Letters in Applied Microbiology 1998, 27, 116–120

The use of alkaline phosphatase-labelled oligonucleotideprobes as culture confirmation reagents for the identification ofcommercially important bacteria

D.J. Glover 1 and W.J. Harris 2

1The National Collections of Industrial and Marine Bacteria Ltd, Aberdeen, and 2Department of Molecular and CellBiology, Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, UK

1704/97: received and accepted 5 December 1997

D.J. GLOVER AND W.J. HARRIS. 1998. A range of rRNA-targeted alkaline phosphatase-labelled oligonucleotide probes was tested for use as culture confirmation reagentsfor the rapid identification of micro-organisms. The probes were specific to clinicallyimportant bacteria (Helicobacter pylori and Mycobacterium tuberculosis), fish and shellfishpathogens (Renibacterium salmoninarum and Vibrio vulnificus), food spoilage bacteria(Listeria spp. and L. monocytogenes), for bacteria of biotechnological importance (Streptomycesspp.) and for bacteria associated with the oil industry (Sulphate-reducing bacteria,SRB). A universal bacterial probe and a eukaryotic probe were included in the study aspositive and negative controls, respectively. A total of 93 bacterial strains was screened.With the exception of a large number of cross-reactions of the SRB probe (specificity valueof 29·4%) and a single cross-reaction of the R. salmoninarum probe (specificity value of97·7%), dot blot analysis indicated that each probe hybridized 100% specifically to theorganisms tested. A simple culture confirmation method was then developed usingthese probes to enable the identification of bacterial colonies using a simple hybridizationprocedure.

INTRODUCTION

In microbiological processes, such as the spread of disease byair or water, production and spoilage of food and drink,wastewater treatment, bioremediation of pollution and metalextraction, there is a need to determine the specific presenceof micro-organisms so that a particular treatment may beapplied. In industries where samples are screened for thepresence or absence of specific bacteria, for example Listeriamonocytogenes in the food industry or Escherichia coli in thewater industry, there is generally a two tier system for ana-lysis. Samples are primarily screened for the presumptivepresence of the target bacterium and this is followed by aconfirmation of any presumptive positives. Currently, tra-ditional bacterial culture techniques are employed and thetime taken to make a presumptive identification is dependenton the length of incubation required for colonies to develop.Following suggested guidelines, this can take up to 10 d for

Correspondence to : Dr David J. Glover, The National Collections ofIndustrial and Marine Bacteria Ltd, 23 St Machar Drive, Aberdeen,AB24 3RY (e-mail : ncimb@abdn.ac.uk).

© 1998 The Society for Applied Microbiology

organisms such as Legionella (Anon. 1992), 4 d for organismssuch as L. monocytogenes (Anon. 1993) and 18 h for E. coli(Anon. 1994). It is therefore important to make a positiveidentification quickly for some or all of the presumptivepositive colonies.

The current methods employed for colony confirmationinvolve the further culturing of the chosen colonies onchromogenic media which can generally take at least a further24 h incubation time. Alternatively, latex agglutination testsare used. These methods can be subjective when more thanfive colonies are present and furthermore, each separate col-ony requires the use of one test device which can be expensivewhere a high volume of samples is screened.

A range of alkaline phosphatase-labelled oligonucleotideprobes, known as RiboTags, which are specific for the identi-fication of a number of commercially important bacteria,have been produced and tested. Using these probes, a simplemethod of rapid culture confirmation has been developed.The method has been designed to enable any number ofpresumptive colonies to be confirmed as positive or negativeusing a single, simple hybridization procedure. Furthermore,

IDENTIFICATION OF COMMERCIALLY IMPORTANT BACTERIA 117

the whole procedure can be performed in under 6 h of whichless than 1 h is spent hands on.

MATERIALS AND METHODS

Bacterial strains

The bacterial strains used in the sensitivity study, and theirsources, are listed in Table 1. Strains used for specificitytesting are not shown, but organisms which are both closelyand distantly related phylogenetically to the target strains areincluded. Bacteria were obtained in the form of lyophilizedcells and were cultured on both solid and liquid media priorto growth for nucleic acid isolation.

