- Email: [email protected]
A PCR-based strategy for simple and rapid identification of rough presumptive Salmonella isolates
Journal of Microbiological Methods 35 (1999) 77–84
Journal of Microbiological Methods
A PCR-based strategy for simple and rapid identification of rough presumptive Salmonella isolates a, a b J. Hoorfar *, D.L. Baggesen , P.H. Porting a
¨ , DK-1790 Copenhagen, Denmark Danish Veterinary Laboratory, 27 Bulowsvej b Statens Serum Institut 3 , 5 Artillerivej, DK-2300 Copenhagen, Denmark
Received 27 August 1998; received in revised form 30 October 1998; accepted 2 November 1998
Abstract The purpose of the present study was to investigate the application of ready-to-go Salmonella PCR tests, based on dry chemistry, for final identification of rough presumptive Salmonella isolates. The results were compared with two different biotyping methods performed at two different laboratories. The sensitivity of the BAX Salmonella PCR test was assessed by testing a total of 80 Salmonella isolates, covering most serogroups, which correctly identified all the Salmonella strains by resulting in one 800-bp band in the sample tubes. The specificity of the PCR was assessed using 20 non-Salmonella strains, which did not result in any DNA band. A total of 32 out of the 36 rough presumptive isolates were positive in the PCR. All but one isolate were also identified as Salmonella by the two biochemical methods. All 80 Salmonella strains were also tested in the two multiplex serogroup tests based on PCR beads. All strains belonging to the serogroups B, C 1 , C 2 -C 3 , and D were grouped correctly. Among the 32 rough presumptive isolates identified, 19 isolates resulted in a band of 882 bp (serogroup B), 11 isolates resulted in a band of 471 bp (serogroup C 1 ), and two isolates showed a band of 720 bp (serogroup D). In conclusion, rough presumptive Salmonella isolates can be conveniently confirmed to the serogroup-level, using the pre-mixed PCR tests. The system can be easily implemented in accredited laboratories with limited experience in molecular biology. 1999 Elsevier Science B.V. All rights reserved. Keywords: Salmonella; PCR; Rough isolates; Rapid method
1. Introduction Isolation and identification of Salmonella continues to be an important issue in clinical and applied microbiology. The challenge is met by continuous development of new media and diagnostic tests (Hanai et al., 1997). The final identification and characterization of Salmonella enterica, which is *Corresponding [email protected]
author.
Fax:
1 45-3-530-0120;
e-mail:
based on biochemical reactions followed by serotyping (Popoff and Le Minor, 1997), is usually performed in reference laboratories. However, a minor proportion of presumptive S. enterica isolates, identified by biochemical testing, may lack the O-antigens (rough isolates), or may lack both O- and H-antigens. In addition, reference laboratories occasionally receive strains from other laboratories for verification which have been initially identified as Salmonella spp., but cannot be identified by subsequent serotyping procedures. At the Danish Vet-
0167-7012 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0167-7012( 98 )00108-0
78
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
erinary Laboratory, nearly 10% of | 15 000 strains obtained for serotyping during 1997 required verification to the species level. Most of these strains were obtained from a production environment, such as slaughterhouses, feed mills and food production units. Biochemical substrate utilization is the basis of species identification for Salmonella, although considerable variation can be seen in the biotyping pattern (Brenner, 1984). The majority of Salmonella are recognized as non-lactose fermenters (Lac 2) and hydrogen-sulfide producers (H 2 S 1 ). Thus, conventional approaches require confirmatory testing of all H 2 S 1 and / or Lac 2 colonies, which extend the time for identification. The majority of the H 2 S 1 and Lac 2 colonies turn out not to be Salmonella enterica, but related species such as Proteus or Citrobacter. New genetic methods such as polymerase chain reaction (PCR) are, contrarily, not dependent on substrate utilization or the expression of antigens (Mullis et al., 1986), thereby circumventing the phenotypic variations seen in both biochemical pattern and lack of detectable antigens (Vaneechoutte and van Eldere, 1997). In addition, PCR can be designed to identify Salmonella to the serogroup level, using the rfb gene cluster which contains the genes responsible for biosynthesis and assembly of the O-antigen repeat unit (Lee et al., 1992; Luk et al., 1993). Several Salmonella-specific PCR-assays have been published (reviewed by Olsen et al., 1995). However, it can be difficult to establish tests that can provide reproducible results within and among diagnostic laboratories, due to the wellknown risk of contamination (carry-over problem), presence of DNA-polymerase inhibitors, or variations in the performance of different thermocyclers (Wilson, 1997). The problems are coming into prominence in the light of increasing demand for quality assurance of the end-user laboratories (Dragon et al., 1993; Kitchin and Bootman, 1993). Thus, the purpose of the present study was to investigate the usefulness of ready-to-go Salmonella PCR tests, as a further tool, for final identification of, often problematic, rough presumptive Salmonella isolates. The results were compared with two different phenotypic identification methods performed at two different laboratories.
2. Materials and methods
2.1. Bacterial strains A total of 80 Salmonella strains from different serogroups were examined. The strains were veterinary or food isolates obtained from the collection at the Department of Microbiology, Danish Veterinary Laboratory. A total of 20 non-Salmonella strains examined were from the culture collection of Gothenburg University (CCUG, Gothenburg, Sweden). A total of 36 rough presumptive Salmonella isolates studied were submitted by other food or environmental laboratories to our laboratory for final characterization, including serotyping, due to initial difficulties in identification. The isolation and identification of the 36 isolates submitted by other laboratories were done essentially as described in the ISO method (Anon., 1993). In general, the isolates submitted to our laboratory were identified by typical morphology on selective-indicative medium (Brilliant Green, Rambach, etc.), and by reaction in triple-sugar-iron (TSI) agar, lysin-ironagar and by positive agglutination in a poly-O antiSalmonella antiserum. The serological typing of the Salmonella in our laboratory was according to the Kauffmann–White scheme (Popoff and Le Minor, 1997) with anti-H and anti-O sera (Statens Serum Institut, Denmark). Isolates which were submitted as Salmonella with typical morphology on BrilliantGreen agar were defined as presumptive Salmonella. If the isolates did not react with any anti-Salmonella antisera, they were then presumed non-serotypeable. However, the species had to be verified before the final identification. The isolates which did or did not react with the H-antisera and were auto-agglutinated with the standard O-antisera were defined as rough, although here the species had to be verified.
2.2. Method 1 At the Danish Veterinary Laboratory, the 36 rough presumptive Salmonella isolates used in this study were characterized using the API 20E, system for the identification of Gram-negative bacilli (Cat. no. ´ 20100, bioMerieux sa, Marcy-l’Etoile, France). The
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
test was performed according to the manufacturer’s instructions.
