Journal of Microbiological Methods 96 (2014) 6–11

Contents lists available at ScienceDirect

Journal of Microbiological Methods

journal homepage: www.elsevier.com/locate/jmicmeth

Evaluation of eight agar media for the isolation of shiga toxin—Producing

Escherichia coli
Alexander Gill a,⁎, George Huszczynski b, Martine Gauthier b, Burton Blais b
a
Bureau of Microbial Hazards, Health Canada, Frederick Banting Building, Tunney's Pasture, Ottawa, Ontario K1A 0K9, Canada
b
Ontario Laboratory Network, Canadian Food Inspection Agency, 960 Carling Avenue, Bldg. 22, C.E.F., Ottawa, Ontario K1A 0C6, Canada

a r t i c l e i n f o a b s t r a c t

Article history: The growth characteristics of 96 shiga toxin-producing Escherichia coli (STEC) strains representing 36 dif-
Received 11 September 2013 ferent O-types (including priority O types O26, O45, O103, O111, O121, O145 and O157) on commercial
Received in revised form 29 October 2013 and in-house agar media were studied. The ability of the strains to grow on agar media with varying selective
Accepted 29 October 2013
supplement formulations was evaluated using MacConkey Agar (MAC); Rainbow® Agar O157 (RBA); Rainbow®
Available online 5 November 2013
Agar O157 with manufacturer-recommended selective supplements (RBA-NT); Rainbow® Agar O157 with
Keywords:
USDA-recommended selective supplements (RBA-USDA); CHROMagar STEC™ (CH STEC); Tryptone Bile agar
Shiga toxin/verotoxin containing cefixime and tellurite (TBA-CT); Tryptone Bile agar containing cefixime, tellurite, eosin and methylene
Escherichia coli blue (TBA-EM); and VTEC agar. All of the strains were able to grow on MAC, RBA and VTEC agar, whereas a number
Growth of strains (including some non-O157 priority O types) were unable to grow on the highly selective media CH STEC,
Agar media RBA-NT, RBA-USDA, TBA-EM and TBA-CT. Only RBA-NT and CH STEC exhibited significant inhibition of background
Selective supplements flora from ground beef enrichment. Significant inhibition of background flora from beef trim enrichment was
observed with RBA-NT, RBA-USDA, CH STEC, TBA-EM and VTEC agar. With exception of E. coli O157, several dif-
ferent colony morphologies were observed on the differential plating media among strains of the same O type,
indicating that this colony morphology is not a reliable means of identifying target STEC. These results suggest
that an approach to maximize the recovery of target STEC from beef enrichment cultures is dual plating on lesser
(RBA, MAC, VTEC agar) and more highly (RBA-NT, CH STEC) selective agars.
Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.

1. Introduction
Shiga toxin-producing Escherichia coli (STEC) (also known as
verotoxin producing E. coli or enterohemorrhagic E. coli) are the
enteric E. coli pathotypes of the greatest public health significance
in the industrialized world due to their low infectious dose, potentially
severe patient outcomes and limited treatment options (Melton-Celsa
et al., 2012).
Development of methods for the detection and isolation of STEC has
focused on E. coli O157:H7 and its nonmotile variant (E. coli O157) as
this serotype has been responsible for a number of major foodborne
outbreaks and is the cause of the majority of reported cases in some
regions, including Canada and the USA (CDC, 2012; PHAC, 2010).
Other serotypes of STEC (non-O157 STEC) are estimated to account for
50–60% of cases of STEC illness in Canada and the USA (CDC, 2012;
Thompson et al., 2005; Chui et al., 2011). The serotypes of STEC that
predominate as a cause of human illness vary between geographic
regions (EFSA, 2007; Johnson et al., 2006).
⁎ Corresponding author. Tel.: +1 613 952 8894.
E-mail address: alex.gill@hc-sc.gc.ca (A. Gill).
The diversity of STEC serotypes causing disease is high, at least 82 dif-
ferent O-types of STEC have been reported as clinical isolates from three
or more outbreaks or unlinked sporadic cases (Bettelheim, 2007). This
situation is probably a consequence of the mobility of the verotoxin
genes, which are encoded by a lysogenic phage (Kaper et al., 2004).
Thus, there is a high probability of novel STEC pathogens emerging,
such as the verotoxin-positive enteroaggregative E. coli O104:H4 respon-
sible for the 2011 outbreak in Germany (Beutin and Martin, 2012).
Isolation of STEC from enrichment broth cultures of foods or other
sample types requires an agar medium for the isolated growth of indi-
vidual colonies. Such a medium should support the growth of the target
organism and ideally have selective and differential characteristics to
support the rapid identification of pathogen colonies in the presence
of other microbial flora. There is a variety of commercially available
media for E. coli O157 which meets these requirements (Heuvelink,
2012). These media are commonly based on differential characteristics
typical to E. coli O157 but unusual in other E. coli, notably the absence
of sorbitol fermentation and β-D-glucuronidase activity. Selectivity is
achieved by the relatively high resistance of E. coli O157 to some antimi-
crobials, such as novobiocin and tellurite, compared to other E. coli.
These differential and selective characteristics are not shared by other
STEC (Gill et al., 2012) and to date the only characteristic known to

0167-7012/$ – see front matter. Crown Copyright © 2013 Published by Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.mimet.2013.10.022
 A. Gill et al. / Journal of Microbiological Methods 96 (2014) 6–11 7

Table 1
Characteristics of STEC strains and their percentage recovery on selective agar media compared to Brain Heart Infusion agar.a

Serotypea Strain ID Isolation stx1 stx2 eae hlyA MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC

O26:NM 11-6009 Human + − + − 106% 124% 133% 51% 104% 102% 114% 120%
O26:H11 05-7321 Human + − + + 119% 94% 121% 188% 106% 91% 104% 121%
O26:H11 05-6544 Human + − + + 123% 157% 20% 117% b0.1%% 87% 97% 150%
O26:H11 03-2816 Human + − + − 133% 123% 111% 52% 119% 101% 103% 188%
O26:H11 01-5870 Human + − + + 118% 95% 103% 47% 62% 90% 72% 108%
O26:H11 02-6737 Human + + + + 33% 67% 52% 95% 52% 57% 95% 81%
O26:H11 99-4610 Human + − + + 133% 100% 108% 108% 62% 67% 100% 100%
O26:H21 11-5130 Human + − + − 85% 72% 83% 71% 70% 83% 98% 102%
O26:H21 11-5593 Human + + − + 95% 158% NG NG NG 71% 58% 105%
O26:H21 11-5805 Human + − + + 113% 103% 105% 175% 103% 89% 89% 92%
O45:H2 04-2445 Human + − + + 135% 93% 71% 55% 68% 132% 103% 132%
O45:H2 05-6545 Human + − + + 81% 73% 49% 40% 7% 76% 90% 71%
O45:H2 85-X-40c R3 Cattle + − + + 102% 131% 63% 35% 19% 95% 126% 84%
O45:H2 3267-95 Human + − + + 116% 113% 60% 22% 18% 68% 84% 108%
O45:H2 3285-96 Human + − + + 72% 126% 126% 27% 65% 81% 102% 63%
O45:H2 89-39 Human + − + + 94% 111% 52% 33% 19% 20% 98% 127%
O103:H2 99-2076 Human + − + + 53% 41% NG NG NG NG NG 94%
O103:H2 06-0434 Human + − + + 86% 110% 100% 96% 45% 103% 83% 117%
O103:H2 04-2446 Human + − + + 79% 74% 77% 130% 32% 81% 121% 145%
O103:H2 95-266 Human + − + + 167% 189% NG NG NG 189% 87% 122%
O103:H2 09-5066 Human + − + + 81% 104% 13% 23% b0.1% 71% 85% 81%
O103:H2 11-5806 Human + − + + 110% 95% 46% 79% 46% 115% 115% 103%
O103:H11 04-3973 Human + − + + 89% 85% 75% 100% 69% 109% 106% 115%
O103:H21 11-4211 Human + − + + 81% 96% 56% 100% 30% 107% 93% 93%
O103:H21 11-5595 Human + − + + 126% 92% 97% 100% 92% 113% 139% 105%
O103:H25 03-2444 Human + − + + 67% 83% 3% 25% b0.1% 37% 45% 92%
O111:NM 05-4161 Human + + + − 126% 174% 53% 63% 42% 111% 153% 137%
O111:NM 98-8338 Human + − + + 109% 98% NG NG NG b0.1% b0.1% 89%
O111:NM CFS3 Human + + + + 96% 82% 7% 96% 20% 51% 65% 76%
O111:NM 00-4748 Human − + + − 2% 8% NG b0.1% NG b0.1% b0.1% 33%
O111:NM 00-4440 Human + − + + 34% 68% 28% 60% 6% 38% 49% 36%
O111:NM 11-5592 Human + − + + 82% 158% 85% 103% 35% 127% 61% 88%
O111:NM 03-3991 Human + − + + 100% 71% 32% 29% 26% 65% 65% 68%
O111:H8 3331-00 Unknown + + + + 100% 100% 30% 39% 9% 52% 74% 113%
O111:H8 3413-07 Unknown + + + + 103% 132% NG 67% NG b0.1% 132% 156%
O111:H11 OLC-455 Unknown + − + + 87% 93% 50% 60% 27% 53% 83% 73%
O121:NM 03-4064 Human − + + + 179% 53% b0.1% 94% 42% 128% 137% 133%
O121:H1 11-5594 Human − + + + 214% 145% 73% 111% 73% 159% 168% 145%
O121:H1 11-5597 Human − + + + 75% 100% 49% 67% 31% 82% 88% 88%
O121:H10 96-0120 Unknown − + − − 90% 115% NG b0.1% NG b0.1% 100% 85%
O121:H19 11-2925 Human − + + + 89% 88% 85% 115% 84% 68% 78% 91%
O121:H19 11-3925 Human − + + + 111% 100% 32% 6% 11% 61% 70% 74%
O121:H19 11-4440 Human − + + − 113% 105% 118% 71% 125% 95% 105% 20%
O121:H19 03-2832 Human − + + + 108% 119% 33% 77% 8% 64% 106% 81%
O121:H19 00-5288 Human − + + + 123% 74% 51% 18% 4% 118% 111% 102%
O121:H19 03-2642 Human − + + + 118% 97% 21% 80% 5% 33% 92% 105%
O145:NM 04-1449 Human + − + + 116% 100% 68% 67% 48% 55% 52% 106%
O145:NM 04-7099 Human + − + + 103% 125% 67% 49% 47% 122% 94% 106%
O145:NM 03-6430 Human + − + + 136% 130% 70% 52% 70% 115% 130% 121%
O145:NM 03-4699 Human + − + + 111% 100% 29% 43% 20% 95% 78% 93%
O145:NM 2454-01 Human + − + + 98% 115% 63% 82% 13% 67% 71% 81%
O145:NM VT113-5 Human − + + + 36% 62% 53% 32% 6% 53% 54% 7%
O145:H2 A9619.C2 Human + − − − 33% 58% NG NG NG NG NG 38%
O145:H2 75-83 Human − + + + 103% 69% 53% 42% NG 56% 83% 94%
O145:H25 2769 Human − + + + 117% 117% 60% 39% 33% 81% 90% 88%
O157:H7 ATCC 35150 Human + + + + 83% 72% 29% 34% 14% 108% 100% 75%
O157:H7 1011-84 Human + + + + 100% 103% 18% 51% 9% 36% 45% 115%
O157:H7 HCO 59 Human + + + + 104% 85% 30% 35% 11% 67% 81% 107%
O157:H7 11-1024 Human − + + + 143% 152% 19% 78% 5% 81% 129% 86%
O157:H7 11-1865 Human − + + + 81% 88% 38% 75% 12% 88% 62% 112%
O157:H7 EDL933 Human + + + + 95% 18% 108% 29% 8% 60% 80% 90%
O157:H7 Sakai Human + + + + 112% 160% 36% 21% 17% 112% 93% 105%
O157:NM ER63-94 Human + + + + 87% 67% 29% 78% 15% 98% 79% 71%
O157:NM E32511 Human − + + + 122% 83% 65% 53% 9% 78% 96% 122%
O157:NM 87-1215 Human + + + − 114% 129% 114% 86% 57% 171% 86% 57%
O1:H20 91-0812 Unknown + − − + 119% 113% NG 19% b0.1% 59% 120% 107%
O5:NM 03-2682 Human + − + + 97% 109% 91% 92% 51% 74% 87% 98%
O6:H34 03-5166 Human − + − − 123% 94% NG NG NG 60% 125% 114%
O7:H4 92-0249 Unknown − + − − 170% 128% NG b0.1% NG 87% 113% 143%
O8:H19 09-1764 Unknown + + − + 109% 123% b0.1% NG b0.1% 85% 88% 106%
O46:H38 97-0757 Human + + − + 83% 81% 32% 47% 23% 73% 87% 108%
O55:H7 05-0376 Human + − + − 127% 107% NG NG NG 83% 82% 110%
O69:H11 11-5596 Human + − + + 143% 129% 114% 83% 129% 57% 86% 71%
O76:H19 09-0523 Unknown + + − + 100% 120% NG NG NG 117% 92% 80%

