See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/228482363

Manipulation of different media and methods for cost-effective

characterization of Escherichia coli strains collected from different habitats

Article in Pakistan Journal of Botany · September 2006

CITATIONS READS

10 3,124

3 authors:

Rubina Arshad Shafqat Farooq

Pakistan Atomic Energy Commission AM agricon (SMC-PVT) Limitedon
32 PUBLICATIONS 273 CITATIONS 72 PUBLICATIONS 1,182 CITATIONS

SEE PROFILE SEE PROFILE

Syed Shahid Ali

University of Lahore
98 PUBLICATIONS 784 CITATIONS

SEE PROFILE

All content following this page was uploaded by Shafqat Farooq on 29 May 2014.

The user has requested enhancement of the downloaded file.

Pak. J. Bot., 38(3): 779-789, 2006.

MANIPULATION OF DIFFERENT MEDIA AND METHODS

FOR COST-EFFECTIVE CHARACTERIZATION OF
ESCHERICHIA COLI STRAINS COLLECTED
FROM DIFFERENT HABITATS
RUBINA ARSHAD, SHAFQAT FAROOQ AND SAYYED SHAHID ALI*
Nuclear Institute for Agriculture and Biology (NIAB), P.O. Box 128,
Jhang Road, Faisalabad-Pakistan and *Department of Zoology, University of the Punjab,
Quaid-e-Azam Campus, Lahore, Pakistan
Abstract
Rapid and reliable identification of about 400 isolates of Escherichia coli collected from soil,
water, plants and animal faeces was made by membrane filtration (MF), culture media and
biochemical methods. Utilization of three types of selective and differential agar media
(MacConkey, Eosin Methylene Blue: EBM and Endo agar) rather than one increased the chances of
successful isolation/identification. The identified bacteria were re-confirmed through the use of
biochemical (IMViC) tests, and diagnostic kits made for this purpose. The results obtained after
comparative studies indicated that isolation media and methods used in the present study are not
only simple and reliable for large-scale bacterial identification but at the same time are more cost-
effective compared to commercially available diagnostic kits. It is anticipated that by using such
methods, isolation and identification of E. coli can be done effectively without importing expensive
diagnostic kits, which is most often difficult especially in the developing countries and thus
becomes limiting factor for microbiological investigations.

Introduction
Escherichia coli, an important bacterial species belonging to the family
Enterobacteriaceae, is the most frequently encountered microorganism in the food
industry. Its presence has also been detected in soil, plants and animal faeces and in water
where it could serve as one of the factors affecting animal and human health (Cundell,
1981). Detection and accurate identification of E. coli is therefore, enormously important,
for which different media and methods are often used depending upon the source of
bacterial collection. For example, for the recovery of E. coli and other coliforms from
drinking water, membrane filtration (MF) method with an agar medium containing a
chromagen and a fluorogen, is usually considered superior to MF method with a modified
Endo (mEndo) agar and nutrient agar medium (Brenner et al., 1993; Brenner et al.,
1996). These media contain chemicals that produce some characteristic change in the
colonies or medium around the colonies of specific bacteria within the group. Selective
media often contain inhibitors, which prevent the growth of unwanted and allowing the
growth of the desired bacteria, while differential media are used to distinguish between
bio-chemically and morphologically related groups of bacteria. Selective and differential
agars are used for isolation of E. coli from faeces of dairy herds (Wallace & Jones, 1996)
because isolation from faeces requires a medium that must comprise bile salt and crystal
violet in addition to lactose sugar, which differentiates pathogenic Entero-bacteria
(Salmonella and Shigella) that are usually lactose negative and can therefore, be easily
distinguished from non-pathogenic (E. coli) Entero-bacterial species (Orskov, 1986). In
addition to these components, there are many other tests that are required for accurate
identification of E. coli, which complicate the task further.
*Corresponding author: Telephone #: 92-041-2654221-28; Fax: 92-041-2654213
E-mail: [email protected], [email protected]
780 RUBINA ARSHAD AND SHAFQAT FAROOQ

