Res. Microbiol.

(~) INSTITUT PASTEUR/ELSEVIER Paris 1992

1992, 143, 183-189

Evaluation of a chemiluminescent DNA probe assay for the rapid confirmation of Listeria monocytogenes O. O k w u m a b u a (l), B. S w a m i n a t h a n (2), P. E d m o n d s (l), J. W e n g e r (2), J. H o g a n (3) a n d M. A l d e n o)

(°Dept. of Biology, Georgia Institute of Technology, Atlanta, GA 30332, re)Meningitis and Special Pathogens Branch, Division of Bacterial Dideases, Center for Infectious Diseases, Centers for Disease Control, Atlanta, GA 30333, and m Gen-Probe inc., San Diego, CA 92!21 (USA) SUMMARY

A Listeria monocytogenes-specific, acridinium-ester-labelled DNA probe was evaluated in a chemiluminescent homogeneous protection assay (HPA} for the rapid confirmation of suspect L, monocyto~enes colonies from blood agar plates. The HPA uses an acridinium-ester-labelled chemiluminescent DNA probe in a free-solution hybridization format. After the DNA probe hybridized with the target ribosomal RNA, the acridinium label on the unhybridized probe was inactivated by a chemical differential hydrolysis step. Formation of a hybrid between probe and target was detected in a luminometer after the addition of a detection reagent. The assay can be completed in 30 to 45 min and allows for simultaneous processing of several (50-100l samples. The probe showed 100% se.~sitivity and 100% specificity for L. monocytogenes when evaluated in the HPA against L. monocytogenes, other Listeria species and other Gram-positive bacteria. The lower detection limit of the HPA was between 104 and 10 s cells, in an evaluation with 296 bacterial colonies isolated from food, the HPA colony confirmation showe~ 100% agreement with conventional biochemical characterization. HPA will be useful for the rapid confirmation of L. monocytogenes isolated from food and clinical specimens. Key-words: DNA, Listeria monocytogenes, Chemiluminescence; Acridinium, Probe, Diagnosis, Food and clinical specimens, Rapid method.

INTRODUCTION

Fleming et al., 1985; Linnan et al., 1988; Schlech et al., 1983), m u c h effort has been focused on the

Since t h e association o f 4 m a j o r o u t b r e a k s o f h u m a n listeriosis with the c o n s u m p t i o n o f contam i n a t e d perishable a n d semi-perishable food products in the 1980s (Anon, 1988; Bula et al., 1988 ;

d e v e l o p m e n t o f m e t h o d s for the rapid detection a n d c o n f i r m a t i o n o f Listeria monoeytogenes in foods, .although several rapid m e t h o d s have been developed, they suffer from certain disadvantages.

Submitted June 17, 199!, accepted September 6, 1991.

Correspondingauthor: B. Swaminathan, Centers for Disease Control, 1-2243, MS D-If Atlanta, GA 30333 (USA). Use of trade names is ['or identification only and does not implyendorsementby the Public Health Serviceor by the U.S. Department of Health and Human Services.

O. OKWUMABUA E T AL.

184

MATERIALS AND METHODS

The conventional approach to laboratory confirmation of L. m o n o c y t o g e n e s from primary isolation plates requires a battery of sequential biochemical tests followed by serotyping. This takes from 2 to 4 days to complete (Cassiday et aL, 1990). Current DNA-probebased assays are cumbersome and require several days for completion (Chenevert et al., 1989; Datt a e t aL, 1987; Klinger et al., 1988). A commercially available assay that uses monoclonai antibodies (Butman et al., 1988) and a commercial non-isotopic D N A probe assay (K!~,nger et al., 1988) detect all Listeria species but are not specific for L. monocytogenes. A colony confirmation assay developed by Datta et al. (1987) uses a 32P-labelled D N A probe and requires 2 days for completion (including 16 h for the hybridization and overnight for autoradiography).

