CURRENTMICROBIOLOGY,Vol. 10 (1984), pp.

23-28
Current
Microbiology
9 Springer-Verlag 1984

A General Method for Plasmid Isolation in Lactobacilli

Todd R. Klaenhammer
Department of Food Science, 339 Schaub Hall, North Carolina State University, Raleigh, North Carolina 27650, USA

Abstract. A simple procedure for rapid isolation and detection of plasmid D N A from
Lactobacillus species is described. Using an alkaline-detergent lysis method, plasmid D N A was
released and characterized from cells treated with either mutanolysin or l y s o z y m e for 1 h at 0~
Treatment of cells with either e n z y m e at 37~ for 1 h was detrimental to plasmid isolation and
characterization in some Lactobacillus species. The procedure was effective with small
volumes of cells and allowed rapid characterization of plasmid D N A in Lactobacillus
plantarum, Lactobacillus acidophilus, Lactobacillus helveticus, and Lactobacillus bulgaricus
strains.

Recent progress in the genetics of lactic acid bacter- tive using either l y s o z y m e or mutanolysin to
ia, particularly the group N streptococci, has estab- achieve cell lysis. Although mutanolysin was more
lished the critical involvement of plasmid D N A in effective, plasmid detection in the lactobacilli was
numerous activities related to fermentative per- dependent more on the temperature conditions es-
formance [2, 5, 10]. Unfortunately, plasmid genet- tablished for e n z y m e activity than on the specific
ics of the lactobacilli have not kept pace with the enzyme used.
advances in molecular genetics, and little informa-
Materials and Methods
tion is presently available on plasmid determinants
Strains and culture conditions. All Lactobacillus strains were
or genetic transfer mechanisms for a bacterial genus
propagated in MRS broth (Difco Laboratories, Detroit, MI) at
critical to many industrial food fermentations. Un- 37~ and maintained as described previously [11]. Escherichia
dermining progress in Lactobacillus genetics has coli V517 was propagated in nutrient broth (BBL, Cockeysville,
been the difficulty in achieving cell lysis with lyso- MD) at 37~ Growth conditions and methods for plasmid
zyme [4, 18] and developing reliable procedures for isolation from E. coli V517 were described previously [12, 15].
Strains used in this study, and their sources, are given in Table 1.
plasmid D N A isolation. Although plasmid D N A has
been reported in Lactobacillus acidophilus [13, 21], Cell lysis. Log phase cells from 200-ml MRS broth cultures (0.1%
Lactobacillus reuteri [21], Lactobacillus casei [3], inoculum, 37~ were harvested by centrifugation when the
Lactobacillus fermentum [7], and Lactobacillus optical density (OD) at 650 nm reached 1.0. Pelleted cells were
washed once in 20 ml of TES buffer [16] and resuspended in 2 ml
helveticus [20, 21], detection has been limited to a
of 25% (wt/vol) sucrose in 50 mM Tris-HCl, pH 7.5. Two 925-~1
select number of strains within each species [13, aliquots were removed and placed in an ice bath. Mutanolysin
21]. Considering the ubiquity of plasmid D N A (Miles Laboratories, Elkhart, IN) or lysozyme (Sigma Chemical
throughout Gram-positive and Gram-negative bac- Company, St. Louis, MO) was added to one aliquot to a final
teria, the inability to demonstrate plasmid D N A concentration of 75 /zg/ml (75 /zl of a 1-mg/ml suspension in
throughout Lactobacillus species could reflect the sterile distilled H20). Sterile distilled H20 (75 ~1) was added to
the second aliquot and both aliquots placed at 37~ Samples
ineffectiveness of current plasmid isolation proce- containing 50/zl were removed from each aliquot at 0, 30, and 60
dures for this genus. min and added to cuvettes containing 2 ml of 50 mM Tris-HCl,
The reported sensitivity of Gram-positive bac- pH 7.0. After mixing, the OD65 o w a s determined. The percent
teria to N-acetylmuramidase from Streptomyces relative lysis at each sampling period was calculated from the
globisporus, mutanolysin [14, 17, 22], suggests that OD65o of control samples (no enzyme) and OD65 o of enzyme-
treated samples. The OD65 o of control aliquots approximated 0.5
D N A yields from lactobacilli could be enhanced and 0.6 and remained stable throughout the 60 min incubation.
upon employing this e n z y m e in plasmid isolation
procedures. This report describes a general method Plasmid isolation. Lactobacillus cultures were propagated over-
for plasmid isolation in Lactobacillus species effec- night (9-12 h) in 10 ml of MRS broth. A sample of 4 ml of culture
24 CURRENT MICROBIOLOGY,Vol. 10 (1984)

