Note: Descriptions are shown in the official language in which they were submitted.
~ ~314248 3
DNA Probe~ æor Identification and Detection of
Campyobacter jejuni and Campylo~acter coli ba-~ed on Major
OutQr Membrane Protein Gene.
The invention described herein was made in the course of work under
a grant from the Natural Sciences and Engineering Research Council
of Canada (Strategic Grant - Biotechnology).
BACKGROUND OF THE INVENTION
The bacterium Campylobacter jejuni is a major cause of bacterial
gastroenteritis in both developed and developing countries (Blaser,
M. J. and Reller, L. B. 1981. New Eng. J. Med. 305:1444-1452;
Skirrow, M. B. 1982. J. Hyg. Camb. 89:175-184). In contrast
Campylobacter coli is responsible for 2-5% of cases of
Campylobacter diarrhea in developed countries such as Canada
(Karmali, M. A., Penner, J. L., Fleming, P. C., Williams, A. and
Hennessy, J. N. 1983. ~. Infect. Dis. 147:243-246), although it
is responsible for a larger proportion of cases in developing
nations such as Africa (Georges-Courbot, M. C., Baya, C., Beraud,
A. M., Meunier, D. M. and Georges, A. J. 1986. J. Clin. Microbiol.
23:592-594).
C. jejuni and C. coli should be differentiated from one another in
the hospital clinical microbiology laboratory and particularly in
the reference laboratory. However, C. jejuni and C. coli are
differentiated only on the basis of a single biochemical test, the
ability of C. jejuni to hydrolyse hippurate, in contrast to C. coli
which does not (Hwang, M. A. and Ederer, G. M. 1975. J. Clin.
Microbiol. 1:114-115). Moreover, some variants of C. jejuni are
hippurate negative, making differentiation based on hippurate
hydrolysis difficult and not necessarily reliable (Penner, J. L.
1988. Clin. Microbiol. Rev. 1:157-172). The Campylobacter genus
contains not only C. jejuni and C. coli but numerous other
Campylobacter species (14 at last count). Some of these organisms
are found in human stool specimens where they occasionally cause
disease. In addition, many others are found in animal feces where
,~
~i
131~2~8 4
they apparently cause no harm. DNA hybridi~ation has been used in
research laboratories to identify and classify strains of
Campylobacter species (Belland, R. J. and Trust, T. J. 1982. J.
Gen. Microbiol. 128:2515-2522; Hebert, G. A., Edmonds, P. and
Brenner, D. J. 1984. J. Clin. Microbiol. 20:138-140; Ng, L.-K.,
Stiles, M. E. and Taylor, D. E. 1987. Molec. Cell. Probes
:233-243). DNA probes have usually consisted of total chromosomal
DNA isolated from a known species often tagged with a 32P-labelled
nucleotide. It would be useful to obtain specific DNA probes for
identification of C. jejuni and C. coli, since the two species show
a significant amount of DNA homology with one another which results
in cross hybridization. Moreover, it is preferable to use a probe
for a known and stable gene for identification of a particular
species. Therefore, specific DNA probes for identification and
differentiation of C. jejuni and C. coli would be useful in
microbiological laboratories for testing these organisms directly.
The other major reason for developing DNA probes to detect C.
je juni and C. coli depends on the biology of these microorganisms.
Both species are microaerophilic and require a low concentration of
oxygen for growth, 7% compared with the 21% 2 normally found in
air. Therefore, special conditions are required for their
isolation from stool specimens and their growth in the laboratory.
Anaerobic jars containing a special gas mixture or dedicated
incubators are required. A number of selective media, which
contain various antibiotics to kill other bacteria present in
feces, have been devised for growth and selection of C. jejuni and
C. coli (Hutchinson, D. N. and Bolton, F. J. 1984. J. Clin. Path.
37:956-957; Skirrow, M. B. 1977. Br. Med. J. 2:9-11). Although the
procedures are not technically difficult for the trained
microbiologist, they are expensive. Therefore DNA probes which
could be used to detect C. jejuni and C. coli directly in stool
specimens, without need to culture the organisms first, would be
useful and could have commercially significant implications.
