DISSERTATIONES MEDICINAE UNIVERSITATIS TARTUENSIS

110
DISSERTATIONES MEDICINAE UNIVERSITATIS TARTUENSIS
110

EVALUATION OF TECHNOLOGICAL
AND FUNCTIONAL PROPERTIES
OF THE NEW PROBIOTIC
LACTOBACILLUS FERMENTUM ME-3

EPP SONGISEPP

TARTU UNIVERSITY

PRESS
Department of Microbiology, University of Tartu, Estonia

Dissertation is accepted for the commencement of the degree of Doctor of

Medical Sciences on May 18, 2005 by the Council of the Faculty of Medicine,
University of Tartu, Estonia

Opponent: Professor Seppo Salminen PhD, Department of Biochemistry and

Food Chemistry, University of Turku, Finland

Commencement: June 20, 2005

Publication of this dissertation is granted by University of Tartu

ISSN 1024–395X
ISBN 9949–11–084–X (trükis)
ISBN 9949–11–085–8 (PDF)

Autoriõigus Epp Songisepp, 2005

Tartu Ülikooli Kirjastus

www.tyk.ee
Tellimus nr 220
To my mother
 CONTENTS

LIST OF ORIGINAL PUBLICATIONS........................................................ 10

ABBREVIATIONS ........................................................................................ 11

INTRODUCTION .......................................................................................... 13

LITERATURE REVIEW ............................................................................... 14

1. Functional foods and probiotics ........................................................... 14
2. Lactic acid bacteria............................................................................... 16
2.1. Taxonomy of Lactobacillus spp. .................................................. 16
2.2. Metabolism ................................................................................... 17
3. Prebiotics.............................................................................................. 22
4. Synbiotics ............................................................................................. 23
5. Strategy of selection of probiotic strains .............................................. 23
5.1. Strain origin, identification and typing ......................................... 24
5.2. Safety assessment ......................................................................... 26
5.3. Functional features ....................................................................... 27
5.4. Clinical testing.............................................................................. 29
5.4.1. Oxidative stress markers of the human body...................... 31
5.5. Technological aspects of probiotics .............................................. 32
5.5.1. Product manufacturing........................................................ 32
5.5.2. Viability of probiotics and interaction with starter
cultures ............................................................................... 33
6. Origin and history of Lactobacillus fermentum ME-3 ......................... 34
7. Unsolved problems............................................................................... 34

AIMS OF THE STUDY ................................................................................. 35

MATERIALS AND METHODS ................................................................... 36

8. Origin of bacterial strains ..................................................................... 37
8.1. Lactobacillus strains ..................................................................... 37
8.2. Identification of Lactobacillus fermentum ME-3 ......................... 37
8.3. Basic characteriation of Lactobacillus fermentum ME-3 ............. 39
8.4. Pathogenic target bacteria............................................................. 39
9. Properties of L. fermentum ME-3......................................................... 40
9.1. Resistance to low pH, bile and heat in vitro ................................. 40
9.2. Growth in cell suspensions ........................................................... 41
9.3. Antagonistic activity...................................................................... 41
10. Growth and survival in different products............................................ 43
10.1. Preparation and survival in probiotic cheese.............................. 43

7
 10.2. Preparation and survival in fermented goat milk ....................... 44
10.3. Preparation and survival in HELLUS fermented milk products 44
11. Total antioxidative activity of ME-3 .................................................... 45
11.1. In vitro testing of antioxidative activity ..................................... 45
11.2. Measurement of selected oxidative stress markers
in humans ................................................................................... 46
12. Detection of health effects of ME-3 ..................................................... 46
12.1. Design of human volunteer trials ............................................... 46
12.1.1. Safety study with probiotic capsule. .............................. 46
12.1.2. Functional efficacy trials................................................ 47
12.2. Microbiological analyses of feces ............................................... 48
12.3 Identification of lactobacilli ......................................................... 48
13. Statistical Analysis ............................................................................... 49

RESULTS AND DISCUSSION..................................................................... 50

14. Basic characterization of L. fermentum ME-3...................................... 50
15. Technological properties of L. fermentum ME-3 ................................. 52
15.1. Acid tolerance ............................................................................ 52
15.2. Dependence of heat resistance on pH.......................................... 52
15.3. Growth in cell suspensions.......................................................... 54
15.5. Antagonistic activity.................................................................... 55
15.5.1. Antagonistic activity between L. fermentum
ME-3 and starter cultures .............................................. 55
15.5.2. Antagonistic activity between L. fermentum
ME-3 and non-starter lactobacilli.................................. 56
15.6. Survival in milk products ............................................................ 57
15.6.1. Probiotic cheese ............................................................. 57
15.6.2. Fermented goat milk ...................................................... 57
15.6.3. Fermented milk products................................................ 57
15.7. Survival and growth of L. fermentum ME-3
in different juices ....................................................................... 58
16. Functional properties of L. fermentum ME-3 ....................................... 58
16.1. Stability in GI conditions ............................................................ 58
16.2. Stability of probiotic properties of L. fermentum
ME-3 in products ....................................................................... 60
16.2.1. Probiotic cheese ............................................................. 60
16.2.2. Fermented milk products................................................ 61
16.2. 3. Capsule and juices......................................................... 63
17. Health effects of ME-3: human volunteer trials ................................... 64
17.1. Safety trial with probiotic capsule.............................................. 64
17.2. Functional efficacy trials with fermented goat milk and
capsules ...................................................................................... 65

8
GENERAL DISCUSSION ............................................................................. 69
18. Technological properties of L. fermentum ME-3 ................................. 69
18.1. Suitability of ME-3 for various delivery vehicles...................... 69
18.1.1. Impact on s the sensory properties of the
product........................................................................... 69
18.1.2. Survival in products ....................................................... 70
18.1.3. Stress responses.............................................................. 71
19. Stability of the functional properties of ME-3 in products................... 73
19.1. Antioxidative activity .................................................................. 73
19.2. Antagonistic activity.................................................................... 74
20. Health effects of ME-3 ......................................................................... 75

CONCLUSIONS ............................................................................................ 80

REFERENCES ............................................................................................... 82

SUMMARY IN ESTONIAN ......................................................................... 91

ACKNOWLEDGMENTS .............................................................................. 95

PUBLICATIONS ........................................................................................... 97

CURRICULUM VITAE................................................................................. 171

9
 LIST OF ORIGINAL PUBLICATIONS
This thesis is based on the following original publications. Additional data are
also presented.

I Mikelsaar, M., Zilmer, M., Kullisaar, T., Annuk, H. and Songisepp, E.

Strain of microorganism Lactobacillus fermentum ME-3 as novel anti-
microbial and antioxidative probiotic. International Patent application 2001,
WO03002131 (http://ep.espacenet.com).

II Annuk, H., Shchepetova, J., Kullisaar, T., Songisepp, E., Zilmer, M.,
Mikelsaar, M. Characterization of intestinal lactobacilli as putative
probiotic candidates. Journal Applied Microbiology 94, 403–412 (2003).

III Songisepp, E., Kullisaar, T., Hütt, P., Elias, P., Brilene, T., Zilmer, M.,
Mikelsaar, M. A New Probiotic Cheese with Antioxidative and Anti-
microbial Activity. Journal of Dairy Science 87, 2017–2023 (2004).

IV Kullisaar, T., Songisepp, E., Mikelsaar, M., Zilmer, K., Vihalemm, T., Zil-
mer, M. Antioxidative probiotic fermented goats milk decreases oxidative
stress-mediated atherogenicity in human subjects. British Journal of
Nutrition 90, 449–456 (2003).

V Songisepp, E., Kals, J., Kullisaar, T., Hütt, P., Mändar, R., Zilmer, M.,
Mikelsaar, M. Evaluation of the functional efficacy of a probiotic in healthy
volunteers. Nutrition Journal, submitted.

10
 ABBREVIATIONS

API 50CHL Analytical Profile Index of 50 Carbohydrates by Lactobacillus

(isolates)
AP-PCR Arbitrarily Primed Polymerase Chain Reaction
ATCC American Type Culture Collection
CFU Colony Forming Unit
DBRP Double blind randomized placebo controlled study
DNA Deoxyribonucleic Acid
DSM Deutsche Sammlung von Mikroorganismen und Zellkulturen
FAO Food and Agriculture Organization of the United Nations
FF Functional Foods
FHEL Facultatively Heterofermentative Lactobacilli
FOS Fructooligosaccharides
GI Gastrointestinal
GRAS Generally Recognized As Safe
GSH Reduced glutathione
GSSG Oxidized glutathione
IMF Intestinal Microflora
ITS-PCR Internal-Transcribed Spacer Polymerase Chain Reaction
LAB Lactic Acid Bacteria
LA-test Linolenic Acid Test
LDL Low Density Lipoprotein
MIC Minimal Inhibitory Concentration
Mn-SOD Mn-Superoxide Dismutase
MRS de Man-Rogosa-Sharpe
NADH Reduced Nicotinamide-adeninedinucleotide
OHEL Obligately Heterofermentative Lactobacilli
OHOL Obligately Homofermentative Lactobacilli
Ox LDL Oxidized Low Density Lipoprotein
PFGE Pulsed-Field Gel Electrophoresis
PBS Phosphate-Buffered Saline
PUFA Polyunsaturated Fatty Acid
RAPD Randomly Amplified Polymorphic DNA
rRNA Ribosomal Ribonucleic Acid
ROS Reactive Oxygen Species
SDS-PAGE Sodium Dodecyl Sulphate-Polyacrylamide Gel-electrophoresis
SOD Superoxide Dismutase

11
TAA Total Antioxidative Activity
TAS Total Antioxidative Status
TGSH total glutathione
WHO World Health Organization
WIPO World Intellectual Property Organization

12
 INTRODUCTION
During the past decades lifestyle in the developed industrial countries has
changed, regarding living standard, hygiene, diet, usage of antibiotics and other
antimicrobial substances. Prevalence of chronic diseases like different allergies
and gut-associated diseases (e.g. ulcerative colitis, Crohn disease, inflammatory
bowel disease) are of rising importance in the industrial world today. The
balance of gut microbiota is considered to grant colonization resistance against
infectious agents and promote antiallergenic processes, stimulate immune
defense and reduce hypersensitivity reactions, incl. food allergy (Isolauri et al,
2001). Probiotics, live beneficial microbes, are aimed to improve the imbalance
of the indigenous microbiota. In Europe, public interest in probiotics started to
increase at the end of the last century and is still growing. About 65% of the
European functional food market is covered nowadays with dairy probiotic
products (Stanton et al., 2001).
In Estonia, investigation of lactobacilli in the human microbiota reached the
high level at University of Tartu already in the second half of the 20th century
(Voronina, 1968; Mikelsaar, 1969; Lenzner, 1973) and has proceeded till
nowadays (Naaber, 1997; Mikelsaar et al., 1998; Annuk, 2002; Mikelsaar et al.,
2002; Naaber and Mikelsaar, 2004). The technological aspects of food related
lactobacilli have been thoroughly studied at Tallinn Technical University (Kask,
2003; Laht, 2003). Since 2001, the Department of Microbiology of the Uni-
versity of Tartu has participated in the EU 5th Framework Programme
PROEUHEALTH (The Food, Gastrointestinal Tract Functionality and Human
Health Cluster), a cooperative investigation of several European universities
aimed to develop new probiotics.
The Lactobacillus fermentum strain ME-3 (previously designated as
822-1-1 and E-3) of healthy human origin was isolated from an Estonian
1-year-old child during studies on development of allergy in two differently
industrialized countries like Estonia and Sweden (Björksten et al., 1994; Sepp,
1998). Its probiotic properties – antioxidative and antimicrobial activity were
assessed at the University of Tartu, Department of Microbiology and Depart-
ment of Biochemistry (Annuk et al., 1999; Mikelsaar et al., 2001). The strain is
deposited in the culture collection (Deutsche Sammlung von Mikroorganismen
und Zellkulturen (DSMZ), DSM 14241 and patented (Application No. 0356/
01PV to the Estonian Patent Agency). The International Bureau of the World
Intellectual Property Organization (WIPO) approved the patent application WO
03002131) and the WIPO experts recognized its novelty in 2003.
In order to be used as functional food or food additive, labeled and mar-
keted according to EU regulations (Commission of the European Communities,
2003) the scientifically proven health claims of L. fermentum ME-3 are
necessary. The present thesis specified the technical applications of the strain in
different food products, its functional properties and the functional efficacy of
L. fermentum ME-3 in healthy human volunteers.

13
 LITERATURE REVIEW

1. Functional foods and probiotics

Although the primary purpose of food is to provide enough nutrients to fulfil
body requirements, various functions of the body are modulated by diet. In
order to compensate for deficiency of certain nutrients in the diet due to changes
in nutritional habits of developed industrial countries, the concept of functional
food has been developed. A food can be regarded as functional if it is
satisfactorily demonstrated to affect beneficially one or more target functions in
the body, beyond adequate nutritional effect, in a way which is relevant to either
an improved state of health and well-being and/or reduction of disease risk "
(ILSI Europe, 1999)
Functional food (FF) is intended for a population generally in normal health
and must demonstrate beneficial effects in amounts that are usually consumed
in the diet. FF is a natural food, to which a component has been added/removed
or a food in which the bioavailability of the components has been modified by
technological or biotechnological means (Korhonen, 2002). FF can be classified
into different groups according to their effect: fat replacers, probiotics, pre-
biotics and dietary fibres, antioxidants, vitamins, polyphenols, plant sterols,
polyunsaturated fatty acids and minerals.
The most promising targets for FF are the GI functions and particularly
control of nutrient bioavailability (Roberfroid, 2000). However, FF can affect
different systems in the body: GI functions (e.g. balanced colonic microflora,
control of transit time and mucosal motility, bowel habits; modulation of
epithelial cell proliferation, balance of redox and antioxidant systems, meta-
bolism of macronutrients, especially amino acids, carbohydrates and fatty acids).
The term “functional food” originates from the 1980s (Sanders, 1999). In
1991, a legal status to functional foods was granted in Japan, indicating foods
for special health use. The first FF probiotic fermented milk drink Yakult has
been available in Japan already since 1935 (Karimi and Peña, 2003).
The term “probiotic” was derived from Greek and means “for life.” Since
the first reference to the positive effecs of beneficial bacteria (Vergin, 1954),
different definitions have been proposed for probiotics. Fuller (1992) defined a
probiotic as “a live microbial feed supplement, which beneficially affects the
host animal by improving its intestinal microbial balance.” According to the
expert panel commissioned by the Food and Agriculture Organization of the
United Nations (FAO) and the World Health Organization (WHO) the present-
day interpretation of probiotics is “live microorganisms which when adminis-
tered in adequate amounts confer a health benefit on the host” (FAO/WHO,
2002).
Probiotics may be administered as a component of FF or as food additives
(e.g. capsules, tablets).

14
 Some authors have intepreted probiotics as “microbial cell preparations or
components of microbial cells that have a beneficial effect on the health and
well-being of the host.” Bacterial cell-wall components, heat-killed whole cells
or metabolites can have a specific probiotic effect, for example, improvement of
lactose digestion or treatment of acute or chronic diarrhoea (Ouwehand and
Salminen, 1998; Romond et al., 1998; Salminen et al., 1999; Simakachorn et
al., 2000; Xiao et al., 2002). Inactivation of probiotics by heat, UV or γ-irra-
diation sustains more or less their ability to adhere to the intestinal mucus in
vitro (Ouwehand et al., 2000). Adherence is considered one of the selection
criteria and pre-requisite for probiotic effects. It has been demonstrated in vitro
that above mentioned inactivation did not negatively affect the ability of
probiotics to bind carcinogens. Yet it is still debatable as, whether a probiotic
must definitely be alive upon digestion (Ouwehand et al., 1999; Sanders and
Huis int Veld, 1999).
Common probiotics include: 1) Lactobacilli such as Lactobacillus acido-
philus, L. johnsonii, L. casei, L. delbrueckii ssp. bulgaricus, L. reuteri, L. brevis,
L. cellobiosus, L. curvatus, L. fermentum, L. plantarum; 2) Gram-positive cocci
such as Lactococcus lactis ssp. cremoris, Streptococcus salivarius ssp. Thermo-
philus, Enterococcus faecium, S. diaacetylactis, S. intermedius; and 3) Bifido-
bacteria such as Bifidobacterium bifidum, B. adolescentis, B. animalis, B. infantis,
B. longum, B. thermophilum (Collins et al., 1998; Gibson, 1999; Mercenier et al.,
2002). Also other microbial species, besides lactic acid bacteria (LAB), like
Bacillus subtilis, Propionibacterium spp. and yeasts (Saccharomyces boulardii)
have been accepted and used as probiotics (Chukeatirote, 2002; Jan et al. 2002).
The mechanism of the action of probiotics (e.g. bifidobacteria and lacto-
bacilli) relies on their metabolic end products, mainly organic acids may lower
the human gut pH at which pathogenic microbes are not able to compete
effectively. Other factors are occupation of normal colonization sites by
probiotics, competition for available nutrients and production of antimicrobial
substances. The second generation of probiotics is genetically modified micro-
organisms providing the host with some necessary components, e.g. production
of immunomodulators (e.g. interleukines) or Helicobacter pylori and rotavirus
antigens (Mercenier et al, 2004).
Probiotic products may be conventional foods (yoghurt, cheese, milk)
(Holzapfel et al., 2001; Temmerman et al., 2002; Yeung et al., 2002) consumed
for nutritional purposes, but also for the probiotic effect; food supplement/
fermented milks or ”medical foods” (e.g. food formulation is a delivery vehicle
for probiotics or their fermentation endproducts are the primary purpose);
dietary supplements: capsules, tablets, liquids, powder (Ross, 2000, Kaur et al.,
2002; Temmerman et al., 2002). Probiotic preparations used as food supplement
can consist of one single strain (e.g. Yacult, Japan – L. casei Sirota) or there are
mixed cultures of two (e.g. Bacilac, Belgium – L. acidophilus plus
L. rhamnosus) or even more (e.g. food supplement VSL-3, Italy containes
8 LAB species) strains.

