Beneficial Microbes, March 2011 ; 2(1): 79-90Wqgeningen AcodemicP u b li s h e r s

Safety and persistence of orally administered human Lactobacillus sp. strains in

healthy adults

P. Hüttl, P. Kõlll, ļ. Stsepetovar, B. Alvarez2, R. Mįindarr, K. Krogh-Andersen2, H. Marcotte2, L. Hammarström2and M. Mikelsaarl

lDepartment of MicrobioĮogy, Facutty of Medicine, ĮIniversity of Tartu, RavįĮ.a 19, 5041l Tartu, Estonįa;2Dįvįsįon of CĮinicaĮImmunoĮogy, Department of Laboratory Medicįne, KaroĮįnska Instįtutet at KaroĮįnska University HospitaĮ Hudd'inge,141 86 StockhoĮm, Sweden; pirje,hutt@ut.ee

Received: 11 Įune 20t0 l Accepted: 16 November 2010@ 2011 Wageningen Academic Pubļishers

Abstract

The aim of the study was to evaļuate the safeŲ and persistence of selected LactobacįĮĮus strains in the gastrointestinaltract (GiT) of healthy adult volunteers after oral consumption of high doses of lactobacilļį to identify potentialcandidates for probiotic and biotechnological applications. In the first phase of the study, nine individųals consumedcapsules containing Į actobaciĮĮus gasserį \77 and,ĖI6B7, LactobacįĮ,Įus acidophilus 821-3, LactobacilĮ'us paracasei3l7 and LactobacįĮ'Į,usfermentum 338-1-1 (each daily dose 1x10r0 cfu) for 5 consecutive days. Data on gut health,blood parameters, and liver and kidney function were collected. The persistence of LactobacįĮIus strains wasassessed by culturing combined with arbitrarily primed polymerase chaįn reaction (AP-PCR) and PCR-denafuringgradient gel electrophoresis (PCR-DGGE) on days 0, 5, 8, 10 and 20 from faecal samples. All strains survivedgastrointestinal passage and were detected on the Sth day. L, acidophilus 82I-3 was detected in four volunteers onthe 8th day (4.3 to 7.0logro cfu/g) and in two on the 10th day (8.3 and 3.9 logro cfu/g, respectiveŲ. In the secondphase of the study, five additįonal voļunteers consumed L. acidophiĮu.s 821-3 (daily 1x1010 cfu) for 5 consecutivedays. The strain was subsequently detected in faeces of aļļ individuals using real-time PCR on the 10th day (range4.6-6.7; median 6.0 log'o cell/g) in both phases of the study for at ļeast 5 days after discontinuation of consumption.The administration of high doses of different Lactobacįl.Į.us strains did not result in any severe adverse effects inGIT and/or abnormal values of blood indices. Thus, the strain Į. acidophilus 821-3 is a promising candidate forprobiotic and biotechnological applications. Further studies will be performed to confįrm the strain persistenceand safeŲ in a larger number of individuals.

I(eywords: probiotic, survival, colonisation, cuļtivation, real-time PCR

1. lntroduction

Lactobacilļi are microorganisms that form part of thehuman microbiota, having an important role in the first

,line of defence against opportunisüc and invasive pathogens(Lidbeck and Nord, 1993; Stecher and Hardt, 2008). SomeLactobacįĮĮus strains have been used as probiotics in foodproducts and dietary supplements for decades (Reid et aĮ.,

2008). Probiotics are defined by the FAO/WHO (2001) aslive microorganisms which, when administered in adequatenumbers, confer a heaļth benefįt on the host. The health-

promoting effects of some Lactobaciļlus strains havebeen associated with high antimicrobial activiŲ againstpathogens, antioxidaüve properties and immunostimulatorypotential (Annuk et aĮ.' 2003;Hutt et al',2006; Mikelsaarand Zilmeņ 2009, Saarela et aL,2}}O'Strahinic et 4l,,2007:.Timmerman et al., 2007 ; Verdenelli et aĮ.' 2009)'

During recent years, different genetically-engineered, e.g.second generation, probiotics ( designer probiotics') havebeen developed for the application oflactobaciļli as vehiclesfor delivery of both active and passive immuniŲ (Detmer

/SSN 7BZ6-2833prtnĻ /SSN 1876-2891 online, Dol 10.3920/BM2010'0023 79

P' Hüftetal'

and Glenting, 2006; Seegers,2002). The transformation ofLactobacįllus strains for delivery ofantibodies has thus beenproposed both for prophylaxis and treatment of selectedinfections (Ift üger et aĮ', 2002; P ant et aĮ' 2006).

HoweveĻ before application as a probiotic and/or biologicaldelivery system, the safeŲ and persistence of high dosesof the selected LactobacįlĮ.us strains in the gastrointestinaltract (GIT) ofhealthyadultvolunteers has to be assessed.Lactobaciļli have generally been regarded as safe, but thereare several theoretical concerns in the safeŲ of putativeprobiotics such as the development of infection (such as

bacteremia or endocarditis), toxic or metabolic effectson the GIĮ the transfer of antibiotic resistance to thegastrointestinaļ microbiota (Snydman, 2008), and adverseeffects mediated by immunomoduļation. In addition, theputative probiotic strains should be accurately identified(Huys et aĮ.,2006; Vankerckhoven et aĮ. 20o8). Antibioticresistance pattern, tolerance to acid, bile and pancreaticjuice, absence of haemoĶic activiŲ, and safeŲ shouļd alsobe proven in animaļ models (FAo/\)flHo, 2002;Frias et aĮ'2009; Pavan et aĮ.,2003; Vesterlund et aĮ'' 2007).

Prior to this study, f'ivė LactobacįĮĮus strains had beenisolated from the gastrointestinal tract of healthy Estonianchildren (Mikelsaar et aĮ',,2002; Sepp eü aĮ., L997) andselected based on the antibiotic resistance pattern, auto-aggregation abiliņ tolerance to acid, bile and pancreaticjuice and absence of haemoĶic activiŲ (Ī{õI| et al,,2010).According to their properties, these five strains wereconsidered as potential probiotics and candidates forbiotechnological applications such as delivery of therapeuticmolecuļes in the gastrointesünaļ tract. To further investigate

the suitabiliŅ ofthe strains for thįs purpose, the presentstudy was conducted to assess the safety, survival, andpersistence of the selected LactobacįĮ.Įzs strains in thegastrointestinal tract in healthy volunteers upon oralconsumption.

2. Materialand methods

Lactobacillus strains

Strains used for randomĮy amplified polymorphic DNA-polymerase PCR (RAPD PCR) and the colonisationstudies are listed in Table 1. For the colonisation study,five Lactobacįl'Įus strains previously isolated from thegastrointestinal tract of healthy Estonian children(MiĮ<elsaar et aĮ',2002;Sepp et aĮ,,1997) and subsequenģcharacterised (I(õļļ e, al',20t0) were used. The strainswere re-identified by sequencing the 165 RNA gene asLacto bącįļļus gasseri l77, Lactob aciĮlus acidophiĮ'us 82ļ-3,Lacto b acįĮĮ.us gasseri Ėt6B7, Lactob acįlļus paracasei 3I7and LactobaciĮĮus fermentum 338-L-t (Kõll ef aĮ.,2070)'The safeŲ of these strains has previously been confirmedusing an animaļ model (I(õll et aĮ'' 20t0).