Bacteria and growth conditions

The bacteria used in this study included 5 different serotypes of L. monocytogenes (patient and food isolates), other Listeria species, and bacteria from several genera (table I). The bacteria were subcultured from the preserved stocks (stored frozen in sheep blodd at - 70°C) onto trypticase soy agar plates containing 5°70sheep blood and incubated aerobically at 35°C (25°C for Brochothrix thermosphacta) for 24 h. Preparation of cultures for the determination of the lower limit of detection of HPA

L. monocytogenes serotypes 1/2a (G-2228), 1/2b (G-3134), and 4b (G-2609) were grown for 16 to 24 h in brain-heart infusion broth (BBL Microbiology Systems, Cockeysville, MD) at 35°C with constant shal.'.ing at 155 rpm in an environmental incubator shaker (model G-24, New Brunswick Scientific Co., Edison, N J). The bacterial cells were pelleted by centrifugation for 5 m'~n at 6,500 g (Sorvall Superspeed, RC 2-B, Dupont, Wilmington, DE). The bacterial cell pellets were resuspended in 0.01 M phosphatebuffered saline (PBS) pH 7.2 and adjusted to approximately 8 x l0 s cells/ml (A540 = 0.64) using a "Spectronic 20" spectrophotometer (Bausch and Lomb, Rochester, NY). Serial 1/10 dilutions of the standardized cell suspension were made in PBS and used to determine the minimum detectable limit of the homogeneous protection assay.

Recently, a new assay format involving a chemiluminescent acridinium-ester-labelled D N A probe, termed homogeneous protection assay (HPA), was developed by Gen-Probe Inc. (San Diego, CA) for the confirmation of L. monocytogenes on primary isolation plates. H P A is carried out in a free-solution hybridization format. After hybridization of the D N A probe to its target on the ribosomal RNA, the chemiluminescent label on the unhybridized probe is selectively inactivated by a differential hydrolysis step. The chemiluminescent label associated with the hybridized probe is minimally affected. Formation of a hybrid between probe and target is detected after the addition of a detection reagent in a luminometer.

Colony confirmation from food

Foods obtained from the refrigerators of confirmed listeriosiz patients were. analysed in the Epidemic Investigations Laboratory at CDC as part of an active surveillance program to study the role of foods in sporadic human listeriosis. Thc foods were homogenized 1:1 (wt/vol) in 0.1070 peptone (Difco Laboratories, Detroit, MI). The homogenates were analysed for Listeria monocytogenes by the US Food and Drug Administration method (Lovett ei aL, 1987), US Department of Agriculture method (McClain and Lee, 1988) anti the Nethe:'land.~n~hod (van Netten et al., 1989), as ~a:scribed elsewhere (Cas-

In this article, we describe the evaluation of this new assay with respect to its specificity, sensitivity and minimum detectable limit. We also compared H P A with conventional biochemical and serologic methods for the confirmation of presumptively positive Listeria colonies on blood agar plates.

CFU HPA

= colony-formingunit. = homogeneouspro~'ectionassay.

[

I

RLU

= relativelight unit.

A CHEMILUMINESCENT DNA PROBE FOR L1STERIA MONOCYTOGENES Table L Results of the homogenous protection assay for the bacterial strains tested, showing sensitivity and specificity for L. monocytogenes. Bacterium L. monocytogenes serotype l/2a serotype l / 2 b serotype 1/2c serotype 3b serotype 4b L. seeligeri L. welshimeri L. murrayi L. innocua L. ivanovii L. grayi Jonesia denitrificans Brochothrix thermosphacta Erysipelothrix rhusiopathiae Ccrynebaeterium diphtheriae C. striatum Staphylococcus aureus Streptococcus agalactiae C. aquaticum C. minutissimunt Rhodococcus equi Lactobacillus casei Leuconostoc dextranicum Bacillus subtilis Lactococcus lactis

Probe reactivity 6/6 7/7 !/I i. 1 7/7 0/4 0/2 ~,,'2 0/3 0/2 0/2 O~1 0/1 0/2 0/2 0/1 O~1 O~l 0/1 0/2 O~1 O~1 0/1 O~1 0/1

Probe reactivity = number of strains reacting reactingwith probe/number of strains tested.

siday et aL, 1990). The primary isolation plates were examined for Listeria-like colonies, and as many as 10 suspect Listeria-like colonies from each primary isolation plate were streaked to trypticase soy agar plates containing 5°7o sheep blood and incubated for 16 to 24 h at 35°C. These colonies were used to perform the HPA.