(continued on next page)

8 A. Gill et al. / Journal of Microbiological Methods 96 (2014) 6–11

Table 1 (continued)
Serotypea Strain ID Isolation stx1 stx2 eae hlyA MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC

O91:H21 85-489 Human − + − + 113% 94% b0.1% NG NG 42% 83% 110%

O98:H29 09-5073 Human + − + + 107% 97% 79% 81% 90% 134% 72% 100%
O100:NM 96-0429 Unknown − + − − 124% 119% 105% 117% 80% 104% 109% 114%
O104:H4 11-3088 Human − + − − 92% 127% 42% 50% 35% 115% 108% 121%
O104:H21 3024-94 Unknown − + − + 110% 89% NG NG NG 55% 93% 85%
O108:H11 11-3580 Human − + + + 110% 78% 24% 24% 29% 60% 63% 84%
O110:H28 09-409 Unknown − + − + 75% 100% NG b0.1% NG 65% 80% 55%
O113:H21 04-1450 Human − + − + 137% 130% NG NG NG 33% 67% 133%
O115:H18 03-3645 Human + + − + 93% 100% NG 66% NG 69% 66% 66%
O117:H7 02-0035 Human + − − − 111% 115% NG NG NG 63% 105% 109%
O123:H2 11-4968 Human + − + + 68% 97% 40% 85% 51% 62% 78% 81%
O128:H2 1759 Human + − − + 86% 100% NG NG NG 57% 95% 93%
O128:H10 2611-99 Human + − − − 115% 110% NG NG NG b0.1% 101% 118%
O130:H11 09-1765 Unknown − + − + 125% 99% NG b0.1% NG 87% 86% 111%
O146:H21 02-1628 Human + − − + 133% 110% NG NG NG 116% 96% 133%
O153:H25 09-1775 Unknown + + − + 100% 133% NG b0.1% NG 67% 142% 87%
O156:NM 92-0389 Unknown − + − − 113% 115% NG NG NG NG 75% 112%
O165:H25 00-4540 Human − + + + 118% 136% NG NG NG 91% 118% 123%
O177:NM 03-3974 Human − + + + 78% 93% 24% 35% 34% 68% 80% 93%
O179:H8 09-415 Unknown − + − + 108% 138% NG b0.1% NG 70% 104% 114%
O181:H49 09-412 Unknown − + − + 119% 138% NG NG NG 50% 117% 152%
Orough:H21 11-6008 Human + + − + 103% 81% NG NG NG 69% 66% 147%

MAC: MacConkey Agar.

RBA: Rainbow® Agar O157.
RBA-NT: Rainbow® Agar O157 with 10 mg/l novobiocin and 0.8 mg/l potassium tellurite.
RBA-USDA: Rainbow® Agar O157 with 5 mg/l novobiocin, 0.05 mg/l cefixime trihydrate and 0.15 mg/l potassium tellurite.
CH STEC: CHROMagar STEC™.
TBA-CT: Tryptone Bile agar with 0.05 mg/l cefixime and 0.5 mg/l of potassium tellurite.
TBA-EM: TBA-CT with 0.4 g/l eosin Y, 0.06 g/l methylene blue and 1 g/l lactose.
VTEC: VTEC agar (Gill et al., 2012).
a
+: positive, −: negative, NG: no growth on streaked or enumeration plates, b0.1% growth on streaked plate but recovery on enumerated plates below 0.1% of BHI.