2.3. Method 2 The same 36 rough presumptive Salmonella isolates were cultured from Columbia agar (Cat. no. CM331, Oxoid, Hampshire, UK) onto blood agar (Columbia agar added 5% sterile calf blood), incubated at 378C overnight. A colony was then transferred to an agar slab and submitted to Statens Serum Institut, Copenhagen, Denmark. The strains were identified, blindly, to the species level using conventional biochemical methods (Lautrop et al., 1979; Nissen, 1984) with the following reactions: adonitol, dulcitol mannitol, inositol, salicin, lactose, sucrose, indole, D-glucose, ammonium citrate, Voges Proskauer, urease, O-nitro-phenyl-b-D-galactoside (ONPG), and p-nitrophenyl-b-D-glucopyranosiduronic acid (PGUA).
2.4. Salmonella PCR All the isolates were streaked from Columbia agar to blood agar, incubated at 378C overnight and then treated according to the instruction for BAX Salmonella PCR kit (Cat. no. 17720519, Qualicon Ltd., Warwick, UK) as follows: 200 ml lysis buffer from the kit, containing a protease, were added to a micro-tube containing one colony mass and incubated in a 378C water bath for 20 min, and thereafter transferred to a 958C water bath for 10 min in order to inactivate the residual protease. In another room, totals of 50 ml of these samples were added to each sample tube and control tube
79
each containing a PCR tablet. The tablets contained all reagents necessary for PCR, such as primers, MgCl 2 , dNTP and DNA polymerase. The control tablets contained the same set of salmonella primers, but a different target DNA as control. The PCR tubes were placed into a DNA thermal cycler (PTC-200, MJ Research, Watertown, MA, USA). The reaction was run at 948C for 2 min (primary denaturation); 35 cycles of 948C for 15 s (denaturation) and 728C for 3 min (annealing and extension); and 728C for 7 min (final extension). Products were stored at 48C. The resulting PCR product was a distinctive band of 800 bp as revealed in 1.5% agarose gel electrophoresis (Mullis et al., 1986). The control PCR tubes also resulted in a distinctive band of 800 bp.
2.5. Serogroup PCR Oligonucleotide primers were designed (Table 1) from the sequence of the rfb (O antigen) gene cluster specific for serogroup C 1 (Wang et al., 1992; Lee et al., 1992). For PCR analysis of serogroup B, C 2 -C 3 and D, the primers were as published previously (Kongmuang et al., 1994) (Table 1). All primers ˚ were purchased from DNA Technology (Arhus, Denmark), which used conventional phosphoramidite chemistry for synthesis, and reverse phase-fast cartridge chemistry for purification. Two multiplex PCR tests were designed, one for detection of serogroup B and D (B 1 D PCR), and one for detection of serogroup C 1 and C 2 -C 3 (C PCR). Only the isolates with positive reaction in the Salmonella PCR were further tested in the serogroup tests. The positive isolates from here were tested in
Table 1 Serogroup-specific PCR primers for amplification of Salmonella rfb genes Serogroup
Primer
Amplicon size (bp)
B
59-AGA ATA TGT AAT TGT CAG-39 59-TAA CCG TTT CAG TAG TTC-39 59-GGT TCC ATA AGT ATA TCT-39 59-CTG GAT ACG AAC CCG TAT-39 59-ATG CTT GAT GTG AAT AAG-39 59-CTA ATC GAG TCA AGA AAG-39 59-TCA CGA CTT ACA TCC TAC-39 59-CTG CTA TAT CAG CAC AAC-39
| 882
C1 C 2 -C 3 D
| 471 | 820 | 720
80
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
the B 1 D PCR. In the case of negative results, the isolates were further tested in the C PCR. Five ml of each enzyme-treated sample (the same as for the Salmonella PCR) were added to a microtube, containing a PCR bead (Ready-to-Go PCR Beads, 0.2 ml, Cat. no. 27-9553-01, Pharmacia, Uppsala, Sweden). The primers were added so that the final reaction mixture volumes for both PCR were 25 ml, containing 1 pmol / ml of each primer. The cycle condition for both PCR was 948C for 10 min (primary denaturation); 40 cycles of 948C for 40 s (denaturation), 538C for 30 s (annealing) and 728C for 2 min (extension); and 728C for 10 min (final extension). Products were stored at 48C. The gel electrophoresis detection was identical to the Salmonella PCR.
strains belonging to the serogroup B, C 1 , C 2 -C 3 , and D were grouped correctly (Table 2). All the other Salmonella strains, not from these serogroups, were negative in the serogroup tests. The 20 non-Salmonella strains (Table 3) were not tested in the serogroup tests, since the tests were only optimized to distinguish within the Salmonella enterica species. Among the 32 rough presumptive isolates identified as Salmonella (Table 4), 19 isolates resulted in a band of 882 bp (serogroup B), 11 isolates resulted in a band of 471 bp (serogroup C 1 ), and two isolates showed a band of 720 bp (serogroup D).
2.6. Sensitivity and specificity
In order to increase the sensitivity, specificity and speed of detection of Salmonella, several different DNA methods have been developed (Gopo et al., 1988; Tsen et al., 1989; Cano et al., 1992). However, due to the lack of common genes for toxins or other virulence factors, the approach for isolation of specific DNA probes has been to select randomly cloned chromosomal fragments. So far, six different specific DNA-probes and seven PCR tests for detection of Salmonella have been published based on this technique (Olsen et al., 1995). Furthermore, ribosomal RNA-directed oligonucleotide probes have been used successfully in a single-phase hybridization assay to detect a large number of serovars of Salmonella, except those belonging to subspecies V (Wilson, 1997). Although the target sequence of the Salmonella PCR kit applied in the present study is not disclosed, the results presented indicate a high specificity for Salmonella. Thus, the sensitivity and specificity of the test were found to be 100%. In the present study, testing the presumptive but non-serotypeable Salmonella isolates showed excellent agreement between the results of Salmonella PCR and both biochemical methods applied. However, one isolate (no. 8, Table 4) was identified as Salmonella only in the API 20E-method. This isolate was found to be Citrobacter spp., using sequence analysis method (data not shown). One of the important potential applications of PCR is identification testing. However, the availability of
The sensitivity was calculated on the basis of the number of Salmonella isolates identified as positive in the PCR test. The specificity was calculated on the basis of the number of non-Salmonella isolates identified as negative in the PCR test.
3. Results
3.1. Salmonella PCR The sensitivity of the PCR was assessed by testing a total of 80 Salmonella isolates, covering most serogroups, which correctly identified all the Salmonella strains by resulting in one 800-bp band in the sample tubes (Table 2). The specificity of the PCR was assessed using 20 non-Salmonella strains, which did not result in any DNA band (Table 3). A total of 32 out of the 36 rough presumptive isolates were identified as Salmonella in the PCR and the biochemical methods (Table 4, Fig. 1). However, one isolate (no. 8, Table 4) was positive only by the API-method, suggesting a false-positive identification.