185

serial dilutions of standardized bacterial suspensions was transferred to 1.5-ml microcentrifuge tubes and centrifuged for 5 min in a tabletop microcentrifuge (Fisher Scientific, Pittsburgh, PA). The supernatant was discarded, and 50 ~.1 of reagent 1 was added to each tube. After vortexing for 15 to 30 s, the tubes were incubated for 10 rain in a 37°C water bath. Fifty microlitres of reagent 2 solution was added and the tubes were incubated for 1 h in a 60°C water bath. After incubation, the entire contents of the microfuge tubes (100 ~1) were transferred to 12 x 75 mm polypropylene tubes. Reagent 3 (300 0tl)was added to each tube, followed by incubation for 10 min in a 60°C water bath. The tubes were read immediately in a luminometer as described below. Colony confirmation by HPA

A colony confirmation H P A test kit was supplied by Gen-Probe and was used in accordance with the manufacturer's instructions. The test kit contained lyophilized probe tube, specimen diluent (reagent 1), probe diluent (reagent 2), selection reagent (reagent 3), and detection reagents (reagents 4 and 5). The protocol is outlined in fi~ur¢ i. Briefly, sample preparation consisted of adding 50 izl of reagent 1 to a lyophilized probe tube and transferring one colony from a blood agar plate to the reconstitvted probe tube. The tube was incubated for 5 m..in in a 37°C water bath to lyse the bacteria. The hybridization step consisted of the addition of 50 rtl of reagent 2 to the tube and incubation in a 60°C water bath for 15 min. Three hundred tnicrolitres of reagent 3 was added, vortexed briefly, and incubated in a 60°C water bath for 5 min. The light emitted from the reaction (chemiluminescence) was measured in a luminometer (Leader 1, Gen-Probe, San Diego, CA) equipped with an automated reagent injector. Upon insertion of the assay tube into the sample compartment, the "! ea~,:, i '~ luminometer injects 200 ~tl of reagent 4. After about 2 s delay, 200 ~tl of reagent 5 is automatically injected and the light output is measured quantitatively. The test results were scored as positive when the relative light unit (RLU) output was equal to or greater than 50,000.

Determination of the minimum detectable limit of HPA

gESULTS

Reagents for the initial evaluation of H P A were provided by Gen-Probe Inc. The reagents included lysis mix (reagent 1), probe solution (reagent 2), activator solution (reagent 3), and detection reagents (reagents 4 and 5) (see Arnold et al. (1989) for details on these reagents). They were used in accordance with the instructions supplied by the manufacturer (Gen-Probe, San Diego, CA). One millilitre of the

In evaluation with pure cultures, the H P A s h o w e d 100% sensitivity a n d 100070 specificity for L. monocytogenes. T h e chemiluminescent signal (RLU) for 22 strains o f L. monocytogenes r a n g e d f r o m 778,489 to 1,193,324 with a m e a n o f 982,070. The chemiluminescent signal (RLU) for 32 o t h e r bacteria tested ranged f r o m 840 to

186

O. O K W U M A B U A E T A L .

1. SAMPLiPREPARATION ' J

Add 50~L

~

'~"] ~"] ~'] ~-] ~--I [~-I

Incubate 5 minutes at 37°C. to iyse organisms

reconstituted probe

~ ~ ~ ~-] N N

Add 50juL ProbeDiluent (Reagent21 to reconstituted probe tube 3. DETECTION

. . . .

Transfer 1 colony to the tube

Specimen Diluent(Reagent 1) to lyophilized probe tube

2. HYBRIDIZATION

~

~

Incubate 15 minutes at 60°C.

~-1~ ~ F1~ ~1

)

Add 300juL SelectionReagent(Reagent31 Vortex briefly

Incubate5 minutes at 60°C.

%

Read in a tube luminometer

Fig. 1. Homogeneous protection assay procedure for colony confirmation of L. rnonocytogenes from primary isolation plates.

2,750 with a mean of 1,448. No serotype-bias was observed in reactivities with L . m o n o c y t o g e n e s (table I). The lower detection limit o f the assay was between 104 and 105 C F U / m l (fig. 2). Table II shows the comparative culture and probe assay results for 296 colony confirmation tests performed on isolates from food. The H P A colony confirmation assay showed 100070 agreement with conventional biochemical characterization. Using the biochemical characterization and serotyping as the standards of reference, no false or discordant results were obtained with the HPA.