Oligonucleotide probes and primers

Nucleotide sequences of the probes and primers used in thisstudy are shown in Table 2. These were synthesized by OswelDNA Service, Southampton, UK. Alkaline phosphatase-labelled oligonucleotide probes are supplied as culture con-firmation reagents under the name of NCIMB RiboTags.

Preparation of membrane dot blots using PCR-amplified DNA

DNA was isolated from all bacterial strains as described byCarter et al. (1995). The universal eubacterial 16S primers(Edwards et al. 1989) were used to amplify rDNA from mostof the strains tested. Genus-specific 16S primers (Tevere etal. 1996) were used to amplify Mycobacterium DNA, and 23Sprimers (Ludwig et al. 1993) were used to amplify rDNAfor the screening of the Vibro vulnificus-specific probe. Theamplified DNA (1 ml) was spotted onto a nylon membrane(Hybond-N, Amersham International) and fixed by u.v.cross-linking for 2 min. The membranes were air dried andstored in a sealed container at room temperature until hybrid-ization.

Culture confirmation test

Bacterial cultures were grown on a 0·45 mm nitrocellulosemembrane (Millipore) which was placed directly onto an agarplate. This was incubated so that colonies could develop. Thecolonies were lysed using a sodium dodecyl sulphate/sodiumhydroxide procedure and the nucleic acid was fixed to themembrane as described by Sambrook et al. (1989). In somecases, the membranes were incubated at 55 °C in proteinaseK solution (5 mg ml−1 proteinase K in 1×SSC plus 0·2%SDS ; 1×SSC is 0·15 mol l−1 NaCl plus 0·015 mol l−1 Nacitrate) for 30 min to remove cellular proteins which werefound to interfere with the colorimetric detection reagents.At this stage, membranes could be air dried and stored in a

© 1998 The Society for Applied Microbiology, Letters in Applied Microbiology 27, 116–120

sealed container at room temperature. Prior to hybridization,the membranes were pre-washed (5×SSC, 0·5% SDS,1 mmol l−1 EDTA) for 30 min at 50 °C.

Nucleic acid hybridization

Membrane dot blots were pre-hybridized for 1 h at 37 °C in4×SSC (2×SSC for Renibacterium salmoninarum probe) plus10×Denhardts (1×Denhardts is 0·02% Ficoll, Type 400 ;0·02% polyvinylpyrrolidone and 0·02% bovine serum albu-min, Fraction V). NCIMB RiboTags were added to freshhybridization solution at a concentration of 1 pm ml−1 andallowed to hybridize for 1 h. Blots were washed at varioustemperatures and stringencies as indicated in Table 2.

RESULTS AND DISCUSSION

Screening of oligonucleotide probes

This screen was carried out to assess the sensitivity andspecificity of the alkaline phosphatase-labelled oligo-nucleotide probes for use as culture confirmation reagents.Initially, a library of PCR products from each of the strainswas generated and used in the primary screening of a rangeof oligonucleotide probes of various specificities. This wascarried out due to the ease of preparation of a large numberof membrane dot blots using PCR products from organismswhich had vastly differing incubation times. The results ofthis screen are shown in Table 1.

All of the probes showed sensitivity values of 100% oneach of the strains tested, the sensitivity being defined as thepercentage of true positives correctly identified. The resultsof the specificity screen (not shown) showed that specificitiesranged from 100% for most of the probes, to 97·7% for theRenibacterium salmoninarum probe and 29·4% for the SRBprobe. The Renibacterium salmoninarum probe cross-hybrid-ized with Arthrobacter globiformis, a strain whose targetsequence is known to differ by only one base (Mattson et al.1993). From a diagnostic perspective, this is not thought tocause a problem as there have been no reported cases wherethis organism is present in the salmon kidney, the environ-ment where R. salmoninarum is found (Mattson et al. 1993).

