FEMS Microbiology Letters 185 (2000) 169^174
www.fems-microbiology.org
Multilocus enzyme analysis in aerobic and anaerobic bacteria using
gel electrophoresis^nitrocellulose blotting
Marie-Laure Combe, Jean-Franc°ois Lemeland, Martine Pestel-Caron, Jean-Louis Pons *
G.R.A.M. (Groupe de Recherche sur les Antimicrobiens et les Microorganismes, EA 2656), U.F.R. Mëdecine-Pharmacie de Rouen, 22 Boulevard Gambettta,
76183 Rouen, France
Received 11 August 1999; received in revised form 14 February 2000; accepted 22 February 2000
Abstract
An optimized multilocus enzyme electrophoresis method, which involves polyacrylamide^agarose gel electrophoresis followed by
electrophoretic transfers on nitrocellulose sheets, was developed for the analysis of enzyme polymorphism in several aerobic and anaerobic
bacterial species including Staphylococcus aureus, Streptococcus pneumoniae, S. agalactiae, Klebsiella pneumoniae and K. oxytoca,
Clostridium bifermentans and C. sordellii, and Prevotella bivia. Serial electrophoretic transfers (during 5^15 min each) from a single
polyacrylamide gel could be achieved for most enzymes studied, and allowed an increased definition of enzyme bands on nitrocellulose as
compared to migration gels. Four enzymes, which could not be blotted in such conditions, could still be stained in gels after blotting. Thus,
the method allowed the combined analysis of several enzymes after a single gel electrophoresis separation. The analysis of enzyme
polymorphism in the various species studied raised the interest of polymorphic loci such as esterase or glutamic-oxaloacetic transaminase
for epidemiologic studies. The method characterized a genetic diversity of enzyme loci of S. pneumoniae higher than previously reported,
and is thus convenient for the analysis of genetic relationships between related isolates. Since the present method reduces the tediousness of
multilocus enzyme electrophoresis and requires experimental conditions that are not specific for the bacterial population studied, it may be
proposed for rapid population genetics analysis of a wide variety of bacteria. ß 2000 Federation of European Microbiological Societies.
Published by Elsevier Science B.V. All rights reserved.
Keywords : Multilocus enzyme polymorphism; Electrophoresis; Genetic diversity
1. Introduction
Multilocus enzyme electrophoresis (MLEE) has been
frequently used in the past 10 years to study bacterial
genetic diversity : the characterization of electrophoretic
variants of a given enzyme re£ects allelic variations at
the corresponding structural gene locus, and the combined
analysis of electrophoretic polymorphism of a set of metabolic enzymes de¢nes electrophoretic types which correspond to genotypes [1]. Since (i) metabolic enzymes are
not subject to direct selective pressure and thus to evolutionary convergence and (ii) individual enzyme loci are
independent features, the multilocus analysis of enzymes
(typically encoded by housekeeping genes) provides phylogenetic information and founds implications in bacterial
* Corresponding author. Tel. : +33 235-14-84-52;
Fax: +33 235-14-84-44 ; E-mail : [email protected]
Abbreviations : MLEE, multilocus enzyme electrophoresis
population genetics [2], systematics [3] and epidemiology
[4].
Electrophoretic characterization of enzyme allelic polymorphism is generally performed using starch gel electrophoresis [1], or polyacrylamide gel electrophoresis [5].
However, the electrophoretic resolution in starch gels is
not always optimal, and the analysis of sizeable numbers
of enzymes in large numbers of bacterial strains, as required in population genetics, may be rather time-consuming and tedious with such methods. The need for an alternative method allowing multiple enzyme analysis from
single electrophoretic gels was previously pointed out,
but the method proposed, based on the application of
indicator agarose gels on separation polyacrylamide gels,
was only evaluated for Staphylococcus aureus strains [6].
