2
(Otto, 1994), and vegetables (Hopkins, 1993) prompted researchers to look
at environmental
sources as vectors for human infection. Previous efforts suggest the presence
of H. pylori in
ground water, surface water, and other drinking water (Mazari-Hiriart, 2001;
Yingzhi, 2002;
Hegarty, 1999; Hulten, 1998; Hulten, 1995; Shahamat, 1993), therefore implying
a waterborne
route of transmission to humans. The complex methods used in these studies
(polymerase chain
reaction; immunomagnetic separation; autoradiography; enzyme immunoassay)
are expensive to
run, subject to numerous interferences and do not differentiate between viable
and non-viable
organisms. A low cost and effective test to isolate viable H. pylori, similar
to selective media
used for Salmonella or E. coli, from ground and surface water would enable
the drinking water
industry to routinely screen samples. This study focused on efforts towards
development of a
plating media that selects viable H. pylori from real world water samples
containing mixed
microbial populations.
MATERIALS & METHODS
Media development
Bacterial strain management
Five clinical infection (no environmental isolates are available in U.S.
or European type
culture collections) Helicobacter pylori strains were obtained from the American
Type Culture
Collection (ATCC; Manassas, VA) or the Wisconsin State Laboratory of Hygiene
culture
collection (WSLH; Madison, WI; Table 1) and cultured on Brain Heart Infusion
agar (BHI;
Becton Dickinson, Sparks, MD) supplemented with 7% calf serum. The plates
were incubated
under a microaerophillic gas mixture (5% C02; 10% H2; 85% N2; Praxair, Inc.,
Danbury, CT) at
37oC. Frozen stock cultures were prepared by picking isolated colonies from
agar plates and
homogenizing in sodium phosphate buffer to about 107 colony forming units
(cfu)/ml using a
McFarland nephelometer. Each of the five isolates was then frozen in BHI
broth containing 10%
glycerol. Sufficient quantities were prepared to complete the entire study
in order to avoid
multiple passing of strains, which can sometimes lead to phenotypic variability.
The non-
Helicobacter bacteria used to create the complex spiked water samples (Table
1) were obtained
from the WSLH culture collection, grown on BHI agar slants at 35oC and then
stored at 4oC for
up to 3 weeks before re-culturing. The bacteria are listed in table 1.
TABLE 1. Bacterial strains and sources used in preparing water samples containing
known levels of contaminants.
Helicobacter pylori aATCC 43504
Helicobacter pylori ATCC 700392
Helicobacter pylori ATCC 49503
Helicobacter pylori bWSLH 95-10882
Helicobacter pylori WSLH 409013
Acinetobacter WSLH
Aeromonas WSLH
E. coli WSLH
Pseudomonas aeruginosa WSLH
Enterobacter cloacea WSLH
Enterococcus faecalis WSLH
Bacillus WSLH
a American Type Culture Collection.
b Wisconsin State Laboratory of Hygiene culture collection.
				
      
      
				

					
3
Preparation of conventional media
Conventional dehydrated media preparations were chosen for selection of H.
pylori from
mixed microbial populations based on the published clinical literature. The
five types of media
chosen for evaluation were BHI + 7% calf serum (Osaki, 1998), Brucella Agar
(Poms, 2001),
Campylobacter Agar Kit Skirrow (Corry, 1995), Columbia Blood Agar Base (Baronn,
1994;
Becton Dickinson, Sparks, MD), and HPSPA (Jiang, 2000; Stevenson, 2000).
Table 2. compares
ingredients of these media to demonstrate the common and unique features
of each. The
inclusion of either Helicobacter pylori selective supplement (Oxoid Limited,
Hampshire,
England) or Campylobacter selective supplement S (Becton Dickinson) provided
antibiotics to
prevent background while permitting H. pylori growth. Positive control consisted
of the five
conventional media without the selective supplements while negative controls
consisted of
uninoculated plates. All media were prepared according to manufacturer's
or authors'
instructions.
TABLE 2. Formulations of conventional media used to culture and select Helicobacter
pylori.
Brain Heart Infusion Brucella Agar Columbia Agar aHPSPA Campylobacter Agar
10 g proteose peptone 10 g peptamin 10 g pantone 15 g spec. peptone 15 g
proteose peptone
NA beef heart inf. 10 g tryptone 10 g bitone 2 g porcine mucin 2.5. ml liver
digest
NA calf brains inf. 2 g yeast extract 3 g beef heart dig. 5 g yeast extract
5 g yeast extract
2 g dextrose 1 g dextrose 1 g dextrose 5 g beef extract
5 g NaCl 5 g NaCl 5 g NaCl 5g NaCl 5g NaCl
15 g agar 15 g agar 15 g agar 15 g agar 12 g agar
2.5 g disodium phos. 0.1 g Na bisulfate 0.5 g ferrous sulfate
70 ml calf serum w/fe 70 ml calf serum w/fe
0.6 g urea
0.5 g Na pyruvate
bselective supplement selective supplement selective supplement selective
supplement cselective supplement
a Formula published by Jiang and Doyle. J. Clin. Microbiol. 2000.
b Vancomycin (10 mg/L), Cefsulodin (5 mg/L), Trimethoprim (5 mg/L), and Amphotericin
B (5 mg/L).
c Vancomycin (10 mg/L), Trimethoprim (5 mg/L), and Polymyxin B (2500 IU/L)
Initial conventional test media screen
Evaluation of the efficacy of the conventional media formulations used in
clinical microbiology
laboratories was carried out as follows. Each strain listed in Table 1 was
separately cultured on
each of the five media listed in Table 2 in order to evaluate the growth
and inhibition spectra of
individual formulations. First, pure colonies of each strain were picked
from solid growth media
(BHI agar plates) and homogenized in sodium phosphate buffer (4%). Serial
dilutions of each
pure homogenate were immediately spread (0.1ml/plate) onto each type of solid
media with and
without selective supplement added. All plates were incubated under microaerophillic
conditions at 37oC for up to seven days and examined to determine the presence
or absence of
colonies on the media compared to the positive control of a non-selective
media.
Selectivity of media formulations for H. pylori from samples spiked with
a cocktail of non
H. pylori bacteria.
Well water samples containing native flora (Table 5; identified using API
20 E
identification system; Biomerieux Vitek, Inc., Hazelwood, MO) were further
spiked with the
seven strains of background bacteria described above, and the H. pylori,
to represent a highly
contaminated water sample (10,000 cells per strain per 100 mls). The spiked
water sample was
then serially diluted and 0.1 ml of each dilution was spread onto duplicate
plates of each of the