Emergence of Raoultella ornithinolytica isolated from chicken products in Alexandria, Egypt

Background and Aim: Raoultella ornithinolytica is one of the emerging gram-negative bacteria, which associated with foodborne illness. Researches affirmed that distinguish between R. ornithinolytica and Klebsiella oxytoca are difficult, as they are phylogenetic related. The evolution of multidrug resistance of Raoultella strains gained more concern for recognition of the pathogen which supports in controlling the disease and minify its threat. This study sought to find a reliable tool for the identification of Raoultella ornithinolytica, isolated from chicken product samples, and assessed the resistance profile of R. ornithinolytica using antibiogram sensitivity tests. Materials and Methods: Forty samples of chicken products were collected between January and September 2019 from different markets in Alexandria Governorate, Egypt. The products included nuggets, strips, burgers, luncheon meats, pane, frankfurters, and minced chicken meat. The samples were transferred to the Reference Laboratory. The samples were subjected to isolation, biochemical reaction testing, phenotypic system analytical profile index (API) E20, and a detection of antimicrobial susceptibility test. Phenotypic identification was confirmed through matrix-assisted laser-desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS). Results: Thirty-three bacterial isolates (82.50%) out of 40 samples were isolated into pure cultures from the chicken samples. Three isolates (9.09%) were positive for R. ornithinolytica, while 30 isolates (90.91%) exhibited growth characters for different pathogens (Escherichia coli, Enterobacter aerogenes, Proteus vulgaris, R. ornithinolytica, and Klebsiella pneumoniae). The isolates of R. ornithinolytica were resistant to five types of antibiotics and sensitive to two types of antibiotics. Conclusion: This study reported the first case of R. ornithinolytica found in chicken products in Egypt. Phenotypic system API 20E and MALDI-TOF MS were found to be reliable tools for confirming the diagnosis of R. ornithinolytica. As it provides rapid identification with high sensitivity and specificity for R. ornithinolytica, which often do not require a molecular procedure for confirmation.


Introduction
Raoultella ornithinolytica is recognized as histamine-producing Gram-negative bacteria, which usually inhabits aquatic environments such as saturated soil, water systems, and fish [1][2][3][4][5][6]. These bacteria belong to Enterobacteriaceae family, with distinctive biochemical characteristics that can assist with their discrimination from other phenotypically related species [2,[7][8][9][10][11]. It has low nutritional demands and has the ability to survive when food is scarce. Moreover, the growth temperature required is remarkably variable from 4°C to 40°C [1]. In the past, R. ornithinolytica was classified as a member of Klebsiella genus and reclassified in 2001 as a new genus Raoultella, based on the sequencing of the 16S rRNA and the rpoB gene [1,12].
The use of a standard classical approach is considered a useful tool in the identification of R. ornithinolytica [13,14]. However, matrix-assisted laser-desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) and sequence analysis are more highly regarded as they offer the potential for extraordinary insight into pathogens [15].
This study sought to find a reliable tool for the identification of R. ornithinolytica, isolated from chicken product samples, and then assessed the resistance profile of R. ornithinolytica using antibiogram sensitivity tests.

Ethical approval
No ethical approval was needed to perform this study. However, the samples were treated according to the national and international criteria.  Available at www.veterinaryworld.org/Vol.13/July-2020/30.pdf

Study duration, location, and data collection
This study took place from January 2019 to September 2019. Forty samples of chicken products were collected from different markets located in or near the center of Alexandria city. The products included nuggets, strips, burgers, luncheon meats, pane, frankfurters, and minced chicken meat. Samples were transported on ice to the Reference Laboratory for Quality Control on Poultry Production, Animal Health Research Institute, Dokki, Giza, Egypt. These samples were sealed in sterile bags and stored at −86°C until testing.