3.2. Serogroup PCR All the 80 Salmonella strains were also tested in the two multiplex serogroup PCR tests. All the
4. Discussion
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
81
Table 2 Evaluation of Salmonella and serogroup PCR tests on 80 isolates of Salmonella a Serotype (ref. no.)
Serogroup
Sal. PCR
Serogroup b PCR
Serotype (ref. no.)
Serogroup
Sal. PCR
Serogroup b PCR
Agona (S19) Bredeney (S3) Coeln (S39) Derby (S17) Essen (S78) Heidelberg (S88) Indiana Kiambu (S55) Saint Paul (S11) Typhimurium (S61) Typhimurium (S121) Typhimurium (S89) Cholerasuis (S106) Infantis (S126) Lille (S118) Mbandaka (S92) Montevideo (E30) Ohio (E32) Oranienburg (S115) Pottsdam (S132) Tennesse (S109) Virchow (S114) Albany (S16) Dunkwa (S46) Goldcoast (S12) Hadar (E23) Manhattan (S47) Muenchen (S48) Baildon (S95) Dublin (S17) Eastbourne (S75) Enteritidis (E8) Enteritidis (S83) Enteritidis (S130) Lundby (S96) Mathura (S97) Moscow (S76) Næstved (S73) Rostock (S74) Adelaide (S71)
B B B B B B B B B B B B C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 D D D D D D D D D D D E1
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P
B B B B B B B B B B B B C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 D D D D D D D D D D D 2
Amsterdam (S10) Anatum (S15) Elisabethville (S40) Falkensee (S1) Give (S13) Lexington 15 1 , 34 1 (S80) Muenster (S9) Weybridge (S136) Broughton (S26) Cannstatt (S14) Krefeld (S5) Niloese (S31) Taksony (S2) Senegal (S41) Veneziana (S64) Farmsen (S7) Havana (S23) Jukestown (S6) Poona (S122) Worthington (S24) Madelia (S49) Hviftingfoss (S30) Carmel (Div778 / 84513.2) IV 16:v 4 z 32 : (S124) Fluntern (S128) Minnesota (S58) Ruiru (S44) Chicago (S52) Wayne (S79) IV 38:z 4 , z 23 : (DS1371 / 90) Hofit (DS130 / 871362) Johannesburg (Div15 / 62) Egusi (S150) II 42:g.t: (DS617 / 91) Kingabwa (S36) Guinea (Div810 / 84823) Bergen (S67) II 47:d:z 39 (S116) IIIb 48:r:z (DS210 / 89) II 50:b:z 6 (DS732 / 91)
E1 E1 E1 E1 E1 E1 E1 E1 E4 E4 E4 E4 E4 F F G G G G G H I J I K L L M N P Q R S T U V X X Y Z
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
a
P, positive; 2 , negative in the serogroup PCR tests. Results of testing in two separate PCR tests for serogroups B 1 D and C 1 1 C 2 -C 3 .
b
methods and reagents for the identification of Salmonella has been historically limited to biochemical and serological approaches. New tests based on novel reagents tend to supplement existing tests, instead of replacing them (Vaneechoutte and van Eldere, 1997). The emerging exception is nucleic acid technology, which is replacing biochemical and
agglutination tests (Vaneechoutte and van Eldere, 1997). The most serious danger to PCR is contamination of specimens with either post-amplification products from previous analyses, the so-called carry-over problem, or contamination of negative specimens with positive specimens prepared at the same work-
82
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
Table 3 Evaluation of BAX PCR on 20 non-Salmonella strains a Strain (reference number)
Bax PCR
Hafnia alvei (497) Citrobacter amalonaticus (4860) Citrobacter braakii (30792) Citrobacter koseri (4859) Serratia marcescens (1647) Shigella flexneri (9566) Proteus mirabilis (138) Klebsiella oxytoca (383) Enterobacter aerogenes (1429) Edwardsiella hoshinae (20937) Enterobacter aerogenes (1429) Enterobacter amnigenus (14182) Enterobacter asburiae (25588) Enterobacter gergoviae (14557) Enterobacter sakazakii (14558) Enterobacter taylorae (18765) Koserella trabulsii (18772) Morganella morganii ss morgan (6328) Obesumbacterium proteus (2078) Pantoea agglomeranes (539)
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
a
2 , negative, Culture Collection, University of Gothenburg (CCUG).
station. Several extensive guidelines have been developed to prevent false-positive results (Kitchin and Bootman, 1993). Application of pre-mixed PCR stable at ambient temperature, based on a novel dry chemistry technique (Ramanujam et al., 1993; Blair et al., 1994), can substantially minimize the risk of carry-over contamination. This in turn can facilitate the implementation of PCR diagnostic tests in accredited laboratories. In particular, the test can simplify the identification procedure by testing presumptive Salmonella colonies directly from indicative agar plates. Another advantage of the PCR kit used is that it does not require further investment in amplicon detection equipment, outside those already available in a conventional PCR laboratory. In addition, due to the flexibility of the sample set-up, it can be used by both small and large laboratories, and it provides the identification results 24 h earlier than biochemical testing. Occurrence of false-negative results, mainly due to the presence of enzyme-inhibitors or poor quality of target DNA, can be controlled by construction of an
internal control sequence, amplified by the same set of primers as the target sequence. Although the Salmonella PCR, used in the present study, included a control sequence, the performance was done in a separate tube than the sample tube. This type of control does not necessarily reflect the amplification condition of the samples, performed in another tube. The correct control would, therefore, have been to include the control sequence, though with a shorter size than 800 bp, in the sample tube. Furthermore, the fact that the Salmonella kit does not require the presence of control amplicon for the interpretation of positive results, implies insignificance of the control included. Some of the control bands were indeed weak, similar to the one in lane 11 (Fig. 1). Another consideration with the PCR kit applied is the low sensitivity of the gel electrophoresis technique in general, which has a detection limit of 10 8 pmol, compared to e.g. fluorescence detection, which is 100 times more sensitive (Kricka, 1992). In addition, the gel-electrophoresis detection, except for the advantage of being available in every PCR laboratory, is time-consuming, laborious and difficult to automate. Furthermore, it is possible that a rough strain would be identified by method 1 and method 2, but not be ‘serotyped’ by PCR, because the gene(s) for the O-antigen have been deleted or mutated. The results of serogroup PCR tests, when used in combination with the Salmonella PCR kit, indicate possibility of further characterization of rough isolates to the serogroup level, particularly in tracing an infection outbreak. The serogroup distribution of the isolates verified was similar to that of our serotypeable isolates, which is usually dominated by serogroup B and D. It is noteworthy that we did not optimize the serogroup PCR tests to be used for direct verification of the suspect isolates, but only on isolates initially found to be positive in the Salmonella PCR. Preliminary efforts in such optimization showed difficulties in finding target sequences to be specific for both Salmonella and the relevant serogroup, as the serogroup sequences cross-reacted with some of the closely-related non-Salmonella strains shown in Table 3. The PCR tests described are incorporated as a further tool in the routine Salmonella identification at
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
83
Table 4 Evaluation of Salmonella and serogroup PCR tests on rough presumptive Salmonella isolates a Sample no.
