was more favorable for transformation than a neutral pH (Fig. 5). At pH 5.5, the number of transformants recovered was twofold higher than
that at pH 7.4. Thus, the transformation system appears quite sensitive over a significant pH range. Effect of length of recovery time. Five hours was initially chosen for the recovery time as this time represents approximately two generations for A. actinomycetemcomitans and should allow for maximal expression of the antibiotic resistance marker on the plasmid. The optimal recovery period following electroporation was determined by varying the recovery time from 1 to 5 h (Fig. 6). The number of
±
E IL U. -
c 0
E 0
ca
DNA concentration (ug/ml) FIG. 3. Effect of DNA concentration on transformation. DNA concentrations of 0.1 to 10 ,ug/ml were added to the bacterial samples. Concentrations of DNA were added in a constant volume of 1 ,ul. Symbols indicate number of transformants per milliliter recovered at each DNA concentration tested. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
VOL. 59, 1991
106 z a L-
E
10-
C) 0 0
10 4-
0
a-
10 3,3
.
4625
Perhaps the plasmid was lost in the absence of selective pressure. Effect of freezing on competence. The long generation time of A. actinomycetemcomitans strains and the length of time required for recovery made for a long experimental day. Hence, the effect on transformation efficiency of freezing competent cells was determined. Since EPB contains 15% glycerol, it was reasoned that cells could be prepared to this stage in the procedure and then quickly frozen in a dry ice-ethanol bath prior to storage at -70°C. The transformation efficiencies of freshly prepared ceUs and frozen competent cells were comparable (data not shown). We have tested cells that have been frozen for up to 2 months and have not seen a loss in transformation efficiency. However, once the cells have been thawed, they cannot be refrozen without a , _._,_._,_._,_._,_.significant loss in competence. 8Statistical analysis. The mean and standard deviation used I. 2 . to describe the results were calculated from at least duplicate experiments, using the software package Lotus 1-2-3. Bacterial density at 600 nm
FIG. 4. Effect of bacterial density on transformation. Logarithmic-phase cultures were resuspended in EPB to densities corresponding to the OD indicated. Symbols indicate number of transformants per milliliter per microgram of DNA recovered at each density tested. Standard deviation bars are not visible for certain data points because the limits of the standard deviation bar do not extend beyond the limits of the square used to depict the data point.
transformants recovered was optimal at 3 h and gradually declined thereafter. CFU of the same samples plated on nonselective media were determined to ascertain whether this decline was due to a decrease in total bacterial numbers. The results shown in Fig. 5 indicate that the bacteria continued to grow for a period of 4 h prior to a decrease in CFU. The number of transformants decreased after 3 h, suggesting that cell death was not responsible for the decrease in the number of transformants observed at 4 h.
z 0
E 0
0 0 I-
4
5
6
7
8
pH FIG. 5. Effect of pH of the EPB on transformation. A. actinomycetemcomitans cells were resuspended in EPB without phosphates at each of the pHs indicated. Symbols indicate the number of transformants per milliliter per microgram of DNA recovered at each pH tested. Standard deviation bars are not visible for certain data points because the limits of the SD bar do not extend beyond the limits of the square used to depict the data point.
DISCUSSION In this report, we describe an efficient method for DNA transformation of A. actinomycetemcomitans and the construction of a stable chimeric plasmid that can be used as an E. coli-A. actinomycetemcomitans cloning vector. The plasmids utilized in these studies are the first plasmids shown to transform several A. actinomycetemcomitans strains. An earlier study reported the transformation of an A. actinomycetemcomitans strain with a broad-host-range plasmid (21). To our knowledge, this work has not yet been published in full and the transformation efficiency is unclear. The shuttle vector pDL282 was capable of transforming to a high efficiency all A. actinomycetemcomitans strains tested. pDL282 has several useful features as a potential cloning vector, such as the following: a unique restriction site within the bla gene for marker inactivation (ScaI), a relatively small size (5.7 kb), and a gene for the phenotype Spr for selection. Since pDL282 replicates equally well in E. coli, this vector will prove to be useful in the study of A. actinomycetemcomitans genes. The transformation of A. actinomycetemcomitans was accomplished by electroporation, and the protocol was standardized by using A. actinomycetemcomitans ATCC 29522 and the shuttle vector pDL282. In general, the electroporation protocols were based on methods available for A. pleuropneumoniae (8), Campylobacterjejuni (13), and E. coli (5). Observations on the growth and metabolic requirements for A. actinomycetemcomitans made in this laboratory significantly influenced the optimization of the electroporation protocol. Several factors were investigated to optimize the transformation protocol. The stage of growth of A. actinomycetemcomitans influenced electroporation efficiencies. A. actinomycetemcomitans cultures in the early logarithmic phase of growth transformed to the highest efficiency. Bacterial cells harvested in the stationary phase of the growth cycle remained transformable, although at much lower efficiencies, suggesting that A. actinomycetemcomitans was transformable at every stage of growth tested. These results indicate that ATCC 29522, with a generation time of 120 min, needs to undergo at least two cycles of cell division before it can attain its highest transformation efficiency. Varying the concentration of DNA added to an electroporation mixture yielded a typical saturation curve. We routinely obtained 105 transformants by using 2.5 jig of pDL282 DNA per ml. These data will be important for the
4626
SREENIVASAN ET AL.