15
 2. Lactic acid bacteria
2.1. Taxonomy of Lactobacillus spp.

Lactic acid bacteria are Gram-positive non-sporing bacteria, which are

physiologically diverse and include the genera of Lactobacillus, Leuconostoc,
Pediococcus, Streptococcus, Enterococcus, Lactococcus, Oenococcus, Weis-
sella, Carnobacterium, Tetragenococcus, Vagococcus and Bifidobacterium
(Kandler and Weiss, 1986; Klein et al., 1998).
The classification of LAB has mainly remained unchanged since the work
of Orla-Jensen (1919). According to growth temperature, the genus Lacto-
bacillus has been divided into three subgenera termed “Thermobacterium”,
“Streptobacterium”, and “Betabacterium” (Fig. 1). In addition, lactobacilli are
divided into three biochemically diverse groups based on their fermentation
pathways of different carbon sources (pentoses and hexoses) as follows:
obligately homofermentative lactobacilli (OHOL), facultatively heterofermen-
tative lactobacilli (FHEL), and obligately heterofermentative lactobacilli
(OHEL) (Kandler and Weiss, 1986; Hammes and Vogel, 1995; Klein et al.,
1998).
However, the type of carbohydrate fermentation as the traditional basis of
grouping of LAB is not strictly related to the evolution of the organisms.
Phylogenetic classification and identification on the species level relies today
largely on molecular methods based on highly conserved regions in a microbial
genome like 16S ribosomal DNA (16S rDNA) genes (Song et al., 2000). The
studies based on 16S rRNA sequences have shown close relationships between
the genera Lactobacillus, Leuconostoc, Pediococcus, Oenococcus and Weis-
sella.
The principal groupings today are summarised in Table 1 as follows: (1) the
L. delbrueckii group, which contains several obligately homofermentative
lactobacilli (L. delbrueckii with subspecies, L. gasseri, L. acidophilus, L. helve-
ticus, L. johnsonii and L. jensenii) and a few facultatively heterofermentative
lactobacilli; (2) the L. casei – Pediococcus group, comprising the remaining
obligately homofermentative, all heterofermentative and most of the facul-
tatively heterofermentative lactobacilli; and (3) the Leuconostoc group inclu-
ding the species from the above mentioned genera (Stiles and Holzapfel, 1997
modified after Kandler and Weiss 1986; Hammes et al., 1992; Collins et al.,
1998; Klein et al., 1998; Gomes and Malcata, 1999; Sanders, 1999; Holzapfel et
al., 2001; Mercenier et al, 2002).

16
 2.2. Metabolism

Lactobacilli have complex nutritional requirements for organic substrates,

nutritional requirements for amino acids, peptides, vitamins, salts, fatty acid or
fatty acid esters, minerals and fermentable carbohydrates (Kandler and Weiss,
1986). Lactobacilli can adapt to various environmental and nutritional
conditions and change their metabolism accordingly. Metabolically, lactobacilli
are mostly microaerophilic, but they are able to grow at variable oxygen tension
from aerobic to anaerobic.
Most species prefer mesophilic growth temperatures; optimum temperature
is generally between 30–40°C. Further, lactobacilli grow best in a slightly acidic
environment, optimal growth pH from 5.5 to 6.2.
Lactobacilli are capable to degrade different carbohydrates and related
compounds, while the end products are dependent on the fermentation type of
the species (Fig. 1). Lactic acid is the predominant end product, however, under
certain conditions additional products may be acetate, ethanol, succinate or CO2
(Botazzi, 1983; Kandler and Weiss, 1986; Hammes et al., 1992; Klein et al.,
1998).
At the enzyme level, lactobacilli of the OHOL and OHEL groups differ
with respect to presence or absence of fructose diphosphate (FDP) aldolase or
phosphoketolase. Lactobacilli from the OHOL group do not possess FDP
aldolase and are thus unable to ferment pentoses. On the other hand, the
representatives of the OHEL group possess phosphoketolase to break down
pentoses, yielding equimolar amounts of lactic and acetic acids. However, the
FHEL group of lactobacilli, possesses an inducible phosphoketolase with
pentoses acting as inducers. They are thus able to ferment pentoses upon
adaptation to lactic acid and acetic acid, whereas hexoses are homofermen-
tatively metabolised (Kandler and Weiss, 1986; Axelsson, 1998).
The main end product of hexoses – glycolysis or the Embden-Meyerhof
pathway (homolactic fermentation) – is lactic acid, characteristic of the lacto-
bacilli of both the FHEL and OHOL groups. Hexoses other than glucose enter
the major pathways after isomerization and/or phosphorylation. In addition to
lactic acid, other end products are also produced (CO2, acetate, ethanol), mainly
by OHEL (Axelsson, 1998).

17
 Glucose fermentation

D-, L-, DL- D-, L-, DL- DL-lactic acid,

lactic acid lactic acid CO2, acetic acid,
ethanol
Growth at 45°C + + ±
Growth at 15°C –(+)* +(–)** +(–)**
Fermentation of hexoses + + +
Gas from glucose – + +
Fermentation of pentoses – + +
Gas from pentoses – + +
NH3 from arginine –(+)* – +(–)**

Subgenera Thermobacterium Streptobacterium Betabacterium

Fermentation (OHOL group) (FHEL group) (OHEL group)
pathway Incl. L. fermentum
*
Mostly negative, with a few exceptions; **mostly positive, with a few exceptions
Figure 1. Differentiation of lactobacilli after Botazzy, 1983, modified.

Different LAB species may use different pathways depending on conditions and
enzymatic capacity (Kandler and Weiss, 1986; Axelsson, 1998). Change of
LAB metabolism in response to various conditions results in production of
different end products. Mostly, it can be attributed to altered pyruvate meta-
bolism. Pyruvate, intermediately formed in both above-mentioned pathways,
may partly undergo several conversions, producing aroma compounds like
diacetyl and acetoin (2,3-butanediol) or acetate, formate and ethanol.
In some LAB species or strains carbohydrates (especially sucrose) may
contribute to formation of dextrans (slime), important in yoghurt production.
Protein utilisation. Lactobacilli have very limited capacity to synthesize
amino acids from inorganic nitrogen sources, depending on amino acids present
in the growth environment (Axelsson, 1998; Christensen et al., 1999).

18
Table 1. Grouping of lactic acid bacteria (Stiles and Holzapfel, 1997; modified after
Kandler and Weiss 1986; Hammes et al., 1992; Collins et al., 1998; Klein et al., 1998;
Gomes and Malcata, 1999; Sanders, 1999; Holzapfel et al., 2001; Mercenier et al,
2002).
Phylogenetic Fermentation Fermentation Fermentation
group group 1 group 2 group 3
Obligately Facultatively Obligately
homofermenters heterofermenters heterofermenters
Lactobacillus *L. acidophilus, L. acetotolerans,
delbrueckii L. amylophilus, L. hamsteri
group *L. amylovorus,
*L. crispatus,
*L. delbrueckii ssp.
bulgaricus, delbrueckii
and lactis,
* L. gallinarium,
*L. gasseri,
L. helveticus,
L. jensenii,
*L. johnsonii,
*L. kefiranofraciens,
*L. kefirgranum
L. casei – L. aviarius ssp. L. agilis, *L. brevis,
Pediococcus araffinosus and L. alimentarius, *L. buchneri,
group aviarius, *L. bifermentans, *L. fermentum,
*L. fraciminis, *L. casei, L. fructivorans,
L. mali, *L. coryniformis ssp. L. hilgardii,
* L. ruminis, coryniformis and *L. kefir,
L. salivarius ssp. torquens, L. malefermentans,
salicinus and *L. curvatus, *L. oris,
salivarius, L. graminis, * L. parabuchneri,
L. sharpae, L. homohiochii, * L. panis,
Pediococcus L. intestinalis, *L. pontis,
damnosus, L. murinus, *L. reuteri,
P. dextrinum, *L. paracasei ssp. *L. sanfrancisco,
P. parvulus paracasei and L. suebicus,
tolerans, L. vaccinostercus,
L. pentosus, L. vaginalis,
*L. plantarum, Pediococcus
L. paraplantarum, pentosaceus
*L. rhamnosus,
*L. sake (L.
bavaricus),
Pediococcus
acidilactici

19
Table 1.
Phylogenetic Fermentation Fermentation Fermentation
group group 1 group 2 group 3
Obligately Facultatively Obligately
homofermenters heterofermenters heterofermenters
Leuconostoc L. fructosus,
group Weisella confusa
(L. confusus),
*W. (L.) viridescens,
W. (L.)
halotolerans,
W. (L.) hilgardii,
W. (L.) kandleri,
W. (L.) minor,
W. hellenica,
W. (Leuconostoc)
paramesenteroides,
Leuc. amelibiosum,
Leuc. argentinum,
Leuc.lactis,
Leuc. pseudo-
mesenteroides,
Leuc. carnosum,
Leuc. geldium,
Leuc. fallax
Bold – Lactobacillus species used as probiotics
* – Lactobacillus species of food origin
Underlined– Lactobacillus species isolated from human sources

The proteinases of lactobacilli are chromosomally determined. The proteinases

and peptidases of lactobacilli are either bound to cell wall or are intracellular.
They can also act on substrates after cell lysis. Their action site on various milk
protein fractions determines the specificity of proteinases. Proteinases can
hydrolyse either several fractions of casein or be fraction specific. The degra-
dation products of proteins are peptides of various lengths, which are
transported into the cell where intracellular peptidases degrade the peptides into
amino acids (Hammes et al., 1992). The peptidases of lactobacilli can be mono-
mers, trimers, tetramers or multimers with molecular size ranging from 29 to
98 kDa (Christensen et al., 1999). Peptidases have been found mostly in various
species belonging to the OHOL or FHEL fermentation group of dairy-
associated lactobacilli. The proteolytic system of LAB, especially non-starter
lactobacilli present in cheese, contributes to maturation and flavour/aroma
composition of the cheese.

20
Table 2. Bioactive peptides from milk after Meisel and Bocklemann (1999), Bos et al.
(2000) and Boland et al. (2001).
Peptide Origin* Bioactivity
α-Casomorphin αS1-#CNf90-96 and fragments Opioid agonist
β-Casomorphin β-CN 60-70 and fragments Opioid agonist, anti-diarrhea
β-CN 1-4 (f 60-63)
α-Lactorphin α-Lactalbumin 50-53 Opioid agonist
β-Lactorphin β-Lactoglobulin 102-105 Opioid agonist
Lactoferricin Lactoferrin Antimicrobial
Fe transport/regulation
α-Casokinin α-s 1-CN 23-27
β-Casokinin β-CN 177-183
ACE inhibitor**
β-CN 58-72
β-CN f 6-14, f7-14, f73-82
κ-CNf38-39, f25-34, f24-26
Casocidin I α-s 1-CN (f165-203) Antimicrobial (to E. coli, S. aureus)
Caseino α-s 1-CN 43-58; 59-79 Ca2+ uptake, bone/dental
phosphopeptide β-CN 1-25 recalcification
Immuno-modu- α-s 1-CN 194-199 Stimulator of macrophages and
latory peptide β-CN 63-68; 191-193 T lymphocytes
Isracidin N–terminal αs1-CN B (f1-23) Antimicrobial (to S. aureus, Candida
albicans), immunomodulator
Casoplatelin κ-CN 106 and fragments Antithrombotic
Casoxin κ-CN 33-38 ad fragments Opioid agonist
Casoxin C Ileum contracting properties
Glycoma- Appetite suppressant, bifidogenic
cropeptide factor, prebiotic, antimicrobial
Lactoferroxin Lactoferrin+αs1-CN fragment Opioid agonist
Lactoferrin Immune enhancer, prebiotic,
anticancer, antimicrobial (to E. coli,
S. aureus, S. albus, S. mutans,
V. cholerae, C. albicans)
Lactoferricin From lactoferrin Antimicrobial (to E. coli, Klebsiella
pneumonia, S. enteritidis,
S. haemolyticus, S. thermophilus,
Corynebacterium ammoniagenes)
*
The original protein and the peptide sequence
**
ACE – angiotensin converting enzyme
#
CN – casein

21
On the other hand, amino acids, especially arginine, present in the growth
environment (e.g. originating from the primary breakdown of milk casein
during cheese ripening), can be used as an alternative energy source by
lactobacilli (Laht, 2003). Energy is derived through substrate level
phosphorylation, ornithine, CO2 and NH3 being the end products of the process
(Axelsson, 1998). The ability of a lactobacillus species to split arginine and the
appearance of NH3 in the growth environment can be used as one of the
parameters for the fermentation group and species level identification of
Lactobacillus spp (Fig. 1).
The breakdown of casein by lactobacilli during milk fermentation or by
human digestive enzymes after consumption produces a variety of hormone-like
substances or bioactive peptides. Different health promoting activities of
bioactive peptides have been described (Table 2).
Different aspects of metabolism are important in elaborating technical
aspects for probiotic strains to be incorporated into different products. Products
in which bioactive peptides are used are rare in the market.
In the European market, the Valio’s bioactive peptides mediated blood
pressure lowering milk-based drink Evolus® is available. The bioactive
peptides in this product are generated by L. helveticus during fermentation
(Seppo et al., 2002, Seppo et al., 2003).

3. Prebiotics
Prebiotics are defined by Gibson and Roberfroid (1995) as “nondigestible food
ingredients that target certain components within the microbiota of the human
large intestine”. These prebiotics are fermented by one or a limited number of
potentially beneficial bacteria form the resident colonic microflora. A prebiotic
is expected to improve the composition of the colonic microbiota and through
this serve as beneficial to the host health (Gibson, 1999).
The two basic types of fermentations taking place in the gut are
saccharolytic fermentation and proteolytic fermentation. The main end products
of carbohydrate metabolism are the short chain fatty acids: acetate, propionate
and butyrate. These may be further metabolised systematically or locally to
generate energy for the host. The end products of the proteolytic fermentation
include more or less toxic compounds as amines, ammonia and phenolic
compounds. Fermentation in the gut can be modulated towards saccharolytic by
prebiotic consumption.
Research into prebiotics stems from interest in the dietary fibre shown since
the beginning of the 1970s. Much of the interest is aimed at non-digestible
oligosaccharides (fructooligosaccharides, trans-galactooligosaccharides, iso-
maltooligosaccharides, xylooligosaccharides, soyoligosaccharides, glucooli-
gosaccharides and lactosucrose).

22
 More than 36 000 plants worldwide contain FOS; some common sources of
inulin are onion (2–6%), garlic (9–16%), leek (3–10%), banana (0.3–0.7%),
asparagus (10–15%), Jerusalem artichokes (15–20%), chicory (13–20%), and
even wheat (1–4%). Yet the levels are too low for a significant GI tract effect
(Crow, 2004). Consumption of more than 4 grams of FOS daily is needed to
induce changes in LAB levels in the gut, though estimated daily consumption
differs in the US and Europe (Roberfroid, 2000; Gibson, 2001).
Prebiotics are increasingly used in development of new food product, e.g.
drinks, yoghurts, biscuits and table spreads (Gibson and Roberfroid, 1995;
Gibson, 1999). Several prebiotics are available in Europe.
The positive effects of prebiotic consumption are: improvement of bowel
habit; reduction of diarrhoea and constipation; modulation of lipid metabolism
by normalizing cholesterol values; reduction of osteoporosis by improved
mineral absorption; reduction of allergy risk through immune system
modulation; reduction of colon cancer risk (Roberfroid, 2000; Conway, 2001).
Unfortunately, many of the above mentioned health claims still require further
research.

4. Synbiotics
Bifidobacteria and lactobacilli are the most frequent target organisms for
prebiotics. These genera are most commonly used as probiotics too. Probiotics
and prebiotics used in synergistic combination are termed synbiotics. Synbiotics
are mixtures that improve the survival and implantation of live microbial dietary
supplements in the GI tract, either by stimulating growth or by metabolically
activating the health promoting bacteria (Kaur et al., 2002).
Although there is growing interesting development of new FF with
synbiotics, combination of prebiotics and probiotics into a synbiotic has been
studied to a limited extent and needs further investigations, because of the afore
mentioned different substrate requirements for individual probiotic LAB species
and strains. Only a few human studies have been carried out on the
effectiveness of synbiotics (Morelli et al., 2003).