Table 1. Bacterial strains Table 2. Effect of mutanolysin or lysozyme on the percent

relative lysis a of Lactobacillus species
Organism Source
Lactobacillus species
Escherichia coli V517 F. L. Macrina et al. [15] Time
Lactobacillus acidophilus NCSU a [13] (min) L. aci- L. plan- L. hel-
C7 at dophilus tarum veticus
Lactobacillus acidophilus NCSU [13] Treatment 37~ PA3 C7 YIT 352 261 103
PA3
Lactobacillus acidophilus Miles Laboratories, Elkhart, Mutanolysin 30 98.4 98.2 79.7 91.4 96.9 92.3
MI IN (75/xg/ml) 60 99.9 99.9 92.2 99.9 99.9 99.9
Lactobacillus acidophilus J. L. Johnson et al. [8] Lysozyme 30 50.0 51.4 0.6 0.8 44.8 38.5
11088 (75/xg/ml) 60 76.6 69.1 15.6 1 4 . 5 36.7 53.8
Lactobacillus bulgaricus 10 NCSU
Lactobacillus helveticus 261 NCDO 9 aPercent relative lysis: calculated as percent loss in optical
Lactobacillus helveticus/ju- NCDO density at 650 nm.
gurt 103
LactobaciIlus lactis 1438 NCDO (also ATCC " 12315)
centrifugation, the aqueous phase above a white protein inter-
Lactobacillus plantarum NCDO
face was removed, added to a second Eppendorf tube, and
352
extracted once with 600/~1 of chloroform:isoamyl alcohol (24:1,
Lactobacillus plantarum Yakult Institute for Microbio-
vol/vol). After 5 rain at room temperature, the sample was again
YIT0068 logical Research, Tokyo,
centrifuged for 5 rain. After removal of the aqueous phase (500
Japan
/xl), 1 ml of cold (-70~ 95% ethanol was added and the sample
a Culture Collection at North Carolina State University, Depart- held at -70~ for 1 h. DNA was pelleted by centrifugation for I0
ment of Food Science. min in an Eppendorf microfuge. Ethanol was removed by
o National Collection of Dairy Organisms, University of Read- aspiration and the pellets dried under vacuum for 10 min. DNA
ing, Shinfield, England. pellets were dissolved in 20 /xl TES buffer and analyzed by
c American Type Culture Collection, Rockville, MD. agarose gel electrophoresis [16].
Eiectrophoresis was conducted on 0.7% agarose gels (Sea-
kern ME agarose, FMC Corp., Rockland, MD) in Tris-borate
was placed in a sterile polypropylene culture tube (17 x 100 mm, buffer, pH 8.2 [16]. Samples of 5-20 #1 were mixed with an equal
Fisher Scientific, Raleigh, NC) and centrifuged at 6000 g for 10 volume of agarose beads [19], loaded onto gels and subjected to
min at 4~ The pellet was resuspended in 10 ml of fresh MRS electrophoresis at 100 V for 3-5 h. Gels were stained and
broth and the cell suspension incubated for 2 h at 37~ Cells photographed as described previously. Agarose beads were
were then harvested by centrifugation and resuspended in 1 ml of prepared by adding 20 mg agarose and 1 mg bromophenol blue to
cold (4~ 25% (wt/vol) sucrose in 50 mM Tris-HCl, 5 mM Na2 10 ml of 10 mM Tris-HC1, 20 mM Na2 EDTA, 10% glycerol, pH
EDTA, pH 7.5. After holding the cell suspension for 10 rain in an 8.0. The mixture was autoclaved, cooled to room temperature,