Īable 1. Bacterial strains used in this study.

L. acidonhitus 821-3 Human intestine Kõll ef a/'. 2010

JCM 1021 Rat intestine JCM

JCM 1023 Rat intestine JCM

JGM'1028 Human intestine JCM

Source of referencez

Kõļl efal' 2010

,Īartu University

Īarļį'University

Tārtu'Uniūėrsity

Īartu UniyersiŲ

Īartū Universiņl

Įi-ģql', .

Xõll eįaį, 2010

,Tartu UniversiŲ

Ī .Uņiversity

AT.GĘ,,. ,

::,,1: ll:l::: l. :

Āiēē,,.,.',.

Ļ gassel

JCM 1229 Human intestine JCM

JOMau Vaginaltract jcķr

',lctvt sgņz Human JcMJcM 11047 Human įntestine JCM

JCM 20315 Unknown JCM

.'įāo,:., .':oaiį'āüį;;4čhirlänšen; oįnmaiķAĪCC 4356Ī Human AĪcc17'Ī Human intestine Koll ef aį, 2010

E16B7 Human intestine Köll et al.,2010

ĪU171 Human intestine Tartu Universitv

ĪU172 Human intestine Tartu Universitv

.-:,niįį]ģįsz- Vaginaiįil:,.AĪcc..'' ''

L. paracasei 311, Human intestine

ĪU194 Human intestine

ĪU199 Human intestine

L. ptanttarum TU191 Human intestine

ĪU192 Human intestine

TU193 Human intestine

ATCC 14917 Pickled cabbage

L. fermentum 338-1-1 Human intestine1.'i. l.::"':]iū1įr ' ', "',įį;r',iįt rtma

TU'154 Human intestine

beets

'Ėaįtniįe' Rl gņ', ",,,,ķur, giain māšh,

1 The strains used in the colonisation study are highlighted in bold and

the oļher strains were used for the develooment of the real time PCR.2AĪ_CG

= Ameilcan Īype Culture Collection; JCM = Japanese Collection

of Mļc1o:o1gaņ!sms;,1,::_':l ,.,,1:,., .:,' , , ,' , '. . ]'i. ]:.:::]],, .i':,i

Preparation of capsules

Lactobacilli were grown in de Man-Rogosa-Sharpe (MRS)medium (oxoįd, Basingstoke, Ui() at 37 'C under micro-aerobic conditions, lyophilised and encapsulated (1010 cfuper capsule). The viability of the encapsulated ļactobacilliwas checked by cultivation before use in the human trials.

Study population

Inclusion criterįa for the subjects were: a desire toparticipate, age20-68years, and no known health problems.Participants were asked to maintain their normal diet butrefrain from consuming probiotic products. Excļusioncriteria included a history of gastrointestinaļ disease,food allergy and acute infection, use of any antimicrobial

80 Beneficial Microbes 2(1 )

agent within the preceding month or use of any regularconcomitant medication including non-steroidal anti-inflammatory drugs and antioxidant vitamins, pregnancyand breastfeeding. Fifteen healthy volunteers (six men,nine women, mean age 36 years) volunteered to participatein the study (interventional, single-arm, open trial). Allparticipants signed a written informed consent and wereinformed that they could withdraw from the study at anytime. The Ethicaļ Committee of Tartu UniversĘ approvedthe study protocol. The trįaļ was registered in CurrentControlled Trials (ISRCTN30946841).

Study protocol

The studywas divided into two phases, the first one aimingto identify the best colonising LactobaciĮĮas strain amongfive candidates and the second phase aimed to repeat thecolonisation study with the best persisting strain. The firstphase ofthe colonisation study included a conventionalplating method and molecular methods (arbitrarily primedpolymerase chain reaction (AP-PCR), PCR-denaturinggradient gel electrophoresis (PCR-DGGE) and real-timePCR). In the second phase, the faecaļ samples were analysedby real-time PCR only.

Frsfphase

The first phase involved ten volunteers (5/5; men/women).They all received capsules containing 5 freeze-driedputative probiotic strains Z. gasseri l77 , L. gasserį Ė16B7 ,

L. acidophiĮus 82L-3, L. fermentum 338-L-t and L. paracasei317 in a daily dose of 1010 cfu each for 5 consecutive days.Faecal samples were collected at days 0, 5, 8, 10 and 20.

The volunteers completed a self-reporting questionnairecontaining questions on welfare, nutritionaļ habits, andhabitual gastrointestinal symptoms (abdominal pain,flatulence, bloating and stool frequency) during the trial(Svedlund ef a/., 1988). The impact of the consumption ofthe five selected strains on the voļunteers'health was alsoevaluated. Blood samples were coļlected at recruitmentand at the end ofthe trial (day 20) to confirm absence ofinfection/inflammation: white blood cell counts (\fBC),high-sensitive C-reactive protein (hs-CRP), and functionindices of liver and kidney (aspartate aminotransamįnase(ASAT), aļanine aminotransaminase (ALAT), albumin andserum creatinine) were measured.

Secondphase

The second phase included five new healthy adultvolunteers (ll4; manlwomen). They consumed capsulescontaining freeze-dried Į, acidophiĮus 821-3 in a daily doseof 1010 cfu for 5 consecutive days. Inflammatory indices(!fBC, hs-CRP) were tested before the start of the secondphase to exclude infection. The other blood anaĶes listed

Safety and perstsįence of Lactobacillus sp. sįralns

above were not measured in the second phase due to theabsence of abnormal values of blood indices in the firstphase using the same strain. Faecal samples were colĮectedand questionnaires were completed as described above.

Collection of faecal samples

Approximately 2 g of voided stool was collected in sterileplastic containers at the beginning on day 0 and after endof administration on the 5th, 8th, 10th, and 20th day. Thefaecal samples were initially stored 1 day maximum in a

freezer at -20'C or delivered to the laboratory immediately,where they were subsequently frozen at -80 "C. Samplesfrom both phases of the study were sent to the Departmentof Laboratory Medicine at l(arolinska Institute on dry icefor real-time PCR analysis.

Microbiological analyses of faeces

The counts oftotaļ faecaļ lactobaciļli, isolates of indigenouslactobacilļi, and the consumed LactobacįĮĮus strainsduring the trial were evaļuated in the faecal samples by aconventionaļ cuļtivation method. Serial dįlutions (10-2- 10_9)

of the weighed faecal samples were prepared with sterilesaline and 0.05 ml aliquots were seeded onto MRS andRogosa agar medium (Songisepp et al,,2005). The pĮateswere incubated' at37 'C for 3 days microaerobicalĮy in a

l0o/o Co2environment (incubator IG 150, ļouan, France).