5. Strategy of selection of probiotic strains

Introducing a new probiotic into the market involves a step-wise process in
order to obtain a functional and safe product. Though the genus Lactobacillus
has a Generally Recognized As Safe (GRAS) status and a long history of safe
use for food fermentation, several criteria must be taken into consideration to
select and evaluate a concrete putative probiotic Lactobacillus strain (Collins et
al., 1998; FAO/WHO, 2002; Reid et al., 2003). The properties of a putative

23
probiotic must be thoroughly described in vitro as well as in vivo animal studies
and in clinical trials (Fig. 2).
Selection criteria for probiotics are an area of much debate and should be
taken into account when defining appropriate strains. The following criteria
have been suggested for use in probiotic strain selection (Sanders and Huis in`t
Veld, 1999; Saarela et al., 2000; FAO/WHO, 2002; Reid et al., 2003):
1. General aspects (origin, identity)
2. Safety
3. Functional features
4. Technological aspects
Today few probiotics have been tested according to all recommended aspects in
the scheme by FAO/WHO.

5.1. Strain origin, identification and typing

A probiotic strain retains its functionality in an environment similar to that from
which it was originally isolated. Therefore, a probiotic strain aimed for human
use should be preferably isolated from the healthy human GI tract.
Increasing use of probiotic lactobacilli strains in fermented foods reguires
careful strain identification. A probiotic organism must be differentiable from
the starter microbes of the food product to grant their quality and functionality.
It is also important to demonstrate the survival of the ingested probiotic strain
during transit and its influence on the GI tract microbiota. Besides strain
identity, it is important to link a strain to a specific health effect as well as to
enable accurate surveillance and performance of epidemiological studies.
Traditional methods used for detection of probiotics in the human GI tract
include identification using colony morphology, fermentation patterns,
serotyping and combinations of these methods. Though classical morphological
and biochemical identification will always play an important role, neither is
very definitive because bacteria may exhibit metabolic variation depending on
growth conditions, substrate availability or gene expression.
Many approaches have been developed recently for molecular finger-
printing of lactobacilli strains. Genetic typing allows rapid differentiation of
strains for distinguishing probiotic additives from starter and non-starter
microbes present in food products.
Plasmid profiling, ribotyping (O’Sullivan, 2001), polymerase chain reaction
(PCR) based methods (Yeung et al., 2002; Brandt and Alatossava, 2003), as
arbitrarily primed AP-PCR, triplet arbitrarily primed (TAP)-PCR (O'Sullivan,
2001; Matsumiya et al., 2002) or multiplex PCR (Song et al., 2000), partial 16S
rDNA sequencing (Yeung et al., 2002), randomly amplified polymorphic DNA
analysis (RAPD) (Tilsala-Timisärvi and Alatossava, 1998), pulse-field gel
electrophoresis (PFGE) (O'Sullivan, 2001) have been explored for distinction of
lactobacilli strains from different environments.

24
 Besides molecular DNA-based typing, extraction of whole-cell proteins,
followed by sodium dodecyl sulphate-polyacrylamide gel-electrophoresis (SDS-
PAGE) separation, has been found to be a reliable and rapid way to characterize
a large number of strains (Reuter et al., 2002).

Strain identification by phenotypic and genotypic methods

• Genus, species, strain
• Deposit strain in international culture collection

Functional characterization
• In vitro tests Safety assessment
• Animal studies • In vitro and/or animal
• Phase 1 human study

Double blind, randomised, placebo-controlled Preferably

(DBPC) phase 2 human trial or other appropriate second
design with sample size and primary outcome Independent
appropriate to determine if strain/product is DBPC study to
efficacious confirm results

Technological properties:
Survival in product
manufacturing and shelf
life

Phase 3, effectiveness
trial is appropriate to
compare probiotics with
standard treatment of a Probiotic
specific condition

Labelling
• Contents - genus, species, strain designation
• Minimum numbers of viable bacteria at the end of shelf-life
• Proper storage conditions
• Corporate contact details for consumer information

Figure 2. FAO and WHO (2002) guidelines for probiotics in food, modified

25
 5.2. Safety assessment
Though lactobacilli and bifidobacteria are historically associated with food, they
are normal commensals of the mammalian microflora and their pathogenic
potential is considered quite low. However, as probiotics are, after all, viable
microorganisms, there is the possibility that they could cause infections in
immunocompromised host.
Lactobacillus spp. related systemic (Saxelin et al., 1996; Soleman et al.,
2003) and local infections (Mackay et al., 1999; Rautio et al., 1999) have been
reported in several studies. Other species used as probiotics (such as ente-
rococci, yeasts) pose greater threat than LAB (Sanders and Huis in`t Veld,
1999; Reid et al., 2003). Precautions should be considered for persons with
lowered immune functions.
Before incorporating a probiotic strain into a food product, it should be
tested for safety, to exclude, for example, haemolytic activity and toxin
production in vitro or on animal models. Besides, post-market epidemiological
surveillance of adverse effects should be carried out (FAO/WHO, 2002).
Safety aspects associated with probiotic microbes include the following
specifications (Saarela et al., 2000): healthy human origin, non-pathogenicity
(no thrombocytic activity; no degradation of host mucins, no platelet aggre-
gation properties), no history of association with diseases; not deconjugating
bile salts, no transmissible antibiotic resistance.
Bile acids synthesized in the liver and excreted into the duodenum in the
conjugated form can be chemically modified (deconjugated) in the colon by GI
microbes. Though both conjugated bile and deconjugated bile have anti-
microbial properties, the deconjugated form is more toxic to microbes (Floch et
al., 1972; Stewart et al., 1986). Therefore, it is considered important that the
consumed probiotic lacks the ability to deconjugate bile. The properties of
probiotics to resist bile as well as their conjugated bile salt hydrolase activity
have been until now considered independent properties (Moser and Savage,
2001).
There exists highly varied range of species-specific natural antibiotic
resistance among lactobacilli and bifidobacteria (Yazid et al., 2000; Mändar et
al., 2001; Danielsen and Wind, 2003), mostly non-transmissible. Though plas-
mid based antibiotic resistance is not very common among lactobacilli strains, it
still can occur. On one hand, the antibiotic resistance of a probiotic Lacto-
bacillus strain is favourable, as probiotics are often consumed after antibiotic
therapy to establish the microbial balance. On the other hand, this arises the
need to confirm whether the antibiotic resistance of the probiotic strain is of
chromosomal origin or it is carried by plasmid and is therefore putatively
transferable (Salminen et al., 1998; Saarela et al., 2000).

26
 5.3. Functional features
Several functional aspects are important while selecting a novel probiotic strain
(Saarela et al., 2000). The probiotic strain must essentially be able to survive in
the GI tract (e.g. tolerate acid and bile, adhere to epithelial cells) in order to be
effective in therapeutic actions and carry on normal metabolic activity after
consumption.
Bile and acid tolerance. Environmental factors like low pH in the stomach,
presence of bile acid in the duodenum and intestinal enzymes affect the viability
as well as the adhesive properties of a probiotic introduced into the human GI
tract (Ouwehand et al., 2001). Mechanical factors that may affect the binding
and temporal persistence of a probiotic strain in vivo are peristalsis and mucus
secretion. Probiotic strains that are able to survive and grow at the physiological
levels of bile and low pH in vitro are more likely to survive in the intestinal
transit.
Adhesion and antimicrobial activity. Among human microbiota, lactobacilli
are considered the important colonization resistance-granting bacteria that fight
infectious agents. Adhesion to host gut epithelial cells and intestinal mucus is an
important property of a probiotic strain for temporary colonization of the GI
tract and stimulation of beneficial effects. Adhesion of a probiotic strain might
be closely related to one of the functionally beneficial properties of probiotics –
antimicrobial activity. Probiotic microorganisms can coaggregate with
pathogens or attach to enterocytes (competitive exclusion) and thus inhibit the
binding of enteric pathogens to the intestinal mucosa. Another strategy of
competitive exclusion of a pathogen is the production of inhibitory compounds
like bacteriocins (antibacterial proteins), low molecular mass non-proteinaceous
compounds like different acids (lactic acid, acetic acid, succinic acid) and toxic
oxygen metabolites like hydrogen peroxide, which in combination with the
lactose peroxidase–thiocyanate milk system exerts a bacteriocidal effect on
most pathogens (Klaenhammer, 1988; Mishra and Lambert, 1996; Helander et
al., 1997). The antagonistic activity of a probiotic strain against pathogens such
as Clostridium difficile, Salmonella sp., Helicobacter pylori, Listeria mono-
cytogenes, Escherichia coli should be detected in different milieus according to
the action site of a certain strain (Annuk, 2002; Naaber et al., 2004; Hütt et al.,
2005).
Immunostimulatory properties. By selecting a probiotic strain the
immunostimulatory effect of the strain has to be taken into account. It has been
proved in human studies that probiotics can have positive effects on the immune
system of their host, which has not been linked with inflammatory response, or
any other harmful effects (Salminen et al., 1998; Saarela et al., 2000). However,
there are differences between probiotic bacteria in respect to their
immunomodulatory properties, which should be evaluated for a particular
probiotic strain.

27
 Antimutagenic and anticarcinogenic properties. Another desired health-
promoting properties of microbes (those of food origin or members of the
intestinal microflora) are the ability to neutralize carcinogenic or mutagenic
compounds through modulating the procarcinogenic enzymes of the gut,
through suppression of tumours by immune stimulation or through binding and
degradating carcinogens (Saarela et al., 2000). Although the antimutagenic and
anticarcinogenic properties of probiotic microorganisms have been demonstra-
ted in vitro and on animal models of elucidation, the problem reguires further
clinical evidence.
Antioxidative properties. Many microbes possess enzymatic or non-
enzymatic antioxidative mechanisms to neutralize the oxidative damage caused
by oxygen semi-reduced radical forms.

Table 3. Microbial defence mechanisms for coping with ROS.

Mechanism of action LAB1, 2, 3 ME-34 Neisseria Neisseria
meningitidis5 gonorrhoeae5
SOD O2• detoxification + Mn-SOD
SodC, B SodB low
MntABC Mn accumulation +C +ABC
scavenge O2•–, H2O2
GSH HO• detoxification + +
Gpx/ Peroxide splitting + + gpx A –
GRed
Catalase H2O2 splitting + +high
Sco + +
Chelating Pro-oxidant removal +
Cu2+ Fe2+
NADH Reduction of O2 and +
oxidase H2O2 splitting to H2O
/NADH
peroxidase
H2O2 + +
MsrA/B +/– +
Ccp H2O2 reduction to H2O – +Anaer
HO2• scavenge +
Inhibition of lipid + +
peroxidation
SOD – superoxide dismutase, GSH – reduced glutathione, NADH – reduced nicotinamide-
adeninedinucleotide; Gpx – glutathione peroxidase; GRed – reductase; Sco – thiol: disulfide
oxidoreductase; MntABC – ATB-binding cassette (ABC)-type manganese (Mn) uptake system;
Msr– methionine sulfoxide reductase; Ccp – cytochrome c peroxidase.
1
Lin and Yen, 1999; 2 Lin and Chang 2000; 3 Amanatidou et al., 2001; 4 Kullisaar et al., 2002; 5
Seib et al., 2004.

28
Especially pathogens have to cope with the defence mechanisms of the host and
have therefore developed various strategies forsurvival in the conditions of
oxidative stress and for scavenging reactive oxygen species (ROS) and nitric
oxide species (Seib et al., 2004).
Although, there is increasing evidence regarding the antioxidative effect of
LAB in vitro, the high values seem to be a strain-specific property among
lactobacilli (Lin and Yen, 1999a; Lin and Yen, 1999b; Lin and Chang, 2000;
Stecchini et al., 2001). And although different antioxidative ways to maintain
ROS have been discovered in both pathogens and lactobacilli, as described in
Table 3, the mechanisms require further clarification. Besides, little information
is available about clinical trials where the effect of antioxidative probiotics on
human health parameters is evaluated.

5.4. Clinical testing

The most important proof of probiotic functional efficacy and safety can be
tested with volunteer trials and clinical studies on children and adults. Both
volunteer and clinical trials should be randomized and double-blinded.
Clinical evaluations are usually designed to compare a test strain to a
placebo or control; rarely the efficacy of more than one strain is compared
(Sanders and Huis in`t Veld, 1999; Pathmakanthan et al., 2000). Clinical trials
for probiotic evaluation are divided into three phases (Fig. 2): phase one
(safety), phase two (efficacy) and phase three (effectiveness) (FAO/WHO,
2002; Reid et al., 2003).
Phase one clinical studies on healthy human volunteers are closely
associated with safety evaluation of the tested probiotic strain. A phase-one
study is carried out to test whether a probiotic is well tolerated, and to establish
the putative side effects of consumption of a certain strain containing a probiotic
product.
Phase-two studies compare the efficacy of probiotic strain/product with that
of a placebo on healthy human volunteers. The format of probiotic delivery and
the proper dose of the probiotic are calculated in the course of this phase.
During clinical studies, pharmacological properties of a probiotic strain are
evaluated, such as survival and activity in the human intestine, fecal recovery,
and dose-response relationship. There have been tested the significant
improvement in health condition, well-being, quality of life or reduced risk of
disease (Reid et al., 2003). However, most the above parameters are subjective
and therefore relatively difficult to evaluate objectively.
The main aim of clinical trials is to assess the effectiveness of a probiotic
product together with a standard therapy for a particular disease. Usually the
study group receives a probiotic as an adjunct to standard therapy, while the
control group is treated according to a standard therapy protocol (Reid, 2001).

29
 There are several more or less scientifically proved health effects of
probiotics, which cover improvement of different functions: like alleviation of
lactose intolerance and constipation, treatment and prevention of GI disorders,
reduction of diarrhoeas of different origin and food allergy, reduction of risk of
various diseases like colon cancer or atherosclerosis, regulation of immuno-
modulatory effects and cholesterol control (Ouwehand and Salminen, 1998;
Salminen, 2001).
Several confounding factors (Table 4) may be important and shoud be
considered in administering a probiotic.

Table 4. Factors affecting the performance of a probiotic strain in GI tract (Goldin

et al., 1992; Reid, 2001; Naaber and Mikelsaar, 2004).
Modulator Effect
Intrinsic properties of the strain Antagonistic activity
Bacteriocins
Antioxidative activity
Metabolic status of the probiotic strain Metabolically active
during ingestion Metabolically inactive (frozen, powdered)
Ecosystem in different parts of host GI Low pH in stomach
tract Bile Intestinal enzymes
Individuality of indigenous microflora
Peristalsis Mucus
Dose At least 109 CFU
Duration of administration At least one week
Time and way of consumption Active part of the day
Before bedtime
During meals
Between meals
Nature of food
Growth promoting, “bifidogenic” factor
Milk Protection by buffering capacity
Interference of adhesion
Nature of the probiotic carrier Fermentation products
Presence of supportive cultures
Capsule Coating
Protectants

However, several above modulators need to be defined more precisely by

specific research.

30
 5.4.1. Oxidative stress markers of the human body
Oxidation is essential to living organisms for energy production. Reactive
oxygen species (ROS) are either free radicals, reactive anions containing
oxygen atoms, or molecules containing oxygen atoms that can either produce
free radicals or are chemically activated by them, for example hydroxyl radical,
superoxide, hydrogen peroxide, and peroxynitrite. The main source of ROS in
vivo is aerobic respiration, although ROS are also produced by oxidation of
fatty acids, stimulation of phagocytosis by pathogens or lipopolysaccharides
and tissue specific enzymes. ROS are important in the metabolism of various
compounds. However, the abundance of ROS generated within the body from
different external and internal sources (UV, pollutants or O2 involving bio-
chemical reactions in vivo) cause oxidative stress (Vervaart and Knight, 1996;
Diplock et al., 1998). Oxidative stress is imposed on cells as a result of one of
the three factors: 1) an increase in oxidant generation, 2) a decrease in antioxi-
dant protection, or 3) a failure to repair oxidative damage. The main damage to
cells results from the ROS-induced alteration of macromolecules such as
polyunsaturated fatty acids in membrane lipids, essential proteins, and DNA.
Additionally, oxidative stress and ROS are considered to be important in
pathophysiology of a variety of human diseases, such as some forms of cancer,
cardiovascular diseases, rheumatoid arthritis, and aging or neurodegenerative
diseases like Alzheimer's disease or Parkinson's disease (Diplock et al., 1998;
Eisen, 2002; Esch and Stefano, 2002).
Since ROS are very short-lived, it is very difficult to measure them and
their effect directly. There are several methods for measurement of oxidative
stress markers from human blood and urine. Polyunsaturated fatty acid (PUFA)
containing lipids (e.g. in cell membranes) are very susceptible to peroxidation
upon exposure to ROS (Vervaart and Knight, 1996). Lipid peroxidation is a
pathophysiological process causing apoptosis, but it may also be involved in
tissue damage in inflammation, ageing, cancer, and toxicity of xenobiotics
(Shan et al., 1990; Karelson et al., 2001). Isoprostanes are prostaglandines-like
compounds that are produced upon lipid peroxidation. Human body fluids (e.g.
urine) usually contain low levels of F2-isoprostanes (8-epi-prostaglandin F2α)
that arise by ROS oxidation of phospholipides containing arachidonic acid
(Diplock et al., 1998). Peroxidation of arachidonyl phospholipids results in
formation of the positional peroxyl isomers of arachidonic acid. These inter-
mediates lead to formation of isoprostanes (Morrow et al., 1992). Measurement
of F2-isoprostanes represents a useful approach to assessment of lipid peroxi-
dation and oxidative stress in vivo of the whole body (Diplock et al., 1998).
Glutathione (L-γ-glutamyl-L-cysteinylglycine) is a tripeptide composed of
cysteine, glutamic acid and glycine, which have two biologically important
structural features: a thiol (SH) group and γ-glutamyl linkage (Shan et al.,
1990). In humans it is present in the millimolar range mainly in the red blood
cells, liver, pancreas, kidneys, spleen, eyes, lungs and intestinal cells (Meister

31
and Anderson, 1983). In cells, total glutathione can be free or bound to proteins.
Free glutathione is present mainly in its reduced form, which can be converted
to the oxidised form during oxidative stress, and can be reverted to the reduced
form by the action of enzyme glutathione reductase. The GSH is the crucial
cellular non-enzymatic antioxidant. The oxidised form of glutathione (GSSG)
becomes toxic even at low levels; therefore the glutathione red-ox ratio
(GSSG/GSH) is maintained as low as possible in the cells (Pastore et al., 2003).
In the case of inflammation this balance is shifted towards the oxidized form,
indicating non-physiological intracellular oxidative stress. Measurement of the
various forms of glutathione concentrations in biological samples is important
for the understanding of GSH homeostasis in health and disease. Because blood
glutathione concentrations may reflect the glutathione status in other less
accessible tissues, measurements of both GSH and GSSG in blood have been
considered a useful indicator of disease risk in humans. Low GSH and a high
GSSG/GSH ratio have been found in the blood of patients with various diseases
(Pastore et al., 2003).

