New Microbiologica, 40, 4, 269-278, 2017, ISN 1121-7138

Application of MALDI-TOF MS fingerprinting as a quick tool for identification and clustering of foodborne pathogens isolated from food productsAyman Elbehiry1,3, Eman Marzouk2, Mohamed Hamada4, Musaad Al-Dubaib5, Essam Alyamani6, Ihab M. Moussa7, Anhar AlRowaidhan2, Hassan A. Hemeg8

1Department of Bacteriology, Mycology and Immunology, Faculty of Veterinary Medicine, Sadat City University, Egypt; 2Department of Medical laboratories, College of Applied Medical Science, Qassim University, Kingdom of Saudi Arabia; 3Department of Public Health, College of Public Health and Health Informatics, Qassim University, Kingdom of Saudi Arabia; 4Department of Food Hygiene & Control, Faculty of Veterinary Medicine, Sadat City University, Egypt; 5Department of Veterinary Medicine, College of Agriculture and Veterinary Medicine, Qassim University, Kingdom of Saudi Arabia; 6National Centre for Biotechnology, KACST, Kingdom of Saudi Arabia; 7Department of Botany and Microbiology, College of Science, King Saud University, Saudi Arabia; 8Department of Medical technology/ Microbiology, College of Applied Medical Science, Taibah University, Madina, Saudi Arabia

INTRODUCTION

The spoilage of food products with pathogenic bacteria and fungi represents a threat to public health worldwide (McAuley et al., 2014). Foods can be contaminated by nu-merous types of pathogenic microorganisms, such as bac-teria, molds, yeasts and parasites (Böhme et al., 2013). The most common foodborne pathogens responsible for the majority of foodborne illnesses are Campylobacter jejuni, Escherichia coli O157:H7, Staphylococcus aureus, Listeria monocytogenes, Bacillus cereus, Salmonella enterica, Vibrio spp., Shiga toxin-producing Escherichia coli (STEC) and Clostridium perfringens (Scallan et al., 2011; Zhao et al., 2014). The higher quantities of street foods and ready-to-

Corresponding author:Ayman ElbehiryE-mail: aymanella2007@yahoo.com or ar.elbehiry@qu.edu.sa

©2017 by EDIMES - Edizioni Internazionali Srl. All rights reserved

eat foodstuffs represent an alarm to public health orga-nizations for assurance of food safety (Lee et al., 2014). Therefore, it is very important to examine food products for the existence of foodborne pathogens to confirm a safe food supply and to diminish the incidence of foodborne illnesses (Law et al., 2014). Although phenotypic, immu-nologic and genotypic are common methods for identifica-tion of foodborne pathogens, they can be laborious, with a high percentage of incorrect identification, time-con-suming and difficult to implement in routine food analysis (Kramer et al., 2009; Wenning et al., 2014). Several studies have been carried out in the last decade to decrease the time and the amount of manual labor using alternative techniques for accurate identification of foodborne patho-gens (Jasson et al., 2010; Wenning et al., 2014). Recently, matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS) has been consid-ered as an excellent tool in different research laborato-ries for detection and discrimination of various types of microorganisms like bacteria and fungi. (Liu et al., 2007; Welker et al., 2011; Wieser et al., 2012; Böhme et al., 2016; Elbehiry et al., 2016).

Key words:Identification, Typing, Foodborne pathogens, MALDI-TOF MS fingerprinting.

SUMMARY

Foodborne pathogens can be associated with a wide variety of food products and it is very important to identify them to supply safe food and prevent foodborne infections. Since traditional techniques are time-consuming and laborious, this study was designed for rapid identification and clustering of foodborne pathogens isolated from various restaurants in Al-Qassim region, Kingdom of Saudi Arabia (KSA) using matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS). Sixty-nine bacterial and thirty-two fungal isolates isolated from 80 food samples were used in this study. Preliminary identification was carried out through culture and BD Phoenix™ methods. A confirmatory identification technique was then performed using MALDI-TOF MS. The BD Phoenix results revealed that 97% (67/69 isolates) of bacteria were correctly identified as 75% Enterobacter cloacae, 95.45% Campylobacter jejuni and 100% for Escherichia coli, Salmonella enterica, Staphylococcus aureus, Acinetobacter baumannii, and Klebsiella pneumoniae. While 94.44% (29/32 isolates) of fungi were correctly identified as 77.77% Alternaria alternate, 88.88% Aspergillus niger and 100% for Aspergillus flavus, Penicillium digitatum, Candida albicans and Debaryomyces hansenii. However, all bacterial and fungal isolates were 100% properly identified by MALDI-TOF MS fingerprinting with a score value ≥2.00. A gel view illustrated that the spectral peaks for the identified isolates fluctuate between 3,000 and 10,000 Da. The results of main spectra library (MSP) dendrogram showed that the bacterial and fungal isolates matched with 19 and 9 reference strains stored in the Bruker taxonomy, respectively. Our results indicated that MALDI-TOF MS is a promising technique for fast and accurate identification of foodborne pathogens.

Received March 13, 2017 Accepted June 26, 2017

FUll PaPer

A. Elbehiry, E.Marzouk, M. Hamada, et al.270

The fast and consistent detection of foodborne patho-gens is highly important to confirm the safety and quality of food products. Nevertheless, the use of MALDI-TOF MS to identify foodborne pathogens has seldom been reported to date (Böhme et al., 2016). In recent years, the Food and Drug Administration (FDA) confirmed that MALDI-TOF MS is an approved technique for identifi-cation and discrimination of various pathogens through ionization of the extracted molecules of entire cell cul-ture (Cheng et al., 2016). MALDI-TOF MS has more ad-vantages than the other microbiological methods as it is accurate, less expensive, faster and needs less technical skill. Therefore, MALDI-TOF MS represents a powerful technique in food microbiology laboratories, where the time factor represents an important issue (Nicolaou et al., 2012; Fournier et al., 2013; Pavlovic et al., 2013; Ca-rannante et al., 2015). Several foodborne bacterial spe-cies were identified using MALDI-TOF MS such as Acine-tobacter baumanii (Alvarez-Buylla et al., 2012; Sousa et al., 2014), Campylobacter species (Khot and Fisher, 2013) and Escherichia coli (Christner et al., 2014; Khot and Fisher, 2013; Novais et al., 2014). Furthermore, previous studies used MALDI-TOF MS for fast identification of various types of bacteria like Salmonella enterica (Dieck-mann et al., 2011; Sparbier et al., 2012), Staphylococcus aureus (Josten et al., 2013) and Acinetobacter baumannii (Marí-Almirall et al., 2016).Since filamentous fungi are ubiquitous in nature with an unexpected capability to decompose plants, they cause spoilage of various food products. Therefore, rapid and accurate detection of filamentous fungi in food has re-cently aroused much recent (Lima and Santos, 2017). Ap-plication of MALDI-TOF MS has moreover developed to identify different species of fungi, such as yeasts (Sendid et al., 2013; Becker et al., 2014). In addition, a few genera of filamentous fungi were correctly identified as Asper-gillus species (Alanio et al., 2011; DeCarolis et al., 2012); Penicillium species (Del Chierico et al., 2012) and Fusar-ium species (DeCarolis et al., 2012; Del Chierico et al., 2012) which are considered the most commonly encoun-tered foodborne fungal pathogens. From the previously mentioned data, our study was designed for fast identifi-cation and clustering of certain types of bacteria and fun-gi isolated from various food samples in the Al-Qassim region, KSA, using MALDI-TOF MS fingerprinting, a po-tent, cost-effective, fast, and robust method in food mi-crobiology laboratories, where the time factor represents an important issue.