In order to avoid such difficulties, pharmaceutical companies e.g. BBL have

introduced biochemical kits, which can be effectively used without any confusion. In
developing countries however, availability of such kits is beset with several factors. For
example, importing the kits takes very long time and thus hinders their timely
availability. Provision of foreign exchange is another limiting factor. Storage process
with the local supplier (in most of the cases) is not very effective and cost of
identification is also very high. Such limitations hinder active and efficient bacterial
collection particularly in underdeveloped and developing countries like Pakistan, which
are otherwise diversity rich countries. In the present paper, comparative identification and
characterization made with three types of selective and differential agar media
(MacConkey, EBM, and Endo), biochemical test like iMViC (Stanier et al., 1989) and
diagnostic kits (Feng & Hartman, 1982) of E. coli collected from different local habitats
is being presented in order to demonstrate the cost-effectiveness and reliability of
isolation, identification and characterization of bacteria with some other methods in case
the kits are not available.

Materials and Methods

Sources of microorganisms: Bacterial populations to be examined were collected from

various sources such as water, soil, plants and animal faeces at random and over many
months.

Collection of samples and their processing: Water samples were collected from
irrigation channels and processed within 30 minutes after collection. Density of E. coli
was determined by adding 2 ml of freshly collected channel water to 10 ml of sterile
distilled water and analyzed immediately. Soil samples were collected in sterile Petri
plates from different irrigated fields, processed and analyzed immediately after they were
received in the laboratory. One gram of fresh soil was shaken manually in 10 ml of sterile
distilled water for few minutes. Serial dilutions (1:10) were prepared in sterile distilled
water and allowed to settle for some time. Plant material (leaves, petals and stem pieces)
were also collected and examined immediately after washing different organs with sterile
distilled water separately. Faecal samples were obtained from buffalo, cow, goat, horse,
donkey, mule and humans. For each animal, samples were collected from fresh faeces at
different localities around Faisalabad. All faecal samples were placed in separate sterile
Petri plates in a refrigerator and processed within 4 hours. One gram of faecal sample was
suspended in 10 ml of sterile phosphate buffer (pH 7.2). The buffer comprised 1.25 ml of
solution A [3.4 g KH2PO4 + 50 ml reagent grade water (pH 7.2 with 1N NaOH) and
diluted to 100 ml with reagent grade water] and 5.0 ml of solution B [3.8 g MgCl2 + 100
ml reagent grade water diluted to one litre with reagent grade water]. Serial dilutions
were prepared in the same buffer.

Direct isolation: Plant samples collected from various sources were spread across the
surface of nutrient agar plates. These plates were then incubated for 18 hours at 37ºC.
After overnight growth, bacterial colonies were picked and maintained on nutrient agar
slants for biochemical tests.
COST-EFFECTIVE CHARACTERIZATION OF E. COLI 781

Membrane filtration (MF) method: Approximately 5 ml of water, soil and faecal

samples were passed through millipore pre-filtration pad to remove dust particles. The
pre-filtrate was then filtered through 0.45 µm-pore-size cellulose nitrate sterile membrane
filter (MF). The membrane filter funnel and flask (filtration assembly fabricated locally)
were autoclaved before each experiment.

Lactose fermentation test: Culture media containing lactose (a fermentable

carbohydrate), and a dye/pH indicator (capable of detecting changes in pH occurring
during growth), were used for differentiating the non-pathogenic coliforms (lactose
positive) from the potentially pathogenic bacteria (lactose negative).

Selective and differential media: Three types of media were used for isolation and
identification of E. coli, which included moderately selective and differential medium
(MacConkey agar), differential medium (Eosin Methylene Blue or EMB agar), and
selective medium (Endo agar).