5.5. Technological aspects of probiotics

The aspects related to the production of probiotic food and food additives are of
utmost importance in providing products of good biological and technological
quality. Technological aspects include: viability during processing, good
sensory properties, phage resistance and stability in final formulation and during
storage

5.5.1. Product manufacturing

Difficulties with the production of probiotic products are due to the human
origin of probiotic strains. Several challenges can arise because the environment
within the human GI tract and that in food may be quite different.
Fermented dairy products, especially yoghurts and yoghurt-like products are
most widely used probiotic carriers (Sanders and Huis in’t Veld, 1999;
Holzapfel et al., 2001; Yeung et al., 2002). There is a technological reason for
this: many dairy products have already been optimized for survival of starter
lactobacilli and are relatively easily adapted to grant survival of probiotic strains
as well. Cheese is used as a probiotic vehicle to a less extent than fermented
milk products. Additionally, probiotics can be applied in non-dairy foods such
as juices and cereals or in a freeze-dried form in special formulations like
capsules, powders and tablets.
The fermentation technology used during strain or product manufacturing is
of major importance for microbiological stability. During product manu-
facturing the chemical composition of the fermentation medium (availability of
nutrients, carbohydrate source, and presence of inhibitors in the food matrix

32
such as NaCl), cultivation conditions (inoculation level, incubation temperature,
fermentation time) or final acidity and flavour additives can affect the probiotic
strain. Also subsequent handling of the product (e.g. cooling the product after
fermentation) and packaging (Lee and Wong, 1998, De Vuyst, 2000) can affect
the viability of the probiotic strain.
At the end of fermentation and during shelf life of products several stressors
occur simultaneously (e.g. carbon source starvation combined with low pH)
(Champomier-Vergès et al., 2002). The property of a probiotic strain to tolerate
very low or high temperatures and/or dehydration is quite important, as
probiotic cultures as food additives are mostly produced in a frozen and freeze-
dried or spray-dried form. Heat tolerance of a probiotic strain favours its
survival in conditions of temperature variatons during product manufacturing
when technology foresees a short-term heat treatment (e. g. spray-drying) of a
raw material with added probiotic cultures. These technological properties are
strain-specific and need to be evaluated separately for every strain.

5.5.2. Viability of probiotics and interaction with starter cultures

Stability of commercial probiotic strains is important in ensuring that stated
levels of viable cells are delivered in probiotic products. According to the
FAO/WHO, the suggested minimum numbers of probiotic bacteria at the end of
the products shelf life and at the time of consumption should be minimally
106–107 CFU per g of food (De Vuyst, 2000; Reid, 2001). The minimum thera-
peutic dose per day is 108–109 viable cells (Reid, 2001), which can be gained
through consumption of 100 g of the product. Besides, a probiotic strain should
not only be viable but also maintain its probiotic characteristics throughout
product manufacturing and storage.
As microbial interactions can be either beneficial or antagonistic, the
suitability of a probiotic strain with starter microbes should be tested befo-
rehand in order to obtain the most suitable combination for a particular product
and to avoid undesirable changes in the composition of the product`s microflora
during manufacture and storage.
Heterofermentative LAB as weak lactic acid producers can create some
unwanted by-products as glucose is metabolised to both lactic acid and acetic
acids. The latter gives an undesirable “vinegary” sharp taste. Carbon dioxide,
produced by heterofermentative strains, may disrupt the food matrix. Therefore,
it is important that the probiotic culture used in fermented products contributes
to good sensory properties, e.g. absence of off-flavour or texture.
To avoid problems with slow acidification and formation of unwanted
byproducts, as well as to control the flavour and aroma of the product, probiotic
bacteria are combined with a support or starter culture suited for the fermen-
tation of the specific product (Sanders and Huis in’t Veld, 1999; Saxelin et al.,
1999; Saarela et al., 2000).

33
 In selecting a suitable starter, its negative impact on probiotic survival
should also be taken into consideration. Survival of a probiotic may be
influenced by the metabolites of the starter cultures such as lactic or acetic acid,
hydrogen peroxide or bacterocins (Saarela et al., 2000).

6. Origin and history of Lactobacillus fermentum ME-3

Strain origin. The Lactobacillus fermentum strain ME-3 (previously designated
as 822-1-1 and E-3) was isolated from a fecal sample of one-year-old healthy
Estonian child during a comparative study of the lactoflora of Estonian and
Swedish children and some of the properties have been described in different
previous studies (Sepp et al., 1997; Mikelsaar et al., 2001; Annuk, 2002;
Mikelsaar et al., 2002; Annuk et al., 2003). The L. fermentum ME-3 is depo-
sited in the culture collection (Deutsche Sammlung von Mikroorganismen und
Zellkulturen (DSMZ), DSM 14241).
Antimicrobial activity. ME-3 has strong antimicrobial activity against
Gram-positive and Gram-negative entero- and uropathogens (Annuk et al.,
1999) and moderate activity against Helicobacter pylori (Hütt et al., 2005).
L. fermentum ME-3 has been tested for production of H2O2 in a qualitative assay
as well as by a quantitative method (Kullisaar et al., 2002).
Antioxidative properties. The cells and cell lysate of L. fermentum ME-3
have high antioxidative potency. The cells have Mn-superoxide dismutase (Mn-
SOD) actvity, contain reduced glutathione and scavenge hydroxyl and peroxyl
radicals. In addition, the cells of ME-3 have high values of total antioxidative
activity (TAA%) (Kullisaar et al., 2002).

7. Unsolved problems
Several “bottle-necks” are still unsolved in probiotic development and
application. Besides the mentioned strain origin and safety, problems connected
with the technological accuracy of production of FF or food additives
containing particular probiotic strains have to be efficiently solved. Only in a
few investigations the technological conditions for the processing of a probiotic
product are selected as a result of profound research.
Selection of markers characterizing specific probiotic-induced functional
processes of host is of utmost importance. To prove probiotic health claims,
there are no definite indications to measure the positive effect on the
physiological and biochemical indices of human health. It has not yet been
assessed if the presence of a probiotic strain in feces could be an important
marker for a probiotic positive effect.
Moreover, no systematic studies have been performed to approve the
functional efficacy of different probiotic formulations on the antioxidative
defence system of the healthy host.

34
 AIMS OF THE STUDY
The general goal of the research was to assess if the Lactobacillus fermentum
strain ME-3 is suitable as a component of functional food or food additive with
antioxidative health claim in normal population.

The following objectives were designed to be achieved:

1. To assess in vitro various technological properties of L. fermentum ME-3:

• effect of high acidity on the viability of the strain;
• heat resistance of the strain in neutral and acidic environments;
• survival in limited nutritional conditions;
• possible interactions between ME-3 and starter cultures or non-starter
lactobacilli;
• viability of L. fermentum ME-3 in various fermented milk products, in
juices and in capsules.

2. To study the viability of L. fermentum ME-3 on the basis of a simulated

gastrointestinal digestion model with pepsin, hydrochloric acid, pancreatin and
bile.

3. To estimate the stability of the functional properties of L. fermentum ME-3 in

different fermented milk products, juices and capsulated formulations.

4. To evaluate the safety and health improvement properties of ME-3 in human

volunteer trials with healthy persons:
• detect the tolerability of the strain and possible side effects;
• estimate the recovery of the strain in the human gastrointestinal tract by
bacteriological and molecular methods;
• measure the effect of ME-3 consumption on total fecal lactoflora count;
• measure the effect on reduction of oxidative stress in the humans through
testing the key markers of blood and urine;
• optimize the dose of the probiotic strain in two different formulations.

35
 MATERIALS AND METHODS
A summary of the materials and methods used in this study is presented in
Table 5 and, in addition, a detailed description is available in the following
section and, for particular cases, in Papers I to V.

Table 5. Study subjects and microbial strains.

Study subjects Type of study Presented in:
Intrinsic antibiotic resistance Paper I
Production of metabolites by fermentation Paper I
Antimicrobial activity in vitro against 3 E. coli Paper I
strains, S. Typhimurium, S. Enteritidis, S. aureus, Paper II
The strain of S. sonnei
Lactobacillus Lactic acid, acetic acid, HCl, pancreatin, pepsin and
fermentum ME-3 bile tolerance in vitro
Heat resistance in skim milk in physiological saline Present study
and in different juices
Growth on starter Probat 505 and supportive
culture L. plantarum LB-4 suspensions in limited
nutritional conditions
Survival of probiotic additive L. fermentum ME-3
throughout product’s shelf life
Probiotic Stability of antimicrobial activity of L. fermentum
fermented milk ME-3 reisolates from products against 3 E. coli
products of the strains, S. Typhimurium, S. Enteritidis, S. aureus, Present study
HELLUS brand S. sonnei
Stability of total antioxidative activity of L.
fermentum ME-3 reisolates from products
Survival of probiotic additive L. fermentum ME-3
throughout storage and ripening of cheese
Probiotic cheese Stability of antimicrobial activity of L. fermentum
based on smear- ME-3 throughout storage and ripening of cheese
ripened semi-soft against 3 E. coli strains, S. Typhimurium, Paper III
cheese Pikantne S. Enteritidis, S. aureus, S. sonnei
Stability of total antioxidative activity of L.
fermentum ME-3 throughout storage and ripening
of cheese
Survival of probiotic additive L. fermentum ME-3
Probiotic open- throughout cheese storage and ripening Present study
texture cheese Stability of total antioxidative activity of
Atleet L. fermentum ME-3 throughout storage and
ripening of cheese

36
Table 5. (Continuation)
Study subjects Type of study Presented in:
Healthy human Effect of consumption of fermented goat milk
volunteers (n=21) containing ME-3 on the composition human fecal Paper IV
lactoflora Paper V
Healthy human Antiatherogenicity of probiotic fermented goat milk
volunteers (n=21)
Healthy human Safety of ME-3 consumption Paper V
volunteers (n=22)
Healthy human Effect of ME-3 capsule consumption on the Present study,
volunteers (n=22 composition human fecal lactoflora and on the Paper V
and n=25) antioxidative markersof the human body.
Different doses and regimens

8. Origin of bacterial strains (Paper IV, present study)

8.1. Lactobacillus strains

All lactobacilli strains used in this study belonged to the culture collection of
the Department of Microbiology of the University of Tartu.
The probiotic strain L. fermentum ME-3 was previously isolated from the
GI tract of a 1-year-old healthy Estonian girl. L. buchneri S-15 originated from
a 2-years-old healthy Swedish infant (Mikelsaar et al., 2002). L. plantarum LB-
4 is an original isolate from cheese whey. A total of 17 non-starter lactobacilli
strains isolated from fresh goat or cow milk, cheese and cheese whey were used,
including L. acidophilus (3 strains), L. plantarum (4 strains), L. casei (4 strains),
L. fermentum (3 strains), L. brevis (3 strains).
Furthermore, four reference strains were used in the study: L. buchneri
ATCC 4005, L. brevis ATCC 14869, L. reuteri DSM 20016 and L. fermentum
ATCC 14931.

8.2. Identification of Lactobacillus fermentum ME-3

(Paper I, present study)

Identification of L. fermentum ME-3 on the species level. The Lactobacillus sp.

strain was identified from feces according to morphological and cultural
properties: gas formation from glucose, hydrolysis of arginine, negative catalase
activity (Lencner et al., 1984; Kandler and Weiss, 1986). The strain was
identified on the basis of carbohydrate fermentation patterns with the API
50CHL System (bioMérieux, Marcy l’Etoile, France) and Internal-Transcribed
Spacer Polymerase Chain Reaction (ITS-PCR).

37
 ITS-PCR, followed by enzymatic restriction Taq I was used to confirm the
identification of the species. The DNA extraction from Lactobacillus isolates
was prepared as described by Alander et al. (1999) using lysozyme (Serva,
Sweden; 20 mg/ml), mutanolysin (Sigma; 0.5 mg/ml) and proteinase K
solutions (Fermentas, Lithuania; 14.6 mg/ml) and the work was carried out with
the help of J. Shchepetova (extraordinary research fellow at the Department of
Microbiology, University of Tartu). The DNA amplification was performed
according to Jacobsen et al. (1999) in a reaction volume of 50 µl containing
1xTaq polymerase buffer (Fermentas, Lithuania), 1.5U Taq polymerase
(Fermentas), 0.5 µM of each primer (16S–1500F and 23S–32R; DNA Techno-
logy AS) (Jacobsen et al., 1999), 200 µM deoxynucleoside triphosphates, 2 mM
MgCl2 and 2 µl of extracted DNA.
Subsequently, the PCR product was restricted as described by Zhong et al.
(1998) using a Taq I restriction enzyme (Fermentas). DNA fragments were
separated by electrophoresis (1.5h, 100 V) on a 2% agarose gel in a 1xTBE
[Tris(Hydroxymethyl)aminomethane-borate/disodium ethylendiamine tetra-
acetate] buffer. A size marker 100 bp DNA Ladder Plus (Fermentas) was also
separated simultaneously on the same gel.
The banding pattern of the isolates was visually compared with that of the
above-mentioned Lactobacillus reference strains.
Rapid identification of L. fermentum ME-3. Colonies of less than 24-hours-
old culture of ME-3 are relatively big, flat and rough with irregular edges and of
different morphology. In the Gram preparation more elongated and yet
undivided cells can be seen compared to the older culture. Over 24-hours-old
colonies of ME-3 are greyish-white, convex with regular edges. Microscopic
evaluation after Gram staining shows regular, Gram-positive plump rods, which
are variable in length, mostly occurring in parallel pairs.
ME-3-like colonies were reisolated from different environments (feces,
food products). For rapid identification of ME-3, three biochemical tests – gas
from glucose, growth at 15°C and lysozyme production was carried out by a
modified scheme of Lencner et al. (1984). The latter was elaborated for rapid
differentiation of L. fermentum species from the other representatives of the
OHEL group.
Lysozyme production was detected on a modified de Man-Rogosa-Sharpe
(MRS) medium containing 10 % of the inactivated Micrococcus lysodeikticus
culture. The transparent zone around selected potential ME-3 colonies, grown in
microaerobic environment, indicated positive lysozyme production due to cell
wall lysis of micrococci by enzyme (Lenzner and Lenzner, 1982). Growth at
15°C was estimated in MRS broth after 7 days of incubation. Turbidity in the
growth medium was considered a positive test result.
The identification of strain L. fermentum ME-3 was confirmed by AP-PCR.

38
 8.3. Basic characteriation of Lactobacillus fermentum ME-3
(Papers I, II, III)

The antibiotic resistance of L. fermentum ME-3 was determined in

cooperation with Dr. R. Mändar, Department of Microbiology. Ampicillin,
cefoxitin, gentamicin, ciprofloxacin, tetracycline, ofloxacin, aztreonam,
trimethoprim-sulfamethoxazole and vancomycin susceptibility of the strain was
tested using antibiotic strips of the E-test (AB Biodisk, Solna, Sweden). A
thioglycolate broth for suspending bacteria (McFarland 0.5 turbidity standard),
Wilkins-Chalgren (Oxoid Ltd. Basingstoke, Hampshire, UK) agar plates with
5% horse blood was used (Mändar et al., 2001). After 36h of incubation at 37°C
microaerobically the MIC-s (minimal inhibitory concentration) and breakpoints
(susceptible/resistant) were determined in accordance with the NCCLS
guideline (Tenover et al., 1999).
Two antibiotics – erythromycin and metronidazole – were tested using the
Kirby-Bauer disk-diffusion test. BBL (Becton Dickinson, Cockeysville, USA)
Sensi-Disk Susceptibility Test Disks were used (erythromycin 15µg,
metronidazole 5µg). The inoculums, agar plates and incubation conditions were
similar to those used for MIC tests.
The profile of metabolites. The production of organic acids and ethanol
(mg/ml) was estimated by gas chromatography as described by Holdeman et al.
(1977) in cooperation with J. Shchepetova. The gas chromatograph (Hewlett-
Packard model 6890, USA) was equipped with a hydrogen flame ionization
detector and an auto sampler (model 7683). The HP Chemical Station for GC
System (A.06 revision) was used. Analyses were performed following
cultivation of the Lactobacillus in modified the MRS broth for 48h in a 10%
CO2 environment.

