INTERNATIONAL JOURNAL of SYSTEMATIC BACTERIOLOGY Vol. 24, No.

2
April 1974, p. 221-224 Printed in U.S.A.
Copyright 0 1974 International Association of Microbiological Societies

Mechanism of Action of Eosin-Methylene Blue Agar in the

Differentiation of Escherichia coli and Enterobacter
aerogenes
R. S. HORVATH and M. E. ROPP

Department of Biology, Bowling Green State University, Bowling Green, Ohio 43403

The mechanism involved in the action of eosin-methylene blue agar (EMB) in

differentiating Escherichia coli from Enterobacter aerogenes was investigated
using prototrophic strains of these organisms and a Lac- auxotroph, E. coli
X-96 1. The final pH of the EMB agar resulting from growth of these organisms
at the expense of lactose was shown to play a critical role in the differential
action. An EMB complex, which formed under acidic conditions, appeared to be
involved in the differential action of this medium. The molecular weight of the
complex indicated an eosin t o methylene blue ratio of 1 t o 1. Ultraviolet and
infrared spectral data indicated the occurrence of an amide bond between the
dyes when complexing occurred. On this basis, a mechanism for complexing of
eosin and methylene blue, under acidic conditions, was proposed.

The use of eosin-methylene blue (EMB) agar The mechanism by which EMB agar differ-
as a differential medium for the colon-typhoid- entiated E. coli and Enterobacter aerogenes
dysentery group has become well established remained unknown. T o demonstrate the mecha-
since its introduction by Holt-Harris and nism of this differential action, the effect of pH
Teague (2) in 19 16. The medium contained the was investigated. The molecular structure of the
dyes eosin and methylene blue, which yielded a dye complex was determined, and a mechanism
sharp distinction between lactose- and non- for formation of the dye complex was pro-
lact ose-fermenting organisms. posed.
Two years later, Levine (3) employed a
modification of the original EMB medium and
concluded that the modified EMB could be MATERIALS AND METHODS
used t o differentiate lactose fermenters from Bacterial strains and media. The strain of E.
non-lac tose ferm enters. aerogenes was supplied by M. M. Brent, and the
Although the medium had been used exten- culture of E. coli wild type was obtained from S.
sively since its introduction, the mechanism of Harmon, both of the Department of Biology, Bowling
its differential action was not known. Wilson Green State University. The culture of E. coli X-961, a
(6), in a review of differential staining by lactose-negative auxo troph, was supplied by R. Curtiss
mixtures of eosin and methylene blue, called I11 of Oak Ridge Laboratories, Oak Ridge, Tenn.
attention to a dye complex which formed Stock cultures were maintained on cystine Trypti-
case agar stabs (Difco Laboratories, Detroit, Mich.) at
under acidic conditions. It was proposed that
15 C. Experimental cultures were grown at 37 C for
such a complex might be responsible for the 18 to 24 h in broth containing 1% lactose, 1%
differential action of EMB agar. peptone, and 0.2% dipotassium phosphate. EMB broth
Wynne, Rode, and Hayward (7) assumed that contained 0.04% eosin Y (Matheson Coleman and Bell
the differential action of EMB was a direct Co., Cincinnati, Ohio) and 0.0065% methylene blue
function of the acid produced during fermenta- (National Aniline and Chemical Co., N.Y.) in addition
tion. Pathogens belonging to the genera Sal- to the above constituents. Solid media were prepared
monella and Shigella were compared with the by adding 2% agar to the basal formula. All media
nonpathogens Escherichia coli and two En tero- were sterilized by autoclaving at 121 C and 15 pounds
bacter species, and it was concluded that the pressure for 15 min.
Methods. A Corning pH meter, model 7 (Corning
pathogens could not ferment lactose t o produce Scientific Instruments) with Beckman pH electrodes
a sufficiently low pH to result in colored (Beckman Co., Chicago, Ill.) was employed for pH
colonies on EMB, whereas the nonpathogens determinations. The relative pH produced by each
could reduce the pH t o a level at which colored organism on solid media was determined by use of the
colonies resulted. indicator bromothymol blue. The medium contained
221
222 HORVATH AND ROPP INT. J. SYST. BACTERIOL.