Method 1
Method 2
Salmonella PCR
Serogroup PCR
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
P P P P P P P 2 P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
P P P P P P P 2 P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
B C1 C1 B B B B ND B B C1 C1 C1 C1 B B B C1 B B B B B D B C1 C1 C1 D C1 B B B ND ND ND
a
P, positive; ND, not determined; 2 , negative; method 1, API 20 E biochemical identification; method 2, conventional biochemical typing.
the Danish Veterinary Laboratory, although their application for the purpose of detection is currently being investigated. In conclusion, rough presumptive Salmonella isolates can be conveniently confirmed to the serogroup-level, using the pre-mixed PCR tests. The system can be easily implemented in accredited laboratories with limited experience in molecular biology. However, further studies are warranted to assess the application of Salmonella PCR in complex biological samples, since inhibitory substances inher-
ent to various samples can interfere with the amplification (Lantz et al., 1994; Wilson, 1997).
Acknowledgements The authors thank Qualicon Europe Ltd. for providing the BAX PCR kits, Ms. Rikke Pedersen for excellent technical assistance and Dr. Peter Ahrens for the sequence analysis.
84
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
Fig. 1. Analysis of PCR-amplified 800-bp fragment by 1.5% agarose gel electrophoresis. Pure colonies of rough presumptive Salmonella isolates were treated with protease and added to sample tubes (A) and control tubes (B). PCR conditions are given in the text. Lane 1, molecular weight marker, lanes 2 through 12, enzymatic treated samples of rough suspect isolates (Table 4, samples no. 15 to 25).
References Anon., 1993. Microbiology — General Guidance on Methods for the Detection of Salmonella. International Standard ISO 6579, Geneva, Switzerland. Blair, P., Ramanujam, R., Burdick, B.A., 1994. Wax-embedded PCR reagents. PCR Methods Appl. 4, 191–194. Brenner, D.J., 1984. Family I. Enterobacteriaceae. In: Krieg, N.R., Holt, J.G. (Eds.), Bergey’s Manual of Systematic Bacteriology, 9th ed. Williams and Wilkins, Baltimore, MD. Cano, R.J., Torres, M.J., Klem, R.E., Palomares, J.C., Casadesus, J., 1992. Detection of salmonellas by DNA hybridization with a fluorescent alkaline phosphatase substrate. J. Appl. Bacteriol. 72, 393–399. Dragon, E.A., Spadoro, J.P., Madej, R., 1993. Quality control of polymerase chain reaction.. In: Persing et al. (Eds.), Diagnostic Molecular Microbiology. American Society for Microbiology, Washington, DC, pp. 160–168. Gopo, J.M., Melis, E., Filipska, E., Meneveri, R., Filipski, J., 1988. Development of a Salmonella-specific biotinylated DNA probe for rapid routine identification of Salmonella. Mol. Cell. Probes 2, 271–279. Hanai, K., Satake, M., Naakanishi, H., Venkateswaran, K., 1997. Comparison of commercially available kits for detection of Salmonella strains in foods. Appl. Environ. Microbiol. 63, 775–778.
Kitchin, P.A., Bootman, J.S., 1993. Quality assurance of the polymerase chain reaction. Med. Virol. 3, 107–114. Kongmuang, U., Luk, J.M., Lindberg, A.A., 1994. Comparison of three stool-processing methods for detection of Salmonella serogroup B, C2, and D by PCR. J. Clin. Microbiol 32, 3072–3074. Kricka, L.J., 1992. Nonisotopic DNA Probe Techniques. Academic Press, New York, p. 11. ¨ ˚ ¨ Lantz, P.G., Hahn-Hagerdal, B., Radstrom, P., 1994. Sample preparation methods in PCR-based detection of food pathogens. Trends Food Sci. Technol. 5, 384–389. Lautrop, H., Høiby, N., Bremmelgaard, A., Korsager, B., 1979. Bakteriologiske undersøgelsesmetoder (Bacteriological investigations), in Danish. FADL’s Forlag, Copenhagen, Denmark. Lee, S.J., Romana, L.K., Reeves, P.R., 1992. Sequence and structural analysis of the rfb (O antigen) gene cluster from a group C1 Salmonella enterica strain. J. Gen. Microbiol. 138, 1843–1855. Luk, J.M.C., Kongmuang, R., Reeves, P.R., Lindberg, A.A., 1993. Selective amplification of abequose and paratose synthase genes (rfb) by polymerase chain reaction for identification of Salmonella major serogroups (A, B, C2, and D). J. Clin. Microbiol. 31, 2118–2123. Mullis, K., Faloona, F., Scharf, S., Siaki, R., Horn, G., Erlich, H., 1986. Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp. Quant. Biol. 51, 263–273. Nissen, B., 1984. Microtest for rapid identification of Enterobacteriaceae. APMIS 92, 239–245. Olsen, J.E., Aabo, S., Hill, W., Notermans, S., Wernars, K., Granum, P.E., Popovic, T., Rasmussen, H.N., Olsvik, Ø., 1995. Probes and polymerase chain reaction for detection of foodborne bacterial pathogens. Int. J. Food Microbiol. 28, 1–78. Popoff, M.Y., Le Minor, L., 1997. Antigenic Formulas of the Salmonella Serovars, 7th Revision. Pasteur Institute, Paris. Ramanujam, R., Koelbl, J., Ting, E., Jolly, J., Burdick, B., 1993. Room-temperature-stable PCR reagents. PCR Methods Appl. 3, 75–76. Tsen, H.Y., Chen, M.H., Sheih, J.S., Wang, S.J., Hu, N.T., 1989. Possible use of a 1.8 kb DNA fragment for the specific detection of Salmonella in foods. J. Ferment. Bioeng. 68, 1–6. Vaneechoutte, M., van Eldere, J., 1997. The possibilities and limitations of nucleic acid amplification technology in diagnostic microbiology. J. Med. Microbiol. 46, 188–194. Wang, L., Romana, L.K., Reeves, P.R., 1992. Molecular analysis of a Salmonella enterica group E1 rfb gene cluster: O antigen and the genetic basis of the major polymorphism. Genetics 130, 429–443. Wilson, I.G., 1997. Inhibition and facilitation of nucleic acid amplification. Appl. Environ. Microbiol. 63, 3741–3751.