13142~8 5
DESCRIPTION OF THE PRIOR ART
Present methods for identifying and detecting the presence of C.
je juni may be found in: Claus P., Moseley, S. L., and Falkow, S.
1982. Prog. Food Nutr. Sci. 7:139-142; Korolik, V., Coloe, P. J.,
and Krishnapillai, V. 1988. J. Gen. Microbiol. 134:521-529, and
Picken, R. N., Wang, Z. and Yang, H. L. 1987. Molec. Cell. Probes
1:245-259.
The United States Patent 4, 358, 535. November 9, 1982. Falkow,
S. and Moseley, S. L. concerns specific DNA probes in Diagnostic
Microbiology.
BRIEF SUMMARY OF THE INVENTION
The subject invention concerns two novel DNA probes, one
specifically for identification of C. je juni (pDT1720) and the
other for identification of both C. jejuni and C. coli (pDT1719).
These probes will enable workers to identify C. jejuni and/or C.
coli and differentiate them after microorganisms have been cultured
and identified as Campylobacter species. The same probes may be
used to identify C. jejuni and C. coli directly in stool specimens
without the need to culture these organisms. Both probes were
selected from a C. jejuni Agtll library as genes which encoded
proteins that reacted with antiserum to a C. jejuni 46 kilodalton
major outer membrane protein.
DESCRIPTION OF THE DRAWING
The Drawing (Figure 1) depicts the endonuclease restriction map of
C. jejuni chromosomal DNA fragments inserted into the plasmid pUC13
to produce recombinant plasmids pDT1719 and pDT1720. Only the map
of the inserted fragments is shown. The size of the inserts is in
base pairs.
DESCRIPTION OF THE SPECIFIC EMBODIMENTS
The subject invention, which is in the field of molecular
biology, concerns DNA probes useful to identify and differentiate
C. jejuni and C. coli, both with cultured organisms or directly in
stool specimens.
,~ ~
~ ,
131~248
These organisms are an important cause of bacterial
gastroenteritis and are difficult to differentiate in the clinical
laboratory with certainty. It is important to differentiate C.
jejuni and C. coli from one another for epidemiologicaI purposes
and to differentiate them from other Campylohacter species as well
as other bacterial species for epidemiological studies and for the
information of the consulting physician. The current high cost of
isolation and growth of C. jejuni and C. coli from stool specimens
is due to the microaerophilic nature of these organisms. This
invention when used to detect these microorganisms directly in
stool specimens will circumvent the requirement for special jars or
a dedicated incubator, expensive media containing antibiotics and
other expensive supplements. It may take up to 72 hours to detect
C. jejuni and C. col i in stool specimens by culture methods. This
time period can be reduced to 18 hours by use of the DNA probes
directly with technology available today. This time could be
reduced further with future developments in DNA hybridization
technology.
These DNA probes of the subject invention could be used to
detect campylobacters in other specimens besides stool specimens,
such as blood and serum samples. Other useful applications include
detection of these organisms in food samples (e.g. chicken, turkey
and game birds), milk and water samples. Since humans become
infected with campylobacter by consuming these items, the DNA
probes would be useful to screen these products at the level of the
food industry to determine the degree of contamination of food by
Campylobacters.
Before detailing the construction and identity of the novel
DNA probes of the subject invention, there are disclosed the
materials and methods employed.
(1) MEDIA
C. jejuni and C. col i were grown from pure culture on Mueller
Hinton agar (Oxoid Ltd., Basingstoke, U.K . ) . Escherichia col i Y10
was grown on L agar (Difco Laboratories, Detroit, Michigan,
U.S.A.).
131424s~ 7
Medium for isolation of C. jejuni and C. coli from feces was the
blood-free charcoal based medium of Hutchinson, D. N. and Bolton,
F. J. (1984. J. Clin. Path. 37:956-957) which contain 35~g
cephaperazone/ml and was supplied by Oxoid Ltd., Basingstoke, U.K.