be common to all STEC and absent from other E. coli is verotoxin
production.
Few outbreaks of illness due to non-O157 STEC contaminated beef
have been reported, compared to E. coli O157, but the development of
methods for non-O157 STEC in beef is a priority due to the high rates
of carriage of non-O157 STEC by cattle and their presence in raw beef
(Gill and Gill, 2010). In an attempt to mitigate public health impacts of
non-O157 STEC in beef, the United States Department of Agriculture's
Food Safety and Inspection Service (USDA-FSIS) has implemented test-
ing of manufacturing trim and raw ground beef for the presence of the
so-called priority STEC serogroups O26, O45, O103, O111, O121, and
O145, in addition to E. coli O157:H7 (USDA, 2012).
While several agar media have been developed or proposed for the
isolation of STEC from foods, few studies have been published on the
comparative performance of these media and those that have tended
to focus on a relatively small number of STEC strains of a limited range
of serotypes. For this reason we have conducted a study to evaluate
the ability of eight different agar media to support the growth of 96
STEC strains, representing 36 different O-types. The media were also
evaluated to assess their ability to suppress the growth of background
microflora from ground beef and beef trim.
2. Material and methods
2.1. Bacterial cultures
The 96 strains of STEC used in this study and their virulence charac-
teristics are listed in Table 1. Unless otherwise noted the strains used
were provided from the collection of the Public Health Agency of
Canada, National Microbiology Laboratory, Winnipeg, Manitoba. Strains
85-X-40c R3, 3267–95, 3285–96, 89–39, 2454–01, 2769, 2454–01,
EDL933, ER63-94, E32511, Sakai, 1759, 2611–99 were kindly supplied
by Dr. Roger Johnson of the Public Health Agency of Canada, Laboratory
of Food-Borne Zoonoses, Guelph, Ontario. Strains CFS3, 3331–00,
3413–07 were kindly supplied by Dr. P. Micheal Doyle of the University
of Georgia, Griffin, Georgia. Strain 11–3088 was kindly supplied by
Dr. Vanessa Gray Allen of the Public Health Ontario Laboratories,
Toronto, Ontario. The remaining strains were from the culture collection
of the Health Canada, Bureau of Microbial Hazards, Ottawa, Ontario. The
presence of STEC virulence genes stx1, stx2, eae, and EHEC hlyA was
determined by the PCR protocol of Paton and Paton (2003). The cultures
were stored as frozen glycerol stocks. For experimental use, the cultures
were streaked onto brain heart infusion (BHI) agar (Difco, BD, Sparks,
MD) and incubated at 37 °C for 24 h.
2.2. Recovery of STEC strains on selective agar
Single colonies of the strains were individually inoculated into 10 ml
of modified tryptic soy broth (mTSB: tryptic soy broth [30 g/l; Difco],
bile salts no. 3 [1.5 g/l; Difco, BD], K2HPO4 [1.5 g/l; Sigma, St. Louis,
MO], pH 7.4), supplemented with 10 mg/l vancomycin (Sigma) and
3 mg/l cefsulodin (Sigma) after 4 h (mTSB-VC broth) as described in
Gill et al. (2012), and incubated for 18–24 h. at 42 °C.
The following selective agar media were evaluated for their ability
to support the growth of STEC strains. MacConkey Agar (MAC; Difco);
Rainbow® Agar O157 (RBA; Biolog Inc., Hayward, California); Rain-
bow® Agar O157 supplemented, with 10 mg/l novobiocin (Sigma)
and 0.8 mg/l potassium tellurite (Sigma) (RBA-NT) as recommended
by the manufacturer for samples with high background flora; Rainbow®
Agar O157 supplemented with 5 mg/l novobiocin, 0.05 mg/l cefixime
trihydrate (Oxoid) and 0.15 mg/l potassium tellurite (RBA-USDA) as
recommended for the USDA STEC method MLG 5B.03 (USDA,
2012); CHROMagar STEC™ (CHROMagar Microbiology, Paris, France)
(CH STEC); Tryptone Bile agar (Oxoid, Basingstoke, Hampshire, UK)
containing 0.05 mg/l cefixime and 0.5 mg/l of potassium tellurite
(TBA-CT); Tryptone Bile agar containing 0.05 mg/l cefixime, 0.5 mg/l
of potassium tellurite, 0.4 g/l eosin Y (Sigma), 0.06 g/l methylene blue
(Fisher Scientific, Ottawa, Ontario) and 1 g/l lactose (TBA-EM); and
VTEC agar (Gill et al., 2012), 17.5 g/l vegetable special infusion powder
(Sigma), 2 g/l meat extract (Sigma), 10 g/l proteose peptone (Oxoid),
10 g/l D-sorbitol (Sigma), 5 g/l sodium chloride (Sigma), 2.5 g/l sodium
phosphate dibasic (Sigma), 30 mg/l neutral red (Sigma), 1 mg/l crystal
violet (Sigma), 1 g/l bile salts #3 and 15 g/l agar, supplemented with
10 mg/l vancomycin and 3 mg/l cefsulodin.
 A. Gill et al. / Journal of Microbiological Methods 96 (2014) 6–11 9

Table 2
Proportion of STEC strains whose growth is inhibited on selective agar media tested.a