The SRB probe was originally designed to target thed-group of purple bacteria, of which SRBs form a major part(Amann et al. 1990). There was, however, generalized cross-reaction with a large number of species tested. Analysis ofthe Ribosomal Database Project using Checkprobe (Maidaket al. 1994) has shown that there are many identical targetsequences for this probe from both closely and distantlyrelated organisms. It was also found that the nucleotidesequence for many other organisms differed by just one ortwo bases towards the centre of this region. The cross-reaction could not be overcome by altering the stringency of

118 D.J. GLOVER AND W.J. HARRIS

Tab

le1

Bac

terial

stra

insan

dhy

brid

izat

ion

resu

ltsw

ith

olig

onuc

leot

ide

prob

esus

edin

this

stud

y

Pro

besp

ecifi

city

*

Str

ain

and

Helicob

acter

Myc

obac

terium

Ren

ibac

terium

Vib

rio

List

eria

List

eria

Strep

tom

yces

Org

anism

sour

ce†

Uni

vers

alEuk

aryo

tic

pylo

ritu

berc

ulos

issa

lmon

inar

umvu

lnifi

cus

spp.

mon

ocyt

ogen

essp

p.SRB

Helicob

acterpy

lori

NC

TC

1163

7¦

−¦

−−

−−

−−

¦N

CT

C11

638

¦−

¦−

−−

−−

−¦

NC

TC

1163

9¦

−¦

−−

−−

−−

¦N

CT

C11

961

¦−

¦−

−−

−−

−¦

NC

TC

1238

5¦

−¦

−−

−−

−−

¦N

CT

C12

368

¦−

¦−

−−

−−

−¦

Myc

obac

terium

BC

GBRI

ND

ND

ND

¦N

DN

DN

DN

DN

DN

Dtu

berc

ulos

isBC

GPA

SN

DN

DN

D¦

ND

ND

ND

ND

ND

ND

Ren

ibac

terium

NC

IMB

2235

¦−

−−

¦−

−−

−¦

salm

onin

arum

NC

IMB

2169

¦−

−−

¦−

−−

−¦

NC

IMB

1111

¦−

−−

¦−

−−

−¦

NC

IMB

1112

¦−

−−

¦−

−−

−¦

NC

IMB

1113

¦−

−−

¦−

−−

−¦

NC

IMB

1114

¦−

−−

¦−

−−

−¦

NC

IMB

1115

¦−

−−

¦−

−−

−¦

NC

IMB

1116

¦−

−−

¦−

−−

−¦

Vib

rio

vuln

ificu

sN

CIM

B20

46¦

−−

−−

¦−

−−

¦N

CIM

B21

36¦

−−

−−

¦−

−−

¦N

CIM

B21

37¦

−−

−−

¦−

−−

¦N

CIM

B21

38¦

−−

−−

¦−

−−

¦N

CIM

B21

39¦

−−

−−

¦−

−−

¦List

eria

mon

ocyt

ogen

esst

.1a

NC

IMB

5000

7¦

−−

−−

−¦

¦−

¦st

.1a

NC

TC

1035

7¦

−−

−−

−¦

¦−

¦st

.2N

CT

C53

48¦

−−

−−

−¦

¦−

¦sv

.4b

NC

TC

1199

4¦

−−

−−

−¦

¦−

¦L.i

nnoc

uaN

CIM

B50

094

¦−

−−

−N

D¦

−−

¦L.i

vano

vii

NC

IMB

5009

6¦

−−

−−

ND

¦−

−¦

L.s

eelig

eri

NC

IMB

5000

0¦

−−

−−

−¦

−−

¦L.w

elsh

imer

iN

CIM

B50

001

¦−

−−

−−

¦−

−¦

Strep

tom

yces

spec

tabi

lisN

CIM

B97

33¦

−−

−¦

−−

−¦

¦S.g

riseus

NC

IMB

1302

3¦

−−

−¦

−−

−¦

¦S.c

inna

mon

eus

NC

IMB

8851

¦−

−−

¦−

−−

¦¦

S.p

urpu

reus

NC

IMB

1131

1¦

−−

−¦

−−

−¦

¦Strep

tom

yces

spp.