Studying genetic relationships within the genus Klebsiella,
we described another method based on the combination of
polyacrylamide^agarose gel electrophoresis with serial
electrophoretic transfers onto nitrocellulose sheets, which
allows the simultaneous analysis of several enzymes from a
single separating gel [7]. The aim of the present work was
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 ( 0 0 ) 0 0 0 9 7 - 5
FEMSLE 9334 29-3-00
170
M.-L. Combe et al. / FEMS Microbiology Letters 185 (2000) 169^174
to establish that this time-saving method is suitable for the
analysis of enzyme polymorphism in a wide variety of
bacteria.
2. Materials and methods
2.1. Bacterial strains and preparation of protein extracts
Bacterial strains included in this study belonged to the
Gram-positive aerobic species Staphylococcus aureus,
Streptococcus pneumoniae [8], and Streptococcus agalactiae, the Gram-negative aerobic species Klebsiella pneumoniae and K. oxytoca [7], the Gram-positive anaerobic species Clostridium bifermentans and C. sordellii [9], and the
Gram-negative anaerobic species Prevotella bivia (Table
1).
Protein extracts were prepared from bacterial cells as
previously described [7]. Brie£y, washed cells were collected by centrifugation and were disrupted by sonication
¢ve times for 20 s at 0³C (after preliminary treatment with
lysostaphin for 45 min at 37³C in the case of S. aureus).
The crude protein extracts were collected after removing
cellular debris by centrifugation; their protein concentration was determined by the method of Bradford [10] and
adjusted to 20 g l31 by speed-vacuum evaporation and
resuspension of the pellet with the required volume of
100 mM sodium phosphate bu¡er (pH 7.2).
Table 1
Bacterial strains studied
Bacterial species
Total number
of isolates
Source (number)
of isolates
Staphylococcus aureus
Streptococcus pneumoniae
Streptococcus agalactiae
67
86
67
Klebsiella pneumoniae
42
Klebsiella oxytoca
24
Clostridium bifermentans
23
blood (67)
nasopharynx (86)
blood (56)
cerebrospinal £uid (11)
urine (31)
blood (2)
expectoration (2)
suppuration (6)
ATCCa (1)
urine (17)
blood (2)
expectoration (2)
suppuration (3)
I.P.b (8)
feces (2)
lake sediments (11)
unknown (2)
I.P. (19)
feces (2)
unknown (1)
suppuration (6)
ATCC (1)
unknown (6)
Clostridium sordellii
22
Prevotella bivia
13
a
b
ATCC, American Type Culture Collection.
Institut Pasteur Paris (Laboratoire des Anaërobies).
2.2. Polyacrylamide^agarose gel electrophoresis,
electrophoretic transfer on nitrocellulose, and detection
of enzymes
The MLEE method used in this study, initially developed for Klebsiella sp. strains [7], was further evaluated for
the species mentioned above: thin horizontal slab gels
(1 mm thick) contained 0.8% (w/v) agarose, 6% (w/v) polyacrylamide, 1.6% (w/v) N,NP-methylene-bisacrylamide in
gel bu¡er (37.5 mM Tris^30 mM glycine, pH 8.7), and
were polymerized by 0.03% (v/v) TEMED and 0.012%
(w/v) ammonium persulfate. Agarose (indubiose
A37HAA) was purchased from IBF Biotechnics (Villeneuve la Garenne, France) and the other gel components
from Eastman Kodak Co. (Rochester, NY, USA). Electrophoresis was performed in a 2117 Multiphor System
(LKB Products, Bromma, Sweden) with 10 mM Tris^330
mM glycine bu¡er (pH 8.7) as electrode bu¡er. Bacterial
extracts (6 Wl) were loaded on the gel with a rubber guide,
and a voltage of 4 V cm31 was applied for 30 min; electrophoresis was then run at 8 V cm31 under refrigeration
until the dye marker (bromophenol blue) had run 8.5 cm
(approximately 3 h).