Provisional identification of lactobacilli isolates was basedon a graļn-positive rod-shaped nonsporing cell morphologyand negative catalase reaction. Further įdentificationincluded biochemical characteristics and/or API 50 CHL(BioMėrieux, Marcy-l'Etoile, France). A total of.727 iso|ateswere proüsionally identified as lactobacilli and were furtheranalysed for fermentation Ųpe. The ability of isolates togrow in MRS broth for 24h Īn a tīo/o Co, environment at15'C and 37'C and to produce gas in MRS agar with 17o

glucose was also assessed. The fermentation of glucosewįthout gas production, growth at 37 'C and no growthat 15 "C identifies obligate homofermentative lactobacilli;growth both at 15'C and 37'C without gas production ischaracteristic of facultative heterofermentative lactobacilli,whereas gas producüon at 37 "C andvariable growth at 15 "Care characteristic of obligate heterofermentative lactobaciļli.The count of' Lactobacįl'ļzs species was expressed as logrocolony forming units per gram of faeces (logro cfu/g). Thedetection ļevel of lactobacilli was 3.0 logro cfu/g faeces.

AP-PCR typing

The putative Z actobacįĮ.Į.us isolates were Ųped by AP-PCR.Pure cuļtures were cultivated on MRS agar microaerobically24h at 37 'C in l0o/o CO- Genomic DNA was extractedwith the QIAamp DNA Mīni IĢt 50 (Qiagen GmbH, Hilden,Germany) according to the manufacturer's instructions.

Beneficial Microbes 2(1) 81

P. Hüft et al.

AP-PCR typing was performed using three primers:ERIClR (5'-ATGTAAGCTCCT GGGGATTC AC-3' ),

ERIC2 (5'-AAGTAAGTGACTGGGGTGAGCG -3') andHayford primer (5'-ACGCGCCCT-3) (DNA TechnologyA/S, Aarhus, Denmark) as described previousĮy (Alanderet aĮ'' , t999; Hayford et aļ., t999; Zhong et al', 1998). ThePCR products were resolved by electrophoresis on a 2o/o

agarose gel containing ethidium bromide (0.1 įļllrr'ļ) in tris/acetic acid/EDTA (TAE) electrophoresis buffer (Bio-RadLaboratories, Hercules, CA, USA) at constant voltage of100 V. A 1 kb ļadder (Fermentas, Vilnius, Lithuania) wasused as a marker. The banding patterns of isolates werevisualised with UV light and compared with that of theconsumed Lactob acįļlus strains.

PCR.DGGE

PCR amplification

The DNA of faecal samples was extracted using QIAampDNA Stool Mini l(it (Qiagen GmbH). The amount ofDNA was determined visually after electrophoresis ona !.2o/o agarose gel containing ethidium bromide. PCR toinvestigate the LactobacįIļus-specific GIT microbiota wasperformed using the 165 rDNA primers Univ-515r-GC(5'-CGCCGGGGGCGCGCCCCGGGCGGGGCGGGGGCACG GGGGGATCGTATTACCGCGGCTGGCA-3)and Lab-O 1 59f (5'-G GAAACAGRTG CTAATACCG-3').Nested PCR was performed with these primers onpreviously generated products from amplification with7f (AGAGTTTGATCTACTGG CTCAG) and Lab67 7 -r(5' - CACCGCTACACATG GAG-3 ) (Heilig et aĮ.' 2002).

DGGĖ analysis of PCR products

The DGGE analysis of PCR amplicons was performed usinga Dcode- System apparatus (Bio-Rad). Polyacrylamide gels(8%lwtlvol] acrylamide-bisacrylamide 137 .5:ll) in 0.5xTAE buffer with a denaturing gradient were prepared witha gradient mixer and Econopump (Bio-Rad). Gradientsfrom 30 to 600/o urea-formamide were employed for theseparation of the products amplified with primers specificfor LactobaciĮlus spp. Gels were analysed according toPearson correlation, using BioNumerics 2.5 (Applied Maths,St. Martens Latem, Belgium) software (HeĪ|iget aĮ,,2002).

Real-Īime PcR

Total DNA from faeces was extracted using the

QIAamp DNA Stool Minikit (Qiagen GmbH)with somemodifications in the protocol. Faecal samples weighing200 mg each were resuspended in 1:10 voļume of cold lxPBS and homogenised using a mixer mill (Retsh MM301;Retsh GmbH, Haan, Germany) at2}Hertz for 3 min. Thesuspension was centrifuged for 30 s at 1009 to removedebris. Supernatants were collected and centrifuged for 5

min at 1,60009. Pellets were washed once with l volumeof lx PBS and once with I volume of 10 mM Tris pH8,10 mM EDTA. Faecal samples were then resuspendedin 10 mM Tris pH8, 10 mM EDTA, 10 mg/ml Ķozyme,400 u/mĮ mutanolysin and 40 pglml RNase, and incubatedfor 15 min at 37 'C. After the incubation, the faecaļsuspensions were transferred to screw-cap tubes containing300 mg of zirconium beads (0.1 mm) and 1.4 ml of ASLbuffer (QIAamp DNA Stool Mini l(it), and were beatenin a Fast Prep at 6.5 ms-1 for 3x45 sec with cooling onice between runs. After that, samples were centrifuged at16,0009 for 1 min to pellet stool particles and DNA wasextracted from sųpernatants using the QIAamp DNA StoolMini kit following the manufacturers' instructions.

Real-time PCR protocols for detection and quantification ofLactobacilli

The RAPD technique using primer 1254 (5' -CCGCAGCCAA-3') (Torriani et aĮ,, t999) was initially applied to detectstrain-specific sequences of Į, acidophiĮus821-3. The RAPDband patterns of 11 different L. acidophiĮus strains andother t7 LactobacįĮ.Ius strains (TabĮe 1) were compared anda specific band of 250 bp was obtained for L' acidophiĮus821-3. The sequence ofthe 250 bp RAPD fragment showedhomologywith an ORF that encodes a phage major capsideprotein of L. gasseri ATCC 33323 (GenBank accessionnumber YP_815279) andL. reuterį NĪCC 55730 (GenBanļ<

accession number 4B043800). In order to obtain a longersequence to design strain-specific primers for the real-time PCR, inverse PCR was performed to determine thesequence ofthe surrounding region ofthis 250 bp fragment.As a result, the sequence of a 3.3 kb region containing theORFs that encoded putative phage proteins was obtained(GenBank accessįon number HM598408).

Strain-specific primers and a Taqman-MGB (Taqman-minor groove binding) probe (I(utyavin et aĮ'' 200o)were designed within the 3.3 kb region using the PrimerExpress 2.0 software (Applied Biosystems, FosterCity, CA, USA). The pair of primers q821-3D-Fw(5'-CTTCACGGGCAGGAATACTACAC-3) I q82t -3D -Rv (5,-GATAACGAATAATAAAGGAAAGCTGCTT-3)was selected based on its specificity. The specificity ofthese primers were tested in PCR and real-time PCRusing as negative controls, DNA from 28 lactobacilļistrains (Table 1) and four DNA samples extracted fromthe faeces of 4 volunteers at day 0 of the first phase ofthe study. The designed Taqman-MGB probe q821-3-P(5'-CGTTACGGTCAAACTA-3') was located in the ORFthat encoded a protein that showed homology with a

putative phage head-tail adaptor of L. gasseri NĪCC33323(GenBank accession number YP-815280).