MATERIALS AND METHODS

Sample collection and preparationEighty food samples were collected from various restau-rants in Al-Qassim region, KSA during the summer season in 2015. The food samples used in the current study are listed in Table 1. Firstly, all samples were collected sep-arately from the restaurants in sterile plastic bags and transported to the Laboratory of Microbiology, College of Public Health, Qassim University for microbiologi-cal analysis. Secondly, from each sample, 50 grams were aseptically homogenized and 450 ml of phosphate buff-ered saline (PBS) was then added. A sterile dilution was obtained using sterile distilled water. From each dilution, 1 ml was plated according to the method described previ-ously (Swanson et al., 1992).

Isolation of foodborne pathogens All food samples were cultured on selective media (Table 2) and then the media were cultivated at 37oC. The culture media used in the present study for isolation and prelim-inary identification of bacteria in the food samples were CHROMagar Acinetobacter®, Methylumbeliferyl β-D-glu-curonide (MUG) Nutrient Agar, Xylose Lysine Deoxycho-late (XLD) Agar, modified Charcoal-Cefoperazone-De-oxycholate Agar (mCCDA), Baird-Parker Agar (BPA),

Table 1 - Eighty food samples isolated randomly from var-ious restaurants.

Type of food sample No. of samples

Shawrmas with salads 13

Matazeez (A Middle Eastern hearty stew of lamb and vegetables with whole wheat)

12

Burger with salads 8

Jareesh (boiled, cracked, or coarsely-ground  wheat mixed with meat)

8

Kabsa (rice, meat, vegetables, and a mixture of spices)

8

French fried potatos 8

Soup (meat soap) 7

Humas (chick pea dip) 6

Tortilla (soft, thin flatbread prepared from finely ground wheat flour)

5

Mozzarella 5

Total 80

Table 2 - Different selective media used for isolation of food-borne pathogens.

Microorganism Medium

Bacteria

Escherichia coli Methylumbeliferyl β-D-glucuronide (MUG) Nutrient Agar

Salmonella species Xylose Lysine Deoxycholate (XLD) Agar

Campylobacter species Modified charcoal-cefoperazone-deoxycholate agar (MCCDA)

Staphylococcus aureus Baird-Parker Agar

Acinetobacter baumannii CHROMagar Acinetobacter®

Enterobacter cloacae HiCrome™ Cronobacter spp. Agar

Klebsiella pneumoniae m-Fecal Coliform Agar (M-FC Agar)

Fungi

Alternaria alternata Czapek-Dox Iprodione Dichloran Agar (CZID)

Aspergillus niger Sabouraud Dextrose Agar, Potato Dextrose agar

Aspergillus flavus Aspergillus Differentiation Agar (AFPA)

Penicillium digitatum Potato Dextrose agar (PDA)

Candida albicans Eosin Methylene Blue Agar (Levine)

Debaryomyces hansenii 5-bromo-4-chloro-3indolyl-D-galactopyranoside (Gal agar)

Application of MALDI-TOF MS fingerprinting as a quick tool for identification and clustering of foodborne pathogens isolated from food products 271

Figure 1 - The strategy of MALDI-TOF-MS used for de-tection and clustering of dif-ferent pathogens isolated from food products (modified from Angelakis et al., 2014).

Mannitol Salt Agar (MSA), HiCrome™ Cronobacter spp. Agar and Modified-Fecal Coliform Agar (M-FCA). While the media used for isolation of fungi were Czapek-Dox Ip-rodione Dichloran Agar (CZID), Sabouraud Dextrose Agar (SDA), Potato Dextrose Agar (PDA), Aspergillus Differen-tiation Agar (AFPA), Eosin Methylene Blue Agar (Levine) and5-bromo-4-chloro-3indolyl-D-galactopyranoside (Gal Agar). All media used in the present study were purchased from Sigma-Aldrich, USA.

Identification by Becton Dickinson (BD) PhoenixAll isolates of bacteria and fungi (yeasts and molds) were identified by the Phoenix system (Becton Dickinson Diag-nostics, USA) using the BD Phoenix ID panels according to the manufacturer’s guidelines (Posteraro et al., 2013).

Identification by MALDI-TOF MS fingerprintingSample preparation for bacteriaWe applied MALDI biotyper device (Bruker Daltonics, Bre-men, Germany) for fast and accurate identification of bacte-rial colonies isolated from different food samples. As shown in Figure 1, the strategy used for mass spectrometry iden-tification of food samples was based on the spectra score value. When discordant results were obtained (value <2), 16S rRNA gene sequencing most often confirmed the mass spectrometry identification (Angelakis et al., 2014). The pro-cedure of formic acid extraction was achieved according to the manufacturer’s guidelines for the identification of bac-terial and fungal isolates. Briefly, after overnight culture, a fresh colony incubated for 18-24 h at 37°C was inoculated onto two spots of target plate and then enclosed with one µl of matrix solution (saturated α-cyano-4-hydroxycinnamic acid in 50% acetonitrile and 2.5% trifluoroacetic acid). The spectra were directly produced by applying new software, namely, Compass Satellite software and the identification was conducted with a Microflex LT device. 

Bacterial test standard (BTS) preparation (positive control) Fifty µl of the standard solvent solution were inoculated onto the in vitro diagnostic product (IVD) BTS pellet and

dissolved thoroughly at 25°C. The IVD BTS solution was then centrifuged at 13,000 rpm for two minutes (Centri-fuge 5430, Eppendorf, Germany). Finally, five µl of the supernatant were transferred into screw cap, microtubes and stored at -18°C for further investigation.

Preparation of samples for filamentous fungiAs different types of fungi have a strong cell wall, the protocol for extraction of protein was a very important step for the intracellular detection of fungal proteins. Fil-amentous fungi samples were prepared according to the technique recommended by Bruker Daltonics. In brief, the inoculated tubes with a sufficient amount of biologi-cal material were mixed by Rotator SB2 (Bibby Scientif-ic Limited, UK). The cultivated tubes were then removed from the rotator and placed on a bench for 10 minutes to settle the filamentous fungi sediment at the bottoms of the tubes; 1.5 ml from the filamentous fungi sediment were then harvested and transferred into an Eppendorf tube. Centrifugation was carried out for 2 min at 13,000 rpm and then a distinct pellet was obtained. After discarding the supernatant carefully, 1 ml of HPLC-grade water (Sig-ma-Aldrich, USA) was added and centrifugation for 2 min was carried out. Three-hundred µl of HPLC-grade water and 900 µl of absolute ethanol were then added and mixed thoroughly. The supernatant was discarded carefully and centrifugation was done again for a few seconds. The re-maining ethanol was then removed entirely. The pellet was left at 37°C for 5-10 minutes and according to the pellet size, 70% formic acid was added (very small pellets re-quired 10 µL to 20 µl, while a big pellet needed 100 µl of 70% formic acid). The pellets were suspended thoroughly and then the same volume of acetonitrile was added to the tube and melted carefully. The tube was then centrifuged at 13,000 rpm. Finally, 1 µl of the supernatant was relocat-ed onto a MALDI target plate and then mixed with 1 µl of the matrix HCCA solution.

Data analysis and clustering According to the instructions of Bruker Daltonics, the score value of unidentified spectrum in the range from zero to

A. Elbehiry, E.Marzouk, M. Hamada, et al.272

three was determined by matching the unidentified spec-trum with the stored spectrum in the Bruker database. The accurate identification of the field isolates is carried out when the score value ranges from 2.30 to 3.00. Neverthe-less, species and genus levels are detected when the score value ranges from 2.00 to 2.29 and 1.700 to 1.999, respec-tively. In contrast, the identification is not reliable when the score value ranges from 0.00 to 1.69. The spectra created by compass software were measured in a m/z range between 3000 and 20000 Daltons (Da). According to the library of MALDI biotyper, which contains 5989 bacterial and fungal sub-species, a dendrogram was generated from the mini-mal spanning tree (MSP) data set. The MSP dendrogram is based on the evaluation of the main spectra of investigating the various species. Primarily, the main spectra of the MAL-DI biotyper taxonomy were compared with the spectra re-sulting in a matrix of cross-wise identification scores. This matrix was applied to estimate the distance level for each pair of main spectra. A dendrogram was created according to the distance level between species.

RESULTS

As shown in Table 3, 69 bacterial isolates were isolated from pure cultures of food samples. The isolated bacte-ria were named as follow: 31.88% (22/69 isolates) Cam-pylobacter jejuni, 24.63% (17/69 isolates) Escherichia coli, 13.04% (9/69 isolates) Salmonella enteritidis, 11.59% (8/69 isolates) Staphylococcus aureus, 8.69% (6/69 isolates) Acinetobacter baumannii, 5.79% (4/69 isolates) Entero-bacter cloacae and 4.34% (3/69 isolates) Klebsiella pneu-monia. While 32 isolates of fungi (27 molds and 5 yeasts) were isolated from all food samples as follows: 28.12% (9/32 isolates) Alternaria alternate, 25% (8/32 isolates) As-pergillus niger, 18.75% (6/32 isolates) Aspergillus flavus

and 12.5% (4/32 isolates) Penicillium digitatum were iso-lated as molds. Moreover, 9.37% (3/32 isolates) Candida albicans and 6.25% (2/32 isolates) Deboryomyces hansenii were isolated as yeasts.

BD Phoenix identification of foodborne pathogens One hundred and one foodborne pathogens (69 bacterial and 32 fungal isolates) were investigated by the BD Phoe-nix™ Automated System, which uses modified tradition-al, fluorogenic, and chromogenic substrates for microbial identification. As shown in Table 4, 67 out of 69 bacterial species (97%) were correctly identified as 95.45% (21/22 isolates) Campylobacter jejuni, 100% (17/17 isolates) Esch-erichia coli, 100% (9/9 isolates) Salmonella enterica, 100% (8/8 isolates) Staphylococcus aureus, 100% (6/6 isolates) Acinetobacter baumannii, 75% (3/4 isolates) Enterobacter cloacae and 100% (3/3 isolates) Klebsiella pneumonia, while 94.44% (29/32 isolates) of fungi were correctly iden-tified as 77.77% (7/9 isolates) Alternaria alternate, 88.88% Aspergillus niger (7/8 isolates), 100% (6/6 isolates) Aspergil-lus flavus, 100% (4/4 isolates) Penicillium digitatum, 100% (3/3 isolates) Candida albicans and 100% (2/2 isolates) De-baryomyces hansenii.

Proteomic identification of foodborne pathogens In the present study, all bacterial and fungal isolates iso-lated from various food samples were identified by MAL-DI-TOF-MS fingerprinting and the spectra obtained were compared with the spectra stored in the Bruker database. As shown in Table 4, all foodborne bacteria (69 isolates) and fungi (32 isolates) were 100% correctly identified by MALDI-TOF-MS fingerprinting with a score value ≥2.00 except one isolate of Aspergillus niger which was identified with a score value <2.00.All spectra were analyzed by MALDI Biotyper compass

Table 3 - List of foodborne pathogens grown on selective agar media isolated from various restaurants in Al-Qassim region, KSA.

Microorganism Food samples No. of isolates % of isolates

Bacteria

Campylobacter jejuni Shawrmas with salads, Burger with salads, French fried potato, Matazeez and Tortilla

22 31.88

Escherichia coli Jareesh, Kabsa, Matazeez, Shawrmas with salads and Burger with salads

17 24.63

Salmonella enterica Shawrmas with salads and Burger with salads, Jareesh and Kabsa 9 13.04

Staphylococcus aureus Mozzarella, Burger with salads, French fried potato and Soup and Tortilla

8 11.59

Acinetobacter baumannii Shawrmas with salads and Humas 6 8.69

Enterobacter cloacae Jareesh, Matazeez 4 5.79

Klebsiella pneumoniae Burger with salads and Humas 3 4.34

Total 69 100%

Fungi

Alternaria alternata Jareesh, Homos , Tortilla, Burger with salads and Matazeez 9 28.12

Aspergillus niger Jareesh, Kabsa, Shawrmas with salads and Burger with salads, Tortilla 8 25.00

Aspergillus flavus Jareesh, Burger with salads and Matazeez 6 18.75

Penicillium digitatum Matazeez, Burger with salads 4 12.50

Candida albicans Shawrmas with salads, Jareesh, and Mozzarella 3 9.37

Debaryomyces hansenii Matazeez and Mozzarella 2 6.25

Total 32 100%

Application of MALDI-TOF MS fingerprinting as a quick tool for identification and clustering of foodborne pathogens isolated from food products 273

software and from 10 to 20 protuberant ion peaks were observed in the line spectra from the zone ranged from 2000 to 12000 Da. Higher strength peaks were detected between 3000 and 10000 Da that matched with 19 refer-ence strains of bacteria (Figure 2) and 9 reference strains of fungi (Figure 3) stored in the Bruker taxonomy. In this study, 7 species of bacteria and 6 species of fungi were properly identified (100%) with a score value rangiong

from 2.00 to 3.00. The various species of bacteria and fungi were recognized by matching their spectral profiles stored in the MALDI Biotyper database, which contains more than 5989 strains of the nationwide type culture collection. With a score value ranging from 2.00 to 2.29, 72.46% (50/69 isolates) of the tested bacteria were cor-rectly identified as follows: 7 Campylobacter jejuni, 12 Escherichia coli, 7 Salmonella enterica, 5 Staphylococcus

Table 4 - Identification of foodborne pathogens by BD Phoenix and MALDI TOF MS fingerprinting systems for 69 bacterial and 32 fungal isolates.

Microorganism Total numberBD Phoenix ID System MALDI-TOF MS Fingerprinting

Number CI (%) Number CI (%)

Bacteria

Campylobacter jejuni 22 21 95.45 22 100

Escherichia coli 17 17 100 17 100

Salmonella enterica 9 9 100 9 100

Staphylococcus aureus 8 8 100 8 100

Acinetobacter baumannii 6 6 100 6 100

Enterobacter cloacae 4 3 100 4 100

Klebsiella pneumoniae 3 3 100 3 100

Total 69 67 97 69 100

Fungi

Alternaria alternate 9 7 77.77 9 100

Aspergillus niger 8 7 88.88 8 100

Aspergillus flavus 6 6 100 6 100

Penicillium digitatum 4 4 100 4 100

Candida albicans 3 3 100 3 100

Debaryomyces hansenii 2 2 100 2 100

Total 32 29 94.44 32 100

CI, correctly identified rate

Table 5 - Score values of foodborne pathogens identified by MALDI-TOF MS.

Microorganism Number of isolatesScore values

0.00-1.69 1.70-1.99 2.00-2.29 2.30-3.00

Bacteria

Campylobacter jejuni 22 0 0 17 5

Escherichia coli 17 0 0 12 5

Salmonella enterica 9 0 0 7 2

Staphylococcus aureus 8 0 0 5 3

Acinetobacter baumannii 6 0 0 4 2

Enterobacter cloacae 4 0 0 3 1

Klebsiella pneumoniae 3 0 0 2 1

Total 69 0 0 50 19

Fungi

Alternaria alternate 9 0 0 5 4

Aspergillus niger 8 0 1 4 3

Aspergillus flavus 6 0 0 3 3

Penicillium digitatum 4 0 0 1 3

Candida albicans 3 0 0 2 1

Debaryomyces hansenii 2 0 0 2 0

Total 32 0 1 17 14

A. Elbehiry, E.Marzouk, M. Hamada, et al.274

Figure 2 - Comparison of mass spectrum protein pro-files of bacterial isolates iso-lated from food samples with 7 reference strains stored in the database of MALDI Biotyper Compass Satellite software. Blue means the spectrum stored in the database used for pattern matching; in the up-per half of the spectrum, green indicates matched peaks, red mismatched peaks, and yellow intermediate peaks.

Figure 3 - Comparison of mass spectrum protein pro-files of fungal isolates isolat-ed from food samples with 6 reference strains stored in the database of MALDI Biotyper Compass Satellite software. Blue means the spectrum stored in the database used for pattern matching; in the up-per half of the spectrum, green shows matched peaks, red mismatched peaks, and yellow intermediate peaks

Application of MALDI-TOF MS fingerprinting as a quick tool for identification and clustering of foodborne pathogens isolated from food products 275

aureus, 4 Acinetobacter baumannii, 3 Enterobacter cloa-cae and 2 Klebsiella pneumoniae (Table 4). Moreover, with the same score range, 53.12% (17/32 isolates) of the test-ed fungi were identified as 5 Alternaria alternate, 4 Asper-gillus niger, 3 Aspergillus flavus, 1 Penicillium digitatum, 2 Candida albicans and 2 Deboryomyces hansenii, while 27.53% (19/69 isolates) of bacterial isolates and 43.75% (14/32 isolates) of fungal isolates were correctly identi-

fied with a score value ranging from 2.30 to 3.00. Howev-er, only one isolate of Aspergillus niger was identified with a score value less than 2.00. A gel view was created by MALDI Biotyper machine for 69 well-characterized iso-lates of bacteria and 32 isolates of fungi. The maximum strength peak intensities for all spectra were concentrat-ed mainly between 3000 and 10000 Da (Figure 4). To es-tablish if the MALDI Biotyper Compass Software could

Figure 5 - The MSP dendro-gram for foodborne pathogens isolated from food samples. (A) 69 bacterial isolates ex-hibited a strong relation with 19 reference strains in the da-tabase; (B) 32 fungal isolates showed a close relation with 9 reference strains

Figure 4 - Gel view of protein spectra for some foodborne isolates isolated from various restaurants. The yellow color of spots was the gathering of protein spectra with various contents and higher strength peaks were detected between 3000 and 10000 Da.

A. Elbehiry, E.Marzouk, M. Hamada, et al.276

differentiate closely related strains at the genus species levels, spectra of all recognized isolates were examined as illustrated above and the spectra were then utilized to create a new cross-wise minimal spanning tree (MSP) dendrogram.

Clustering of identified foodborne pathogens As shown in Figure 5A, the bacterial spectra were created by Compass Satellite software to create MSP dendrogram. The findings of the dendrogram demonstrated that all identified isolates of Campylobacter jejuni, Escherichia coli, Salmonella enterica, Staphylococcus aureus, Acinetobacter baumannii, Enterobacter cloacae and Klebsiella pneumoni-ae were closely related to 19 reference strains stored in the Bruker library. From the created dendrogram, it was ob-served that all species of Salmonella enterica, Escherichia coli, Klebsiella pneumoniae and Enterobacter cloacae were matched together at the distance level of 600, whereas Campylobacter jejuni, Acinetobacter baumannii and Staph-ylococcus aureus strains were matched at the distance lev-el of 900. The results of the dendrogram for fungal isolates revealed that Alternaria alternate, Aspergillus niger, Asper-gillus flavus, Penicillium digitatum, Candida albicans and Deboryomyces hansenii were closely related to 9 reference strains stored in the Bruker library (Figure 5B). It was dis-played that Aspergillus niger and Aspergillus flavus were related to each other at the distance level of 600, while Penicillium spp., Deboryomyces hansenii, Alternaria alter-nate and Candida albicans were matched together at the distance level of 900.

DISCUSSION

In our research laboratory MALDI-TOF-MS fingerprinting has been established as a reliable tool for identification of various microorganisms such as bacteria, fungi and yeasts at the genus and species level. Mass spectrometry identifi-cation of foodborne pathogens has many advantages such as accuracy, rapidity, simplicity, sensitivity and reproduc-ibility. Analysis of samples is very fast (approximately 2 hours to examine a full 96-well target plate), which is considered a necessary factor for food quality and safe-ty, principally in certain conditions where the outbreak of food poisoning needs fast detection. Moreover, low reagent costs (about $0.50 per sample) represent a very strong benefit in the food safety supply chain (Cherkaoui et al., 2010; Pavlovic et al., 2013). Therefore, MALDI-TOF-MS offers the chance of solving the problem of foodborne pathogens.In the present work, 69 foodborne bacterial and 32 fun-gal isolates were isolated from 80 different restaurants in Al-Qassim region, KSA during the summer season, 2015. A limitation of the current study was that food samples were examined only for bacteria and fungi not for the ex-istence of their toxins. Fast identification of foodborne pathogens is considered a very important issue to confirm the safety of different food products (Böhme et al., 2016). MALDI-TOF MS is considered as an alternative technique for the primary recognition of microbial threats, which may infect different food products (Singhal et al., 2015; Cheng et al., 2016). The species of bacteria isolated in this study included Campylobacter jejuni, Escherichia coli, Salmonella enterica, Staphylococcus aureus, Acinetobacter baumannii, Enterobacter cloacae and Klebsiella pneumo-nia. All these isolates were correctly identified 100% by

MALDI-TOF MS fingerprinting at the genus and species levels with a score value ≥2.00, whereas 97% of bacterial isolates and 94.44% of fungal isolates were correctly iden-tified by the BD Phoenix system.The results of the current study were similar to several studies using MALDI-TOF MS as a confirmation tool for identification of foodborne pathogens like Campylobacter species (Bessède et al., 2011; Zautner et al., 2013), Esche-richia coli (Khot and Fisher, 2013; Christner et al., 2014; Matsumura et al., 2014). Moreover, Salmonella enterica (Dieckmann and Malorny, 2011; Sparbier et al., 2012), Staphylococcus aureus (Josten et al., 2013) and Enterobacter cloacae (Khot and Fisher, 2013; Almuzara et al., 2015) were identified by MALDI-OF MS fingerprinting, which also identified a large number of foodborne pathogens, such as Escherichia, Yersinia, Proteus, Morganella, Salmonella, Staphylococcus, Micrococcus, Lactococcus, Pseudomonas and Listeria (Mazzeo et al., 2006). Some pathogenic mi-croorganisms isolated from food products such as seafood were also correctly identified by MALDI-TOF MS (Böhme et al., 2010; Böhme et al., 2011; Böhme et al., 2013). They identified the major significant species of bacteria isolat-ed from food of sea origin, such as Acinetobacter species, Campylobacter species, Staphylococcus species, Pseudo-monas species, and Vibrio species. In another work, An-geletti et al. (2015b) and Kilic et al. (2016) used MALDI-TOF-MS for identification of Klebsiella pneumonia strains isolated from various clinical samples. They indicated that using of MALDI Biotyper is an excellent tool for accurate identification of all strains at the species level with a score value ≥2. Sakarikou et al. (2017) also investigated 175 Klebsiella pneumonia strains isolated directly from blood culture samples by MALDI-TOF-MS. They found that all strains (100%) were correctly identified at the species level and the MALDI-TOF-MS technology appeared accurate, cost-effective and fast. Identification of various microor-ganisms using protein fingerprinting is considered one of the common and superior microorganism identification tool used at the current time. It depends mainly on the spectral variations of bacterial and fungal strains to detect an unidentified type by the evaluation of their spectral side view stored in the Bruker database (Mazzeo et al., 2006; Böhme et al., 2013; El Behiry et al., 2014; Elbehiry et al., 2016). After comparison between two MALDI-TOF instru-ments, the MALDI Biotyper from Bruker Daltonics and the Vitek-MS from bioMérieux for identification of Viri-dans Group Streptococci (VGS) isolates at the species-lev-el, Angeletti et al. (2015a) found that both instruments are reliable, cost saving and rapid for identification of VGS even within the Mitis group.Thirty-two isolates of fungi namely Alternaria alternate, Aspergillus niger, Aspergillus flavus, Penicillium digitatum, Candida albicans and Deboryomyces hansenii were also identified properly by MALDI-TOF-MS. Similar results were obtained by Dhiman et al. (2011), Sendid et al. (2013) and Becker et al. (2014), who identified different species of fungi by MALDI-TOF-MS fingerprinting. In addition, Angelleti et al. (2015c) isolated Candida species from two hospitals in Rome (nosocomial isolates), then identified and clustered by MALDI-TOF-MS. They established that the various Candida species were properly identified and clustered independently, authorizing the capability of this method to distinguish between different species of Candi-da. A few genera of filamentous fungi such as Aspergillus species (Alanio et al., 2011; DeCarolis et al., 2012); Penicil-

Application of MALDI-TOF MS fingerprinting as a quick tool for identification and clustering of foodborne pathogens isolated from food products 277

lium spp. (Del Chierico et al., 2012) and Fusarium species (DeCarolis et al., 2012; Del Chierico et al., 2012) which are considered the most commonly encountered foodborne fungal pathogens were also identified by MALDI-TOF MS. Conversely, Kim et al. (2016) indicated that MALDI-TOF MS could not accurately recognize some uncommon Can-dida species. It appears that the spectra for some Candida species were not included in the library software or had inadequate and phenotypic variation. A further main objective of our study was to assess the ability Microflex LT database for the clustering of food isolates potentially existing in the food products. The per-formance of Microflex LT for clustering of 69 bacterial and 32 fungal isolates were evaluated as a complement typing method, especially for screening purposes. The results of the MSP dendrogram showed that most of the identified bacteria and fungi were highly related with 19 and 9 ref-erence strains stored in the Bruker taxonomy, respectively. Christensen et al. (2012) indicated that the MSP dendro-gram is considered a reliable tool for illustrating the capa-bility of MALDI-TOF MS to visualize the degree of similar-ities and differences between species when more isolates are considered.All bacterial and fungal isolates used in our study were identified by MALDI-TOF MS fingerprinting (protein fin-gerprinting) in less than 2 h. By contrast, identification by colorimetric or molecular methods can take several hours or days (Böhme et al., 2016; Elbehiry et al., 2016). There-fore, in our study when we compared the MALDI-TOF-MS with the colorimetric method (BD Phoenix System) for routine identification of various foodborne pathogens, the MALDI-TOF-MS proved more reliable, rapid and sensitive than the BD Phoenix System. Throughout this study, we verified that MALDI-TOF MS fingerprinting combined with the compass software was not only a quick, inexpensive, and precise tool for identification of foodborne pathogens but also for clustering of different bacteria and fungi.One of the possible drawbacks of MALDI-MS analysis is the device cost. Various laboratories that apply MAL-DI-TOF MS fingerprinting for identification of food prod-ucts may be satisfied with the use of this device. However, considering the extremely low cost of consumables and the possibility of reducing labor costs, the total per sample cost may really be decreased in the long run compared to other techniques. MALDI-TOF MS fingerprinting is con-sidered one of the chemotaxonomic approaches since it is based mainly on the similarities and differences in certain biomarkers. Not only is comparison with reference data-bases compulsory for obtaining excellent identification re-sults, but the value of the reference data, reference protein fingerprints, and the algorithm for identification are also very important (Pavlovic et al., 2013). The algorithm for automated recognition of different isolates in the MAL-DI Biotyper was developed in terms of a strong and re-liable identification process. A restricted number of high strength peaks, mostly due to ribosomal proteins (Ryzhov and Fenselau, 2001; Pavlovic et al., 2013) have been select-ed as biomarkers.

CONCLUSIONS

Based on our findings, various foodborne pathogens were correctly identified and typed by MALDI-TOF MS finger-printing. Campylobacter jejuni, Escherichia coli, Salmonel-la enterica and Staphylococcus aureus were the most com-

mon isolated pathogenic bacteria, while Aspergillus niger, Aspergillus flavus and Candida albicans were the major fungal pathogens isolated from the food samples. MAL-DI-TOF MS protein fingerprinting is considered one of the most powerful techniques used for fast and accurate iden-tification of foodborne pathogens. This technology is char-acterized by its simple procedures and shortened analysis time, which play an important role in the maintenance of a high-level of food safety.

Conflict of interestThe authors confirm that this article content has no con-flict of interest.

AcknowledgmentsThe authors extend their appreciation to the Deanship of Scientific Research at King Saud University for funding the work through the research group project No.: RGP-VPP-162. We also thank Dr. Ilias for revising the manu-script.

ReferencesAlanio A., Beretti J.L., Dauphin B.,   Mellado E.,  Quesne G.,  Lacroix

C., Amara A., Berche P., Nassif X., Bougnoux M.E. (2011). Matrix-as-sisted laser desorption ionization time-of flight mass spectrometry for fast and accurate identification of clinically relevant Aspergillus spe-cies. Clin. Microbiol. Infect. 17, 750-5.

Almuzara M., Barberis C., Traglia G., Famiglietti A., Ramirez M.S., Vay C. (2015). Evaluation of matrix-assisted laser desorption ionization-time-of-flight mass spectrometry for species identification of nonfermenting gram-negative bacilli. J. Microbiol. Methods 112, 24-27.

Alvarez-Buylla A., Culebras E., Picazo J.J. (2012). Identification of Acine-tobacter species: is Bruker biotyper MALDI-TOF mass spectrometry a good alternative to molecular techniques? Infect. Genet. Evol. 12, 345-349.

Angelakis E.,  Yasir M.,  Azhar E.I.,  Papadioti A.,  Bibi F.,  Aburizaiza A.S.,  Metidji S.,  Memish Z.A.,  Ashshi A.M.,  Hassan A.M.,  Harakeh S., Gautret P., Raoult D. (2014). MALDI-TOF mass spectrometry and identification of new bacteria species in air samples from Makkah, Saudi Arabia. BMC Res. Notes 7, 892-899.

Angeletti S., Dicuonzo., Avola A., Crea F., Dedej E., Vailati F., Farina C., De FlorioL. (2015a). Viridans Group Streptococci Clinical Isolates: MAL-DI-TOF Mass Spectrometry versus Gene Sequence-Based Identifica-tion. PLoS One 10, e0120502.

Angeletti S., Dicuonzo G., Lo Presti A., Cella E., Crea F., Avola A., Vitali M.A., Fagioni M., De Florio L. (2015b). MALDI-TOF mass spectrom-etry and blakpc gene phylogenetic analysis of an outbreak of carbap-enem-resistant K. pneumoniae strains. New Microbiol.  38, 541-550. 

Angeletti S., Lo Prestib A., Cellab E., Dicuonzoa G., Creaa F., Palazzottic B.,  Dedeja E., Ciccozzib M., De Florioa L. (2015c). Matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF MS) and Bayesian phylogenetic analysis to characterize Candida clin-ical isolates. J. Microbiol. Methods. 119, 214-222

Becker P.T., de Bel A., Martiny D., Ranque S., Piarroux R., Cassagne C., Detandt M., Hendrickx M. (2014). Identification of filamentous fungi isolates by MALDI-TOF mass spectrometry: clinical evaluation of an extended reference spectra library. Med. Mycol. 52, 826-834.

Bessède E., Solecki O., Sifré E., Labadi L., Mégraud F. (2011). Identifica-tion of Campylobacter species and related organisms by matrix assist-ed laser desorption ionization-time of flight (MALDI-TOF) mass spec-trometry. Clin. Microbiol. Infect. 17, 1735-1739.

Böhme K., Antelo C.S., Fernández-No I.C., Quintela-Baluja M., Bar-ros-Velázquez J., Cañas, B., Calo-Mata, P. (2016). Detection of Food-borne Pathogens Using MALDI-TOF Mass Spectrometry. Antimicrobi-al Food Packaging 15, 203-214.

Böhme K., Fernández-No I.C., Barros-Velázquez J., Gallardo J.M., Ca-lo-Mata P., Cañas B. (2010). Species differentiation of seafood spoilage and pathogenic gram-negative bacteria by MALDI-TOF mass finger-printing. J. Proteome Res. 9, 3169-3183.

Böhme K., Fernández-No I.C., Barros-Velázquez J., Gallardo J.M., Cañas B., Calo-Mata P. (2011). Rapid species identification of seafood spoil-age and pathogenic Gram-positive bacteria by MALDI-TOF mass fin-gerprinting. Electrophoresis 32, 2951-2965.

Böhme K., Fernández-No I.C., Pazos M., Gallardo J.M., Barros-Velázquez J., Cañas B., Calo-Mata P. (2013). Identification and classification of seafood-borne pathogenic and spoilage bacteria: 16S rRNA sequenc-ing versus MALDI-TOF MS fingerprinting. Electrophoresis 34, 877-887.

A. Elbehiry, E.Marzouk, M. Hamada, et al.278

Carannante A., De Carolis E., Vacca P., Vella A., Vocale C., De Frances-co M.A., Cusini M., Del Re S., Conte I.D., Cristaudo A., Ober P., San-guinetti M., Stefanelli P. (2015). Evaluation of matrix-assisted laser desorption ionization-time of flight mass spectrometry (MALDI-TOF MS) for identification and clustering of Neisseria gonorrhoeae. BMC Microbiol. 15, 142-148.

Cheng K., Chui H., Domish L., Hernandez D., Wang G. (2016). Recent de-velopment of mass spectrometry and proteomics applications in iden-tification and typing of bacteria. Proteomics Clin. Appl. 10, 346-357.

Cherkaoui, A., Hibbs, J., Emonet, S., Tangomo, M., Girard, M., Francois, P., Schrenzel, J. (2010). Comparison of two matrix-assisted laser desorp-tion ionization-time of flight mass spectrometry methods with conven-tional phenotypic identification for routine identification of bacteria to the species level. J. Clin. Microbiol. 48, 1169-1175.

Christensen J.J., Dargis R., Hammer M., Justesen U.S., Nielsen X.C., Kemp M. (2012). Matrix-assisted laser desorption ionization-time of flight mass spectrometry analysis of Gram-positive, catalase-negative cocci not belonging to the Streptococcus or Enterococcus genus and bene-fits of database extension. J. Clin. Microbiol.  50, 1787-1791.

Christner M., Trusch M., Rohde H., Kwiatkowski M., Schlüter H., Wolters M., Aepfelbacher M., Hentschke M. (2014). Rapid MALDI-TOF mass spectrometry strain typing during a large outbreak of Shiga-toxigenic Escherichia coli. PLoS One 9, e101924.

DeCarolis E., Posteraro B., Lass-Florl C., Vella A.,  Florio A.R.,  Torelli R., Girmenia C., Colozza C., Tortorano A.M., Sanguinetti M., Fadda G. (2012). Species identification of Aspergillus, Fusarium and Muco-rales with direct surface analysis by matrix-assisted laser desorption ionization time-of flight mass spectrometry. Clin. Microbiol. Infect. 18, 475-484.

Del Chierico F., Masotti A., Onori M.,  Fiscarelli E., Mancinelli L., Ricciotti G., Alghisi F., Dimiziani L., Manetti C., Urbani A., Muraca M., Putig-nani L. (2012). MALDI-TOF MS proteomic phenotyping of filamentous and other fungi from clinical origin. J. Proteomics 75, 3314-3330.

Dhiman N., Hall L., Wohlfiel S.L., Buckwalter S.P., Wengenack N.L. (2011). Performance and cost analysis of matrix-assisted laser desorption ion-ization time-of flight mass spectrometry for routine identification of yeast. J. Clin. Microbiol. 49, 1614-1616.

Dieckmann R., Malorny B. (2011). Rapid screening of epidemiologically important Salmonella enterica subsp. enterica serovars by whole-cell matrixassisted laser desorption ionization-time of flight mass spec-trometry. Appl. Environ. Microbiol. 77, 4136-4146.

El Behiry A., Zahran R.N., Marzouk E., Al-Dabib M. (2014). Phenotypical and mass spectral assessment methods for identification of some con-tagious mastitis pathogens. American J. of Microbiol. 5, 1-10.

Elbehiry A., Al-Dubaib M., Marzouk E., Osman S., Edrees H. (2016). Per-formance of MALDI biotyper compared with Vitek™ 2 compact sys-tem for fast identification and discrimination of Staphylococcus spe-cies isolated from bovine mastitis. Microbiology Open 5, 1061-1070.

Fournier P.E., Drancourt M., Colson P., Rolain J.M., La Scola B., Raoult D. (2013). Modern clinical microbiology: new challenges and solutions. Nat. Rev. Microbiol. 11, 574-585.

Jasson V., Jacxsens L., Luning P., Rajkovic A., Uyttendaele M. (2010). Al-ternative microbial methods: An overview and selection criteria. Food Microbiol. 27, 710-730.

Josten M., Reif M., Szekat C., Al-Sabti N., Roemer T., Sparbier K., Kostrze-wa M., Rohde H., Sahl H.G., Bierbaum G. (2013). Analysis of the ma-trixassisted laser desorption ionization-time of flight mass spectrum of Staphylococcus aureus identifies mutations that allow differentiation of the main clonal lineages. J. Clin. Microbiol. 51, 1809-1817.

Khot P.D., Fisher M.A. (2013). Novel approach for differentiating Shigella species and Escherichia coli by matrix-assisted laser desorption ion-izationtime of flight mass spectrometry. J. Clin. Microbiol. 51, 3711-3716.

Kilic A., Dogan E.,  aya S., Oren S., Tok D., Ardic N., Baysallar M. (2016). Rapid Identification of Klebsiella pneumoniae by Matrix-Assisted La-ser Desorption/Ionization-Time of Flight Mass Spectrometry and De-tection of Meropenem Resistance by Flow Cytometric Assay. J. Clin. Lab. Anal.  30, 1191-1197.

Kim T., Kweon O.J., Kim H.R., Lee M. (2016). Identification of Uncommon Candida Species Using Commercial Identification Systems. J. Microbi-ol. Biotechnol. 26, 2206-2213.

Kramer M., Obermajer N., Matijasic B.B., Rogelj I., Kmetec V. (2009). Quantification of live and dead probiotic bacteria in lyophilised prod-uct by real-time PCR and by flow cytometry. Appl. Genetics and Molec-ular Biotechnol. 84, 1137-1147.

Law J.W., Ab Mutalib N., Chan K., Lee L. (2015). Rapid methods for the detection of foodborne bacterial pathogens: principles, applications, advantages and limitations. Frontiers 5, 1-19.