(2) BACTERIAL STRAINS
The Escherichia coli strains used for the construction of the ~gtll
library was Y1090 (Young, R. A., Bloom, B. R., Grosskinsky, C. M.,
Ivanyi, J., Thomas, D. and Davis, R. W. 1985. Proc. Nat. Acad.
Sci. U.S.A. 82:2583-2587) and for subsequent plasmid cloning E.
coli JM83 (Vieira, J. and Messing, J. 1982. Gene. 19:259-268).
The series of strains of different Campylobacter species (C.
jejuni, C. coli, C. fetus subps. fetus, C. fetus subsp. venerealis,
C. laridis, C. hyointestinalis, C. sputorum, C. pylori) are
maintained in the Diane E. Taylor culture collection at the
University of Alberta. Other species were obtained from the
Department of Medical Microbiology and Infectious Diseases,
University of Alberta and were Pseudomonas aeruginosa, Salmonella
typhimurium, Shigella sonnei, Bacillus subtilis and Staphylococcus
aureus.
(3) PREPARATION OF ANTISERUM
A polyclonal antibody was prepared in rabbits against a 46
kilodalton (kDa) major outer membrane protein (MOMP) of C. jejuni
UA580. The MOMP was prepared from C. jejuni UA580 by extraction
with the detergent Triton X-100~ as described previously (Huyer,
M., Parr, T. R., Jr., Hancock, R. E. and Page, W. J. 1986. FEMS
Microbiol. Lett. 37:247-250). The anti-MOMP antibody was adsorbed
three times against E. coli JM83 and was tested for its activity
against a range of Campylobacter species (Taylor, D. E. and Chang,
N. 1987. Molec. Cell. Probes 1:261-274)-
(4) CONSTRUCTION OF THE C. JEJUNI ~GT11 LIBRARYC. jejuni DNA from strain UA580 was partially digested with the
restriction endonuclease Sau3A1. The resulting DNA fragments were
13l~2~8
"filled-in" using DNA polymerase I holoenzyme supplied by
Boehringer-Mannheim Ltd., Montreal, Quebec to produce blunt-ended
DNA fragments. EcoR1 linkers (supplied by Alberta Regional DNA
Synthesis Laboratory, Calgary, Alberta) were added to the
fragments. The DNA fragments were ligated to dephosphorylated,
EcoR1 cleaved, arms of the cloning vector ~gtll. The DNA was
packaged into bacteriophage lambda heads using the methods
described by the manufacturer of the packaging extracts (Promega
Biotech. Inc., California).
(5) SCREENING OF C. JEJUNI LIBRARY
The ~gtll recombinant library was screened using the anti-MOMP
antiserum by a previously published procedure (Young, R. A., Bloom,
B. P., Grosskinsky, C. M., Ivanyi, J., Thomas, D. and Davis, R. W.
1985. Proc. Nat. Acad. Sci. U.S.A. 82:2583-2587). An alkaline
phosphatase conjugated goat anti-rabbit IgG was used to detect
positive plaques using 5-bromo-4-chloro-3-indoyl phosphate
(BCIP)/nitroblue tetrazolium (NBT) as the colour substrate.
(6) RESTRICTION ENDONUCLEASE DIGESTION AND ISOLATION OF DESIRED
FRAGMENTS
Restriction endonucleases were purchased from Boehringer Mannheim
Canada or Bethesda Research Laboratories Ltd. Digestions were
carried out according to supplier~s instructions. Separation of
fragments was achieved by agarose gel electrophoresis in TBE buffer
(9OmM TRIS, 0.89 M borate, 2mM EDTA). Isolation of the desired
fragment was achieved using low melting point agarose (Biorad
Laboratories Inc.) and the resulting DNA was then purified by
phenol and chloroform extractions ~Maniatis, T., Fritsch, E. F. and
Sambrook, J. 1982. Molecular Cloning: a laboratory manual. Cold
Spring Harbor Laboratory, N. Y.).
(7) PLASMID DNA ISOLATION
Plasmid DNA was isolated by ethidium bromide-cesium chloride
density gradient centrifugation using standard DNA methodology
~`
131~2~
(Birnboim, H. C. and Doly, J. 1979. Nucleic Acids Res.
7:1513-1523).