O-type No. strains MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC

O26 10 0 0 1 1 1 0 0 0
O45 6 0 0 0 0 0 0 0 0
O103 10 0 0 2 2 2 1 1 0
O111 10 0 0 3 1 1 0 0 0
O121 10 0 0 1 0 1 0 0 0
O145 9 0 0 1 1 2 1 1 0
O157 10 0 0 0 0 0 0 0 0
Others 31 0 0 20 15 20 1 0 0
Total 96 0 0 28 (29.2%) 20 (20.8%) 27 (28.1%) 3 (3.1%) 2 (2.1%) 0
a
MAC: MacConkey Agar; RBA: Rainbow® Agar O157; RBA-NT: Rainbow® Agar O157 with 10 mg/l novobiocin and 0.8 mg/l potassium tellurite; RBA-USDA: Rainbow® Agar O157 with
5 mg/l novobiocin, 0.05 mg/l cefixime trihydrate and 0.15 mg/l potassium tellurite; CH STEC: CHROMagar STEC™; TBA-CT: Tryptone Bile agar with 0.05 mg/l cefixime and 0.5 mg/l of
potassium tellurite; TBA-EM: TBA-CT with 0.4 g/l eosin Y, 0.06 g/l methylene blue and 1 g/l lactose; VTEC: VTEC agar (Gill et al., 2012).

Cultures grown in mTSB-VC were streaked onto the eight selective
agar media, which were then incubated for 24 h at 37 °C. The plates
were examined for growth and the colony morphology recorded. To
determine the relative enumerative recovery on the selective agars,
the cultures were serially diluted in phosphate buffered saline (PBS)
and the dilutions manually spread plated onto the eight selective agar
media and BHI agar. The plates were incubated for 24 h at 37 °C and
the colonies counted. The relative efficiency of recovery was calculated
as the percentage of colonies on the selective agar compared to BHI.
Replicate experiments were not performed systematically as little
variation is expected with a descriptive study of this type. The primary
potential sources of error were identified as a failure to inoculate
media or pipetting errors leading to inaccurate reporting of growth inhi-
bition. These errors could be detected by observation of turbidity in the
culture test tubes and comparison of the culture dilutions plated for
enumerative recovery. A small number of errors were detected and
the experiments with those strains repeated and reported.
2.3. Growth of beef microflora on selective agar
Experiments were conducted to evaluate the ability of the selective
agar to inhibit the growth of background microflora present in enrich-
ment cultures from ground beef and beef trim that were not inoculated
with STEC. The ground beef had a 15% fat content and was purchased
from an Ottawa area supermarket, Trim samples were sourced from
an Ontario beef processing plant. Samples of meat 25 g were suspended
in 225 ml of mTSB-VC and incubated for 18–24 h at 42 °C. The post-
incubation enrichment broth was then serially diluted at a 1:10 ratio
in PBS (1 ml into 9 ml) from 10−1 to 10−7 and all dilutions manually
spread plated on the eight selective agar and BHI agar. The agar media
were incubated as described above and the total colonies on each
plate enumerated. This experiment was performed with meat from
three separate lots of ground beef and beef trim.
CH STEC included O26 (1/10), O103 (2/10), O111 (1/10), O121 (1/10)
and O145 (2/9). Strains inhibited on RBA-NT included O26 (1/10),
O103 (2/10), O111 (3/10), O121 (1/10) and O145 (1/9).
For those STEC strains which were able to grow on the selective
media, the average number of colonies counted compared to BHI agar
was 100% for MAC, RBA and VTEC agar (Table 3). The average number
of colonies counted on TBA-EM was 91% and for TBA-CT 78% compared
to BHI agar. The average number of STEC colonies counted on RBA-NT
was 58% and RBA-USDA 60% compared to BHI agar, down from 100%
for RBA without additional antimicrobials. CH STEC was the medium
which was most inhibitory to STEC growth, with an average of 40% of
colonies compared to BHI agar and O26 was the only serogroup for
which the colony count was greater than 50% of the BHI agar count.
The appearance of STEC colonies on the different selective media is
described in Table 4. The colour of the colonies of all strains was highly
uniform on MAC, TBA-CT and TBA-EM. Colonies of all strains on TBA-CT
were white, and almost all were pink on MAC and TBA-EM. Colonies of
strains of the same serogroup were not a consistent colour on RBA, RBA-
NT and RBA-USDA, With the exception of the O157 serogroup, which
were blue-grey. Two or more colours of colonies were observed for
strains all serogroups on CH STEC and for all serogroups on STEC agar
except O26, O103 and O145, which were uniformly pink.
The number of colonies recovered from enrichment broths of
ground beef and beef trim in mTSB-VC on the selective agar media
was variable among the three samples (Table 5). This probably reflects
variation in the composition of the initial flora of the meat samples
prior to enrichment. For ground beef enrichment the number of colo-
nies recovered was only significantly less (Student's t-test p b 0.05)
than BHI agar on RBA-NT and CH STEC. Significantly lower numbers of
colonies (p b 0.05) were recovered from beef trim enrichment com-
pared to BHI agar on RBA-NT, RBA-USDA and CH STEC.