INFECT. IMMUN. 10 11
lo'U
z
0
E L0
10 10 0 6I
*10 9
.108
-.
0 c 0
10 7
co
E
0 C
LL
105
10 3
* Transformants -0- Viable bactena
-06
I-
cc
0 CD 0 cc .
.0 .0 CI
-1n 5
0
1
2
3
4
5
6
Hour FIG. 6. Effect of length of recovery time following electroporation on transformation. Immediately following electroporation, A. actinomycetemcomitans cells were placed in prewarmed TSBYE without antibiotics and incubated for the times indicated. The closed symbols indicate the number of Spr transformants per milliliter per microgram of DNA, while the viable count on nonselective media is indicated by the open symbols.
construction of gene libraries and in isolating specific transformants occurring at low frequencies. Additional factors that influenced electroporation efficiency included the density of the bacterial suspension, the pH of the EPB, and the duration of the recovery period following electroporation. As expected, more transformants were recovered by increasing the bacterial density used for electroporation. Saturation occurred at a bacterial density of 6.0 at 600 nm. The pH of the EPB was also a factor affecting electroporation; twice the number of transformants were recovered with EPB at pH 5.5 versus pH 7.4. The recovery period following the delivery of the electrical pulse also plays a role in transformation efficiency. The maximum number of transformants was recovered following 3 h of recovery. In contrast, the total number of viable bacteria in the culture increased for 4 h after electroporation. These data suggest that the transformants may lose their plasmid in the absence of selective pressure. Transformation efficiency with logarithmic-phase bacterial cultures suspended in EPB and frozen to -70°C was comparable to that obtained with fresh bacterial cultures. The A. pleuropneumoniae plasmid pYGlO was able to transform several A. actinomycetemcomitans strains. This plasmid has been reported to transform A. pleuropneumoniae and E. coli (8), suggesting that pYGlO is a broad-hostrange plasmid. The role of E. coli and A. actinomycetemcomitans modification systems on transformation was examined by using the shuttle vectors pDL282 and pYG10. We routinely obtained equal numbers of A. actinomycetemcomitans or E. coli transformants (105 per ,ug of DNA) with pDL282 isolated from the heterologous strain. In the case of pYG10, plasmid DNA isolated from A. actinomycetemcomitans was more efficient at transforming both E. coli and A. actinomycetemcomitans, suggesting a role for an A. actinomycetemcomitans modification system. However, the modification system of A. actinomycetemcomitans does not limit transformation of E. coli with either pDL282 or pYG10. The structures of pDL282 and pYGlO isolated from A. actinomycetemcomitans and E. coli transformants were found to be intact by
single and double restriction enzyme digestion, suggesting that the plasmids did not undergo any deletions or rearrangements following transformation. The structural stability of these plasmids is of significance in applying these plasmids for routine DNA manipulations in either A. actinomycetemcomitans or E. coli. Surprisingly, pDL282 isolated from A. actinomycetemcomitans transformed A. actinomycetemcomitans 10-fold less than pDL282 isolated from E. coli and transformed into E. coli. The reason for this result is unclear. The convenience of using DNA obtained from an E. coli donor for high-efficiency transformation of A. actinomycetemcomitans will prove invaluable in utilizing the extensive tools of E. coli genetics for the analysis of A. actinomycetemcomitans genes. Our laboratories are investigating the amount of passenger DNA that can be cloned and stably maintained in pDL282. We are also adapting pDL282 as a potential transposon delivery vehicle. The availability of these genetic tools will accelerate a molecular understanding of A. actinomycetemcomitans virulence factors. REFERENCES 1. Anderson, D., and L. L. McKay. 1983. Simple and rapid method for isolating large plasmid DNA from lactic streptococci. Appl. Environ. Microbiol. 46:549-552. 2. Birkedal-Hansen, H., P. W. Claufield, Y. M. Wannemuehier, and R. Pierce. 1982. A sensitive screening assay for epitheliotoxins produced by oral microorganisms, abstr. 125. Abstr. Annu. Meet. J. Dent. Res. 61:192. 3. Birnboim, H. C., and J. Doly. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res. 7:1513-1521. 4. Christersson, L. A., B. Albini, J. J. Zambon, U. M. E. Wikesjo, and R. J. Genco. 1987. Tissue localization of Actinobacillus actinomycetemcomitans in human periodontitis. i. Light, immunofluorescence and electron microscopic studies. J. Periodontol. 58:529-539. 5. Dower, W. J., J. F. Miller, and C. W. Ragsdale. 1988. High efficiency transformation of E. coli by high voltage electroporation. Nucleic Acids Res. 16:6127-6145. 6. Dunny, G. M., L. L. Lee, and D. J. LeBlanc. 1991. Improved electroporation and cloning vector system for gram-positive bacteria. Appl. Environ. Microbiol. 57:1194-1201.