Successive copies of the separation gel were made by
electrophoretic transfers onto nitrocellulose sheets (BA
85, Schleicher and Schuell, Keene, NH, USA), using a
horizontal `sandwich' of the gel with one sheet of nitrocellulose between ¢lter papers soaked in electrode bu¡er
(1 mM Tris^33 mM glycine, pH 8.7) and the graphite
cathode and anode plates of a Multiphor II Novablot
System (LKB Products). Electrophoretic transfer from
the gel to each nitrocellulose sheet was run for 5 min at
25 mA or 15 min at 100 mA, depending on the enzyme
tested. The following enzymes were selectively stained with
their corresponding substrate according to established
methods [1,5]: alcohol dehydrogenase (ADH), adenylate
kinase (ADK), L-naphthylacetate esterase (EST), glucose6-phosphate dehydrogenase (G6P), glutamate dehydrogenase (GDH), glutamic-oxaloacetic transaminase (GOT),
hexokinase (HEX), lactate dehydrogenase (LDH), malate
dehydrogenase (MDH), nucleoside phosphorylase (NSP),
6-phosphogluconate dehydrogenase (6PG), peptidase
(PEP) (L-phenylalanyl-L-leucine peptidase, or L-valyl-L-leucine peptidase for Clostridium sp. or L-phenylalanyl-L-proline peptidase for P. bivia), phosphoglucose isomerase
(PGI) and phosphoglucomutase (PGM).
2.3. Statistical analysis of genetic diversity of enzyme loci
Statistical analysis of genetic diversity of enzyme loci
was achieved using a computer program designed by
J.-C. Pi¡aretti (Lugano, Switzerland) from the original
program of Selander et al. [1] for MLEE. Distinctive electromorphs of each enzyme were numbered in order of
increasing rate of anodal migration, electrophoretic types
(ETs) were recorded as the combination of electromorphs
FEMSLE 9334 29-3-00
M.-L. Combe et al. / FEMS Microbiology Letters 185 (2000) 169^174
171
Fig. 1. Electrophoretic polymorphism of NSP in Prevotella sp. strains, revealed on nitrocellulose after electrophoretic blotting (25 mA for 5 min).
terial species considered. Thus, to assess the general suitability of the present method, we analyzed a set of widely
diverse bacterial species. The detectability of metabolic
enzymes after polyacrylamide gel electrophoresis followed
by sequential electrophoretic transfer on nitrocellulose
sheets is presented in Table 2. Compared to 3 mm thick
gels previously used for enterobacteria [5] or Clostridium
perfringens [11], the 1 mm thick polyacrylamide^agarose
gels presently used could be loaded with samples of small
volume (6 Wl instead of 200 Wl) and allowed a sharper
de¢nition of enzyme bands. In addition, their handling
characteristics make the moving step on electrode graphite
plates for electrophoretic blotting easy.
The transferability of enzymes detected in gels was then
tested, trying to perform short-time transfers allowing
for each isolate, and genetic diversity coe¤cients (H)
were calculated for each enzyme locus from allele frequencies, using the formula H = (13gxi 2 )[n/(n31)], where xi
is the frequency of the ith allele and n is the number of
ETs.
3. Results and discussion
3.1. Reliability of the method
The aim of the present work was to examine if the
MLEE method that we originally developed for K. pneumoniae and K. oxytoca [7] was only applicable to the study
of these bacteria, or could be proposed whatever the bac-
Table 2
Detection of metabolic enzymes in aerobic or anaerobic bacterial species after polyacrylamide gel electrophoresis and electrophoretic transfer on nitrocellulose
Enzyme
S. aureus
gel/nc
ADH
ADK
EST
G6P
GDH
GOT
HEX
LDH
MDH
NSP
6PG
PEP
PGI
PGM
+b /+
+/3
+/+
+/+
+/+
3
3
+/+
3
+/+
+/+
+/+
+/+
3
a
S. pneumoniae
S. agalactiae
K. pneumoniae, K. oxytoca
C. bifermentans, C. sordellii
gel/nc
gel/nc
gel/nc
gel/nc
gel/nc
+/+
+/3
+/+
+/+
+/+
+/3
+/3
+/+
3
+/+
+/+
+/+
+/+
+/3
+/+
+/3
+/+
3
+/+
+/3
+/3
+/+
3
+/+
3
+/+
+/+
+/3
+/+
+/3
+/+
3
+/+
+/3
3
+/+
+/+
+/+
+/+
+/+
+/+
+/3
3c
+/3
+/+
3
+/+
+/3
3
3
3
+/+
3
+/+
+/+
+/3
3
+/3
+/+
3
+/+
3
+/3
3
+/+
+/+
3
+/+
+/+
3
a
Gel, polyacrylamide running gel; nc, nitrocellulose.