Bacterial culture: The filter was placed on a plate containing 10 ml of MacConkey agar
followed by incubation at 44ºC for 24 hours. The presumptive positive colonies of E. coli
were again transferred on to MacConkey agar and incubated for another 18 hours at 37ºC.
Two to three well-grown colonies were picked up randomly and streaked on the surface
of EMB agar to obtain well-isolated single colony. Plates were incubated overnight at
37ºC and each single colony was subsequently isolated, maintained on nutrient agar
slants. All these colonies were further confirmed by inoculating on the Endo agar plates
by streak-plate method. These plates were incubated for 24 hours at 35ºC. Further
confirmation was made through conventional biochemical tests (IMViC) that is Indole
(I), Methyle red (M), Voges-Proskauer (Vi), and Citrate (C) as described by Stanier et al.,
(1989) and were used primarily to distinguish between Escherichia coli and Enterobacter
aerogenes.

Test Kit: Bactident® E. coli rapid identification kit (E.C Merck) was also used along
with conventional identification methods based on series of biochemical tests. The
principle of the test is based on the enzyme profile of the E. coli, particularly on the
detection of the enzyme β -D-glucuronidase and tryptophanase (indole formation). One
pack of test kit contained 50 strips, 50 reaction vessels and reagents required to perform
50 tests. Bacterial strain was suspended in a reaction cuvette containing 200 µl of
deionized water. The Bactident® E. coli test strip was inserted in the reaction vessel and
was incubated at 37ºC for 30-120 min. The presence of β-D-glucuronidase was assessed
under a long wavelength (360 nm) UV lamp. One drop of KOVAC’s reagent was added
to the cuvette for detecting indole formation.
Results

Samples processed through standard membrane filtration technique (Fig. 1a-c) were
tested through lactose fermentation test. The identification and differentiation of E. coli
colonies on MacConkey agar is shown in Fig. 2a. Three types of colonies appeared on
MacConeky agar that are bright pink colonies with red halo (E. coli), yellow colonies
with red border and pinkish red colonies with yellow border, which indicated the
presence of Klebsiella and Enterobacter, respectively in the test sample (Fig. 3).
782 RUBINA ARSHAD AND SHAFQAT FAROOQ

c
Fig. 1. Membrane filtration technique for the isolation of E. coli (a) pre-filtration of sample; (b)
filtration; (c) membrane filter on MacConkey agar.
COST-EFFECTIVE CHARACTERIZATION OF E. COLI 783

Fig. 2. E. coli colonies on (a) MacConkey agar; (b) EMB agar (c) Endo agar.
RUBINA ARSHAD AND SHAFQAT FAROOQ

(*&**) Pink colonies were further confirmed through streaking on EMB agar and IMViC tests
Fig. 3. Differentiation of gram-negative bacilli based on lactose fermentation test.
784
COST-EFFECTIVE CHARACTERIZATION OF E. COLI 785

Table 1. The IMViC reactions for distinguishing E. coli and Enterobacter aerogenes.
IMViC Tests
Strain Voges- Citrate
Indole Methyl red
proskauer
Escherichia coli Dark red ring Magenta red Brown colour No visible
(+) colour (+) (-) growth (-)
Enterobacter aerogenes No red ring Yellow Pink/crimson Visible
(-) colour (-) colour (+) growth (+)
As described by Stanier et al., 1989.