Lee N., Kwon K.Y., Oh S.K., Chang H.J., Chun H.S., Choi S.W. (2014). A multiplex PCR assay for simultaneous detection of Escherichia coli O157:H7, Bacillus cereus, Vibrio parahaemolyticus, Salmonella spp., Listeria monocytogenes, and Staphylococcus aureus in Korea ready-to-eat food. Foodborne Pathog. Dis. 11, 574-580.

Lima N., Santos C. (2017). MALDI-TOF MS for identification of food spoil-age filamentous fungi. Current Opinion in Food Science 13, 26-30.

Liu H., Du Z., Wang J., Yang R. (2007). Universal sample preparation meth-od for characterization of bacteria by assisted laser desorption ion-ization-time of flight mass spectrometry. Appl. Environ. Microbiol. 73, 1899-1907.

Marí-Almirall M.1., Cosgaya C.1., Higgins P.G., Assche V., Telli M., Huys G., Lievens B., Seifert H., Dijkshoorn L., Roca I., Vila J.L. (2016). MAL-DI-TOF/MS identification of species from the Acinetobacter bauman-nii (Ab) group revisited: inclusion of the novel A. seifertii and A. dijk-shoorniae species. Microbiol. Infect. 23, 210-28.

Matsumura Y., Yamamoto M., Nagao M., Tanaka M., Machida K., Ito Y., Takakura S., Ichiyama S. (2014). Detection of extended-spec-trum-β-lactamase-producing Escherichia coli ST131 and ST405 clonal groups by matrix-assisted laser desorption ionization-time of flight mass spectrometry. J. Clin. Microbiol. 52, 1034-1040.

Mazzeo M.F., Sorrentino A., Gaita M., Cacace G., Di Stasio M., Facchiano A., Comi G., Malorni A., Siciliano R.A. (2006). Matrix-assisted laser desorption ionization-time of flight mass spectrometry for the discrim-ination of food-borne microorganisms. Appl. Environ. Microbiol. 72, 1180-1189.

McAuley C.M., McMillan K., Moore S.C., Fegan N., Fox E.M. (2014). Prev-alence and characterization of foodborne pathogens from Australian dairy farm environments. J. Dairy Sci. 97, 7402-7412.

Nicolaou N., Xu Y., Goodacre R. (2012). Detection and quantification of bacterial spoilage in milk and pork meat using MALDI-TOF-MS and multivariate analysis. Anal. Chem. 84, 5951-5958.

Novais Â., Sousa C., de Dios Caballero J., Fernandez-Olmos A., Lopes J., Ramos H., Coque T.M., Cantón R., Peixe L. (2014). MALDI-TOF mass spectrometry as a tool for the discrimination of high-risk Escherichia coli clones from phylogenetic groups B2 (ST131) and D (ST69, ST405, ST393). Eur. J. Clin. Microbiol. Infect. Dis. 33, 1391-1399.

Pavlovic M., Huber I., Konrad R., Busch U. (2013). Application of MAL-DI-TOF MS for the  Identification of Food Borne Bacteria. Open Mi-crobiol. J. 7, 135-141.

Posteraro B., De Carolis E., Vella A., Sanguinetti M. (2013). MALDI-TOF Mass Spectrometry in the Clinical Mycology Laboratory: Identification of Fungi and Beyond. Expert. Rev. Proteomics 10, 151-164.

Ryzhov V., Fenselau C. (2001). Characterization of the protein subset de-sorbed by MALDI from whole bacterial cells. Anal. Chem. 73, 746-50. 

Sakarikou C., Ciotti M., Dolfa C., Angeletti S., Favalli C. (2017). Rapid de-tection of carbapenemase producing Klebsiella pneumoniae strains derived from blood cultures by Matrix-Assisted Laser Desorption Ion-ization-Time of Flight Mass Spectrometry (MALDI-TOF MS). BMC Microbiol. 17, 54-61.

Scallan E., Hoekstra R.M., Angulo F.J., Tauxe R.V., Widdowson M.A., Roy S. L., et al. (2011). Foodborne illness acquired in the United States- major pathogens. Emerg. Infect. Dis. 17, 7-15.

Sendid B., Ducoroy P., Francois N., Lucchi G., Spinali S., Vagner O., Damiens S., Bonnin A., Poulain D., Dalle F. (2013). Evaluation of MAL-DI-TOF mass spectrometry for the identification of medically-import-ant yeasts in the clinical laboratories of Dijon and Lille hospitals. Med. Mycol. 51, 25-32.

Singhal N., Kumar M., Kanaujia P.K., Virdi J.S. (2015). MALDI-TOF mass spectrometry: an emerging technology for microbial identification and diagnosis. Front Microbiol. 5, 791-806.

Sousa C., Botelho J., Silva L., Grosso F., Nemec A., Lopes J., Peixe L. (2014). MALDI-TOF MS and chemometric based identification of the Acineto-bacter calcoaceticus-Acinetobacter baumannii complex species. Int. J. Med. Microbiol. 304, 669-677.

Sparbier K., Weller U., Boogen C., Kostrzewa M. (2012). Rapid detection of Salmonella sp. by means of a combination of selective enrichment broth and MALDI-TOF MS. Eur. J. Clin. Microbiol. Infect. Dis. 31, 767-773.

Swanson K.M., Busta F.F., Peterson E.H., Johnson M.G. (1992). Colony count methods. In Vanderzant C, Spilitstoesser F (4th ed) Compendi-um of methods for the microbiological examination of foods. Ameri-can Public Health Association, Washington, DC USA. 75-95.

Welker M., Moore E.R.B. (2011). Application of whole-cell matrix-assisted laser desorption/ionization time-of-flight mass spectrometry in sys-tematic microbiology. Syst. Appl. Microbiol. 34, 2-11.

Wenning M., Breitenwieser F., Konrad R., Huber I., Busch U., Scherer S. (2014). Identification and differentiation of food-related bacteria: A comparison of FTIR spectroscopy and MALDI-TOF mass spectrome-try. J. Microbiol. Methods. 103, 44-52.

Wieser A., Schneider L., Jung J., Schubert S. (2012). MALDI-TOF MS in microbiological diagnostics-identification of microorganisms and be-yond (minireview). Appl. Microbiol. Biotechnol. 93, 965-974.

Zautner A.E., Masanta W.O., Tareen A.M., Weig M., Lugert R., Groß U., Bader O. (2013). Discrimination of multilocus sequence typing-based Campylobacter jejuni subgroups by MALDI-TOF mass spectrometry. BMC Microbiol. 13, 247-255.

Zhao X., Lin C.W., Wang J., Oh D.H. (2014). Advances in rapid detection methods for foodborne pathogens. J. Microbiol. Biotechn. 24, 297-312.