3. Results Table 3
Percentage recovery of STEC strains on selective agar media compared to Brain Heart Infu-
sion agar, excluding inhibited strains.a
The percentage recovery of each STEC strain on the eight selective
agar media is presented in Table 1. The ability of the agar media to sup- O-type MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC
port the growth of strains of the 7 STEC O-types designated a priority by O26 106% 109% 93% 100% 75% 85% 97% 118%
the USDA (O26, O45, O103, O111, O121, O145, O157) is summarized in O45 93% 108% 70% 35% 33% 79% 101% 98%
Table 2. Three of the media tested MAC, RBA and VTEC agar were able to O103 94% 97% 58% 82% 39% 103% 97% 107%
O111 84% 98% 41% 57% 24% 50% 68% 87%
support the growth of all 96 STEC strains tested. TBA-EM did not sup-
O121 122% 100% 51% 64% 43% 81% 106% 92%
port the growth of two strains (O103:H2 99–2076 and O145:H2 O145 95% 97% 58% 51% 34% 81% 82% 82%
A9619.C2). TBA-CT did not support the growth of the same O103 and O157 104% 96% 49% 54% 16% 90% 85% 94%
O145 strains as TBA-EM, and in addition no growth was observed Others 110% 110% 55% 44% 47% 72% 93% 106%
Total 104% 103% 58% 60% 40% 78% 91% 100%
for O156:NM 92–0389. RBA-USDA did not support the growth of 20
a
(20.8%) strains tested, including strains of the serogroups O26 (1/10), MAC: MacConkey Agar; RBA: Rainbow® Agar O157; RBA-NT: Rainbow® Agar O157
O103 (2/10), O111 (1/10) and O145 (1/9). RBA-NT and CH STEC sup- with 10 mg/l novobiocin and 0.8 mg/l potassium tellurite; RBA-USDA: Rainbow® Agar
O157 with 5 mg/l novobiocin, 0.05 mg/l cefixime trihydrate and 0.15 mg/l potassium
ported the growth of the least number of STEC strains, with 28 tellurite; CH STEC: CHROMagar STEC™; TBA-CT: Tryptone Bile agar with 0.05 mg/l
(29.2%) strains on RBA-NT and 27 (28.1%) strains on CH STEC failing cefixime and 0.5 mg/l of potassium tellurite; TBA-EM: TBA-CT with 0.4 g/l eosin Y, 0.06 g/l
to develop visible colonies after 24 h incubation. Strains inhibited on methylene blue and 1 g/l lactose; VTEC: VTEC agar (Gill et al., 2012).
10 A. Gill et al. / Journal of Microbiological Methods 96 (2014) 6–11

Table 4
Appearance of STEC colonies on selective agar.a

Serogroup MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC agar

O26 Pink (10) Mauve (5) Mauve (5) Mauve (7) Mauve (8) White (9) Pink (10) Pink (10)
Blue-grey (5) Blue-grey (4) Pink (2) Pink Pin point
O45 Pink (6) Pink (3) Pink (3) Pink (4) Mauve (3) White (6) Pink (6) Pink (5)
Mauve (1) Mauve Mauve Pink White
Blue-grey (2) Blue-grey (2) White White
Blue-grey
O103 Pink (10) Pink (4) Pink (2) Pink (3) Mauve White (9) Pink (9) Pink (10)
Blue-grey (3) Blue-grey (3) Blue-grey (2) Pink (7)
White White (2) White
Mauve Mauve Mauve
Light pink
O111 Pink (10) Blue-grey (8) Blue-grey (6) Blue-grey (6) Pink (3) White (7) Pink (7) Pink (9)
White Pin Point White Mauve (4) Pin point (3) Mauve White
Mauve Mauve Pin point (2)
Pin Point
O121 Pink (10) Mauve (4) Mauve (4) Mauve (2) Pink (6) White (9) Pink (10) Pink (9)
Blue-grey (2) Blue-grey (2) Pink (7) Mauve (3) Pin point Mauve
Pink (3) Pink (2) Pin point
White Pin point
O145 Pink (9) Blue-grey (3) Blue-grey (2) Blue-grey (3) Pink (6) White (8) Pink (8) Pink (9)
Mauve (6) Mauve (6) Mauve (4) Mauve (2)
Pink
O157 Pink (10) Blue-grey (10) Blue-grey (10) Blue-grey (10) Pink (6) White (10) Pink (10) Light pink (9)
Mauve (4) Pink
Others Pink (30) Pink (9) Pink (3) Pink (8) Pink (8) White (22) Pink (31) Pink (29)
Light pink (1) Mauve (15) Mauve (3) Mauve Mauve (3) Pin point (8) Light pink (2)
Blue-grey (5) Blue-grey (5) Blue-grey (3)
Light pink
Pin point (3)

CH STEC: CHROMagar STEC™; TBA-CT: Tryptone Bile agar with 0.05 mg/l cefixime and 0.5 mg/l of potassium tellurite.
TBA-EM: TBA-CT with 0.4 g/l eosin Y, 0.06 g/l methylene blue and 1 g/l lactose; VTEC: VTEC agar (Gill et al., 2012).
a
MAC: MacConkey Agar; RBA: Rainbow® Agar O157; RBA-NT: Rainbow® Agar O157 with 10 mg/l novobiocin and 0.8 mg/l potassium tellurite; RBA-USDA: Rainbow® Agar O157 with
5 mg/l novobiocin, 0.05 mg/l cefixime trihydrate and 0.15 mg/l potassium tellurite.

4. Discussion characteristics common in the subgroup. These include hemolysin

The isolation of STEC from foods remains a challenge because STEC are
very often present with other non-pathogenic E. coli or Enterobacteriaceae
and no selective condition has been identified which is unique to the
STEC pathotype of E. coli. Since no unique differential characteristic
has been identified, other than verotoxin expression, this means that
isolation requires that potentially large numbers of colonies must be
screened individually for VT genes or expression. Diverse STEC have
been successfully isolated from foods, by identifying VT positive colonies
grown on a relatively permissive agar medium by colony immunoblot
(Atalla et al., 2000) or colony hybridization (Barlow et al., 2006). But
the lack of uptake of these methods reflects the technical challenges in
adopting these methods outside of a research laboratory.
Alternatively, isolation media for specific subgroups of STEC
have been developed which are based on selective or differential
activity (Beutin et al., 1989), carbohydrate fermentation patterns
(Possé et al., 2008) or antimicrobials (Hiramatsu et al., 2002;
Tzschoppe et al., 2012). This approach may be of utility when seeking
the isolation of a STEC strain of known characteristics, but given the di-
versity of STEC strains there is serious concern that STEC pathogens may
not be recovered on agar media containing inhibitory levels of antimi-
crobial agents or which use differential characteristics that may screen
out the pathogen. It should be noted that the diversity of colony mor-
phologies observed on the agar media in this study, even among strains
of a given O type, indicates that colony appearance will not provide a
reliable means of identifying target STEC in the presence of other E. coli.
The problem of media being too selective was demonstrated in this
study by the results obtained with CH STEC, RBA-NT and RBA-USDA.
These media were unable to support the growth of a significant propor-
tion of STEC strains (CH STEC 28.1%, RBA-NT 29.2%, RBA-USDA 20.8%),