Detection (+) or no detection (3) of enzyme on running gel or on nitrocellulose.
c
Lack of detection of enzyme by both methods in all strains studied.
b
FEMSLE 9334 29-3-00
P. bivia
172
M.-L. Combe et al. / FEMS Microbiology Letters 185 (2000) 169^174
Fig. 2. Electrophoretic polymorphism of PEP in C. bifermentans and C. sordellii strains, revealed on nitrocellulose after electrophoretic blotting (100
mA for 15 min).
multiple enzyme analysis from single running gels. The
electrophoretic transfer conditions required did not vary
with the bacterial species: whatever the bacterial species
examined, a reliable detection of enzymes on nitrocellulose
was obtained after transfers performed at 25 mA for 5 min
(G6P, MDH, NSP, 6PG, PGI) (Fig. 1) or at 100 mA for
15 min (ADH, EST, GDH, LDH, PEP) (Fig. 2). Up to
¢ve transfers could be achieved without a¡ecting the enzyme detection in gels or on nitrocellulose sheets. Staining
on nitrocellulose led to an increased de¢nition of enzyme
bands as compared to staining of the electrophoretic gel,
especially for NSP and PGI which cannot be stained directly on running gels but require an agar overlay [1].
Electrophoretic transfer of ADK, GOT, HEX, and
PGM was not successful with such short-time transfer
conditions, whatever the bacterial species, but these enzymes could still be detected in gels after serial electrophoretic transfers for transferable enzymes (Fig. 3).
Thus, combining the analysis of one enzyme on the electrophoresis gel and several enzymes on nitrocellulose after
transfers from the same gel, the method made it possible
to routinely analyze four enzymes after one electrophoretic
run, and thus to establish signi¢cant multilocus enzyme
pro¢les from only two or three running gels, resulting in
a signi¢cant time saving.
3.2. Analysis of enzyme locus polymorphism
The method was further evaluated for its ability to analyze genetic diversity of enzyme loci in various bacteria.
The genetic diversity of the loci analyzed, expressed by
both number of alleles and genetic diversity coe¤cients,
is presented in Table 3. Both parameters are necessary to
re£ect genetic polymorphism of a given locus, since di¡erent numbers of electromorphs in two enzymes may, in
fact, re£ect comparable genetic diversity, depending on
Table 3
Genetic diversity of 14 enzyme loci in aerobic or anaerobic bacterial species
Enzyme locus
S. aureus
S. pneumoniae
S. agalactiae
K. pneumoniae,
K. oxytoca
C. bifermentans,
C. sordellii
P. bivia
Mean
ADH
ADK
EST
G6P
GDH
GOT
HEX
LDH
MDH
NSP
6PG
PEP
PGI
PGM
0.50a (3)b
0.11 (2)
0.66 (3)
0.39 (3)
0.00 (1)
^
^
0.53 (4)
^
0.00 (1)
0.11 (2)
^
0.11 (2)
^
-c
0.32
0.55
0.69
0.31
0.66
0.26
^
^
0.43
0.27
0.61
0.07
^
0.48
0.38
0.49
^
0.50
0.46
0.25
0.25
^
0.07
^
^
0.00
0.39
0.63
^
0.58
^
0.64
0.41
^
^
0.77
0.48
^
0.76
0.00
^
^
0.58
^
^
0.68
0.69
^
^
^
0.65
^
0.44
0.73
0.80
^
0.69
^
^
0.69
^
^
^
0.00
0.51
^
0.54
0.39
^
0.54
0.42
0.57
0.54
0.47
0.56
0.26
0.39
0.89
0.36
0.38
0.59
0.22
0.60
(3)
(3)
(5)
(3)
(6)
(4)
(2)
(2)
(3)
(2)
(3)
(2)
(4)
(4)
(4)
(2)
(3)
(2)
(1)
(3)
(6)
(6)
(4)
(4)
(7)
(4)
(7)
(1)
a
Genetic diversity coe¤cient (H) per enzyme locus.