0.0016% bromothymol blue, 1% peptone, 0.2% bromothymol blue as an indicator of final pH

dipotassium phosphate, 2% agar, and either 1% of the culture medium indicated that E. coli did
lactose, 1% glucose, or no sugar. The color of the dye
produce a strongly acidic condition during
at the end of 24 h of growth at 37 C served as angrowth at the expense of lactose, whereas E.
indication of final pH. The effect of pH on formation
aerogenes yielded a much less acidic medium.
of the dye complex was investigated by employing
The reaction exhibited by E. coZi X-961 was
buffer concentrations of 0.0, 0.2, 0.4, 0.5, 0.6, 0.7,
similar to that obtained with E. aerogenes. This
0.8, 0.9, and 1.0% in the EMB medium. All solutions
observation was verified by measurement of pH
were examined visually for the presence of a precipi-
tate, and the pH was determined as above. The pH after growth of E. coZi and E. aerogenes in
lactose broth. E. coli reduced the pH of this
necessary to complex the dyes was also determined by
titration of EMB broth with 0.01 N HC1. medium from the initial pH of 7.1 t o 4.9 in 24
The molecular weight of the dye complex was h, whereas E. coli X-961 yielded a pH of only
determined by the molal freezing point depression ( 5 )
6.4, and E. aerogenes reduced the pH to 5.3 in
using d-camphor as the solvent. The dye complex was
the same time period. These results were
prepared by reducing the pH to 3.5 with 1.0 N HC1.
consistent with those obtained by Wynne et al.
The precipitate was harvested by centrifugation at
10,000 X g for 10 min and washed 10 times with (7) and established a quantitative difference in
the final pH resulting from fermentation of
distilled water to remove traces of unreacted eosin and
methylene blue. For determination of the freezinglactose by the microbial species in question.
point depression, a Buchi melting point instrument Titration with 0.01 N HCl indicated that the
(Rinco Instrument Co., Greenville) was employed. complex formed at a pH of about 4.9, thereby
Ultraviolet spectra of the complex, eosin methylene
establishing the significance of the pH differ-
blue, and a noncomplexed eosin-methylene blue ence observed above.
solution were determined using a Beckman DBG If pH is the critical factor in formation of the
spectrophotometer (Beckman Co., Chicago, Ill.). The
dye complex, growth of E. aerogenes on
infrared spectrum of the complex was examined as a
KBr macropellet with a Perkin-Elmer 337 double- unbuffered EMB agar should result in a reaction
typical of E. coli. The effect of pH on
beam spectrophotome ter (Perkin-Elmer Co ., Norwalk,
formation of the dye complex was observed by
Conn.), as were the spectra of eosin and methylene
blue. varying the buffer concentration from 0.0 to
1.0%. Growth of E. aerogenes on EMB agar
containing n o buffer did result in the formation
RESULTS AND DISCUSSION of a green metallic sheen, and the pH of the
corresponding EMB broth was found t o be 4.8.
Because the differentiation of pathogenic This sheen was not produced by E . aerogenes
microorganisms from nonpathogens by EMB on media containing a phosphate buffer. E.
agar was dependent on the amounts of acidic coli, however, was capable of producing the
products resulting from the fermentation of typical reaction at buffer concentrations up to
lactose (7), it appeared that this mechanism 0.5%. Thus it appeared that pH was the critical
might also account for the differentiation of E. factor in the formation of the green sheen
coli and E. aerogenes. Results obtained with observed with E. coli on EMB agar.