Bacteriological examination of R. ornithinolytica
The samples were pretreated according to a method described previously [16]; briefly, samples were thawed at room temperature, macerated into small pieces using sterile blades, and homogenized using a sterile mortar and pestle. The samples were incubated under aerobic conditions using pre-enrichment (buffered peptone water) at 37°C for 24 h (HiMedia ® , India). Twenty-five grams of the chicken product sample were pre-enriched into 225 ml buffered peptone water. Then, 0.1 ml and 1 ml of pre-enriched aliquots were transferred into 10 ml Rappaport and Vassiliadis broth for the enrichment and then incubated at 42°C for 24 h (HiMedia ® , India). The enriched aliquot samples were seeded onto MacConkey and Xylose Lysine Deoxycholate agar (XLD) and incubated at 37°C for 24-48 h to observe colonial growth (HiMedia ® , India) [17].

Recommended biochemical tests panel for the identification of R. ornithinolytica
The four biochemical tests utilized are illustrated in Table-1 to provide an example of the test variability that occurs using the standard laboratory methods for each test since complete standardization of these biochemical methods has not yet been fully elucidated. Thus, some reports may not provide sufficient data [7,8,16].

Identification using a phenotypic system analytical profile index (API) 20E strips
Phenotypic system API 20E strips are considered a well-established method for accurate identification, this method depends on the standardized extensive databases [18], the selected isolates and reference strains were then evaluated using a (identification of products) testing kit API 20E (bioMérieux) to minimize the misidentified results of R. ornithinolytica strains with Klebsiella oxytoca using a conventional laboratory technique. Preparation procedures occurred based on the manufacturer's recommendation protocol, and the results were analyzed using the API 20E WEB™ service.

Identification of R. ornithinolytica by MALDI-TOF MS fingerprinting
MALDI-TOF MS fingerprinting (Brüker Daltonik, GmbH, Bremen, Germany; Biotyper 3.0 database) was performed according to the following steps. The samples were prepared based on the manufacturer's guidelines for the identification of Gram-negative bacteria using a formic acid method; the fresh isolates were inoculated into spots on the target plate. Then, 1 μl of 70% formic acid was placed in the microbial spot and left to dry at room temperature. Thereafter, 1 μl of MALDI matrix solution was added (saturated α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 2.5% trifluoroacetic acid) [19]. Data analysis of MALDI-TOF MS required a fingerprint comparison with a database of reference spectra according to the manufacturer's recommendations through the use of various algorithms.
The antibiotic susceptibilities of R. ornithinolytica isolates were also determined and interpreted in accordance with CLSI, 2018. Briefly, the samples were subjected to 15 distinct antibiotics (FLX, NA, CIP, OTC, TE, AMK, S, ERY, BAC, CS, C, AMX,  (Table-2). Antibiotic sensitivity disks were applied to each prepared plate and the plates were incubated at 37°C for 24 h. Then, the zones of inhibition were measured.
Table-1 and Figure-2 refer to the results of the individual biochemical reactions (oxidase, Triple Sugar Iron Agar (TSI), Lysine Iron Agar (LIA), and urease) for the expected positive R. ornithinolytica isolates, which were then identified as R. ornithinolytica by use of an API 20E test. The phenotypic system API 20E confirmed that the three pre-identified samples had proved to be R. ornithinolytica based on their positive results for ornithine decarboxylase, ODC+, which is the primary identification test that distinguishes between R. ornithinolytica and K. oxytoca (Figure-3). Moreover, the three ODC+ samples were subjected to MALDI-TOF MS fingerprinting as a confirmation method for the results obtained by API 20E. MALDI-TOF MS verified the diagnosis of R. ornithinolytica. The obtained spectra were compared with a Biotyper database, which revealed that all three samples that were pre-identified using conventional techniques were 100% correctly identified as R. ornithinolytica, with a score value < 2.00. These results demonstrated that the most effective methods used for the identification of R. ornithinolytica were the phenotypic system API 20E combined with MALDI-TOF MS.
The three R. ornithinolytica samples were subjected to antibiotic susceptibility tests to evaluate their resistance profiles using antibiotic disks (Oxoid,  Available at www.veterinaryworld.org/Vol.13/July-2020/30.pdf U.K, depicted in Figure-4) [20,21]. The samples were exposed to 15 discrete antibiotics (Table-2), the first sample demonstrated resistance to five types of antibiotic disks (FLX, ERY, P, AMC, and BAC), intermediate sensitivity was found to two antibiotics (AMX and AMP), and sensitivity to eight antibiotics (C, CS, NA, CIP, TE, OTC, AMK, and S). The second sample demonstrated resistance to 10 antibiotics (FLX, AMP, ERY, P, TE, OTC, AMC, AMX, S, and BAC). Intermediate sensitivity was found to only one antibiotic (NA) and sensitive to four antibiotics (C, CIP, CS, and AMK). The third sample demonstrated resistance to 10 antibiotics (FLX, AMP, ERY, P, tetracycline, OTC, AMC, AMX, S, and BAC). Intermediate sensitivity was found for two antibiotics (CIP and AMK) and sensitivity to three antibiotics (C, CS, and NA).