Number of alleles per enzyme locus.
c
Enzyme not detected or not extensively studied in the corresponding bacterial species.
b
FEMSLE 9334 29-3-00
(4)
(5)
(5)
(5)
(3)
(6)
(7)
(3)
(5)
(1)
(2)
(5)
(2)
(4.0)
(2.8)
(4.0)
(4.0)
(3.7)
(4.8)
(3.0)
(3.5)
(4.0)
(2.7)
(2.0)
(4.5)
(2.3)
(5.0)
M.-L. Combe et al. / FEMS Microbiology Letters 185 (2000) 169^174
173
Fig. 3. Electrophoretic polymorphism of GOT in S. pneumoniae strains, revealed on polyacrylamide migration gel.
the relative frequency of mobility variants: for example,
NSP and PEP shared about the same genetic diversity
coe¤cient in P. bivia (0.51 and 0.54 respectively), although
they exhibited two and ¢ve alleles respectively. In contrast,
two enzyme loci exhibiting the same number of alleles may
present major di¡erences of genetic diversity (for example
EST and G6P in S. aureus).
Some enzyme loci, such as EST, GOT or GDH, were
found to be polymorphic in most species and are thus
especially informative in epidemiologic studies, as we previously showed for the characterization of an epidemic
multiresistant clone of S. pneumoniae in children attending
a day-care center [8]. Conversely, some other loci such as
PGI exhibited a limited polymorphism in most species;
such loci, re£ecting phylogenetic stability of chromosomal
genes, may thus be considered in evolutionary genetics
studies. The genetic diversity of loci such as NSP varied
with the bacterial species considered (genetic diversity coe¤cient ranging from 0.00 in S. aureus to 0.65 in C. sordellii), re£ecting di¡erent levels of mutation or recombination amongst species.
The polymorphism of the enzyme loci analyzed was also
examined in view of previous studies performed with
starch gel electrophoresis. The numbers of alleles of loci
analyzed in S. aureus and in S. agalactiae were in keeping
with previous reports [12,13]: for example, PGI, HEX,
and PGM of S. agalactiae exhibited one, two, and three
alleles respectively in both the present work and a previous
study based on 277 isolates [13]. On the other hand, our
data concerning S. pneumoniae, although obtained from a
bacterial population restricted to isolates from children in
a single day-care center [8], revealed a higher genetic diversity for ¢ve enzyme loci than previously reported in a
study of pneumococcal isolates originating from di¡erent
continents [14]. Thus, owing to a better resolution of
discrete mobility variants, our method is especially well
suited to the analysis of relationships between potentially
related isolates.
MLEE is acknowledged as a reference method for
studying population genetics of bacteria, and has been
shown to be reliable for analyzing the clonal structure of
bacterial populations [2], global molecular epidemiology
and spread of resistant bacteria [12,14], clonal characteristics of pathogenic bacteria [13], or taxonomic relationships between closely related species [9]. However, it is
sometimes considered tedious and time-consuming, since
population genetics studies require the analysis of sizeable
numbers of loci in large numbers of isolates. The method
evaluated in the present work decreases the number of gels
which have to be run for multilocus enzyme polymorphism analysis, resulting in signi¢cant time saving, and
requires conditions independent of the bacterial population to be analyzed. Thus, it may be proposed for population genetics analyses concerning various Gram-positive
or Gram-negative, aerobic or anaerobic bacteria.
Acknowledgements
We wish to thank Richard Sesboue« for helpful technical
discussions, and R.D. Arbeit, G. Leluan and Emmanuelle
Martin for supplying strains.