On EMB agar, E. coli appeared as 2-3 mm diameter colonies with fluorescent blue-
black color reflecting greenish metallic sheen when exposed to light (Fig. 2b) and a dark
or black center in transmitted light, while Enterobacter appeared as 4-6 mm diameter
colonies with gray-brown center (or sometimes black) without any metallic sheen in
transmitted light. On Endo agar plates, presumptive E. coli appeared as red colonies with
a permanent metallic sheen (Fig. 2c). Lactose-negative and weakly lactose-positive E.
coli does not show any fuchsin sheen.
The results of IMViC reactions for distinguishing E. coli and Enterobacter
aerogenes are shown in Table 1. A dark red colour in the amyl alcohol surface layer
constituted a positive indole test and confirmed the presence of E. coli. In Methyl Red
(MR) test, a magenta red color indicated the presence of E. coli (positive test) and yellow
color its absence (negative test). The bacterial strains producing red color were referred to
as MR positive (confirmed E. coli). Likewise strains producing a yellowish color were
termed MR negative (non-E. coli) and were identified as Enterobacter aerogenes. The
Voges-Proskauer (VP) reaction was a qualitative test performed to detect the presence of
acetyl methyl carbinol: the end products of glucose fermentation. Since the bacterial
strains that were tested did not produce a pink color in the presence of alkali, therefore,
these were considered as VP negative (E. coli confirmed).
A total of 50 selected strains with diagnostic characteristics of E. coli were re-
confirmed with the test kit by determining the profile of enzyme β-D-glucuronidase and
tryptophanase (indole formation), which is characteristic of E. coli. The results show that
100% of the strains were positive for these enzymes. The presence of β-D-glucuronidase
was indicated by the appearance of light blue fluorescence in UV light of long
wavelength (360 nm) and indole formation was indicated by the appearance of red colour
on addition of KOVAC’s reagent (positive reaction for E. coli).
Table 2 shows the distribution pattern of 474 bacterial strains isolated from 90
samples collected from different sources including water channels, soil from irrigated
fields, different types of plants and feces of animals and human. Among them, 424 E. coli
strains were confirmed through manipulation of different media/method and biochemical
methods described above. Of the 90 samples collected, 41 (45%) samples were from
fecal matter of animals, from where about 159 E. coli strains were isolated (Table 3).
Remaining samples were from soil (13), plants (16) and water (20), which helped in
isolation of an additional 100, 50 and 115 E. coli strains, respectively. Of the 424 E. coli,
215 strains (37% of the total) were collected only from irrigation channels and soil from
irrigated fields (Fig. 4) whereas the recovery rate from water, soil and plants was 27%,
24% and 12%, respectively.
786 RUBINA ARSHAD AND SHAFQAT FAROOQ

Table 2. Bacterial strains isolated from various sources.

Samples examined (n) Strains (n)
Source Total
E. coli Non-E. coli
Water 20 115 - 115
Soil 13 100 - 100
Plants 16 50 50 100
Animal feces 41 159 - 159
Total 90 424 50 474

Table 3. Escherichia coli isolated from faeces of different animals.

Samples examined E. coli isolated
Source
(n) (n)
Buffalo 9 27
Cow 5 25
Goat 7 25
Horse 3 15
Donkey 10 35
Mule 4 15
Human 3 17
Total 41 159

Fig. 4. The percent distribution of E. coli strains isolated from different sources.

Discussion

In the present study, membrane filtration (MF), culture media, biochemical methods
and diagnostic kits were used for identification of E. coli. Lactose-fermenting intense
pink colonies (presumptive E. coli) with red halo: a characteristics of E. coli
(MacConkey. 1905) were observed on MacConkey agar. In addition to E. coli, other
colonies such as Enterobacter, Klebsiella were also present that were eliminated in EBM
agar as it provides favorable conditions for growth of E. coli and improves its
proliferation in particular compared to other lactose-positive bacteria. Since, E. coli
produced large amounts of acid from lactose and gave the colonies a very dark, metallic
sheen, which Klebsiella and Enterobacter were lacking may be due to accumulation of
less acid, and gave the colonies a purple center and pink periphery which helped
COST-EFFECTIVE CHARACTERIZATION OF E. COLI 787