When bacterial colonies from 4-day-old plates were tested in the H P A , the results were not significant1.', different from those obtained with 16 to 24-h-old colonies (data not shown), suggesting that age o f culture may not significantly affect results. DISCUSSION Although radioisotope-labelled probes offer great sensitivity in hybridization experiments, they present safety and disposal problems to the end user. Therefore, they are not the labels o f

A C H E M I L U M I N E S C E N T D N A P R O B E FOR LISTERIA MONOCYTOGENES

187

~ ~1.2

~"E ~.0 ~,~

,- .~ 0.8

"~ ~,o.~

~,-N~ . ~ ~.~ ~

~

8X10

8X102

8~0~~

8x'~o~8~0,8x;~

Colony Forming Units /Jsteria monocytogenes/ml © serotype 1/2a (G-2228) ~ =- serotype 1/2b(G-3134) ±, serotype 4b (G-2609)

Fig. 2. Minimal detectable limit of the HPA for serotypes of L. monocytogenes. L. innocua (F-4247) was used as the negative control and relative light units less than 2 × 103 were obtained.

Table 11. Comparison of HPA and biochemical/serological results for the confirmation of bacterial colonies

as L. monocytogenes (*).

Colonies isolated from

Positive tests biochemical/ serological

HPA

Negative tests biochemical/ serological

HPA

Beef Lunch Meats Dairy products Sea food Fruits Vegetables Mist (**)

47 8 0 4 0 22 0

47 8 0 4 0 22 0

96 0 12 20 12 21 54

96 0 12 20 12 21 54

Total

81

81

215

215

t*~Suspectcolonieson Listeria selectiveagar platesinoculatedwith enrichmentcultures of naturally contaminatedfoods. ~**lBananabar, snack cakes.

choice for DNA-probe assays ir~tended to be performed in clinical and quality control laboratories. Recently, a great deal o f research effort has been directed towards the development o f nonisotopic labels for D N A probes. In the past, direct primary labels (alkaline phosphatases, horseradish peroxidase) and indirect primary labels (biotin, dioxigenin) have been used to label

DNA for the development of chromogenic and chemiluminescence assays (Bronstein et al., 1989; Leary and Ruth, 1989; Clyne et al., 1989). The direct primary labels react with their corresponding substrate to give a chromogenic, fluorogenic, or chemiluminescent product. The indirect primary labels require an additional reaction step (eg, streptavidin-alkaline phospha-

188

O. OKWUMABUA E T AL.

tase for biotin, peroxidase-labelled antibody for digoxigenin) before the reaction of substrate. Chemiluminescent acridinium ester compounds were first used to label antibodies for immunoassays (Law et al., 1989; Septak, 1989). They have recently been used for the labelling of oligonucleotides (Arnold et aL, 1989; Septak, 1989; Weeks et al., 1983). Acridinium est:-r derivatives can be easily incorporated into. oligonucleotides througk the primary aliphatic amino grou0 on the nucleetide. Other advantages of acridinium-ester-labelledDNA probes are long shlef-life of reagem, short assay time, high detection sensitivity, compatibility with objective light-measuring system~, and amenability to automation of procedures. HPA is a unique type of DNA-probe assay in which separation of excess unhybridized probe is not required. The principle of HPA involving acridinium-ester labelled DNA probe is described in detail elsewhere (Arnold et al., 1989). HPA involves direct chemical labelling of DNA with acridinium ester during DNA synthesis in such a way that the acridinium ester is precisely positioned in an interior location on the probe to selectively protect it upon hybridization of the probe to its target. This technology allows for differential chemical hydrolysis of the acridinium ester label such that there is a rapid loss of chemiluminescence associated with unhybridized probe while chemiluminescence associated with hybridized probe is essentially unaffected (Arnold et al., 1989). Therefore, separation steps are not required to remove unhybridized probe before detection of the probetarget complex. Further, because the hybridization occurs with target and probe DNA in solution, the efficiency of hybridization is very high. The hybridization of the probe to its target is approximately 90e/0 complete in 15 rain (Hogan, J. unpublished data). The results of our evaluation of the GenProbe HPA for the rapid confirmation of L. monocytogenes colonies from culture showed that the assay worked well (tables I and II). In one instance, during the application of HPA to confirm L. monocytogenes colonies from food isolates, 5 colonies from a food enrichment