(8) ISOLATION AND CLONING OF MOMP DNA
DNA from recombinant lambda phage identified as positive using the
anti~MOMP serum was isolated and purified using Lambda Sorb~ phage
absorbent (Promega Biotech Inc. CA.), as described by the
manufacturer. The cloned DNA fragment were isolated from EcoRI
digests of the recombinant phage DNA using low melting point
agarose. The purified fragments were cloned into the EcoRI site of
the cloning vector pUC13.
(9) PREPARATION OF MOMP PROBE DNA
MOMP DNA was isolated for use as probe from EcoRI digests of pUC13
subclones using low melting point agarose. DNA was labelled with
both radioactive and nonradioactive labelling methods. The
radioactive label used was [~-32P]-dATP and the probe DNA was
labelled by nick translation. The nonradioactive probe DNA was
labelled using the nonradioactive DNA labelling kit obtained from
Boehringer Mannheim Ltd. This kit labels DNA using
digoxigenin-labelled dUTP, which was incorporated into the probe
and was detected using alkaline phosphatase conjugated
anti-digoxigenin antiserum and 5-bromo-4-chloro-3-indolyl phosphate
(BCIP)/nitroblue tetrazolium (NBT) as the color substrate.
(10) HYBRIDIZATIONS OF PROBES TO CAMPYLOBACTER DNA
Total DNA was extracted from various Campyl obacter sp. using a
method described previously (Ezaki, T., Takeuchi, N., Liu, S., Kai,
A., Yamamoto, H. and Yabuuchi, E. 1988. Microbiol. Immunol.
32:141-150). The DNA samples were then digested using Sau3AI,
fractionated by agarose gel electrophoresis, and transferred to
nitrocellulose membranes as described previously (Southern, E. M.
1975. J. Mol. Biol. 93:502-517). Hybridizations were carried out
in the presence of 50% formamide at 37C or 42C using
radioactively labelled probe DNA. The membranes were washed twice
in 2xSSC and 0.1% SDS at 65C for 15 min. The membranes were
~'
1314248
exposed to X-ray film (Kodak~ XAR 5) at -70C with an intensifying
screen (DuPont Cronex~ Lightning Plus).
(11) DIRECT PROBING OF STOOL SAMPLES
A total of 140 stool samples were obtained from the University of
Alberta Hospital and the Enteric Section of the Provincial
Laboratory of Northern Alberta. Seventy of the stool samples had
previously been reported to be culture positive for Campylobacter
species using the blood-free charcoal based medium of Hutchinson
and Bolton (10) which contains 35 ~g/ml cefaperazone. All except
one Campylobacte~ spp. (69 samples) were identified as C. jejuni by
hippurate hydrolysis (9) the other was identified as C. coli.
Seventy stool samples which were culture negative for Campylobacter
spp. were also tested. The negative stool samples were stored at
5C prior to testing. Some culture positive stool samples were
tested after storage at 5C, others were frozen at -20C for
several weeks. Freezing did not appear to affect the test results.
Samples were tested in batches of 10. Pure cultures of C. jejuni
and C. coli were tested concurrently with each batch of stool
specimens as positive controls as well as a fecal specimen
containing Salmonella spp. as a negative control. Approximately
200 ~l of each stool sample was suspended in 1.0 ml of
Mueller-Hinton broth and vortexed until suspensions had an even
consistency. Following this, 0.5 ml of the sample was transferred
to 2.0 ml of Mueller-Hinton broth and mixed thoroughly. This
suspension was passed through a Whatman~ #3 filter and a
nitrocellulose membrane filter contained in ~ Millipore Swinex~ 25
filter unit under vacuum. In some cases pressure was applied using
a 10 ml syringe to force the samples through the filter units. The
Whatman #3 filter effectively removed some of the particulate
matter in the stool samples while allowing any bacteria in the
suspension to pass onto the nitrocellulose filter. The filter
units were then filled with a lysing solution of 0.5M NaOH and 1.5M
NaCl and the filters were soaked in this solution for 10 minutes.