8.4. Pathogenic target bacteria

The reference strains of Escherichia coli K12, E. coli ATCC 700336, E. coli
ATCC 700414, Shigella sonnei ATCC 25931, Staphylococcus aureus B46,
Salmonella Enteritidis ATCC 13076 and two clinical isolates of Salmonella
enterica ssp. enterica serovar. Typhimurium were used as the target bacteria.

39
 9. Properties of L. fermentum ME-3
9.1. Resistance to low pH, bile and heat in vitro
(Paper V, present study)

Resistance to low pH. The ability of L. fermentum ME-3 to tolerate low pH was
tested in hydrochloric (HCl), lactic and acetic acid environments.
L. fermentum ME-3 was precultured on a MRS agar medium. The overnight
grown cells were harvested and suspended in saline (0.9% NaCl). The amount
of 0.5 ml of suspension, according to McFarland 4 standard suspension (109
CFU/ml), was inoculated into 4.5 ml of the MRS broth, pH adjusted from 2.0 to
4.0 with either HCl, DL-lactic or acetic acid.
Bile tolerance. Bile tolerance of L. fermentum ME-3 was detected in the
MRS broth with 0.3 to 2.0% of ox gall (Sigma-Aldrich Chemie GmbH, Stein-
heim, Germany). Survival was detected on the MRS agar medium after
2, 4, 6, 8 and 24h. The samples were serially diluted in saline, plated on MRS
agar and incubated for 24h in a variable atmosphere incubator (IG 150, Jouan,
France) with the following microaerobic atmosphere CO2/O2/N2: 10/5/85 at
37°C.
Survival in simulated gastric digestion. The cumulative effect of low pH,
digestive enzymes and bile was tested in simulated gastric digestion. The cell
suspension of the overnight L. fermentum ME-3 culture was prepared as
described above and added to 10 ml of phosphate-buffered saline (PBS) with
1.5 M NaCl (Naxo Ltd, Tartu, Estonia) and pepsin (3g/l, Sigma, EC3.4.23.1),
where pH was adjusted to pH2.5 with HCl.
In parallel the effect of simulated gastric duodenal juice on the capsulated
lyophilized L. fermentum ME-3 culture was tested with two different methods.
One capsule was dissolved in 4.5 ml of pepsin containing PBS (pH2.5). Second,
the content of a capsule was added to the mixture together with 0.3ml of skim
milk to simulate consumption of milk. All variants were incubated at 37°C for 3
hours, after which the cells were centrifuged and resuspended in PBS (pH 7.5)
containing 0.1% of pancreatin (BMP Production Gmbh, 19370 Parchirn,
Germany) and 0.1% of ox gall (Sigma-Aldrich Chemie GmbH, Steinheim,
Germany) and incubated for 3 more hours. Aliquots were taken for enumeration
of viable cells at 0, 1.5h, 3h, 4.5h and 6 h from the beginning of the trial. The
harvested cells of the overnight L. fermentum ME-3 culture (109 CFU/ml) in a
PBS buffer (pH7.5) without enzymes and bile served as control.
Heat resistance. ME-3 was precultured overnight on the MRS agar
medium. The amount of 0.17 ml of ME-3 cell suspension according to
McFarland 4 was inoculated into 1.5 ml of 10% reconstituted sterile skim milk
(Oxoid Ltd. Basingstoke, Hampshire, UK) or into 0.9% NaCl preheated to
required temperatures (73°C, 85°C or 95°C). The samples were taken after 5
and 10 minutes.

40
 The heat resistance of ME-3 in an acidic environment was estimated in two
ways. The strain was precultured on the MRS agar medium and the suspension
adjusted to McFarland 4 in physiological saline was prepared as described
above. The survival of ME-3 in three different juices: multivitamin nectar (pH
3.57), orange juice (pH 3.85) and tropical drink with carrot (pH 3.05), produced
by Tallinn Dairy Ltd. from natural juice concentrates (no stabilizers) was
estimated during 4 minutes at 96°C.
Additionally, the heat resistance of a frozen (–80°C) ME-3 skim milk
suspension (5x109 CFU/ml) in three juices was estimated as follows. The frozen
suspension was melted at room temperature (22°C). The melted suspension was
added to a juice (0,4 ml/l) at different times (0h, 1h, 2h, 3h) after melting and
the mixture was kept at 96°C during 15 s. ME-3 mixture in juice without heat
treatment served as a control.
All experiments were carried out in a dry-block thermostat (Biosan
TDB120, MyLab, SeouLin Bioscience Ltd, Seoul, Korea).
All samples were cooled, serially diluted in saline, plated on the MRS agar
and incubated for 48 h at 37°C microaerobically.

9.2. Growth in cell suspensions (Present study)

L. fermentum ME-3, lactococci from the starter culture of Pikantne cheese

Probat 505 (Danisco A/S, Copenhagen, Denmark) and L. plantarum LB-4 were
precultured for 24h on the modified MRS agar medium without triammonium-
citrate and sodium-acetate (pH 7.2) microaerobically. Two suspensions in 0.9%
NaCl (9.9 ml each), adjusted to McFarland 4, were prepared from the cheese
starter culture and L. plantarum LB-4. One of the suspensions of either culture
was thermally inactivated by boiling for 2 minutes and cooled.
All suspensions were inoculated with 0.1 ml L. fermentum ME-3 suspension
(105 CFU/ml) to a final count of about 103 CFU/ml. The suspensions were
incubated for 7 days at 37°C and the changes in cell count were evaluated by
making serial dilution from the suspensions and inoculating 0.1 ml of each
dilution on the modified MRS agar medium. A suspension of ME-3 in
physiological saline served as control. Different isolated were identified by
colony morphology and were microscopically evaluated after Gram staining.

9.3. Antagonistic activity (Papers I, II, III, present study)

On agar, the antagonistic activity of the L. fermentum ME-3 culture and its
isolates from fermented milk products, cheese and capsules against pathogens
was assessed using a streak line method on the modified MRS agar medium
without triammonium-citrate and sodium-acetate (pH 7.2). A single line of ME-

41
3 culture, grown in the MRS broth for 48h, was seeded in the middle of an agar
plate and cultivated for 48h at 37°C in a 10% CO2 environment and
subsequently inactivated using chloroform gases for 2h. The target pathogens
were cultured in a peptone broth (CM9 Unipath Basingstoke, Hampshire, U.K.)
for 24h at 37°C and seeded in duplicate perpendicular to the streak line of
lactobacillus. Following incubation of the plates at 37°C for 24h, the width of
the zone of inhibition (mm) of the target bacteria extending from the culture line
of the lactobacilli seeded in the middle of the agar plate was measured
(Mikelsaar et al., 1987). The width of the growth inhibition zone 0 to 13.0 mm
was considered low; 13.0 to 25.0 mm – intermediate and >25.0mm – high (Hütt
et al., 2005).
The antagonistic activity of the non-starter lactobacilli isolates from cheese
and cheese whey, cow milk or goat milk was measured using the same method.
Non-starter lactobacilli were precultured in the MRS broth for 48h and the test
was performed on the MRS agar medium. The plates were incubated at 37°C
microaerobically for 48h. The inhibitory activity of L. fermentum ME-3 on the
growth of the cheese starters (Probat 505 and starter mix CHN 11, CHN 19 and
Flora Danica, Christian Hansen Holding AS, Denmark) and the yoghurt starter
Jo-Mix VK 1-30 (Danisco A/S, Copenhagen, Denmark) was tested on modified
MRS (without either tri-ammonium-citrate or sodium-acetate, pH 7.2). All
starters were tested as mixed cultures and no different species were isolated
beforehand.
Milk and broth. The antagonistic activity of L. fermentum ME-3 in milk was
assessed against starter cultures and against the pathogens E. coli K12, S. sonnei
ATCC 25931, S. aureus B46, and two strains of Salmonella Typhimurium. The
target pathogens were cultured in a peptone broth.
The starter cultures were precultured in sterilized (120°C for 15 min) milk
(fat content 2.5%) for 48h at 37°C. L. fermentum ME-3 was precultured as
described above. Milk was inoculated with equal aliquots of ME-3 and starter
culture suspensions adjusted to McFarland 4 (109 CFU/ml) and co-incubated.
Serial dilutions were made and plated onto the modified MRS agar medium and
incubated microaerobically. The identification of Lactobacillus and cocci was
microscopically confirmed by cell morphology after Gram staining.
The survival of target pathogens was detected on peptone agar aerobically
for 24 h at 37°C. Different colonies were counted.
The antagonistic activity of ME-3 against pathogens in broth was
determined in the modified MRS broth and in milk. Equal aliquots of pathogens
and L. fermentum ME-3 suspensions were co-incubated. Subsequently, the
number of colony forming units (CFU/ml) of pathogens was semi-quantitatively
determined on peptone agar (Gould, 1965).

42
 10. Growth and survival in different products
Capsule. Gelatine coated capsules were manufactured by the Tallinn
Pharmaceutical Company Ltd. The probiotic capsules contained freeze-dried L.
fermentum ME-3 (109 CFU) with the addition of 250 mg of saccharose and
microcellulose. Identical placebo capsules contained only saccharose and
microcellulose. All capsules were stored at +4°C.
Survival of ME-3 in capsule was measured by dissolving the content of one
capsule aseptically in 2 ml of a 0.9% NaCl solution. The suspension was
vortexed, serially diluted and plated 0.1 ml on the MRS agar medium (Oxoid
Ltd. Basingstoke, Hampshire, U.K.). The plates were incubated for 48 hours at
37°C microaerobically (10% CO2) and the colonies were counted. Viable cell
count per capsule was calculated.
Juices. The ability of L. fermentum ME-3 to grow in different juices was
estimated in multivitamin nectar, orange juice and tropical drink with carrot
produced by Tallinn Dairy Ltd.
A suspension according to McFarland 4 was prepared from L. fermentum
ME-3, precultured for 24 h microaerobically on the MRS agar. The juices were
inoculated with L. fermentum ME-3 to a final count of about 5x105 CFU/ml and
incubated for 48 h at 37°C. Changes in cell count were evaluated at the
beginningof the trial, at 24h and 48h by the serial dilution method as described
above.
Survival of L. fermentum ME-3 in the juices was estimated during one
month. The juices were inoculated with an approximate concentration of viable
cells 107 CFU/ml of the strain and stored at 4°C. Samples were taken at 48 h, 2
weeks and one month from the beginning of the trial, and were serially diluted
as described above.

10.1. Preparation and survival in probiotic cheese

(Paper III, present study)

The Vana-Kuuste Dairy Ltd developed a probiotic cheese, based on the

Estonian open texture smear-ripened semi-soft cheese Pikantne as described in
Paper III.
Probiotic cheese, based on the Estonian open texture semi-hard cheese
Atleet (a cheese from the Svecia group) was developed by Võru Juust Ltd. from
11.7 tons of whole milk with 170 l of the precultured starter mix CHN 11, CHN
19 and Flora Danica (Christian Hansen Holding AS, Denmark) containing
Lactococcus lactis ssp. lactis; Lactococcus lactis ssp. cremoris; Lactococcus
lactis ssp. lactis biovar diacetylactis; Leuconostoc mesenteroides ssp. cremorsi.
The amount of 0.4% of L. fermentum ME-3 (109 CFU/ml) was added together
with the starter.

43
 Inoculation rate for both cheeses with ME-3 was 108 CFU per gram of
cheese. The ripening of Pikantne cheese lasts 30 days and the shelf life is also
30 days. For Atleet cheese the respective periods are 45 days and 8 months. The
survival of the strain ME-3 in both probiotic cheeses and its number per gram
during ripening and storage was analysed on the 10th, 24th, 38th, 54th and 66th day
of cheese preparation for Pikantne cheese and on the 12th, 20th, 29th, 40th, 63rd,
118th, 217th and 242nd day for probiotic Atleet cheese.

10.2. Preparation and survival in fermented goat milk

(Papers IV, V, present study)

Combining the probiotic ME-3 strain with two supportive lactobacilli cultures
of different origin allowed development of experimental fermented probiotic
goat milk. L. buchneri S1-5 tempered the specific taste of goat milk.
L. plantarum LB-4 was included as an efficient producer of exopolysaccharides,
which gives the fermented milk cream-like consistence and delicate acidity.
L. fermentum ME-3 was used as the probiotic additive. Each Lactobacillus
strain was incubated for 48 h in the MRS medium at 37°C in microaerobic
conditions. Goat milk was inoculated with a 2% mixture of Lactobacillus
strains and incubated at 37ºC for 24 hours. The ready product was cooled and
stored at 4°C. The count of probiotic L. fermentum ME-3 amounted to 3x109
CFU/ml of the product.
To measure the viable cell count of ME-3 in fermented goat milk, samples
were taken at the end of fermentation (and before cooling the product), after
24h, 48h, 72h and 7 days of preparation when the product was stored at 4˚C.
The amount of 0.5 ml of fermented milk was serially diluted in saline and plated
on the MRS agar medium and incubated for 48 h at 37°C in microaerobic
conditions.

10.3. Preparation and survival in HELLUS fermented

milk products (Present study)

HELLUS is a brand name for four different fermented milk products (yoghurt,
kefir, quark, sour cream), containing L. fermentum ME-3 as a probiotic additive,
produced by Tallinn Dairy Ltd.
Yoghurts (fat content 3%) were produced using 0.1% of the thermophilic
starter cultures Lactobacillus delbrueckii ssp. bulgaricus and Streptococcus
thermophilus.
Yoghurts with four different flavours (flavour additive content 15%) were
prepared: “Rhubarb-oat” with rhubarb-oat jam; “Pasha” with apricot-orange jam
and 8% of raisins; “Wild berries” with youngberry, blueberry and raspberry jam

44
or “Cowberry-muesli” containing whole grains of wheat, barley and oat and
cowberry jam.
Using the mesophilic cultures of Lactococcus lactis ssp. lactis and L. lactis
ssp. cremoris as the starters, three different flavoured quarks (fat content 4%)
were produced with 19% of different jams: “Tropical” with peach jam,
“Raspberry” and”Wild strawberry.”
Mesophilic aromatic L. lactis ssp. lactis, L. lactis ssp. cremoris, L. lactis ssp
diacetylactis and Leuconostoc mesenteroides ssp. cremoris were used as the
sour cream (fat content 20%) starter cultures.
Mixture of mesophilic lactic acid bacteria L. lactis ssp. lactis, L. lactis ssp.
cremoris, L. lactis ssp diacetylactis and Leuconostoc mesenteroides ssp.
Cremoris, Lactobacillus kefir, two yeasts Klyveromyces marxianus and
Saccharomyces unisporus were used as starter cultures for the production of
kefir (fat content 2.5%). All starters originated from Danisco A/S.
The freeze probiotic ME-3 culture (5x109 CFU per ml of saline) was
activated in a small amount of milk and added in vat simultaneously with starter
cultures. The dose depended on the product’s fat content: 400 ml per ton for yog-
hurt and kefir, or 500 ml per ton of product for sour cream and quark dessert.
The viability of ME-3 in HELLUS brand products (yoghurt, kefir, quark
dessert, sour cream) was analyzed at different times of the product’s shelf life.
The products were analysed during 6 months. The shelf life of kefir and quark is
15 days that of yoghurt and sour cream 20 days. The amount of 0.5 ml of a
product was serially diluted in saline and plated on the MRS agar medium and
incubated for 48 h at 37°C in microaerobic conditions.

11. Total antioxidative activity of ME-3

11.1. In vitro testing of antioxidative activity
(Paper III, present study)

The antioxidative activity of L. fermentum ME-3 was measured in vitro. The

antioxidative influence of ME-3 was estimated in body fluids in trials with
healthy human volunteers. All further work was carried out in cooperation with
Dr. T. Kullisaar from the Department of Biochemistry, University of Tartu.
The linolenic acid test (LA-test) was used to assess the total antioxidative
activity (TAA) of L. fermentum ME-3, the reisolates of ME-3 from cheese;
HELLUS brand fermented milk products, probiotic capsule and juices
following the procedure as described in Paper III.
The TAA of samples was expressed as the percentageof inhibition of the
linolenic acid (LA) peroxidation per sample. The high numerical value (%)
indicates the high total antioxidative activity of the sample. Peroxidation of
LA-standard in isotonic saline served as control.