ice bath, 75 /xl of mutanolysin or lysozyme (1 mg/ml in 50 mM and then passed repeatedly (3x) through a tuberculin syringe.
Tris-HCl, 5 mM Na2 EDTA, pH 7.5) was added and the sample Agarose beads were stored at 4~
held in an ice bath for 1 h. For some experiments, cell-enzyme
samples were held at 37~ for 1 h. Cells were pelleted by
centrifugation and the supernatant discarded. A lysis solution Results
(500/xl) was added directly to the cell pellet and the pellet gently
disrupted with a plastic disposable pipet. During the lysis proce- Lysis by mutanolysin or lysozyme. L y s i s o f t h r e e
dure, the pH of the cell suspension, following addition of the Lactobacillus species was examined after treatment
lysis solution, should approximate pH 12.2. To achieve this pH, w i t h e i t h e r m u t a n o l y s i n o r l y s o z y m e ( T a b l e 2). I n
the volume of 10 N NaOH added to the lysis solution can be all six s t r a i n s e x a m i n e d , m u t a n o l y s i n a c t e d q u i c k l y
varied from 8 to 12/xl. The lysis solution was a modification of
a n d w a s m o r e e f f e c t i v e t h a n l y s o z y m e in g e n e r a t i n g
that reported by Kado and Liu [9] and was composed of 50 mM
Tris, 5 mM Na2 EDTA, 50 mM glucose, and 3% SDS. Immedi- osmotically fragile cells. Lactobacillus plantarum
ately before use, 10 p.1 of 10 N NaOH was mixed with 1 ml of s p e c i e s w e r e p a r t i c u l a r l y i n s e n s i t i v e to l y s o z y m e .
lysis solution. The sample was heated at 62~ for 1 h and then These data are consistent with previous reports
allowed to cool slowly to room temperature (approximately 15 n o t i n g t h e r e s i s t a n c e o f l a c t o b a c i l l i t o l y s o z y m e [4,
min). A quantity of 50/xl 2 M Tris, pH 7.0, was added and mixed, 18] a n d s e n s i t i v i t y o f G r a m - p o s i t i v e b a c t e r i a to
followed by addition of 70 /xl 5 M NaC1. The sample was
transferred by disposable pipet to a 1.5-ml polypropylene Eppen- m u t a n o l y s i n [14, 17, 22].
doff centrifuge tube. The appearance of lysates at this step were
often viscous, but remained turbid. Deproteinization of the Piasmid detection. N o t i n g t h e e f f e c t i v e n e s s o f m u -
sample was conducted with 500/xl of 3% NaCl-saturated phenol tanolysin on generation of osmotically fragile lacto-
(redistilled, Bethesda Research Laboratories, MD). After addi-
b a c i l l i , L. acidophilus P A 3 a n d C 7 w e r e e x a m i n e d
tion of phenol, the tube was shaken vigorously for 3 s and held at
room temperature for 5 rain. To facilitate phase separation, 300 for plasmid DNA by a cleared lysate procedure for
/xl chloroform was added (repeat shaking for 3 s) prior to l a c t o b a c i l l i [13] m o d i f i e d b y s u b s t i t u t i o n o f 7 5 / x g / m l
centrifugation for 5 rain in an Eppendorf 5414 microfuge. After m u t a n o l y s i n f o r l y s o z y m e . F i g u r e 1 s h o w s t h a t L.
T. R. Klaenhammer: Plasmid Method for Lactobacilli 25