The lysing solution was removed by vacuum. The nitrocellulose
filters were then removed from the filter units and transferred to
~'
131~2~8
11
a neutralizing solution (0.5 M Tris-HCl pH 7.5, 1.5 M NaCl) for 10
minutes. This was followed by a 10 minute wash in 2 X SSC. At
this point the filters could be dried for screening later or
screened immediately. The filters were sealed in plastic bags with
approximately 1.0 ml/filter hybridization solution (50~ formamide,
5 X SSC, 0.02% SDS, 0.1% Sarkosine, 5% blocking reagent) as
described in the protocol given by the manufacturer of the
non-radioactive labelling kit (Boehringer Manneheim Ltd.) and were
pre-hybridized in this solution for a minimum of 1 hr at 37C.
Fresh hybridization solution was added containing freshly heat
denatured non-radioactively labelled probe (approximately 0.1 ~g of
probe DNA/filter). The filters were incubated at 37C for 1 to 2
hrs. The filters were then rinsed briefly in 2 x SSC and placed in
20 mls of 0.5% blocking reagent solution (150mM NaCl, 100mM
Tris-HCl pH 7.5, 0.5% w/v Boehringer Mannheim blocking reagent) at
room temperature for 30 minutes. The filters were then incubated
in a 1:5000 dilution of alkaline phosphatase conjugated
anti-digoxigenin antiserum in TBS (150mM NaCl, 100mM Tris-HCl pH
7.5) for 1 hr at room temperature. The filters were washed three
times for 5 mins in TBS at room temperature and placed in the color
developing solution. Color was allowed to develop for at least one
hour or to a maximum of 24 hours in reduced light. Filters were
air dried and stored in reduced light. Most filters were examined
after 18 hours. A strong positive hybridization signal was denoted
by a deep purple color, a weak one by a pale mauve. Negative
hybridization was denoted by a filter the color of the fecal
material itself, which ranged from white, yellow, brown or even
pink, if the stool specimen contained blood.
EXAMPLE 1 - IDENTIFICATION AND DIFFERENTIATION OF C. JEJUNI AND C.
CO~I USING MOMP DNA PROBES
From the ~gtll library prepared with DNA from C. jejuni UA580,
positive plaques were detected with a polyclonal antiserum to a 46
kDa MOMP from C. jejuni UA580. Bacteriophage lambda was isolated
from the plaques and propagated in E. coli Y1090. DNA fragments
. ~ ,
.~,
1 3 1 ~ 2 4 8 12
from C. jejuni UA5800 were then cloned into pUC13 using the EcoRI
site within the polylinker to give recombinant plasmids pDtl719 and
pDT1720 containing DNA fragments of 1845-bp and 1475-bp,
respectively. The restriction endonuclease maps of the two
fragments are shown in the Drawing. Although the restriction
endonuclease maps of the two fragments indicate that the two
fragments are not closely related, hybridization experiments with
dot blots and Southern blots demonstrated some cross-hybridization
under low to moderate stringency conditions (37C), whereas at 42C
(high stringency condition) very little cross-hybridization was
observed.
Although E. coli containing the ~gtll vector into which the C.
jejuni DNA fragments had been ligated, expressed a polypeptide that
was detected in immunoblots using anti-MOMP antibody, the pUC13
derivatives pDT1719 and pDT1720 did not specify any MOMP related
polypeptides. This is expected since ~gtll is an expression vector
whereas pUC13 is not.
Initial studies to determine probe specificity employed probes
labelled by nick translation with 32P-dATP. The probe prepared
from pDT1720 hybridized only with C. jejuni DNA under conditions of
both high stringency (42C) and moderate stringency (37C). The
probe prepared from pDT1719 hybridized with C. jejuni and C. coli
DNA and weakly with DNA from two out Gf five strains of C. laridis
when hybridizations were performed at 37C. Raising the
temperature of hybridization to 42C to give conditions of higher
stringency reduced the degree of hybridization of pDT1719 with C.
coli and eliminated hybridization with C. laridis. Neither of the
probes hybridized with other Campylobacter species including C.
fetus subsp fetus, C. fetus subsp venerealis, C. hyointestinalis,
C. sputorum, C. upsaliensis or C. pylori. Neither did they
hybridize with other unrelated bacteria, including E. coli,
Salmonella typhimurium, S. typhi, Klebsiella spp., Shigella sonnei,
Bacil l us subtilis nor Staphylococcus aureus.