45
 11.2. Measurement of selected oxidative stress markers in humans
(Paper IV, V)

In urine, changes in the concentrations of the oxidative stress marker

8-isoprostanes (ng/ml) were assessed by a competitive enzyme-linked immu-
noassay (ELISA) (BIOXYTECH 8-Isoprostane Assay, Cat No 21019) as
described in Paper IV.
Blood serum was analysed for total antioxidative activity TAA, total
antioxidative status TAS and glutathione red-ox (GSSG/GSH). The TAA of the
serum was assessed by the linolenic acid test (LA-test) described in Paper IV.
The TAS of the serum was measured with a commercially available kit (TAS,
Randox Laboratories Ltd. Ardmore, UK) as described in Paper IV, water-
soluble vitamin E (Trolox) serving as the standard. This method is based on
inhibition of absorbance of the ferrylmyoglobin radicals of 2,2’-azinobis-
ethylbenzothiazoline 6-sulfonate (ABTS+) generated by activation of met-
myoglobin peroxidase with H2O2.
The cellular oxidative stress markers as total glutathione and oxidized
glutathione were measured using the method of Griffith (1980) as described in
Paper IV. Glutathione content was calculated on the basis of a standard curve
generated using a known concentration of glutathione. The amount of GSH
(µg/ml) was calculated as a difference between total glutathione (TGSH) and
GSSG (TGSH– GSSG = GSH). The glutathione red/ox ratio was expressed as
GSSG/GSH.

12. Detection of health effects of ME-3

12.1. Design of human volunteer trials (Papers IV, V)

12.1.1. Safety study with probiotic capsule.

A double-blind, 10-day randomized placebo controlled study (DBRP) was
carried out with 7 men and 15 women, mean age 42 years (range 24–64) to
evaluate the possible side effects of L. fermentum ME-3 consumption. Healthy
volunteers were selected according to their self-assessment and wish to
participate in the study and were randomly allocated by the computer. The
members of the study group (3 male and 8 female) took daily three probiotic
capsules containing 109CFU per capsule (daily dose 3x109 CFU); the placebo
group (4 males and 7 females) received identical capsules without any probiotic
strain. The participants of the trial were daily questioned about their general
welfare, intestinal function (gut gas production, stool frequency) and putative
adverse effects. Fecal samples were collected before and after the trial to assess
the survival of the probiotic strain in the human GI tract and the effect of the

46
strain on the fecal lactoflora. Urine samples were collected before and after the
trial to estimate the effect of ME-3 consumption on 8-isoprostanes
concentrations.

12.1.2. Functional efficacy trials

Two healthy volunteer (n 45) trials (an open placebo controlled study and a
DBRP study) were carried out to evaluate the functional efficacy of L.
fermentum ME-3 in the human body. The inclusion criteria included the wish to
participate, absence of known health problems and absence of medical
conditions requiring drug therapy, non-use of other yoghurts or special diets.
The subjects with a history of GI tract disease or food allergy, use of any
antimicrobial agent within the last month or use of any regular concomitant
medication were excluded. The blood samples (6 ml) from the antecubital vein,
and fecal and urine samples were collected before and at the end of all clinical
trials.
Open placebo controlled fermented goat milk trial. The study participants
were 5 men and 16 women, mean age 50 years (range 35–60). During three
weeks of the trial, the study group (3 men and 13 women) consumed daily 150
ml of fermented goat milk. The average daily dose of the probiotic Lacto-
bacillus strain was 3x1011CFU per person.
The control group (1 man and 4 women) consumed the same dose of fresh goat
milk.
Probiotic capsule trial. The DBRP study was carried out as follows. The
study group consisted of 15 men and 9 women, mean age 52 years (range 40–
60) who were allocated according to their wish to participate and divided
randomly into two groups by an independent person who used a computer
program. The members of the study group (8 male and 4 female) took three
probiotic containing capsules (2.6x108CFU per capsule) two times daily (daily
dose 1.5x109CFU) during three weeks. The placebo group (7 men and 5
women) received identical capsules without the probiotic strain. Participants
were recruited for endothelium testing by pulse wave analysis and by analysis
of the biomarkers (total of 17 parameters). Changes in the selected biomarkers
after 3-week consumption of ME-3 are described in the present study. This
study was carried out in cooperation with the Dept. of Biochemistry and the
centre of endothelial research of Tartu University Clinics.
The fecal samples of all participants for assessing changes in the fecal
lactoflora and the persistence of the ingested probiotic strain were collected
before and at the end of the trials. Several laboratory indices of blood and urine
were measured before and after the consumption of ME-3. Changes in human
body oxidative stress markers as total antioxidative activity (TAA); total
antioxidative status (TAS) and glutathione red-ox ratio (GSSG/GSH) from
blood serum and 8-isoprostanes in urine were reported.

47
 The participants of all trials, including the safety study gave informed
consent to the study protocols approved by the Ethics Committee of the
University of Tartu.

12.2. Microbiological analyses of feces

(Papers IV, V)

The fecal samples were collected at the beginning (day 0) and at the end of the
trials (day 21 in the case of the goat milk trial with healthy volunteers, day 10 in
the case of safety trial and day 21 in the case of the probiotic capsule efficacy
trial) and stored at –80°C unti analysed. The serial dilutions of the weighed
fecal samples were prepared with phosphate buffer (pH 7.2) and 0.05 ml of each
dilution was plated onto the MRS agar medium (Mikelsaar et al. 1972). The
plates were incubated at 37°C for 4 days microaerobically. Representative
colonies were selected on the basis of colony morphology, cell microscopy and
Gram staining. Detection level was 1x103 CFU/g feces.
The relative amount of lactobacilli colonizing the GI tract of the persons in
the study groups were expressed as a proportion of total count (%), using the
Bioquant statistical program (Mändar et al., 1992), which gives output data as
an absolute count (log10 CFU/g) for every micro-organism and the percentage of
different species in the total count.

12.3 Identification of lactobacilli

(Papers III, IV, V, present study)

Lactobacillus sp. isolates from different environments (fermented milk

products, cheese and feces) were identified according to their morphology and
the properties of the culture: carbohydrate fermentation patterns, gas formation
from glucose, hydrolysis of arginine, negative catalase activity (Lencner et al.,
1984; Kandler and Weiss, 1986).
Putative ME-3 isolates were typed according to arbitrarily primed poly-
merase chain reaction (AP-PCR). Genomic DNA was extracted from 24h old
cultures, cultivated on MRS agar microaerobically with the QIAamp DNA Mini
Kit 50 (QIAGEN GmbH., Hilden, Germany) according to the manufacturer`s
instructions. AP-PCR typing was done with two primers: ERIC1R
(5'-ATGTAAGCTCCT GGGGATTCAC-3') and ERIC2
(5'-AAGTAAGTGACTGGGGTGAGCG -3') (DNA Technology A/S, Aarhus,
Denmark). A 30 µl volume of the reaction mixture consisted of 10xPCR buffer
(Fermentas, Vilnius, Lithuania), 2.5 mM MgCl2 (Fermentas, Vilnius,
Lithuania), 200µM deoxynucleoside triphosphate mixture (dATP, dGTP, dTTP
and dCTP, Amersham Pharmacia Biotech, Freiburg, Germany) 0.60µg of each

48
primer and 2.5U Taq DNA Polymerase (Fermentas, Vilnius, Lithuania,) and 5
µl of extracted DNA according to Matsumiya et al. (2002). The PCR mixture
was subjected to thermal cycling 35 cycles of denaturation at 95°C for 1 min,
annealing at 35°C for 1 min, and extension at 74°C for 2 min, with a final
extension at 74°C for 5 min with the PTC-200 thermal cycler (Eppendorf AG,
Hamburg, Germany). The PCR products were separated by electrophoresis in a
horizontal 2% agarose gel containing 0.1µl/ml ethidium bromide in a Tris-acetic
acid–EDTA (TAE) buffer (40mM Tris, 20mM boric acid, 1mM EDTA, pH 8.3)
(Bio-Rad Laboratories, Hercules, USA) at constant voltage of 120V. A 1kb
ladder (GeneRuler, Fermentas, Vilnius, Lithuania) was used as the base pair
size marker. The banding patterns of the isolates were visualised with UV light
and compared with those of the L. fermentum ME-3 strain.

13. Statistical Analysis

The computer program Sigma Stat for Windows 2.0 (Jandel Coprporation,
USA) was applied. The count of the fecal lactoflora was compared by using
Student’s t-test and Mann-Whitney rank sum test. Changes in oxidative stress
markers of blood sera (TAA, TAS and glutathione red-ox ratio) and urine
(8-isoprostanes) were evaluated by Student’s t-test, paired t-test and the Mann-
Whitney rank sum test. The choice of the tests was made automatically
according to the distribution of the data. Both microbial and biochemical
markers are presented as mean and standard deviation.
One-way ANOVA test was performed to compare the effect of different
formulations on TAA, TAS and the fecal lactoflora.
Differences were considered statistically significant if the value was
p <0.05.

49
 RESULTS AND DISCUSSION

14. Basic characterization of L. fermentum ME-3

L. fermentum ME-3 was identified on the species level according to the API
CHL 50 System kit (ID% 99.6, one test contra). The strain ferments the
following sugars: ribose, galactose, D-glucose, D-fructose, D-mannose,
esculine, maltose, lactose, melibiose, saccharose, D-raffinose, D-tagatose and
gluconate. The identification of ME-3 on the species level was confirmed by
ITS-PCR (Fig. 3, a) and by AP-PCR (Fig. 3, b).

(a) 1 2 3 4 5 6 7 8 (b) 1 2 3 4 5 6 7

Figure 3. Identification of ME-3 on the species level according to (a) ITS-PCR:

Starting from left (1) fecal isolate; (2) L. fermentum ME-3; (3) fecal isolate; L.
fermentum ATCC 14931; (4) isolate from fermented milk; (5) L. reuteri DSM 20016;
fecal isolate, (8) DNA 100 kb Ladder Plus and (b) according to AP-PCR on the species
level Starting from left (1) DNA 1 kb Ladder; (2) L. buchneri ATCC 4005; (3) L. brevis
ATCC 14869; (4) L. reuteri DSM 20016; (5) L. fermentum ATCC 14931; (6) L.
fermentum ME-3

The production of organic acids (mg/ml) in different growth environments was

estimated by gas chromatography. L. fermentum ME-3 has heterofermentative
metabolism, the fermentation profile being dependent on the environmental
conditions (Table 6).

50
Table 6. Concentration of lactic acid, acetic acid, succinic acid and ethanol (mg/ml) in
microaerobic and anaerobic environments during 24 and 48 h
Growth Lactic acid Acetic acid Succinic acid Ethanol
environment 24h 48h 24h 48h 24h 48h 24h 48h
Microaerobic 10.6 11.1 0.8 0.9 1.8 1.9 9.8 7.5
Anaerobic 8.2 8.8 1.0 1.0 0.5 0.9 7.0 33.3

Besides lactic acid and acetic acid, succinic acid was produced micro-
aerobically, however, an additional large amount of ethanol was produced
additionally in anaerobic conditions.
Antibiotic resistance. By disk diffusion test and an E-test, L. fermentum
ME-3 was found to have natural resistance to metronidazole, ofloxacin,
aztreonam, cefoxitin and trimethoprim-sulfmethoxazole.
Rapid identification of L. fermentum ME-3. Gas production from glucose is
characteristic of heterofermentative lactobacilli. L. fermentum ME-3 is a
vigorous producer of gas from glucose. Additionally, two more parameters were
selected for the rapid identification of L. fermentum ME-3. The feature of
distinguishing L. fermentum best from the other species of the OHELgroup is
lysozyme production (Table 7). Slow growth at 15°C distinguishes ME-3 from
most of the other L. fermentum strains.
Confirmation of rapid identification of L. fermentum ME-3 by molecular
typing.
The identification of putative ME-3 and ME-3-like isolates from feces by
rapid methods was confirmed by AP-PCR (Fig.3). Out of the 12 fecal isolates
banding patterns none had similar to those of ME-3.

Table 7. The biochemical properties of ME-3

Biochemical L. fermentum L. fermentum *Fecal Fecal Fecal Fecal Fecal
tests ME-3 ATCC 14931 isolate isolate isolate isolate isolate
1 2 3 4 5
Growth at + – – – – – –
15°C
Lysozyme + + + + + + +
production
* Fecal isolates obtained from the probiotic efficacy trial with healthy human

51
 1 2 3 4 5 6 7 8 9 10 11 12 13 14

Figure 4. Distinguishing of L. fermentum ME-3 from the other L. fermentum strains

isolated from feces. Starting from left (1) DNA 1 kb Ladder; (2) L. fermentum ME-3;
(3-14) fecal lactobacillus isolates obtained from the probiotic capsule efficacy trial
participants

15. Technological properties of L. fermentum

ME-3 (Present study)
15.1. Acid tolerance

In food products, survival of a probiotic strain may be influenced by acidic

stress caused by accumulation of the metabolic end products of starters or by
strain itself.
During 24h, L. fermentum ME-3 tolerated well both lactic acid and acetic
acid at pH 4.5 and pH 4. In more acidic conditions (pH<3.5), a decrease in cell
count from 1 log (lactic acid) to 1.5 log (acetic acid) was seen after 6 hours.
Thus, ME-3 was able to retain its viable count for at least 6 h in an environment
with relatively low pH values caused by lactic or acetic acid.

15.2. Dependence of heat resistance on pH

The heat resistance of L. fermentum ME-3 was tested to evaluate the potential
use of the strain in probiotic juices, where technology foresees pasteurisation
and addition of the strain to hot juice. The ability of ME-3 to resist high
temperatures was tested in skim milk at pH 7.0 and in different juices at pH
3.05…3.85.

52
 12
(a) (b)
Orange juice Multivitamin nectar Tropical carrot drink
10

8
CFU log10

0
0 5 10 before after before after before after
Time (minutes) heating heating heating heating heating heating

73°C, saline Skim milk Frozen culture 1 hour

85°C, saline Skim milk 2 hours 3 hours
95°C, saline Skim milk

Figure 5. Heat resistance of L. fermentum ME-3 (a) in skim milk and in isotonic saline
at 73°C, 85°C and 95°C and (b) frozen ME-3 culture in orange juice, in multivitamin
nectar and in tropical carrot drink. The time intervals between different stages of
melting of the lactobacillus culture were: frozen culture, 1h, 2h and 3h.

In a neutral environment the decline of viable cell count of ME-3 was fast at all
temperatures. The performance of the strain was somewhat more stable in saline
at all tested temperatures in comparison with skim milk (Fig. 5). The decrease
in viability in saline was the highest at 95°C – approximately 7 logs. In skim
milk ME-3 appeared to be more susceptible to the lowest tested temperature
(73°C), where the decrease in viability was nearly for 8 log CFU. After 10 min
from the beginning of the heat treatment the strain was not detectable at any
temperature.
For the estimation of the heat resistance of ME-3 in an acidic environment,
a suspension of L. fermentum ME-3, precultured on the MRS agar medium, was
added to three different juices. The strain was found to survive the co-action of
acidity and high temperature for 4 minutes with less than 1 log CFU of loss in
viable cell count when an active Lactobacillus culture was used (Table 8).

53
Table 8. Heat resistance of ME-3 suspension by (log10 CFU/ml) in different juices
Juice pH Before heat After heat Decrease
treatment treatment
Orange juice 3.8 6.9 6.1 0.8
Multivitamin nectar 3.5 6.9 5.9 1.0
Tropical drink with carrot 3.0 6.9 6.1 0.8

The heat resistance of the frozen (–80°C) L. fermentum ME-3 skim milk
suspension (5x109 CFU/ml) after melting and addition to the juice was found to
be dependent on the time interval between the addition of Lactobacillus to juice
and heat treatment (Fig. 5) but not on a particular juice or juice drink. The
smallest loss in the viable cell count of L. fermentum ME-3 (decrease from 1 log
to1.5 log) was noted when the Lactobacillus and juice were heat-treated
immediately after mixing and in the case of orange juice, also 3 hours from
preparation. When L. fermentum ME-3 containing juice was heat-treated 1h and
2h after mixing, the strain did not survive in any of the 3 tested variants.

15.3. Growth in cell suspensions

Several environmental factors limit the survival and growth of a probiotic

additive in cheese, after free carbohydrates are metabolized during the first days
of cheese ripening. Cell lysis products of starter microbes can be used as a
potential source of nutrients and energy.
There was approximately a 2 log and 2.5 log increase in the cell count of L.
fermentum ME-3 during 7 days in thermally inactivated cultures of non-starter
lactobacilli or starter lactococci, accordingly (Fig. 7).
In the live suspension of a starter culture and L. fermentum ME-3, the count
of the probiotic ranged from initial 4 log to 6.5 logs. A rapid decrease in the
count of starter lactococci was registered at the same time (Fig. 6, a). In the
mixture containing live non-starter lactobacilli and L. fermentum ME-3, a slight
increase in the probiotic count and a sharp, 3 log decrease in the count of the
non-starter lactobacillus strain occurred during the first 5 days of the test,
followed by an increase in the count of the non-starter lactobacillus strain (Fig.
6, b).