NC

IMB

8592

¦−

−−

¦−

−−

¦¦

Desul

fovi

brio

vulg

aris

NC

IMB

8303

¦−

−−

−−

−−

−¦

Des.d

esul

furica

nsN

CIM

B83

07¦

−−

−−

−−

−−

¦D

es.g

igas

NC

IMB

9332

¦−

−−

−−

−−

−¦

Desul

folobu

spro

pion

icus

NC

IMB

1290

7¦

−−

−−

ND

−−

−¦

*ND

,Not

dete

rmin

ed;¦

,pos

itiv

e;−

,neg

ativ

e.†

Gen

omic

DN

Afr

omth

etw

oM

ycob

acterium

tube

rcul

osis

stra

inski

ndly

dona

ted

byD

rK

enFor

bes,

Dep

artm

entof

Med

ical

Mic

robi

olog

y,U

nive

rsity

ofA

berd

een.

© 1998 The Society for Applied Microbiology, Letters in Applied Microbiology 27, 116–120

IDENTIFICATION OF COMMERCIALLY IMPORTANT BACTERIA 119

Table 2 Nucleotide sequences of probes used in this study—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

Probe specificities Sequence (5?-3?) Wash* Reference—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––Universal probe ACG GGC GGT GTG T(G/A)C A Giovannoni et al. 1990Eukaryotic probe TAG AAA GGG CAG GGA A Giovannoni et al. 1990Helicobacter pylori GGA CAT AGG CTG ATC TCT TAG C A Morotomi et al. 1989Mycobacterium tuberculosis GGT GGA AAG CGC TTT AGC GGT A Tevere et al. 1996Renibacterium salmoninarum TAC GGC GCG GAG AAC GTG GAA TGT CCC CCA B Mattson et al. 1993Vibrio vulnificus ACC GTT CGT CTA ACA CAT A Aznar et al. 1994Listeria spp. CTT TGT ACT ATC CAT TGT A B Wang et al. 1991Listeria monocytogenes ATA GTT TTA TGG GAT TAG C B Wang et al. 1991Streptomyces spp. GCG TCG AAT TAA GCC GCC ACA B Stackebrandt et al. 1991Sulphate-reducing bacteria CGG CGT CGC TGC GTC AGG B Amman et al. 1990—––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––

* Wash conditions varied depending on the length and G¦C content of each oligonucleotide probe. A, Wash three times for 1 min in4×SSC plus 0·5% sodium dodecyl sulphate (SDS) at room temperature. B, Wash two times for 10 min in 0·5×SSC plus0·1% SDS at 37 °C.

hybridization. This is not thought to present a problem whenthese probes are used as culture confirmation reagents forassessment of anaerobic environments. For example, anenrichment of micro-organisms under sulphate-reducingconditions would preclude the growth of aerobic organisms.

Culture confirmation test

Following the initial large screen using PCR-generatedrDNA, a range of representative organisms was chosen andthe efficacy of the probes as culture confirmation reagentswas confirmed. The results of one such confirmation test,using the Vibrio vulnificus-specific oligonucleotide probe, isshown in Fig. 1. The application of a hybridization test is

Fig. 1 A membrane dot blot resulting from the hybridization ofa RiboTag specific for the shellfish pathogen, Vibrio vulnificus.The bacterial strains were : (A1) Vibrio vulnificus, (A2) Listonellaanguillarum, (A3) Listeria monocytogenes, (A4) L. welshimeri, (A5–A8) L. monocytogenes, (B1–B4) Legionella pneumophila, (B5)Vibrio carchariae, (B6–B8) Vibrio vulnificus, (C1) Aeromonassalmonicida, (C2) Escherichia coli, (C3) Legionella micdadei, (C4–C8) L. pneumophila, (D8) Vibrio vulnificus

© 1998 The Society for Applied Microbiology, Letters in Applied Microbiology 27, 116–120

simple and requires a series of incubation periods. The totaltime for a positive identification from a bacterial colony isunder 6 h, of which less than 1 h is spent hands on. Further-more, this is a unique culture confirmation test as it is possibleto identify, positively or negatively, any number of bacterialstrains once they have formed colonies after enrichment on amembrane, using a single hybridization procedure.