Fig. 1. Electrophoresis of plasmid DNA from Lactobacillus

acidophilus PA3 (B) and C7 (C). DNA was isolated as described Fig. 2. Electrophoresis of plasmid DNA from Lactobacillus
previously [13] using mutanolysin for 1 h at 37~ prior to lysis acidophilus C7. Conditions of plasmid isolation were as de-
with sodium dodecyl sulfate. Track A contains mobility plasmids scribed in Materials and Methods using: mutanolysin, 0~ 1 h
from Escherichia coli V517. (A and F); lysozyme, 37~ 1 h (C); mutanolysin, 37~ 1 h (D);
and lysozyme, 0~ 1 h (E). Track B contains mobility plasmids
from Escherichia coli V517. "Chr" indicates contaminating
chromosomal fragments (also indicated on Figs. 4 and 5).
acidophilus PA3 (Tract B) contained two plasmid
D N A species corresponding to the 13.7 and 6.3
megadalton molecules previously reported [13].
Alternatively, L. acidophilus C7 (Tract C), reported
devoid of plasmid D N A using l y s o z y m e [13], again
showed no detectable plasmid D N A when mutano-
lysin was employed. Therefore, despite excellent
cell lysis conditions, plasmid D N A could not be
extracted from C7 using mutanolysin in the cleared
lysate procedure described previously [13]. H o w e v -
er, L. acidophilus C7 does contain a single 38
megadalton plasmid detectable when mutanolysin
incubation was conducted at 0~ for 1 h rather than
at 37~ for 1 h (Fig. 2A). The effect of incubation
temperature, during l y s o z y m e or mutanolysin di-
gestion, on plasmid yields of L. acidophilus C7 is
shown in Fig. 2 (C-F). Using the procedure de-
scribed herein, incubation at 37~ for 1 h with
mutanolysin or l y s o z y m e resulted in a failure to
detect the 38 megadalton plasmid. Alternatively,
incubation at 0~ for 1 h with either e n z y m e result- Fig. 3. Electrophoresis of plasmid DNA from Lactobacillus
ed in a detectable plasmid band. These data demon- helveticus 261. Conditions of plasmid isolation were as described
strate that extended 37~ incubation during cell wall in Materials and Methods using: lysozyme, 37~ 1 h (A);
digestion is detrimental to isolation of the large mutanolysin, 37~ 1 h (B); mutanolysin, 0~ 1 h (C); lysozyme,
plasmid in L. acidophilus C7. Although l y s o z y m e 0~ 1 h (D). Track F contains CsC1-EB purified plasmid DNA.
Tracks E and G contain mobility plasmids from Escherichia coli
was not as effective as mutanolysin in preparing V517.
osmotically fragile cells in C7, plasmid D N A was
readily isolated f r o m cells digested with 75 /~g/ml Further studies were conducted with L. helve-
l y s o z y m e at 0~ for 1 h. ticus 261 and L. plantarum 352 to determine the
26 CURRENTMICROBIOLOGY,Vol. 10 (1984)

Fig. 5. Electrophoresis of plasmid DNA from Lactobacillus

lactis 1438 (A); Lactobacillus acidophilus M1 (B); Lactobacillus
Fig. 4. Electrophoresis of plasmid DNA from Lactobacillus plantarum YIT0068(C); Lactobacillus bulgaricus 10 (D); Lacto-
plantarum 352. Conditions of plasmid isolation were as de- bacillus helveticus 103 (E); Lactobacillus acidophilus 11088(F);
scribed in Materials and Methods" using: lysozyme, 37~ 1 h and Escherichia coli V517 (G). Conditions of plasmid isolation in
(A); lysozyme, 0~ 1 h, (B); mutanolysin, 37~ 1 h (C); and lactobacilli were as described in Materials and Methods using
mutanolysin, 0~ 1 h (D). Track E contains mobility plasmids mutanolysin 0~ 1 h.
from Escherichia coli V517.
750 Ixg/ml gave results identical to those with mu-
effects of incubation temperature and type of lytic tanolysin (data not shown). The stair-step bands
enzyme on plasmid yields. Plasmid D N A from L. isolated from L . p l a n t a r u m 352 and L . h e l v e t i c u s
h e l v e t i c u s 261 was isolated under all treatment 261 during e n z y m e incubation at 37~ for 1 h were
conditions. Either lysozyme or mutanolysin at 0~ carried with covalently closed circular (ccc) D N A
for 1 h gave good plasmid D N A yields (Fig. 3C and through cesium chloride-ethidium bromide (CsC1-
D) with patterns comparable to cesium chloride- EB) purification (data not shown). Therefore, these
ethidium bromide purified plasmid D N A from L. molecules maintained their superhelical conforma-
h e l v e t i c u s 261 (Fig. 3F). Incubation at 37~ for 1 h tion and were not simply degradative products
with lysozyme or mutanolysin resulted in a "stair- generated solely by nicking of ccc DNA. This
step" pattern between plasmids of molecular banding pattern was strikingly similar to topoiso-
weights (x 106) ranging from 1 to 10 (Fig. 3A and B). mers of ccc D N A generated by nicking and religa-
It was also observed that the lysozyme sample tion of nicked molecules [6]. The specific cause and
incubated at 37~ for 1 h was missing the two large nature of these stair-step bands in the lactobacilli
molecular weight plasmids. Similar results were were not further investigated. H o w e v e r , the results
obtained with L . p l a n t a r u m 352 (Fig. 4). Mutanoly- demonstrated that extended 37~ incubation with
sin at 0~ for 1 h resulted in good yields, whereas mutanolysin or lysozyme altered the banding pat-
incubation at 37~ yielded a stair-step pattern and tern of small plasmids and either eliminated large
loss of large plasmid molecules. Use of lysozyme at plasmids or reduced them to nondetectable levels.
75/~g/ml was totally ineffective for plasmid isolation
from L . p l a n t a r u m 352. Failure to detect plasmid Plasmid distribution in Lactobacillus species. Six
D N A could be directly attributed to the resistance strains of lactobacilli were examined for plasmid
o f L . p l a n t a r u m 352 to 75/xg/ml lysozyme (Table 2). DNA using the mutanolysin (O~ 1 h) procedure to
However, increasing the lysozyme concentration to determine the usefulness of this method for rapid
T. R. Klaenhammer: Plasmid Method for Lactobacilli 27