54
 ME-3 + inactivated starter
9 (a) viable starter (+ME-3)
8 ME-3 (+ viable starter)
control (ME-3)
7
6
5
log10

4
3
2
1
0
0 1 2 5 6 7
Time (hours)

(b) ME-3 + inactivated NSLAB

10 viable NSLAB (+ME-3)
9 ME-3+ (viable NSLAB)
8 control (ME-3)
7
6
log10

5
4
3
2
1
0
0 1 2 5 6 7
Time (days)

Figure 6. Growth of L. fermentum ME-3 in limited nutritional conditions on the cell

suspension of non-starter lactobacillus (NSLAB) (a) and lactococci (b)

15.5. Antagonistic activity

15.5.1. Antagonistic activity between L. fermentum ME-3 and

starter cultures

The compatibility of the probiotic strain to starter organisms should be assessed

in order to avoid undesirable changes in the composition of the product`s
microflora and unwanted changes in the sensory properties of the product.

55
 L. fermentum ME-3 revealed low inhibitory activity against the tested
starters (Table 9) on an agar medium. None of the tested starters inhibited the
growth of ME-3.

Table 9. Antagonistic activity of L. fermentum ME-3 against different starters on agar,

expressed by the values of the inhibition zone and the survival of starters and L.
fermentum ME-3 in milk samples
Starter growth LAB** count
Tested starter inhibition (mm) (log10/ml)
by ME-3 ME-3 Starter

Cheese starters: Probat 505 2.2±2.5* 8 7.7

CHN 11, CHN 19, 10.6±1.2 7.7 7.1
Flora Danica
Yoghurt starter: Jo-Mix VK 1-30 2.1±1.0 8.7 8.0
* Width of the growth inhibition zone on agar (Hütt et al., 2005): 0…13 mm – low; 13… 25 mm
– intermediate; >25 mm – high;
**Tested combination in milk: Probat 505+ME-3; CHN 11, CHN 19, Flora Danica+ME-3; Jo-
Mix VK 1-30+ME-3

No antagonistic activity was detected between the tested starters and L. fer-
mentum ME-3 in milk, either. After 48h, both cultures were detectable with
equally high numbers in the growth medium (Table 9).

15.5.2. Antagonistic activity between L. fermentum ME-3 and

non-starter lactobacilli

Interactions between a probiotic additive and the non-starter lactoflora, present

in cheese milk can affect the survival of the probiotic strain and the sensory
properties of the final product.
The antagonistic activity of L. fermentum ME-3 against non-starter lacto-
bacilli of milk and cheese origin and vice versa was measured on MRS agar
using the streak line method. A total of 17 strains of non-starter lactobacilli
(6 strains of the OHEL group; 7 strains of the FHEL group and 3 strains of the
OHOL group), isolated from cheese, cheese whey; cow or goat milk, were
tested.
The tested non-starter lactobacilli strains revealed low (0–3mm) anta-
gonistic activity against L. fermentum ME-3. The probiotic strain showed low
activity against the strains from the OHOL (4–9mm) and OHEL (9–13mm).
The activity of ME-3 was against the FHEL group was moderate (14–18mm),
though the inhibitory effect was relatively high in comparison with the
antagonistic activity of non-starter lactobacilli against ME-3.

56
 15.6. Survival in milk products

15.6.1. Probiotic cheese

The inoculation rate of L. fermentum ME-3 was 108 CFU per gram of cheese.
ME-3 withstood the manufacturing of the tested cheese varieties and storage at
low temperatures (Fig. 9). The viable cell count of the probiotic was found to be
relatively high at the end of storage (1x107 to 5x107 CFU per g of cheese) in
both cheeses. However, a decline starting from the 3rd week from preparation
was followed by a rise of the viable count to the initial level (5x107 CFU/g)
afterwards.
In Atleet cheese, another decline in the viable count was registered towards
the end of the shelf life between day 120 and 240 from preparation.

15.6.2. Fermented goat milk

The fermented goat milk was experimentally prepared for the trial with healthy
volunteers in order to establish the health effects and safety of ME-3
consumption. The study group was supplied with the fresh product once a week.
The viable cell count of ME-3 in fermented goat milk were determined at
the end of fermentation, 24h, 48h, 72h and 7 days from preparation, when the
product was stored at 4oC. The cell count varied from one preparation of the
fermented goat milk to another 1x109...5x109 CFU/ml. The viable count of
ME-3 in fermented goat milk were found to remain stable during 7 days of
storage at 4°C.

15.6.3. Fermented milk products

The viable cell count of ME-3 in the HELLUS brand yoghurt, kefir, quark, sour
cream were analysed at different times of the product`s shelf life (kefir and
quark, 15 days; yoghurt and sour cream, 20 days). The identification of ME-3
isolates and the genetic stability in fermented milk products was verified by AP-
PCR (Figure 10). Besides, ME-3 is easily differentiated from the other species
of the OHEL fermentation group.
Most products contained about 6 log of ME-3, except for yoghurt for which
the cell count were 7 log. The survival of ME-3 in the fermented milk products
of the HELLUS brand was stable in all tested products throughout shelf life. No
decline was detected in viable count.
The identification of putative ME-3 isolates from different environments on
the strain level was verified by AP-PCR (Fig. 7).

57
 1 2 3 4 5 6 7 8

Figure 7. Molecular finger-prints of

L. fermentum ME-3 isolates from
different products and carriers by
AP-PCR. Starting from left
(1) L. fermentum ME-3;
ME-3-like profile in the following
products: (2) Yoghurt, (3) Quark
dessert, (4) Sour cream; (5) Kefir;
(6) Cheese Atleet; (7) Probiotic
capsule; (8) M – DNA Alu I Ladder.

15.7. Survival and growth of L. fermentum ME-3 in different juices

If incorporated into juices, the probiotic strain has to survive till the end of shelf
life without causing loss of quality of products.
Multiplication of L. fermentum ME-3 was established in two of the three
tested juices, orange juice and multivitamin nectar, where the viable cell count
rose during 4-log (multivitamin nectar) and 5.5-log (orange juice) 48h. In
tropical drink with carrot, a 2-log decline in the initial count of L. fermentum
ME-3 was detected.
During a 1-month period, the survival of L. fermentum ME-3 at 4°C was
better for orange juice and multivitamin nectar, where loss in viability was
lower (approximately 0.5 log) than in the tropical drink with carrot where loss
in viable cells was 2.3 log.

16. Functional properties of L. fermentum ME-3

(Present study)

16.1. Stability in GI conditions

The ability of L. fermentum ME-3 to survive the passage through the upper
parts of the GI tract in the presence of acid and bile stress was tested in vitro in
two different manners.
The bile tolerance and acid tolerance of L. fermentum ME-3 were measured
separately in the MRS broth. No growth was detected in the presence of HCl in

58
the environment, though ME-3 tolerated well as low as pH 2.5 without loss in
viability during 24h. At pH2.0 the strain survived well for 6 hours.
L. fermentum ME-3 was found to tolerate all the tested bile concentrations
without loss in viable count. The acid and bile tolerance of the active overnight
culture and freeze-dried capsulated culture of L. fermentum ME-3 (Fig. 8) was
tested on simulated gastric digestion model, applying low pH and pepsin
followed by bile and pancreatin.
Different behaviours of L. fermentum ME-3 were noted on the comparison
of two tests. The viable count of ME-3 were relatively stable at least for 6 h at
low pH and even longer at different bile concentrations in MRS, a decrease of
0.5-log to 1.5-log of viable count of ME-3 was noticed during first 1.5 h from
the beginning at the presence pH2.5 and pepsin.

5
log10

3 control
ME-3
2
capsule
1 capsule+skim milk
0
0 1,5 3 4,5 6
Time (hours)
Figure 8. Survival of L. fermentum ME-3 on the basis of the in vitro gastric digestion
model at pH2.5 in an HCl and pepsin environment (solid line), dashed line shows
following treatment with 0.15% bile and 0.1% pancreatin.

At the end of the experiment, freeze-dried ME-3 with skim milk proved to be
the most stable and was found to resist better the stresses according to the the in
vitro gastro-intestinal model compared with the active overnight ME-3 culture
and the capsulated probiotic without skim milk.

59
 16.2. Stability of probiotic properties of
L. fermentum ME-3 in products

Probiotic strain should maintain its probiotic characteristics throughout the

manufacturing and storage of the product. The stability of the antioxidative and
antimicrobial activity of L. fermentum ME-3 was measured in isolates from
fermented milk products of the HELLUS brand, cheese, capsule and juices.

16.2.1. Probiotic cheese

Antioxidativity. The TAA values of ME-3 incorporated into the smear-ripened

cheese Pikantne decreased during the first days of maturation, dropping to the
level of 15% on the 12th day (Fig. 9, a). Further, the reisolates of the strain
showed a tendency of increase in TAA throughout the trial: 17% on the 24th day
and 20% on the 38th day. However, at the end of the experiment (66th day) the
TAA of the cells of the probiotic Lactobacillus attained the same value as the
original culture of ME-3.
In the Atleet cheese, the TAA value of the ME-3 isolates started to decrease
from the first week of preparation, reached the lowest level by the 40th day and
remained stable towards the end of the trial (Fig. 9, b).
Antagonistic activity. The stability of antimicrobial activity in cheese was
measured in the probiotic Pikantne cheese throughout its ripening and shelf life
(Paper III). In comparison with the original culture of ME-3, the reisolates of
the probiotic additive from Pikantne cheese revealed some decrease in
antagonistic activity. L. fermentum ME-3 showed some decrease in antagonistic
activity against both S. aureus and Gram-negative pathogens during the
ripening period and storage, except for the cystitis causing strains E. coli ATCC
700414 and S. sonnei ATCC 25931 which were equally well suppressed by the
L. fermentum ME-3 reisolate from the cheese, being the same level on the 66th
day from the preparation of the cheese as in the case of the original pure culture
of ME-3.

60
 9 (a) 30
8
25
7
6 20
CFU log10

TAA (%)
5
15
4
3 10
2
5
1
0 0
01 10 2 24 3 4 3 85 56
4 676
T im e (d a ys)

(b)
9 30
8
25
7
6 20
CFU log10

TAA (%)
5
15
4
3 10
2
5
1
0 0
01 12
2 20
3 29
4 40
5 6 63
7 8 118
9 217 11
10 242
T im e (days)

Viable count antioxidativity

Figure 9. Survival and stability of antioxidativity of ME-3 in (a) Pikantne and (b) Atleet
cheese during maturation and storage.

16.2.2. Fermented milk products

Antioxidative activity. The TAA of the original ME-3 culture is 26.0±6.1%. The
ME-3 isolates from four tested product types displayed different TAA values:
the isolates from sour cream (25.0±13.2) being the best, followed by quark
dessert (20.0±5.7), yoghurt (16.0±3.2) and kefir (12.0±2.6). Inside the yoghurts
and quark desserts, differences in TAA were noted in flavour additives.
ME-3 isolates from wild strawberry (21.0±4.6) and raspberry (20.5±7.7)
jam-flavoured quark desserts showed the highest TAA values, followed by the
tropical fruit (15.0±7.0) flavoured variety. Inside the yoghurts, the ME-3
isolates from the wild berry- (18.0±1.4) and pasha-flavoured (16.0±1.4)
varieties displayed the highest antioxidativity, followed by the probiotic isolates
from the rhubarb-oat (13.0±4.2) and cranberry-muesly (12.0±3.8) flavoured
yoghurts (Fig. 10).

61
 50

Yoghurt quak dessert

40

* * ** * ** * *
TAA (%)

30

20

10

0
re f ir am a at es es l y rr y rr y ic al
lt u ke re sh - o er r i be be
u c p a b a rb b y -m
u
w p
3c ur l d rr str
a
ra
s p
tro
E- so
Rh
u
w i nbe d
M l
cr
a
wi

Figure 10. Stability of the TAA of ME-3 in fermented milk products. Significant
difference from the TAA of the original ME-3 culture: *p≤0.001; **p<0.05 (Mann-
Whitney rank sum test)

Antagonistic activity. The L. fermentum ME-3 isolates from quark dessert and
yoghurt revealed somewhat lowered antagonistic activity on the agar medium
against Staphylococcus aureus. In the sour cream isolates the decrease was
significant (p<0.05) in comparison with the base values of the original ME-3
culture (Fig. 11).

30
** ** * * **

25
Pathogen inhibition zone (mm)

20

15

10

0
M E - 3 c u ltu r e y o gh u rt q ua rk d e ssert k e fir so u r crea m

S . a u re u s B 4 6 G r a m - n e g a tive p a th o g e n s

Figure 11. Antagonistic activity of the ME-3 isolates from the fermented milk products
of the HELLUS brand. Significant difference from the antagonistic activity of the
original ME-3 culture: *p≤0.001; **p<0.01 (Mann-Whitney rank sum test)

62
ME-3 isolates from the yoghurts, flavoured with rhubarb-oat and wild-berries,
expressed significantly reduced (p≤0.001) activity against both S. aureus and
Gram-negative bacteria. In quark desserts, significantly lowered antimicrobial
activity was detected in the ME-3 isolates from the wild strawberry- and
raspberry-jam flavoured varieties (p<0.01).

16.2.3. Capsule and juices

The TAA value of the L. fermentum ME-3 capsule isolates after reactivation
had remained on the same level as that of the original Lactobacillus culture
(26±6.1 of ME-3; 21±10.6 of capsule isolate).
The antagonistic activity of the L. fermentum ME-3 isolates from capsule
after reactivation was found to be significantly higher (p<0.001) than that of the
ME-3 culture before freeze-drying and capsulation (Fig. 12) against all tested
pathogens on agar medium.
The TAA of the L. fermentum ME-3 isolates from all the tested juices was
measured after 1 month. All isolates showed relatively high TAA values
(21...31), those of the isolates from orange juice being the highest.

40

35
pathogen inhibition zone (mm)

30

25

20

15

10

0
ME-3 culture capsule

S. aureus B46 E.coli K12

E. coli ATCC 700336 E. coli ATCC 700414
S. Typhimurium S. Enteritidis ATCC 13076
S. sonnei ATCC 25931

Figure 12. Antagonistic activity of the L. fermentum ME-3 isolates from capsule

63
 17. Health effects of ME-3: human volunteer trials
(Papers I, IV, V)

17.1. Safety trial with probiotic capsule

The duration of this trial was 10 days. The daily dose was 3 capsules and the
dose of the probiotic was 109 CFU per capsule. No dropouts from the safety
trial were registered. In the study group no differences in the general welfare
and intestinal function in comparison to the placebo group were found, nor were
adverse effects seen during the trial. After the 10-day consumption of ME-3, the
total count of the fecal Lactobacillus species increased significantly in the study
group (Fig. 13). However, no significant changes in L. fermentum count were
detected during the trial and the probiotic strain was not recovered from feces
after consumption (Fig. 15).
In the study group the consumption of the probiotic capsule lowered the 8-
isoprostanes concentrations in urine (Table 11). No change in the 8-isoprostanes
level in the placebo group was detected in comparison with the indices before
treatment.

12 stu d y gro u p c o ntr ol stu d y gr o u p pla c eb o stu d y gro u p p la c eb o

* *
10
* * * * *

8
CFU log10

G oa t m ilk tria l P rob iotic ca p s ule sa fety tr ia l P rob iotic ca ps u le e ffica y tr ia l

D ay 0 A fter treatm en t

Figure 13. Box plots of changes in fecal lactobacilli count during the human volunteer
trials. Data are median count (—) and distribution (box display 25th–75th quartile area,
bars 10th–90th percentage area. Significant difference from the pre-treatment values:
*p<0.05; **p<0.01 (Paired t-test).

64
 18.2. Functional efficacy trials with fermented goat milk and
capsules

Changes in total LAB count. The consumption of ME-3 fermented milk and
ME-3 capsule significantly increased the total count of lactobacilli in feces in
comparison with to the initial levels. On the contrary, in the group of volunteers
consuming the non-fermented goat-milk the count in total fecal LAB even
decreased significantly during the 3-week trial (Fig. 13). The increase of total
fecal LAB observed for the study group of the capsule trial was not different
from that for the placebo group.
Prolonged consumption of the probiotic with different vectors altered also
the proportion of fecal lactobacilli species: in both trials the proportion of fecal
lactobacilli species decreased with the appearance of new species (Fig. 15).
Recovery of the probiotic strain. In the goat milk group, L. fermentum as a
species appeared in fecal samples of all individuals after consumption of goat
milk fermented with L. fermentum ME-3 (p<0.001) while in 12 persons it was
not found before the trial. AP-PCR confirmed the recovery of ME-3 in the feces
of all members of the study group (Fig. 14). In the probiotic capsule trial, the
ME-3 strain was not detectable among the L. fermentum isolates either by using
bacteriological methods or AP-PCR.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18

Figure 14. Confirmation of the survival of L. fermentum ME-3 in the GI tract of the
persons of the fermented goat milk group. M – DNA 1kb Ladder, Line 2 – ME-3, Line
3…18 – ME-3 like profile in the feces of different persons.

However, in different trials, L. fermentum did not predominate over the other
Lactobacillus species in any of the participants (Fig. 15) at the end of treatment.
Although more persons were colonized and there was a tendency for increase in
L. fermentum count, the proportion of the species in total lactobacilli count was
not increased.