Table 5
Recovery of flora from beef enriched in modified tryptic soy broth with 10 mg/l vancomycin and 3 mg/l cefsulodin for 24 h at 42 °C.a

Log CFU/ml

O-type BHI MAC RBA RBA-NT RBA-USDA CH STEC TBA-CT TBA-EM VTEC

Ground beef 8.5 8.6 8.7 b2.0 7.4 b2.0 8.5 8.1 8.1
Ground beef 8.9 8.8 8.8 2.3 8.0 3.5 8.9 8.9 8.9
Ground beef 8.8 8.7 8.8 4.7 8.6 4.3 8.7 8.8 8.8
Mean 8.8a 8.7a 8.8a 3.0b 8.0a 3.3b 8.7a 8.6a 8.6a
Beef trim 8.4 8.5 8.5 3.3 5.6 5.0 8.1 7.3 7.3
Beef trim 8.8 8.6 8.6 7.5 7.6 6.9 8.5 7.4 7.4
Beef trim 8.6 8.5 8.6 4.5 4.8 5.0 8.3 4.3 4.3
Mean 8.6a 8.5ac 8.5a 5.1bc 6.0b 5.6b 8.3a 6.3b 6.3b

BHI: Brain Heart Infusion Agar; MAC: MacConkey Agar; RBA: Rainbow® Agar O157; RBA-NT: Rainbow® Agar O157 with 10 mg/l novobiocin and 0.8 mg/l potassium tellurite; RBA-USDA:
Rainbow® Agar O157 with 5 mg/l novobiocin, 0.05 mg/l cefixime trihydrate and 0.15 mg/l potassium tellurite; CH STEC: CHROMagar STEC™; TBA-CT: Tryptone Bile agar with 0.05 mg/l
cefixime and 0.5 mg/l of potassium tellurite; TBA-EM: TBA-CT with 0.4 g/l eosin Y, 0.06 g/l methylene blue and 1 g/l lactose; VTEC: VTEC agar (Gill et al., 2012).
a
Averages with different superscripts are significantly different by Student's t-test (p b 0.05).
 A. Gill et al. / Journal of Microbiological Methods 96 (2014) 6–11
including strains belonging to 5 of the 6 priority serogroups (O26, O45,
O103, O111, O121, O145) designated for beef testing in the US (USDA,
2012). Inhibition of a significant proportion of STEC strains was also re-
ported in two previous publications which evaluated the performance
of CH STEC and identified resistance to tellurite as the primary selective
principle of the medium (Hirvonen et al., 2012; Tzschoppe et al., 2012).
Hirvonen et al. (2012) reported that CH STEC inhibited the growth of 93
of 362 (25.7%) STEC strains tested and the media was not inhibitory of
the single strains tested of enteropathogenic E. coli, enterotoxigenic
E. coli and enteroaggregative E. coli. The inhibited STEC included O26
(2/20), O103 (19/30) and non-sorbitol fermenting O157:H7 and NM
(4/183) (Hirvonen et al., 2012). Tzschoppe et al. (2012) reported that
CH STEC inhibited the growth of 34 of 133 (25.5%) STEC strains, including
O103 (9/18), O121 (2/4), O145 (1/10), sorbitol fermenting O157 (5/5)
and diverse serotypes (17/17).
When used for the purpose of isolation of target bacteria from
enrichment broth, selective agents added to agar media improve the
probability of recovering the target bacteria by reducing the numbers
of non-target flora which can grow on the media. In this study RBA-
NT and CH STEC were the only agar media which had significantly
lower numbers of colonies from ground beef enrichment (mTSB-VC at
42 °C) compared to non-selective BHI agar. However, RBA-NT, RBA-
USDA, CH-STEC, TBA-EM and VTEC agar were equally inhibitory to the
growth of background microflora from beef trim enrichment broth.
These results suggest that depending on the composition of the initial
microflora of the sample and the enrichment conditions, selective agar
may not always improve the probability of target recovery.
5. Conclusions
The essential requirement of any agar media for the isolation of
bacteria is the ability to support the growth of the target organism, and
secondary to that is the ability to suppress the growth of non-target
organisms. Consequently, we conclude that due to the significant propor-
tion of STEC strains whose growth is inhibited on these media, including
strains of priority serogroups, that the agar media RBA-USDA, RBA-NT,
and CH STEC are unsuitable for application to the isolation of STEC from
beef as the sole isolation media. Nonetheless, RBA-USDA, RBA-NT, and
CH STEC might be applied to beef samples containing high levels of
background microflora, if paired with an agar media capable of
supporting the growth of a wide range of STEC, such as MAC, RBA
or VTEC agar.
Acknowledgements
The authors would like to thank Mahdid Meymandy of Health
Canada and Mylène Deschênes of the Canadian Food Inspection Agency
for their technical assistance.
11
References
Atalla, H.N., Johnson, R., McEwen, S., Usborne, R.W., Gyles, C.L., 2000. Use of a shiga toxin
(Stx)-enzyme-linked immunosorbent assay and immunoblot for detection and isolation