Journal of Microbiological Methods
A PCR-based strategy for simple and rapid identification of rough presumptive Salmonella isolates a, a b J. Hoorfar *, D.L. Baggesen , P.H. Porting a
¨ , DK-1790 Copenhagen, Denmark Danish Veterinary Laboratory, 27 Bulowsvej b Statens Serum Institut 3 , 5 Artillerivej, DK-2300 Copenhagen, Denmark
Received 27 August 1998; received in revised form 30 October 1998; accepted 2 November 1998
Abstract The purpose of the present study was to investigate the application of ready-to-go Salmonella PCR tests, based on dry chemistry, for final identification of rough presumptive Salmonella isolates. The results were compared with two different biotyping methods performed at two different laboratories. The sensitivity of the BAX Salmonella PCR test was assessed by testing a total of 80 Salmonella isolates, covering most serogroups, which correctly identified all the Salmonella strains by resulting in one 800-bp band in the sample tubes. The specificity of the PCR was assessed using 20 non-Salmonella strains, which did not result in any DNA band. A total of 32 out of the 36 rough presumptive isolates were positive in the PCR. All but one isolate were also identified as Salmonella by the two biochemical methods. All 80 Salmonella strains were also tested in the two multiplex serogroup tests based on PCR beads. All strains belonging to the serogroups B, C 1 , C 2 -C 3 , and D were grouped correctly. Among the 32 rough presumptive isolates identified, 19 isolates resulted in a band of 882 bp (serogroup B), 11 isolates resulted in a band of 471 bp (serogroup C 1 ), and two isolates showed a band of 720 bp (serogroup D). In conclusion, rough presumptive Salmonella isolates can be conveniently confirmed to the serogroup-level, using the pre-mixed PCR tests. The system can be easily implemented in accredited laboratories with limited experience in molecular biology. 1999 Elsevier Science B.V. All rights reserved. Keywords: Salmonella; PCR; Rough isolates; Rapid method
1. Introduction Isolation and identification of Salmonella continues to be an important issue in clinical and applied microbiology. The challenge is met by continuous development of new media and diagnostic tests (Hanai et al., 1997). The final identification and characterization of Salmonella enterica, which is *Corresponding [email protected]
author.
Fax:
1 45-3-530-0120;
e-mail:
based on biochemical reactions followed by serotyping (Popoff and Le Minor, 1997), is usually performed in reference laboratories. However, a minor proportion of presumptive S. enterica isolates, identified by biochemical testing, may lack the O-antigens (rough isolates), or may lack both O- and H-antigens. In addition, reference laboratories occasionally receive strains from other laboratories for verification which have been initially identified as Salmonella spp., but cannot be identified by subsequent serotyping procedures. At the Danish Vet-
0167-7012 / 99 / $ – see front matter 1999 Elsevier Science B.V. All rights reserved. PII: S0167-7012( 98 )00108-0
78
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
erinary Laboratory, nearly 10% of | 15 000 strains obtained for serotyping during 1997 required verification to the species level. Most of these strains were obtained from a production environment, such as slaughterhouses, feed mills and food production units. Biochemical substrate utilization is the basis of species identification for Salmonella, although considerable variation can be seen in the biotyping pattern (Brenner, 1984). The majority of Salmonella are recognized as non-lactose fermenters (Lac 2) and hydrogen-sulfide producers (H 2 S 1 ). Thus, conventional approaches require confirmatory testing of all H 2 S 1 and / or Lac 2 colonies, which extend the time for identification. The majority of the H 2 S 1 and Lac 2 colonies turn out not to be Salmonella enterica, but related species such as Proteus or Citrobacter. New genetic methods such as polymerase chain reaction (PCR) are, contrarily, not dependent on substrate utilization or the expression of antigens (Mullis et al., 1986), thereby circumventing the phenotypic variations seen in both biochemical pattern and lack of detectable antigens (Vaneechoutte and van Eldere, 1997). In addition, PCR can be designed to identify Salmonella to the serogroup level, using the rfb gene cluster which contains the genes responsible for biosynthesis and assembly of the O-antigen repeat unit (Lee et al., 1992; Luk et al., 1993). Several Salmonella-specific PCR-assays have been published (reviewed by Olsen et al., 1995). However, it can be difficult to establish tests that can provide reproducible results within and among diagnostic laboratories, due to the wellknown risk of contamination (carry-over problem), presence of DNA-polymerase inhibitors, or variations in the performance of different thermocyclers (Wilson, 1997). The problems are coming into prominence in the light of increasing demand for quality assurance of the end-user laboratories (Dragon et al., 1993; Kitchin and Bootman, 1993). Thus, the purpose of the present study was to investigate the usefulness of ready-to-go Salmonella PCR tests, as a further tool, for final identification of, often problematic, rough presumptive Salmonella isolates. The results were compared with two different phenotypic identification methods performed at two different laboratories.
2. Materials and methods
2.1. Bacterial strains A total of 80 Salmonella strains from different serogroups were examined. The strains were veterinary or food isolates obtained from the collection at the Department of Microbiology, Danish Veterinary Laboratory. A total of 20 non-Salmonella strains examined were from the culture collection of Gothenburg University (CCUG, Gothenburg, Sweden). A total of 36 rough presumptive Salmonella isolates studied were submitted by other food or environmental laboratories to our laboratory for final characterization, including serotyping, due to initial difficulties in identification. The isolation and identification of the 36 isolates submitted by other laboratories were done essentially as described in the ISO method (Anon., 1993). In general, the isolates submitted to our laboratory were identified by typical morphology on selective-indicative medium (Brilliant Green, Rambach, etc.), and by reaction in triple-sugar-iron (TSI) agar, lysin-ironagar and by positive agglutination in a poly-O antiSalmonella antiserum. The serological typing of the Salmonella in our laboratory was according to the Kauffmann–White scheme (Popoff and Le Minor, 1997) with anti-H and anti-O sera (Statens Serum Institut, Denmark). Isolates which were submitted as Salmonella with typical morphology on BrilliantGreen agar were defined as presumptive Salmonella. If the isolates did not react with any anti-Salmonella antisera, they were then presumed non-serotypeable. However, the species had to be verified before the final identification. The isolates which did or did not react with the H-antisera and were auto-agglutinated with the standard O-antisera were defined as rough, although here the species had to be verified.
2.2. Method 1 At the Danish Veterinary Laboratory, the 36 rough presumptive Salmonella isolates used in this study were characterized using the API 20E, system for the identification of Gram-negative bacilli (Cat. no. ´ 20100, bioMerieux sa, Marcy-l’Etoile, France). The
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
test was performed according to the manufacturer’s instructions.