discriminating E. coli from the accompanying Klebsiella and Enterobacter. The growth
of other Gram-positive microorganisms was inhibited by the dyes (Eosin and Methylene
Blue) contained in EMB agar (Levine, 1921). Despite this, confusion can be created in
identifying E. coli and Enterobacter, both of which appeared with dark/black and
gray/brown (or gray/black), respectively upon viewing under transmitted light.
Presumptive positive E. coli were therefore, confirmed again both culturally through the
use of Endo agar and bio-chemically through IMViC tests.
On Endo agar plates (Endo, 1904), presumptive E. coli appeared as red colonies with
a permanent metallic sheen. This is because Endo agar contains Sodium sulfite, which
after reacting with aldehyde and acid (liberated by E. coli through metabolizing lactose)
produced fuchsin-sulfite, which turns the colonies red. In the case of E. coli, this reaction
is so intense that the fuchsin crystallizes out giving the colonies a permanent greenish
metallic sheen (fuchsin sheen). Lactose-negative and weakly lactose-positive bacteria do
not show any fuchsin sheen. The Indole test, which confirmed the presence of E. coli is
based on the enzyme tryptophanase (indole formation), which is a characteristic of E. coli
only, and is not found in Enterobacter aerogenes (Stanier et al., 1989) and thus served as
confirmatory test for E. coli. In Methyl-red test, the red color was an indication of
substantial acid production: a characteristic of mixed acid type fermentation in E. coli.
The purpose of using Methyl red was to use it as pH indicator to determine the pH of
dextrose broth culture after 2-4 days’ incubation. This indicator is yellow at pH 4.5 or
higher (negative test) and red at lower pH values. Test is said to be positive when the
accumulation of acidity is sufficient to turn the indicator red: a characteristic of only the
E. coli.
In Voges-proskauer test, in the presence of alkali, the acetyl methyl carbinol is
oxidized to di-acetyl, which in turn reacts with some constituents of the peptone to give it
a pink color. When acetyl methyl carbinol is produced, the bacterial strain is said to be
VP positive. Since E. coli lacks this ability hence it appeared negative.
“Bactident® E. coli rapid identification kit was used to compare the validity,
reliability and cost-effectiveness of conventional screening procedures based on different
agars and series of biochemical tests and to reconfirm the identified E. coli. Test kit that
takes only 2-3 minutes to set up reaction and 30-120 minutes to produce results detected
the presence of β-D-glucuronidase and tryptophanase (indole formation) in all the 50
strains selected for this test thus confirming them pure E. coli cultures. This comparative
study showed that the selective and differential agars, used in the present study for the
isolation and identification of E. coli, are economical and reliable.
Bactident® E. coli kit serves as a screening test for E. coli therefore, such kits are
usually desirable in the clinical microbiological laboratory for rapid diagnosis. In contrast
for screening of a large number of samples for culture collection, this kit is not
economical because one kit, containing 50 strips and 50 reaction vessels, costs about US$
135 which is enough for only a maximum of 50 tests. With differential and selective
media, hundreds of isolates were screened for the identification of E. coli. Thus the use of
kits for this purpose was not considered cost-effective. Compared to short shelf life of the
kits, media used in the present study were technically simple and relatively inexpensive
and were also devoid of problems related with shelf-life. The use of agars also reduced
the cost; e.g. 500 g of EMB, MacConkey and Endo agar (BBL) cost about US$ 80 each
or less than that and can be used to perform hundreds of tests. Moreover compared to
these expensive kits, the agars were within the range of most of the laboratories in
788 RUBINA ARSHAD AND SHAFQAT FAROOQ

Pakistan and other countries like Pakistan, where such items suffer inadequate supply
and/or storage facilities. In addition to these kits, Immuno-Magnetic Separation (IMS),
visual immuno-precipitate assay (VIP) and visual immunoassay (VIA) are also being
used in food testing, clinical, veterinary and environmental sciences (Chapman, 2000;
Chapman et al., 1994; Warburton, 1996; Warburton, 1997) and for isolation of
pathogenic and non pathogenic bacteria from foods, vegetables, and dairy products
(Warburton, 2001). Although these assays and kits are more sensitive, easy to use and
rapid in the detection of E. coli, nevertheless, they are expensive and may not prove
economical when used for testing large number of samples. Compared to this, membrane
filtration method used in the present study is easier than isolating and identifying E. coli
by diagnostic kits (Fig. 5) and is also cheaper. For example, ONPG Discs (BBL) for
detecting lactose fermenters is commercially available and a pack of six discs costs US$
278. Identification of E. coli by performing indole and glucuronidase test through
“ColiScreen” kits (Hardy Diagnostics, online data) is a rapid and reliable method of E.
coli identification. Nevertheless, one test kit costs about US$ 19, which is sufficient for
only 20 tests and should be refrigerated on arrival. Sim Plates used for total coliforms and
E. coli count are also very expensive and their shelf life is only six months (BioControl,
online data).
The present study has been conducted stepwise and highlighted the importance of
cost-effectiveness in identification of large number of samples. We believe that since this
area is continuously developing and a universally accepted method is yet to be developed,
hence media and methods used in the present study will encourage resource deficient
laboratories especially in the third world countries, which are reluctant to undertake
extensive microbiological tasks only due to inadequate and untimely supply of expensive
diagnostic kits.