The primers and the Taqman-MGB probes used todetermine total counts of lactobacilļi and Į. acidophiĮus

I

I

82 Beneficial Microbes 2(1 )

were described previously (Haarman and l(nol, 2006). Forquantification of total lactobacilli the primers F_alļlactjS(5'-TGGATGCCTTGGCACTAGGA-3')

",rO O_u111"c*-IS

(5'-AAATCTCCG GATCAAAGCTTACTTAT-3') and theprobe P_alllactJs (5'-TATTAGTTCCGTCCTTCATC-3')were ųsed. For quantification of totaļ Į. acidophiĮus,the primers F_acid_IS (5'-GAAAGAGCCCAAACCAAGTGATT-3') and R_acid_IS (5'-CTTCCCAGATAATTCAACTATCGCTTA-3') and the probe P_acid_IS (5'-TACCACTTTG CAGTCCTACA-3') were used.

All the probes were labelļed with the 5' reporter dye6-carboxy-fluorescein (FAM) and the 3' quencher NFQ-MGB (Applied Biosystems). The PCR reactions wereperformed in a total volume of 20 ķil containing 0.5 U ofAmpliTaq Gold' DNA Polymerase (AppĮied Biosystems),0.2 U of AmpErase UNG, 2 gI of 10x Taqman Buffer A,200 pM of each dNTB 2 į of purified DNA and optimisedconcentrations of primers, probe, and MgClr. Theamplification and detection program contained: one cycleof 50 'C for 2 mĪn, one cycle of 95 "C for 10 min, and 40cycĮes of95 "C for 15 s and 60 'C for 1 min. Standard curveswere made using ten fold dilutions of plasmid pGEM-FTDNA (Zhao et a1.,2009) containing the correspondingamplicon ranging from 106 to 10 copies. Each sample wasanalysed in duplicates. The reproducibilĘ of each assay wasdetermined calculating the coefficients of variation (CV)from duplicates of three independent runs testing DNAisoļated from faeces spiĮ<ed with a bacterial monocuļture.

For the quantification of the strain Į, acidophilus 82t-3, the optimal concentrations of primers, probe, andmagnesium were 300 nM forward primeņ 900 nM reverseprimeņ 200 nM probe, and 5 mM MgCl'. Under theseconditions, the detection limit of the assaywas 20 copies perreaction, corresponding to 5x104 cell/g faeces. Regardingthe reproducibility, the CV was 0,53o/o, ln the case of theassays for the quantification oftotal lactobacilli and totaļ Į.acidophiĮus the optimal conditions were 900 nM forwardprimeņ 900 nM reverse primeņ 200 nM probe, and 5 mMMgClr. The detection ļimit and CV were 10 copies perreaction (5x103 ceĮ/g faeces) and 0'99% for totaļ lactobacilliand 20 copies per reaction (5x 104 cell/g faeces) and 0.91%

for total Į, acidophiĮus.

To validate the reaĻtime PCR method for detection of Į'acidophiĮus 821-3, the faecal samples from the first phaseof colonisation study were analysed and the collected datawas compared to the resuļts obtained with other molecularmethods.

Statįstical analysis

Statisticaļ analysis was performed using R 2.10.1 (ALanguage and Environment, http://www.r-project.org).A statisticaļ evaļuation ofthe significance ofthe dįfferences

Safety and persisįence of Lactobacįllus sp. sįrams

in the numbers of lactobaciĮi obtained at different samplingtimes was performed by using the'Wilcoxon signed ranktest. Aļļ measurements of cļinicaļ data were given as meansand standard deviations. For evaluation of the changesin clinical parameters at different time points, the pairedt-test or'Wilcoxon signed rank test were applied accordingto the data distribution. The Fisher exact test was used todetermine the between-group differences in gastrointestinalsymptoms. Differences were considered statisticallysignificant if the P-value was <0.05.

3. Results

First phase of colonisation study: identification of thebest coloniser

In the first phase ofthe study, ten volunteers consumedcapsules containing a mixture of five Į actobacįĮl.us strainsin order to select the best colonising strain. One person wasexcluded at the start of the study due to signs of infectionin the blood analyses.

AII consumed strains survived the gastrointestinal passageand were detected by conventional cultivation of faecalsamples and confirmed by molecuļar methods (AP-PCR andPCR-DGGE) on the 5th day of consumption. Nonetheless,there were differences in persistence among the selectedstrains of human origin. orrly two (L. acidophiĮu.s 821-3 andL. paracasei 317) strains persisted for up to five days afterdiscontinuing the administration, albeit with diminishingcounts. The different detection methods showed resultsmostly concordant in prevalence of' L. acidophįĮus 82L-3and L. paracasei 3t7 (Figure 1A and 1B). The best temporalcolonįsers werc L. acidophiļus 82I-3 and L. paracaseį 3I7that were found in 7 out of 9 subjects on the 5th day and 4and 3 (respectively) out of9 voļunteers even on the 8th day(Figure 1 and 2). The counts ofthe two strains detected onthe 10th day were quite variable. The strain Į. acidophiĮus821-3 was detected in two volunteers (8.3 logro cfu cell/gand 3.9 logro cfu cell/g respectively) and L, paracasei317 was detected in one volunteer (4.3 logro cfu cell/g)on that day (Figure 2D). The volunteers colonised by Į,acidophiĮus 82!-3 andL. paracaseį3!7 on the 10ü day werealso colonised by these two strains on the 5th and 8th day.

At baseline, the total number of faecal i.raig".rou,ļactobacilĮi varied between 4.3 and 6.3 logro cfu/g offaeces.The counts of indigenous lactobacilli were significantlyincreased on the last day of administration (day 5, median6.6 logro cfulg) in comparison to baseline values (day0) (median 5.0 logro cf:ulg). The total count returned tobaseline vaļues 3 days after stopping the administrationof lactobaciļli (day 8) (median 5.3 logro cfu/g).

We observed an increase in the intensity of some bands as

well as new bands in the DGGE profiles in faecal samples

Beneficial Microbes 2(1 ) 83

P' Hüft et al.

Day 0

Figure 1, Prevalence of recovered Lactobacillusstrains at five sampling times using (A)conventional plating method with typingAP-PCR and (B)by PCR-DGGE. The prevalence is defined as the percent of individuals in whom the strain was detected in faeces.None of the strains was detected on day 0 and 20.

811..:100

t'-,'..:,...,..'.-..,-ņ

.; -7O

āš '-'(į) vvo-^ācUE 40,

#so'.t.'::'t.20'

.:.'1'i.'.,10l:] ],:j::l:Ī:

4,,,,,.i00,

,.:'j90,:'...r':,89'Q70Ļ^^gį oU'

õ ,,'"ē''..,40.,#.,,.3,.;;';.,',,.';.2į,.,:.1::1,,1O

::::::r,:::':::::0:

o'i-t,

8

t

9.

c..]:

Ä',

3

Day 10 Day20

Figure 2. counts of recovered lactobacilli using a conventional plating method with typing AP-PCR. (A) Before the adminįstration(day 0), (B) on the last day of administration (day 5), (C) 3 days after consumption (day 8) and (D) 5 days after the last consumption(day 10) (range, median) in the first phase of study. Limit of detection, 3.0 log,o cfu/g faeces (dashed line). Each dot represents

the log,o cfu/g faeces of lactobacilli for each individual. The median is indicated by a dash. None of the strains was detected on

day 20 (not shown).

84 Beneficial Microbes 2(1 )

coļlected on the 5th day of administration of lactobacilliindicating an increased complexiŲ of the LactobacįĮ.ĮascommuniŲ after application of the mixture of strains (data

not shown).

All haematological and functional indices of liver and Į<idney

remaįned normal in all participants who completed the trial(Table 2). According to the semį-assisted questionnaire, thestudy subjects tolerated the consumption of lactobacilļiwell though some individual differences were noted. Mildabdominal symptoms (abdominal pain, flatulence, orabdominal bloating) were reported by three persons at thebeginning of the study. Two participants had complaints ofabdominal pain accompanied with flatulence and bloatingon day 3; a single participant marked all three ļistedabdominal symptoms on day 6 (one day after tļte cessationof consumption). These mild symptoms seemed not to berelated to the temporary colonisation of the volunteers bythese strains. There was no dįfference in the presence of theLactobacįĮl.us strains on day 5 between indiūduals with andwithout abdominal complaints. Furthermore, although thethree participants were still colonised with Į. acidophiĮus821-3 on day 8,4 other participants colonised on the sameday did not experience any symptoms (Table 3).

Given that L. acidophiĮu.s 821-3 was the best temporarycoļoniser in the first phase of the study, a real-time protocolfor detection of this strain in faeces was developed. In termsof prevalence, the real_tįme-PCR showed a good correĮationwith conventįonal cultivation combined with AP-PCR orPCR-DGGE (Table 3). HoweveĻ real-time PCR was moresensitive and detected slightly more positive individualsthan a combination of conventįonaĮ methods and AP-PCR(Table 3). Thus, the absence of detection of L. acid'ophiĮus82t-3 and L. paracasei 317 in two individuals out of nineon day 5 using cultivation (Figure 2B) is probably due to the

Safety and persistence of Lactobacillus sp. sfra,hs

lower sensitiviŲ of the method. Furthermore, the counts ofL' acidophilus 821-3 isolates measured with reaĻtime PCRwere significantly hģer (up to 2 logro cfu higher) comparedto the results obtained by conventional cultivation on thelast day of consumption (day 5) (median 8.0 logro cell/g vs,5.8 logro cfu/g, P=0.004) and on the Sth day (5.4logro cell/gvs. <3.0 logro cfu/g P=0.022, respectively). The counts ofL' acidophiĮus 82t-3 were decreased at the 10th day asdetected by both methods (Table 3).

Although real-time PCR method seemed to be highlyspecific, an amplicon was detected in one of the subjectson day 0 (before consuming the capsules containing thestrain) and on day 20 of the same volunteer (Table 3). Thesequence of this amplicon presented 100% homoļogy withthe target sequence which is localised within a putative openreading frame that is homologous to ORFs found in otherļactobacilļi, so it wouĮd be possible to find this sequence inother intestinal strains. Nevertheless, these'false' positiveresults did not critically affect the data; therefore, thedeveloped method could be used to detect this strain infaeces in the second phase of the colonisation study.

Second phase of colonisation study: temporalcolon isati on wįth Įacfobac ilius acido phil u s 821 -Į

In the second phase of the study, the best temporarycolonising strain, L. acidophįįus 82I-3, was consumed as a

single preparation. At the starting point, the five volunteersselected for the second phase did not show any symptoms/indices of inflammation. The count of total lactobacillį,L' acidophiĮu.s as a species and the strain Į. acidophiĮus821-3 were determined by real-time PCR (Figure 3). Į.acidophilus 821-3 was detected in all five subjects on day5 (median 6.8 logro celllg), on day 8 (median 6.0 logrocelllg), and on day 10 (median 6.0logro cell/g) (Figure 3A).

Table 2. Haematological and biochemical indices of subiects in the first phase of colonisation study (mean t SD).

Aļ beginn|ng ] At end

3:0į1 q 61į03 3,įp,8x109/lĮ.ļil,],. ffi::il;;i: 991*991 19,1191i

9 1t-9 11 9 1'9 l-19111 , ,

iilĻ',, ., .. įĮįį* ,' .Ī*i,;!x ' i*i:ļ

^l...,rine(pmo|),].....,,;I|oļ,,.'...'

Beneficial Microbes 2(1) 85

P. Hüft et al'

Table 3. Detection of Lactobacillus acidophilus 821-3 by three different molecular methods in the first phase colonisation study.Data are expressed in log.,o cfu/g faeces forAP.PCR and log,o cell/g for real.time PCR.

Subject Postfeeding periodc

Dāy,8 Day 20

'P!!,..i,,;,

'į į! :;,;..

::.' ,,;:,':.:

'w;t.t,,'t,:Pķ: '::.::.::.::.

.P5,..,1,'..

.P'..lĘ'iP:ļ';:,';';;,;i,.

'Pgi,,.'

,P,9ir''....P1e1lālehēė

Medįānl:: :]l

A.,

c.J4'5

7:;ļa

ta 5,5

1ļ9

0

a

4ļ9

0

It was still found in one subject 15 days after cessation ofconsumption (day 20) (5.1 logro celllg). Compared to thefirst phase, it persisted in a higher proportion of indiūdualsfive days after discontinųation of consumption (4 out of9 positive individuals in the first phase vs. 5 out of 5 inthe second phase according to real time PCR detection).Taken togetheĻ the results from both phases showed that

the L. acidophįlus 82l-3 strain was detectable in 2lĪ4persons by real-time PCR even 15 days after cessation ofconsumption. The values of total counts of L. acid,ophiĮus(Figure 3B) and lactobacilļi (data not shown) did notchange significantly during the second phase of the study.None of the individuaļs showed any symptoms/indices ofinflammation following consumption of L. acidophiĮus

.,1.,.:ė.o(),

r','1 ,1 t'5.,:::::l: '::l:O.,::.ta ::::4.--.

I

t:

Äo

86 Beneficial Microbes 2(1)

821-3, neither reported any symptoms during feeding ofL' acidophiĮus 82t-3.

4. Discussion and conclusion

During previous years, a biotechnological approach,e.g. transformation of LactobacįĮĮus strains for deliveryof therapeutics, has been proposed for prophylaxis andtreatment of seļected infections (Detmer and Glenting,2006; IĢüger et aĮ'' 2002). Before being used as deliverysystem for therapeutics or as probiotics, howeveĻ thesafety of consumption of high doses of the lactobacilļi andpersistence ofthe strains in the gastrointestinaļ tract has tobe assessed in healthyvolunteers (Snydman, 2008).

In order to seļect a safe and good colonising strain oflactobacilli, a two phase study with human volunteers wasperformed. In the first phase, the volunteers were orallygiven five strains of human origin previously been chosenamong several candidates due to their resistance to acidand biĮe in vitro, antibiotic resistance profile, and safeŲ inan animal modeļ (I(õļl et aĮ.' 2010). The results indicatethatL' acidophįĮus 82L-3 was the best strain in temporallycolonising the gastrointestirial tract when given in a mixture.The coļonising ability ofĮ. acidophilus 821-3 was confirmedin the second phase when the strain was administered alone.

Even wįth the low numbers of volunteers involved, therewas a trend towards a longer persistence of Į. acidophiĮus821-3 when fed as a singĮe strain compared to when fedas part of a mixture. These results suggest that in the firstphase, competįtion with co-administered lactobaciĮi mighthave reduced the colonisation byĮ. acidophiĮus 82t-3.Howeveņ clinical trials with larger number of indivįdualswill be necessary to confirm these results.

The L. acidophilus 82!-3 strain was shown to persist for atleast five days after discontinuation of consumption whichis good in comparison to other published studįes. In thedose-response studies of Saxelin et aĮ', (1995), the authorsshowed that the well-known probiotic LactobacįlĮ'us GG,administered orally in high doses (capsules 1.2x1010 cfu/day), could be detected only for a short period (up to threedays) after cessaüon of consumption. Similar to our resuļts,

ļacobsen et al, (1999) re-isolated the se|ected LactobacįĮ.Įus

rhamnosus strains from faeces 5 days after cessation ofthe 18-day consumption period of' a LactobacįĮ'lus stra\ns

mixture. HoweveĻ one study shows poor survival andpersistence of the particular probiotic Į ' acidophiĮusNCFB1748 after consumption oftriple-strain yoghurt for 10 days.

Namely, according to RAPD PCR resuļts the survival of theL. acidophilus was detected in 3 of the 14 subjects, whilepersistence in I ofthe 14 subjects 4 days after cessation ofthe triple-strain yoghurt (Mättö et aĮ.' 2006),

The mechanism undeĄing the difference of persistence oflactobacilli in the GIT has not been elucidated yet. It may

Safety and persisfence ofLactobacillus sp. sįralns

be influenced by the origin of ļactobacilli and particularstrain properties, either ofhuman characteristics or foodmatrices (Walteņ 2008). In our study, the latter did notinfluence the results as we used capsulated strains.

An increase of total ļactobacilli counts after consumptionof either a capsulated probiotic and/or a fermented byprobiotic goat milk yoghurt has been shown previously(I(ullisaar et aĮ'.,2003; Songisepp et al.' 2005). Recently,Dommels et aĮ. (2009) described that consumption of aļow-fat probiotic spread containing L, rhamnosus GGor L. reuterį DSM 17938, increased the totaļ counts ofļactobacilli. Similarly, in our study, the consumption ofcapsulated, putatively probiotic lactobacilļi significantlyincreased the indigenous lactobacilli counts and even thespecies diversiŲ of ļactobaciļli. HoweveĻ the increase inindigenous lactobacilli was not observed when feeding Į.acidophiĮus as a single strain suggesting that this properŲmight be dependant of the strain or the dose used.

During the current human colonisation study, the FAO/WHO recommendatįons to assess the safeŲ of consumptionof lactobacilli were followed. Several previous volunteertriaļs have applied different indices (haematologicalvalues, tests for liver and renal function) for recordingadverse effects and tolerance following consumption ofprobiotics (Vanļ<erckhoven et aĮ.,2008; Vlieger et aĮ.'2009)''We seļected several inflammatory (WBC, hs-CRP) andbiochemical įndices (ALAT, ASAT, albumin, creatįnine) toexcļude infection and detrimentaļ effects on liver and kidneyfunction as the most important clinical indices for healthstatus. AlthoughS}o/o of the volunteers in the first phase ofour study reported mild abdominal symptoms, the feedingof high doses of the mixture of selectedLactobacįĮĮus strainsdid not induce any severe adverse effects in the GIT and didnot cause any change in blood and biochemicaļ indices. Inthe second phase of the study, no symptoms were observedand it could not be not ruled out that the mild symptomsobserved in the first triaļ were caused by the co-feeding withother lactobacilli. In spite of the low number of participantsin each phase ofthe study (9 in the first phase and 5 in thesecond phase), the resuļts of the safety assessments fromeach phase, taken together with the in vįtro and animaļstudies reported before (I(õļl e, aĮ.,2ot0), suggest thatthe tested strains can be considered as potentially safe forhuman consumption. Nevertheless, further safeŲ studieswith larger amounts of human volunteers would be neededto confirm this assumption.

In previous studies, specific and sensitive quantitative PCRmethods for detecüon of probiotic bacteria from faeces have

also been applied (Ahlroos and Ļnldgnen,2o}9; Bartoschet aL,2}}5Brigidi et al'' 2000). Similar to these methods,the real-time PCR assay developed to detect the strain Į.acidophiĮus 821-3 showed a high specificiŲ, sensitiviŲ andreproducibiliŲ. 1ü7-hen real-time PCR assay was compared

Beneficial Microbes 2(1 ) 87

P. Hüft et al.

to conventionaļ cuļtivation followed by AP-PCR, the countsof the strain Į . acidophiĮus 821-3 determined by real-timePCR were significantly higher. These data are in goodaccordance with previous studies on the topic (Ahlroosand Tynkkynen,2009; Dommels et a1.,2009). Real-timePCR methods quantify bacteriaĮ ļoads inferred from thenumber of copies of a particular DNA target sequence.Thus, this method does not differentįate between dead andļive bacteria which leads to an overestimation of the numberof viable cells. Furthermore, in relation to conventionalcultivation methods, not all the colonies may originate froma single ūable ceļļ, which leads to an underestimation of thenumber of viable cells. In order to obtain a more accuratemeasurement of viabļe cells using quantitative PCR,Cenciarini-Borde et aļ. (2009) have suggested treating ofbacteria with propidinium or ethidium monoazide prior tothe DNA isolation for selective suppression of amplificationof DNA released from dead cells.

In conclusion, a combined approach using culture-based andmolecular methods is suitable to detect the gastrointestinalsurvival and persistence after discontinuation of theconsumption of the probiotįc candidate Į , acidophiĮus 82L-3. Thįs strain can be considered as a good temporal coloniserif administered in high doses either as a single strain orin a mixture with other lactobacilli. The consumption ofhigh doses of dįfferent strains of LactobacįĮlus species,including Į. acidophiĮus 821-3, did not result in any severegastrointestinaĮ adverse effects and/or abnormal vaĮues ofblood indices in human volunteers. Thus, the L' acidophiĮus821-3 strain is a promising candidate for probiotic andbiotechnological application and further studies wiļl beperformed to confirm the persistence and safeŲ in a largernumber of indiūduals.

Acknowledgments

'We tharü< I(ai Truusaļu for critical reading of the manuscript,

Imbi Smidt and }anne Üļ<sti for technical assistance. Thiswork was supported by Grant No. 05042 and 6782 fromthe Estonian Science Foundation, Estonian Target fundingNo. 0418 from the Estonian Ministry of Education and theCommission of European Union (LACTOBODY 202162).8.Āļvarez was holding a postdoctoraļ felļowship from FICYT(Fundaci6n para el Fomento en Asturias de la Investigaci6nCientifica Aplicada y Įa Tecnologia) with reference numberPOST07-24.

References

Ahlroos, T. and Tynkkynen, S., 2009. Quantitative strain-specificdetection of LactobacįįĮus rhamnoszs GG in human faecal samples

by reaĻtime PCR. JournaĮ of Applied Microbiology 106: 506-5t4.

Aļandeņ M., Satokari, R., I(orpela, R., Saxelin, M., \Įlpponen-Salmela,

T., Mattila-Sandholm, T. and Von Wright, A., L999. Persistence

of colonization of human colonic mucosa by a probiotic strain,

LactobacįĮĮus rhamnosus GG, after oral consumption. Applied and

Environmental Microbiology 65: 35L-354.

Annuk, H., Shchepetova, ļ., I(ullisaar, T., Songisepp, Ė., Zilmer, M.

and Mikelsaaņ M., 2003. Characterization of intestinal lactobacilli

as putative probiotic candidates. Įournal ofApplied Microbiology94:403-4L2.

Anonymous, 2004. Method manuaļ, Cobas Integra 4001700/800.

Version 2. Roche Diagnostics GmbH, Mannheim, Germany.

Bartosch, S., woodmansey, E.J., Patersoņ, I.C., McMurdo, M.E. and

Macfarlane, G.T.,2005. Microbiological effects of consuming a

synbiotic containing Bifidobacter įum bifidum' Bifidobacter ium

Įactis, and ohgotructose in eldeģ Persons, determined by reaĻtime

polymerase chain reaction and counting ofviable bacteria. CĮinical

Infectįous Diseases 40: 28-37.

Brigidi, P., Vitali, 8., Swennen, E., Altomare, L., Rossi, M. and

Matteuzzi, D., 2000. Specific detection of Bifidobacterįum StĪains

in a pharmaceutical probiotic product and in human feces bypolymerase chain reaction. Systematic and Applied Microbiology23:39I-399.

Cenciarini-Borde, C., Courtois, S. and La Scola, B., 2009. Nucleic acids

as viabiliŲ markers for bacteria detection using molecular tools.

Future Microbiology 4: 45-64.

Detmeņ A. and Glenting, I.,2006. Live bacterialvaccines _ a review

and identi.fication of potential hazards. Microbiaļ Celļ Factories 5: 23.

Dommels, Y.E., I(emperman, R.A., Zebregs, Y.E., Draaisma, R.B., Įol,

A., Wolvers, D.A., Vaughan, E.E. and Aļbers, R., 2009. Survival ofLactobaciĮlus reuterį DSM !7938 and LactobacįĮĮas rhamnosus GGin the human gastrointestinal tract with daily consumption of a

low_fat probiotic spread. Applied and Eņüronmental Microbiology75:6t98-6204.

Food and Agriculture Organization of the United Nations/WorldHealth Organization (FAO/WHO), 2001. Health and nutritionalproperties of probiotics in food including powder milk wįth ļive

lactic acid bacteria. Report of a joint FAO/WHO expert consultation

on evaluation ofhealth and nutritional properties ofprobioticsin food includįng powder milk with live lactic acid bacterįa.

Cördoba, Argentina. Available at: http://www.who.int/foodsafeŲ/

publications/fs_management/en/probiotics.pdf.

Food and Agriculture Organization of the United Nations / World

health Organization (FAo/WHo)' 2002. Guįdelines for the

evaluation of probiotics in food. Avaiįable at: http://www.who.

int/foodsafeŲ/fs-management/en/probiotic-gūdelines.pdf.

Frias, R., Ouwehand,4., Spillmann, T., Vankerckhoven, V., Hewicker-

Trautwein, M., Salminen, S. and Gueimonde, M., 2009. Effect ofclinical and probiotic LactobacįIlus rhamnosus strains on intestinal

permeabiliŲ and bacteriaļ translocation in healthy and colitic rats.

Food Research International 42: 636-640.

Haarman, M. and l(noĻ J.,2006. Quantitative real_time PCR analysis

of feca| LactobaciĮĮus species in infants receiving a prebiotic infant

formula. Applied and Enüronmental Microbioļo gy 72: 2359-2365'

88 Beneficial Microbes 2(1 )

Hayford, A.E., Petersen, A., Vogensen, F.I(. and Įakobsen, M.' 1999.

Use of conserved randomly amplified polymorphic DNA (RAPD)

fragments and RAPD pattern for characterization of LactobacįĮĮus

fermentum įn Ghanaian fermented maize dough. Applied and

Environmental Microbiology 65: 3213-322L.

Heil, W., I(oberstein, R. and Zawta, 8., 2008. Reference ranges foradults and children. Pre-analņical considerations. 6th ed. Roche

Diagnostics GmbH, Mannheim, Germany.

Heilig, H.G., Zoetendal, E.G., Vaughan, E.E., Marteau, P., Aldcermans,

A.D. and De Vos, W.M,, 2002. Molecuļar dtversiŲ oĪ LactobacįĮĮus

spp. and other lactic acid bacteria in the human intestine as

determined by specific amplification of 165 ribosomal DNA. Applied

and Environmental Microbiology 68 It4-I23.Hütt, P., Shchepetova, }., Loivukene, I(., I(ullisaaņ Ī and Mikelsaaņ M.,

2006. Antaģonistic actiūŲ of probiotic lactobaciĮi and bi_fidobacteria

against entero_ and uropathogens. Įournal ofApplied MicrobiologyIO0: L324-L332.

Huys, G., Vancannelt, M., D'Haene, I(., Vankerckhoven, V, Goossens,

H. and Swings, J., 2006. Accuracy of species identĘ of commercial

bacterial cuļtures intended for probiotic or nutriüonal use. Research

in Microbiology 157: 803-810.

]acobsen, C.N., Rosenfeldt, Nielsen, V, Hayford, A.E., Molleņ P.L.,

Michaeļsen, I(.F., Paerregaard, A., Sandstrom, B., Tvede, M. and

ļakobsen, M., 1999. Screening of probiotic actMties of forŲ-seven

struns of LactobaciĮĮus spp.by in vifro techniques and evaluation ofthe colonization abilĘ offive selected straįns in humans. ApPliedand Environmental Microbiolo gy 65 4949-4956.

I(õll, P., Mändaņ R., Smidt, I., Hütt, P., Truusalu, I(., Mikelsaaņ R._H.,

Shchepetova, ļ., I(rogh_Andersen, I(., Marcotte, H., Hammarström,

L. and Mikelsaaņ M., 2010. Screening and evaluation of human

intestinal lactobacilļi for the development of novel gastrointestinal

probiotics. Current Microbiology 61: 560-656.

I(rügeņ C., Hu Y., Pan, Q., Marcotte, H., Hultberg, A., Delwaņ D., Van

Dalen, P.Į., Pouwels, P.H., Leeņ R.J., Ī(eĮy, C.G., Van Dollenweerd,

C., Ma, J.I(. and Hammarstrom,L.,20o2. Įn slfz delivery of passive

immuniŲ by lactobacilli producing single_chain antibodies. Nature

Biotechnology 2 0: 7 O2-7 06.

I(ullisaar, T., Songisepp, E., Mikelsaaņ M., Zilmer, Ī(., Mhalemm, T.

and Zilmeņ M., 2003. Antioxidative probiotic fermented goats'

milk decreases oxidative stress_mediated atherogeniciŲ in human

subjects. British ļournaĮ of Nutrition 90: M9-456.

I(uŲavin, I.V., Afonina, I.A., Mills, A., Gorn, W., Lukhtanoų E.A.,

Belousov, E.S., Singeņ M.ļ., Walburger, D.I(., Lokhoų S.G., GalĮ,

A.A., Dempcy, R., Reed, M.W., MeyeĻ R.B. and Hedgpeth, ļ.,2000.3'-minor groove binder_DNA probes increase sequence specificiŲat PCR extension temperatures. NucĮeic Acids Rese arch29: 655-66I.

Lewis S,, 2001. Reference ranges and normal values. In: Lewis, S.,

Bain, B. and Bates, I. (eds.) Dacie and Lewis practical haematology.

Churchill LMngstone, London, UI(.