2.3. Method 2 The same 36 rough presumptive Salmonella isolates were cultured from Columbia agar (Cat. no. CM331, Oxoid, Hampshire, UK) onto blood agar (Columbia agar added 5% sterile calf blood), incubated at 378C overnight. A colony was then transferred to an agar slab and submitted to Statens Serum Institut, Copenhagen, Denmark. The strains were identified, blindly, to the species level using conventional biochemical methods (Lautrop et al., 1979; Nissen, 1984) with the following reactions: adonitol, dulcitol mannitol, inositol, salicin, lactose, sucrose, indole, D-glucose, ammonium citrate, Voges Proskauer, urease, O-nitro-phenyl-b-D-galactoside (ONPG), and p-nitrophenyl-b-D-glucopyranosiduronic acid (PGUA).
2.4. Salmonella PCR All the isolates were streaked from Columbia agar to blood agar, incubated at 378C overnight and then treated according to the instruction for BAX Salmonella PCR kit (Cat. no. 17720519, Qualicon Ltd., Warwick, UK) as follows: 200 ml lysis buffer from the kit, containing a protease, were added to a micro-tube containing one colony mass and incubated in a 378C water bath for 20 min, and thereafter transferred to a 958C water bath for 10 min in order to inactivate the residual protease. In another room, totals of 50 ml of these samples were added to each sample tube and control tube
79
each containing a PCR tablet. The tablets contained all reagents necessary for PCR, such as primers, MgCl 2 , dNTP and DNA polymerase. The control tablets contained the same set of salmonella primers, but a different target DNA as control. The PCR tubes were placed into a DNA thermal cycler (PTC-200, MJ Research, Watertown, MA, USA). The reaction was run at 948C for 2 min (primary denaturation); 35 cycles of 948C for 15 s (denaturation) and 728C for 3 min (annealing and extension); and 728C for 7 min (final extension). Products were stored at 48C. The resulting PCR product was a distinctive band of 800 bp as revealed in 1.5% agarose gel electrophoresis (Mullis et al., 1986). The control PCR tubes also resulted in a distinctive band of 800 bp.
2.5. Serogroup PCR Oligonucleotide primers were designed (Table 1) from the sequence of the rfb (O antigen) gene cluster specific for serogroup C 1 (Wang et al., 1992; Lee et al., 1992). For PCR analysis of serogroup B, C 2 -C 3 and D, the primers were as published previously (Kongmuang et al., 1994) (Table 1). All primers ˚ were purchased from DNA Technology (Arhus, Denmark), which used conventional phosphoramidite chemistry for synthesis, and reverse phase-fast cartridge chemistry for purification. Two multiplex PCR tests were designed, one for detection of serogroup B and D (B 1 D PCR), and one for detection of serogroup C 1 and C 2 -C 3 (C PCR). Only the isolates with positive reaction in the Salmonella PCR were further tested in the serogroup tests. The positive isolates from here were tested in
Table 1 Serogroup-specific PCR primers for amplification of Salmonella rfb genes Serogroup
Primer
Amplicon size (bp)
B
59-AGA ATA TGT AAT TGT CAG-39 59-TAA CCG TTT CAG TAG TTC-39 59-GGT TCC ATA AGT ATA TCT-39 59-CTG GAT ACG AAC CCG TAT-39 59-ATG CTT GAT GTG AAT AAG-39 59-CTA ATC GAG TCA AGA AAG-39 59-TCA CGA CTT ACA TCC TAC-39 59-CTG CTA TAT CAG CAC AAC-39
| 882
C1 C 2 -C 3 D
| 471 | 820 | 720
80
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
the B 1 D PCR. In the case of negative results, the isolates were further tested in the C PCR. Five ml of each enzyme-treated sample (the same as for the Salmonella PCR) were added to a microtube, containing a PCR bead (Ready-to-Go PCR Beads, 0.2 ml, Cat. no. 27-9553-01, Pharmacia, Uppsala, Sweden). The primers were added so that the final reaction mixture volumes for both PCR were 25 ml, containing 1 pmol / ml of each primer. The cycle condition for both PCR was 948C for 10 min (primary denaturation); 40 cycles of 948C for 40 s (denaturation), 538C for 30 s (annealing) and 728C for 2 min (extension); and 728C for 10 min (final extension). Products were stored at 48C. The gel electrophoresis detection was identical to the Salmonella PCR.
strains belonging to the serogroup B, C 1 , C 2 -C 3 , and D were grouped correctly (Table 2). All the other Salmonella strains, not from these serogroups, were negative in the serogroup tests. The 20 non-Salmonella strains (Table 3) were not tested in the serogroup tests, since the tests were only optimized to distinguish within the Salmonella enterica species. Among the 32 rough presumptive isolates identified as Salmonella (Table 4), 19 isolates resulted in a band of 882 bp (serogroup B), 11 isolates resulted in a band of 471 bp (serogroup C 1 ), and two isolates showed a band of 720 bp (serogroup D).
2.6. Sensitivity and specificity
In order to increase the sensitivity, specificity and speed of detection of Salmonella, several different DNA methods have been developed (Gopo et al., 1988; Tsen et al., 1989; Cano et al., 1992). However, due to the lack of common genes for toxins or other virulence factors, the approach for isolation of specific DNA probes has been to select randomly cloned chromosomal fragments. So far, six different specific DNA-probes and seven PCR tests for detection of Salmonella have been published based on this technique (Olsen et al., 1995). Furthermore, ribosomal RNA-directed oligonucleotide probes have been used successfully in a single-phase hybridization assay to detect a large number of serovars of Salmonella, except those belonging to subspecies V (Wilson, 1997). Although the target sequence of the Salmonella PCR kit applied in the present study is not disclosed, the results presented indicate a high specificity for Salmonella. Thus, the sensitivity and specificity of the test were found to be 100%. In the present study, testing the presumptive but non-serotypeable Salmonella isolates showed excellent agreement between the results of Salmonella PCR and both biochemical methods applied. However, one isolate (no. 8, Table 4) was identified as Salmonella only in the API 20E-method. This isolate was found to be Citrobacter spp., using sequence analysis method (data not shown). One of the important potential applications of PCR is identification testing. However, the availability of
The sensitivity was calculated on the basis of the number of Salmonella isolates identified as positive in the PCR test. The specificity was calculated on the basis of the number of non-Salmonella isolates identified as negative in the PCR test.
3. Results
3.1. Salmonella PCR The sensitivity of the PCR was assessed by testing a total of 80 Salmonella isolates, covering most serogroups, which correctly identified all the Salmonella strains by resulting in one 800-bp band in the sample tubes (Table 2). The specificity of the PCR was assessed using 20 non-Salmonella strains, which did not result in any DNA band (Table 3). A total of 32 out of the 36 rough presumptive isolates were identified as Salmonella in the PCR and the biochemical methods (Table 4, Fig. 1). However, one isolate (no. 8, Table 4) was positive only by the API-method, suggesting a false-positive identification.
3.2. Serogroup PCR All the 80 Salmonella strains were also tested in the two multiplex serogroup PCR tests. All the
4. Discussion
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
81
Table 2 Evaluation of Salmonella and serogroup PCR tests on 80 isolates of Salmonella a Serotype (ref. no.)