of stx-producing Escherichia coli from naturally contaminated beef. J. Food Prot. 63,
1167–1172.
Barlow, R.S., Gobius, K.S., Desmarchelier, P.M., 2006. Shiga toxin-producing Escherichia coli
in ground beef and lamb cuts: results of a one-year study. Int. J. Food Microbiol. 111,
1–5.
Bettelheim, K.A., 2007. The non-O157 Shiga-toxigenic (verocytotoxigenic) Escherichia
coli; under-rated pathogens. Crit. Rev. Microbiol. 33, 67–87.
Beutin, L., Martin, A., 2012. Outbreak of shiga toxin-producing Escherichia coli (STEC)
O104:H4 infection in Germany causes a paradigm shift with regard to human pathoge-
nicity of STEC strains. J. Food Prot. 75, 408–418.
Beutin, L., Montenegro, M.A., Orskov, I., Orskov, F., Prada, J., Zimmermann, S., Stephan, R.,
1989. Close association of verotoxin (Shiga-like toxin) production with enterohemolysin
production in strains of Escherichia coli. J. Clin. Microbiol. 27, 2559–2564.
CDC, 2012. Foodborne Diseases Active Surveillance Network (FoodNet): FoodNet Surveil-
lance Report for 2011 (Final Report). U.S. Department of Health and Human Services,
CDC, Atlanta, Georgia.
Chui, L., Lee, M.-C., Malejczyk, K., Lim, L., Fok, D., Kwong, P., 2011. Prevalence of shiga
toxin-producing Escherichia coli as detected by enzyme-linked immunoassays and
real-time PCR during the summer months in Northern Alberta, Canada. J. Clin.
Microbiol. 49, 4307–4310.
European Food Safety Authority, 2007. Scientific opinion of the panel on biological
hazards on a request from EFSA on monitoring of verotoxigenic Escherichia coli.
EFSA J. 579, 1–61.
Gill, A.O., Gill, C.O., 2010. Non-O157 verotoxigenic Escherichia coli and beef: a Canadian
perspective. Can. J. Vet. Res. 74, 161–169.
Gill, A., Martinez-Perez, A., McIlwham, S., Blais, B., 2012. Development of a method for the
detection of verotoxin producing Escherichia coli in food. J. Food Prot. 75, 827–837.
Heuvelink, A.E., 2012. Culture media for the isolation of diarrhoeagenic Escherichia coli
from foods. Ch. 16 In: Corry, J.E.L., Curtis, G.D.W., Baird, R.M. (Eds.), Handbook of
Culture Media for Food and Water Microbiology, pp. 321–356.
Hiramatsu, R., Matsumoto, M., Miwa, Y., Suzuki, Y., Saito, M., Miyazaki, Y., 2002. Charac-
terization of Shiga toxin-producing Escherichia coli O26 strains and establishment of
selective isolation media for these strains. J. Clin. Microbiol. 40, 922–925.
Hirvonen, J.J., Siitonen, A., Kaukoranta, S., 2012. Usability and performance of CHROMagar
STEC in detection of shiga toxin-producing Escherichia coli strains. J. Clin. Microbiol.
50, 3586–3590.
Johnson, K.E., Thorpe, C.M., Sears, C.L., 2006. The emerging clinical importance of non-
O157 Shiga toxin-producing Escherichia coli. Clin. Infect. Dis. 43, 1587–1595.
Kaper, J.B., Nataro, J.P., Mobley, H.L., 2004. Pathogenic Escherichia coli. Nat. Rev. Microbiol.
2, 123–140.
Melton-Celsa, A., Mohawk, K., Teel, L., O'Brien, A., 2012. Pathogenesis of shiga-toxin
producing Escherichia coli. Curr. Top. Microbiol. Immunol. 357, 67–103.
Paton, A.W., Paton, J.C., 2003. Detection and characterization of STEC in stool samples
using PCR. An overview. Methods Mol. Med. 73, 45–54.
Possé, B., De Zutter, L., Heyndrickx, M., Herman, L., 2008. Novel differential and confirma-
tion plating media for Shiga toxin-producing Escherichia coli serotypes O26, O103,
O111, O145 and sorbitol positive and negative O157. FEMS Microbiol. Lett. 282,
124–131.
Public Health Agency of Canada, 2010. National enteric surveillance program (NESP).
Annual Summary 2010. (http://www.nml-lnm.gc.ca/NESP-PNSME/surveillance-2010-
eng.html. Accessed 11 March 2013).
Thompson, L.H., Giercke, S., Beaudoin, C., Woodward, D., Wylie, J.L., 2005. Enhanced sur-
veillance of non-O157 verotoxin-producing Escherichia coli in human stool samples
from Manitoba. Can. J. Infect. Dis. Med. Microbiol. 16, 329–334.
Tzschoppe, M., Martin, A., Beutin, L., 2012. A rapid procedure for the detection and isola-
tion of enterohaemorrhagic Escherichia coli (EHEC) serogroup O26, O103, O111,
O118, O121, O145 and O157 strains and the aggregative EHEC O104:H4 strain from
ready-to-eat vegetables. Int. J. Food Microbiol. 152, 19–30.
USDA, 2012. MLG 5B.03 detection and isolation of non-O157 shiga toxin-producing
Escherichia coli (STEC) from meat products. http://www.fsis.usda.gov/PDF/MLG-5B.
pdf (Accessed 17 March 2013).