VOL. 59, 1991
TRANSFORMATION OF A. ACTINOMYCETEMCOMITANS
7. Genco, R. J., and J. Slots. 1984. Host responses in periodontal diseases. J. Dent. Res. 63:441-451. 8. Lalonde, G., J. F. Miller, and L. S. Tompkins. 1989. Transformation of Actinobacillus pleuropneumoniae and analysis of R factors by electroporation. Am. J. Vet. Res. 50:1957-1960. 9. LeBlanc, D. J., L. N. Lee, A. R. Abu-AM-Jaibat, P. K. Sreenivasan, and P. M. Fives-Taylor. Submitted for publication. 10. LeBlanc, D. J., L. N. Lee, and J. M. Inamine. 1991. Cloning and nucleotide base sequence analysis of a spectinomycin adenyltransferase AAD(9) determinant from Enterococcus faecalis. Antimicrob. Agents Chemother. 35:1804-1810. 11. Maniatis, T., E. F. Fritsch, and J. Sambrook. 1982. Molecular cloning: a laboratory manual. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 12. Meyer, D., P. Sreenivasan, and P. Fives-Taylor. 1991. Evidence for invasion of a human oral cell line by Actinobacillus actinomycetemcomitans. Infect. Immun. 59:2719-2726. 13. Miller, J. F., W. J. Dower, and L. S. Tompkins. 1988. Highvoltage electroporation of bacteria: genetic transformation of Campylobacter jejuni with plasmid DNA. Proc. Natl. Acad. Sci. USA 85:856-860. 14. Miller, J. H. 1972. Experiments in molecular genetics. Cold Spring Harbor Laboratory, Cold Spring Harbor, N.Y. 15. Nowotny, A., U. H. Behling, B. Hammond, C.-H. Lai, M. Listgarten, P. H. Pham, and F. Sanavi. 1982. Release of toxic microvesicles by Actinobacillus actinomycetemcomitans. Infect. Immun. 37:151-154.
4627
16. Robertson, P. B., M. Lantz, P. T. Marucha, K. S. Kornman, C. L. Trummel, and S. C. Holt. 1982. Collagenolytic activity associated with Bacteroides species and Actinobacillus actinomycetemcomitans. J. Periodontal. Res. 17:275-283. 17. Schenker, B. J., M. E. Kushner, and C.-C. Tsai. 1982. Inhibition of fibroblast proliferation by Actinobacillus actinomycetemcomitans. Infect. Immun. 38:986-992. 18. Slots, J., and R. J. Genco. 1984. Black-pigmented Bacteroides species, Capnocytophaga species, and Actinobacillus actinomycetemcomitans in human periodontal disease: virulence factors in colonization, survival, and tissue destruction. J. Dent. Res. 63:412-421. 19. Tsai, C.-C., B. J. Shenker, J. M. DiRienzo, D. Malamund, and N. S. Taichman. 1984. Extraction and isolation of a leukotoxin from Actinobacillus actinomycetemcomitans with polymyxin B. Infect. Immun. 43:700-705. 20. Van Dyke, T. E., E. Bartholomew, R. J. Genco, J. Slots, and M. J. Levine. 1982. Inhibition of neutrophil chemotaxis by soluble bacterial products. J. Periodontol. 53:502-508. 21. Yip, J., D. Furgang, E. Weidman, P. Goncharoff, D. Figurski, and D. H. Fine. 1990. Transfer of the promiscuous plasmid RK2 from E. coli to Actinobacillus actinomycetemcomitans (Aa). Program Abstr. 68th Int. Assoc. Dent. Res., abstr. 1959. J. Dent. Res. 69:353. 22. Zambon, J. J. 1985. Actinobacillus actinomycetemcomitans in human periodontal disease. J. Clin. Periodontol. 12:1-20.
