Sie sind auf Seite 1von 35

See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/337889239

Gram-negative bacteria from fish, meat and chicken UV-VIS


spectrophotometry for determination of caffeic acid Journal of Food Safety
and Food Quality Archiv für Lebensmittelhyg...

Article · December 2019

CITATIONS READS

0 249

1 author:

Neslihan Gundogan
Gazi University
9 PUBLICATIONS   151 CITATIONS   

SEE PROFILE

All content following this page was uploaded by Neslihan Gundogan on 10 January 2020.

The user has requested enhancement of the downloaded file.


Journal of Food Safety
and Food Quality
Archiv für Lebensmittelhygiene  No.

Volume 70
4
Highlights in this issue July/August 2019
  Salmonellae isolated from fish Pages 91–124
in Brazil Page  94 ISSN 0003-925 X

 Gram-negative bacteria from fish,


meat and chicken Page  99

 UV-VIS spectrophotometry for


determination of caffeic acid Page 111

 Physical properties of pigeon pea Page 117

Foto: Wibowo Djatmiko/wikipedia.de


92 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

+++ Nachrichten aus Forschung, Politik und Industrie +++


(Die Verantwortlichkeit für die Texte liegt ausschließlich bei den Instituten, Ministerien und werbenden Unternehmen.)

OPSON VIII: Behörden untersuchen ben wurden Arabica- durch Robustabohnen ausgetauscht. Liegen
alle Ergebnisse vor, wird auf europäischer Ebene ein Gesamtbild zu
Betrug bei Kaffee Ausmaß und Strukturen beim Kaffeebetrug erstellt.
Bei der diesjährigen von Europol und INTERPOL koordinierten
Operation OPSON VIII haben sich dreizehn europäische Staaten Laborkooperation
zusammengetan, um gemeinsam etwaige Betrugsfälle bei Kaffee Bei der diesjährigen Schwerpunktaktion gab es erstmalig eine Labor­
zu verfolgen. In Deutschland, Portugal und der Schweiz wurden in kooperation auf nationaler und europäischer Ebene. Dies stärkt die
neun Fällen preisgünstigere Robustabohnen in „Arabica“-Kaffee zwischenbehördliche Zusammenarbeit auch im Bereich der Analytik.
nachgewiesen. Dies teilte das Bundesamt für Verbraucherschutz Analytikfragen spielen eine Schlüsselrolle bei der effektiven Bekämp­
und Lebensmittelsicherheit (BVL) heute in Berlin mit, welches die fung von Food Fraud (Lebensmittelkriminalität). Das C ­ hemische
Schwerpunktaktion koordinierte. In Deutschland beteiligten sich und Veterinäruntersuchungsamt (CVUA) Karlsruhe hielt Labor­
neben den Lebensmittelüberwachungsbehörden der Bundesländer kapazitäten für die Ermittlung des 16-O-Methylcafestol-­Gehalts mit­
auch der Zoll und das Bundeskriminalamt an der Operation. tels NMR-Analyse (Kernspinresonanzspektroskopie) ­bereit. Diese
Während der sechswöchigen Kernphase der Operation von Anfang Kapazitäten konnten nicht nur von den deutschen Behörden, sondern
Januar bis Mitte Februar prüften die Lebensmittelüberwachungs­ von allen Teilnehmerstaaten der Schwerpunktaktion in Anspruch
behörden in Deutschland über 1,5 Tonnen gerösteten sowie geröste­ ­genommen werden. Außerdem hat sich, initiiert durch das BVL, eine
ten und gemahlenen Kaffee. Bei den 134 Kontrollen wurde der als weitere Laborkooperation zur „Herkunftsbestimmung“ ergeben. Die
100 % Arabica deklarierte Kaffee auf einen Nachweis von Robusta­ Herkunftsanalyse bei Kaffee ist schwierig. Es fehlt an eindeutigem
bohnen hin untersucht. Der Zoll unterstützte die Operation mit der Referenzmaterial. Zwei deutsche Laborgerätehersteller sowie drei
Auswertung und Bereitstellung von Einfuhrdaten zu Kaffee. deutsche Untersuchungsämter haben deshalb 106 Proben analysiert,
Von der Lebensmittelüberwachung konnte in drei Fällen (ca. deren Herkunft sich über mehrere Kontinente und zahlreiche Staaten
2 % der untersuchten Proben) eine Irreführung festgestellt werden. verteilte. Die Ergebnisse der Herkunftsbestimmung stehen noch aus.
Dabei reichten die ermittelten Robusta-Gehalte von ca. 7 % bis hin Sie sollen vor allem dem Aufbau eines Probenpools und damit der
zu 100 %. Die Arabicapflanze (Hochlandkaffee) ist im Anbau an­ Etablierung einer Analysenmethode dienen.
spruchsvoller als die widerstandsfähigere Robustapflanze. Auch die Zusätzlich unterstützte ein namhaftes Kaffeeunternehmen die
Aufbereitung des Rohkaffees unterscheidet sich. Arabicabohnen europaweite Schwerpunktaktion mit der Bereitstellung von authen­
­erzielen daher höhere Preise als Robustabohnen. Sind die Bohnen tischen Mustern von Kaffeeproben für einen Authentizitätsabgleich.
optisch noch zu unterscheiden, wird eine genaue Bestimmung schwie­
rig, sobald der Rohkaffee geröstet und gemahlen wurde. Zum Nach­ Weitere Schwerpunktaktionen
weis von Anteilen der Robusta-Kaffeebohne wurden die Proben auf Neben der Schwerpunktaktion zu Kaffee, die vom BVL koordiniert
den Gehalt von 16-O-Methylcafestol untersucht. Dieser Inhaltsstoff wurde, gab es erstmals bei OPSON zwei weitere Schwerpunktaktio­
ist nur in Kaffeebohnen der Art Robusta enthalten. nen. Die Europäische Kommission koordinierte eine Schwerpunkt­
aktion zur Bekämpfung des Betrugs bei Bio-Lebensmitteln, an der
Ermittlungen dauern an sich 16 europäische Staaten beteiligt haben. Großbritannien und
In allen drei Fällen erfolgte der Eintrag in der Rösterei. Hinweise weitere neun Staaten haben Angebote von Fatburner-Produkten ver­
auf eine Beteiligung weiterer Lebensmittelunternehmer oder auf folgt, die den Inhaltsstoff DNP (2,4-Dinitrophenol) enthielten.
verfälschte Rohware gibt es nicht. Die Ermittlungen der Lebens­
mittelüberwachung dauern teilweise noch an. Ein Fall wurde bereits Weitere Informationen (Quelle):
an die Staatsanwaltschaft übergeben. Auch in der Schweiz und in Bundesamt für Verbraucherschutz und
­Portugal wurden Fälle von Verfälschungen aufgedeckt. In sechs Pro­ Lebensmittelsicherheit (BVL) | www.bvl.bund.de

Mikroorganismen auf Mikroplastik tikeln. Der Name „piscicida“ bedeutet fischtötend, denn der Erreger
kann die Haut von Fischen durch die Bildung von Giftstoffen schä­
Organismen können auf Mikroplastik wachsen und sich im Gewässer digen. Diese Gifte können bei Massenentwicklung die Gesundheit
anreichern. Forschende vom Leibniz-Institut für Gewässerökologie von Mensch und Tier stark gefährden. Die Forschenden wählten zur
und Binnenfischerei (IGB) und dem Leibniz-Institut für Ostseefor- Dichteabschätzung der unterschiedlichen Mikroorganismen eine Me­
schung (IOW) zeigen in einer Studie, dass auch potenziell giftbil- thode zur Quantifizierung spezifischer ribosomaler RNA. Bei dieser
dende Mikroorganismen wie der Dinoflagellat Pfiesteria piscicida Methode wird nicht die tatsächliche Zellzahl der besiedelnden Orga­
Plastikteilchen besiedeln und dort etwa fünfzig Mal so hohe Konzen- nismen erfasst. Sie gilt aber als guter Indikator dafür, in welchem Maß
trationen erreichen wie im umgebenden Wasser und etwa zwei bis bestimmte Organismen eine mikrobielle Lebensgemeinschaft prägen.
drei Mal so hohe Dichten wie auf vergleichbaren Treibholzpartikeln. „Mikroplastik kann ein bedeutender Lebensraum und ein Trans­
portmittel für Mikroorganismen sein – auch für giftige oder schädigen­
Ein Plastikteilchen von einem Gramm Gewicht kann mehr lebende de. Wir konnten in unseren Untersuchungen feststellen, dass Mikroor­
Organismen beherbergen als eintausend Liter Seewasser, in denen es ganismen, beispielsweise potenziell giftige Dinoflagellaten wie Pfiesteria
schwimmt. Bisher ist kaum erforscht, in welchem Ausmaß Mikroorga­ piscicida, sich auf Plastikteilchen anreichern und dort höhere Dichten
nismen Mikroplastik im Brackwasser besiedeln und welche Arten da­ als auf Treibholzteilchen oder im umgebenden Wasser erreichen,“ er­
bei dominieren. Ein Team aus Gewässerforscherinnen und Gewässer­ läutert die Erstautorin der Studie, Maria Therese Kettner vom IGB, die
forschern untersuchte die natürliche Besiedlung von Mikroplastik aus Ergebnisse. Der Leiter der Studie, IGB-Forscher Hans-Peter Grossart,
Polyethylen (PE) und Polysterol (PS) mit eukaryotischen Mikroorga­ spricht eine weitere Problematik an: „Im Gegensatz zu natürlichen Subs­
nismen. Eukaryotische Mikroorganismen sind beispielsweise Plankton­ tanzen wie Holz oder Algenkolonien, zerfallen die Mikroplastikpartikel
arten, die – anders als Bakterien und Viren – einen Zellkern besitzen. nur extrem langsam und können so die anhaftenden Lebewesen über
Die Forschenden inkubierten wenige Millimeter große PE- und weite Strecken transportieren“. Schwimmendes Plastik könnte damit
PS-Teilchen für 15 Tage an verschiedenen Stellen in der Ostsee, der zur Ausbreitung von verschiedensten Organismen, darunter invasive,
Warnow und in einer Kläranlage mit der natürlichen Mikrofauna. An­ parasitäre oder pathogene Arten, beitragen. „Allerdings verändern sich
schließend untersuchten sie mittels Sequenzanalysen die komplexen die Gemeinschaften auf Mikroplastikpartikeln auch häufig, wenn sie
Lebensgemeinschaften auf dem Mikroplastik. Rund 500 verschiedene ‚auf Reisen sind‘ und passen sich ihrer neuen Umgebung an“, so Meeres­
Eukaryoten-Arten tummelten sich auf den winzigen Teilchen. mikrobiologe Matthias Labrenz. „Daher benötigen diese Aspekte noch
Die Top-20-Liste der Mikroplastik Mikroorganismen wurde von weitere Untersuchungen“, so der IOW-Forscher abschließend.
einer potenziell giftigen Planktonart, dem Dinoflagellaten Pfiesteria
piscicida angeführt. Er erreichte etwa fünfzig Mal so hohe Dichten Weitere Informationen (Quelle):
wie wie im umgebenden Wasser und etwa zwei bis drei Mal so hohe Leibniz-Institut für Gewässerökologie und
Dichten wie auf vergleichbaren im Wasser schwimmenden Holzpar­ Binnenfischerei (IGB) | www.igb-berlin.de
Journal of Food Safety and Food Quality Juli/August 2019
70. Jahrgang
Archiv für Lebensmittelhygiene Seiten 91–124

Offizielles Organ des Arbeitsgebietes Lebensmittelhygiene der Deutschen Veterinärmedizinischen Gesellschaft e. V.

ORIGINAL ARTICLES ORIGINALARBEITEN

Adelino Cunha-Neto, Pedro Panzenhagen, Adelino Cunha-Neto, Pedro Panzenhagen,


Larrayane Carvalho, Dália Rodrigues, Larrayane Carvalho, Dália Rodrigues,
Carlos Conte-Júnior, Eduardo Figueiredo Carlos Conte-Júnior, Eduardo Figueiredo
Occurrence and antimicrobial resistance p
­ rofile Vorkommen und Antibiotikaresistenzprofil
of Salmonella isolated from native fish von Salmonellen aus in Brasilien geschlachteten
slaughtered and commercialised in Brazil ................. 94 und vermarkteten einheimischen Fischen ................ 94

Neslihan Gundoğan, M. Burcu Külahcı, Neslihan Gundoğan, M. Burcu Külahcı,


Ethem Serhat Yavaş Ethem Serhat Yavaş
Amino acid decarboxylase activity, biofilm Aminosäuren-Dekarboxylase-Aktivität, Biofilm­
formation and antibiotic resistance of bildung und Antibiotikaresistenz gramnegativer
gram-negative bacteria isolated from Bakterien isoliert aus Meeresfischen, Kalbs- und
marine fish, calf meat and chicken ............................. 99 Hähnchenfleisch ............................................................ 99

Danijela A. Kostic, Snezana Mitic, Milan Mitic, Danijela A. Kostic, Snezana Mitic, Milan Mitic,
Emilija Pecev Marinkovic, Ivana Rasic Misic, Emilija Pecev Marinkovic, Ivana Rasic Misic,
Biljana Arsic, Gordana Stojanovic Biljana Arsic, Gordana Stojanovic
A new kinetic method using UV-VIS Ein neues kinetisches Verfahren zur Bestimmung
spectrophotometry for determination von Kaffeesäure in Propolis mittels UV-VIS-
of caffeic acid in propolis ...........................................  111 Spektrophotometrie ...................................................  111

Chandan Solanki, Dhritiman Saha, Ranjeet Singh Chandan Solanki, Dhritiman Saha, Ranjeet Singh
Moisture dependent physical Feuchtigkeitsabhängige physikalische
properties of dehusked unsplitted Eigenschaften von geschälten, nicht geteilten
pigeon pea ...................................................................  117 Straucherbsen ..............................................................  117

Schriftleitung Dir. u. Prof. Dr. L. Ellerbroek, Bundesinstitut Dr. C. Nguyen-The, INRA, Avignon, Frankreich
für Risikobewertung, Berlin, Deutschland
Prof. Dr. med. vet. Corinna Kehrenberg,
Dr. K. Nöckler, Bundesinstitut für Risiko-
­Justus-Liebig-Universität, Institut für Tier-
LVetD Dr. D. Horn, CVUA-RRW, AöR, bewertung (BfR), Berlin, Deutschland
ärztliche Nahrungsmittel­kunde, Professur
für Lebensmittelsicherheit und Verbraucher- Krefeld, Deutschland
schutz, Frankfurter Straße 92, 35392 Gießen, Prof. Dr. Mehmet Musa Özcan, Selcuk Uni­
Deutschland versity, Faculty of Agriculture, Department
Min.Rat Dr. H. Kobelt, Bundesministerium
of Food Engineering, Konya, Türkei
Prof. Dr. E. Haunhorst, Niedersächsisches für Ernährung, Landwirtschaft und
­Landesamt für Verbraucherschutz und Verbraucherschutz, Bonn, Deutschland
Prof. Dr. M. Prieto Maradona,
­Lebensmittelsicherheit, Postfach 39 49, Universidad de León, Spanien
26029 Oldenburg, Deutschland Prof. Dr. W. Luf, Institut für Milchhygiene,
­Veterinärmedizinische Universität Wien,
Dr. A. Rampp, Bayerisches Landesamt
Editorial Board Österreich
für ­Gesundheit und Lebensmittelsicherheit,
Dr. O. Andreoletti, INRA, Ecole Nationale Oberschleißheim, Deutschland
­Vétérinaire de Toulouse, Frankreich Prof. Dr. A. Martínez López, Instituto de
­Agroquímica y Tecnología de Alimentos Prof. Dr. F. Smulders, Institut für
Dr. U. P. Brunner, Tierärztliche Vereinigung (CSIC), Valencia, Spanien Fleisch­hygiene, Veterinärmedizinische
für Lebensmittelsicherheit und Tiergesund- ­Universität Wien, Österreich
heit, Schaffhausen, Deutschland
Prof. Dr. Dr. h.c. E. Märtlbauer, Lehrstuhl
Prof. Dr. M. Bülte, Institut für Tierärztliche für Hygiene und Technologie der Milch, Prof. Dr. R. Stephan, Institut für
Nahrungsmittelkunde, Justus-Liebig-­ ­Ludwig-Maximilians-Universität München, ­Lebensmittelsicherheit und -hygiene,
Universität Gießen, Deutschland Deutschland Universität Zürich, Schweiz

Zum Titelbild:
Blütenstand und Frucht der Straucherbse (Cajanus cajan) in Westtimor. Foto: Wibowo Djatmiko/wikipedia.de
94 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Arch Lebensmittelhyg 70, 1


) Department of Food Technology, Faculty of Veterinary, Fluminense Federal University,
94–98 (2019) Rio de Janeiro, Brazil; 2) Department of Food and Nutrition, Faculty of Nutrition, Fede-
DOI 10.2376/0003-925X-70-94 ral University of Mato Grosso, Mato Grosso, Brazil; 3) Chemistry Institute, Food Science
­Program, Federal University of Rio de Janeiro, Rio de Janeiro, Brazil; 4) National Reference
© M. & H. Schaper GmbH & Co. Laboratory for Cholera and Enteric Diseases, Oswaldo Cruz Institute, Rio de Janeiro, Brazil
ISSN 0003-925X

Korrespondenzadresse: Occurrence and antimicrobial resistance


pedrop@id.uff.br
­profile of Salmonella isolated from native fish
slaughtered and commercialised in Brazil
Vorkommen und Antibiotikaresistenzprofil von Salmonellen aus in Brasilien
geschlachteten und vermarkteten einheimischen Fischen
Adelino Cunha-Neto¹,²), Pedro Panzenhagen¹), Larrayane Carvalho²), Dália Rodrigues 4),
Carlos Conte-Júnior¹,³), Eduardo Figueiredo²)

Summary Brazil is one of the largest freshwater fish producers worldwide, producing and supply-
ing thousands of tons to the entire population. Fish-carrying Salmonella has been im-
plicated in foodborne disease worldwide. In this context, this study aimed to investigate
the occurrence of Salmonella, identify the serotypes present in fish samples slaught-
ered and commercialized in Brazil, and evaluate the antimicrobial resistance profiles of
the isolated strains. Fifty-two samples of commercialized native fish were evaluated by
classical microbiological culture and by a multiplex PCR. Salmonella was isolated and
detected in three (5.76%) of the 52 analyzed fish samples. We identified the presence
of two uncommon serovars in fish samples: S. Abony and S. Schwarzengrund. This is
a novel worldwide report on the occurrence of S. Abony in freshwater fish. All strains
demonstrated a single resistance to sulphamethoxazole + trimethoprim. This study is
crucial for Salmonella surveillance in the entire country and can provide data to for-
mulate control measures for the effective treatment and prevention of foodborne and
zoonotic pathogens.

Keywords: Salmonella Abony, Salmonella Schwarzengrund, sulphamethoxazole,


trimethoprim, foodborne disease

Zusammenfassung Brasilien ist einer der größten Süßwasserfischproduzenten der Welt und produziert und
liefert Tausende von Tonnen an die gesamte Bevölkerung. Mit Salmonellen belaste-
te Fische waren weltweit an lebensmittelbedingten Erkrankungen beteiligt. Das Ziel
dieser Studie war es, das Auftreten von Salmonellen in Brasilien geschlachteten und
vermarkteten Fischproben zu untersuchen, die Serotypen von zu identifizieren und die
antimikrobiellen Resistenzprofile der isolierten Stämme zu bewerten. 52 kommerziell
gefangene, einheimische Fischproben wurden bakteriologisch-kulturell und mit Multi-
plex-PCR untersucht. Salmonellen wurden in drei (5,76%) der 52 Fischproben nachge-
wiesen. Es wurden zwei seltene Serovare in den Fischproben nachgewiesen: S. Abony
und S. Schwarzengrund. Dies ist ein neuer, weltweiter Bericht über das Vorkommen
von S. Abony in Süßwasserfischen. Alle Stämme zeigten ausschließlich eine Resistenz
gegenüber Sulfamethoxazol + Trimethoprim. Diese Studie ist für die Salmonellenüber-
wachung im gesamten Land von Bedeutung und kann Daten liefern, um Kontrollmaß-
nahmen für die wirksame Behandlung und Prävention von Lebensmittel- und Zoonose-
erregern zu entwickeln.

Schlüsselwörter: Salmonella Abony, Salmonella Schwarzengrund, Sulfamethoxazol,


Trimethoprim, Lebensmittelvergiftung
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 95

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Introduction mum) X Pacu-caranha (Piaractus mesopotamicus), and 6


of the hybrid Tambatinga – Tamabaqui (Colossoma ma-
Infectious diseases transmitted between animals and hu­ cropomum) + Pirapitinga (Piaractus brachypomus) were
mans can be acquired by direct contact with animals, by analyzed for the presence or absence of Salmonella.
environmental exposure, or by eating contaminated food
(European Food Safety et al., 2015). Salmonella enterica Isolation and identification of Salmonella species
is one of the main microorganisms carried by products of Samples were collected and stored aseptically in thermal
animal origin, which are responsible for Foodborne Di­ boxes with ice and transported to the Federal University of
seases (FBD). Also, this microorganism was classified as Mato Grosso for microbiological and molecular analyses.
the seventh major cause of diarrheal diseases in the world The isolation method was based on the protocol recom­
(Havelaar et al., 2015). Salmonella is a gram-negative ba­ mended by the International Standardization Organizati­
cillus belonging to the Enterobacteriaceae family, clas­ on ISO-6579 (ISO, 2017). Briefly, 25 grams of each sample
sified according to the Kauffmann-White scheme in two were inoculated in Buffered Peptone Water (Himedia®,
species, bongori and enterica (Issenhuth-Jeanjean et al., Mumbai, India), incubated at 37°C for 24h, enriched in
2014). The latter is divided into six subspecies: enterica, Rappaport-Vassiliadis Broth (Oxoid®, United Kingdom),
salamae, arizonae, diarizonae, houtenae and indica. The incubated at 42°C for 24h and then in Muller Kauffmann
enterica subspecies presents approximately 2,600 seroty­ Novobiocin Tetrathionate Broth (Himedia®, Mumbai,
pes, of which 99% can cause infections in both animals ­India) at 37°C for 24h, with subsequent plating on Xylo­
and humans (Issenhuth-Jeanjean et al., 2014). se Lysine Deoxycholate Agar (Himedia®, Mumbai, India)
Fish do not represent a Salmonella reservoir, and the and Rambach Agar (Merck, Darmstadt, Germany), in­
presence of this microorganism in this matrix might be in- cubated at 37°C for 24h. Typical colonies were selected,
dicative of fecal contamination in the aquatic environment. purified on Nutrient Agar and subsequently inoculated on
Also, the detection of Salmonella in processed products API 20E (BioMérieux®, Lyon, France). Strains showing a
thus suggests cross-contamination during fish processing typical Salmonella reaction were subjected to serum-ag­
stages, such as evisceration, cut into big steaks, ribs and glutination by the anti-salmonela polyvalent O serum.
fish fillets. Fish-carrying Salmonella has been implicated
in foodborne disease (FDB) cases by 1% in the European Multiplex-PCR
Union according to the World Health Organization. From The strains biochemically identified as Salmonella were
1996 to 2014, fish were ranked as the main FBD-causing inoculated in 10 mL of BHI Broth (Brain Heart Infusion)
microorganism vehicle from imported food in the United and incubated at 35°C for 24h. A 1.5 mL aliquot was then
States of America (Gould et al., 2017). In Brazil so far, centrifuged at 14,000 x g for 5 minutes; the pellet was dis­
there are very few studies regarding the occurrence of Sal- solved in 500 μL Mili-Q water, and heated at 100°C for 10
monella originated from fish. For this reason, health aut­ minutes on a heating plate (BioGPRO, Brazil), then co­
horities were not capable to publish any official national oled at 4°C for 10 minutes. The lysate was then centrifuged
report. This study may be essential to provide data to help at 14,000 x g for 5 minutes, and 200 μL of the supernatant
Brazilian authorities to determine the real occurrence of was removed, maintained in a freezer and subsequently
FBD caused by Salmonella from fish. In addition, investi­ subjected to multiplex PCR. The reaction was performed
gating the antimicrobial resistance profile of Salmonella in a total volume of 25 μL containing 1U Taq Polymerase
strains isolated from fish is necessary since infection by (Invitrogen®), 1x Taq buffer (5 mM KCl Tris-HCl, pH 8.5)
strains carrying resistance to the antimicrobials of choice 1.5 mM MgCl 2, 0.1 mM dNTP’s (Promega®), 0.9 MM pri­
for salmonellosis treatment causes severe impacts on pub­ mer Inv-A (Fratamico, 2003) and 0.4 μM of IE-1 (Wang
lic health. and Yeh, 2002) and Flic-C (Paião et al., 2013) primers (In­
The state of Mato Grosso, Brazil is the third largest na­ vitrogen®). The conditions were based on the study per­
tional freshwater fish producer, producing approximately 49 formed by Paião et al. (2013). The m-PCR assay was per­
thousand tons in 2016. The abundance of fresh water and fa­ formed with an initial denaturation step for 5 minutes at
vorable climate result in higher fish production in this region 95°C, followed by 30 one-minute cycles at 95°C, 1 minute
and justify our interest in evaluating Salmonella occurrence at 58°C, and 30 seconds at 75°C, with a final extension step
in this food matrix. Moreover, the perception of the frequen­ at 72°C for 7 minutes. The PCR products were analyzed by
cy and identification of Salmonella serovars in fish from the­ electrophoresis on 1.5% agarose gels, with TBE buffer (45
se regions are unrevealed and will fill a gap in this unders­ mmol L –1 Tris pH 8.3, 45 mmol L –1 borate, and 2 mmol L –1
tanding. In this context, we aimed to reveal the frequency EDTA) as the running buffer. The gels were then stained
of Salmonella serotypes isolated from Brazilian native fish with Gel Red (Invitrogen®) and photo-documented (Mini­
specimens commercialized in the States of Mato Grosso and Bis-Pro DNT, Bio-Imaging Systems®).
characterize their antimicrobial susceptibility profiles.
Antimicrobial Susceptibility Test
Antimicrobial susceptibility tests were performed with
Materials and methods the disk-diffusion test according to Clinical and Labo­
ratory Standards Institute (CLSI, 2017). Salmonella iso­
Sample collection lates were inoculated into Mueller-Hinton broth (Hime­
The 52 samples of native fish, eviscerated or not, fresh, dia, Pennsylvania, US) and incubated at 37°C overnight
chilled and frozen were randoly colectted from processing (16–18 h). The suspensions were adjusted to the turbidity
facilities and retail trade in municipalities of the state of of 0.5 McFarland scale (about 108 cfu/ml) and streaked
Mato Grosso – Brazil. Forty samples of Tambaqui (Colos- onto Mueller-Hinton agar (Himedia). Antimicrobial disks
soma macropomum), two of Pirarucu (Arapaima gigas), (Cecon®, Brazil) were dispensed onto the surface of the
three of Pintado (Pseudoplastystoma corruscans), one of inoculated agar plates and the plates were incubated at
the hybrid Tambacu – Tamabaqui (Colossoma macropo- 37°C for 16 to 20h. The following 15 antimicrobials were
96 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

tested: ampicillin (10µg), aztreonam (30µg), cephalothin access to these fish aquatic environments and processing
(30µg), cefoxitin (30µg), ceftiofur (30µg), chloramphenicol stages once the samples we collected samples from the
(30µg), flor­fenicol (30µg), streptomycin (300µg), gentami­ retails. Therefore, the accurate information about the
cin (10µg), nalidixic acid (30µg), ciprofloxacin (5µg), enrof­ primary source of Salmonella cannot be clarified in this
loxacin (5µg), tetracycline (300µg), sulfamethoxazole-tri­ study. Besides, the small number of samples analyzed in
methoprim (25µg) and nitrofurantoin (300µg). The results this study might be a concerning issue to determine the
were recorded as susceptible, intermediate or resistant by real occurrence of Salmonella in this specimen. Neverthe­
measuring the inhibition zone diameter according to the less, the detection of samples contaminated with Salmo-
interpretative standards of CLSI in Enterobacteriaceae nella reported here revealed a population health hazard
section M02-07 (2017). The reference strain Escherichia by the consumption of contaminated fish with potential
coli ATCC 25922 was used as the quality control organism microbial pathogens.
and included with each batch of isolates tested. Data from Some authors commonly agreed that the presence of
all quality-control E. coli were within appropriate CLSI Salmonella in eviscerated fish, steaks, fish ribs and salted
quality ranges (CLSI, 2017). cod in the supermarkets of cities in the central, northeast
and southeast regions of Brazil indicates a cross-conta­
Salmonella serotyping mination during the fish processing (Almeida et al., 2004;
Salmonella serotype identification was carried out at the Duarte et al., 2010; Mayrla et al., 2017). One of our pre­
National Reference Laboratory Diagnosis of Enteric Bac­ vious studies has provided strong evidence that Salmo-
teria, Oswaldo Cruz Institute, Oswaldo Cruz Foundation nella contamination in the fish environment and proces­
(FIOCRUZ), by detection of somatic and flagellar anti­ sing industry in Mato Grosso State might have originated
gens, using polyvalent and monovalent antisera, with or from the poultry production waste. In a paper published
without flagellar phase induction (Voss-Rech et al., 2015). at Poultry Science (Cunha-Neto et al., 2018), we demon­
Serotypes were assigned according to the Kauffmann-Whi­ strated that the occurrence of Salmonella spp. in chicken
te scheme (Grimont and François-Xavier, 2007). carcasses in Mato Grosso State was 3.7% (31/850) and the
serovars Abony and Schwarzengrund comprised 25.8%
and 9.7% of the strains isolated, respectively. The Mato
Results Grosso State has one of the highest poultry production in
Brazil, so it is much more evident that poultry production
Out of the 52 fish samples analyzed, 5.76% (3/52) were was the ­primary source of fish contamination than con­
contaminated with Salmonella. They comprised one sam­ versely. Also corroborating with our hypothesis, there is
ple of frozen eviscerated fish, one of rib toothpick and an­ a recently gradual reduction of the most common Salmo-
other one of frozen fillet of Pintado (Table 1). The isolates nella serotypes in the poultry industry such as Enteritidis
­showed the presence of Inv-A amplicon which confirmed (Voss-Rech et al., 2015). This reduction is a consequence
they belong to the genus Salmonella spp. Additionally, of the Salmonella control program of immunization with
we use Multiplex-PCR for fast detection and screening of inactivated vaccines implemented by the Brazilian biose­
serovars Salmonella Typhimurium (ST) and Enteritidis curity authorities. Hence, when some serovars decrease,
(SE). However, the technique did not amplify the genes others less frequently such as Schwarzengrund and Abony
FliC e IE-1 which means there were no ST and SE among find a free niche to grow with less competition resulting
the isolates. Subsequently, serotyping identified two in a faster and better replication rate. It may be possible
strains as S. Abony and S. Schwarzengrund but was not that higher level of these serovars originated from poultry
able to serotype the third strain isolated from the frozen production resulted in transmission towards the fish pro­
fillet of Pintado (Table 1). Antimicrobial resistance was duction.
tested against 15 agents. The
three strains isolated presen­ TABLE 1: Distribution and antibiotic resistance profile of Salmonella serotypes by fish speci-
ted the same unique profile mens and post-processing presentation of the sample from the retail trade in Mato
of antimicrobial resistance Grosso, Brazil.
against sulfamethoxazole-tri­ Name Sample Number of Number of PCR (invA) Serovars AMR
methoprim (SUT). type Samples Isolates (%) detection profile
Tambatinga1 Frozen eviscerated fish   4 0 negative - -
Tambatinga Fish with viscera and refrigerated   2
1
0 negative - -
Discussion
Tambacu2 Fish with viscera and refrigerated   1 0 negative - -
Salmonella presence in fish Tambaqui Frozen eviscerated fish 23 1 (4.35%) positive Schwarzengrund SUT
could be explained due to the (Colossoma macropomum)
contamination of the aquatic Tambaqui Freshly gutted fish   1 0 negative - -
environment by sewage, ani­ (Colossoma macropomum)
mal or human waste; or due Tambaqui Frozen toothpick rib 16 1 (6.25%) positive Abony SUT
to the handling process from (Colossoma macropomum)
harvest areas to the fish mar­ Pintado Frozen fish fillet   3 1 (33.3%) positive Salmonella spp. SUT
ket resulting in deterioration (Pseudoplastystoma corruscans)
of the quality of fish; or may
Pirarucu Frozen fish fillet   2 0 negative - -
also occur by the cross-con­
(Arapaima gigas)
tamination during the pro­
cessing and preparation of Total 52 3 (5.76%)
the fish fillet. As a limitation ¹ Tamabaqui (Colossoma macropomum) x Pirapitinga (Piaractus brachypomus); ² Tamabaqui (Colossoma macropomum) x Pacu-caranha (Piaractus mesopotamicus);
of our study, we did not have SUT (trimethoprim+sulfamethoxazole)
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 97

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Salmonella Schwarzengrund is cited among the top 20 Brazil, from fish found in the retail trade in China, and
serotypes responsible for FBD cases on the Asian con­ fish caught in Nigeria, Africa, and fish purchased from the
tinent (Hendriksen et al., 2011). Also, this serotype was Turkish trade (Ertas Onmaz et al., 2015; Raufu et al., 2014;
­related to cases of FBD in humans involving pet food in Ribeiro et al., 2010; Yang et al., 2015). Brazil is a count­
the United States and Canada (Behravesh et al., 2010; Li ry of continental dimensions, with producers of chicken
et al., 2012). However, the presence of S. Schwarzengrund and swine in all its regions. The possible explanation for
in fish samples is sporadic around the world. Few isolations the common SUT resistance among Salmonella isolates
were reported in Africa and also in rations for fish produ­ should be attributed to the extended use of trimethoprim
ced and imported by Norway (Lunestad et al., 2007; Trao­ + sulfamethoxazole to treat infections by enterobacteria
ré et al., 2015). A 9 year (1990–98) study of the presence and as a growth promoter in the Brazilian animal produc­
of Salmonella in imported and domestic seafood products tion. The indiscriminate use of SUT during the swine and
carried out by the US FDA indicated that Salmonella chicken production in Mato Grosso State might have se­
Schwarzengrund is rarely detected in seafood (Heinitz et lected resistant strains which were transmitted to the fish
al., 2000). In our acknowledge, the presence of S. Schwarz­ production in the region and resulted in the isolation of
engrund in fish have not been reported yet in Brazil. This Salmonella strains with this resistance profile.
evidence also corroborates with the hypothesis discussed
above which the transmission of this serovar is likely co­
ming from poultry toward fish production.
We detected no reports about the occurrence of the S. Conclusion
Abony in fish worldwide. However, this serotype has been
described occurring in surface waters of lakes and rivers This study provides interesting data that are critical for
in northern Greece (Arvanitidou et al., 1997), and in mud, assessing and controlling the risk associated with the pre­
water and shrimp ponds of crustacean in southwest Asia sence of Salmonella in fish. The consumption of fish infec­
(Reilly and Twiddy, 1992). Our study seems to be the first ted with Salmonella can instigate a public health problem,
to report the isolation of S. Abony from frozen fish world­ so we suggest that fish should be appropriately processed
wide. Currently, we have the interest in understanding before consumption. Finally, this study is crucial for Sal-
whether if this Abony serovar origin is much more related monella surveillance in the entire country and can provide
to the hypothesis of transmission from poultry production data to formulate control measures for the effective treat­
or if the Brazilian water of lakes, ponds and rivers harbor ment and prevention of foodborne and zoonotic patho­
Salmonella as reported in northern Greece and southwest. gens.
Shortly, an extensive study of Salmonella in aquaculture
in the various States of Brazil will be performed by our
research group to provide solid answers to confirm our
hypothesis. Also, the Whole Genome Sequencing of the Acknowledgments
S. Abony strain is ongoing to compare with the genome of
others exemplars isolated from poultry origin.
The isolates of Salmonella spp., S. abony and S. The authors would like to thank the Fundação de Ampa­
Schwarzengrund were tested against seventeen antimicro­ ro à Pesquisa do Estado do Rio de Janeiro (process no.
bials. The antimicrobials tested were carefully selected to E-26/201.185/2014, FAPERJ, Brazil) and the Conselho
include the most commonly used veterinary drugs in the Nacional de Desenvolvimento Científico e Tecnológico
Brazilian animal production and the first-choice drugs for (process no. 311422/2016-0, CNPq, Brazil), for financial
the treatment of human enteric infections. The isolates support.
showed a single resistance against trimethoprim + sulfa­
methoxazole – SUT (Table 1). The resistance against this
antimicrobial seems to be very common in Salmonella
Conflict of interest
isolated from fish worldwide. Similar studies also detec­
ted resistance against trimethoprim + sulfamethoxazole
in isolates from fish farmed in tanks in Rio de Janeiro, The authors declare there is no conflict of interest.
98 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

References Issenhuth-Jeanjean S, Roggentin P, Mikoleit M, Guibourdenche


M, de Pinna E, Nair S, Fields P I, Weill F-X (2014): Supplement
Almeida ES de, Sigarini C de O, Valente AM, Andrade PF, Olivei- 2008–2010 (no. 48) to the White-Kauffmann-Le Minor scheme.
ra LAT de, Franco RM, Carvalho JCA do P (2004): Presença Res Microbiol 165: 526–530.
de microrganismos indicadores de condições higiênicas, e de Li X, Bethune LA, Jia Y, Lovell RA, Proescholdt TA, Benz SA,
patógenos em bacalhau saithe (Pollacius virens) salgado seco, Schell TC, Kaplan G, McChesney DG (2012): Surveillance of
comercializado no município de Niterói, Rio de Janeiro, Brasil. Salmonella Prevalence in Animal Feeds and Characterization
Rev Bras Ciênc Veterinária 11: 171–173. of the Salmonella Isolates by Serotyping and Antimicrobial Su­
Arvanitidou M, Papa A, Constantinidis TC, Danielides V, Kat- sceptibility. Foodborne Pathog Dis 9: 692–698.
souyannopoulos V (1997): The occurrence of Listeria spp. and Lunestad BT, Nesse L, Lassen J, Svihus B, Nesbakken T, Fossum
Salmonella spp. in surface waters. Microbiol Res 152: 395–397. K, Rosnes JT, Kruse H, Yazdankhah S (2007): Salmonella
Behravesh CB, Ferraro A, Deasy M, Dato V, Moll M, Sandt C, in fish feed; occurrence and implications for fish and human
Rea NK, Rickert R, Marriott C, Warren K, Urdaneta V, Sa- health in Norway. Aquaculture 265: 1–8.
lehi E, Villamil E, Ayers T, Hoekstra RM, Austin JL, Ostroff Mayrla CSD de O, Yohanna FA, Julyanna de LM, Izabel CR da
S, Williams IT, the Salmonella Schwarzengrund Outbreak S, Daniel OF, Daniela CO (2017): Microbiological evaluation
Investigation T (2010): Human Salmonella Infections Linked and development of quality index method (QIM) scheme for
to Contaminated Dry Dog and Cat Food, 2006-2008. PEDIA­ farmed pintado fish (Pseudoplatystoma corruscans). Afr J
TRICS 126: 477–483. ­Microbiol Res 11: 426–432.
CLSI (2017): Performance Standards for Antimicrobial Suscepti­ Paião FG, Arisitides LGA, Murate LS, Vilas-Bôas GT, Vilas-Boas
bility Testing (27th ed.). Wayne, PA: Clinical and Laboratory LA, Shimokomaki M (2013): Detection of Salmonella spp,
Standards Institute. Salmonella Enteritidis and Typhimurium in naturally infected
Cunha-Neto A da, Carvalho LA, Carvalho RCT, dos Prazeres broiler chickens by a multiplex PCR-based assay. Braz J Mi­
Rodrigues D, Mano SB, Figueiredo EE de S, Conte-Junior crobiol 44: 37–42.
CA (2018): Salmonella isolated from chicken carcasses from a Raufu IA, Lawan FA, Bello HS, Musa AS, Ameh JA, Ambali AG
slaughterhouse in the State of Mato Grosso, Brazil: antibiotic (2014): Occurrence and antimicrobial susceptibility profiles of
resistance profile, serotyping, and characterization by repetiti­ Salmonella serovars from fish in Maiduguri, sub-Saharah, Nige­
ve sequence-based PCR system. Poult Sci 97: 1373–1381. ria. Egypt J Aquat Res 40: 59–63.
Duarte DAM, Ribeiro AR, Vasconcelos AMM, Silva JVD (2010): Reilly PJA, Twiddy DR (1992): Salmonella and Vibrio cholerae in
Ocorrência de Salmonella spp. e staphylococcus coagulase po­ brackishwater cultured tropical prawns. Int J Food Microbiol
sitiva em pescado no nordeste, Brasil. São Paulo 3. 16: 293–301.
Ertas Onmaz N, Abay S, Karadal F, Hizlisoy H, Telli N, Al S Ribeiro RV, Reis EMF, Reis CMF, Freitas-Almeida AC, Rodrigu-
(2015): Occurence and antimicrobial resistance of Staphylococ- es DP (2010): Incidence and antimicrobial resistance of entero­
cus aureus and Salmonella spp. in retail fish samples in Turkey. pathogens isolated from an integrated aquaculture system: In­
Mar Pollut Bull 90: 242–246. cidence of enterobacteria in an aquaculture system. Lett Appl
European Food Safety A, European Centre for Disease P, Control Microbiol 51: 611–618.
(2015): The European Union summary report on trends and Traoré O, Nyholm O, Siitonen A, Bonkoungou IJO, Traoré AS,
sources of zoonoses, zoonotic agents and food-borne outbreaks Barro N, Haukka K (2015): Prevalence and diversity of Salmo-
in 2013: EU summary report on zoonoses, zoonotic agents and nella enterica in water, fish and lettuce in Ouagadougou, Burki­
food-borne outbreaks 2013. EFSA J 13: 3991. na Faso. BMC Microbiol 15.
Fratamico P M (2003): Comparison of culture, polymerase chain Voss-Rech D, Vaz CSL, Alves L, Coldebella A, Leão JA, Rodrigu-
reaction (PCR), TaqMan Salmonella, and Transia Card Salmo- es DP, Back A (2015): A temporal study of Salmonella enterica
nella assays for detection of Salmonella spp. in naturally-conta­ serotypes from broiler farms in Brazil. Poult Sci 94: 433–441.
minated ground chicken, ground turkey, and ground beef. Mol Wang S-J, Yeh D-B (2002): Designing of polymerase chain reac­
Cell Probes 17: 215–221. tion primers for the detection of Salmonella enteritidis in foods
Gould L H, Kline J, Monahan C, Vierk K (2017): Outbreaks of and faecal samples. Lett Appl Microbiol 34: 422–427.
Disease Associated with Food Imported into the United States, Yang X, Wu Q, Zhang J, Huang J, Chen L, Liu S, Yu S, Cai S
1996–20141. Emerg Infect Dis 23: 525–528. (2015): Prevalence, enumeration, and characterization of Sal-
Grimont PAD, François-Xavier W (2007): Antigenic formulae of monella isolated from aquatic food products from retail mar­
the Salmonella serovars (9th ed.). Paris: Institut Pasteur. kets in China. Food Control 57: 308–313.
Havelaar AH, Kirk MD, Torgerson PR, Gibb HJ, Hald T, Lake
RJ, Praet N, Bellinger DC, Silva NR de, Gargouri N, Spey-
broeck N, Cawthorne A, Mathers C, Stein C, Angulo FJ, Dev-
leesschauwer B, Group on behalf of W H O F D B E R (2015):
World Health Organization Global Estimates and Regional
Comparisons of the Burden of Foodborne Disease in 2010.
PLOS Med 12: e1001923.
Address of corresponding author:
Heinitz ML, Ruble RD, Wagner DE, Tatini SR (2000): Incidence
of Salmonella in Fish and Seafood. J Food Prot 63: 579–592. Pedro Panzenhagen, D.V.M., M.Sc., Ph.D.
Hendriksen RS, Vieira AR, Karlsmose S, Lo Fo Wong DMA, Jen- Postdoc researcher
sen AB, Wegener HC, Aarestrup FM (2011): Global Monito­ Department of Food Technology and Inspection –
ring of Salmonella Serovar Distribution from the World Health Veterinary Faculty
Organization Global Foodborne Infections Network Country Fluminense Federal University – 64, Vital Brazil Street
Data Bank: Results of Quality Assured Laboratories from 2001 – Niterói, Rio de Janeiro,
to 2007. Foodborne Pathog Dis 8: 887–900. Brazil
ISO (2017): ISO 6579-1:2017(en), Microbiology of the food chain
pedrop@id.uff.br
– Horizontal method for the detection, enumeration and se­
rotyping of Salmonella – Part 1: Detection of Salmonella spp.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 99

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Arch Lebensmittelhyg 70, Department of Biology, Faculty of Science, Gazi University, Teknikokullar, Ankara 06500,
99–110 (2019) Turkey
DOI 10.2376/0003-925X-70-99

© M. & H. Schaper GmbH & Co. Amino acid decarboxylase activity, biofilm
ISSN 0003-925X formation and antibiotic resistance of
Korrespondenzadresse: gram-negative bacteria isolated from
gundogan@gazi.edu.tr
marine fish, calf meat and chicken
Aminosäuren-Dekarboxylase-Aktivität, Biofilmbildung und
Antibiotikaresistenz gramnegativer Bakterien isoliert aus Meeresfischen,
Kalbs- und Hähnchenfleisch
Neslihan Gundoğan, M. Burcu Külahcı, Ethem Serhat Yavaş

Summary The present study was carried out to test amino acid decarboxylase activity, biofilm for-
mation and antibiotic resistance of 404 Gram-negative bacteria isolated from ­marine
fish, minced veal and chicken. The following isolates were identified: Esherichia coli,
Enterobacter cloacae, Klebsiella oxytoca, Citrobacter freundii, Hafnia alvei, Serratia
marcescens, Pantoea agglomerans, Serratia fanticola, Proteus vulgaris, Citrobacter
amalonaticus, Rahnella aquatilis, Morganella morganii, Escherichia vulneris, Klebsiella
pneumoniae, Providencia rettgeri, Aeromonas hydrophila, Aeromonas caviae, Pseudo-
monas aeruginosa, Pseudomonas putida, Pseudomonas fluorescens, Pseudomonas
oryzihabitans, Acinetobacter lwoffii, Acinetobacter baumannii and Shewanella putre-
faciens. Two E. coli O157 isolates were isolated from minced veal. Decarboxylase acti-
vity was quite common for Gram-negative bacteria and over 70% of isolates could de-
carboxylate at least one amino acid, and lysine was the most frequently decarboxylated
amino acid. According to our results, 60.3% and 62.6% of the Gram-negative bacteria
produced slime and biofilm, respectively. In the antimicrobial susceptibility test, the
isolates were highly resistant to ampicillin, and ß-lactamase inhibitors. Multiple antibio-
tic resistance indices are ranged from 0.29 to 0.64, suggesting exposure to antibiotic
contamination. One hundred forty four (35.6%) out of 404 isolates were identified as
extended spectrum ß-lactamase (ESBL)-producers.

Keywords: Amino acid decarboxylase, antibiotic resistance, biofilm, chicken, fish,


Gram-negative bacteria, minced veal

Zusammenfassung Die vorliegende Studie wurde durchgeführt, um die Aminosäure-Decarboxylase-Akti­


vität, die Biofilmbildung und die Antibiotikaresistenz von 404 gramnegativen Bakterien­
stämmen zu untersuchen, die aus Meeresfischen, Kalbshack- und Hähnchenfleisch
isoliert wurden. Folgende Isolate wurden identifiziert: Escherichia coli, Enterobacter
cloacae, Klebsiella oxytoca, Citrobacter freundii, Hafnia alvei, Serratia marcescens,
Pantoea agglomerans, Serratia fanticola, Proteus vulgaris, Citrobacter amalonati-
­
cus, Rahnella aquatilis, Eitrobacter amalonaticus, Aeromonas hydrophila, Aeromonas
caviae, Pseudomonas aeruginosa, Pseudomonas putida, Pseudomonas fluorescens,
­
Pseudomonas oryzihabitans, Acinetobacter lwoffii, Acinetobacter baumannii und She-
wanella putrefaciens. Zwei E. coli O157-Isolate wurden aus Kalbshackfleisch isoliert.
Die Decarboxylase-Aktivität war bei gramnegativen Bakterien verbreitet, über 70% der
Isolate konnten mindestens eine Aminosäure decarboxylieren. Lysin war die am häu-
figsten decarboxylierte Aminosäure. Nach unseren Ergebnissen produzierten 60,3%
und 62,6% der gramnegativen Bakterien Schleim bzw. Biofilm. Im antimikrobiellen
Empfindlichkeitstest waren die Isolate gegen Ampicillin- und ß-Lactamaseinhibitoren
hochresistent. Mehrere Antibiotikaresistenzindizes lagen im Bereich von 0,29 bis 0,64,
was auf eine Exposition gegenüber Antibiotikakontamination hindeutete. 144 (35,6%)
von 404 Isolaten wurden als Extended-Spectrum Beta-Lactamasen (ESBL) Produzenten
mit erweitertem Spektrum identifiziert.

Schlüsselwörter: Aminosäuredecarboxylase, Antibiotikaresistenz, Biofilm, Fisch,


gramnegative Bakterien, Kalbshackfleisch
100 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Introduction nems and aminoglycosides, is frequently observed among


Enterobacteriaceae, Pseudomonas spp., Acinetobacter
The microbiological content of raw meat purchased by spp. and Shewanella spp. (Matyar et al., 2008). Most Aero-
consumers depends mostly on the slaughter process, sani­ monas spp. strains are typically resistant to penicillin, am­
tation during processing and packaging, maintenance of picillin, carbenicillin, methicillin, erythromycin, clinda­
adequate cold chain storage from the processing to retail mycin and vancomycin (Arslan and Kucuksarı, 2015). The
and to the consumer and finally sanitation during hand­ presence of extended spectrum beta lactamase (ESBL)
ling at the retail end. Enterobacteriaceae are a large family and metallo beta lactamase (MBLS) genes among bacteri­
of Gram-negative bacteria that includes a number of im­ al communities is of great concern, as they confer resistan­
portant foodborne pathogens such as Salmonella, Yersinia ce to beta-lactam antibiotics as well as aminoglycosides,
ente­rocolitica, pathogenic Escherichia coli (including E. fluoroquinolones, and trimethoprim-sulfamethoxazole.
coli O157:H7), Shigella spp. and Cronobacter spp. Coli­ Antibiotic resistance has turned into a global public health
form bacteria within this family namely Enterobacter, problem in all over the world. Excessive or incorrect use
Klebsiella, Citrobacter, Serratia and Escherichia are con­ of antimicrobials in human and veterinary medicine and
sidered as indicator organisms to define sanitary quality without proper prescription are mediated on the develop­
of food and water (Tekiner and Özpinar, 2016). Klebsiel- ment of antibiotic resistance. Antibiotics are extensively
la spp, Serratia spp. and Citrobacter spp. are regarded as used in Turkey and this situation remains uncontrolled at
opportunistic pathogens, especially in clinical settings. both the community and hospital levels. Turkey has been
Enterobacteriaceae species are inhabitants of soil, water, identified as the country with the highest antibiotic use out
plants and the intestinal tract of a wide range animals. That of 42 countries in the broader European region (Kuzucu et
means they could enter into the food chain and contribute al. 2011; Karabay et al. 2016).
to disease and spoilage. E. coli, Citrobacter, Enterobacter, The aim of this study was to identify Gram-negative
Klebsiella and Serratia are the most prevalent bacteria iso­ bacteria that were isolated from marine fish, raw minced
lated from beef, pork and poultry (Schwaiger et al., 2012; veal and chicken breasts, and to determine amino acid de­
Kilonzo-Nthenge et al., 2013;Wong et al., 2015). Bacteria carboxylase activity, slime formation, biofilm formation,
in the genera Citrobacter, Enterobacter, Klebsiella, Ser- antibiotic resistance and extended spectrum beta lactama­
ratia, Shewanella, Pseudomonas, Photobacterium, Aero- se (ESBL) production that may influence food safety.
monas, Acinetobacter, Morganella and Vibrio have been
found in the main spoilage flora of fresh seafood products
(Chakravarty et al., 2015). Materials and Methods
The amino acid decarboxylase activity of numerous
bacteria generates high level of biogenic amines in conta­ Sample collection
minated foods. The decarboxylation of histidine, tyrosine, Ninety samples of marine fish (Black sea anchovy, Eng-
lysine and ornithine yields to histamine, tyramine, cada­ raulis encrasicolus), 90 samples of raw minced veal and 90
verine and putrescine, respectively, which are the food samples of chicken breasts were purchased from different
amines. The formation of biogenic amines in food is im­ fish markets and butcher shops in Ankara, Turkey, bet­
portant for health associated with food spoilage. Amino ween June 2012 and December 2013. Samples were collec­
acid decarboxylases are present in many microorganisms ted in sterile polyethylene packs, placed on ice, immedia­
of food concern. They have been found in genera of the tely transported to the laboratory, and processed within 2
family Enterobacteriaceae, such as Citrobacter, Klebsiel- h after collection.
la, Escherichia, Proteus, Salmonella and Shigella and
Micrococcaceae, such as Staphylococcus, Micrococcus Sampling, Isolation and Identification
and Kocuria (Marino et al., 2000). Furthermore, species Twenty-five grams of each food sample was homogenized
of the genera Bacillus, Pseudomonas, Photobacterium, as with 225 ml of sterile buffered peptone water (Merck,
well as many lactic acid bacteria (LAB) belonging to the Darmstadt, Germany) and homogenized for 2 min using
genera Lactobacillus, ­E nterococcus, Carnobacterium, Pe- a stomacher (Lab. Lemco 400. Of each prepared sample,
diococcus, Lactococcus and Leuconostoc are able to de­ 0.1 ml was evenly spread on MacConkey agar (Merck). The
carboxylate amino acids (Zaman et al., 2010; Tembhurne inoculated media was incubated aerobically at 37 °C for
et al., 2013). 24–48 h. After incubation, at least five red (lactose posi­
The formation of bacterial biofilm on the surface of tive) and colorless (lactose negative) colonies were picked
food processing equipment increases the threat of a cross­ from the plates and restreaked on fresh MacConkey agar
over contamination of the product. This can have an effect to purify. Pure isolates were characterised by colony and
on the quality and safety of the final product, especially cell morphology, Gram staining, oxidase and catalase ac­
if pathogenic bacteria or spoilage organisms become do­ tivity, OF glucose and gelatin liquefaction tests and indol
minant in the biofilm. Several types of food-contamina­ reaction (Matyar et al., 2008). Isolates were then identified
ted-bacteria are found to be biofilm-forming, including using the BBL® Crystal™ ENF system (Becton Dickinson
L. monocytogenes, Vibrio spp., Salmonella spp., Bacillus and Company, Maryland, USA).
spp., Aeromonas spp., and Pseudomonas spp. It was noted For the determination of E. coli O157:H7 serotype, 25 g
that the presence of Pseudomonas spp. would significantly of each sample was homogenised in tryptone soya broth
enhance the colonization of L. monocytogenes on stain­ (Oxoid) supplemented with novobiocin (20 mg/L) and
less steel (Mahapatra et al., 2015). ­incubated at 37 °C for 24 h. The enrichment samples were
Food-related and environmental bacteria resistant to streaked onto sorbitol MacConkey agar (Merck, Darm­
antibiotics represent a major threat to humans, because stadt, Germany) plates supplemented with cefexime (0.5
they can act as a reservoir for the maintenance and spread mg/L) and potassium tellurite (2.5 mg/L), and incubated at
of antibiotic resistance genes. Multidrug-resistance, inclu­ 37 °C for 24 h. After incubation, the plates were checked
ding resistance to ß-lactams, fluoroquinolones, carbape­ for the presence of sorbitol-negative, colourless colonies
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 101

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

1–2 mm in diameter. Subsequently, these presumptive co­ pneumoniae (ATCC 700603) sensitive to all antimicrobial
lonies were confirmed serologically using an E. coli O157 drugs being tested was used as a control. For each isolate, a
latex agglutination test (Oxoid) and H7 antisera (Denka standard inoculum was prepared by adjusting the bacterial
SeikenCo., Tokyo, Japan), as described by the manufactu­ suspension in Lactose Broth (LB; Merck) to a final optical
rers (AOAC, 1998). density of 0.5 McFarland units.

Amino Acid Decarboxylase Activity Detection of ESBL by Double Disk Synergy Test
Amino acid decarboxylase activity of the isolates was ESBL was detected by a double disk synergy technique in
­qualitatively assessed by observing their ability to grow which an augmentin disk (20 µg of amoxicillin and 10 µg
on modified Niven agar (0.5 % tryptone, 0.5 % yeast ex­ of clavulanic acid) was placed in the center of a plate, and
tract, 0.5 % Na CI, 0.1 % CaCO3, 3 % agar, and 0.006% cefotaxime (30 µg), ceftazidime (30 µg), cefepime (30 µg),
bromcresol purple, pH 5.3) containing 1 % of each precur­ and aztreonam (30 µg) disks were placed 30 mm (center to
sor amino acid; L-histidine hydrochloride, L-lysine hydo­ center) from the augmentin disk. The enhancement of the
chloride, L-ornithine hydochloride and L-tyrosine hydro­ zone of inhibition of any one of the four drug disks toward
chloride (Sigma, St. Louis, MO) (Niven et al., 1981). The the disk containing clavulanic acid suggested the presen­
inoculated plates were incubated at 37 °C for 24–72 h. A ce of ESBLs. Escherichia coli NCTC 10418 was used as
colour change from yellow to purple indicated a positive an ESBL-negative control, and K. pneumoniae ATCC
reaction, i. e. that the respective amino acid decarboxylase 700603 was used as an ESBL-positive control (Gundogan
was present. et al., 2011).

Slime Formation Multiple Antibiotic Resistance (MAR) Index


Production of slime from all isolates was studied by culti­ The MAR index was calculated as the ratio (a/b) between
vation of the isolates on Congo Red Agar (CRA). CRA the number of antibiotics to which the isolate was resistant
plates (sucrose 50 g (Sigma), brain heart infusion broth (a) and the total number of antibiotics tested (b). A MAR
37 g (Oxoid, Basingstoke, Hampshire, UK), agar 10 g, index value >0.2 is observed when the isolates are exposed
­congo red 0.8 g (Sigma, St. Louis, MO), distilled water to high risk sources of human or animal contamination,
1000 ml) were incubated at 37 °C for 24 h. After incubati­ where antibiotics use is common. In contrast, a MAR in­
on, bright black colonies were established as slime positive dex value ≤0.2 is observed when antibiotics are seldom or
(Gundogan et al., 2013). never used (Matyar et al., 2014).

Biofilm-forming ability Statistical Analysis


Biofilm-forming ability was measured by determination of Chi-square (x 2) tests were used to determine statistically
adhesion to polystyrene microtiter plates according to the significant differences in the prevalence of Gram-negative
protocol of Christensen et al., (1985). Briefly, isolates were bacteria in food samples. P values of less than 0.05 were
inoculated in Tryptic Soy Broth (TSB; Oxoid) and incu­ considered significant.
bated for 18 h at 37 °C. Afterwards a 1:40 dilution in TSB
supplemented with 0.25 % glucose, 200 µl of each dilution
was distributed in flat-bottom 96-well polystyrene plates Results and discussion
(Oxyvital, Hong Kong, China). The plates were incubated
for 18 h at 37 °C, washed 3x with phosphate buffer saline Prevalence of Gram-negative Bacteria in Marine fish,
(PBS), pH 7.0, air-dried for 1 h at 60 °C and stained with Minced veal and Chicken
0.25 % crystal violet for 1 min. After washing, optical den­ Raw meat can be contaminated with a variety of micro­
sity (OD) of each well content was measured at 570 nm organisms, including those capable of spoiling the product
using an automated microplate reader (Thermo Scientific during storage, and certain foodborne pathogens (Schwai­
Multiskan Microplate Reader GIO de Vita E C; Rome, ger et al., 2012; Kilonzo-Nthenge et al., 2013; Wong et al.,
Italy). We defined the cut-off OD (ODc) for the microti­ 2015). In this study, 404 Gram-negative bacteria isolated
ter-plate test as three standard deviations above the mean and identified from 270 samples of marine fish, minced
OD of the negative control. The adherence ability of the veal, and chicken breasts were investigated for amino
tested strains was classified into four categories based on acid decarboxylase activity, slime and biofilm production,
the OD: “OD≤ODc: non–adherent, ODc<OD≤2XODc: ESBL production and antibiotic resistance. We found that
weakly adherent, 2XODc<OD≤4XODc: moderately ad­ there was a significant difference in the Gram-negative
herent, 4XODc<OD: strongly adherent”. All tests were bacterial contamination levels among fish, minced meat
carried out three times and the results were averaged. and chicken, with the highest contamination level seen in
fish (P <0.05).
Antibiotic susceptibility testing Enterobacteriaceae are useful marker for the identifi­
Susceptibility testing of the isolated organisms was done by cation of either fecal contamination to raw meats during
a disk diffusion method using the Kirby-Bauer technique the slaughter processs or secondary contamination along
and following the recommendations of the Clinical and La­ the processing chain (Tekiner and Özpinar, 2016). In this
boratory Standards Institute (CLSI, 2006). All disks were study, out of the 404 Gram-negative bacterial isolates, 309
obtained from Bioanalyse (Ankara, Turkey): amikacin (30 (76.5%) isolates were belonging to the family Enterobacte-
µg), gentamicin (30), imipenem (10 µg), ertapenem (10 µg), riaceae. As shown in Table 1, the most predominant speci­
ampicillin (10 µg), amoxicillin-clavulanic acid (20 and 10 es present in the isolates was Escherichia coli (53 isolates),
µg ), ampicillin-sulbactam (each 10 µg), piperacillin-tazo­ followed by Enterobacter cloacae (45 isolates), Klebsiella
bactam (100 and 10 µg), ceftazidime (30), cefotaxime (30), oxytoca (35 isolates), Citrobacter freundii (26 isolates),
cefepime (30 µg), ceftriaxone (30 μg), ­ciprofloxacin (5 µg), Hafnia alvei (24 isolates), Serratia marcescens (24 isola­
and aztreonam (30 µg). A standard r­ eference strain of K. tes), Pantoea agglomerans (15 isolates), Serratia fantico-
102 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

TABLE 1: Gram-negative bacteria isolated from fish, minced veal and coli O157 was detected in 7.6%–8.8% of ground
chicken. beef, minced meat and ground meat (Sarimehme­
Bacteria Fish Meat Chicken Number of % toglu et al., 2009; Cetin et al., 2010; Temelli et al.,
total isolates 2012). These values are higher than that obtained
from this study. The intestinal tract of cattle was
Escherichia coli 25 10 18 53 13.1
reported as the principal reservoir of E. coli O157
Enterobacter cloacae 13 14 18 45 11.1 (Gundogan and Avci, 2014). Therefore, preventing
Aeromonas hydrophila 43 – – 43 10.6 faecal material from contaminating meat is an im­
Klebsiella oxytoca  7 16 12 35  8.6
portant step in reducing the prevalence of E. coli
O157 in raw meat and its products.
Citrobacter freundii 10  6 10 26  6.4
In our country, the largest proportion of fish yi­
Hafnia alvei  8  8  8 24  5.9 elds were from the Black Sea where anchovy (Eng-
Serratia marcescens  4 12  8 24  5.9 raulis encrasicolus) was the dominant fish in caught
with about 51 percent of total between 2003 and
Pantoea agglomerans  4  5  6 15  3.7
2012 (Turkish Statistical Institute, 2013). Black sea
Serratia fanticola  4  2  6 12  2.9 anchovy is most commonly consumed fish species
Proteus vulgaris  4  6  2 12  2.9 in Turkey due to its cheaper price compared to ot­
Citrobacter amalonaticus  5  7 – 12  2.9 her fish species (Gucu et al., 2017). The following 95
Gram-negative isolates Aeromonas hydrophila (43
Rahnella aquatilis  3  4  5 12  2.9
isolates), Aeromonas caviae (6 isolates), Pseudomo-
Pseudomonas aeruginosa 11 – – 11   2.7 nas aeruginosa (11 isolates), Pseudomonas putida
Morganella morganii  9  1  1 11  2.7 (6 isolates), Pseudomonas fluorescens (6 isolates),
Escherichia vulneris  6  2  2 10  2.4 Pseudomonas oryzihabitans (5 isolates), Acineto-
bacter lwoffii (6 isolates), Acinetobacter baumannii
Klebsiella pneumoniae  4  2  3  9  2.2
(6 isolates), and Shewanella putrefaciens (6 isola­
Providencia rettgeri  5  3  1  9  2.2 tes) have been isolated from the fish only (Table 1).
Shewanella putrefaciens  6 – –  6  1.4 In the present study, Aeromonas hydro­phila were
found the most frequently comparing with species
Pseudomonas putida  6 – –  6  1.4
from Enterobacteriaceae, Pseudomonas and Aci-
Pseudomonas fluorescens  6 – –  6  1.4 netobacter (P <0.05). Similarly, Grigoryan et al.
Aeromonas caviae  6 – –  6  1.4 (2013) reported that although bacteria from Enter-
Acinetobacter lwoffii  6 – –  6  1.4 obacter, Klebsiella, Citrobacter, Serratia, Pseudo-
monas, Alcaligenes and Vibrio genera have been
Acinetobacter baumannii  6 – –  6  1.4
isolated and identified from rainbow trout, bacteria
Pseudomonas oryzihabitans  5 – –  5  1.2 form genera Aeromonas have been found in prevai­
Total 206 98 100 404 ling quantities (96%). Aeromonas spp. have emer­
ged as an important human pathogens because of
la (12 isolates), Proteus vulgaris (12 isolates), Citrobacter diarrhea related to foodborne outbreaks. These bacteria
amalonaticus (12 isolates), Rahnella aquatilis (12 isolates), are also predominantly pathogenic to aquatic animals,
Morganella morganii (11 isolates), Escherichia vulneris especially fish (Arslan and Küçüksari, 2015).
(10 isolates), Klebsiella pneumoniae (9 isolates), and Pro- Compared to our results, higher contamination ra­
videncia rettgeri (9 isolates). Similar to our results, a large tes of fish with Aeromonas spp. were reported as 44.1%,
number of Enterobacteriaceae species isolated from fish, 42.8% and 77.9%, respectively by Arslan and Kucuksari
red meat and chicken meat products have been reported (2015), Yucel and Erdogan (2010) and Yucel et al. (2005)
in Turkey (Matyar 2007; Ondes and Ozpinar, 2016; Teki­ in Turkey. In Malaysia A. hydrophila isolates were isolated
ner and Ozpinar, 2016), In USA (Kilonzo-Nthenge et al., from 11.5% of the fish (Radu et al., 2003), while in Brazil,
2013), in Germany (Gwida et al., 2014), in Nepal (Shrestha the percentage of these bacteria was 50% (Da Silva et al.,
et al. 2017), and in China (Ye et al., 2018). 2010). There may be several reasons for these variations,
In this study, the important foodborne pathogens such such as differences in geographic location and season and
as Salmonella, Yersinia enterocolitica, Shigella spp. and difference in fish species studied. The fish samples were
Cronobacter spp. were not detected in any of the food obtained from several sources and storage conditions
samples. However; two organisms of concern were E. coli which bring about different results. According to Mol and
and K. pneumoniae, opportunistic pathogens of humans Saglam (2004), fish boxes are generally laid on the floor,
and animals responsible for a wide range of infections, and this is a major cause of bacterial contamination in Tur­
such as urinary tract infections, pneumoniae, wound in­ kish fish markets. Furthermore, the transportation of fish
fections and septicemia (Gundogan et al., 2011; Gundo­ from seaside cities to Ankara will take at least 5 hours.
gan and Avci, 2014). The presence of E. coli in foods is a During the transportation, sprinkling of fish with conta­
matter of concern because some strains may be pathgenic minated water, packing it with contaminated ice, coupled
(Gundogan and Avci, 2014).We isolated two (2.2%) E. coli with unhygienic handling may explain the high prevalence
O157 isolates from 90 minced veal. E. coli O157:H7 sero­ of Gram-negative bacteria in fish in the markets.
types, identified as enterohaemorrhagic E. coli (EHEC)
and grouped as verotoxin-producing E. coli (VTEC), are Decarboxylase Activity of Gram-negative Bacteria
recognised as the primary cause of haemorrhagic colitis The detection of bacteria possessing amino acid decar­
(HC) and the diarrhoea-associated form of haemoly­ boxylase activity is of main importance to assess the risk
tic-uremic syndrome (HUS) (Gundogan and Avci, 2014). of foods to contain biogenic amine and to prevent their ac­
In some studies conducted in different cities of Turkey, E. cumulation in food products. Formation of biogenic ami­
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 103

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

nes in foods is important for health and also for unfavor­ trefaciens and some isolates of Acinetobacter spp. In this
able effects on flavor (Marino et al., 2000; Maifreni et al., study, only few isolates of the Enterobacteriaceae species
2013). Therefore, we studied 404 Gram-negative bacteria could decarboxylate histidine-ornithine. Isolates from Ae-
isolates for their capability to decarboxylate tyrosine, or­ romonas spp. and Pseudomonas spp. were able to decar­
nithine, lysine and histidine (Table 2). Decarboxylase ac­ boxylate histidine-ornithine or lysine-tyrosine (Table 2).
tivity was quite common for Gram-negative bacteria and Histamine is considered to be the most active amine and
over 70% of the isolates could decarboxylate at least one is related to almost all food amines poisoning incidences.
amino acid. Lysine was the most frequently decarboxyla­ However, the occurrence of putrescine and cadaverine
ted amino acid, followed by ornithine, tyrosine and histidi­ which may enhance the toxicity of histamine should not be
ne. This is an important food safety concern, considering underestimated (Zaman et al., 2010). The result observed
the these isolates were potential cadaverine, putrescine, is that, even if the microorganisms had the capability to
tyramine and histamine producers in fish, veal and chi­ produce more than one decarboxylase, the decarboxyla­
cken. Durlu-Ozkaya et al. (2001) reported that conversion ting activity is highest towards lysine and ornithine with
of ornithine, lysine, tyrosine and histidine respectively, to consequent production of cadaverine and putrescine, re­
putrescine, cadaverine, tyramine and histamine was found spectively which was in ­agreement with the results repor­
in ≤ 82% of Enterobacteriaceae isolates. Authors indica­ ted by Marino et al. (2000) and Maifreni et al. (2013).
ted that high levels of these biogenic amines in ground Previous studies have also revealed the capability of
meat and meat products can be an indicator of the hygie­ Gram-negative bacteria to produce decarboxylase en­
nic quality of meats. Marino et al. (2000) and Maifreni et zymes was variable. Durlu-Ozkaya et al. (2001) showed
al. (2013) reported that most of the Enterobacteriaceae that E. coli and M. morganii possess both histidine and
species were shown to have the ability to decarboxylate lysine decarboxylases, P. mirabilis has both histidine and
mainly lysine and ornithine, which were consistent with ornithine decarboxylases and Enterobacter spp. have both
our results. lysine and ornithine decarboxylases. Ozogul and Ozogul
The results obtained from this study also indicated that (2005) reported that E. coli, K. oxytoca, H. alvei and P.
depends on the isolates, the ability of microorganisms to vulgaris possess histidine decarboxylase activity while
decarboxylate amino acids were highly variable. The ab­ Pseudomonas spp., K. oxytoca and H. alvei could de­
ility to decarboxylate the two amino acids (histidine-tyro­ carboxylate both histidine and lysine. In their study, all
sine or lysine-ornithine or tyrosine-ornithine) was present Gram-negative rods decarboxylated ornithine but none of
in all isolates of Enterobacteriaceae, in Shewanella pu- the Acinetobacter spp. isolates had lysine, ornithine and

TABLE 2: Amino acid decarboxylase activity of gram-negatvie bacteria.


Species his lys tyr orn his-tyr his-orn lys-tyr lys-orn tyr-orn his-lys his-tyr lys-tyr- his-lys- Total
tyr orn orn tyr-orn
E. coli (n=53) 7 12 11 7 3 2 2 1 3 1 2 – 1 52
E. cloacae (n=45) 1 18 9 5 3 – 1 1 2 1 – – – 41
A. hydrophila (n=43) 1 1 1 1 – 6 2 – – 1 – 2 1 16
K. oxytoca (n=35) 2 9 4 8 2 – 1 1 1 1 2 – – 31
C. freundii (n=26) 7 1 2 5 1 – – 1 1 – – 1 1 20
H. alvei (n=24) 1 2 2 8 1 1 1 2 1 – 1 1 1 22
S. marcescens (n=24) 1 7 2 2 2 1 1 1 1 – – – – 18
P. agglomerans (n=15) – – 1 – 1 – – 1 1 – – – – 4
S. fanticola (n=12) – – 1 1 1 1 – 1 1 – – – – 6
P. vulgaris (n=12) 1 – 1 – 1 1 – 1 1 – – – – 6
C. amalonaticus (n=12) 1 1 – 1 1 1 – 1 1 – – – – 7
R. aquatilis (n=12) – 1 – – 1 – – 1 1 – – – – 4
P. aeruginosa (n=11) 1 – – 1 – 1 2 – – 1 – – 1 7
M. morganii (n=11) 2 – 1 – 1 1 – 1 1 – – 1 1 9
E. vulneris (n=10) – 1 – – 1 – – 1 1 – – – – 4
K. pneumoniae (n=9) 1 1 1 1 1 1 – 1 1 – – – – 8
P. rettgeri (n=9) – – – – 1 – – 1 1 – – – – 3
S. putrefaciens (n=6) 1 – – 1 1 – – 1 1 – – – 1 6
P. putida (n=6) – 1 1 – – 1 1 – – 1 – – 1 6
P. fluorescens (n=6) 1 1 – – – 1 1 – – 1 – – 1 6
A. caviae (n=6) – 1 1 – – 1 1 – – 1 – – – 5
A. lwoffii (n=6) – 1 – – – – – 2 3 – – – – 6
A. baumannii (n=6) – 1 – – – – – 2 2 – – – – 5
P. oryzhabitans (n=5) 1 1 – – – 1 1 – – 1 – – – 5
Total 29 60 38 41 22 20 14 21 24 9 5 5 9 297
his: histidine; lys: lysine; orn: ornithine; tyr: tyrosin
104 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

histidine decarboxylase activity. However, in this study, who found that slime activity in 45.2% of A. caviae isolates
Acinetobacter spp. isolates have the capacity to decarbo­ but none of the A. hydrophila isolates produced slime whi­
xylate these amino acids. Tembhurne et al. (2013) reported le the A. hydrophila isolates in this study had slime activi­
that 63 out of 202 Enterobacteriaceae isolates from Indian ty. Some of the previous studies have shown that the slime/
Mackerel gave positive results in the Niven medium, indi­ biofilm formation is largely dependent on the origin of the
cating histidine decarboxylase activity. In the same study, isolates as well as temperature and time, and associated
Shewanella putrefaciens was found as non-Enterobacte- with nutrient content of the growth medium (Orozova et
rial histamine-pro­ducing bacteria, which was similar to al., 2009; Reynisson et al. 2009). Meanwhile we found that
our result. Other study showed histidine decarboxylase none of the P. agglomerans, S. fanticola, C. amalonati-
activity in 84% of 152 Gram-negative bacteria from fish cus, R. aquatilis, E. vulneris and P. rettgeri isolates had
using modified Niven method whereas E. coli isolates were slime activity. However, E. coli (66%), E. cloacae (62.2%),
detected as non-histamine producers (Bjornsdottir et al., K. oxytoca (54.3%), C. freundii (53.8%), H. alvei (66.7%),
2009). This result is not in agreement with the result of our S. marcescens (54.2%), P. vulgaris (66.7%) M. morganii
study. We observed that few isolates from Klebsiella spp., (81.8%) and K. pneumoniae (77.8%) isolates had a great
Pseudomonas spp., Aeromonas spp., E. coli, C. freundii, tendency to produce slime. Furthermore, except S. marce-
H. alvei, M. morganii, and S. putrefaciens also decarbo­ scens, S. fanticola, R. aquatilis, M. morganii and P. rettgeri
xylated three or four amino acids. Our results provide isolates, remeaning isolates were characterised by modera­
new information regarding the decarboxylase activity of te to strong biofilm-forming ability (Table 3). The species
S. fanticola, R. aquatilis, E. vulneris, K. pneumoniae, P. in the biofilm originated from the all samples studied and
rettgeri and A. baumannii (Table 2). Differences between therefore be expected to play a role in biofilm formation in
the reults may be due to the difference methods that were food contact surfaces. This is not suprising because simi­
used for the detection of amino acid decarboxylase-pro­ lar results have already been reported by several authors.
ducing bacteria. Bagge et al. (2001) found that S. putrefaciens, a fish spoi­
Nevertheless, in this study, it is observed that a high lage bacterium, is able to attach and form biofilms on food
proportion of Enterobacteriaceae, Aeromonas spp., and processing surfaces. Møretrø et al. (2013) showed that
Pseudomonas spp. isolates had amino acid decarboxylase Enterobacter spp., Pseudomonas spp., Citrobacter spp.,
activity. The presence of microbial populations with de­ Acinetobacter spp., Serratia spp. and Listeria monocyto-
carboxylase activity and availability of free amino acids genes isolates from meat abattoir process surfaces were
are considered the most important factors affecting the strong biofilm producers. Also Liu et al. (2015) observed
production of biogenic amines in raw and proces­
sed foods (Ozogul and Ozogul 2005). The results TABLE 3: Slime formation and biofilm-forming ability of gram-negative
obtained regarding isolation of these organisms bacteria.
from raw meats highlight the need to improve hy­ Bacteria Slime Biofilm-forming ability
gienic practices to prevent further proliferation of formation Absent Weak Moderate Strong
decarboxylase positive microflora on fish, veal and E. coli 53/35 26 – 15 12
chicken.
E. cloacae 45/28 19 20  3  3
A. hydrophila 43/43 21 – 13  9
Slime and biofilm formation of
Gram-negative bacteria K. oxytoca 35/19  1 17  9  8
Microorganisms in food are able to form biofilms C. freundii 26/14 12  3 10  1
on the food and food processing equipment surfa­ H. alvei 24/16  5 – 12  7
ces. Biofilms can also be transferred onto food, such
as fish, meat and poultry, when these foods come in S. marcescens 24/13  6  8 10 –
contact with contaminated surfaces (Silagyi et al., P. agglomerans 15/-  3 –  9  3
2009). The present study showed that slime and bio­ S. fanticola 12/- – 12 – –
film-forming Gram-negative bacteria contaminate
P. vulgaris 12/8  7 –  4  1
fish, meat and chicken (Table 3). According to our
results, 244 (60.4%) and 253 (62.6%) out of the 404 C. amalonaticus 12/-  
4 –  
8 –
Gram-negative bacteria had slime and biofim for­ R. aquatilis 12/- 10 – –  2
mation, respectively. All of the Pseudomonas spp., P. aeruginosa 11/11  3  3  4  1
Aeromonas spp., Acinetobacter spp., and S. putre-
M. morganii 11/9 – 11 – –
faciens isolates produced slime.
A recent study showed that slime-producing E. vulneris 10/-  5  3  1  1
Pseudomonas spp. isolates were the most abundant K. pneumoniae 9/7  4  3  1  1
bacteria on slaughterhouse surfaces after cleaning P. rettgeri 9/-  
9 – – –
and sanitizing treatments (Bakhtiary et al. 2016).
The supportive activity of Pseudomonas isolates S. putrefaciens 6/6  3  1  1  1
for the attachment and biofilm formation of S. au- P. putida 6/6  3  1  1  1
reus and L. monocytogenes has also been reported P. fluorescens 6/6  3 –  2  1
(Bakhtiary et al. 2016). On the other hand, Arslan
A. caviae 6/6 –  3  2  1
et al. (2011) showed that Pseudomonas spp. isolates
did not produce slime. Furthermore, Orozova et al. A. lwoffii 6/6  4 –  1  1
(2009) reported that Aeromonas isolates were ne­ A. baumannii 6/6  1  1  3  1
gative for slime production. The results of our study P. oryzhabitans 5/5  2 –  2  1
do not confirm these findings. Also, our results only
Total 404/244 151 86 111 56
partially agree with Arslan and Kucuksari (2015)
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 105

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

that P. lundensis isolated from spoiled Chinese pork had The results for 404 isolates that were tested against 14
a high capacity to produce biofilms and was able to adhere antimicrobial agents are presented in Table 4.
to the contact surfaces. According to Bitrian et al.(2012), Aminoglycosides are broad-spectrum antibiotics of
Yaron and Römling (2014) and Bakhtiary et al. (2016), in high potency that have been used for the treatment of seri­
meat-processing environments, all surfaces and materials ous Gram-negative infections (Arslan et al., 2011; Gundo­
are likely to be colonized by microorganisms if sanitation gan et al., 2011). According to the results reported here, all
procedures are inadequate and/or insufficient. Moreo­ the isolates of Enterobacteriaceae, Aeromonas spp., Pseu-
ver the attachment properties and the biofilm formation domanas spp., Acinetobacter spp., and S. putrefaciens were
of bacteria on surfaces facilitate cross-contamination. If susceptible to aminoglycosides (amikacin and gentamicin),
pathogens are present, consumption of the contaminated which is in agreement with previously published data for
foods may pose a health risk (Reynisson et al., 2009). In Enterobacteriaceae, Aeromonas spp., Pseudomonas spp.,
addition, biofilm formation creates major problems in the and A. baumanii isolates isolated from food (Arslan et al.,
food industry, because biofilms represent an important 2011; Schwaiger et al., 2012; Kilonzo-Nthenge et al., 2013).
source of contamination, increased food spoilage and can Resistance to amikacin and gentamicin was most common
support microbial growth. Therefore, for quality and safe­ in clinical isolates of Aeromonas spp., Acinetobacter spp.,
ty of foods, preventive and control strategies like hygienic Pseudomonas spp. and Enterobacteriaceae (Kucukates,
plant lay-out and design of equipment, choice of materials, 2005). However, Chakravarty et al. (2015) reported that
correct use and selection of detergents and disinfectants prevalence of gentamicin resistance of coliform bacteria
coupled with physical methods can be suitably applied for in Indian foods was 50%. Furthermore, Gundogan and
controlling biofilm formation on food-contact surfaces Avci (2014) observed that 53.7% of gentamicin resistan­
(Reynisson et al., 2009; Mahapatra et al., 2015). ce amongst E. coli isolates. Arslan and Küçüksari (2015)
reported low levels of resistance to gentamicin (4.1%) and
Antimicrobial Resistance amikacin (4.1%) in Aeromonas spp. isolates while high le­
During the past decade, multidrug resistance in Gram-ne­ vel of gentamicin resistance was reported in Aeromonas
gative bacteria is increasing throughout the world (Wood­ spp. isolates (54%) by Yucel et al. (2005).
ford et al., 2014). This increase is mainly the result of an Carbapenem antibiotics are the last treatment option
increased prevalence of ESBL-producing Enterobacteria- for severe, life-threatening infections caused by multi­
ceae and non-lactose fermenting bacteria such as Pseudo- ple-drug resistant pathogens. Carbapenem-resistant Ente­
monas and Acinetobacter species (Tekiner and Ozpinar, ro­bacteriaceae strains and non-fermentative gram-ne­
2016). gative bacilli isolated from human infections have been

TABLE 4: Antibiotic resistance of gram-negative bacteria.


Species AMK GEN IMP ETP AMP AMC SAM PIT CAZ CTX CEP CRO CIP AZT
E. coli – – – 3.8*
100 71.7 56.6 52.8 – – 7.5 13.2 17 20.8
E. cloacae – – – 8.9 84.4 77.8 68.9 60 13.3 – 11.1 – – –
A. hydrophila – – – – 100 72 72 51.1 74.4 55.8 67.4 55.8 – –
K. oxytoca – – – 8.6 100 74.3 54.3 51.4 14.3 14.3 8.6 14.3 17.1 –
C. freundii – – – – 100 73 53.8 57.7 15.3 11.5 11.5 15.3 – –
H. alvei – – – 8.3 100 70.8 58.3 58.3 12.5 – – – – –
S. marcescens – – – 8.3 70.8 70.8 50 70.8 – 16.7 16.7 – – –
P. agglomerans – – – – 33.3 26.7 20 20 6.7 – – – – –
S. fanticola – – – – 16.7 16.7 25 16.7 16.7 – – 8.3 – –
P. vulgaris – – – – 75 25 16.7 16.7 – – – – – –
C. amalonaticus – – – – 100 8.3 16.7 16.7 16.7 16.7 – – – –
R. aquatilis – – – – 50 8.3 8.3 16.7 – – – – – –
P. aeruginosa – – – – 100 72.7 54.5 81.8 63.6 63.6 72.7 54.5 – –
M. morganii – – – – 72.7 72.7 54.5 72.7 – – – – – –
E. vulneris – – – – 100 20 10 20 – – – – – –
K. pneumoniae – – – 11.1 66.7 77.8 55.6 77.8 11.1 – – – 11.1 22.2
P. rettgeri – – – – 66.7 11.1 11.1 11.1 – – – – – –
S. putrefaciens – – – – 100 83.3 50 50 66.7 83.3 66.7 66.7 – –
P. putida – – – – 50 16.7 16.7 16.7 66.7 66.7 66.7 83.3 – –
P. fluorescens – – – – 50 83.3 66.7 66.7 83.3 83.3 66.7 66.7 – –
A. caviae – – – – 100 83.3 50 50 66.7 66.7 83.3 83.3 – –
A. lwoffii – – – – 83.3 16.7 16.7 16.7 66.7 83.3 66.7 66.7 – –
A. baumannii – – – – 100 83.3 66.7 83.3 83.3 83.3 83.3 66.7 – –
P. oryzhabitans – – – – 20 20 20 20 80 80 80 80 – –
*Percentage of resistant isolates; AMK: amikacin; GEN: gentamicin; IMP: imipenem; ETP: ertapenem; AMP: ampicillin; AMC: amoxycillin/clavulanic acid; SAM: ampicillin /sulbactam; PIT: piperacillin/tazobac-
tam; CAZ: ceftazidime; CTX: cefotaxime; CEP: cefepime; CRO: ceftriaxone; CIP: ciprofloxacin; AZT: aztroenam
106 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

reported in many parts of the world (Woodford et al., Schwaiger et al., 2012; Gundogan and Avci, 2014; Arslan
2014; Webb et al., 2016), including Turkey (Kuzucu et al. and Küçüksari, 2015; Wong et al., 2015).
2011; Karabay et al. 2016). In the present study, all isolates Ciprofloxacin is a broad spectrum fluoroquinolone
were susceptible to imipenem, whereas, 3.8% of E. coli, antibacterial agent. The observed resistance of E. coli,
8.9% of E. cloacae, 8.6% of K. oxytoca, 8.3% of H. alvei, K. oxytoca and K. pneumoniae to ciprofloxacin was 17%,
8.3% of S. marcescens, and 11.1% of K. pneumoniae isola­ 17.1% and 11.1 %, respectively. No resistance to ciptof­
tes were resistant to ertapenem. However, in our previous loxacin was observed in other isolates. Gundogan et al.,
studies, carbapenems were the most effective antibiotics 2011, 2013, 2014) also reported low resistance rates of
for Klebsiella spp. and E. coli isolates, in which 100% of Klebsiella spp. (16%) and E. coli (29.4%-31.1%) to cipro­
the isolates were s­usceptible to meropenem and imipe­ floxacin. Compared to our results, higher rates of cipro­
nem (Gundogan et al., 2011; Gundogan and Avci 2014). floxacin resistance have been reported by Benameur et
A high incidence of imipenem resistance (95%) was also al. (2018), who observed that 90.47% of K. pneumoniae
documented both in coliform bacteria (E. coli, C. diver- and 85.10% of E. coli isolates isolated from poultry were
sus, E. cloacae, K. oxytoca, S. fonticola, K. pneumoniae, resistant to ciprofloxacin. Chakravarty et al. (2015) sho­
E. aerogenes) and in fish pathogens (A. hydrophila, A. ca- wed high prevalence of ciprofloxacin-resistant E. coli in
viae, P. oryzihabitans) isolated from Turkish trout farms foods. In other study, more than 63% of Enterobacteria-
(Capkin et al., 2015). Resistance to carbapenems, such as ceae isolates showed ­resistance to ciprofloxacin (Ye et al.,
ertapenem (40.82%), meropenem (31.36%), and imipe­ 2013). Ciprofloxacin resistance in Klebsiella and E. coli
nem (10.65%) in E. coli isolates have also been reported is predominantly due to a chromosomal mutation in the
by Chakravarty et al. (2015) in India. Some studies have gyrA gene, which codes for the target of quinolone activity
reported the low prevalence of carbapenem resistance in (Pehlivanlar-Onen et al., 2015). As resistance to ciproflo­
A. lwoffii (Wang et al., 2012), A. baumannii (Zhang et xacin emerged, resistance to ß-lactam antibiotics became
al., 2013) Aeromonas spp. (Arslan and Kücüksari, 2015), prominent. This resistance was largely a result of ESBLs,
Pseudomonas spp. (Wong et al., 2015), and Enterobacte- which mediate resistance to newer ß-lactam agents, such as
riaceae (Ye et al., 2018). ceftazidime, ceftriaxone, cefotaxime, and aztreonam, that
Gram-negative rods are frequently associated with have an oxyamino group (Pehlivanlar-Onen et al., 2015).
­resistance to ß-lactam antibiotics due to a constitutively Cephalosporins are an important class of antibacterial
expressed ß-lactamase. We detected resistance to ampi­ agents in use for both humans and animals. According to
cillin in all of the E. coli, A. hydrophila, K. oxytoca, C. the results reported here, resistance to ceftazidime, cefo­
freundii, H. alvei, C. amalonaticus, P. aeruginosa, E. taxime, cefepime and ceftriaxone was observed for >54 %
vulneris, S. putrefaciens, A. caviae and A. baumanii iso­ of the Aeromonas spp., Pseudomonas spp., Acinetobacter
lates. Resistance rates of other isolates to ampicillin varied spp., and S. putrefaciens isolates. Enterobacteriaceae iso­
between 16.7% and 84.4%. This is not surprising because lates showed resistance to ceftazidime, cefotaxime, cefepi­
ß-lactams are commonly used antibiotics for the treatment me and ceftriaxone in the range of 0–16.7%. In a previous
of Gram-­negative bacterial infections in humans and an­ ­studies conducted by Tekiner and Ozpinar (2016) and Ye
imals (Ondes and Ozpinar, 2016). Similar findings have et al. (2018) on the resistance of Enterobacteriaceae from
been reported in a recent study for Enterobacteriaceae various foods, a high percentage (≥60%) of isolates were
isolates in China, where 97.9% of all isolates were resis­ resistant to cephalosporins. Resistance to these group an­
tant to ampicillin (Ye et al., 2018). Frequent occurrence tibacterial agents for Gram-negative bacteria in aquatic
of ampicillin resistance in members of Enterbacteriaceae, environments reported to be >90% (Matyar et al., 2004;
Aeromonas spp. and Pseudomonas spp. obtained from 2008; 2014; Schwaiger et al. 2012; Wasiński et al., 2014;
various foods have also been described previously (Kilon­ ­Devarajan et al., 2017; Singh et al., (2017). The simulta­
zo-Nthengeet al., 2013; Gundogan and Avci, 2014; Capkin neous resistance of isolates to ß-lactams, may be due to
et al.,2015). the dissemination of antibiotic resistance plasmids in the
Clavulanic acid, sulbactam or tazobactam, which are marine environment, as reported by Matyar et al. (2004).
ß-lactamase inhibitors, regarded as good choice for inhibit Aztreonam is a synthetic monocyclic ß-lactam in the
ESBL-producing Gram-negative bacteria. However, high family of monobactams and is exclusively active (like ami­
rates of resistance to amoxycillin/clavulanic acid (>70%), noglycosides) against the aerobic gram-negative bacilli. We
ampicillin/sulbactam (≥50%), and piperacillin/tazobactam observed that only 22.2 % and 20.8 % of K. pneumoniae and
(≥50%) were observed among E. coli, E. cloacae, A. hy- E. coli isolates, respectively, were resistant to aztroenam.
drophila, K. oxytoca, C. freundii, H. alvei, S. marcescens, According to Gundogan et al. (2011, 2013, 2014), aztroenam
P. aeruginosa, M. morganii, K. pneumoniae, S. putrefa- had moderate activity against Klebsiella spp. (24%–42.9%)
ciens, P. fluorescens, A. caviae and A. baumanii isolates. and E. coli (29.9%). Our results were not in agreement with
In the present study, low incidences of resistance to ß-lac­ the findings of Capkin et al. (2015) who found that a high
tamase inhibitors was found in other isolates (<30 %). incidence of azt­reonam resistance (95%) in Aeromonas spp.,
Singh et al. (2017), in India, investigated fresh seafoods for and Pseudomonas spp. Environmental and food isolates of
the occurrence and antimicrobial resistance patterns of E. coli and Klebsiella spp. showed resistance to aztreonam in
ESBL-producing Enterobacteriaceae. These researchers the range of 76.9–100% (Afifi, 2013; Ibrahimagić et al., 2016).
reported that resistance to amoxicillin-clavulanic acid and Excessive ampicillin usage in Turkey for treatment
piperacillin/tazobactam was seen in 38.46% and 40.82% of infections in humans and animals can be regarded as
of the Enterobacteriaceae isolates, respectively which are one of the major causes of resistance to this antimicrobial
comparable with results of our study. Compared to our among Gram-negative bacteria. Also, there are great ten­
results, lower prevalences of resistance to amoxycillin/cla­ dency towards decreased susceptibility observed for ß-lac­
vulanic acid and piperacillin/tazobactam has been found tams, carbapenems and cephalosporins. Therefore, there
in isolates of E. coli, C. freundii, P. agglomerans, Aero- remains a need for continued surveillance and judicious
monas spp. and Klebsiella spp. (Gundogan et al., 2011; use of these antibiotics.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 107

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Multiple Antibiotic Resistance (MAR) Index varied from 0.29 to 0.36. Our results were in agreement
Increasing levels of MAR among bacteria are resulted with Matyar et al. (2010) who reported that Aeromonas
from widespread use of antibiotics in human and veteri­ isolates resistant to 6 or more antibiotics and MAR index
nary medicine and as growth promoters for intensive li­ values ranged from 0.2 to 0.60. These authors have also
vestock production (Arslan and Kucuksari, 2015). In ad­ detected multiple resistance in Pseudomonas isolates but
dition, food handlers may cross contaminate foods during reported higher MAR index values (0.2–0.73). Matyar
preparation and if they are carriers of MAR bacteria, they (2007) showed that 29.4% and 8% of the Pseudomonas
may contaminate foods themselves (Carvalheira et al., spp. isolates from Turkish sea bass (Dicentrarchus labrax)
2017). Consumable animal products have been suggested resistant to 7 and 8 antibiotics, respectively. In his study,
as a possible source of both resistant bacteria and resistant higher MAR index values (ranging from 0.3–0.8) has been
genes that can be transferred to humans directly (Shrestha reported. Arslan and Kucuksari et al. (2015) reported that
et al., 2017). prevalence of multiple resistance in Aeromonas spp. isola­
If the bacterial isolates were resistant to four or more tes to 3 or more antimicrobial agents was 57.1%, indicating
antibiotics, they were regarded as multi antibiotic resistant fish and ground beef were exposed to significant antibiotic
(MAR) (Matyar et al., 2014). In the present study, MAR contamination.
index values in Gram-negative bacterial isolates ranged In this study, S. putrefaciens showed resistance to 4
from 0.29 to 0.64, showing a resistance to 4–9 antibiotics. (50%), 5 (50%) and 6 (16.7%) antibiotics and MAR index
MAR index values >0.2 indicate that the isolates must ranging from 0.29 to 0.43 (Figure 1). In Korea and Italy,
have originated from an environment where antibiotics ­respectively Kang et al. (2013) and Smaldone et al. (2014)
are often used. Our finding is in agreement with previous detected multiple resistance of S. putrefaciens isolated
reports showing the prevalence of multiple-resistant bac­ from marine fish and shellfish involving resistance to van­
teria in various foods in Turkey and different parts of the comycin, oxacillin, tetracycline, penicillin and ampicillin.
world. Authors suggested that marine environments are import­
Among the multi-resistant isolates, a significant pro­ ant reservoirs for resistance genes, and may be play an
prtion of Acinetobacter spp. isolates were resistant to important role in transfer of drug-resistant genes between
4 (100%), 5 (83.3%), 6 (83.3%), 7 (66.7%) 8 (50%) and 9 bacteria.
(50%) antibiotics. The MAR index ranged from 0.29–0.57 It was observed in our study and above-mentioned stu­
(Figure 1). In Portugal, 51.2% of the Acinetobacter strains dies that there are no geographic borders for the spread
were considered as multidrug-resistant (Carvalheira et al., of antibiotic-resistant bacteria, and that their emergence
2017). In China, isolates from aquaculture products, inclu­ in the food chain threatens the world. MDR bacteria can
ding Acinetobacter spp. were resistant to two (22%), three continue to spread globally via the food chain as well.
(36%), and four (29%) antibiotics (Ye et al., 2013). High
prevalence of multi-resistance in A. baumannii isolates ESBL Production of Gram-negative Bacteria
from chicken, veal, beef, pork and turkey has also been ESBLs are plasmidborne enzymes that can hydroly­
­reported in Switzerland by Lupo et al. (2014). ze cephalosporins and monobactams. They are mainly
Our results revealed that the MAR index ranging from producing in Gram-negative bacilli, particularly several
0.29 to 0.64 for Enterobacteriaceae isolates. They were re­ members of Enterobacteriaceae, including Escherichia,
sistant to 4 (55%), 5 (46.4%), 6 (43%), 7 (35.9%), 8 (35.6%) Klebsiella, Proteus, Citrobacter, Serratia, Salmonella and
and 9 (28.8%) antibiotics. (Figure 1). Higher percentages Shigella (Ondes and Ozpinar, 2016).
reported by Matyar et al. (2008) who showed that 43.3% In the present study, 144 (35.6%) out of 404 isolates
of Gram-negative bacteria isolated from different sources, were identified as ESBL producers. In a previous stu­
including seawater, shrimp and sediment in Turkey, were dies conducted in Turkey, the frequency of ESBL-produ­
resistant to 6 antibiotcs while 56.8 % of them were resis­ cing Enterobacteriaceae ranged between 3.7%–52.7% in
tant to 7 or more antibiotcs. In USA, Kilonzo-Nthenge ground beef, chicken meat, raw milk and cheese (Ondes
et al. (2013) observed that 84.9% of
Enterobacteriaceae isolates isolated
from chicken and beef displayed
MAR to 3 or more antimicrobials.
In their study, 19.2% of isolates sho­
wed MAR to 5 or more antimicro­
bials. Capkin et al. (2015) showed
MAR index values ranging between
0.19−0.83 and 0.42−0.83, respecti­
vely for coliform bacteria and other
Gram-negative bacteria in Turkish
trout. These values are higher than
that obtained from our study.
As it shown in Figure 1, isolates
from Aeromonas spp. showed resis­
tance to 4, 5, 6, 7 and 8 antibiotics
with a frequency of 91.8%, 81.6%,
65.3% ,44.9% and 32.6%, respecti­
vely. The MAR index ranged from
0.29-0.57. Pseudomonas spp. isola­
tes were resistant to 4 (89.3%) and 5
antibiotics (60.8%). The MAR index FIGURE 1: M  ultiple antibiotic resistance (MAR) index of Gram-negative bacteria.
108 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

and Ozpinar, 2016; Tekiner and Ozpinar, 2016). Outside Although H. alvei, S. fanticola, R. aquatilis, M. mor-
of Turkey, various rates of ESBL-producing Enterobac- ganii, E. vulneris, P. rettgeri, P. fluorescens, A. caviae, A.
teriaceae were reported as 8.3%–42.5% in China (Ye et lwoffii and P. oryzhabitans isolates did not produce ESBL
al., 2018), 13.5% in Bosnia and Herzegovina (Ibrahimagić in the present study, it should not be underestimated that
et al., 2016), 36.9% in Nepal (Shrestha et al., 2017), and ESBL encoding genetic elements are transferable between
78.6% in India (Singh et al., 2017) examined raw meat the same and different bacterial species (Tekiner and Oz­
products, seafoods, ready-to-eat (RTE) foods and frozen pinar, 2016). Thus, monitoring of ESBL-producing bacte­
foods. On the other hand, ESBL-producing bacteria were ria should be continued at various level (animals, human,
not detected in the Swedish meat or in the Swed­ish chi­ and environment), while investigating the factors that
cken (Tham et al., 2012). contri­bute to their selection and dissemination.
The results in the work reported here showed that the
majority of the ESBL producers were K. oxytoca (74.3%),
followed by C. freundii (61.5%), E. coli (56.6%), A. hy- Conclusion
drophila (51.1%), P. aeruginosa (45.4%), C. amalonaticus
(41.7%), E. cloacae (33.3%), S. marcescens (33.3%), P. The results obtained in this study in Turkey showed a high
vulgaris (33.3%), K. pneumoniae (33.3%), S. putrefaciens incidence of Gram-negative bacteria in marine fish, veal
(33.3%), P. putida (33.3%) and A. baumannii (33.3%) and and chicken samples. Our results suggest that some of
P. agglomerans (26.7%) isolates (data not shown). Percen­ these bacteria such as E. coli, E. coli O157, and Aeromo-
tages of ESBL producing E. coli and P. aeruginosa, re­ nas spp. represent a potential health risk. This is because
spectively were 45% in Egyptian camel meat (Elhariri et some strains of these organisms are capable of producing
al., 2017), and 55% in Dutch chicken meat (Blaak et al., toxins. Furthermore, bacterial characteristics that influen­
2015) which are comparable with the results of our study. cing the food safety such as biogenic amine production,
How­ever, our results were appeared to be higher than tho­ slime and biofilm formation are quite common within the
se in several studies. For instance, Singh et al. (2017) de­ Gram-negative bacteria. In particular, E. coli, E. cloacae,
tected ESBL-positive K. oxytoxa in 27% of seafood. Duan C. freundii, K. oxytoca, K. pneumoniae, P. aeruginosa, P.
et al. (2006) reported 3.1% prevalence of ESBL producers fluorescens, A. hydrophila, A. caviae, A. baumanni and
among E. coli isolates from cattle in China while Petternel S. putrefaciens isolates were found resistant to clinical­
et al. (2014) reported an incidence of 24% from minced ly important antibiotics and most of them have multiple
meat in Austria. Studies from Turkey reported 21%, 5.5% ­antibiotic resistance (MAR) patterns and produced ESBL,
and 9.1%, respectively of ESBL producing K. oxytoca, thus posing a health risk for the Turkish consumers. The
Citrobacter spp., and E. cloacae isolates in red meat and presence of these bacteria in fish, veal and chicken seemed
chicken meat (Gundogan et al., 2014; Tekiner and Ozpi­ to be related to the unhygienic production processes and
nar, 2016). ESBL detection rates in Citrobacter spp. and storage conditions. Thus, all potential sources of MAR
Proteus spp. isolated from chiken meat were 26.1% and and ESBL-producing bacteria should be considered and
26.3%, respectively (Shrestha et al., 2017). The frequency strategies devised to reduce their presence in foods. Furt­
of ESBL-producing E. coli, Klebsiella spp., Citrobacter hermore, Turkish regulatory agencies should require food
spp., Proteus spp., E. cloacae, P. aeruginosa, M. morganii, processing plants to adopt quality guarantee systems such
P. rettgeri, isolated from water, food and environ­mental as Hazard Analysis and Critical Control Points (HACCP)
samples were 13.6% in Bosnia and Herzegovina (Ibrahi­ system and a better control system to prevent the presence
magić et al., 2016). of these products on the market.
Compared to our results, higher contamination rates of
different types of foods with ESBL-producing E. coli and
Pseudomonas spp. were reported as 77%, 84% and 100% Acknowledgements
by Nahar et al. (2018), Stuart et al. (2012) and Shrestha et
al. (2017), respectively. Foods of animal origin sold in Tur­ This work was financially supported by the Gazi Univer­
key have been found as potential reservoirs for ESBL-pro­ sity Research Fund (Project No:05/2012-67). The authors
ducing Gram-negative bacteria. The National Surveillance wish to thank Mec. Eng. M.Sc. Tuncer Yakut for critical
Network by the Ministry of Health in Turkey (www.uhes. reading of the manuscript.
saglik.gov.tr) has reported an increasing prevalence of
ESBL-producing E. coli (33.2% in 2008 and 48.83% in
2013) and ESBL-producing K. pneumoniae (40% in 2008 Conflict of interest
and 49.69% in 2013). In the current study, fish, veal and
chicken meat were found to be highly contaminated with The authors declare that there are no conflicts of interest
ESBL-producing K. oxytoca (74.3%), and E. coli (56.6%). regarding the publication of this paper.
Results obtained in the present study were not suprising
when compared with our previous study in which ESBL
production was detected in 26% of K. oxytoca and 44.4% References
of E. coli isolates islated from foods of animal origin (Gun­
dogan and Avci 2013). There is a clear tendency towards Afifi MM (2013): Detection of extended spectrum beta-lactamase
increased prevalence of ESBL-producing Gram-negati­ producing Klebsiella pneumoniae and Escherichia coli of en­
ve bacteria in Turkish animal-derived foods. This might vironmental surfaces at upper Egypt. Int J Biol Chem 7: 58-68.
AOAC (1998): Bacteriological Analytical Manual, 8th edn, pp.
lead to a risk for infection and colonisation of the human
4.01–4.29. Revision A, Gaithersburg, MD, USA: AOAC Inter­
­intestinal flora with ESBL-producing bacteria. Transfer of national.
ESBL-producing Enterobacteriaceae to humans via the Arslan S and Küçüksari R (2015): Phenotypic and genotypic viru­
food chain has been reported previously (Tekiner and Oz­ lence factors and antimicrobial resistance of motile Aeromonas
pinar 2016). spp. from fish and ground beef. J Food Saf 35: 551-559.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 109

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Arslan S, Eyi A, Ozdemir F (2011): Spoilage potentials and anti­ Gundogan N, Citak S, Yalcın E (2011): Virulence properties of ex­
microbial resistance of Pseudomonas spp. isolated from chee­ tended spectrum ß-lactamase–producing Klebsiella Species in
ses. J Dairy Sci 94: 5851-5856. meat samples. J Food Protect 74: 559-564.
Bagge D, Hjelm M., Johansen C, Huber I, Gram L (2001): She- Gundogan N, Ataol O, Torlak FO (2013): Determination of some
wanella putrefaciens adhesion and biofilm formation on food virulence factors in Staphylococcus aureus, Enterococcus fae-
processing surfaces. Appl Environ Microbiol 67: 2319-2325. calis and Enterococcus faecium isolated from meat and milk
Bakhtiary F, Sayevand HR, Remely M, Hippe B, Hosseini H, products. J Food Saf 33: 387-393.
Haslberger AG (2016): Evaluation of bacterial contamination Gundogan N and Avci E (2013): Prevalence and antibiotic resis­
sources in meat production line. J Food Qual 39: 750-756. tance of extendedspectrum beta-lactamase (ESBL) producing
Benameur Q, Tali-Maamar H, Assaous F, Guettou B, Benklaouz Escherichia coli and Klebsiella species isolated from foods of
MB, Rahal K, Ben-Mahdi MH (2018): Characterization of qui­ animal origin in Turkey. Afr J Microb Res 7:4059-4064.
nolone-resistant Enterobacteriaceae strains isolated from poul­ Gundogan N and Avci E (2014): Occurrence and antibiotic resis­
try in Western Algeria: First report of qnrS in an Enterobacter tance of Escherichia coli, Staphylococcus aureus and Bacillus
cloacae. Vet World 11: 469-473. cereus in raw milk and dairy products in Turkey. Int J Dairy
Bitrian M, Solari CM, González RH, Nudel CB (2012): Identifica­ Technol 67: 562-569
tion of virulence markers in clinically relevant strains of Acine- Gwida M, Hotzel H, Geue L, Tomaso H (2014): Occurrence of
tobacter genospecies. Int Microbiol 15: 79-88. Enterobacteriaceae in raw meat and in human samples from
Bjornsdottir K, Bolton GE, Mcclellan-Green PD, Jaykus LA, Egyptian retail sellers. Int Schol Res Not 2014 565671 doi:
Green DP (2009): Detection of gram-negative histamine-pro­ 10.1155/2014/565671.
ducing bacteria in fish: a comparative study. J Food Protect 72: Ibrahimagić A, Idrizovic E, Divjan E, Klimenta B (2016): Preva­
1987-1991. lence and antimicrobial resistance of beta-lactamase-producing
Blaak H, Lynch G, Italıaander R, Hamıdjaja RA, Schets FM, De Gram-negative isolates from outpatient clinical and environ­
Roda Husman AM (2015): Multidrug-resistant and extended mental samples in the Zenica-Doboj Canton, Bosnia and Herz­
spectrum beta-lact­amase-producing Escherichia coli in Dutch egovina. J Health Sci 6: 94-99.
surface water and wastewater. PLoS One 10: 1-16 Kang CH, Shin Y, Jeon H, Choi JH, Jeong S, So JS (2013): Antibio­
Capkin E, Terzi E, Altinok I (2015): Occurrence of antibiotic resis­ tic resistance of Shewanella putrefaciens isolated from shellfish
tance genes in culturable bacteria isolated from Turkish trout collected from the West Sea in Korea. Mar Pol Bull 76: 85-88.
farms and their local aquatic environment. Dis Aqua Org 114: Karabay O, Altındis M, Koroglu M, Karatuna O, Aydemir OA,
127-137. Erdem AF (2016): The carbapenem-resistant Enterobacteriace-
Carvalheira A, Casquete R, Silva J, Teixeira P (2017): Prevalence ae threat is growing: NDM-1 epidemic at a training hospital in
and antimicrobial susceptibility of Acinetobacter spp. isolated Turkey. Ann Clin Microbiol Antimicrob 15: 2-6.
from meat. Int J Food Microbiol 243: 58-63.
Kilonzo-Nthenge A, Rotich E, Nahashon SN (2013): Evalua­
Cetin Ö, Bingol EB, Çolak H, Ergün Ö, Demir C (2010): The mi­ tion of drug-resistant Enterobacteriaceae in retail poultry and
crobiological, serological and chemical qualities of mincemeat beef. Poultry Sci 92: 1098-1107.
marketed in İstanbul. Turk J Vet Anim Sci 34: 407-412.
Kucukates E (2005): Antimicrobial resistance among Gram-nega­
Chakravarty MS, Ganesh PRC, Amaranth D, Shanthi Sudha B, tive bacteria isolated from intensive care units in a Cardiology
Subhashini M (2015): Escherichia coli-occurrence in the meat Institute in Istanbul, Turkey. Japan J Infect Dis 58: 228-231.
of shrimp, fish, chicken and mutton and its antibiotic resistan­
Kuzucu C, Yetkin F, Gorgeç S, Ersoy Y (2011): Investigation of
ce. Europ J Exp Biol 5: 41-48.
the susceptibilities of extended-spectrum beta-lactamase-pro­
Christensen GD, Simpson WA, Younger JJ, Baddour LM, Bar-
ducing Escherichia coli and Klebsiella spp. strains to ertapenem
rett FF, Melton DM, Beachey EH (1985): Adherence of coa­
and other carbapenems. Mikrob Bul 45: 28-35.
gulase-negative staphylococci to plastic tissue culture plates: a
Liu YJ, Xie J, Zhao LJ, Qian YF, Zhao Y, Liu X (2015): Biofilm
quantitative model for the adherence of staphylococci to medi­
formation characteristics of Pseudomonas lundensis isolated
cal devices. J Clin Microbiol 226: 996-1006.
from meat. J Food Sci 80: 2904-2910.
Clinical and Laboratory Standards Institute (2006): Performance
Standards for Antimicrobial Susceptibility Testing; 16th Infor­ Lupo A, Vogt D, Seiffert SN, Endimiani A, Perreten V (2014):
mational Supplement. CLSI document M100-S16, Clinical and Antibiotic resistance and phylogenetic characterization of Aci-
Laboratory Standards Institute Wayne, PA. USA. netobacter baumannii strains isolated from commercial raw
meat in Switzerland. J Food Protect 77: 1976-1981.
Da Silva ML, Rogério Matté GLAVUR, Germano PML, Matte
MH (2010): Occurrence of pathogenic microorganisms in fish Mahapatra A, Padhi N, Mahapatra D, Bhatt M, Sahoo D, Jena S,
sold in São Paulo, Brazil. J Food Saf 30: 94-110. Chayani N (2015): Study of biofilm in bacteria from water pipe­
Devarajan N, Kohler T, Sivalingam P, Van Delden C, Mulaji CK, lines. J Clin Diag Res 9: 9-11.
Mpiana PT, Poté J (2017): Antibiotic resistant Pseudomonas Maifreni M, Frigo F, Bartolomeoli I, Innocente N, Biasutti M, Ma-
spp. in the aquatic environment: A prevalence study under tro­ rino M (2013): Identification of the Enterobacteriaceae in Mon­
pical and temperate climate conditions. Water Res 115: 256- tasio cheese and assessment of their amino acid decarboxylase
265. activity. J Dairy Res 80: 122-127.
Duan RS, Sit TH, Wong SS, Wong RC, Chow KH, Mak GC, Yam Marino M, Maifreni M, Moret S, Rondinini G (2000): The capaci­
LT, Ng KY, Yuen DR, Ho PL (2006): Escherichia coli produ­ ty of Enterobacteriaceae species to produce biogenic amines in
cing CTX-M ß-lactamases in food animals in Hong Kong. Mi­ cheese. Lett Appl Microbiol 31: 169-173.
crob Drug Res 12: 145-148. Matyar F, Dincer S, Kaya A, Colak O (2004): Prevalence and re­
Durlu-Özkaya F, Ayhan K, Vural N (2001): Biogenic amines pro­ sistance to antibiotics in Gram negative bacteria isolated from
duced by Enterobacteriaceae isolated from meat products. Meat retail fish in Turkey. Ann Microbiol 54: 151-160.
Sci 58: 163-166. Matyar F (2007): Distribution and antimicrobial multiresistance in
Elhariri M, Hamza D, Elhelw R, Dorgham SM (2017): Exten­ Gram-negative bacteria isolated from Turkish sea bass (Dicen-
ded-spectrum beta-lactamase-producing Pseudomonas aeru­ trarchus labrax L., 1781) farm. Ann Microbiol 57: 35-38.
ginosa in camel in Egypt: potential human hazard. Ann Clin Matyar F, Kaya A, Dincer S (2008): Antibacterial agents and
Microbiol Antimicrob 16: 21-27. heavy metal resistance in Gram-negative bacteria isolated from
Grigoryan K, Badalyan G, Harutyunyan A, Sargsyan M (2013): seawater, shrimp and sediment in Iskenderun Bay, Turkey. Sci
Prevalence of Gram negative and oxıdase posıtıve bacteria in Tot Environ 407: 279-285.
trout processing factory. J Hyg Eng Des 4:10-15. Matyar F, Akkan T, Uçak Y, Eraslan B (2010): Aeromonas and
Gucu AC, Genc Y, Dagtekin M, Sakinan S, Ak O, Ok M, Aydin Pseudomonas: antibiotic and heavy metal resistance spe­
I (2017): On black sea anchovy and its fishery. Rev Fish Sci cies from Iskenderun Bay, Turkey (northeast Mediterranean
Aqua 25: 230-244. Sea). Environ Mon Ass 167: 309-320.
110 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Matyar F, Gulnaz O, Guzeldag G, Mercimek HA, Akturk S, Arkut nation in organic and conventional retail chicken meat. Int J
A, Sumengen M (2014): Antibiotic and heavy metal resistance Food Microbiol 154: 212-214.
in Gram-negative bacteria isolated from the Seyhan Dam Lake Tekiner IH and Ozpinar H (2016): Occurrence and characteristics of
and Seyhan River in Turkey. Ann Microbiol 64: 1033-1040. extended spectrum beta-lactamases-producing Enterobacteria-
Mol S, Saglam OE (2004): Investigating seafood marketing con­ ceae from foods of animal origin. Braz J Microbiol 47: 444-451.
ditions in some important Turkish seafood markets in compa­ Tembhurne M, Ghag A, Sanathkumar H, Nayak BB (2013): Do­
rison with European countries. Turkish J Fish Aq Sci 4: 65-70. minance of Enterobacteria among histamine-producing bacte­
Møretrø T, Langsrud S, Heir E (2013): Bacteria on meat abattoir ria isolated from Indian mackerel. Advan Microbiol 3: 537-542. 
process surfaces after sanitation: characterisation of survival Temelli S, Eyigor AG, Anar Ş (2012): Prevalence of Escherichia
properties of Listeria monocytogenes and the commensal bac­ coli O157 in red meat and meat products determined by VI­
terial flora. Adv Microbiol 3: 255-264. DAS ECPT and LightCycler PCR. Turk J Vet An Sci 36: 305-
Nahar A, Awasthi SP, Hatanaka N, Okuno K, Hoang PH, Hassan 310.
J,  Hinenoya A, Yamasaki S (2018): Prevalence and characte­ Tham, J, Walder M, Melander E, Odenholt I (2012): Prevalence
ristics of extended-spectrum ß-lactamase-producing Escheri- of extended-spectrum beta-lactamase-producing bacteria in
chia coli in domestic and imported chicken meats in Japan. J food. Infect Drug Res 5: 143-147.
Vet Med Sci 80: 510-517. TurkStat (2013): Fishery Statistics. Ankara: Turkish Statistical In­
Niven CF, Jeffrey MB, Corlett DA (1981): Differential plating stitute
­medium for quantitative detection of histamine-producing bac­ Wang Y, Wu C, Q, Qi J, Liu H, Wang Y, He T, Ma L, Lai J, Shen Z,
teria. App Environ Microbiol 41: 321–322. Liu Y, Shen J (2012): Identification of New Delhi metallo-be­
Ondes N and Ozpinar H (2016): Occurrence of ESBL-producing ta-lactamase 1 in Acinetobacter lwoffii of food animal origin.
Enterobacteriaceae in cubed beef samples. Kafkas Univ Vet PloS one, 7, e37152.
Fak Derg 22: 79-83. Wasiński B, Hanna R, Jacek O (2014): Antimicrobial resistance
Orozova P, Barker M, Austin DA, Austin B (2009): Identificati­ of ESBLand AmpC-producing Escherichia coli isolated from
on and pathogenicity to rainbow trout, Oncorhynchus mykiss meat. Bull Vet Inst Pul 58: 567-571.
(Walbaum), of some Aeromonads. J Fish Dis 32: 865-871. Webb HE, Bugarel M, Den Bakker HC, Nightingale KK, Granier
Ozogul F and Ozogul Y (2005): Formation of biogenic amines by SA, Scott HM, Loneragan GH (2016): Carbapenem-resistant
Gram-negative rods isolated from fresh, spoiled, VP-packed bacteria recovered from faeces of dairy cattle in the high plains
and MAP-packed herring (Clupea harengus). Eur Food Res region of the USA. PloS one, 11, e0147363.
Technol 221: 575–581. Wong MHY, Chi Chan EW, Chen S (2015): Isolation of carbape­
Pehlivanlar-Onen S, Aslantas O, Sebnem Yilmaz E, Kurekci C nem-resistant Pseudomonas spp. from food. J. Global Antimi­
(2015): Prevalence of ß-lactamase producing Escherichia coli crob Res 3: 109-114.
from retail meat in Turkey. J Food Sci 80: 2023-2029. Woodford N, Wareham DW, Guerra B, Teale C (2014): Carbapene­
Petternel C, Galler H, Zarfel G, Luxner J, Haas D. Grisold AJ, mase-producing Enterobacteriaceae and non-Enterobacteriaceae
Feierl G (2014): Isolation and characterization of multidrug-­ from animals and the environment: an emerging public health
resistant bacteria from minced meat in Austria. Food Micro­ risk of our own making?. J Antimicrob Chem 69: 287-291.
biol 44: 41-46. Yaron S and Romling U (2014): Biofilm formation by enteric
Radu S, Ahmad N, Ling FH, Reezal A (2003): Prevalence and ­pathogens and its role in plant colonization and persistence.
­resistance to antibiotics for Aeromonas species from retail fish Microb Biotech 7: 496-516.
in Malaysia. Int J Food Microbiol 81: 261-266. Ye L, L, Li X, Shi L, Huang Y, Wang HH (2013): Antibiotic-resis­
Reynisson E, Lauzon HL, Magnússon H, Jónsdóttir R, Ólafsdóttir tant bacteria associated with retail aquaculture products from
G, Marteinsson V, Hreggviðsson GÓ (2009): Bacterial com­ Guangzhou, China. J Food Protect 76: 295-301.
position and succession during storage of North-Atlantic cod Ye Q, Wu Q, S, J, Yang G,Wang J, Xue L, Chen M (2018): Charac­
(Gadus morhua) at superchilled temperatures. BMC Microbi­ terization of extended-spectrum ß-lactamase-producing Ente­
ol 9: 250-262. robacteriaceae from retail food in China. Front Microbiol
Sarimehmetoglu B, Aksoy MH, Ayaz ND, Ayaz Y, Kuplulu O, 9:1709. doi: 10.3389/fmicb.2018.01709.
Kaplan YZ (2009): Detection of Escherichia coli O157:H7 in Yucel N and Erdogan S (2010): Virulence properties and charac­
ground beef using immunomagnetic separation and multiplex terization of Aeromonads isolated from foods of animal origin
PCR. Food Cont 20: 357-361. and environmental sources. J Food Protect 73: 855-860.
Schwaiger K, Huther S, Holzel C, Kämpf P, Bauer J (2012): Pre­ Yucel N, Aslim B, Beyatli Y (2005): Prevalence and resistance to
valence of antibiotic-resistant Enterobacteriaceae isolated from antibiotics for Aeromonas species isolated from retail fish in
chicken and pork meat purchased at the slaughterhouse and at Turkey. J Food Qual 28: 313-324.
retail in Bavaria, Germany. Int J Food Microbiol 154: 206-211. Zaman MZ, Bakar FA, Selamat J, Bakar JA (2010): Occurrence
Shrestha A, Bajracharya AM, Subedi H, Turha RS, Kafle S, Shar- of biogenic amines and amines degrading bacteria in fish sau­
ma S, Neupane S, Chaudhary DK (2017): Multi-drug resistance ce. Czech J Food Sci 28: 440-449.
and extended spectrum beta lactamase producing Gram nega­ WJ, Lu Z, Schwarz S, RM, Wang XM, Si W, Yu S, Chen L, Liu S
tive bacteria from chicken meat in Bharatpur Metropolitan, (2013): Complete sequence of the blaNDM-1-carrying plasmid
Nepal. BMC Res Not 10: 574-579. pNDM-AB from Acinetobacter baumannii of food animal ori­
Silagyi K, Kim SH, Lo YM, Wei CI (2009): Production of biofilm gin. J Antimicrob Chem 68: 1681-168.
and quorum sensing by Escherichia coli O157: H7 and its trans­
fer from contact surfaces to meat, poultry, ready-to-eat deli,
and produce products. Food Microb 26: 514-519.
Singh AS, Lekshmi M, Prakasan S, Nayak BB, Kumar S (2017):
Multiple Antibiotic-Resistant, Extended Spectrum-ß-Lacta­ Address of corresponding author:
mase (ESBL)-Producing Enterobacteria in Fresh Seafood. Mi­ Dr. Neslihan Gundoğan
croorg 5: 53-63. Department of Biology, Faculty of Science
Smaldone G, Marrone R, Cappiello S, Martin GA, Oliva G, Cor- Gazi University, Teknikokullar
tesi ML, Anastasio A (2014): Occurrence of antibiotic resistan­
Ankara 06500
ce in bacteria isolated from seawater organisms caught in Cam­
pania Region: preliminary study. BMC Vet Res 10: 161-168. Turkey
Stuart JC, van den Munckhof T, Voets G, Scharringa J, Fluit A, gundogan@gazi.edu.tr
Leverstein-Van Hall M (2012): Comparison of ESBL contami­
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 111

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Arch Lebensmittelhyg 70, Faculty of Sciences and Mathematics, The University of Nis, Visegradska 33, 18000 Nis,
111–116 (2019) Serbia
DOI 10.2376/0003-925X-70-111

© M. & H. Schaper GmbH & Co. A new kinetic method using UV-VIS
ISSN 0003-925X spectrophotometry for determination
Korrespondenzadresse: of caffeic acid in propolis
danijelaaakostic@yahoo.com
Ein neues kinetisches Verfahren zur Bestimmung von Kaffeesäure
in Propolis mittels UV-VIS-Spektrophotometrie
Danijela A. Kostic, Snezana Mitic, Milan Mitic, Emilija Pecev Marinkovic,
Ivana Rasic Misic, Biljana Arsic, Gordana Stojanovic

Summary The aim of this work is to develop an application kinetic-spectrophotometric procedure


for the determination of caffeic acid (CA) in propolis. The method is based on oxidation
reaction of CA by hydrogen peroxide in the presence of Cu (II) ions in alkaline solution.
The reaction was monitored spectrophotometrically by measuring the rate of change of
absorbance at 345 nm. The optimum operating conditions for reagent concentrations
and temperature were established. Linear calibration curve was obtained in the range
of 1.94 to 19.4 µg/ml with standard deviation from 2.77 to 4.15 %. The optimized
­conditions yielded a theoretical detection limit of 0.6 µg/ml based on 3.3So criterion.
The developed method is sensitive, accurate and reproducible and could be used for
routine analysis of CA in propolis.

Keywords: caffeic acid, interference, kinetic procedure, propolis

Zusammenfassung Ziel dieser Arbeit ist die Entwicklung eines kinetisch-spektrophotometrischen Verfah-
rens zur Bestimmung von Kaffeesäure (KS) in Propolis. Die Methode basiert auf der
Oxidationsreaktion von KS mit Wasserstoffperoxid unter Anwesenheit von Cu (II)-Ionen
in alkalischer Lösung. Die Reaktion wurde spektrophotometrisch überprüft, indem die
Änderungsrate der Absorption bei 345 nm gemessen wurde. Die optimalen Betriebs-
bedingungen hinsichtlich der Konzentration der Reagenzien sowie der Temperatur wur-
den ermittelt. Die lineare Kalibrierungskurve lag im Bereich von 1,94 to 19,4 µg/ml, mit
einer Standardabweichung von 2,77 bis 4,15%. Die optimierten Bedingungen ergaben
eine theoretische Nachweisgrenze von 0,6 µg/ml, basierend auf dem 3.3So-Kriterium.
Die entwickelte Methode ist sensitiv, präzise sowie reproduzierbar und könnte bei der
Routineanalyse von KS in Propolis zum Einsatz kommen.

Schlüsselwörter: Kaffeesäure, Interferenz, kinetisches Verfahren, Propolis


112 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Introduction 1200 (Agilent) series with a semi-preparative diode ar­


ray detector, and a Zorbax Eclipse XDB-C18 Semi-Prep,
Caffeic acid (CA), 3,4-dihydroxycinnamic acid, is one of 5um, 9.4x250 mm column.
the most abundant hydroxycinnamic acids found natural­
ly in fruits, vegetables, cereals, legumes, coffee, and tea. Reagents
It is also found in wine for human consumption as simple Stock solution (0.1 mol/l) of NaOH (Merck, Darmstadt,
esters with quinic acid (Iglesias et al., 2009; Jiang et al., Germany) was prepared in deionized water.
2005). A stock solution (1.0∙10 –3 mol/l) of CA was prepared in
CA and related compounds are well known radical sca­ absolute ethanol from CA powder. A stock solution of Cu
vengers and antioxidants in coffee beans with health-pro­ (II) (1.0∙10 –3 mol/l) was prepared by dissolving CuCl 2 (J. T.
moting attributes (Celik and Erdogan, 2008; Kanimozhi Baker, USA) in water. Hydrogen peroxide solution (0.442
and Prasad, 2015). In addition, CA, multifunctional natu­ mol/l) was prepared from the 34 % reagent and stored at
ral available organic acid plays a significant role in binding 4 ºC.
­metal ions from the natural environment. The compound For the preparation of all solutions, analytical reagent
has two complexing sites in competition: the catechol grade chemicals and deionized water (MicroMed High
group (dihydroxybenzene) and the carboxylic function. ­Purity Water System) were used. All used glassware was
Several coworkers reported the complexation of the washed with aqueous HCl (1:1) and then with distilled
compound with different metal ions in aqueous solutions: ­water, and finally with deionized water.
Al (III) (Cornard and Lapouge, 2006; Cornard et al.,
2006), copper (II), Ni (II), Zn (II), Co (II) and iron (III) General procedure
(Hynes and O’Coinceanainn, 2004). There are also re­ Prior to the first measurement, the instruments were run
ports on the complexation of caffeic acid with polyphenol for 10 min to obtain good mechanical and thermal stabi­
and aromatic compounds investigated by spectroscopic lity.
and computational methods for advanced design and con­ In the reaction mixture four compartments vessel, the
trollable carriers of drugs and food components (Górnas solution of CA was placed in one compartment, NaOH in
et al., 2009). the second, in the third Cu (II), hydrogen peroxide and
Propolis or “bee glue” is the generic name of the resi­ ­water (total volume: 10 ml) in the fourth compartment.
nous product which is collected by bees from various plant The vessel was thermostated at 25.0±0.1°C, and the
sources (Burdock, 1998), and its composition varies with ­reaction was initiated by vigorously shaking the reactants,
the source. Generally, it is composed of 50 % resin and which was followed by transferring the content to a cell,
vegetable balsam, 30 % wax, 10 % essential oils and aro­ and the absorbance at 345 nm was measured using UV/
matics, 5% pollen, and 5% other substances (Siheri et al., VIS spectrophotometer every 30 s for 5–6 min against the
2017). Propolis is well known from ancient times because blank prepared similarly. The rate of the reaction at dif­
of its biological and pharmacological properties, like im­ ferent concentrations of each reactant was determined by
munomodulatory, antitumoral, antimicrobial, anti-inflam­ measuring the slope of the linear part of the curves of the
matory, and antioxidant effects (Alvarez-Suarez, 2017). absorbance time plot.
Methods of the used extraction may influence its activity.
The most used method is solid-liquid extraction (ethanol Sample preparation
in different concentrations, methanol or water). The bio­ Samples of raw propolis were collected during 2017 by
logical activity of propolis is mainly due to the presence scraping the frames of bee hives that originated from three
of flavonoids, especially caffeic acid (CA) and its phenet­ different regions of Serbia: Stara planina, Soko banja and
hyl ester (CAPE) (Santos-Buelga and González-Paramás, Suva planina. The representative sample of a particular
2017). location was obtained through mixing the propolis from
Up to now, our colleagues (Tosic et al., 2017) were in­ ten different apiaries. The samples were frozen and pul­
vestigated the content of minerals in propolis and found verized.
that propolis from Serbia is rich in minerals. Since pro­ Five grams of propolis (in small pieces) was soaked into
polis is rich in phenolic acids, and particularly caffeic acid, 10 ml of water, and then heated at 70 °C for 40 min in a
we decided to investigate the content of caffeic acid in pro­ thermostatic water bath system and filtered yielding the
polis by the application of new and simple UV/VIS met­ refined propolis, which was then placed in a constant tem­
hod. perature drying oven set at 80 °C until the constant weight.
There are published UV-Vis spectrophotometric met­ Thus obtained propolis samples were homogenized using
hods for the determination of CA, but they are not easy an agate homogenizer and then stored in sealed glass vials
to perform and are not sufficiently sensitive, accurate and until further analysis.
selective. The aim of the present work is the development Methanol, ethanol and acetonitrile in different concen­
of a simple, sensitive, specific, spectrophotometric method tration were examined as extracting solvents for caffeic
for the detection of CA in propolis, which does not need acid from propolis. The highest extraction yield was obtai­
sophisticated instruments or special skills. ned with methanol 80 %.
In order to extract phenolic compounds, 3.50 g of the
dried propolis samples was soaked into 10 ml of 80 %
Materials and Methods methanol for 24 h. The obtained solution was extracted
using ultrasound-assisted procedure twice (each time for
Apparatus 20 min), and then evaporated using a rotary evaporator (at
UV/Vis spectrophotometer (Perkin-Elmer Lambda 15) 65 °C) to dryness (Čižmárik and Matel, 1970).
equipped with kinetic accessory and a temperature con­ The obtained sample was treated with 100 ml of the
trolled cell was used. ­mobile phase, filtered through a filter (pore size 0.45 µm)
HPLC analysis was carried out using the HPLC system and injected into the HPLC.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 113

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

HPLC analysis The correlation between the reaction rate and H 2O2
HPLC analysis was carried out using the HPLC system concentration was not linear, so the logarithms of tana
1200 (Agilent) series with a diode array detector, and a were calculated and the gotten values were plotted versus
Zorbax Eclipse XDB-C18 Semi-Prep, 5um, 9.4x250 mm the logarithms of H 2O2 concentration. The order of reac­
column. Solvents used for separation were 0.1 % ortho­ tion was minus 0.7 in the whole interval of H 2O2 concen­
phosphoric acid in water (v/v) (eluent A) and 0.1% ortho­ tration. For further investigation, a concentration of H 2O2
phosphoric acid in methanol (v/v) (eluent B). The gradient of 1.326·10 –2 mol/l was selected.
used was: 0–10 min, linear gradient from 40 % to 50 % B; How reaction rates depend on the concentrations of Cu
10–15 min, linear gradient from 50 % to 60 % B, maintain (II) it was monitored over the range (1∙10 –6 mol/l–5∙10 –6
at 60 % B until 25 min. The flow rate was 1.0 ml/min. De­ mol/l) (Fig. 3).
tection wavelength was 330 nm. The sample injection vo­ For further investigation, a concentration of Cu (II) of
lume was 10 µl. The chromatographic peaks of caffeic acid 2·10 –6 mol/l was selected as the working value. Regarding
were confirmed by comparing their retention times and Cu (II) concentration, the rate of substrate reaction was
UV spectra with that of their reference standards. Working –1 order.
standard solutions were injected into the HPLC and peak The optimal reaction conditions were:
area responses obtained. Standard graphs were prepared CNaOH = 1·10 –2 mol/l, CCu(II) = 2·10 – 6 mol/l, CH2O2 =
by plotting concentration versus area. Quantification was 1.326 ·10 –2 mol/l ,
carried out from integrated peak areas of the samples using
t = 25 ± 0.1 °C,  = 345 nm
the corresponding standard graph (Wang et al., 2004).
Statistical analysis Taking into account the kinetics of the proposed indicator
The data were reported as the mean ± standard devia­ reaction, the kinetic equation (Eq. 1) for the reaction was
tion (SD) for triplicates. The significance of inter-group derived (Ermer, 2001).
differences was determined by the analysis
of variance (ANOVA). The p-value < 0.05
was considered statistically significant (Sta­
tistical Analysis and Reporting System, ser
Guide, version 1.0, 1MB, 1999).

Results

Kinetic studies
The kinetic data were processed using the
differential ­variant of the tangent method
(Svehla, 1993) due to the fact that a linear
correlation exists between the absorbance at
345 nm and time during the first 6 min after
mixing. To determine the lowest possible
detectable concentration of CA, the condi­
tions had to be optimized.

Effect of variables FIGURE 1: D


 ependence of reaction rate on pH. Initial concentrations: CCA =
Constant experimental parameters were 8·10 –5 mol/l, CCu(II) = 1·10 – 6 mol/l, CH2O2 = 8.825·10 –2 mol/l, t = 25 ±
kept while the d ­ ependence of the reaction 0.1 °C,  = 345 nm.
rate on pH in NaOH solution (0.1 mol/l) in
the range of 0.5–2.0∙10 –3 mol/l was studied.
(Fig. 1).
The optimum value of the difference bet­
ween the rates of non-substrate and substra­
te reactions was at a concentration of NaOH
solution 1.0 ∙10 –3 mol/l and it was used for
further work.
However, the correlation between the re­
action rate and pH was not linear, so the lo­
garithms of tana were calcu­lated and the got­
ten values were plotted versus pH. Using the
obtained regression equations, the order of
reaction was determined (0.8) in the NaOH
concentration range (0.5·10 –3 –2.0 ∙10 –3 mol/l).
The reaction rate dependence on the
concentration of H 2O2 was investigated in
the range 0.442–8.84∙10 –2 mol/l. Oxidation
reaction rate dependence of caffeic acid on FIGURE 2: D
 ependence of the reaction rate on H2O2 concentration. Initial con-
­hydroxide peroxide concentration was shown centrations: CCA = 8·10 –5 mol/l, CCu(II) = 1·10 – 6 mol/l, C NaOH = 1·10 –2
in Fig. 2 and it is an exponential function. mol/l, t = 25 ± 0.1 °C, =345 nm.
114 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

dc
–– = k · c 0.8 · c –1 · c – 0.7 · cCA
1
 (1)
dt NaOH Cu H 2O2

Concentrations of CA were varied (1.94–


19.4 µg/ml) and a linear dependence was
found between tg a and the concentration of
CA (Fig. 4).
The kinetic equations (Eq. 2-4) for the
reaction were derived on the basis of the
kinetics of the indicator reaction proposed.

tga = 0.2387 CCA – 0.1214, R2 = 0.9984, t=20±0.1 °C (2)


tga = 0.2625 CCA – 0.1168, R2 = 0.9984, t=25±0.1 °C (3)
tga = 0.287  CCA – 0.0847, R2 = 0.9982, t=30±0.1 °C(4)

where tga corresponds to the slope of the


­linear part of the curve of the absorbance-­
time plot; CCA is concentration of CA ex­
pressed as µg/ml and R is the coefficient of FIGURE 3: D
 ependence of reaction rate on Cu (II) concentration. Initial concen-
the correlation. trations: CCA = 8·10 –5 mol/l, C NaOH = 1·10 –2 mol/l, CH2O2 = 1.326 ·10 –2
mol/l, t = 25 ± 0.1 °C,  = 345 nm.
Parameter value
The limit of detection (LOD) (Prichard and
Bedson, 2001) was found according to Eq. 5,

3.3 · S 0
cL = –––––– (5)
m

where S 0 corresponds to the residual stan­


dard deviation of the calibration line, and m
to the slope of the calibration line. LOD was
found to be 0.6 µg/ml. The limit of quantifi­
cation (LOQ) was found according to Eq. 6.

10 · S 0
cQ = –––––– (6)
m

It was found to be 1.8 µg/ml, which indicates


that the method is sensitive.
The accuracy and the precision of the FIGURE 4: D
 ependence of reaction rate on CA concentration. Initial concentra-
proposed method were investigated by per­ tions: CNaOH = 1·10 –2 mol/l, CCu(II) = 2·10 – 6 mol/l, CH2O2 = 1.326 ·10 –
forming the experiment five times at three 2
mol/l, t = 20± 0.1 °C (red), t = 25± 0.1 °C (blue), t = 30± 0.1 °C
different CA concentration levels (low, (green), =345 nm.
­medium and high) (Table 1).

TABLE 1: Accuracy and precision of the proposed method. TABLE 2: Tolerance ratio for foreign species in the deter-
Taken Found N S RSD %
––µ
x mination of CA (11.65 μg/ml) under optimal
––––– · 100 conditions.
(µg/ml) (µg/ml) µ
  1.94 2.02 5 0.06 2.77 3.96 Taken Foreign species
11.65 12.12 5 0.34 2.82 3.86 Cx/CCA

19.42 20.20 5 0.84 4.15 4.00 100 Na+, K+, Ni2+, Fe2+, Mg2+, Ca2+, NO3–, CH3COO –, CO32–, Cl–
N: number of experiments from each sample 100 talc, fructose, lactose, starch, magnesium stearate, microcrystalline cellulo-
se
Interference studies
10 Se4+, Cd2+, Mn2+, Pb2+, Cu2+, Br–, SO42–
A systematic investigation of the possible interferences by
species accompanying CA in natural pharmaceuticals was Interfere ascorbic acid, glucose, citric acid, gallic acid, coumaric acid, ferulic acid
performed in order to find the selectivity of the method.
For the established level of caffeic acid, 5 % variation of caffeic acid phenethyl ester (CAPE) is not expected due to
the average slope change measured (n=5) was fixed as the the extraction procedure of caffeic acid from propolis and
criterion of interference (Table 2). polarity of the ester (Chen et al., 1996; Russo et al., 2002).
Several ions and some substances-microcrystalline
­cellulose, starch, lactose, fructose, talc, magnesium stea­ Applicability of the proposed method
rate, were shown to have no interference. Contrary, seve­ A point hypothesis test was used to statistically compa­
re interference was observed for ascorbic acid, citric acid, re our results with those of the standard HPLC method
and other phenolic acids (in 1:1 ratio). The interference of (Chen et al., 2012) (Table 3).
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 115

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

TABLE 3: Determination of CA in propolis by the proposed method (kinetic) and stan- from 2.5 to 3.0 %, which is in line
dard HPLC methods. with the results shown by the analy­
Sample Content of CA RSD Recovery Content of CA F-test* t-test** sis of propolis from the territory of
HPLC method (%) (%) kinetic method Croatia (continental part: 0.27–2.67
(µg/ml) (µg/ml) %, Adriatic part: 0–10.11 %) (Kosa­
lec et al., 2003), Romania (1.15–1.54
Stara planina 3.05±0.20 1.86 99.20 3.00±0.15 1.83 1.26
%), Israel (1.23–1.51 %) (Croci et al.,
Suva planina 2.70±0.30 2.11 98.76 2.60±0.20 1.73 1.13 2009), and Turkey (Anatolian part:
Soko banja 2.60±0.25 1.49 99.54 2.50±0.20 2.06 1.36 0.05–1.2 %, Kazan: 0.32 %, Mar­
*Data are based on the mean of five determinations; *Theoretical F-value (v1 = 4, v2 = 4) and t- value (v = 8) at 95 % confidence level are 6.39 and maris: 18.54 %) (Kartal et al., 2002;
2.306, respectively. Uzel et al., 2005). It is noteworthy
that the samples from the areas close
The accuracy was estimated by recovery tests perfor­ to the sea show higher concentrations of caffeic acid in the
med for each analyte. The spiked propolis samples were propolis samples.
prepared and analysed by the proposed HPLC-DAD Propolis is very useful due to its richness in minerals
method. The results are shown in Table 3. The average re­ (Tosic et al., 2017) and natural phenolic compounds, espe­
coveries are in the range 98.76–99.54 % (RSD < 2.11 %); cially caffeic acid that has very beneficial effects on health
the results showed very good recoveries for the proposed (Alvarez-Suarez, 2017).
analytical method.

Conclusions
Discussion
The developed method is sensitive, accurate and repro­
Using the differential tangential method, optimal condi­ ducible and could be used for routine analysis of CA. No
tions for the determination of the micro amounts of CA in significant differences between the performance of the
the solution are determined. proposed and the standard HPLC method confirms the
usefulness of using our method. Therefore, the proposed
CNaOH = 1·10 –2 mol/l, CCu(II) = 2·10 – 6 mol/l, method could be used for the determination of CA in pro­
CH2O2 = 1.326 ·10 –2 mol/l , t = 25± 0.1 °C,  = 345 nm polis.

Under the experimental conditions, concentrations of


CA were 1.94–19.4 µg/ml and a linear dependence was Acknowledgements
found between tg a and the concentration of CA. A linear
­calibration curve is constructed (1.94–19.4 μg/ml with the Ministry of Education, Science and Technological De­
relative standard deviation of less than 4 %, and the detec­ velopment of the Republic of Serbia [172047].
tion limit of 0.6 μg/ml.
Examination of the selectivity of the method, i.e. the
­effects of numerous compounds and ions present in natu­ Conflict of interest
ral and pharmaceutical preparations, does not significant­
ly influence the determination of CA (Tab. 3).
The authors declare that they have no conflict of interest.
The developed method is accurate, sensitive, and re­
producible and could be used for everyday analysis of CA.
In comparison with other spectrophotometric methods,
the method is more sensitive and selective. Propolis com­ References
pounds have been analyzed by other different methods:
Alvarez-Suarez JM (2017): Bee products-chemical and biological
thin-layer chromatography (TLC), gas chromatogra­ properties. Springer International Publishing.
phy-mass spectrometry (GC-MS) (Garcia Viguera et al., Bankova VS, Popov SS, Marekov NL (1982): High-performance
1993), high performance liquid chromatography (HPLC) liquid chromatographic analysis of flavonoids from propolis. J.
(Bankova et al., 1982), LC-MS and micellar electrokinetic Chromatogr. A 242, 135–143.
capillary chromatography (MEKC) (Fontana et al., 2000). Burdock GA (1998): Review of the biological properties and to­
The method was developed for the determination of xicity of bee propolis (propolis). Food Chem. Toxicol. 36(4),
CA in propolis and the results obtained are compared 347–363.
Celik S, Erdogan S (2008): Caffeic acid phenethyl ester (CAPE)
with those obtained by the HPLC method. Statistical
protects brain against oxidative stress and inflammation indu­
analysis of the results (Tab. 3) showed that on the basis of ced by diabetes in rats. Mol. Cell. Biochem. 312 (1-2), 39–46.
calculated F and t values (95 % confidence levels) there Chen F, Song H, Hong Bn GAO, Wang Z, Peng FU, SongWei
are no significant differences between the performance of LU, Yuan W (2012): Simultaneous determination of gallic acid,
the proposed and the standard HPLC method. Therefore, chlorogenic acid and caffeic acid contents in Ganmao’an granu­
the proposed spectrophotometric method could be used les by HPLC method. Pharm. Care Res. 12(6), 427–430.
for the determination of CA in propolis. Comparing to Chen J-H, Shao Y, Huang M-T, Chin C-K, Ho C-T (1996): Inhibi­
HPLC methods the linearity intervals (LOD, LOQ) are tory effect of caffeic acid phenethyl ester on human leukemia
in the range of previously published methods (Spagnol et HL-60 cells. Cancer Lett. 108, 211-214.
Čižmárik J, Matel I (1970): Examination of the chemical compo­
al., 2016). Our method is suitable for the investigations and
sition of propolis I. Isolation and identification of the 3,4-di­
the control in areas without modern and expensive equip­ hydroxycinnamic acid (caffeic acid) from propolis. Experientia
ment and also for fieldwork. 26(7), 713.
The content of caffeic acid in propolis samples from Cornard JP, Caudron A, Merlin J-C (2006): UV-visible and syn­
different locations in the area of Southeast Serbia ranges chronous fluorescence spectroscopic investigations of the com­
116 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

plexation of Al(III) with caffeic acid, in aqueous low acidic Prichard E, Bedson P (2001): Analytical measurement termino­
medium. Polyhedron 25(11), 2215–2222. logy: handbook of terms used in quality assurance of analytical
Cornard J-P, Lapouge C (2006): Absorption spectra of caffeic acid, measurement (valid analytical measurement). Royal Society of
caffeate and their 1:1 complex with Al(III): Density functional Chemistry.
theory and time-dependent density functional theory investiga­ Russo A, Longo R, Vanella A (2002): Antioxidant activity of pro­
tions. J Phys. Chem. A 110 (22), 7159–7166. polis: role of caffeic acid phenethyl ester and galangin. Fitote­
Croci AN, Cioroiu B, Lazar D, Corciova A, Ivanesku B, Lazar MI rapia 73 (Suppl. 1), S21-S29.
(2009): HPLC evaluation of phenolic and polyphenolic acids Santos-Buelga C, González-Paramás AM (2017): Phenolic Com­
from propolis. Farmacia LVII (1), 52–57. position of Propolis. In Bee Products-Chemical and Biological
Ermer J (2001): Validation in pharmaceutical analysis. Part I: An properties. Springer, Cham. 99–111.
integrated approach. J. Pharm. Biomed. Anal. 24(5–6): 755–767. Siheri W, Alenazi S, Tusiimire J, Watson DG (2017): The chemical
Fontana JD, Passos M, dos Santos MHR, Fontana CK, Oliveira and biological properties of propolis. In Bee products-chemical
BH, Schause L, Pontarolo R, Barbirato MA, Ruggiero MA, and biological properties. Cham. 137–178.
Lanças FM (2000): Profiling propolis flavonoids by means of
Spagnol CM, Isaac VLB, Corrêa MA, Salgado HRN (2016): Vali­
micellar electrokinetic capillary chromatography, capillary gas
dation of HPLC–UV assay of caffeic acid in emulsions. J. Chro­
chromatography and bactericidal action. Chromatographia 52,
matogr. Sci. 54(3), 305–311.
147–151.
García-Viguera C, Ferreres F, Tomás-Barberán FA (1993): Study Svehla G (1993): Nomenclature of kinetic methods of analysis.
of Canadian propolis by GC-MS and HPLC. Z. Naturforsch. Pure Appl. Chem. 65, 2291.
48c, 731–735. Tosic S, Stojanovic G, Mitic S, Pavlovic A, Alagic S (2017): Mine­
Górnas P, Neunert G, Baczynski K, Polewski K (2009): Beta-­ ral composition of selected Serbian propolis samples. J. Apic.
cyclodextrin complexes with chlorogenic and caffeic acids from Sci. 61(1), 5–15.
coffee brew: spectroscopic, thermodynamic and molecular mo­ Uzel A, Sorkun K, Oncag O, Cogulu D, Gencay O, Salih B (2005):
delling study. Food Chem. 114(1), 190–196. Chemical compositions and antimicrobial activities of four
Hynes MJ, O’Coinceanainn (2004): The kinetics and mechanisms different Anatolian propolis samples. Microbiol. Res. 160(2),
of reactions of iron(III) with caffeic acid, chlorogenic acid, si­ 189–195.
napic acid, ferulic acid and naringin. J. Inorg. Biochem. 98(8), Wang H, Provan GJ, Helliwell K (2004): Determination of ros­
1457–1464. marinic acid and caffeic acid in aromatic herbs by HPLC. Food
Iglesias J, Pazos M, Andersen ML, Skibsted LH, Medina I (2009): Chem. 87, 307–311.
Caffeic Acid as antioxidant in fish muscle: mechanism of syner­
gism with endogenous ascorbic acid and a-tocopherol. J. Agric.
Food Chem. 57(2), 675–681.
Jiang R-W, Lau K-M, Hon P-M, Mak TCW, Woo K-S, Fung K-P
(2005): Chemistry and biological activities of caffeic acid de­
rivatives from Salvia miltiorrhiza. Curr. Med. Chem. 12 (2),
237–246. Address of corresponding author:
Kanimozhi G, Prasad NR (2015): Anticancer effect of caffeic acid Dr. Danijela A. Kostic
on human cervical cancer cells. In Coffee in health and disease Faculty of Sciences and Mathematics
prevention. Academic Press. The University of Nis
Kartal M, Kaya S, Kurucu S (2002): GC-MS analysis of propolis Visegradska 33
samples from two different regions of Turkey. Z. Naturforsch.
18000 Nis
57c, 905–909.
Kosalec I, Bakmaz M, Pepeljnjak S (2003): Analysis of propolis
Serbia
from the continental and Adriatic region of Croatia. Acta danijelaaakostic@yahoo.com
Pharm. 53, 275–285.

Kontakte
Sie interessieren sich für ein
­Abonnement des „Archivs“?
Telefon (0 51 81) 80 02-50
Ravenstraße 45 · 31061 Alfeld (Leine)
Telefax (0 51 81) 80 02-55 Postfach 16 42 · 31046 Alfeld (Leine)
E-Mail heiko.schaper@p-d-ges.de Telefon (0 51 81) 80 02-0
Telefax (0 51 81) 80 02-55
E-Mail info@p-d-ges.de
Sie interessieren sich für Anzeigen-
oder Beilagenwerbung?
Telefon (0 51 81) 80 02-53
Telefax (0 51 81) 80 02-55
E-Mail anzeigen@p-d-ges.de
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 117

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Arch Lebensmittelhyg 70, ICAR – Central Institute of Post-Harvest Engineering & Technology,
117–123 (2019) Ludhiana - 141 004, India
DOI 10.2376/0003-925X-70-117

© M. & H. Schaper GmbH & Co. Moisture dependent physical properties


ISSN 0003-925X of dehusked unsplitted pigeon pea
Korrespondenzadresse:
chandan4uu12@gmail.com Feuchtigkeitsabhängige physikalische Eigenschaften von geschälten,
nicht geteilten Straucherbsen
Chandan Solanki, Dhritiman Saha, Ranjeet Singh

Summary Pigeon pea (Cajanus cajan, L.) is an important legume crop in semiarid tropics gene-
rally consumed as dried peas for making dhal, flour or green vegetable peas. Dehusked
unsplitted pigeon pea also called gota are obtained during the primary processing for
making dhal. Hence, the knowledge about the physical properties of gota is important
for designing processing machineries. In this regard, the study of the physical pro-
perties of gota (dehusked unsplitted pigeon pea) in the moisture content range of
10.2–28.2% d.b. was conducted. Five different levels of moisture content in the above
range were selected. The average length, width, thickness, arithmetic mean diameter,
geometric mean diameter, sphericity surface area, volume, thousand seed weight, bulk
density, true density, porosity, angle of repose and coefficient of friction on plywood,
mild steel and galvanized iron were determined in this moisture range. The values of
average length, width, thickness, arithmetic mean diameter, geometric mean diame-
ter, sphericity and surface area varied from 5.33 to 5.43mm, 4.39 to 4.45mm, 3.93 to
3.77mm, 4.51 to 4.55mm, 4.88 to 4.5mm, 0.858 to 0.828 and 63.29 to 63.73mm2
respectively. At the given moisture levels, volume, thousand seeds weight ranged from
38.41 to 40.77mm3, 61.94g to 68.93g in the same moisture range, the bulk density
and true density decreased linearly from 859.5 to 781.7 kg/m3 and 1306.6 to 1128 kg/
m3 respectively and porosity decreased non linearly from 34.22 to 30.2%. The static
coefficient of friction of gota over different materials, namely plywood, mild steel and
galvanized iron increased linearly from 0.4491 to 0.8405, 0.4789 to 0.7882 and 0.2851
to 0.6775 respectively. In the selected moisture range, the angle of repose increased
from 28.35° to 36.72°.

Keywords: dehusked unsplitted pigeon pea, gota, physical property,


moisture content

Zusammenfassung Die Straucherbse (Cajanus cajan, L.) ist eine wichtige Hülsenfrucht in semiariden Tropen,
die im Allgemeinen in getrockneter Form zur Herstellung von Dhal, Mehl oder Erbsen-
gemüse verzehrt wird. Während der Primärverarbeitung für die Herstellung von Dhal
werden geschälte, nicht geteilte Straucherbsen, auch Gota genannt, gewonnen. Daher
sind die physikalischen Eigenschaften von Gota für die Konstruktion von Bearbeitungs-
maschinen von großem Interesse. Die Untersuchungen der physikalischen Eigenschaf-
ten der geschälten, nicht geteilten Straucherbsen wurden im Feuchtigkeitsgehaltbe-
reich von 10,2–28,2%, bezogen auf die Trockenmasse, durchgeführt. Es wurden fünf
verschiedene Feuchtigkeitsgehalte im genannten Bereich ausgewählt. Die durchschnitt-
liche Länge, Breite, Dicke, der arithmetische mittlere Durchmesser, der geometrische
mittlere Durchmesser, die sphärische Oberfläche, das Volumen, das Tausendkorn­
gewicht, die Schüttdichte, die wahre Dichte, die Porosität, der Schüttwinkel und der
Reibungskoeffizient auf Sperrholz, Weichstahl und verzinktem Eisen wurden in diesem
Feuchtigkeitsbereich bestimmt. Die Werte der durchschnittlichen Länge, Breite, Dicke,
des arithmetischen Durchschnittsdurchmessers, des geometrischen Durchschnitts-
durchmessers, der Sphärizität und der Oberfläche variierten von 5,33 bis 5,43 mm,
4,39 bis 4,45 mm, 3,93 bis 3,77 mm, 4,51 bis 4,55 mm, 4,88 bis 4,55 mm, 0,858 bis
0,828 bzw. 63,29 bis 63,73 mm². Bei den angegebenen Feuchtigkeitsgraden, Volumen
und der Tausendkernmasse lagen die Gewichte im Bereich von 38,41 bis 40,77 mm3,
61,94 g bis 68,93 g im gleichen Feuchtigkeitsbereich. Die Schüttdichte und die wahre
Dichte nahmen linear von 859,5 auf 781,7 kg/m3 und 1306,6 auf 1128 kg/m3 ab. Die
Porosität nahm nicht linear von 34,22 auf 30,2% ab. Der statische Reibungskoeffizient
der Straucherbsen gegenüber der verschiedenen Materialien (Sperrholz, Weichstahl
und verzinktem Eisen) stieg linear von 0,4491 auf 0,8405, 0,4789 auf 0,7882 bzw.
0,2851 auf 0,6775. Im ausgewählten Feuchtigkeitsbereich erhöhte sich der Ruhewinkel
von 28,35° auf 36,72°.

Schlüsselwörter: Straucherbse, Gota, physikalische Eigenschaft, Feuchtigkeitsgehalt


118 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Introduction Materials and method

Sample preparation
Pigeon pea (Cajanus cajan L.) is a perennial and popu­ Whole pigeon pea seeds were procured from local mar­
lar legume from the family Fabaceae. The dry seeds are ket. The average initial moisture content was found to be
boiled and eaten as a pulse or with other foods. The In­ 10.20% db. The grains were cleaned using cleaner cum
dia’s production of pigeon peas is estimated at 2.56 million grader to separate foreign matter, dust, dirt, twigs, broken
tons (Annual report 2016–17, Department of Agriculture, and immature seeds. Moisture content of the sample was
Cooperation and Farmers’ Welfare). India produces ab­ determined by hot air oven method as described by Nim­
out 90% of world’s total pigeon pea and remaining 10% kar and Chattopadhyay 2001. The grain moisture content
produced by Africa and Caribbean countries (Nwokolo range was selected between 10% to 28% dry basis because
& Smartt, 1996). It is consumed on a large scale main­ the harvesting is being practiced at about 28% and trans­
ly for being a major source of protein. This pigeon pea portation, storage, handling and processing operations of
is harvested at 28% d.b. moisture content to reduce the the crop are being performed at about 10%.
shuttering losses and processing recovery. The seeds are The weight of the samples was recorded on an analytical
usually round or oval (orbicular), vary in size and color, balance (Model: TB403, Denver Instrument) of accuracy
being w­ hite, grayish, red, brown or purplish. The dehus­ 0.001g in triplicate, and their average value was recorded.
ked u­ nsplitted pigeon pea is generally termed as ’Gota‘. The sample was divided into lots that were conditioned for
The quality of any food grains is mainly affected by its moisture content in the range of 10.20–28.20% db by ad­
harvesting moisture content and this moisture content is ding predetermined amounts of distilled water calculated
very much relevant to is processing quality with nutritional from the following relationship:
security.
In the European era it’s known as dehusked unsplitted
pigeon pea grain. Dhal is prepared after the processing of
dehusked splitted pigeon pea. Dhal is the term which is
used after cleaning, grading and most important unit ope­ Where, Q = Quantity of water to be added (g)
ration i.e. milling then after dhal is prepared. The German W = Quantity of sample (g)
part of pigeon pea is Taubenerbse. The physical properties Mi = initial moisture content of the sample (% db.)
of pigeon pea, like other grains, are essential and import­ Mf = desired moisture content of the sample (% db.)
ant for designing the processing equipment for further
handling and post-­harvest processing. Various types of The sample were kept at 5°C in a refrigerator for one
cleaning, grading and separation equipment are desig­ week for uniform distribution of moisture throughout the
ned on the basis of physical properties of seeds (Sahay & ­sample. Before each test, the required quantity of sample
Singh, 1994). Physical properties affect conveying charac­ was taken out of refrigerator and allowed to attain ambient
teristics of solid materials by air or water and cooling and temperature before carrying out the experiment.
heating loads of food materials.
Measurement of Properties
The knowledge on the physical properties of a crop is
essential for proper design of processing equipment. The Size and shape
size distribution and characteristic dimensions of grain is In order to determine the size and shape, fifty gota seeds
important for the design of equipment for cleaning, sor­ were chosen at random from the overall sample. This
ting and separation (Kachru et al., 1994). The bulk density ­random sampling method was similar to the one used
is used to determine the capacity of storage and transport, by Dutta et al. 1988. For each grain, three-dimensional
while true density is useful to design proper separation ­features (length (L), width (W) and thickness (T)) were
equipment. Moreover, porosity of the grain mass deter­ measured with a digital vernier caliper (Absolute Digima­
mines the resistance to airflow during aeration and dry­ tic, Make: Mitutoyo, Model: CD-‘CSX’) with an accura­
ing operation (Brooker et al., 1992; Kachru et al., 1994). cy of 0.01 mm. The arithmetic mean diameter (Da), the
Frictional properties such as angle of repose and coeffi­ ­geometric mean diameter (Dg) and sphericity of gota were
cient of friction are important properties for the design of calculated by the following equations (Mohsenin, 1986);
grain containers and other storage structures (Vilche et
al., 2003). These properties are affected by factors such as
size, form and moisture content of the grain. The review
of literature showed that there is a lack of information on
physical properties of gota for wide ranges of moisture
content.
Hence, the knowledge of these physical properties are
necessary for designing processing machines like cleaner,
grader and dehusker. The properties of different types
of grains and seeds have been determined by other re­
searchers also (Deshpande & Ojha, 1993; Dutta, Nema, Surface area
& Bhardwaj, 1988; Joshi, Das, & Mukherji, 1993; Oloso The equivalent surface area of gota was obtained using the
& Clarke, 1993; Singh & Goswami, 1996; Suthar & Das, geometric mean diameter by analogy of a sphere (Desh­
1996). They studied the physical properties of different pande et al., 1993; Nimkar et al., 2005) as:
­variety of pigeon peas but no one determined the physical
properties of dehusked unsplitted pigeon peas which fur­
ther undergoes splitting to yield dhal. Where, As = Surface area
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 119

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Thousand grain weight Statistical Analysis


Thousand grain weight was determined by randomly se­ The experimental results were subjected to analysis of
lecting 100 grains from the overall sample, measuring their ­variance (ANOVA) using AGRES (version 7.01) software
weight on a digital electronic balance with an accuracy and least significant difference test was used to describe
of 0.001 g, and multiplying by 10 to get the mass of 1000 the means with 95% confidence.
grains (Altuntas et al., 2005).

Density measurements Results and discussion


Bulk density (b) was considered as the ratio between the
mass of a sample of grain and the total volume occupied by Grain Dimensions
it. It was determined using a container of known volume The length (L), arithmetic mean diameter (Da) and geome­
(Deshpande et al., 1993; Vilche et al., 2003). tric mean diameter (Dg) increased significantly (p≤0.01)
True density (t), defined as the ratio between the mass from 5.23 to 5.51 mm, 4.52 to 4.64 mm and 4.49 to 4.59
of the sample grains and the actual volume occupied by it, mm, respectively with increase in moisture content from
was determined for five moisture contents in the range of 10.20 to 18.30 % db. However, the dimensions decreased
10.20% –28.20% db using toluene displacement method to 5.44 mm, 4.56 mm and 4.50 mm, respectively on increa­
with three replications (Singh and Goswami, 1996). This sing the moisture content to 28.20% db. In the moisture
is the most useful method worldwide for determination range between 10.20 to 24.30 % db, the width (W) increa­
of true density in agricultural commodity like all cereals, sed from 4.40 to 4.53 mm then decreased to 4.46 mm with
­pulses and oilseeds. the increase of moisture content to 28.20 % db of dehus­
Porosity () of the grain bed was defined as the frac­ ked unsplitted pigeon peas whereas thickness (T) values
tion of space in a bed of grains that is not occupied by the linearly decreased from 3.94 to 3.77 mm in the entire range
grains. The percentage porosity was calculated using the of this moisture content.
following equation (Mohsenin, 1986): The increase in size could be attributed to the expansi­
on of the grain as a result of moisture absorption in the in­
tercellular spaces inside the grains (Solomon and Zewdu,
2009). The dependence of these properties with moisture
content could be represented by the following equations:
Angle of repose
Angle of repose is the angle with the horizontal at which L = –0.0024m 2 + 0.1014m + 4.4494 R² = 0.9762
the material will stand when piled. The static angle of re­ W = –0.0011m 2 + 0.0475m + 4.0101 R² = 0.8302
pose was determined using a topless and bottomless cy­ T = –0.0002m 2 – 0.0019m + 3.9673 R² = 0.9739
linder of known dimensions. The cylinder was placed at D a = –0.0012m 2 + 0.049m + 4.1424 R² = 0.9866
the center of a raised circular plate and was filled with
Dg = –0.0011m 2 + 0.0443m + 4.1487 R² = 0.9731
dehusked unsplitted pigeon peas. The cylinder was raised
slowly until the grains formed a cone on the circular plate Where, m is the moisture content of grain in % db
of known diameter. To determine the dynamic angle of re­
pose, the dehusked unsplitted pigeon peas were allowed Similar trends were reported for increase in dimensions
to fall freely from a hopper over a disc of known diameter with increase in moisture content upto 18.30 % db. for ja­
to assume a natural slope. The static or dynamic angle of tropha, guna, chickpea, neem nut and barley (Dutta et al.,
repose was then calculated from the measurement of the 1988; Visvanathan et al., 1996; Aviara et al., 1999; Garnay­
height and the radius of the cone (Kaleemullah and Guna­ ak et al., 2008; Solugubik et al., 2013).
sekar, 2002) using the following relationship: Beyond the moisture content between 18.30 to 28.20%
db, the decrease in grain dimension might be due to voids
in the seed which when being filled with water make the
voids contract due to surface tension resulting in reduction
Where, Ø = Angle of repose, degrees
H = Height of cone formed, mm, and in dimensions.
D = Diameter of cone, mm
Sphericity
Static Coefficient of Friction Sphericity () was calculated from the geometric mean
Static coefficient of friction (μ) of gota was determined for diameter and the main axis (L) of gota. The results are pre­
the displacement of grains on three different materials, sented in Fig. 2. Sphericity displayed significant differences
­namely plywood, galvanized iron and mild steel, all with with change in moisture content of the grain. The values of
three replications. A wooden box was filled with gota and the sphericity decreased from 0.859 to 0.828 in the moistu­
placed on different surfaces as mentioned above. The total re range from 10.20 to 28.20 % db. According to Dutta et
weight required to move the box with grains was recorded. al. 1988 and Bal and Mishra 1988, the gota can be consi­
This value was used to calculate the static coefficient of dered as spherical since the sphericity value was more than
friction over different surfaces. 0.70. The decrease in sphericity might have been caused by
The total weight required to make the box with grain sli­ a proportional increase in the length as compare to width
de uniformly over the friction surface was used to measure and thickness of the gota. The relationship between mois­
the dynamic coefficient of friction (Kaleemullah, 1992). ture content (% db) and sphericity () of the grain could be
represented by the following polynomial equation:

 = 0.0002m 2 – 0.0078m + 0.9179 R² = 0.8912


Where, F = Force required just to move the box with grain, N,
and W = Weight of box + grain, N. Where m is the moisture content of grain in % db
120 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

FIGURE 2: Effect of moisture content on sphericity of gota.

crease in moisture content up to 28.20 % db. The relation


between surface area and moisture content is given by the
equation:

AS = –0.0319m 2 + 1.2611 m + 53.609 R 2 = 0.9723

Where m is the moisture content of grain in % db

Sologubik et al., 2013, Garnayak et al., 2008, Pradhan et


al., 2008, Zareiforoush et al., 2009, Altuntas et al., 2005
and Selvi et al., 2006 also reported similar trend in who­
le moisture range for grains of Scarlett barley, jatropha,
karanja kernel, rice, fenugreek and lin grain respectively.

Thousand grain weight


The variation of thousand grain weight of gota with
­moisture content, is presented in Fig. 4, which showed that
thousand grain weight increased from 61.95 g to 68.94 g as
the moisture content increased from 10.20 to 28.20% db
(p≤0.01). It was found to be a linear function of moisture
content and the relationship could be expressed using the
following equation:

Ws = 0.0003m 2 + 0.3768m + 57.981 R² = 0.9858


Where m is the moisture content of grain in % db

Similar linear increase had been noted by Dutta et al. 1988


for gram, Deshpande et al. 1993 for soybean, Singh and

FIGURE 1: E
 ffect of moisture content on (a) length (b) FIGURE 3: E
 ffect of moisture content on surface area of
width (c) thickness (d) arithmetic mean dia- gota.
meter and (e) geometric mean diameter of de-
husked unsplitted pigeon pea.
Similar trend in the entire moisture range 10.20% to
18.30% db were also observed where an initial increase
in sphericity of the grains, followed by its decrease, were
reported for okra, pea and barley (Sahoo and Srivastava,
2002; Yalcin et al., 2007; Sologubik et al., 2013).

Surface Area
The surface area increased from 63.30 to 66.13 m 2 with
increase in moisture content from 10.2% to 18.3% db FIGURE 4: E
 ffect of moisture content on thousand grain
moisture content and then decreased to 63.73 m 2 with in­ weight of gota.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 121

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Goswami 1996 for cumin grains, Altuntas and Yildiz 2007


for faba bean, Esref and Halil 2007 for red kidney bean
and Sologubik et al. 2013 for barley.

Bulk density
Bulk density (b) of gota at different moisture content va­
ried significantly (p≤0.01) from 859.5 to 781.7 kg/m 3 when
moisture content increased from 10.20 to 28.20% db Fig.
5 (a). This behavior could be attributed to the fact that the
increased mass of the sample associated with increased FIGURE 6: E
 ffect of moisture content on porosity of gota.
­moisture was lower than the volume expansion experien­
ced by the grains (Sologubik et al., 2013). This would have Porosity
resulted in having greater compaction (higher bulk densi­ It was observed that when moisture content increased
ty) in dry gota compared with wet gota. The relationship from 10.20 to 28.20% db, porosity decreased significant­
of bulk density (b) of gota and moisture content can be ly (p≤0.01) from 34.22 to 30.20% as shown in Fig. 6. The
expressed by the following equation: relationship between the value of porosity () and the mo­
isture content can be expressed as:
b = –0.0456m 2 – 2.4785m + 887.71 R² = 0.9856
 = 0.0009m 2 – 0.2947m + 37.591 R² = 0.8812
Where m is the moisture content of grain in % db
Where m is the moisture content of grain in % db
Parde et al. 2003 also found similar trend for Koto and
Manisoba cultivar at moisture content (wb.) range of
­ Similar behaviors were reported for beniseeds, pumpkin,
14.8% – 17.9% and 13.0% – 17.0%, respectively. Similar and pigeon pea seeds (Shepherd and Bhardwaj 1986).
relationships have been reported for chickpea (Konak et
al., 2002) and locust bean seed (Sobukola and Onwuka, Angle of repose
2010). The experimental results of angle of repose with respect
to moisture content are shown in Fig. 7, exhibiting a sig­
nificant increase of angle from 28.35° to 36.73° (p≤0.01)
True density with moisture content from 10.20 to 28.20% db. The trend
The true density of gota decreased from 1306.60 to 1128 could be due to the fact that moisture in the surface layer
kg/m 3 as the moisture content of the gota increased from of the grain kept them bound together by surface tension
10.20% to 28.20% db. It could be seen from Fig. 5 (b) that (Pradhan et al., 2008). The angle of repose is of para­
true density had a linear relationship with moisture con­ mount importance in the design of hopper openings, side
tent and could be represented as: walls and storage structures in the bulk of grains per ramp
(Solomon and Zewdu, 2009). The linear relationship bet­
t = 0.0298m 2 – 11.731m + 1429.9 R² = 0.9851 ween the angle of repose and the moisture content can be
described by the following equation:
Where m is the moisture content of grain in % db
 = 0.0203m 2 – 0.3008m + 29.194 R² = 0.9967
Similar trends of bulk density and true density have been Where m is the moisture content of grain in % db
reported for various materials like guna grains (Aviara et
al., 1999), green gram (Nimkar and Chattopadhyay, 2001), Similar behavior of the angle of repose with respect to moi­
cotton (Ozarslan, 2002), Lentil grain (Amin et al., 2004) sture content were observed for buckwheat (Koto, Koban
and barley (Sologubik et al., 2013). and Manisobacvs), barley, sorghum, jatropha and karanja
(Parde et al., 2003; Sologubik et al., 2013; Mwithiga and Si­
funa, 2006; Garnayak et al., 2008; Pradhan et al., 2008).

FIGURE 5: E
 ffect of moisture content on (a) bulk density FIGURE 7: E
 ffect of moisture content on angle of repose
and (b) true density of gota. of gota.
122 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Static Coefficient of Friction le of ­repose and static coefficient of friction of the gota
Variation of static coefficient of friction for gota on three were also significantly influenced by moisture content and
surfaces (Plywood, mild steel, and galvanized iron) with ­increased linearly with increase in moisture content in the
moisture content is shown in fig. 8. The static coefficient experimental moisture range. These properties will be
of friction increased significantly with moisture content very much helpful in designing different machineries for
for all the surfaces. This was due to the increased adhe­ pro­cessing of pigeon pea.
sion between the grain and the material surfaces at hig­
her moisture contents leading to higher values. Similar
results have been reported for faba bean (Altuntas and Conflict of interest
Yildiz 2007). The s­ tatic coefficient of friction ranged from
0.44 to 0.84, 0.47 to 0.78, and 0.28 to 0.67, respectively for There is no conflict among us.
plywood, mild steel, and galvanized iron surfaces in the
experimental ­moisture content range. Variation of static
coefficient of friction with moisture content of gota is ex­ References
pressed as:
Altunas E, Yildiz M (2007): Effect of moisture content on some
µGI = 0.0202m + 0.0396 R² = 0.8788 physical and mechanical properties of faba bean (Viciafabal)
µ MS = 0.0174m + 0.3465 R² = 0.8854 grains. Journal of Food Engineering, 78(1), 174–183.
µ PW = 0.024 m + 0.1659 R² = 0.9649 Altuntas E, Ozgoz E, Taser OF (2005): Some physical properties
of fenugreek (Trigonellafoenumgraceum L.) grains. J. Food
Eng., 71, 37–43.
Parde et al., 2003 found similar increase in the friction
Amin MN, Hossain MA, Roy KC (2004): Effect of moisture con­
coefficient of buckwheat (Koto cultivar) with moisture tent on some physical properties of lentil seed. Journal of Food
content. Other researchers also found that as the moistu­ Engineering, 65(1), 83–87.
re content increased, the static coefficient of friction also Annual report (2016–17): Department of Agriculture, Coopera­
increased (Baryeh, 2002; Altuntas and Yildiz, 2007; Prad­ tion and Farmers’ Welfare, Govt. Of India, as on 22-09-2017.
han et al., 2008). As stated by Thompson and Ross, 1983 Aviara NA, Gwandzang MI, Haque MA (1999): Physical proper­
and Lawton, 1980, at low moisture contents particles of ties of guna grains. J. Agric. Eng. Res., 73, 105–111.
grain tend to be elastic. As moisture content increased, Bal S, Mishra HN (1988): Engineering properties of soybean. In:
Proc. National Seminar on Soybean Processing and Utilization
the grain particles became more elastic and were able to
in India, 22–23 November 1986, Bhopal, India, 146–165.
deform requiring increased force to break the bonds bet­ Baryeh EA, Mangope BK (2002): Some Physical Properties of QP-
ween sliding grain and surfaces. 38 variety pigeon pea. Journal of Food Engineering, 56, 59–65.
Brooker DB, Bakker-Arkema FW, Hall CW (1992): Drying and
Storage of Grains and Oil Grains. Van Nostrand Reinhold,
New York, NY, pp: 450.
Dutta SK, Nema VK, Bhardwaj RJ (1988): Physical Properties
of gram. Journal of Agricultural Engineering Research, 39,
259–268.
Deshpande SD, Ojha TP (1993): Physical properties of soybean.
Journal of Agricultural Engineering Research, 56, 89–98.
Esref I, Halil U (2007): Moisture-dependent physical properties
of white speckled red kidney bean grains. J. Food Eng., 82,
209–216.
Garnayak DK, Pradhan RC, Naik SN, Bhatnagar N (2008): Mois­
ture-dependent physical properties of Jatropha grain (Jatropha
FIGURE 8: E
 ffect of moisture content on static coefficient curcas L.). Ind. Crops Prod., 27(1), 123–129.
of friction of gota. Joshi DC, Das SK, Mukherji RK (1993): Physical properties of
pumpkin seeds. Journal of Agricultural Engineering Research,
54, 219–229.
The coefficient of friction at all moisture contents was Kachru RP, Gupta RK, Alam A (1994): Physico-chemical Cons­
highest on mild steel followed by plywood and galvanized tituents and Engineering Properties of Food Crops. Scientific
iron. This was due to the smoother surface of galvanized Publishers, Jodhpur.
iron as compared to plywood and mild steel. The order Kaleemullah S (1992): The effect of moisture content on the physi­
reported for locust bean seed (Sobukola and Onwuka, cal properties of groundnut kernels. Tropical Sci., 32, 129–136.
2010) is plywood followed by mild steel and galvanized Kaleemullah S, Gunasekar JJ (2002): Moisture dependent physi­
iron sheet. However, Amin et al., 2004 have reported that cal properties of arecanut. Biosyst. Eng., 82(3), 331–338.
Konak M, Carman K, Aydin C (2002): Physical properties of chick
no variation existed between plywood and galvanized iron
pea seeds. BiosystEng, 82, 73–78.
for lentil seeds.
Lawton PJ (1980): Coefficient of friction between cereal grain and
various silo wall materials. J. Agric. Eng. Res., 25, 75–86.
Mohsenin NN (1986): Physical properties of plant and animal
Conclusion ­materials. New York: Gordon and Breach.
Mwithiga G, Sifuna MM (2006): Effect of moisture content on the
The values of physical dimension of gota such as length, physical properties of three varieties of sorghum grains. J. Food
Eng., 75, 480–486.
width, arithmetic mean diameter, geometric mean dia­
Nimkar PM, Chattopadhyay PK (2001): Some physical properties
meter and surface area increased up to moisture content of green gram. J. Agric. Eng. Res., 80(2), 183–189.
18.30% db significantly (p≤0.01) and then decreased, Nimkar PM, Mandwe DS, Dudhe RM (2005): Physical properties
­whereas thickness, sphericity, bulk density and true den­ of moth gram. Biosyst. Eng., 91, 183–189.
sity decreased (p≤0.01) with increasing moisture content Nwokolo E, Smartt J (1996): Food and feed from legumes and oil­
from 10.20% to 28.20% db. Thousand grain weight, ang­ seeds. London: Chapman and Hall.
Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124 123

The contents are protected by copyright. The distribution by unauthorized third parties is prohibited.

Oloso AO, Clarke B (1993): Some aspects of strength properties Sologubik CA, Campanone LA, Pagano AM, Gely MC (2013):
of cashew nuts. Journal of Agricultural Engineering Research, Effect of moisture content on some physical properties of bar­
55, 27–43. ley. Ind. Crops Prod., 43, 762–767.
Ozarslan C (2002): Physical properties of cotton grain. Biosyst. Thompson SA, Ross IJ (1983): Compressibility and frictional co­
Eng., 83, 169–174. efficient of wheat. Trans. ASAE, 26, 1171–1176.
Parde SR, Johal A, Jayas DS, White NDJ (2003): Physical proper­ Vilche C, Gely M, Santalla E (2003): Physical properties of quinoa
ties of buckwheat cultivars. Can. Biosyst. Eng., 45, 3.19–3.22. grains. Biosyst. Eng., 86(1), 59–65.
Pradhan RC, Naik SN, Bhatnagar N, Swain SK (2008): Moistu­ Visvanathan R, Palanisamy PT, Gothandapani V, Sreenarayanan
re-dependent physical properties of Karanja (Pongamiapinna­ VV (1996): Physical properties of neem nuts. Journal of Agri­
ta) kernel. Ind. Crops Prod., 28(2), 155–161. cultural Engineering Research, 63, 19–26.
Sahay KM, Singh KK (1994): Unit operations in agricultural pro­ Yalcin I, Ozarslan C, Akbas T (2007): Physical properties of pea
cessing. New Delhi: Vikas Publishing House Pvt Ltd. (Pisumsativum) grain. J. Food Eng., 79(2), 731–735.
Sahoo PK, Srivastava AP (2002): Physical properties of okra Zareiforoush H, Komarizadeh MH, Alizadeh MR (2009): Effect
grain. Biosyst. Eng., 83(4), 441–448. of moisture content on some physical properties of paddy
Selvi KC, Pinar Y, YeOilolu E (2006): Some physical properties of grains. Res. J. Appl. Sci. Eng. Technol., 1(3), 132–139.
lingrain. Biosyst. Eng., 95(4), 607–612.
Singh KK, Goswami TK (1996): Physical properties of cumin seed.
Journal of Agricultural Engineering Research, 64, 93–98.
Solomon WK, Zewdu AD (2009): Moisture-dependent physical
properties of niger (Guizotiaabyssinica Cass.) grain. Indus.
Crops Prod., 29, 165–170. Address of corresponding author:
Suthar SH, Das SK (1996): Some physical properties of karingda
Dr. Chandan Solanki
seeds. Journal of Agricultural Engineering Research, 65, 15–22.
Shepherd H, Bhardwaj RK (1986): Moisture dependent physical ICAR – Central Institute of Post-Harvest Engineering
properties of pigeon pea. Journal of Agricultural Engineering & Technology
Research, 35, 227–234. Ludhiana (Punjab) 141004
Sobukola OP, Onwuka VI (2010): Effect of moisture content on India
some physical properties of locust bean seed (Parkia fillicoidea chandan4uu12@gmail.com
L.). J Food Proc Eng. doi: 10.111/j.1745-4530.2009.00511.
124 Journal of Food Safety and Food Quality 70, Heft 4 (2019), Seiten 91–124

Impressum Richtlinien für Autoren


Journal of Food Safety
and Food Quality Detaillierte Richtlinien für Autoren sind im Internet
Archiv für Lebensmittelhygiene unter www.journal-food-safety.de oder www.vetline.de/afl/ zu finden.
70. Jahrgang · ISSN 0003-925X Die Schriftleitung behält sich vor, Beiträge zurückzuweisen oder den Autoren Vorschläge
Verlag:  Presse Dienstleistungsgesellschaft für Ände­rungen zu unterbreiten. Alle Beiträge werden von Spezialisten begutachtet („peer-­
mbH & Co. KG · Postfach 16 42, 31046 Alfeld · reviewed“). Die Schriftleitung bestimmt den Zeitpunkt der Veröffentlichung. Beiträge dürfen
Ravenstraße 45, 31061 Alfeld · Tel.: (0 51 81) 80 02-0,
Fax: (0 51 81) 80 02-55 · E-Mail: info@p-d-ges.de
nicht bereits ­veröffentlicht sein und nicht gleichzeitig anderen Verlagen oder sonstigen Stellen
© 2019 M. & H. Schaper GmbH, zum A­ bdruck a­ ngeboten werden.
Postfach 54 29, 30054 Hannover
Schriftleitung:  Prof. Dr. med. vet. Corinna Kehrenberg, Urheber- und Verlagsrechte
Institut für Tierärztliche Nahrungsmittelkunde,
Professur für Lebensmittelsicherheit und Verbraucher­ Alle in dieser Zeitschrift veröffentlichten Beiträge sind urheberrechtlich geschützt. Der Rechts­
schutz, Frankfurter Str. 92, 35392 Gießen; schutz gilt auch gegenüber Datenbanken und ähnlichen Einrichtungen. Kein Teil dieser Zeit­
Prof. Dr. E. Haunhorst, Niedersächsisches Landesamt
für Verbraucherschutz und Lebensmittelsicherheit, schrift darf außerhalb der jeweils geltenden engen Grenzen der Urheberrechtsgesetze der einzel­
­Postfach 39 49, 26029 Oldenburg nen L
­ änder ohne schriftliche Genehmigung des Verlages in irgendeiner Form – durch Fotokopie,
Schriftleitungsassistenz:  Dr. med. vet. Anja Müller, Mikrofilm oder andere Verfahren – reproduziert oder in eine von Maschinen, insbesondere von
I­ nstitut für Tierärztliche Nahrungsmittelkunde,
Justus-Liebig-Universität Gießen,
Datenverarbeitungsanlagen, verwendbare Sprache übertragen werden.
Frankfurter Str. 92, 35392 Gießen;
Dr. Anke Rottinghaus, Niedersächsisches Landesamt
für Verbraucherschutz und Lebensmittelsicherheit,
­Postfach 39 49, 26029 Oldenburg
Geschäftsführer:  Dipl.-Kfm. Ewald Dobler,
Checkliste für Autoren
Dipl.-Vw. Carolin Dobler
  Manuskripte sind dem Verlag in elektronischer Form einzureichen.
Anzeigenverkauf:  Carsten Sadlau, Tel.: (0 51 81) 80 02-53,
Fax: (0 51 81) 80 02-55, E-Mail: anzeigen@p-d-ges.de
Vertrieb:  Heiko Schaper   Abbildungen sind separat als hochauflösende Bilddaten zu liefern.
Herstellung:  Jens Rubrecht
Konten:  Sparkasse Hildesheim, IBAN: DE94 2595 0130   Das Manuskript enthält Wirkungsstätte, Name und Vorname der Autoren.
0010 0084 29, BIC: NOLA DE 21 HIK · Volksbank eG
Seesen, IBAN: DE89 2789 3760 0316 3199 00, BIC:
GENO DE F1 SES  Titel/Title, Zusammenfassungen/Summary und Schlüsselwörter/Keywords sind in
Erfüllungsort und Gerichtsstand für Lieferung ­deutscher und englischer Sprache anzugeben.
und Zahlung:  31061 Alfeld (Leine)
Bezugsbedingungen:  Das „Journal of Food Safety and
Food Quality“ erscheint jährlich sechsmal, seit 2017 aus-   Das Manuskript enthält die vollständige Korrespondenzadresse mit E-Mail-Adresse.
schließlich digital. Es ist verfügbar über www.journal-food-
safety.de. Bezugspreise: 150,– Euro für eine Einzelplatz­
lizenz; 175,– Euro für eine Mehrplatzlizenz für bis zu zehn
 Literatur im Text wird mit Namen und Jahr der Publikation (chronologisch) in
Nutzer; 200,– Euro für eine Campus-Lizenz für unbegrenzt ­K lammern angegeben (Müller und Frank, 1983; Albrecht, 1985; Schmitz et al., 1988)
viele Nutzer. Preis des Einzelheftes 25,– Euro (Hefte bis
zur Ausgabe 6/2016 können ja nach Verfügbarkeit gegen
oder bei Namensnennung des Autors im Text mit dem Erscheinungsjahr in Klammern
zusätzliche Portoberechnung auch gedruckt verschickt ergänzt.
werden; ansonsten ist der Kauf einer digitalen Ausgabe
möglich). Ermäßigter Bezugspreis für Studenten jährlich
98,– Euro. Diese Preise verstehen sich inkl. Mehrwert­  Alle im Text genannten Literaturquellen müssen im Literaturverzeichnis erscheinen
steuer. Abbestellungen sind nur bis sechs Wochen vor Ende und umgekehrt („cross-check“).
des Berechnungszeitraumes ­möglich. Wird das Erscheinen
durch höhere Gewalt oder Streik verhindert, so können
keine Ansprüche an den Verlag geltend gemacht werden.   Im Literaturverzeichnis ist nachfolgende Zitierweise anzuwenden:
Rechtliche Hinweise:  Eingereichte Arbeiten gehen in
allen Teilen ins Eigentum des Verlages über und dürfen
in derselben oder ähnlichen Form nicht anderweitig     – Zeitschriftenartikel: Sudhaus N, Pina-Pérez MC, Martínez A, Klein G (2012):
ange­boten noch andernorts ver­öffentlicht werden. Mit der ­I nactivation kinetics of spores of Bacillus cereus strains treated by a peracetic acid-­
Übergabe des Manuskripts tritt der Autor folgende ­Rechte
an den Verlag ab:
based disinfectant at different concentrations and temperatures. Foodborne Pathog
1. Bestand der Rechte: Der Verfasser versichert, dass er Dis 9: 442–452.
allein berechtigt ist, über die urheberrechtlichen Nut­
zungsrechte an seinem Beitrag einschließlich etwaiger
Bildvorlagen, Zeichnungen, Pläne, Karten, Skizzen und     – Buchkapitel: Shama G, Ahlfeld B (2012): Part I. Introduction to non-equilibrium
Tabellen zu verfügen und dass der Beitrag keine Rechte plasma, cell biology, and contamination: 5 Contamination. In: Laroussi M (ed.),
Dritter verletzt.
2. Dauer der Rechte: In Erweiterung von § 38 UrhG räumt ­Plasma medicine: applications of low-temperature gas plasmas in medicine and bio­
der Verfasser dem Verlag das ausschließliche Verlags­ logy. Cambridge, Cambridge University Press, UK, 99–114.
recht an seinen Beiträgen für die Dauer des gesetzlichen
Urheberrechts­schutzes ein (alternativ: für die Dauer von
drei Jahren ab Erscheinen).     – Buch: Rood JI, McClane BA, Songer JG, Titball RW (1997): The clostridia: mole­
3. Umfang der Rechte: Der Verfasser räumt dem Verlag
auch die folgenden Nutzungsrechte ein:
cular biology and pathogenesis. Academic Press, San Diego, CA, USA.
   a) Das Recht zur Übersetzung in andere Sprachen sowie
das Recht zur sonstigen Bearbeitung, insbesondere     – Dissertation: Unnerstad H (2001): Listeria monocytogenes – strain diversity demon­
zur EDV-gerechten Aufbereitung zum Zwecke der
Nutzung in neuen Medien wie Bildschirmtext, Video­ strated by genotyping. Uppsala, Sweden, Swedish University of Agricultural Sciences,
text, Datenbanken und dgl. sowie zur Erstellung von diss.
­Zusammenfassungen (abstracts) u. zur Herausgabe
als Mikrofilm, Mikrofiches, etc.
   b) Das Recht zur Veröffentlichung von Sonderdrucken     – Tagungsbericht: Baggesen DL, Wingstrand A, Thomsen LK, McFadden C, Nielsen B
und zu sonstiger Vervielfältigung, insbesondere
durch Fotokopie, sowie die von der VG Wort wahrge­ (1997): Salmonella contamination of carcasses from “Salmonella high risk pig herds“
nommenen Rechte einschließlich der entsprechenden and documentation of cross-contamination at abattoirs by use of epidemiological
Vergütungsansprüche.
   c) Das Recht zur Aufzeichnung auf Bild- und Ton­
markers. Proceedings of the International Symposium Salmonella and Salmonellosis,
träger sowie auf maschinenlesbare Datenträger, Ploufragan, France 1997, 321–324.
ferner das Recht zur elektronischen Speicherung in
Daten­banken sowie zur Ausgabe in körperlicher und
­unkörperlicher Form.
   d) Das Recht zur öffentlichen Wiedergabe in unkörper­
licher Form und das Recht zur Weitergabe der dem
Verlag eingeräumten Nutzungsrechte an Dritte. E-Mail-Adresse zur Online-Manuskripteinreichung
Für den Inhalt der Beiträge sind deren Verfasser verant­
wortlich. Die fachliche Aussage der Beiträge drückt nicht   journal-food-safety-quality@gmx.de
immer die Meinung der Schriftleitung aus.

View publication stats

Das könnte Ihnen auch gefallen