References
[1] Selander, R.K., Caugant, D.A., Ochman, H., Musser, J.M., Gilmour,
M.N. and Whittam, T.S. (1986) Methods of multilocus enzyme electrophoresis for bacterial population genetics and systematics. Appl.
Environ. Microbiol. 51, 873^884.
[2] Ochman, H. and Selander, R.K. (1984) Evidence for clonal population structure in Escherichia coli. Proc. Natl. Acad. Sci. USA 81, 198^
201.
FEMSLE 9334 29-3-00
174
M.-L. Combe et al. / FEMS Microbiology Letters 185 (2000) 169^174
[3] Boerlin, P., Rocourt, J. and Pi¡aretti, J.-C. (1991) Taxonomy of the
genus Listeria by using multilocus enzyme electrophoresis. Int. J.
Syst. Bacteriol. 41, 59^64.
[4] Mun¬oz, R., Co¡ey, T.J., Daniels, M., Dowson, C.G., Laible, G.,
Casal, J., Hakenbeck, R., Jacobs, M., Musser, J.M., Spratt, B.G.
and Tomasz, A. (1991) Intercontinental spread of a multiresistant
clone of serotype 23F Streptococcus pneumoniae. J. Infect. Dis. 164,
302^306.
[5] Goullet, P. and Picard, B. (1990) Characterization of enterobacteria
by esterase-speci¢c activity pro¢les. J. Gen. Microbiol. 136, 431^
440.
[6] John, M.A. and Hussain, Z. (1994) Multilocus enzyme electrophoresis using ultrathin polyacrylamide gels. J. Microbiol. Methods 19,
307^313.
[7] Combe, M.-L., Pons, J.-L., Sesboue«, R. and Martin, J.-P. (1994)
Electrophoretic transfer from polyacrylamide gel to nitrocellulose
sheets, a new method to characterize multilocus enzyme genotypes
of Klebsiella strains. Appl. Environ. Microbiol. 60, 26^30.
[8] Pons, J.-L., Mandement, M.-N., Martin, E., Lemort, C., Nouvellon,
M., Mallet, E. and Lemeland, J.-F. (1996) Clonal and temporal patterns of nasopharyngeal penicillin-susceptible and penicillin-resistant
Streptococcus pneumoniae strains in children attending a day care
center. J. Clin. Microbiol. 34, 3218^3222.
[9] Pons, J.-L., Ageron, E., Mangeney, N., Tessedre, A.-C., Niel, P.,
[10]
[11]
[12]
[13]
[14]
Leluan, G. and Grimont, P.A.D. (1997) Genetic relationships of
Clostridium bifermentans and Clostridium sordellii analyzed by multilocus enzyme electrophoresis and DNA-DNA hybridization. Syst.
Appl. Microbiol. 20, 381^386.
Bradford, M.M. (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of
protein-dye binding. Anal. Biochem. 72, 248^254.
Pons, J.-L., Combe, M.-L. and Leluan, G. (1994) Multilocus enzyme
typing of human and animal strains of Clostridium perfringens.
FEMS Microbiol. Lett. 121, 25^30.
Musser, J.M. and Kapur, M. (1992) Clonal analysis of methicillinresistant Staphylococcus aureus strains from intercontinental sources:
association of the mec gene with divergent phylogenetic lineages implies dissemination by horizontal transfer and recombination. J. Clin.
Microbiol. 30, 2058^2063.
Quentin, R., Huet, H., Wang, F.-S., Geslin, P., Goudeau, A. and
Selander, R.K. (1995) Characterization of Streptococcus agalactiae
strains by multilocus enzyme genotype and serotype: identi¢cation
of multiple virulent clone families that cause invasive neonatal disease. J. Clin. Microbiol. 33, 2576^2581.
Sibold, C., Wang, J., Henrichsen, J. and Hakenbeck, R. (1992) Genetic relationships of penicillin-susceptible and -resistant Streptococcus pneumoniae strains isolated on di¡erent continents. Infect. Immun. 60, 4119^4126.
FEMSLE 9334 29-3-00