0.1 1 I I I 1 1 1 1 1 1 I I
0.0 I
3m m am 110 yo 1w ¶40 ¶an ¶a0 110 ?oo lI*

W A V l LlNOlY CNW

FIG. 1. Ultraviolet spectra of the dye complex (solid line) and the eosin-methylene blue uncomplexed
solution (broken line).
 VOL. 24,1974 EMB AGAR AND ORGANISM DIFFERENTIATION 223

O'O t

iii
04
O.*
1.0

2ooo
i
I
1-
I 1

1aao
V
I
1100
1

1100
I
lo00
I
900
I

a00
I

700
I
(00

CY-l

FIG. 2. Infrared spectrum of the eosin-methylene blue d y e complex.

DYI C O M P L I X

FIG. 3. Proposed mechanism o f complexing o f eosin and methylene blue under acidic conditions.

The metallic appearance of E. coli colonies The structure of the complex was inferred
seemed to result from the formation of a dye from ultraviolet and infrared spectral data. As
complex under acidic conditions. The molecu- seen in Fig. 1, a significant difference in the
lar weight of this complex was determined to ultraviolet spectrum of the complex was ob-
be 1,016, indicating that the dyes eosin and served at 236 nm when compared to the
methylene blue occurred in this complex in a spectrum of the uncomplexed solution of eosin
ratio of 1 t o 1. and methylene blue. This new peak indicated
224 HORVATH AND ROPP INT. J. SYST. BACTERIOL.

the presence of an amide structure in the University and from the National Science Foundation
complex, and it would result from a n + 7 ~ * Institutional Committee of Bowling Green State
transition of the carbonyl group (4). University.
The occurrence of an amide bond in the dye
complex was also indicated by an examination REPRINT REQUESTS
of the infrared spectrum of this compound
(Fig. 2). Amides are characterized by a strong Address reprint requests to: Dr. R. S . Horvath,
absorption in the 1,695 to 1,630 cm-' region Department of Biology, Bowling Green State Univer-
due to C = 0 stretching (1). The complex sity, Bowling Green, Ohio 43403.
exhibited a peak at 1,700 cm-l. Interpretation
of this peak as indicating an amide bond was
supported by a peak which occurred at 1,275 LITERATURE CITED
cm-l. This band would result from a CNH
1 . Colthup, N. B., L. H. Daly, and S. E. Wiberley.
vibration in which the nitrogen and hydrogen
1964. Introduction t o infrared and raman spectros-
atoms moved in the same direction relative to copy, p. 263-265, 332, 382-384. Academic Press
the carbon atom (1). A peak of 640 cm-l was Inc., New York.
interpreted as being characteristic of an N H 2. Holt-Harris, J. E., and 0. Teague. 1916. A new
out-of-plane wag whereas a peak at 590 cm-l culture medium for the isolation of Bacillus
was indicative of a C = 0 out-of-plane bend. typhosus from stools. J. Infect. Dis. 18596-600.
Thus, the formation of the dye complex 3. Levine, M. 1918. Differentiation of B. coli and B.
appeared to be dependent on the pH of the aerogenes on a simplified eosin-methylene blue
medium and involved formation of an amide agar. J. Infect. Dis. 23:4347.
bond between eosin and methylene blue, as 4. Rao, C. N. R. 1967. Ultraviolet and visible
spectroscopy, chemical applications, p. 193.
shown in Fig. 3 . The occurrence of this dye Plenum Press, New York.
complex during growth of E. coli on EMB agar 5. Schwarz, J. C. P. 1964. Physical methods in organic
appeared to account for the ability of this chemistry, p. 250-252. Holden-Day, Inc., San
medium t o differentiate between E. coli and E. Francisco.
aeroge nes. 6. Wilson, T. M. 1907. Chemical and staining proper-
ties of derivatives of methylene blue and eosin. J.
Exp. Med. 9:645-670.
ACKNOWLEDGMENTS 7. Wynne, E. S., L. J. Rode, and A. E. Hayward.
1942. Mechanism of the selective action of
This work was supported by a grant from the eosin-methylene blue agar on the enteric group.
Faculty Research Committee, Bowling Green State Stain Technol. 17:ll-20.