plasmid purification and identification in this genus. further, provides the first evidence for the presence
With the exception of L. lactis 1438, plasmid DNA of plasmid DNA in L. plantarum. The availability of
was detected in all strains examined (Fig. 5). a rapid and effective method for plasmid isolation in
the lactobacilli will, hopefully, facilitate future de-
velopments on the plasmid genetics of this genus.
Discussion
ACKNOWLEDGMENTS
Conventional lysis and plasmid purification tech- This work was supported, in part, by the North Carolina Dairy
niques have routinely been ineffective for lactoba- Foundation and Miles Laboratories, Elkhart, IN. Paper 8933 of
cilli. In a previous study from our laboratory, only the Journal Series of the North Carolina Agricultural Research
one of eight strains of L. acidophilus was shown to Service.
harbor plasmid DNA [13]. Similarly, Vescovo et al. Literature Cited
[21] identified plasmid DNA in only 19% of L. 1. Chassy, B. M. 1976. A gentle method for the lysis of oral
acidophilus, 27% of L. helveticus, and 5% of L. streptococci. Biochemical and Biophysical Research
bulgaricus strains examined. Furthermore, plasmid Communications 68:603-608.
DNA was not detected in nine strains of L. plan- 2. Chassy, B. M., Gibson, E. M., Giuffrida, A. 1978. Evidence
for plasmid-associated lactose metabolism in Lactobacil-
tarum examined. Difficulty in isolation of plasmid lus casei subsp, casei. Current Microbiology 1"141-144.
DNA from lactobacilli could be attributed directly 3. Chassy, B. M., Gibson, E., Giuffrida, A. 1976. Evidence for
to the lysozyme insensitivity of this genus [18]. extrachromosomal elements in Lactobacillus. Journal of
Noting the level of resistance, investigators have Bacteriology 127:1576-1578.
reported methods to optimize lysozyme action [1,4] 4. Chassy, B. M., Giuffrida, A. 1980. Method for lysis of Gram-
positive, asporogenous bacteria with lysozyme. Applied
and thus enhance cell lysis and plasmid DNA and Environmental Microbiology 39:153-158.
yields. However, these methods continue to em- 5. Davies, F. L., Gasson, M. J. 1981. Reviews of the progress
ploy a 37~ incubation for 1 h to facilitate cell wall of dairy science: genetics of lactic acid bacteria. Journal
digestion by lysozyme [4]. Data presented in this of Dairy Research 48"363-376.
study demonstrated that extended incubation at 6. Depew, R. E., Wang, J. C. 1975. Conformational fluctua-
tions of DNA helix. Proceedings of the National Academy
37~ is unnecessary and, in fact, may be detrimen- of Sciences, USA 72:4275-4279.
tal to plasmid isolation in some Lactobacillus spe- 7. Ishiwa, H., Iwata, S. 1980. Drug resistance plasmids in
cies. A similar reponse to 37~ incubation with Lactobacillus fermentum. Journal of General Applied
lysozyme was previously reported for a 30 Mdal Microbiology 26:71-74.
plasmid from Streptococcus lactis C2 [12]. The 8. Johnson, J. L., Phelps, C. P., Cummins, C. S., London, J.,
Gasser, F. Taxonomy of the Lactobacillus acidophilus
events reponsible for plasmid loss or alteration group. International Journal of Systematic Bacteriology
noted for the lactobacilli are unknown. However, it 30:53-68.
is apparent that failure to detect plasmid DNA in a 9. Kado, C. I., Liu, S. T. 1981. Rapid procedure for detection
number of Lactobacillus strains could be the direct and isolation of large and small plasmids. Journal of
result of the extended incubation times at 37~ Bacteriology 145:1365-1373.
10. Kempler, G . M . , MeKay, L . L . 1981. Biochemistry and
typically used for plasmid isolation procedures. genetics of citrate utilization in Streptococcus lactis
Compensation for this effect could employ 0~ as subsp, diacetylactis. Journal of Dairy Science 64:1527-
was done in this study, or, alternatively, a substan- 1539.
tial reduction of incubation time using an effective 11. Klaenhammer, T. R., Kleeman, E. G. 1981. Growth charac-
teristics, bile sensitivity, and freeze damage in colonial
cell wall lytic enzyme, such as mutanolysin.
variants of Lactobaeillus acidophilus. Applied and Envi-
The method described in this study was suit- ronmental Microbiology 41:1461-1467.
able using either mutanolysin or lysozyme and was 12. Klaenhammer, T. R., MeKay, L. L., Baldwin, K. A. 1978.
rapid, convenient, and effective. Detergent lysis Improved lysis of group N streptococci for isolation and
under alkaline conditions appeared very effective in rapid characterization of plasmid deoxyribonucleic acid.
the release of plasmid DNA from cells, but did not Applied and Environmental Microbiology 35:592-600.
13. Klaenhammer, T. R., Sutherland, S. M. 1980. Detection of
totally eliminate minor contaminating linear chro- plasmid deoxyribonucleic acid in an isolate of Lactobacil-
mosomal DNA or nicked, open circular, molecules. lus acidophilus: Applied and Environmental Microbiology
However, plasmid patterns were consistent with 39:671-674.
CsC1-EB purified samples, and comparisons of 14. Kondo, J. K., McKay, L. L. 1982. Mutanolysin for improved
plasmid content can easily be made without routine lysis and rapid protoplast formation in dairy streptococci.
Journal of Dairy Science 65:1428-1431.
density gradient purification of DNA samples. The 15. Maerina, F. L., Kopeeko, D. J., Jones, K. R., Ayers, D. J.,
method was applicable to L. acidophilus, L. bulgar- MeCowen, S . M . 1978. A multiple plasmid containing
icus, L. helveticus, and L. plantarum species, and Escherichia coli strain: convenient source of size refer-
28 CURRENTMICROBIOLOGY,VO1. 10 (1984)