Serogroup
Sal. PCR
Serogroup b PCR
Serotype (ref. no.)
Serogroup
Sal. PCR
Serogroup b PCR
Agona (S19) Bredeney (S3) Coeln (S39) Derby (S17) Essen (S78) Heidelberg (S88) Indiana Kiambu (S55) Saint Paul (S11) Typhimurium (S61) Typhimurium (S121) Typhimurium (S89) Cholerasuis (S106) Infantis (S126) Lille (S118) Mbandaka (S92) Montevideo (E30) Ohio (E32) Oranienburg (S115) Pottsdam (S132) Tennesse (S109) Virchow (S114) Albany (S16) Dunkwa (S46) Goldcoast (S12) Hadar (E23) Manhattan (S47) Muenchen (S48) Baildon (S95) Dublin (S17) Eastbourne (S75) Enteritidis (E8) Enteritidis (S83) Enteritidis (S130) Lundby (S96) Mathura (S97) Moscow (S76) Næstved (S73) Rostock (S74) Adelaide (S71)
B B B B B B B B B B B B C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 D D D D D D D D D D D E1
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P
B B B B B B B B B B B B C1 C1 C1 C1 C1 C1 C1 C1 C1 C1 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 C 2 -C 3 D D D D D D D D D D D 2
Amsterdam (S10) Anatum (S15) Elisabethville (S40) Falkensee (S1) Give (S13) Lexington 15 1 , 34 1 (S80) Muenster (S9) Weybridge (S136) Broughton (S26) Cannstatt (S14) Krefeld (S5) Niloese (S31) Taksony (S2) Senegal (S41) Veneziana (S64) Farmsen (S7) Havana (S23) Jukestown (S6) Poona (S122) Worthington (S24) Madelia (S49) Hviftingfoss (S30) Carmel (Div778 / 84513.2) IV 16:v 4 z 32 : (S124) Fluntern (S128) Minnesota (S58) Ruiru (S44) Chicago (S52) Wayne (S79) IV 38:z 4 , z 23 : (DS1371 / 90) Hofit (DS130 / 871362) Johannesburg (Div15 / 62) Egusi (S150) II 42:g.t: (DS617 / 91) Kingabwa (S36) Guinea (Div810 / 84823) Bergen (S67) II 47:d:z 39 (S116) IIIb 48:r:z (DS210 / 89) II 50:b:z 6 (DS732 / 91)
E1 E1 E1 E1 E1 E1 E1 E1 E4 E4 E4 E4 E4 F F G G G G G H I J I K L L M N P Q R S T U V X X Y Z
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
a
P, positive; 2 , negative in the serogroup PCR tests. Results of testing in two separate PCR tests for serogroups B 1 D and C 1 1 C 2 -C 3 .
b
methods and reagents for the identification of Salmonella has been historically limited to biochemical and serological approaches. New tests based on novel reagents tend to supplement existing tests, instead of replacing them (Vaneechoutte and van Eldere, 1997). The emerging exception is nucleic acid technology, which is replacing biochemical and
agglutination tests (Vaneechoutte and van Eldere, 1997). The most serious danger to PCR is contamination of specimens with either post-amplification products from previous analyses, the so-called carry-over problem, or contamination of negative specimens with positive specimens prepared at the same work-
82
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
Table 3 Evaluation of BAX PCR on 20 non-Salmonella strains a Strain (reference number)
Bax PCR
Hafnia alvei (497) Citrobacter amalonaticus (4860) Citrobacter braakii (30792) Citrobacter koseri (4859) Serratia marcescens (1647) Shigella flexneri (9566) Proteus mirabilis (138) Klebsiella oxytoca (383) Enterobacter aerogenes (1429) Edwardsiella hoshinae (20937) Enterobacter aerogenes (1429) Enterobacter amnigenus (14182) Enterobacter asburiae (25588) Enterobacter gergoviae (14557) Enterobacter sakazakii (14558) Enterobacter taylorae (18765) Koserella trabulsii (18772) Morganella morganii ss morgan (6328) Obesumbacterium proteus (2078) Pantoea agglomeranes (539)
2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2 2
a
2 , negative, Culture Collection, University of Gothenburg (CCUG).
station. Several extensive guidelines have been developed to prevent false-positive results (Kitchin and Bootman, 1993). Application of pre-mixed PCR stable at ambient temperature, based on a novel dry chemistry technique (Ramanujam et al., 1993; Blair et al., 1994), can substantially minimize the risk of carry-over contamination. This in turn can facilitate the implementation of PCR diagnostic tests in accredited laboratories. In particular, the test can simplify the identification procedure by testing presumptive Salmonella colonies directly from indicative agar plates. Another advantage of the PCR kit used is that it does not require further investment in amplicon detection equipment, outside those already available in a conventional PCR laboratory. In addition, due to the flexibility of the sample set-up, it can be used by both small and large laboratories, and it provides the identification results 24 h earlier than biochemical testing. Occurrence of false-negative results, mainly due to the presence of enzyme-inhibitors or poor quality of target DNA, can be controlled by construction of an
internal control sequence, amplified by the same set of primers as the target sequence. Although the Salmonella PCR, used in the present study, included a control sequence, the performance was done in a separate tube than the sample tube. This type of control does not necessarily reflect the amplification condition of the samples, performed in another tube. The correct control would, therefore, have been to include the control sequence, though with a shorter size than 800 bp, in the sample tube. Furthermore, the fact that the Salmonella kit does not require the presence of control amplicon for the interpretation of positive results, implies insignificance of the control included. Some of the control bands were indeed weak, similar to the one in lane 11 (Fig. 1). Another consideration with the PCR kit applied is the low sensitivity of the gel electrophoresis technique in general, which has a detection limit of 10 8 pmol, compared to e.g. fluorescence detection, which is 100 times more sensitive (Kricka, 1992). In addition, the gel-electrophoresis detection, except for the advantage of being available in every PCR laboratory, is time-consuming, laborious and difficult to automate. Furthermore, it is possible that a rough strain would be identified by method 1 and method 2, but not be ‘serotyped’ by PCR, because the gene(s) for the O-antigen have been deleted or mutated. The results of serogroup PCR tests, when used in combination with the Salmonella PCR kit, indicate possibility of further characterization of rough isolates to the serogroup level, particularly in tracing an infection outbreak. The serogroup distribution of the isolates verified was similar to that of our serotypeable isolates, which is usually dominated by serogroup B and D. It is noteworthy that we did not optimize the serogroup PCR tests to be used for direct verification of the suspect isolates, but only on isolates initially found to be positive in the Salmonella PCR. Preliminary efforts in such optimization showed difficulties in finding target sequences to be specific for both Salmonella and the relevant serogroup, as the serogroup sequences cross-reacted with some of the closely-related non-Salmonella strains shown in Table 3. The PCR tests described are incorporated as a further tool in the routine Salmonella identification at
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
83
Table 4 Evaluation of Salmonella and serogroup PCR tests on rough presumptive Salmonella isolates a Sample no.
Method 1
Method 2
Salmonella PCR
Serogroup PCR
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36
P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
P P P P P P P 2 P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
P P P P P P P 2 P P P P P P P P P P P P P P P P P P P P P P P P P 2 2 2
B C1 C1 B B B B ND B B C1 C1 C1 C1 B B B C1 B B B B B D B C1 C1 C1 D C1 B B B ND ND ND
a
P, positive; ND, not determined; 2 , negative; method 1, API 20 E biochemical identification; method 2, conventional biochemical typing.
the Danish Veterinary Laboratory, although their application for the purpose of detection is currently being investigated. In conclusion, rough presumptive Salmonella isolates can be conveniently confirmed to the serogroup-level, using the pre-mixed PCR tests. The system can be easily implemented in accredited laboratories with limited experience in molecular biology. However, further studies are warranted to assess the application of Salmonella PCR in complex biological samples, since inhibitory substances inher-
ent to various samples can interfere with the amplification (Lantz et al., 1994; Wilson, 1997).
Acknowledgements The authors thank Qualicon Europe Ltd. for providing the BAX PCR kits, Ms. Rikke Pedersen for excellent technical assistance and Dr. Peter Ahrens for the sequence analysis.
84
J. Hoorfar et al. / Journal of Microbiological Methods 35 (1999) 77 – 84
Fig. 1. Analysis of PCR-amplified 800-bp fragment by 1.5% agarose gel electrophoresis. Pure colonies of rough presumptive Salmonella isolates were treated with protease and added to sample tubes (A) and control tubes (B). PCR conditions are given in the text. Lane 1, molecular weight marker, lanes 2 through 12, enzymatic treated samples of rough suspect isolates (Table 4, samples no. 15 to 25).
References Anon., 1993. Microbiology — General Guidance on Methods for the Detection of Salmonella. International Standard ISO 6579, Geneva, Switzerland. Blair, P., Ramanujam, R., Burdick, B.A., 1994. Wax-embedded PCR reagents. PCR Methods Appl. 4, 191–194. Brenner, D.J., 1984. Family I. Enterobacteriaceae. In: Krieg, N.R., Holt, J.G. (Eds.), Bergey’s Manual of Systematic Bacteriology, 9th ed. Williams and Wilkins, Baltimore, MD. Cano, R.J., Torres, M.J., Klem, R.E., Palomares, J.C., Casadesus, J., 1992. Detection of salmonellas by DNA hybridization with a fluorescent alkaline phosphatase substrate. J. Appl. Bacteriol. 72, 393–399. Dragon, E.A., Spadoro, J.P., Madej, R., 1993. Quality control of polymerase chain reaction.. In: Persing et al. (Eds.), Diagnostic Molecular Microbiology. American Society for Microbiology, Washington, DC, pp. 160–168. Gopo, J.M., Melis, E., Filipska, E., Meneveri, R., Filipski, J., 1988. Development of a Salmonella-specific biotinylated DNA probe for rapid routine identification of Salmonella. Mol. Cell. Probes 2, 271–279. Hanai, K., Satake, M., Naakanishi, H., Venkateswaran, K., 1997. Comparison of commercially available kits for detection of Salmonella strains in foods. Appl. Environ. Microbiol. 63, 775–778.
Kitchin, P.A., Bootman, J.S., 1993. Quality assurance of the polymerase chain reaction. Med. Virol. 3, 107–114. Kongmuang, U., Luk, J.M., Lindberg, A.A., 1994. Comparison of three stool-processing methods for detection of Salmonella serogroup B, C2, and D by PCR. J. Clin. Microbiol 32, 3072–3074. Kricka, L.J., 1992. Nonisotopic DNA Probe Techniques. Academic Press, New York, p. 11. ¨ ˚ ¨ Lantz, P.G., Hahn-Hagerdal, B., Radstrom, P., 1994. Sample preparation methods in PCR-based detection of food pathogens. Trends Food Sci. Technol. 5, 384–389. Lautrop, H., Høiby, N., Bremmelgaard, A., Korsager, B., 1979. Bakteriologiske undersøgelsesmetoder (Bacteriological investigations), in Danish. FADL’s Forlag, Copenhagen, Denmark. Lee, S.J., Romana, L.K., Reeves, P.R., 1992. Sequence and structural analysis of the rfb (O antigen) gene cluster from a group C1 Salmonella enterica strain. J. Gen. Microbiol. 138, 1843–1855. Luk, J.M.C., Kongmuang, R., Reeves, P.R., Lindberg, A.A., 1993. Selective amplification of abequose and paratose synthase genes (rfb) by polymerase chain reaction for identification of Salmonella major serogroups (A, B, C2, and D). J. Clin. Microbiol. 31, 2118–2123. Mullis, K., Faloona, F., Scharf, S., Siaki, R., Horn, G., Erlich, H., 1986. Specific enzymatic amplification of DNA in vitro: the polymerase chain reaction. Cold Spring Harbor Symp. Quant. Biol. 51, 263–273. Nissen, B., 1984. Microtest for rapid identification of Enterobacteriaceae. APMIS 92, 239–245. Olsen, J.E., Aabo, S., Hill, W., Notermans, S., Wernars, K., Granum, P.E., Popovic, T., Rasmussen, H.N., Olsvik, Ø., 1995. Probes and polymerase chain reaction for detection of foodborne bacterial pathogens. Int. J. Food Microbiol. 28, 1–78. Popoff, M.Y., Le Minor, L., 1997. Antigenic Formulas of the Salmonella Serovars, 7th Revision. Pasteur Institute, Paris. Ramanujam, R., Koelbl, J., Ting, E., Jolly, J., Burdick, B., 1993. Room-temperature-stable PCR reagents. PCR Methods Appl. 3, 75–76. Tsen, H.Y., Chen, M.H., Sheih, J.S., Wang, S.J., Hu, N.T., 1989. Possible use of a 1.8 kb DNA fragment for the specific detection of Salmonella in foods. J. Ferment. Bioeng. 68, 1–6. Vaneechoutte, M., van Eldere, J., 1997. The possibilities and limitations of nucleic acid amplification technology in diagnostic microbiology. J. Med. Microbiol. 46, 188–194. Wang, L., Romana, L.K., Reeves, P.R., 1992. Molecular analysis of a Salmonella enterica group E1 rfb gene cluster: O antigen and the genetic basis of the major polymorphism. Genetics 130, 429–443. Wilson, I.G., 1997. Inhibition and facilitation of nucleic acid amplification. Appl. Environ. Microbiol. 63, 3741–3751.