Lidbeck, A. and Nord, C.E., 1993. Lactobacilļi and the normal human

anaerobic microflora. Clinical Infectious Diseases l6 Suppl.4:s181-187.

Safety and perslsįence ofLactobacillus sp. sįrains

Mättö, J., Fondėn, R., Tolvanen, T., Von Wright, A., Vilpponen-Salmela, T., Satokari, R. and Saarela, M.,2006.Intestinal survival

and persistence of probiotic Lactobacįllus and Bifidobacteriumstrains administered in triple-strain yoghurt. International Dairy

Journal 16: 1174-1180.

Mikelsaaņ M., Annuk, H., Shchepetova, Į., Mändar, R., Sepp, E. and

Björksten, B.'2002. Intestinal lactobacilli ofEstonian and Swedish

children. Microbial Ecology in Health and Disease 14:75-80.

MikeĮsaaņ M. and Zilmeņ M.' 2009. LactobaciĮĮus fermentum MĖ-3

- an antimicrobial and antioxidative probiotic. Microbial Ecology

in Heaįth and Disease 21: 1-27.

Pant, N', Huļtberg, A., Zhao, Y., Svensson, L., Pan-Hammarstrom, Q.,

Johansen, I(., Pouwels, P.H., Ruggeri, F.M., Hermans, P., Frenken, L.,

Boren, T., Marcotte, H. and Hammarstrom, L., 2006. Lactobacilli

expressing variable domain of lļama heavy_chain antibody fragments

(lactobodies) confer Protection agaįnst rotaürus_induced diarrhea.

Įournal of Infectious Diseases 194: 1580-1588.

Pavan, S., Desreumaux, P. and Mercenier, A.,2OO3. Use of mouse

models to evaluate the persistence, safeĘ and immune modulation

capaciües oflactįc acid bacteria. Clinical and Diagnostic Laboratory

Immunology l0 696 -7 0L.

Reid, G., Anukam, Ī(. and l(oyama, T., 2008. Probiotic products

in Canada with clinical evidence: what can gastroenterologists

recommend? Canadian Journal of Gastroenterclogy 22: 169-175.

Saarela, M., Mogensen, G., Fonden, R., Matto, ļ. and Mattila-Sandholrn,

T., 2000. Probiotic bacteria: safeŲ, functional and technological

properties. ļournal of Biotechnology 84: 197 -2L5.

Saxelin, M., Pessi, T. and Salminen, S., 1995. Fecal recovery following

oral administration of LactobacįįĮus strain GG (ATcc 53t03) ingelatine capsules to healthy volunteers. International Įournal ofFood Microbiology 25: 199-203.

Seegers, Į.F., 2002. Lactobaciļli as live vaccine delivery vectors: progress

and prospects. Trends in Biotechnology 20: 508-515.

Sepp, E., ļulge, I(., Vasar, M., Naabeņ P., BjorĮ<sten, B. and Mikelsaaņ

M.' |997 ' Īntestinal microflora of Estonian and Swedish infants.

Acta Paediatrica 86 956-961.

Snydman, D.R., 2008. The safeŲ ofprobiotįcs. Clinical Infectious

Diseases 46 Suppl. 2: S104-111; discussion S144-151.

Songisepp, E., I(als, J., I(ullisaaņ T., Mandaņ R., Hutt, P., Zilmeņ M.

and Mikelsaar, M., 2005. Evaluation of the functional efficacy of an

anüoxidative probiotic in healthyvoįunteers, Nutrition )ournal 4: 22.

Stecheņ B. and Hafdt, !f.D.,2008. The roĮe of microbiota in infectious

disease. Trends in Microbiology 16: L07-114.

Strahinic, I., Busarcevic, M., Pavlica, D., Milasin, ļ., Golic, N, and

Topisirovic, L., 2007. Molecular and biochemical characterizations

ofhuman oraį lactobacilli as Putative probiotic candidates. OralMicrobiology and Immunology 22: lLt-LL7.

Svedlund, J., Sjodin, I. and Dotevalļ, G., 1988. GSRS - a clinical rating

scale for gastrointestinal symptoms in patients with irritable bowel

syndrome and peptic ulcer disease. Digestive Diseases and Sciences

33: I29-I34.

Timmerman, H.M., Niers, L.E., Ridwan, B.U., I(oning, C.)., Mulder, L.,

Akkermans, L.M., Rombouts, F.M. and Rijkers, G.T. ,2007.Designofa multispecies probiotic mixture to prevent infectious complications

in critically ill patients. Clinical Nutrition 26:450-459.

Beneficial Microbes 2(1 )po

P. Hütt etal'

Torriani, S., Zapparoli, G. and Dellaglio '

Ė. L999. Use of PCR_based

methods for rapid differentiation of LactobacįlĮus deĮbrueckįį

subsp. buĮgaricus and L. deĮbrueckįį subsp. Įactis. Applied and

Environmental Microbiology 65: 435L-4356.

Vankerckhoven, V., Huys, G., Vancanneyt, CV., Vael, C., I(lare, I.,

Romond, M.-B., Entenza, J.M., Moreillon, P., Wind, R.D., I(nol, J.,

Wiertz, E., Pot, B., Vaughan, E.E., I(ahlmeteņ G' and Goossens,

H., 2008. BiosafeŲ assessment of probiotics used for humanconsumption: recommendations from the EU-PROSAFE project.

Trends in Food Scįence & Technology 19: Io2-LĄ4.

Verdenelļi, M.C., Ghelfi, F., Silvi, S., Orpianesi, C., Cecchini, C. and

Cresci, A., 2009. Probiotic properties of Lactobacįllus rhamnosus

and LactobacįĮįus paracasei isolrated from human faeces. European

ļournal of Nutrition 48: 355-363.

Vesterlund, S., Vankerckhoven, V., Saxelin, M., Goossens, H., Salminen,

S. and ouwehand, A.C.' 2007 . Safety assessmenį of LactobacįĮĮus

strains: presence of putative risk factors in faecal, blood and

probiotic isolates. Internationaļ ļournal of Food Microbiology116:325-331.

Vlieger, A.M., Robroch, A., Van Buuren, S., I(iers, ļ., Rijkers, G.,

Benninga, M.A. and Te Biesebeke, R., 2009. Tolerance and safeŲ

of LactobacįĮĮus paracasei SSp. paracaseį in combination withBifidobacterium animaĮis ssp . Įactis in a prebiotic_containing infant

formula: a randomised controlled trial. Briüsh Īournal of Nutrition102:869-875.

W'alteņ ļ., 2008. Ecological role of lactobacilli įn the gastrointestinal

tract: implications for fundamentaļ and biomedicaļ research. Applied

and Environmental Microbiology 7 4: 4985-4996.

Zhao, Y., Liu, Z., YLĮ, S., wen, S., Hammarström, L. and Rabbani,

H.,2009. Construction of a high efficiency PCR products cloning

T Vector using pGEM-5zf (+). Avicenna Įournal of MedicalBiotechnology L: 37 -39.

Zhong, W., Millsap, I(., Bialkowska-Hobrzanska, H. and Reid, G.,

1998. Differentiation of LactobaciĮĮus speciesby molecular Ųping.Applied and Environmentaį Microbiology 64: 2418-2423.

90 Beneficial Microbes 2(1 )