Southern blot analysis was used to detect probe sequences
present in C. jejuni and C. coli DNA. DNA isolated from eight C.
jejuni strains, and four C. coli strains obtained from different
; ~ ^ '
i ~"
1314248
13
geographic locations. DNA isolated from these strains was digested
with Sau3Al and hybridized with the two probe insert DNAs. The
pDT1719 probe hybridized with two major Sau3A1 DNA fragments of
675-bp and 530-bp in all C. jejuni strains tested. In contrast in
the C. coli Sau3A1 digest this DNA probe hybridized predominantly
to the 530-bp fragment. Hybridizations carried out with the probe
from pDT1720 with Sau3A1 digests of DNA from C. jejuni strains
showed one strongly hybridizing band of 1475-bp in the majority of
the strains. This fragment is the same size as the probe itself.
In two C. jejuni strains (UAl and UA560) another slightly smaller
band (1400-bp) was also found to hybridize with the pDT1720 probe.
~herefore these probes may be useful to compare strain differences
based on restriction fragment length polymorphisms.
EXAMPLE 2 - DETECTION OF C. JEJUNI AND C. COII IN STOO~ SPECIMENS
USING THE MOMP DNA PROBES
The detection limits of pDT1719 and pDT1720 insert DNA were
determined using pure cultures of C. jejuni tUA580) and C. coli
(UA33). The pDT1720 probe was tested with Gnly the C. jejuni
culture whereas the pDT1719 probe was tested with both C. jejuni
and C. coli cultures. Detection limits were tested with both
radiolabelled and non-radiolabelled probes. Cultures were usually
diluted and lysed in situ on nitrocellulose disks. The detection
limits of the pDT1720 probe for C. jejuni was determined to be
approximately 5 x 105 organisms with the 32P-labelled probe and 1-2
x 105 with the Boehringer Mannheim Ltd. non-radioactive probe.
Similar results were obtained using the pDT1719 probe with C.
jejuni. However, for C. coli the detection limit of the pDT1719
probe was approximately 8 x 107 organisms with the radiolabelled
probe and 3 x 107 with the non-radiolabelled probe.
The two probes were used to detect C. jejuni and C. col i
directly in stool specimens. The non-radioactive labelling method
was chosen for this work, as non-radioactive reagents are more
acceptable in clinical laboratories, whereas radioactive methods
are more suited to research environments. A total of 140 stool
specimens were tested with both probes. Seventy of the stools were
1314248
14
culture positive for Campylobacter species ( 69 C. jejuni and 1 C
coli). With the pDT1719 probe 55/70 of the samples gave strong
positive reactions and 7 gave weak positive reactions. A strong
positive hybridization signal was denoted by a deep purple colour,
a weak one by a pale mauve colour. The results correspond to a
sensitivity of 89%. With the pDT1720 probe 58/70 of the samples
gave a strong positive reaction and 6 give weak positive reaction.
The C. coli was among those positively identified with the pDT1719
probe but not with the pDT1720 probe. Therefore the pDT1720 probe
demonstrated a sensitivity of 93% based on 69 culture positive C.
jejuni samples.
Of the seventy stool specimens which were culture negative for
C. jejuni, 11 gave a false-positive reaction. The same negative
stool samples were identified as positive no matter which probe was
used. The approximately 15% rate of false positives could be
explained by the presence of enzymes in the feces, either alkaline
phosphatase or related esterases, which would act on the substrate
directly or otherwise affect the dye and give a false positive
reaction. Other non-radioactive labelling systems may give a lower
proportion of false positives.
The two examples given demonstrate some of the uses of the C.
jejuni and C. coli gene probes based on major outer membrane
protein encoding sequences. It is likely that certain changes and
modifications may be practiced within the scope of the appended
claims.
The embodiments of the invention in which the exclusive
property or privilege is claimed are defined as follows:
