OUP user menu

Development of a multiplex PCR assay for the simultaneous detection and identification of Arcobacter butzleri, Arcobacter cryaerophilus and Arcobacter skirrowii

Kurt Houf, Ann Tutenel, Lieven De Zutter, Jan Van Hoof, Peter Vandamme
DOI: http://dx.doi.org/10.1111/j.1574-6968.2000.tb09407.x 89-94 First published online: 1 December 2000


A multiplex PCR assay with five primers targeting the 16S and 23S rRNA genes was developed for the simultaneous detection and identification of Arcobacter butzleri, Arcobacter cryaerophilus and Arcobacter skirrowii. The selected primers amplify a 257-bp fragment from A. cryaerophilus, a 401-bp fragment from A. butzleri and a 641-bp fragment from A. skirrowii. No PCR product was generated for closely related bacteria including Campylobacter and Helicobacter species. The assay was useful to identify cultures after in vitro cultivation and to detect and identify A. butlzeri and A. cryaerophilus from poultry samples present in 24-h old enrichment in Arcobacter broth with cefoperazone, amphotericin and teicoplanin (CAT)-supplement.

  • Multiplex PCR
  • Poultry
  • Arcobacter

1 Introduction

The genus Arcobacter belongs to the rRNA superfamily VI of the Proteobacteria and comprises four species [1]. Two species, Arcobacter butzleri and Arcobacter cryaerophilus are found in animal livestock [24], food of animal origin [59] and water [1012] and are associated with human diarrheal illness and bacteremia [1319]. Arcobacter skirrowii is also found in animals but has not been isolated from other sources [20]. The fourth species, Arcobacter nitrofigilis, is a nitrogen-fixing bacterium associated with roots of Spartina alterniflora, a salt marsh plant. No association with animal or human infection is known [21]. Arcobacter, like Campylobacter, has been isolated more frequently from poultry than from red meat [22]. Therefore, poultry may be a significant reservoir of Arcobacter. Differences in recovery rates between several surveys of poultry products in different countries have been described [2224]. This could be due to different hygienic conditions in the abattoirs or to differences in the sensitivity of the isolation methods used. Some arcobacters have fastidious growth requirements and because they are rather biochemically inert and morphologically similar to campylobacters, incorrect detection and identification might occur [25]. Differentiation between these genera is based on differences in optimal growth temperature and the ability of Arcobacter species to grow in aerobic conditions [1]. DNA-based assays used for the identification of Arcobacter species, are more rapid and have a higher specificity than conventional identification methods. For routine identification of Arcobacter at the genus level and of A. butzleri, the use of DNA probes and PCR-based methods has been previously described [2630]. The PCR-based identification scheme described by Bastijns et al. [31] required two distinct DNA amplifications at different annealing temperatures, for species level identification of all veterinary Arcobacter species.

The aim of this work was to develop a sensitive multiplex PCR (m-PCR) assay targeting the 16S and 23S rRNA genes for the simultaneous identification of A. butzleri, A. cryaerophilus and A. skirrowii. The assay was used for rapid detection and identification of Arcobacter species in two enrichment broths. Subsequently, the m-PCR was used to identify isolates picked up from agar plates that have been inoculated from the enrichment broths.

2 Materials and methods

2.1 Bacterial strains and culture conditions

Reference strains of Arcobacter species (Table 1) and closely related bacteria (Table 3) were obtained from the BCCM/LMG Bacteria collection, Ghent University (Ghent, Belgium). Arcobacter reference strains were grown in Mueller Hinton broth supplemented with 5% lysed horse blood and incubated for 48 h at 30°C under microaerobic conditions by evacuating 80% of the normal atmosphere and introducing a gas mixture of 8% CO2, 8% H2 and 84% N2 into the jar. Cultivation of the other test organisms (Table 3) was performed according their specific needs.

View this table:

Arcobacter reference strains used in this study

View this table:

Reference strains included for specificity in this study

2.2 Primer design

Five PCR primers, ARCO, BUTZ, SKIR, CRY1 and CRY2, based on the 16S rRNA and 23S rRNA sequences published in GenBank were designed using the GeneCompar 2.1 software package (Applied Maths) (Table 2). The primers were checked against the GenBank database to ensure that they did not have significant identity to sequences from non-Arcobacter organisms.

View this table:

Sequence and origin of the sets of primers

2.3 PCR conditions and electrophoresis

PCR reactions were performed in a reaction mixture (50 μl final volume) containing 2 μl of lysed bacteria, 5 μl of Gibco BRL 10×PCR buffer, 1.5 U of Taq DNA polymerase (Gibco), 0.2 mmol l−1 of each deoxyribonucleotide triphosphate, 1.3 mmol l−1 MgCl2, 5 μl of loading buffer (4 mM cresol red, 60% sucrose) and 50 pmol of the primers ARCO, BUTZ, CRY1, CRY2, and 25 pmol of primer SKIR. PCR involved 32 cycles of denaturation (94°C, 45 s), primer annealing (61°C, 45 s) and chain extension (72°C, 30 s). Prior to cycling, samples were heated at 94°C for 2 min. For all experiments, a Perkin Elmer GeneAmp System 9700 PCR thermocycler was used. Amplified products were detected by electrophoresis in 1.5% agarose in 0.5×Tris–borate–EDTA buffer at 100 V for 40 min. Gels were stained with ethidium bromide. An UV transilluminator (Digi-Doc system, Bio-Rad) with an analyst computer program (Quantity One 4.1 software, Bio-Rad) was used for visualization.

2.4 Analysis of the specificity and sensitivity of the m-PCR

Bacteria used for examination of primer specificity were grown as described above. Cells were subsequently diluted in sterile saline to approximately 106 cfu ml−1. From each cell suspension, 100 μl was transferred to Eppendorf tubes and boiled for 10 min to lyse the bacteria before examination of 2 μl lysate in the m-PCR. The specificity was evaluated by including the nearest phylogenetic neighbors (Campylobacter, Helicobacter, Bacteroides ureolyticus and Wolinella species) and other bacteria commonly present in poultry (Salmonella and Escherichia species) in the m-PCR.

The sensitivity of the m-PCR was determined in two different enrichment broths, described for the recovery of Arcobacter species in poultry samples. The Arcobacter enrichment broth (ASB) developed by de Boer et al. [22], consists of Brucella broth supplemented with 5% lysed horse blood and with piperacillin, cefoperazone, trimethoprim and cycloheximide as selective supplements. The medium described by Atabay et al. [32], consists of Arcobacter broth (Oxoid) with CAT (cefoperazone, amphotericin and teicoplanin) selective supplement. Arcobacter strains used for determining the sensitivity were grown for 48 h as described above and appropriate bacterial concentrations were prepared by serial 10-fold dilutions in sterile saline. To enumerate the bacteria, 100-μl aliquots were spread onto four blood agar plates and incubated at 30°C for 48 h under microaerobic conditions. Estimation of the bacterial concentration was done by calculating the average colony count on the four plates. At the same time, a panel of mixtures of two or three Arcobacter species at several concentrations was prepared.

Samples of 10 g of poultry skin or poultry meat were homogenized in a sterile stomacher bag with 90 ml of the two enrichment broths each for 1 min and seeded with pure cultures to obtain bacterial counts of 10, 102, 103, 104 and 105 cfu g−1 chicken sample and with the mixtures of the two and three Arcobacter species. Non-inoculated controls were included in each experiment. From each sample, two aliquots of 1 ml were transferred to Eppendorf tubes. One of the two spiked samples was sedimented passively at 4°C for 30 min to remove coarse particles and solidified fat. Then, almost 1 ml supernatant fluid was transferred to a new Eppendorf tube. Samples were centrifuged in an Eppendorf centrifuge 5417C at 13 000 rpm for 15 min. The supernatant was discarded and the bacterial pellet was washed three times with 500 μl sterile water. Finally, the bacterial pellet was resuspended in 100 μl of sterile water and boiled for 10 min to lyse the cells. From each sample, 2 μl were used for m-PCR examination.

2.5 m-PCR detection of Arcobacter species in poultry samples

The m-PCR assay was applied to two enrichment broths to determine the performance of the test. For this purpose, 37 neck skin samples and 14 poultry meat samples were collected from seven different abattoirs, each sample representing a different poultry flock. They were obtained directly after chilling, transported to the laboratory under cooled conditions, and examined within 6 h after sampling. Each sample was examined by the methods described by de Boer et al. [22] and Atabay et al. [32]. The m-PCR was applied directly after homogenization of the samples in enrichment broth and on the enrichment broths after 24 h and 48 h of incubation. DNA was extracted from 1 ml of the broths as described above. After 48 h of incubation, 40 μl ASB enrichment broth was pipetted on the center of the surface of a semisolid Arcobacter selective medium (ASM) plate. The ASM plates was prepared by suspending 21 g of Mueller Hinton broth and 2.5 agar no.3 in 960 ml of distilled water. After sterilization and cooling, the same antimicrobial agents as in ASB were added. The plates was incubated for 48–72 h at 24°C. After incubation, ASM plates were examined for the presence of motility zones. For confirmation, some material from the outer edge of the motility zone was streaked on blood agar plates to obtain single colonies and incubated for another 24–72 h at 30°C. At the same time, 100 μl Arcobacter broth enrichment medium was pipetted on a 0.65-μm filter on a blood agar plate. The plate was incubated at 37° under aerobic conditions. After 1 h, the filter was removed and the fluid was spread across the surface of the plate. The plate was further incubated for 48–72 h at 30°C.

The m-PCR assay was also performed on bacterial colonies harvested from the blood agar plates. One colony was suspended in 100 μl distilled water and centrifuged at 13 000 rpm for 15 min. The pellet was resuspended in 50 μl sterile water, boiled for 10 min and 2 μl was added to the m-PCR reaction mixture.

3 Results and discussion

3.1 Specificity of the m-PCR primers

A panel of Arcobacter reference strains (Table 1) and related bacteria (Table 3) was tested in the m-PCR assay using five primers. The ARCO-BUTZ primer pair amplified a 401-bp fragment and was shown to be specific for all the tested A. butzleri strains. With ARCO-SKIR primers, a specific 641-bp fragment was noted for all the tested A. skirrowii strains. For A. cryaerophilus, primer set CRY1 and CRY 2 was used. A. cryaerophilus is known as a genotypically heterogeneous species. Two subgroups were identified within this species based on DNA hybridization [14,33]. Primer combination CRY1 and CRY2 amplified a specific 257-bp fragment for all the tested A. cryaerophilus subgroup 1 and 2 strains. For A. nitrofigilis, Campylobacter, Helicobacter, Wolinella and the other test organisms, no product was amplified.

3.2 Sensitivity of m-PCR with spiked poultry samples

The sensitivity of the m-PCR assay for the detection of Arcobacter species in the presence of chicken skin and chicken meat in two different enrichment broths was studied. When Arcobacter broth with CAT supplement was used, the detection level for A. butzleri and A. cryaerophilus was 103 cfu g−1 of chicken sample. A detection level of 102 cfu g−1 of chicken sample was achieved for A. skirrowii (Fig. 1A). There was no difference in the recovery rate between skin and meat samples. Increasing the number of amplification cycles, as well as the implementation of the commonly used final elongation step (72°C for 7 min) after final extension, did not improve the detection levels for any of the three species. When two or three Arcobacter species were spiked, the detection level for the individual species was not affected (Fig. 1B).


Sensitivity of the m-PCR with spiked poultry samples. A: lane M: 100-bp ladder; lanes 1–5, PCR products amplified from 105 to 10 cfu of A. butzleri strain 10828; lanes 6–10, PCR products amplified from 105 to 10 cfu of A. cryaerophilus strain 7536; lanes 11–15, PCR products amplified from 105 to 10 cfu of A. skirrowii strain 6621. B: lane lane M: 100-bp ladder; lane 1; PCR products amplified from 105 of A. butlzeri and 103 of A. cryaerophilus; lane 2, PCR products amplified from 105 of A. butlzeri and 102 of A. skirrowii; lane 3, PCR products amplified from 103 of A. butlzeri and 105 of A. cryaerophilus; lane 4, PCR products amplified from 105 of A. cryaerophilus and 102 of A. skirrowii; lane 5, PCR products amplified from 105 of A. skirrowii and 103 of A. butzleri; lane 6, PCR products amplified from 105 of A. skirrowii and 103 of A. cryaerophilus; lane 7, PCR products amplified from 103 of A. butzleri, 103 of A. cryaerophilus and 102 of A. skirrowii (strains used were the same as in A).

In spite of the presence of hemoglobin in the enrichment medium of de Boer, there was no inhibition of the PCR reaction and the same detection levels for the three species were achieved as for Arcobacter broth. Use of several washing steps prior to lysing the bacterial cells, may have diluted the hemoglobin concentration to less than the inhibitory level. Direct centrifugation and DNA extraction of 1 ml of the enrichment broths without the removal of coarse particles and solidified fat however, resulted in a substantial inhibition of the m-PCR assay (data not shown).

3.3 m-PCR detection of Arcobacter in poultry samples

The m-PCR assay was applied to detect Arcobacter species in 37 neck skin samples and 14 poultry meat samples. In only five neck skin samples was A. butzleri detected by m-PCR directly after homogenization in both the enrichment broths of the same samples (results not shown).

After enrichment for 24 h in ASB, A. butzleri was detected in 14 poultry samples. However after enrichment for 24 h in Arcobacter broth, 36 samples were positive for A. butzleri, three of which were positive for A. cryaerophilus as well. A. skirrowii was not detected (Table 4). Further incubation for another 24 h and examination of the enrichment broths, only confirmed the results already found and did not reveal new positive samples (Table 4).

View this table:

Number of positive poultry samples examined by two isolation protocols and m-PCR

After 48 h of enrichment in microaerobic conditions, isolation of Arcobacter from the enrichment broths was performed as described by de Boer et al. [22] and Atabay et al. [32]. All single colonies, with a maximum of 15, were harvested from blood agar plates and DNA was extracted as described above. Almost full agreement between m-PCR results from enrichment broth and from single colonies was obtained for all samples (Table 4). From ASM plates, some material from the outer edge of the motility zones were streaked on blood agar plates. The recognition of typical colonies on the blood agar plates was easy, and the accompanying flora was well suppressed. When Arcobacter broth with CAT supplement was used, recognition of the colonies on blood agar plates was often hindered due to growth of other types of bacteria.

The difference in recovery rate of Arcobacter species in poultry between the method of de Boer and Arcobacter broth with CAT supplement is comparable with published data. A. butzleri was detected in 24% (n=224) of retail poultry samples in The Netherlands with the ASB–ASM method [22]. In contrast, Atabay et al. reported the presence of arcobacters in broilers at a much higher contamination level [7,23,32]. Although A. butzleri was the predominant species present, isolation of A. cryaerophilus and A. skirrowii was also possible. With enrichment in ASB and isolation on ASM, no other species than A. butzleri was isolated. When Arcobacter broth with CAT supplement was used, A. cryaerophilus was detected by m-PCR in three neck skin samples and two were confirmed by isolation.

The time for isolation by either method required at least 96 h and further biochemical identification extended the examination time. However, the application of the developed m-PCR after 24 h incubation in Arcobacter broth with CAT-supplement, species-specific detection of Arcobacter was possible in the examined poultry products. Gonzalez et al. applied a genus-specific PCR assay to detect Arcobacter in poultry meat [29]. After 16 h of incubation, an Arcobacter-specific amplicon was generated. Results of the complete assay where available within 2 working days.

In conclusion, the species-specific m-PCR assay described in this paper is a rapid method for the detection of Arcobacter species in poultry products by the m-PCR assay when combined with a 24-h enrichment in Arcobacter broth with CAT supplement. Subsequently, the m-PCR can be used to identify isolates from agar plates, and is a good alternative to biochemical identification.


  1. [1].
  2. [2].
  3. [3].
  4. [4].
  5. [5].
  6. [6].
  7. [7].
  8. [8].
  9. [9].
  10. [10].
  11. [11].
  12. [12].
  13. [13].
  14. [14].
  15. [15].
  16. [16].
  17. [17].
  18. [18].
  19. [19].
  20. [20].
  21. [21].
  22. [22].
  23. [23].
  24. [24].
  25. [25].
  26. [26].
  27. [27].
  28. [28].
  29. [29].
  30. [30].
  31. [31].
  32. [32].
  33. [33].
View Abstract