Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences
Publications of the University of Eastern Finland Dissertations in Forestry and Natural Sciences
isbn 978-952-61-0096-8
Jenni Korhonen
Antibiotic Resistance of Lactic Acid Bacteria
Resistance to antibiotics is a common characteristic in the world of bacteria.
Hitherto, only little attention has been paid to the antimicrobial
susceptibilities of beneficial lactic acid bacteria, on the opposite to several pathogenic bacterial species. However, there is some concern that antibiotic resistance in lactic acid bacteria could be transferred to pathogenic bacterial species, complicating the treatment of a diesease or infection and lead to the spread of antibiotic-resistant bacteria.
Studies were conducted to isolate and identify lactic acid bacteria of animal and human origins, to evaluate their antimicrobial resistance patterns and to propose minimum inhibitory concentrations for Lactobacillus species.
dissertations | 007 | Jenni Korhonen | Antibiotic Resistance of Lactic Acid Bacteria
Jenni Korhonen
Antibiotic Resistance of
Lactic Acid Bacteria
JENNI KORHONEN
Ant ibiot ic Resist ance of Lact ic Acid Bact eria
Pu blications of the University of Eastern Finland Dissertations in Forestry and N atu ral Sciences
7
To be presented by permission of the Facu lty of Forestry and Natural Sciences
Dep artm ent of Biosciences
University of Eastern Finland , Ku op io, on Saturday 24thAp ril 2010, at 12 noon.
University of Eastern Finland for public exam ination in the Au d itoriu m L22, Snellm ania Bu ild ing, ,
University of Eastern Finland Kuopio
2010
Kop ijyvä Ku op io, 2010 Ed itors: Prof. Pertti Pasanen Prof. Tarja Lehto, Prof. Kai Peip onen
Distribu tion:
Eastern Finland University Library / Sales of p u blications P.O.Box 107, FI-80101 Joensu u , Finland
tel. +358-50-3058396 http :/ / w w w .u ef.fi/ kirjasto
Print:
ISBN : 978-952-61-0096-8 ISSN L: 1798-5668
ISSN : 1798-5668
Online:
ISBN : 978-952-61-0097-5 (PDF) ISSN : 1798-5676 (PDF)
Au thor´s ad d ress: University of Eastern Finland
Dep artm ent of Biosciences
P.O.Box 1627
70211 KUOPIO
FIN LAN D
em ail: jenni.korhon en@u ef.fi
Su p ervisors: Professor Atte von Wright, Ph.D.
University of Eastern Finland
Dep artm ent of Biosciences
P.O.Box 1627
70211 KUOPIO
FIN LAN D
em ail: atte.vonw right@u ef.fi
Professor Sinikka Pelkonen, Ph.D.
Research Dep artm ent of Veterinary Bacteriology Finnish Food Safety Au thority Evira
N eu laniem entie 4 70150 KUOPIO FIN LAN D
em ail: sinikka.p elkonen@evira.fi
Review ers: Professor Wolfgang Kneifel
BOKU-University of N atu ral Resou rces and Ap p lied Life Sciences Dep artm ent of Food Science and Technology
VIEN N A AUSTRIA
em ail: w olfgang.kneifel@boku .ac.at
Associate Professor Arthu r Ouw ehand University of Tu rku
Fu nctional Food s Foru m TURKU
FIN LAN D
em ail: arthu r.ou w ehand @u tu .fi
Op p onent: Professor Sven Lind gren N ational Food Ad m inistration UPPSALA
SWEDEN
em ail: sven.e.lind gren@telia.com
ABSTRACT:
Lactic acid bacteria (LAB) are a heterogeneou s grou p of bacteria fou nd w id ely in natu re . They colonize the gastrointestinal and u rogenital tracts of hu m ans and anim als, and are p resent in food s su ch as d airy p rod u cts, ferm ented m eats, fru its and vegetables. LAB are also intentionally ad d ed to several p robiotic p rod u cts becau se of their p otential health benefits. Many LAB sp ecies are generally recognized as safe (GRA S), and several LAB sp ecies have received a Qu alified Presu m p tion of Safety (QPS) statu s given by Eu rop ean Food Safety Au thority (EFSA).
Resistance to antim icrobial d ru gs (antibiotics) is a com m on character istic in the w orld of bacteria.
In the interaction betw een bacteria, genetic m aterial is transferred from one bacteriu m to another, and also genes cod ing for resistance to a certain antibiotic m ay be p assed on to other bacterial sp ecies. Since LAB are natu ral and p rofitable inhabitants in m any environm ents (gastrointestinal tract, several food s), strains w ith resistance to an tibiotics w ou ld not be d etrim ental to the w ellbeing of hu m ans or anim als. H ow ever, there is som e concern that antibiotic resistance in LAB cou ld then be transferred to p ossibly p athogenic bacterial sp ecies, com p licatin g the treatm ent of a d isease or infection and lead to the spread of antibiotic-resistant bacteria.
H itherto, only little attention h as been p aid to the antim icrobial su scep tibilities of LAB (Eu rop ean Com m ission. 2005, Eu rop ean Com m ission 2008).
In ord er to illu strate the cu rrent situ ation of antibiotic resistance p atterns in beneficial LAB, stu d ies w ere first cond u cted to isolate and id entify lactic acid bacteria of anim al and hu m an origins, and second ly to evalu ate their antim icrobial resistance p atterns. Moreover, tentative cu t- off valu es d ivid ing the p op u lations into su scep tible and resistant w ere also p rop osed based on m inim u m inhibitory concentrations (MIC) for a total of fou rteen Lactobacillus sp ecies.
Most of the LAB strains w ere fou nd to be su scep tible to all antim icrobial agents u sed in the stu d ies. The m ost frequ ently fou nd resistance w as against tetracycline, follow ed by resistance against am inoglycosid es. The horizontal transferability of antibiotic resistance betw een LAB (Lactococcus garvieae sp ecies) and p athogenic Listeria monocytogenes w as also achieved , d em onstrating that a gene cod ing for tetracycline resistance tet(S) can be transferred from a fish p athogen to a hu m an p athogen by conju gation in vitro.
Universal Decimal Classification:579.864, 615.015.8, 615.33, 577.18 N ational Library of M edicine Classification: QW 142.5.A 8, QW 45
CA B Thesaurus:lactic acid bacteria; Lactobacillus; drug resistance; antibiotics; tetracycline;
aminoglycoside antibiotics; horizontal transmission; Lactococcus garvieae, Listeria monocytogenes
TIIVISTELMÄ:
Maitohap p obakteerien vastu stu skykyä antibiootteja kohtaan selvitettiin p orsaid en, vasikoid en ja ihm isten su olistosta sekä elintarvikkeista eristetyistä kannoista. Tu losten p eru steella ehd otettiin neljälletoista laktobasillilajille raja-arvoja, joita voitaisiin käyttää jaettaessa kantoja herkkiin ja vastu stu skykyisiin. Eniten vastu stu skykyisyyttä havaittiin tetrasykliinille sekä am inoglykosid eille. Lisäksi havaittiin, että antibioottivastu stu skyky voi siirtyä lajien välillä laboratorio-olosu hteissa Lactococcus garvieae – bakteerin ja ihm iselle tau tia aiheu ttavan Listeria monocytogenes –bakteerin välillä.
Y leinen suomalainen asiasanasto: maitohappobakteerit; resistenssi -- antibiootit
In nature we never see anything isolated, but everything in connection with something else---
Johann Wolfgang von Goethe
ACKN OWLED GEMEN TS
This w ork w as carried ou t in the Dep arm ent of Forestry and N atu ral Sciences, University of Eastern Finland (form erly Dep artm ent of Biosciences, N u trition and Biotech nology, University of Ku op io). My w ork w as financially su p p orted by Eu rop ean Union 6th Fram ew ork Program p roject “Assessm ent and Critical Evalu ation of Antibiotic Resistance Transferability in Food Chain, ACE-ART”, and also by Finnish Cu ltu ral Fou nd ation.
In p articu lar, I w ish to exp ress m y sincere gratitu d e to m y p rincip al su p ervisor, Professor Atte von Wright, for giving m e the op p ortu nity to w ork c om p aratively ind ep end ently and still having his fu ll su p p ort w hen need ed . Thank you for sharing you r tim e and thou ghts w ith m e d u ring these years. Sincere thanks go also to m y other su p ervisor, Professor Sinikka Pelkonen . Sincere thanks to Pau la H yv önen, for her com m ents on the thesis, and you r g u id ance and infinite su p p ort d u ring m y stu d ies and w ork at the d ep artm ent.
I ow e m y thanks to the official referees, Professor Wolfgang Kneifel and Associate Professor Arthu r Ou w ehand , for agreeing to review m y thesis. Thank you for you r constru ctive criticism s. Thanks also to Ow en MacDonald for langu age consu lting.
I w ish to thank all the former “ACE-ARTists”, esp ecially all co-au thors arou nd Eu rop e and also in the d ep artm ent. Thank you for the p leasant collaboration and you r efforts in scientific research. Many of you r com m ents have essentially im p roved the p u blished p ap ers. Sp ecial thanks for thorou gh review ing to Angela van H oek, Geert H u ys and Sigrid Mayrhofer, and also to Mia Egervärn for nice collaboration.
Particu lar thanks to m em bers of ou r “su bu nit”, N u trition and Food Biotechnology, for the p leasant w orking atm osp here. Sp ecial thanks to Carm e Plu m ed Ferrer, Kristiina Kinnu nen and Jou ni H eikkinen, for sharing you r exp ertise on lactic acid bacteria. For their technical assistance, I w ou ld like to thank Mirja Rekola, Elvi Mäkirinne, Riitta Venäläinen and Eeva-Liisa Palkisp ää.
I w ou ld like to thank m y m other and father, Marjatta and Sakari, and m y m other - and father-in-law , Eeva and Veikko, for their care and su p p ort d u ring m y stu d ies and w ork.
A clean hou se, d inner read y and ch ild ren hap p y, after som etim es long hou rs at w ork.
What m ore cou ld you ask!
Finally, I w ish to thank ou r child ren, Aino, Lau ri and Elsa, for bringing m e the joy, and giving m e a p ersp ective of life in general. To m y hu sband Petri, a sp ecial thank you for those constru ctive argu m ents that w e both so m u ch enjoy, no m at ter the su bject! Thank you for you r u ncond itional love and su p p ort d u ring ou r years together.
Ku op io, Ap ril 2010
Jenni Korhonen
ABBREVIATION S
ACE-ART Assessm ent and critical evalu ation of antibiotic resistance transferability in food chain
ATCC Am erican Typ e Cu ltu re Collection BLAST Basic local alignm ent search tool CAMBH cation ad ju sted Mü ller-H inton agar CFU Colony form ing u nit
CLSI Clinical and Laboratorty Stand ard s Institu te EFSA Eu rop ean Food Safety Au thority
EMBL Eu rop ean Molecu lar Biology Laboratory
FEEDAP Panel on Ad d itives and Prod u cts or Su bstances u sed in Anim al Feed GIT Gastrointestinal tract
GRAS Generally recognized as safe LAB Lactic acid bacteria
LAMVAB Lactobacillus Anaerobic MRS w ith Vancom ycin and Brom ocresol green L. Listeria
Lb. Lactobacillus Lc. Lactococcus
LSM Lactic acid bacteria su scep tibility m ed iu m MIC Minim u m inhibitory concentration MRS d e Man, Rogosa and Sharp e MTT Agrifood Research Finland
N CCLS N ational Com m ittee on Clinical Laboratory Stand ard s
OD Op tical d ensity
PCR Polym erase ch ain reaction QPS Qu alified p resu m p tion of safety
RAPD Rand om ly am p lified p olym orp hism DN A rp m Revolu tions p er m inu te
TC transconju gant
TF transform ant
W . W eissella
LIST OF ORIGINAL PUBLICATIONS
This thesis in based on d ata p resented in the follow ing articles, referred to by their Rom an nu m erals.
I Korhonen, J.M., Sclivagnotis, Y. and von Wright, A. (2007)
Characterization of d om inant cu ltivable lactobacilli and their antibiotic resistance p rofiles from faecal sam p les of w eaning p iglets. Journal of A pplied M icrobiology 103: 2496-2503.
II Korhonen, J.M., H assel, A., Kom u lainen, H . and von Wright, A.
Antim icrobial su scep tibility of lactic acid bacteria from fecal sam p les of w eaned calves. M anuscript.
III Korhonen, J.M., van H oek, A.H .A.M., Saarela, M., H u ys, G., Tosi, L.,
Mayrhofer, S. and von Wright, A. (2010) Antim icrobial su scep tibility of Lactobacillus rhamnosus. Beneficial M icrobes 1: 75-80.
IV Korhonen, J.M., Danielsen, M., Mayo, B., Egervärn, M., Axelsson, L.,
H u ys, G. and von Wright, A. (2008) Antim icrobial su scep tibility and p rop osed m icrobiological cu t-off valu es of lactobacilli by p henotyp ic d eterm ination. International Journal of Probiotics and Prebiotics 3: 257-268.
V Gu glielm etti, E., Korhonen, J.M., H eikkinen, J., Morelli, L. and von Wright, A. (2008) Transfer of p lasm id -m ed iated resistance to tetracycline in p athogenic bacteria from fish and aqu acu ltu re environm ents. FEM S M icrobiology Letters 293: 28-34.
Cont ent s
1. Introduction
... 152. Review of the literature
... 172.1 ANTIBIOTIC RESISTANCE IN THE FOOD CHAIN ... 17
2.1.1 M echanisms of antibiotic resistance ... 19
2.1.2 A ntibiotic resistance determinations ... 20
2.2 LACTIC ACID BACTERIA ... 21
2.2.1 Identification of LA B ... 23
2.2.2 LA B of human and animal origins ... 24
2.2.3 Safety of LA B ... 25
2.3 ANTIBIOTIC RESISTANCE OF LAB ... 27
2.3.1 Phenotypic antibiotic resistance of LA B ... 27
2.3.2 Genotypic antibiotic resistance of LA B ... 29
2.3.3 Horizontal transferability of antibiotic resistance from LA B in the food chain . 29
3. Aims of the study
... 314. Materials and methods
... 324.1 BIOLOGICAL ORIGINS OF LAB STRAINS (I-V) ... 32
4.2 ISOLATION AND IDENTIFICATION OF LAB (I-IV) ... 33
4.2.1 Sampling and culturing of LA B from faecal samples (I-III) ... 33
4.2.2 Identification with biochemical methods (I-III) ... 33
4.2.3 Identification with molecular biology methods (I-III, V ) ... 33
4.2.4 LA B strain characterization (I-II) ... 34
4.3 MICROARRAY ASSAY FOR LB. RHAMNOSUS (III) ... 34
4.4 PLASMID TRANSFORMATION AND MATING PROTOCOL OF LACTOCOCCUS
SPECIES (V) ... 35
4.5 ANTIMICROBIAL RESISTANCE DETERMINATIONS (I-IV) ... 36
4.5.1 A gar dilution method (I-III) ... 36
4.5.2 Broth microdilution method (III-V ) ... 37
4.5.3 Etest method (III-IV ) ... 37
5. Results and discussion
... 385.1 SPECIES DISTRIBUTION AMONG THE FAECAL SAMPLES OF PIGLETS AND CALVES (I-II) ... 38
5.2 RAPD FINGERPRINTS OF PIGLET AND CALF ISOLATES (I-II) ... 39
5.3 ANTIMICROBIAL SUSCEPTIBILITY PROFILES OF LACTOBACILLUS SPECIES OF PIGLET ORIGIN (I) ... 44
5.4 ANTIMICROBIAL SUSCEPTIBILITY PROFILES OF LAB SPECIES OF CALF ORIGIN (II) ... 46
5.5 ANTIMICROBIAL SUSCEPTIBILITY PROFILES OF LACTOBACILLUS RHAMNOSUS AND PROPOSED MICROBIOLOGICAL CUT-OFF VALUES (III) 52 5.6 ANTIMICROBIAL SUSCEPTIBILITY PROFILES OF LACTOBACILLUS SPECIES AND PROPOSED MICROBIOLOGICAL CUT-OFF VALUES (IV) ... 53
5.7 ANTAGONISTIC ACTIVITY AND TRANSFORMATION OF LACTOCOCCUS SPECIES (V) ... 58
5.8 TRANSFER OF THE TET(S) GENE FROM LACTOCOCCUS TO L. MONOCYTOGENES (V) ... 59
6. Conclusions
... 607. References ...
621. Introduction
Lactic acid bacteria are a heterogeneou s group of bacteria , m any of them having received a generally recognized as safe (GRAS) or qualified p resu m p tion of safety (QPS) –statu s. These bacteria are w id ely found in natu re, inclu d ing the gastrointestinal and u rogenital tracts of hu m ans and anim als, and are p resent in m any ferm ented food s like salted gherkins, m arinated olives, cap ers and salam i, and d ifferent m ilk based prod u cts su ch as cheeses and yoghu rts. Lactic acid bacteria have trad itionally been associated w ith these d airy p rod u cts and w ith cereal-, vegetable- and m eat-based ferm ented food s, either as intentionally ad d ed starters or d u e to their natu ral p resence lead ing to sp ontaneou s ferm entation. Certain lactic acid bacteria are also u sed as p robiotics ad d ed to confer health benefits to consu m ers or to im p rove anim al p rod u ction. In this respect, lactic acid bacteria sp ecies are econom ically very im p ortant to the food and feed ind u stry.
Du ring the recent d ecad es, there has been concern abou t the p ossibility of the sp read of antibiotic resistance in the environm ent. Accord ing to the Eu rop ean Com m ission (2005), it has been estim ated that som ew here from one to ten m illion tons of antibiotics have been released into the biosp here over the last 60 years. This has lead to a very strong selective p ressu re for the ap p earance of resistant bacterial strains. Mu ch of the concern has been abou t p athogenic bacteria and their antibiotic resistances, since infections cau sed by these resistant m icro-organism s are not only m ore com p licated to treat, bu t the treatm ent is m u ch m ore costly d u e to th e m ore intensive and tim e consum ing care need ed in these cases.
Since lactic acid bacteria are p resent in the gastrointestinal tract in large am ou nts and are also intentionally ad d ed to ou r d iet, concerns have been raised abou t the antibiotic resistance in these beneficial bacterial sp ecies. For exam p le, lactic acid bacteria resistant to certain antibiotics cou ld benefit the host (hu m an or anim al) by help ing to m aintain balance in the gastrointestinal tract in cases of d iarrhea cau sed by antibiotic treatm ent. H ow ever, there is a risk associated w ith the ability of these resistant strains to transm it the resistance factor (gene) to other , p ossibly p athogenic bacteria. This cou ld com p licate the treatm ent of a p atient w ith an antibiotic resistant bacterial infection or d isease. H ence, the p ossibility of the circu lation of genes cod ing for antibiotic resistance also from beneficial lactic acid bacteria, in the food chain via anim als to hu m ans, has been investigated .
At the beginning of 2004, the Eu rop ean Union lau nched a 6th fram ew ork p roject
“Assessm ent and Critical Evalu ation of Antibiotic Resistance Transferability in Food Chain” (ACE-ART), in w hich m ost of the w ork inclu d ed in this thesis, w as cond u cted . The m ajor objective of this p roject w as to critically evalu ate the im p act of antibiotic u se in anim al and plant p rod u ction and in the p rophylaxis and treatm ent of d isease in hu m ans u sing non -p athogenic lactic acid bacteria as m od el organism s. In ord er to obtain a w id e p ersp ective of the cu rrent situ ation of antibiotic resistance p atterns in lactic acid bacteria, a large array of these beneficial bacterial strains belonging to Lactobacillus, Lactococcus and Streptococcus thermophilus, together w ith Bifidobacterium, w ere inclu d ed in the stu d ies of ACE-ART -p roject.
In this thesis, the antibiotic resistance p rofiles from faecal sam p les of w eaning p iglets and calves w ere d eterm ined . In ad d ition, the antibiotic resistance of Lactobacillus rhamnosus strains from variou s sou rces (m ostly hu m an and food origins) w as stu d ied . In su m m ary, antim icrobial su scep tibility and p rop osed m icrobiological cu t-off valu es of lactobacilli by phenotypic d eterm ination w ere p erform ed , containing both p u blished and unp u blished d ata of fou rteen Lactobacillus sp ecies w ith their p henotyp ic antibiotic resistance p henotyp es.
Moreover, in ord er to d etect the horizontal transfer of antibiotic resistance genes in the food chain, in vitro transfer of a p lasm id d erived tetracycline resistance gene, tet(S), w as achieved from Lactococcus garvieae to the p athogenic Listeria monocytogenes.
2. Review of the literature
2.1 ANTIBIOTIC RESISTANCE IN THE FOOD CHAIN
There is a close association betw een the qu antities of antim icrobials being u sed and the rate of d evelopm ent of resistance to these su bstance s and thus the m isu se of antibiotics in hu m an m ed icine is believed to be the p rincip al cau se of the antibiotic resistance p roblem (Singer et al. 2003). Another asp ect, how ever, is the selection of resistant bacteria in the food chain d u e to the heavy u tilization of antim icrobial agents in anim al hu sband ry (Teale 2002).
As early as 1969, the Sw ann rep ort (Anonym . 1969) d rew attention to the p otential transfer of antibiotic resistan t bacteria from anim als to the food chain and to hu m ans. In anim als, antibiotics are u sed for three d ifferent p u rp oses: to treat sick anim als (therap eu tic u se), to p revent infection in anim als (p rop hylactic u se) and to im p rove feed u tilization and p rod u ction, in other w ord s, to convert feed into m ore bod y m ass (grow th p rom oters) ( Barton 2000, Singer et al. 2003). The therap eu tic u se of antibiotics is not qu estioned in farm anim als, bu t the view s of exp erts d iffer on the u se of antibiotics as anim al grow th p rom oters, and there is d ebate w hether the banning of these su bstances w ou ld have beneficial effects on hu m an health. In the Eu rop ean Union, avop arcin (a glycop ep tid e, w hich p rod u ces cross-resistance to vancom ycin), virginiam ycin (a strep togram in), bacitracin (also u sed in hu m an m ed icine), tylosin and sp iram ycin (both m acrolid es) have been banned as feed ad d itives (Teale 2002, N ou siainen et al. 2004), and the EU is m oving tow ard s a total restriction on grow th prom oters, thu s fu lfilling the recom m end ation of the Sw ann rep ort (1969) w here it w as stated that antibiotics u sed in hu m an m ed icine shou ld not be u sed as grow th p rom oters. N onetheless, the recom m end ations of the Sw ann rep ort have not been ad op ted in m any cou ntries ou tsid e the EU (Teale 2002). For exam p le, in the USA a total of 19 d ifferent antibiotics are allow ed to be u sed for grow th p rom otion , inclu d ing several antibiotics, like p enicillin and strep tom ycin, w hich are u sed also in hu m an m ed icine (Singer et al. 2003).
If an antibiotic resistance is to cau se a d isease in h u m ans via the food chain, certain events have to take p la ce (Figu re 1). First, there has to be an antibiotic resistant bacteriu m and a selection p ressu re p resent in a p articu lar grou p of anim als. The bacteriu m has to rem ain in the food p r ocess, throu gh , for exam p le, faecal contam ination in the p rocess or via a recontam ination by im p rop er hand ling, or lack of an ad equ ate tem p eratu re treatm ent in the p rocess. After con su m p tion, the resistant bacterium has to colonize the GI-tract.
(Singer et al. 2003) In conclu sion, the transm ission of antibiotic resistance via the food chain is the sam e as for food borne p athogens. The risk of transm ission
of antibiotic resistance from anim als to hu m ans is consid erably red u ced w ith p rop er food hand ling and good food p rep aration p ractices. (Singer et al. 2003) It shou ld also be noted that the u se of antim icrobials is not restricted to anim al hu sband ry bu t also occu rs in horticu ltu re, for exam p le the u se of am inoglycosid es in ap p le grow ing (Teale 2002).
Figure 1. Possible transmission routes of antibiotic resistance bacteria from animals to humans. Modified from Khachatourians 1998, Anonym 2004 and Claycamp and Hooberman 2004.
In Finland , the u se of antim icrobial d ru gs for anim als has rem ained rather stable for several years; althou gh a slight increase m ay have occu rred . Accord ing to N ational Agency for Med icines, the consu m p tion of veterinary antim icrobials w as 15 100 kg year 2007, calcu lated as w eight of active su bstance. In com p arison, in the year 2001 the consu m p tion w as 13 800 kg. The p ossible reason for the increm ent rem ains to be clarified , i.e. is it becau se of an actu al increase in antim icrobial consu m p tion or sim p ly d u e to statistical variation. The overall good resistance situ ation can be traced to the strict antim icrobial p olicy. H ow ever, p ressu re for increasing the u se of antibiotics in Finland w ill p robably rise in the fu tu re. This is m ainly becau se of increased herd size and the trend tow ard s m ore inten sive livestock p rod u ction.
(Myllyniem i et al. 2007)
2.1.1 Mechanisms of ant ibiot ic resist ance
The genetic basis for the d evelop m ent of antibiotic resistance in bacteria in the food chain is based on tw o facts. Firstly, the bacteria m u st com e into contact w ith the antim icrobial agent in concern, and second ly, resistance against the antibiotic m u st d evelop (Khachatou rians 1998, Levy and Marshall 2004).
Resistance against a certain antim icrobial agent can be inherent in a bacterial sp ecies, this being referred to as intrinsic resistance, or “natu ral resistance”. In this case, the resistance is typ ical for all of the strains of that p articu lar species.
In contrast, the resistance is consid ered as acqu ired , w hen a strain of a norm ally su scep tible sp ecies becom es resistant to an antim icrobial d ru g . (Eu rop ean Com m ission 2008)
The antibiotic resistance genes can b e sp read from one bacteriu m to another throu gh several m echanism s. Intrinsic resistance is estim ated to p resent a m inim al p otential for horizontal sp read (betw een d ifferent bacterial sp ecies), as has been d em onstrated for exam p le w ith the chrom osom al vancom ycin resistance d eterm inant of the Lactobacillus rhamnosus strain GG (Tynkkynen et al. 1998). Sim ilar to intrinsic resistance, acqu ired resistance u su ally p ossesses a low risk of horizontal d issem ination, w hen the resistance is a resu lt of a chrom osom al m u tation. In contrast, acqu ired resistance is consid ered as having a higher p otential for horizontal d issem ination of antibiotic resistance, w hen the resistance genes are p resent on m obile genetic elem ents (plasm id s and transp osons). (Khachatou rians 1998, Eu rop ean Com m ission 2008)
Antim icrobial d ru gs can be d ivid ed into d ifferent grou p s based on their m echanism s of action. Tod ay there are m ore than 250 antibiotics available for therap eu tic u se (m ore than 100 of those are β-lactam s), bu t these d ru gs act only against a few d ifferent bacterial target sites (van d en Bogaard and Stobberingh 1996). These sites of action inclu d e cell w all synthesis, p rotein s ynthesis (targeting for 30S, 50S or tRN A), DN A gyrase or folic acid m etabolism (N eu 1992). The m ajor antibiotic fam ilies and their target of action relevant to this thesis are show n in Table 1.
Table 1. Major antibiotic families and their target of action.
Sites of inhibition Group Antibiotic
Cell wall synthesis β-lactams Amoxicillin (AMO)
Ampicillin (AMP) Penicillin C (PEN) Glycopeptides Vancomycin (VAN)
Protein synthesis Aminoglycosides Gentamicin (GEN) Kanamycin (KAN) Neomycin (NEO) Streptomycin (STR) Chloramphenicols Chloramphenicol (CHL) Tetracyclines Tetracycline (TET) Macrolides Erythromycin (ERY)
Tylosin (TYL) Lincosamides Clindamycin (CLI)
Lincomycin (LIN)
DNA replication/transcription Quinolones Enrofloxacin (ENR)
Folate synthesis Sulphonamides Trimethoprim (TMP)
2.1.2 Ant ibiot ic resist ance det erminat ions
Antim icrobial su scep tibility testing m ay be p erform ed u sing d ifferent p henotypic test m ethod s. In CLSI (Clinical and Laboratory Stand ard s Institu te, form erly N CCLS, N ational Com m ittee on Clinical Laboratory Stand ard s) the ap p roved stand ard s state that the m ethod s of choice are agar d ilu tion and broth m icrod ilu tion (Anonym . 2007). Other w id ely u sed m ethod s includ e the agar grad ient m ethod and com m ercial m ethod s, su ch as Etest, w hich consists of a p red efined grad ient of antibiotic concentrations on a p lastic strip (AbBiom erieu x, Sw ed en).
In ad d ition to phenotyp ic antibiotic resistance d eterm inations, also genotyp ic d etection of particu lar genes cau sing resista nce m ay be p erform ed . These genotyp ic m ethod s inclu d e d ifferent PCR –based m ethod s, sou thern hybrid ization, p lasm id profiling and m icroarray (Aqu ilanti et al. 2007, Am m or et al. 2008).
The situ ation is clearest w hen the phenotypic and genotyp ic resistance p atterns are in agreem ent. H ow ever, a p henotyp ically resistant bacteriu m strain m ay be genotyp ically “su scep tible”. This is u su ally d u e to the fact that ap p rop riate genes are not inclu d ed in the test p atterns, or th ere exist u nknow n resistance genes. Tetracycline, for exam p le, has m ore than 40 d ifferent genes conferring antibiotic resistance d iscovered at the m om ent, and the nu m ber of tetracycline resistance genes continu es to increase (Roberts 2005). With tetracycline, also new m osaic genes have recently been d iscovered (Patterson et al. 2007, van H oek et al. 2008a). In contrast, a suscep tible p henotyp e m ay also carry silent genes, w hich are observed w ith genotyp ing. The silence of antibiotic resistance m ay be cau sed by d ow n -regu lation in a p rom oter region or
by other m echanism s. Desp ite of their “silence”, they still cou ld be a p otential concern since they cou ld be transferred to other sp ecies w here they w ou ld be activated .
2.2 LACTIC ACID BACTERIA
Lactic acid bacteria (LAB) are gram -p ositive, acid -tolerant and non -sp ore form ing cocci and rod s. They are a heterogeneous grou p of bacteria com prising abou t 20 genera w ithin the p hylu m Firm icu tes. From a p ractical p oint of view the genera A erococcus, Carnobacterium, Enterococcus, Lactobacillus, Lactococcus, Leuconostoc, Oenococcus, Pediococcus, Streptococcus, Tetragenococcus, V agococcus and W eissella have been consid ered as the p rincip al LAB. (H olzap fel et al. 2001, Axelsson 2004)
One com m on featu re of the LAB is their ability to p rod u ce lactic acid as a m ajor end p rod u ct of their ferm entation of hexoses. As ferm enting organism s, LAB lack electron transport system s and cytochrom es, and they d o not have a fu nctional Krebs cycle (Batt 2000). Based on end p rod u cts of glu cose m etabolism (Figu re 2), LAB can be d ivid ed into tw o grou p s, nam ely hom oferm entative and heteroferm entative (Jay 2000).
LAB are w id esp read organism s and they m ay be fou nd in m any environm ents rich in carbohyd rates. In ad d ition to carbohyd rates, LAB have com p lex nu tritional requ irem ents for am inoacid s, p ep tid es, fatty acid esters, salts, nu cleic acid d erivatives and vitam ins (Tannock 2004). In short, they have com p lex nu tritional requ irem ents d u e to their lack of m any biosynthetic p athw ays. On the other hand , they are fou nd in a w id e range of d ifferent environm ental niches d u e to their good cap acity for ad ap tation. In food p rod u cts, they are fou nd in d airy p rod u cts, such as yoghu rt and cheese, in ferm ented vegetables (olives, sau erkrau t), in ferm ented m e ats (salam i) and in sou rd ou gh bread (Tannock 2004). They are also associated w ith both terrestrial and m arine anim als (see chap ter 2.2.2, LAB of hum an and anim al origins).
Figure 2. Generalized pathways for the production of fermentation products from glucose by A: homofermentative LAB and B: heterofermentative LAB (Kandler 1983).
Lact obacillus species
The genu s Lactobacillus is the largest LAB grou p com p rising at the m om ent arou nd 140 sp ecies and 30 su bsp ecies (Bernard eau et al. 2008, Claesson et al.
2008). These nu m bers are constantly being revalu ated on the basis of m od ern m olecu lar biology m ethod s and w hole genom e-based techniqu es (Makarova et al. 2006, Felis and Dellaglio 2007).
Based on their ferm entation characters, Lactobacillus sp ecies can be d ivid ed into three grou p s; obligately hom oferm entative, facu ltatively heteroferm entative and obligately heteroferm entative (H am m es and Vogel 1995, Axelsson 2004). Moreover, if one u ses 16S p hylogeny, then the Lactobacillus sp ecies can be d ivid ed into three grou p s: the L. casei-Pediococcus grou p , the Leuconostoc grou p and the Lactobacillus acidophilus/delbrueckii grou p (Collins et al. 1991, H am m es and Vogel 1995, Stiles and H olzap fel 1997). Since then, based again on 16S rDN A sequ ences, it w as p rop osed to d ivid e, the Lactobacillus sp ecies into five grou p s, nam ely Lb. acidophilus, Lb. salivarius, Lb.
reuteri, Lb. buchneri and Lb. plantarum (Schleifer and Lu d w ig 1995). H ow ever, these classification s have generally been consid ered as u nsatisfactory and also the u se of 16S rRN A genes as phylogenetic m arkers has been criticized (Claesson et al. 2008). N ew p rop osals for the classification of the lactobacilli sp ecies claim that the genu s cou ld be d ivid ed into seven or eight grou p s (Dellaglio and Felis 2005, H am m es and H ertel 2006). As com p lete genom e sequ ences becom e available, the high d iversity of Lactobacillus has also been su ggested to requ ire the creation of new , su bgeneric d ivisions (Canchaya et al.
2006, Claesson et al. 2008).
Lact ococcus species
In contrast to Lactobacillus sp ecies, the genu s Lactococcus com p rises at the m om ent only five sp ecies, nam ely Lactococcus garvieae, Lc. lactis (su bsp ecies cremoris, ssp . hordniae and ssp . lactis), Lc. piscium, Lc. plantarum, and Lc.
raffinolactis (http :/ / w w w .ncbi.nlm .nih.gov/ Taxonom y/ ). They can initially be d istingu ished from each other by their p ossible ability to grow at tem p eratu res above 40 °C and in >4 % sod iu m chlorid e (Batt 2000).
Plasm id s are com m on com p onents of Lc. lactis genom es, these being d iverse in size, cop y nu m ber and d istribu tion. Plasm id s often carry som e significant characteristics, su ch as carbohyd rate ferm entation, p roteolysis, p olysaccharid e p rod u ction or bacteriosine p rod u ction. Occasionally som e p lasm id s also encod e the d eterm inants necessary for conju gation, and transfer of conju gal p lasm id s m ay be a significant factor in horizontal gene transfer, likew ise other transp osable elem ents (Cou rtney 2000).
W eissella species
W eissella sp ecies have been classified as Lactobacillus or Leuconostoc sp ecies u ntil the reclassification p rop ose by Collins et al. 1993. W eissella sp ecies are heteroferm entative lactics (Batt 2000, Jay 2000), and com p rise at the m om ent of fou rteen sp ecies (http :/ / w w w .ncbi.nlm .nih.gov/ Taxonom y/ ). W eissella ssp.
have been fou nd in d ifferent ferm ented vegetable based food s, such as
ferm enting cassava (Kostinek et al. 2007) and sau erkrau t (Plengvid hya et al.
2007), in sou rd ou gh (Di Cagno et al. 2006) and in blood sau sage, “Morcilla d e Bu rgos” (Santos et al. 2005) and in “Shochu ”, a trad itional Jap anese liqu or (End o and Okad a 2005). W eissella ssp . have also been isolated in hu m an sam p les from vaginal m icrobiota (N am et al. 2007) and saliva (Kang et al. 2006).
2.2.1 Ident ificat ion of LAB
Reliable id entification of LAB is cru cial in m any tasks, e.g. ap p lications in ind u strial p rocesses (technological p rop erties), bu t also safety and qu ality control p roblem s (Tem m erm an et al. 2004). Previou sly id entification of bacterial sp ecies w as based on the p henotyp ic characterization, bu t m olecu lar biology (genotyp ic) m ethod s, lar gely DN A-based techniqu es, offer m u ch greater d iscrim inatory p ow er, all the w ay to d ifferentiation of ind ivid u al strains. Usu ally a com bination of both p henotypic and genotyp ic id entification techniqu es (p olyphasic ap p roach) is p referred (Tem m erm an et al. 2004, Aqu ilanti et al. 2007).
Phenotyp ic m ethod s inclu d e m orp hological analysis, grow th characteristics and su gar ferm entation p rofiles (e.g. w ith com m ercial API-tests, BioMerieu x, France). H ow ever, these m ethod s have a low taxonom ic resolu tion and often allow d ifferentiation only at the genu s level (Tem m erm an et al. 2004).
On the other hand , no sp ecial laboratory equ ipm ent is requ ired to p erform these tests. Cu ltivation, m icroscop y and su gar ferm entation p rofiles are im p ortant, even thou gh in m ost of the cases, these tests are insu fficient for accu rate sp ecies id entification d u e to the great nu m ber of d ifferent LAB sp ecies w ith sim ilar p henotypic characteristics.
The u se of m olecu lar tools has revolu tionized the id entification of m any bacterial sp ecies, inclu d ing LAB. Many of these techniqu es are based on the p olym erase chain reaction (PCR) u sing oligonu cleotid e p rim ers to am p lify targeted DN A fragm ents. These PCR p rim ers m a y be d esigned to d ifferent taxonom ical levels, from genu s-sp ecific (for exam p le d ifferentiation betw een W eissella and Leuconostoc sp ecies, Schillinger et al. 2008) to the sp ecies-specific level (for exam p le d ifferentiation betw een Lb. brevis, Lb. fermentum and Lb.
parabuchneri, (Coton et al. 2008)), to su b-sp ecies level (for exam p le betw een Lc.
lactis su bsp . lactis and Lc. lactis su bsp . cremoris (Pu et al. 2002)), and also further to the strain-level (for exam p le id entification of Lb. reuteri strain 35 (Cou d eyras et al. 2008)). This ap p roach is su itable w hen the bacterial sp ecies has been p relim inary characterized w ith other, u su ally phenotyp ic, m ethod s.
Another genotyp ic m ethod w hich can be u sed to id entify unknow n bacterial isolates is ribosom al DN A sequ encing, u su ally targeted to 16S or 23S ribosom al su bu nits. The obtained sequ ence is com p ared w ith DN A sequences in online d atabases, su ch as EMBL (Eu rop ean Molecu lar Biology Laboratory) or Genbank u sing a sequ ence search engine like BLAST or FASTA (http :/ / w w w .ebi.ac.u k/ em bl/ ; http :/ / blast.ncbi.nlm .nih.gov/ Blast.cgi/ , (Altschu l et al. 1990).
A w id ely u sed DN A fingerp rint techniqu e based on the PCR –m ethod is rand om ly am p lified p olym orp hic DN A (RAPD), w here short arbitrary p rim ers are u sed to rand om ly am p lify DN A fragm ents. RAPD-PCR has been u sed w ith LAB in several stu d ies and has su ccessfu lly typ ed LAB strains (Urso et al. 2006, Aqu ilanti et al. 2007). Desp ite the u sefu lness of RAPD-PCR for strain typ ing it also has been criticized . Most concerns are related to the rep rod u cibility of t he resu lts. H ow ever, w ith carefu l op tim ization, inclu d ing ensu ring the qu ality of the tem p late DN A, p olym erase enzym e and PCR cond itions, the RAPD -PCR m ethod has been fou nd reliable and rep rod u cible (Ravelo et al. 2003, Foschino et al. 2008).
Accord ing to EFSA, an ind isp ensable p re-requisite for d eterm ining antibiotic resistance w ith accu racy is the correct id entification of the strain of concern at the sp ecies level by m eans of m olecu lar taxonom y m ethod s (Eu rop ean Com m ission, 2008).
2.2.2 LAB of human and animal origins
A com p osition of LAB sp ecies d iffers betw een environm ents. Esp ecially in hu m ans, d etailed rep orts abou t LAB sp ecies and strains, as u sefu l bacteria, have received enorm ou s attention in recent d ecad es. Accord ing to Reu ter et al.
(2001), the m ajor au tochthonou s Lactobacillus sp ecies found both in infants and ad u lts are Lb. ruminis, Lb. salivarius, Lb. reuteri and Lb. gasseri (Reu ter 2001). Dal Bello and co-w orkers (2002) fou nd also food -associated LAB, nam ely Lb. sakei and Ln. mesenteroides as intestinal inhabitants, w hen alternative incubation cond ition s (30 ºC, 2 % O2) w ere u sed (Dal Bello et al. 2003).
The list of d ifferent lactobacilli sp ecies fou nd in hu m ans is inevitably m u ch longer. H ow ever, m ost of the LAB sp ecies fou nd in hu m an intestinal sam p les are transient over tim e, and colonize the intestinal tract for only a short p eriod of tim e (Walter 2005). In ad d ition, variation betw een ind ivid u als m ost p robably influ ences the sp ecies com p osition, d u e to antibiotic treatm ents, d ifferences in the d iet consu m ed or other environm ental factors (Donohu e 2004, Dicksved et al. 2007). Taking into account the recent d evelop m ent in taxonom y of LAB, the com p lexity of the sp ecies com p osite on of LAB w ill m ost p robably increase, not least w hen cu ltu re ind ep end ent m ethod s are u sed to id entify the strain . For exam p le, the PCR-DGGE (d enatu ring grad ient gel electrop horesis) system d escribed by Walter et al. (2001) p erm itted not only the d etection of Lactobacillus sp ecies consid ered to be norm al intestinal inhabitants bu t also several LAB com m only associated w ith food and often u sed as starter organism s. These non -cu ltivable LAB sp ecies inclu d ed lactobacilli (Lb. sakei and Lb. curvatus), p ed iococci (P. pentosaceus), leu conostocs (Ln. mesenteroides) and w eissellas (W . confusa). (Walter et al. 2001).
In p iglets as in hu m ans, the m ost d ominant LAB sp ecies fou nd are Lb.
ruminis, Lb. reuteri/fermentum, Lb. salivarius and Lb. acidophilus/delbrueckii grou p lactobacilli (N aito et al. 1995, Leser et al. 2002, Yin and Zheng 2005, Yu n et al.
2009). This is not su rp rising, since both hu m ans and p igs are m onogastrics. The LAB sp ecies com p osition varies in p iglets d ep end ing to som e extent on w hich m ethod s are being u sed for an alysis (cu ltu re-d ep end ent versu s cultu re-
ind ep end ent m ethod s), and betw een d ifferent gastrointestinal locations. In ad d ition, the d iet and the age of the anim als have an influ ence on the LAB sp ecies com p osition. Lb. reuteri has been observed to be a m ajor com p onent of Lactobacillus sp ecies in intestine of p igs, bu t also in chickens, cattle, d ogs, m ice and rats (Mitsu oka 1992).
The id entification of lactobacilli sp ecies com p osition in calves has not received as m u ch attention as in hu m ans or p igs w hen consid ering LAB isolated from faecal sam p les. Instead , m uch of the w ork on calves and ad u lt cow s has focu sed on w hich LAB sp ecies can be fou nd in the ru m en. The m ost frequ ently fou nd lactobacilli sp ecies from faecal sam p les in calves are Lb.
mucosae and Lb. reuteri, follow ed w ith Lb. acidophilus/delbrueckii grou p lactobacilli (Bu sconi et al. 2008). Strep tococci w ere other im p ortant LAB together in ad d ition to the lactobacilli fou nd in the stu d y of Bu sconi et al.
(2008). A novel find ing in faecal sam p les of calves has been W eissella sp ecies, also Lactococcus sp ecies have been fou nd (see pu blication II, Korhonen et al.).
Interestingly, Lc. garvieae has been associated w ith m astitis in cow s (Devriese et al. 1999).
In fish, the d om inant LAB sp ecies are carnobacteria , for exam p le Carnobacterium divergens and C. piscicola. Lactobacilli and lactococci have also been isolated from fish as w ell as several other LAB like aerococci and strep tococci. (Gatesou p e 2008) H ow ever, LAB norm ally account for only a trivial p ercentage of the intestinal m icrobiota of fish, w hich m ay p artly be d u e to the inap p rop riate cu ltivation m ethod s u sed in these stu d ies. The m ain reason for not isolating LAB from aqu atic anim als m ight be that the LAB from fish are generally slow -grow ing m icroorganism s and for this reason, a longer incu bation tim e of u p to fou r w eeks and a low er incu bation tem p eratu re (4-12 ºC) have been recom m end ed , as w ell as resorting to non-cu ltu rable m ethod s (Ringo and Gatesou p e 1998).
2.2.3 Safet y of LAB
Lactic acid bacteria have been u sed all over the w orld in variou s trad itional and ind u strial food ferm entations. Moreover, LAB have intentionally been ad d ed as p robiotics in ord er to achieve beneficial effects on health of hu m ans and anim als. Mem bers of Lactobacillus and Bifidobacteria sp ecies are generally recognized as safe (GRAS-statu s), since they have a long history of u se (Donohu e 2004).
H ow ever, the hu ge variability of LAB and also other m icroorganism s requ ire a m ore d etailed assessm ent of safety. Du e to the ap p arent need to d evelop a tool for setting p riorities associated w ith the risk of u sing these bacteria in food or feed p rod u ction, EFSA has been consid ering how best to cond u ct a form al assessm ent of safety. Thu s, a system w as p rop osed for a p re- m arket safety assessm ent of selected grou p s of m icroorganism s lead ing to granting a “Qu alified Presu m p tion of Safety (QPS)”. Therefore, EFSA p rop osed that a safety assessm ent of a d efined taxonom ic grou p , su ch as a genu s or grou p of related sp ecies cou ld be m ad e based on establishing id entity, bod y of
know led ge, p ossible p athogenicity and end u se (Eu rop ean Com m ission 2007).
Thu s EFSA has stated (2007) that tod ay a total of 33 Lactobacillus sp ecies can be consid ered to have QPS-statu s (Eu rop ean Com m ission 2007); these sp ecies are show n in Table 2.
Table 2. Lactobacillus species with QPS- status according to EFSA.
Lactobacillus acidophilus Lactobacillus amylolyticus Lactobacillus amylovorus Lactobacillus alimentarius Lactobacillus aviaries Lactobacillus brevis Lactobacillus buchneri Lactobacillus casei Lactobacillus crispatus Lactobacillus curvatus Lactobacillus delbrueckii
Lactobacillus farciminis Lactobacillus fermentum Lactobacillus gallinarum Lactobacillus gasseri Lactobacillus helveticus Lactobacillus hilgardii Lactobacillus johnsonii Lactobacillus kefiranofaciens Lactobacillus kefiri
Lactobacillus mucosae Lactobacillus panis
Lactobacillus paracasei Lactobacillus paraplantarum Lactobacillus pentosus Lactobacillus plantarum Lactobacillus pontis Lactobacillus reuteri Lactobacillus rhamnosus Lactobacillus sakei Lactobacillus salivarius Lactobacillus sanfranciscensis Lactobacillus zeae
In ad d ition to Lactobacillus sp ecies, also other LAB sp ecies have been granted QPS –statu s. They includ e three leu conostocs, (Ln. citreum, Ln. lactis and Ln.
mesenteroides), three p ed iococci (P. acidilactici, P. dextrinicus and P. pentosaceus), Lc. lactis and Streptococcus thermophilus (Eu rop ean Com m ission 2007).
One of the d ifficu lties in the QPS ap p roach is the d ifficu lty to ap ply a strain-to-strain ap p roach to u nd efined m icrobial cu ltu res, as are fou nd in m any ferm ented food s, su ch as trad itional d ried sausages and cheeses; w here the ferm entation is based on the u se of these u nd efined cu ltu res and / or backslop p ing (Rossetti et al. 2009). Since system atic investigation s have rarely been und ertaken, there is no convincing bod y of know led ge available for a given m icroorganism . N onetheless, in m any instances there exists a long and d ocu m ented history of safe u se of LAB in food s.
Som e LAB have also been associated w ith d isease, althou gh this occu rs in very rare cases, w here they can cau se op p ortu nistic in fections in p eop le w ith severe u nd erlying illnesses. LAB have been isolated from end ocard itis, bacterem ia, blood stream and local infections (Ishibashi and Yam azaki 2001, Cannon et al. 2005). In m ost cases, the infective bacteriu m has been show n to be of host origin. H ow ever, there are a few cases w here the infection has also been associated w ith the consu m p tion of p robiotics.
One of the m ost im p ortant safety asp ects of LAB is their resistance tow ard s antim icrobial d ru gs that m ight be transferred to other, p ossib ly p athogenic, bacterial species, see chap ter 2.3.3, H orizontal transferability of antibiotic resistance in the food chain.
2.3 ANTIBIOTIC RESISTANCE OF LAB
Stu d ies of antibiotic resistance of LAB have not been extensivelly investigated u ntil recently, in contrast to the situ ation w ith p athogenic sp ecies and their antibiotic resistance. H ow ever, interest in LAB and their antibiotic resistances has recently gained strength since the resistant d eterm inants are know n to be able to be transferred betw een bacterial sp ecies, also from beneficial bacteria to p athogens (see below ).
2.3.1 Phenot y pic ant ibiot ic resist ance of LAB
The first step in characterizing LAB sp ecies as being either su scep tible or resistant to antibiotics is to d eterm ine the su scep tible/ resistant p atterns w ith p henotypic m ethod s. This is easier said than d one, d u e to the variou s m ethod s available. These inclu d e factors like size of the inocu la and incu bation tim e (Egervärn et al. 2007), d ifferent test m ethod s (broth versu s agar ap p lications), inclu d ing m icrod ilu tion (Ku shiro et al. 2009), Etest (Danielsen and Wind 2003), agar d ilu tion (Florez et al. 2005) and d isc d iffu sion (Gevers et al. 2000).
Moreover, d ifferent grow th m ed ia have been u sed for actu al testing, inclu d ing Iso Sensitest, MRS, M17 and Mü ller-H inton (H u ys et al. 2002, H u m m el et al.
2007).
In the stu d y of Klare and co-w orkers (2005), a new grow th m ed iu m , LSM (LAB su scep tibility test m ed iu m ) w as d evelop ed p articu larly to su p p ort the grow th of Lactobacillus, Lactococcus, Pediococcus and Bifidobacteria bu t w hich w ou ld not have any antagonistic interactions betw een com p onents of the m ed iu m and sp ecific antim icrobial agents (Klare et al. 2005). Su bsequently, several stu d ies on phenotyp ic antibiotic resistance of LAB have been cond u cted u sing LSM broth and agar (Klare et al. 2007, Devirgiliis et al. 2008, H u ys et al.
2008, Devirgiliis et al. 2009, H aakensen et al. 2009, Ku shiro et al. 2009) assisting in the com p arison of resu lts obtained in d ifferent laboratories and on the other hand easing the d eterm ination of com p arable MICs and the safety evalu ation.
The Eu rop ean Food Safety Au thority has recently (Eu rop ean Com m ission 2008) u p d ated the m icrobiological breakp oints that categorize the LAB as either resistant or su scep tible, see Table 3. The m ajor d ifference to the p reviou s m icrobiological breakp oints given by EFSA (Eu rop ean Com m ission 2005) is that the categories have been d ivid ed m ore sp ecifically accord ing to sp ecies. This m eans that the correct taxonom ical id entification is even m ore cru cial, bu t on the other hand it d oes p rovid e tools for assessing m ore precise MIC valu es of the LAB sp ecies u nd er stu d y. When an ad equ ate num ber of strains of a p articu lar species have been stu d ied (in the ACE-ART p roject this w as set as 50 strains) and the MIC valu es d eterm ined , the m icrobiological breakp oints (or ep id em iological cu t-off valu es) can be estim ated . In som e cases, how ever, the d istribu tion of MIC valu es is not “bell-shap ed ” and the MIC d istribu tions cover m ore than op tim al five d ilu tions. This cau ses u ncertainties w ith the p rop osed cu t-off valu es, and thu s one has to ap p ly the cu t-off value only as a conservative estim ation.
Table 3. Microbiological breakpoints for selected LAB (European Commission. 2008).
Species/group AMP VAN GEN KAN STR ERY CLI Q+D TET CHL
Lb. oblig. homoferm. 1 2 16 16 16 1 1 4 4 4
Lb. helveticus 1 2 16 16 16 1 1 4 4 4
Lb. acidoph./delb. 1 2 16 16 16 1 1 4 4 4
Lb. oblig. heteroferm. 2 n.r. 16 16 64 1 1 4 8 4
Lb. reuteri 2 n.r. 8 16 64 1 1 4 16 4
Lb. fermentum 1 n.r. 16 32 64 1 1 4 8 4
Lb. facult. heteroferm. 4 n.r. 16 64 64 1 1 4 8 4
Lb. plantarum 2 n.r. 16 64 n.r. 1 1 4 32 8
Lb. rhamnosus 4 n.r. 16 64 32 1 1 4 8 4
Lb. paracasei 2 n.r. 32 64 n.r. 1 1 4 4 4
Enterococcus 4 4 32 512 128 4 4 4 8 8
Pediococcus 4 n.r. 16 64 64 1 1 4 2 4
Leuconostoc 2 n.r. 16 16 64 1 1 4 8 4
Lc. lactis 2 4 32 64 64 2 4 4 8 8
AMP: ampicillin; VAN: vancomycin; GEN: gentamicin; KAN: kanamycin; STR; streptomycin;
ERY: erythromycin; CLI: clindamycin; Q+D: Quinupristin + dalfopristin; TET: tetracycline; CHL:
chloramphenicol; n.r., not required.
The Lactobacillus sp ecies have been fou nd su scep tible to m any cell w all synthesis inhibitors, like p enicillins and am p icillin (Danielsen and Wind 2003, Cop p ola et al. 2005), in contrast to glycop ep tid es su ch as vancom ycin, m ost Lactobacillus sp ecies, exclu d ing obligate heteroferm entative sp ecies, have been fou nd to be resistant to these typ es of antibiotics. H ow ever, the resistance tow ard s vancom ycin has been d em onstrated being as intrinsic (Tynkkynen et al. 1998) and shou ld not be com p ared w ith transm issible, p lasm id -m ed iated resistance found in enterococci (Leclercq et al. 1992).
Lactobacillus sp ecies are u su ally su sceptible to chloram p henicol, erythrom ycin and clind am ycin, antibiotics that inhibit p rotein synthesis, (Cop p ola et al. 2005, Klare et al. 2007). In ad d ition, resistance against inhibitors of nucleic acid synthesis, su ch as trim ethop rim , seem s to be intrinsic, althou gh fu rther characterizations are requ ired on this top ic (Am m or et al. 2007).
Resistance to tetracycline has been observed m ore often am ong Lactobacillus sp ecies, and it has been show n to have a w id e range of MICs (Korhonen et al. 2008), also w ith a m u ltim od al d istribu tion of MICs, p robably d u e to the extensive variability of tetracycline resistance m echanism s conferring d iverse levels of su scep tibility (Roberts 2005). Esp ecially w ith tetracycline, m olecu lar m ethod s shou ld be ap p lied in ord er to revea l the natu re of resistance, i.e. is it d u e to intrinsic m echanism s, m u tation or ad d ed , m obile genes.
Resistance against am inoglycosid es, su ch as neom ycin, kana m ycin, strep tom ycin and gentam icin has been observed m ore frequ ently am ong lactobacilli (Danielsen and Wind 2003, Cop p ola et al. 2005, Zhou et al. 2005).
The m ed iu m of choice has been d em onstrated to significantly affect to the classification of LAB as being either su scep tible or resistant (H u ys et al. 2002), esp ecially w ith am inoglycosid es and MRS m ed ium .
2.3.2 Genot y pic ant ibiot ic resist ance of LAB
When a strain clearly d eviates from other strains of that p articu lar sp ecies , as d eterm ined w ith p henotyp ic m ethod s, the antibiotic resistance need s to be fu rther stu d ied w ith genotyp ic m ethod s. The potentially transferable genes in LAB have been d escribed in m u ltip le stu d ies and have been review ed in Am m or et al. 2007. Tw o of the m ost com m only observed resistance genes in LAB fou nd so far are tet(M) for tetracycline resistance and erm(B) for erythrom ycin, follow ed w ith cat genes cod ing for chloram p henicol resistance (Lin et al. 1996, Danielsen 2002, Gevers et al. 2003, Catalolu k and Gogebakan 2004).
2.3.3 Horiz ont al t ransferabilit y of ant ibiot ic resist ance from LAB in t he food chain
H orizontal transfer of genetic m aterial m ay hap p en via three d ifferent m echanism s: transd u ction, transform ation or conju gation. In transd u ction, the DN A is transferred from one bacteriu m to another via bacteriophages. The im p ortance of bacteriop hages in d issem inating antibiotic d eterm inants is, how ever, qu estionable, becau se the phages are often highly sp ecies -sp ecific. In transform ation, the DN A is released from a bacteriu m and taken u p by another.
Sim ilar to the situ ation w ith transd u ction, transform ation is not believed to be a very im p ortant m echanism of transfer of antibiotic resistance (Am m or et al.
2007). In contrast, conjugation, i.e. the d irect cell-to-cell contact, p otentially can achieve horizontal gene transfer, as it has been show n to be a m echanism of transfer of genetic inform ation w ith a broad host range (Cou rvalin 1994).
The p ossible transfer of antibiotic resistance genes betw een bacterial sp ecies have been stu d ied m ostly in harm fu l or p athogenic sp ecies , bu t also recently w ith LAB. The vast m ajority of the exp erim ents have been m ad e in vitro, u sing m ethod s su ch as filter-m ating (Gevers et al. 2003, Klare et al. 2007, Ou oba et al. 2008), althou gh these in vitro m ethod s d o not m im ic the circu m stances in natu re, and resu lts obtained cannot be com p ared w ith the resu lts achieved or expected u sing in vivo m ethod s. The transferability of antibiotic resistance genes in the gastrointestinal tract (GIT) from LAB is not straightforw ard , since the GIT is a hostile environm ent to m any allochthonou s bacteria. Moreover, stu d ies m ad e in vivo u su ally are based on “w orst-case scenario”, sim u lating very high d aily intake of food p rod u cts containing the resistant bacteria (Jacobsen et al. 2007).
One w ell-characterized resistance gene in LAB, originally d etected in Enterococcus faecalis is the broad -host range p lasm id p AMβ 1, w hich has been transferred from Streptococcus lactis (Lactococcus lactis at p resent) to Lb. reuteri and betw een lactobacilli (Tannock 1987), and also from Lc. lactis to Enterococccus and Lactobacillus sp ecies (both in vitro and in vivo in the m ou se gastrointestinal tract) (Morelli et al. 1988, Gru zza et al. 1993, Igim i et al. 1996). In the recent stu d y of Devirgiliis et al. (2009), tetracycline-resistant Lb. paracasei strains w ere
id entified in sam p les of m ilk and natu ral w hey starter cu ltu res . A transp oson Tn916 inclu d ing tet(M) w as transferred to E. faecalis strain JH 2-2 in m ating exp erim ents, bu t only at a low conju gation frequ ency (Devirgiliis et al. 2009).
Recently, exp erim ents of antibiotic resistance transferability in vivo w ere also cond u cted from Lb. plantarum to E. faecalis (Jacobsen et al. 2007). The transfer frequ encies have been observed to increase w hen the anim als have received the antibiotic in qu estion at su btherap eu tic levels (Igim i et al. 1996, Salyers and Shoem aker 1996, Licht et al. 2003) in their d rinking w ater or feed , su ggesting that increasing the antibiotic p ressu re can am p lify the transfer of antibiotic resistance betw een bacterial sp ecies.
All of these above stu d ies ind icate that antibiotic resistant factors m ay be transferred from food related bacteriu m sp ecies to other, p otentially p athogenic sp ecies, su ch as enterococci. Another interesting p oint of view is that intestinal bacteria m ight also interact w ith bacteria that are ju st passing throu gh the colon (com m ensal bacteria), allow ing these bacteria to acqu ire and transm it antibiotic resistance genes (Salyers et al. 2004).
3. Aims of t he st udy
The m ain objective of th is thesis w as to ad d ress the antim icrobial resistance p atterns of LAB. The w ork started w ith the isolation and characterization of LAB from anim al and hu m an origins, and continu ed w ith the a ccu rate sp ecies id entification follow ed by the d eterm ination of antim icrobial su scep tibility p rofiles of antim icrobial su bstances from d ifferent classes. H orizontal transferability of antibiotic resistance gene tet(S) cod ing for tetracycline w as also exam ined betw een Lactococcus and Listeria sp ecies.
The sp ecific aim s of the ind ivid u al stu d ies w ere:
To isolate, id entify and d efine the d istribu tion of the Lactobacillus sp ecies in w eaning p iglets (I);
To d eterm ine the su scep tibility p henotyp e to antibiotics in lactobacilli sp ecies isolated from p iglet faecal sam p les (I);
To isolate, id entify and d efine the d istribu tion of LAB sp ecies in you ng calves (II);
To d eterm ine the su sceptibility phenotyp e to antibiotics in LAB isolated from calf faecal sam p les (II);
To d eterm ine the MICs of Lactobacillus rhamnosus strains m ostly from hu m an and food isolates and to stu d y the p revalence of antim icrobial resistance genes w ith m icroarray analysis (III);
To su m m arize p henotyp ic d ata of antim icrobial resistances and to d evise tentative m icrobiological cu t-off valu es for fou rteen Lactobacillus sp ecies (IV);
To stu d y the transfer of p lasm id d erived antibiotic resistance to tetracycline from Lactococcus garvieae to Listeria monocytogenes (V).