Fig. 5. Cost-effectiveness of methods used for identification of E. coli.

References

Brenner, K.P., C.C. Rankin, Y.R. Roybal, G.N. Stelma Jr., P.V. Scarpino and A.P. Dufour. 1993.
New medium for the simultaneous detection of total coliforms and Escherichia coli. Appl.
Environ. Microbiol., 59: 3534-3544.
Brenner, K.P., C.C. Rankin, M. Sivaganesan and P.V. Scarpino. 1996. Comparison of the
recoveries of Escherichia coli and total coliforms from drinking water by the MI agar method
and the U.S. Environmental Protection Agency-approved membrane filter method. Appl.
Environ. Microbiol., 62: 203-208.
Chapman, P.A. 2000. Methods available for the detection of Escherichia coli O157 in clinical, food
and environmental samples. W. J. Microbiol. Biotechnol., 16: 733-740.
Chapman, P.A., D.J. Wright and C.A. Siddons. 1994. A comparison of immunomagnetic separation
and direct culture for the isolation of verocytotoxin-producing Escherichia coli O157 from
bovine feces. J. Med. Microbiol., 40: 424-427.
 COST-EFFECTIVE CHARACTERIZATION OF E. COLI 789

Cundell, A.M. 1981. Rapid counting method for coliform bacteria. Adv. Appl. Microbiol., 27: 169-
183.
Endo, S. 1904. Ober ein Verfahren zum Nachweis von Typhusbacillen. Centralbl. Bakt. I Orig., 35:
109-110.
Feng, P.C.S. and P.A. Hartman. 1982. Fluorogenic assay for immediate confirmation of E. coli.
Appl. Environ. Microbiol., 43: 1320-1329.
Levine, M. 1921. Bacteria fermenting lactose and the significance in water analysis. Bull 62. Iowa
State College Engr. Exp. Station.
MacConkey, A. 1905. Lactose-fermenting bacteria in feces. J. Hyg., 8: 333-379.
Orskov, F. 1986. Escherichia coli. In: The Prokaryotes-A handbook on habitats, isolation, and
identification of bacteria. Vol. II. (Ed.): M.P. Starr, H. Stolp, H.G. Truper, A. Balows, H.G.
Schlegel. Springer-Verlag, Berlin. pp. 1128-1134.
Stanier, R.Y., J.L. Ingraham, M.L. Wheelis and P.R. Painter. 1989. The enteric group and related
eubacteria. In: General Microbiology. Macmillan Education Ltd. pp. 439-452.
Wallace, J.S. and K. Jones. 1996. The use of selective and differential agars in the isolation of
Escherichia coli O157 from dairy herds. J. Appl. Bacteriol., 81: 663-668.
Warburton, D. 1996. Detection of enterohemorrhagic E. coli (EHEC) in food products and food
ingredients by the VIP for EHEC method. In: The Compendium of Analytical Methods. Health
Protection Branch, Polyscience Publications, P.O. Box 1606, Station St-Martin, Laval,
Quebec, Canada H7V 3P9 (http://www.hc-sc.gc.ca/food-aliment).
Warburton, D. 1997. Detection of E. coli O157 by the TERCA® E. coli visual immunoassay (VIA)
method. In: The Compendium of Analytical Methods Health Protection Branch, Polyscience
Publications, P.O. Box 1606, Station St-Martin, Laval, Quebec, Canada H7V 3P9
(http://www.hc-sc.gc.ca/food-aliment).
Warburton, D. 2001. Isolation of E. coli O157 in foods. In: The Compendium of Analytical Methods
Health Products and Food Branch, Polyscience Publications, P.O. Box 1606, Station St-
Martin, Laval, Quebec, Canada H7V 3P9 (http://www.hc-sc.gc.ca/food-aliment).

(Received for publication 29 March 2004)

View publication stats