The results of the screen have shown the potential ofoligonucleotide probes for use as culture confirmationreagents for the identification of micro-organisms. They havethe advantage over biochemical and latex agglutination testsas a single hybridization test enables the rapid identificationof multiple presumptive positive colonies. Work is continuingat NCIMB Ltd to develop further probes for use as cultureconfirmation reagents, and also to use these probes as a basisfor the direct detection and identification of micro-organismsfrom environmental samples without the need for targetenrichment.

REFERENCES

Amman, R.I., Binder, B.J., Olson, R.J., Chisholm, S.W., Devereux,R. and Stahl, D.A. (1990) Combination of 16S rRNA-targetedoligonucleotide probes with flow cytometry for analysing mixedmicrobial populations. Applied and Environmental Microbiology56, 1919–1925.

Anon. (1992) Draft for development DD 211. Method for detectionand enumeration of Legionella organisms in water and relatedmaterials. London : British Standards Institution.

Anon. (1993) BS4285. Microbiological examination for dairypurposes. Section 3.15. Detection of Listeria monocytogenes. Lon-don : British Standards Institution.

Anon. (1994) Methods for the examination of waters and associatedmaterials. The microbiology of water. Part 1—Drinking water.

120 D.J. GLOVER AND W.J. HARRIS

Section 7.7. Count of total coliforms and Escherichia coli by mem-brane filtration. London : HMSO.

Aznar, R., Ludwig, W., Amman, R.I. and Schleifer, K.H. (1994)Sequence determination of pathogenic Vibrio species and whole-cell identification of Vibrio vulnificus with rRNA-targeted oligo-nucleotide probes. International Journal of Systematic Bacteriology44, 330–337.

Carter, J.P., Hsiao, Y.H., Spiro, S. and Richardson, D.J. (1995)Soil and sediment bacteria capable of aerobic nitrate respiration.Applied and Environmental Microbiology 61, 2852–2858.

Edwards, U., Rogal, T., Blocker, H., Embde, N. and Bottger, E.C.(1989) Isolation and direct complete nucleotide determination ofentire genes. Characterisation of a gene encoding for 16S ribo-somal RNA. Nucleic Acids Research 171, 7843–7853.

Giovannoni, S.J., Britschig, T.B., Moyer, C. and Field, K.G. (1990)Genetic diversity in Sargasso Sea bacterioplankton. Nature 345,60–62.

Ludwig, W., Kirchof, G., Klugbauer, N. et al. (1993) Complete23S ribosomal RNA sequences of Gram-positive bacteria with alow DNA G¦C content. Systematic and Applied Microbiology 15,487–501.

Maidak, B.L., Larson, N., McCaughey, J. et al. (1994) The ribo-somal database project. Nucleic Acids Research 22, 3485–3487.

© 1998 The Society for Applied Microbiology, Letters in Applied Microbiology 27, 116–120

Mattson, J.G., Gersdorf, H., Jansson, E., Hongslo, T., Gobel, U.B.and Johanssan, K.-E. (1993) Rapid identification of Renibacteriumsalmoninarum using an oligonucleotide probe complementary to16S rRNA. Molecular and Cellular Probes 7, 25–33.

Morotomi, M., Hoshina, S., Green, P. et al. (1989) Oligonucleotideprobe for detection and identification of Campylobacter pylori.Journal of Clinical Microbiology 27, 2652–2655.

Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Clon-ing : a Laboratory Manual, 2nd edn. Cold Spring Harbor, NY :Cold Spring Harbor Laboratory Press.

Stackebrandt, E., Witt, D., Kemmerling, C., Kroppendstedt, R.and Liesack, W. (1991) Designation of Streptomycete 16S and23S rRNA-based target regions for oligonucleotide probes.Applied and Environmental Microbiology 57, 1468–1477.

Tevere, J.J., Hewitt, P.L., Dare, A. et al. (1996) Detection of Myco-bacterium tuberculosis by PCR amplification with pan-Myco-bacterium primer and hybridisation to an M. tuberculosis-specificprobe. Journal of Clinical Microbiology 34, 918–923.

Wang, R.-F., Cao, W.-W. and Johnson, M.G. (1991) Developmentof a 16S rRNA-based oligomer probe specific for Listeria mono-cytogenes. Applied and Environmental Microbiology 57, 3666–3670.