Discussion
In the past few decades, serious infections have occurred through emerging diseases. R. ornithinolytica infections were once rare in humans; however, recently, this pathogen is emerging and is associated with foodborne illness [22], toxicity, septicemia, bacteremia, enteric fever, scombroid poisoning, and immune deficiency [15,[23][24][25][26][27][28][29][30]. Furthermore, R. ornithinolytica is commonly misidentified in clinical microbiology laboratories, as this bacterium is quite similar to Klebsiella species [31]. Therefore, the correct identification of R. ornithinolytica is important for treating patients and improving the classification and characterization of this bacterial species. In the current study, three strains of R. ornithinolytica were isolated from chicken products ( Figure-1), subjected to four manual laboratory biochemical tests, which included oxidase, TSI, LIA, and urease (Table-1). R. ornithinolytica was identified through the following morphological characters, biochemical reaction, growth temperature, and pigment production [7,8,16]. The first preliminary identification in this study was confused with Klebsiella oxytoca. Subsequent testing using the phenotype system API 20E (bioMérieux), identified these samples as R. ornithinolytica. API 20E (bioMérieux) is considered a discriminatory method which uses ornithine decarboxylase (ODC) tests to differentiate between R. ornithinolytica positive (ODC+) and K. oxytoca negative (ODC−) [18]. The phenotype recognition results were confirmed through an emerging microbial diagnostic technology (MALDI-TOF MS). Then, the samples were subjected to disk diffusion to determine their antibiotic profiles. However, there are many variables that could influence the outcome of the data interpreted through disk susceptibility determinations, such as similarity in minimum inhibitory concentration for the antibiotic disk accompanied by great variation in the sensitivity result within the same species. A possible explanation for this conflict might be attributed to the fact that the resistance of R. ornithinolytica to different types of antibiotics was not increased sufficiently or might be due to a higher microbial load as reported in our study, there were some differences found in the susceptibility ratio between the parents of isolated strains varying from 1:4. Furthermore, the presence of more than 1 isolate from the same sample, taken at different times, showed variability in the susceptibility profile overtime [32].
The results of this study were consistent with that of other recent studies denoting the efficacy and efficiency of MALDI-TOF MS for the identification of Gram-negative bacteria [13,33,34]. Moreover, this study indicated that a replacement for the traditional identification method for R. ornithinolytica isolation in the microbiology laboratory was important. Since this technology was simple, fast, and reliable, API 20E and MALDI-TOF MS should be considered for further studies. However, improvements would be required in sample preparation and the availability of databases specifically designed to identify significant strains.

Conclusion
This is the first study in Egypt that illustrates the isolation of the extremely rare pathogen R. ornithinolytica from the chicken products which demonstrated an excellent prognosis with antibiotic susceptibility, as reported in the literature.

Authors' Contributions
SME created the research and experimental design, performed the laboratory experiment, data analysis, and wrote the manuscript. AAAE supervised the experiment, checked the data analysis, and revised the manuscript. WMMH supervised the experiment, helped in the laboratory work, and revised the manuscript. All authors contributed to the drafting and revision of the manuscript. All authors read and approved the final manuscript.