culture that were presumptively identified as L. monocytogenes (based on haemolytic reaction) were found to be HPA-negative. Upon further characterization, the colonies were identified as L. seeligeri. L. seeligeri and L. monocytogenes cannot be differentiated by their haemolytic activity on blood agar plates. The high sensitivity and specificity of HPA makes this assay a useful alternative to the conventional biochemical tests. Although the lower limit of d~tection of HPA was higher (104 to 105 cells) than expected (103 cells), it was not found to be a critical factor in the assay since the HPA format requires the use of an entire colony. Also, colony size did not affect positive results. Because the assay was specifically optimized for rapid colony confirmation, its minimum detectable limit was probably adversely affected. In conclusion, HPA for colony confirmation of L. monocytogenes shows considerable promise as a rapid alternative to currently used colony confirmation tests. It is sensitive, specific, easy to perform, and can be completed in 30 to 40 min. In contrast, conventional biochemical confirmation tests and other colony hybridization assays may take 1-3 days for completion. The HPA will be very useful to all those who are interested in rapid confirmation of Listeria monocytogenes, including food processors and food regulatory agencies.

Evaluation d'un dosage par sonde ADN chimioluminescente pour la confirmation rapide de |'identit~ de Listeria monocytogenes

Pour la confirmaiton rapide de colonies de Listeria monocytogenes suspect6es sur des plaques de g61ose au sang, nous avons 6valu6, par un dosage de protection homog6ne(HPA) en chimioluminescence, une sonde ADN sp6cifique de L. monocytogenes marqu6e par l'ester d'acridine. Cette sonde est utilis6e pour une hybridation en solution. Apr~s l'hybridation de la sonde ADN avec I'ARN ribosomal cible, le marquage de la sonde non hybrid6e par l'acridine est inactiv6par une hydrolysechimiquediff6rentielle. La formation d'un hybride entre la sonde et la cible est d6tect~e b. l'aide d'un luminom~tre, apr~s addition du r6actif de d~tection. Le dosage peut ~tre ter-

A C H E M I L U M I N E S C E N T DNA P R O B E FOR LISTERIA MONOCYTOGENES mink en 30/t 45 minutes ,:t permet le traitement simultank de nombreux (50-100) kchantillons. La sensibilit6 ¢t la sp6cificit6 de cette sonde sont de 100% pour L. monocytogenes quand riles sont 6valu~es par HPA contre L. mono,Tytogenes ou contre d'autres esp~ces de Listeria ou de bactkries Grsm-positives. La limite infkrieure de dktecfion de I'HPA est de 104-105 cellules. Dans une 6valuation concernant 296 colonies bactkriennes isol6es d'aliments, la confirmation de l'identit6 des colonies par I'HPA est de 100% en accord avec la caract~risation biochimique conventionnelle. L ' H P A se rkv61e une technique utile pour la confirmation rapide de L. monocytogenes isolk d'aliments ou de prklkvements cliniques. Mots-clds: ADN, Listeria monocytogenes, Chimioluminescence; Acridine, Sonde, Diagnostic, Nourriture et sp6cimens cliniques, Mkthode rapide.

References Anon (1988), Listeriose en Sttisse. Bull Bundesamtesfur Gesundheitswesm., 28-29. Arnold, L.J., Hammond, P.W., Wiese, W.A. & Nelson, N.C. (1989), Assay formats involving acridiniumester-labelled DNA probes. Clin. Chem., 35, 1588-1594. Bronstein, I., Voyta, J.C. & Edwards, B. (1989): A comparison of chemiluminescent and colorimetric substrates in a hepatitis B virus DNA hybridization assay. Analyt. Biochem., 180, 95-98. Bula, C., Billie, J. & Mean, F. (1988), Epidemic food-borne listeriosis in western Switzerland. -- 1. Description of the 58 adult cases. Abstracts of the 28th Interscience Conference on Antimicrobial Agents and Chemotherapy, Los Angeles, CA. Butman, B.T., Plank, M.C., Durham, R.J. & Mattingly, J.A. (1988), Monoclonal antibodies which identify a genus-specific Listeria antigen. Appl. environ. MicrobioL, 54, 1564-1569. Cassiday, P.K., Graves, L.M. & Swaminathan, B. (1990), Replica-plating of colonies from Listeria selective agars to blood agar to improve the isolation of Listeria monocytogenes from foods. AppL environ. MicrobioL, 56, 1215-1217. Chenevert, J., Mengaud, J., Gormley, E. & Cossart, P. (1989), A DNA probe specific for L. monocytogenes in the genus Listeria. Int. J. Food MicrobioL, 8, 317-319. Datta, A.R., Wentz, B.A. & Hill, W.E. (1987), Detection