ence plasmid molecules. Plasmid 1:417-420. of DNA fragment lengths from agarose gels. Analytical
16. Meyers, J. A., Sanchez, D., Elwell, L. P., Falkow, S. 1976. Biochemistry 115:113-122.
Simple agarose gel electrophoretic method for the identifi- 20. Smiley, M. B., Fryder, V. 1978. Plasmids, lactic acid produc-
cation and characterization of plasmid deoxyribonucleic tion, and N-acetyl-D-glucosamine fermentation in Lacto-
acid. Journal of Bacteriology 127:1529-1537. bacillus helveticus subsp, jugurti. Applied and Environ-
17. Monsen, T. J., Holm, S. E., Burman, L. G. 1983. A general mental Microbiology 35:777-781.
method for cell lysis and preparation of deoxyribonucleic 21. Vescovo, M., Bottazzi, V., Sarra, P. G., Dellaglio, F. 1981.
acid from streptococci. FEMS Microbiology Letters Evidence of plasmid deoxyribonucleic acid in Lactobacil-
16:19-24. lus. Microbiologica 4:413-419.
18. Neujahr, H . Y . , B6rstad, B., Logardt, I. 1973. Factors 22. Yokogawa, K., Kawata, S., Takemura, T., Yoshimura, Y.
affecting the resistance of Lactobacillus fermenti to lyso- 1975. Purification and properties of lyric enzymes from
zyme. Journal of Bacteriology 116:694-698. Streptomyces globisporus 1829. Agricultural and Biologi-
19. Schaffer, H. E., Sederoff, R. R. 1981. Improved estimation cal Chemistry 39:1533-1543.