189

of hemolytic Listeria monocytogenes by using DNA colony hybridization. Appl. environ. MicrobioL, 53, 2256-2259. Fleming, D.W., Cochi, S.L., MacDonald, K.L,, Brondum, J., Hayes, P.E., Plikaytis, B.D., Holmes, M.B., Audurier, A., Broome, C.V. & Reingold, A.L. (1985), Pasteurized milk as a vehicle of infection in an outbreak of listeriosis. NewEngl. J. h'Ied., 313,404-407. Klinger, J.D., Johnson, A., Croan, D., Flynn, P., Whippie, D., Kimball, A., Lawrie, J. & Curiale, M. (1988), Comparative studies of nucleic acid hybridization assay for Listeria in foods. J. Assoc. off. Anal. Chem., 71,669-673. Law, S., Miller, T., Piran, U., Klukas, C., Chang, S. & Unger, J. (1989), Novel poly-substituted aryl acridinium esters and their use in immunoassay. J. Biolumin. Chemilumin., 4, 88-98. Leary, J.J. & Ruth, J.L. (1989), Non-radioactive labeling of nucleic acid probes, in "Nucleic acid and monoclonal antibody probes" (B. Swaminathan and G. Pr,~kash) (p. 33-37). Marcel Dekker Inc. New York. Linnan, M.J., Mascola, L., Lou, X.D., Goulet, V., May, S., Salminen, C., Hird, D.W., Yonekura, L., Hayes, P., Weaver, R., Audurier, A., Plikaytis, B.D., Fannin, S.L., Kicks, A. & Broome, C.V. (1988), Epidemic listeriosis associated with Mexican-style cheese. New Engl. J. Med., 319, 823-828. Lovett, J., Francis, D.W. & Hunt, J.M. (1987), Listeria monocytogenes in raw milk : detection, incidence a~d pathogenicity. J. Food. Prot., 50, 188-192. McClain, D. & Lee, W.H. (1988), Development of USDAFSIS method for the isolati'~n of Listeria monocytogenes from raw meat. J. Assoc. off. Anal. Che,,w., 7!, 876-879; 892. Septak, M. (1989), Acridinium-ester-labelled DNA oligonucleotide probes. J. Biolumin. Chemih~min., 4, 351-356. Schlech, W.F., Lavigne, P.M., Bortolussi, R.A., ~llen, A.C., Wort, A.J., Hightower~ A W., J~hn~on, S.E., King, S.H., Nicholls, E.S. & Broome, C.V. (1983), Epidemic listeriosis: evidence for transmission by food. New Engl. J. Med., 308, 203-206. Clyne, J.M., Running, J.A., Stempien, M., Stephens, R.S., Akhavan-Tafti, H., Schaap, A.P. & Urdea, M.S. (1989), A rapid chemiluminescent DNA hybridization assay for the detection of Chlamydia trachomatis. J. Bio,'umin. Chemilumin., 4, 357-366. Van Net;.en, P., Perales, I., van de Moosdik, A., Curtis, G.D.W. & Mossel, D.A.A. (1989), Liquid and solid differential media for the detection and enumeration of L. monocytogenes and other Listeria species, hit. J. Food MicrobioL, 8, 299-316. Weeks, I., Beheshti, I., McCapra, F., Campbell, A K. & Woodhead, J.S. (1983), Acridinium esters as highspecific-activity labels in immunoassay. Clin. Chem., 29, 1474-1479.