65
 Antioxidative effect of ME-3. The consumption of ME-3 in both formu-
lations had a positive effect on the blood oxidative stress markers TAA and
TAS (Table 10).
Consumption of goat milk and fermented goat milk enhanced TAA and
TAS in both the study and control groups. There was an additional increase in
both indices in the fermented goat milk group, which was not statistically
significant. A significant increase in TAA and TAS values occurred also during
the consumption of the probiotic strain in the capsulated form. No change was
detected in the placebo group.
Compared with the baseline values, the consumption of ME-3 in goat milk
reduced urine 8-isoprostanes concentrations but no changes were seen in the
probiotic capsule trial (Table 10).
The decrease in the glutathione red-ox ratio was significant in both the
study and control groups in the goat milk trial. The goat milk fermented with L.
fermentum ME-3 had no statistically significant additional effect. When the
probiotic was consumed in the capsulated form, no significant decrease was
noted in the glutathione red-ox ratio (Table 10).

66
 12% 0% 18%
12%

L. plantarum L. casei
L. salivarius 16% L. plantarum L. casei 21% 22%
L. buchneri 14% 18% 20%
L. acidophilus 10% L. brevis 22%
L. acidophilus 14%

L. brevis Therm obacterium sp

20% 21%. L. fermentum Thermobacterium sp.
10% 13% 29% 28%
L. ferm entum
21% 18
28%
%

a)a) b)
13%

29% 0%
24%
0%
L. acidophilus L.salivarius
5% 12% L. casei
0% 21%
29% L. buchneri 3%
L. plantarum
L. acidophilus 17%
21% L. plantarum
L. casei 14%
9% 27%
L. ferm entum
4% Therm obacterium sp.
L. brevis 22% 24%
L. brevis 2% 12%
14%
L. buchneri 22% Therm obacterium sp.
43% L. ferm entum
17%

d)
c)
c) 9% 20%

0% 15%
1% 15%

L. fermentum L. buchneri 5%
11% L. acidophilus
L. plantarum L. acidophilus 13%
12% 11%
8% 34%
L. plantarum L. casei
15%
L. fermentum L. casei 13% 19% 14%
11% 11%

Thermobacterium sp.
14%
Thermobcterium sp L. brevis
L. brevis 9% 55% 15%
L. salivarius 17%
14%
16%

e) f) 19%
e) 33%

Figure 15. Changes in the proportions of lactobacilli species in the fecal lactoflora
during volunteer trials: ME-3 group, goat milk trial (a), control group, goat milk trial
(b), ME-3 group, safety trial (c) placebo group, safety trial (d), probiotic capsule trial (e)
and placebo group, capsule trial (f).
Pie area: values before treatment; pointed with arrows: values after treatment; in arrow
callouts: new species additionally isolated after treatment.

67
Table 10. The effect of L. fermentum ME-3, administered with different carriers on systemic oxidative stress markers (mean ± SD).
8-isoprostanes p Glutathione redox ratio p TAS (mmol/l) p TAA (%) P
Groups (ng/ml) in urine value (GSSG/GSH) in sera value in sera value in sera value
Before After Before After Before After Before After
Goat milk trial, 5.5±0.4 5.0±0.5 <0.01 0.15±0.01 0.11±0.04 <0.01 0.82±0.14 1.14±0.08 <0.005 38.0±3.0 45.0±3.0 0.01
ME-3
Control 4.5±0.8 5.2±0.3 NS 0.14±0.03 0.11±0.02 <0.01 0.86±0.11 1.06±0.04 <0.01 41.0±6.0 45.0±5.0 0.01
Capsule trial, 3.6±1.3 3.0±1.6 NS 0.07±0.02 0.06±0.01 NS 0.82±0.08 0.86±0.07 <0.05 37.8±3.0 39.5±3.3 0.005
ME-3
Placebo 3.3±2.8 3.4±1.9 NS 0.06±0.01 0.06±0.01 NS 0.78±0.18 0.79±0.069 NS 38.9±2.5 38.9±3.3 NS
NS – no statistical difference

68
 GENERAL DISCUSSION
A probiotic strain meets different challenges on its way to the target site in the
human body. Therefore, a putative probiotic should be tested beforehand in
vitro, only after which its suitability could be confirmed in human trials.
The purpose of the present study was to evaluate the effect of different
environments on the functional properties and viability of L. fermentum ME-3.
The safety and efficacy of ME-3 consumption during healthy volunteer trials
were assessed.
L. fermentum ME-3 possesses two main probiotic properties: antimicrobial
activity and antioxidative activity. The stability of these functional properties of
L. fermentum ME-3 in the course of technological process was evaluated using
elaborated simple tests by us.
There are three main technological aspects concerning suitability of
probiotic strains for incorporation into food products:
1. Probiotic strains must be suitable for incorporation into various foods
without a negative impact on activity product`s sensory properties.
2. Probiotic strains must be suitable for manufacturing under industrial
conditions.
3. Probiotic strains must retain viability and stable functionality during
storage in food products.

The main functional aspects concerning a probiotic in host are as follows:

1. Probiotic strain has to survive in the GI tract
2. Probiotic strain has to express some measurable human health promoting
properties.

18. Technological properties of L. fermentum ME-3

18.1. Suitability of ME-3 for various delivery vehicles

18.1.1. Impact on s the sensory properties of the product

Lactobacilli using the heterofermentative pathway can produce unwanted by-

products like acetic acid or carbon dioxide, which may negatively affect the
texture or flavour and aroma formation of a certain food product. L. fermentum
ME-3 belongs to the obligately heterofermentative group of lactobacilli and is a
vigourous producer of carbon dioxide from glucose as well as a relatively
efficient producer of lactic acid and acetic acid. Therefore the suitability of ME-
3 as a probiotic additive for incorporation into different fermented milk
products was tested in the present study. ME-3 was incorporated during pilot
plant experiments into two cheese varieties, and into yoghurt, kefir, quark

69
dessert and sour cream. No negative impact on the organoleptic quality of any
product was detected. All products were tested by food experts and proved to be
commercially acceptable, while no off-flavour or -texture, caused by ME-3
metabolism was noticed.

18.1.2. Survival in products

Probiotic products are carriers of strain’s which should enter the human GI tract
alive to express their specific functional properties.
The survival of the strain was satisfactory in all above products, although
slight differences were detected between different products. The viability of
ME-3 in fermented milk products of the HELLUS brand was found to be stable
throughout the entire product`s shelf life. In yoghurts the cell count higher than
7 logs were detected already one week after production, although the initial
amount of inoculation has been the lowest. Obviously, the strain multiplied
during the manufacturing and/or storage of yoghurt. All other products
contained about 6 logs of ME-3. Somewhat lower count compared with the
initial inoculation rate was detected only for quark desserts. Apparently, a quark
dessert environment does not support the survival of the strain.
The survival of L. fermentum ME-3 in both cheese varieties was similar
(107 to 5x107 CFU/g of cheese). The property to tolerate lower temperatures
distinguishes ME-3 from some other L. fermentum strains. This made possible
for the strain to survive at a relatively low temperature of cheese ripening and
even lower temperature of storage (6o–8°C) for long time (maximum 8 months).
The behaviour of L. fermentum ME-3 was similar in both cheese varieties in
spite of differences in technological process, maturation time as well as in the
duration of shelf life: a decline in viable count was noted at almost the same
time from cheese preparation, followed by a rise in cell count afterwards. This
phenomenon can be explained by the reorganization of the metabolism on other
potential growth substrates by ME-3 after the depletion of milk sugars shortly
after preparation of cheese. Non-starter lactobacilli can use sugars released via
the enzymatic hydrolysis of casein, lactate, free amino acids or a variety of
other compounds present in the cheese environment as the alternative energy
source for carbohydrates (Laht, 2003). Like most heterofermenters, L.
fermentum ME-3 utilizes arginine and is probably able to use it as an additional
energy source. Besides, according to our data, ME-3 is able to survive by using
the microbial cell lysis products of cheese starters. ME-3 also expresses weak
proteolytic activity, however, this issue needs further investigation.
Additionally, the potential use of ME-3 for the production of probiotic
juices was tested in laboratory scale studies. During 1-month refrigerated
storage of probiotic juice, ME-3 retained its viability in orange juice and in
multivitamin nectar; at the same time a loss of viability was detected in the

70
tropical drink with carrot. The cause for this could be the higher acidity and the
chemical composition (the juice contained 9% mixture of orange, lemon and
carrot juices) of the latter product in comparison with 100% of orange juice or
multinectar drink, a mixture containing six different juices.
Although good survival during refrigerated storage in orange juice and
multivitamin nectar is a favourable feature, multiplication of L. fermentum ME-
3 in the same juices may potentially negatively affect the quality parameters like
the taste and appearance of the product in the future, when the consumer stores
probiotic juice at room temperature.

18.1.3. Stress responses

Any extreme change in environmental conditions causes stress for an organism.

Probiotic lactobacilli can be affected by various stresses due to industrial
process and following storage in the final product. Therefore, the resistance of
L. fermentum ME-3 to different possible stresses: heat and cold shock, acidic
environment and nutrient limitation, was assessed.
In food products the survival of a probiotic strain might be influenced by
the acidic stress caused by a decrease in pH due to fermentation of sugars by
starters or by the strain itself. Typical acidity of cultured milk is between pH3.5
and 4.5, containing 0.5–1.5% (w/v) lactic acid. The H+ ions can cause a
decrease in the viability of cells. Acidity has negative impact on bacterial
physiology by altering enzymatic activities, leading to dissipation of the proton
motive force and expression of acid response proteins (Champomier-Vergès et
al., 2002). Lactobacilli have an intrinsic property to tolerate relatively lower pH
values in comparison with some other microbes especially pathogens.
Nutrient limitation. In cheese, non-starter lactobacilli are submitted to
growth-restricting conditions, like carbon source starvation after free
carbohydrates are metabolized during the first days of cheese ripening. For non-
starter bacteria, probiotic additives included other energy sources like microbial
metabolites and microbial cell lysis products are available. It is proposed that
the rate of starter cultures cell lysis might influence the growth of the secondary
microflora as well as of probiotic additives during ripening of cheese (Rapposch
et al., 1999). In present study the ability of ME-3 to grow in microbial cell
suspensions as a potential source of nutrients and energy was tested on an
intact, thermally inactivated cheese starter and on non-starter lactobacilli
suspensions. We found that ME-3 was able to survive and grow in limited
nutritional conditions in vitro and to use microbial cell lysis products as an
energy source at least during 7 days. This could be one explanation for the
multiplication of the strain in cheese during its ripening and storage.
Co-action of stresses. A probiotic strain can be submitted to several stresses
at the same time. The resistance of ME-3 to the co-action of sharp temperature
changes and low pH was tested in 3 juice drinks (pH ranging from 3.05 to 3.85)

71
during different time intervals. Differences in pH or in the composition of juices
did not seem to affect the resistance of the strain to sharp temperature variation.
The active ME-3 strain tolerated well the co-action of acidity and high
temperature for 4 minutes at 96°C with a loss in viable cell count less than
1 log. Surprisingly, the heat treatment at the same temperature and even lower
values (73°C and 85°C) was less tolerated in neutral conditions at pH 7.0 in
skim milk and in physiological saline, where the decrease in the viable cell
count was sharper (from 5 log to 8 log) compared with the juices. In both above
described tests the culture submitted to heat treatment was in the active form.
Probably, the chemical composition of these juices enhanced the stress
tolerance of ME-3 more than reconstituted skim milk or saline.
The survival of the strain was tested in even more harsh conditions: frozen
skim milk suspension was let melt at room temperature and was then added to
the juices at different times and kept at 96°C during 15 seconds. The test was to
clarify the optimal time interval for the maximum possible survival of the
frozen culture during the production of probiotic juice, when added to
pasteurized juice. The lowest loss in the viable cell count (decrease from 1 log
to1.5 log) of ME-3, comparable that in the above heat treatment test with an
active culture, was noted when Lactobacillus and juice mixture were heat-
treated immediately just after mixing and, in the case of orange juice, also 3
hours after the preparation. On the other hand, the time 1–2h interval from
between the melting of the frozen culture and heat treatment appeared to be
lethal for the strain.
Cold shock is known to alter the liquid crystalline nature of the cytoplasmic
membrane by transforming it to a gel phase state, DNA supercoiling and
mRNAs encoding cold shock response proteins (Champomier-Vergès et al.,
2002). The actual stage of recovery from cold shock seems to be crucial for the
resistance of varying stresses affecting the strain. Sublethal exposure to one
stress is known to induce genetic responses in bacteria that can lead to elevated
tolerance and gives cross protection against other stresses (Klaenhammer and
Kullen, 1999). This could be one factor that affected the survival of ME-3
during following heat challenge.
Heat resistance was tested to evaluate the possibility to produce ME-3
containing probiotic juices, where technology foresees addition of the strain to
freshly pasteurised juice and following cooling of the final product. The results
show that despite some decrease in viable count, ME-3 could even survive
pasteurization in juices. Some technical advantages like encapsulation of the
strain can increase the viability of the strain and make pasteurized juices
suitable carrier`s for ME-3.

72
 19. Stability of the functional properties
of ME-3 in products

19.1. Antioxidative activity

The antioxidative activity among lactic acid bacteria seems to be strain specific.
Kaizu with colleauges (1993) showed that only about 3% of tested strains
revealed remarkable antioxidative properties.
L. fermentum ME-3 expresses high antioxidative activity in vitro. The strain
retained its high antioxidative property in all fermented milk products, though
in most products it was somewhat lower than the initial value of ME-3 original
culture. In the HELLUS brand, the antioxidativity of ME-3 was clearly
influenced by the acidity of a particular product and a flavour additive.
Acidity. The products tested by us can be categorized into acidic (yoghurts,
titrable acidity up to 1.1%), slightly acidic (sour cream, quark, 0.5%) and acid-
alcoholic (kefir). Lactic acid content appears to affect the TAA of ME-3, as in
products with lower acidity like sour cream and quark desserts the strain
expressed higher antioxidativity compared with yoghurts. The relatively lower
TAA values, of kefir, compared to the majority of other HELLUS brand
products can be due to kefir`s acidity and higher ethanol content. According to
literature, ethyl alcohol content in kefir can be up to 1% (Botazzi, 1983; Otles
and Cagindi, 2003).
Flavour additive. Besides lactic acid, flavour additives influenced
antioxidativity. The isolatef of ME-3 from yoghurts with cranberry-muesly and
rhubarb-oat flavours revealed the lowest TAA values. L. fermentum possesses
Mn-SOD, an enzyme related to the antioxidativity of the strain (Kullisaar et al.,
2002). Fibre- and oxalate rich additives (e.g. oat, wheat, barley, raisins and
rhubarb) or low-pH additives (e.g. orange jam in pasha flavoured yoghurt) have
the ability to absorb or enhance the absorbance of divalent microelements
including Mn. Through this; the activity of Mn-SOD is probably lowered.
Milling of additive cereals (e.g. rhubarb-oat yoghurt additive) increases the
microelement-binding reaction.
Cheese environment appeared to be suitable for ME-3, as the probiotic
strain survived well the cheese manufacturing and retained both its viability and
high antioxidativity. The tested cheese varieties Atleet and Pikantne were
similar with the respect to fat content (45% and 50% in dry matter,
respectively), but differed in all acpects of the manufacture and length of shelf-
life. The incorporation of ME-3 into cheese probably causes stress to the
probiotic as its cell count and TAA decreased was during cheese ripening. The
unique property to tolerate low temperatures distinguishes ME-3 from the other
L. fermentum strains (Hammes et al., 1992; Kandler and Weiss, 1986) and
makes possible for the strain to survive and multiply at relatively low
temperature of cheese ripening and even lower temperature of storage. The

73
TAA values, which decreased rapidly during the initial phase of the maturation
of Pikantne cheese, gained the high values of the original ME-3 culture towards
the end of shelf-life. In Atleet cheese, on the other hand, the TAA values
remained stable after the initial drop to the value of 18%. Fluctuations in viable
count and TAA values probably reflected the adaptation of the probiotic strain
to a certain cheese environment.
The TAA of the L. fermentum ME-3 isolates from all tested juices was
measured after 1 month of refrigerated storage. All isolates were found to reveal
relatively high TAA values (21... 31%), the isolates from orange juice having
the highest value. Although the survival of the strain in tropical carrot drink was
the lowest among the tested juices, the loss in viable count was compensated by
stable antioxidative activity, which remained at the same level as that of the
initial ME-3 culture.
Surprisingly, freeze-drying did not affect negatively the TAA of the strain,
although the range of the antioxidative activity of the ME-3 isolates from
capsule was more variable during several following isolation tests, than that of
the isolates from food products. Still, the mean value of TAA was only
somewhat lower (21±11%) than the base value of the Lactobacillus culture.

19.2. Antagonistic activity

The antagonistic activity of the ME-3 isolates from cheese and majority of the
HELLUS brand products against tested pathogens was significantly decreased
in comparison to the original probiotic culture. No correlation was detected
between products titrable acidity and antimicrobial activity. However, the
flavour additives seemed to influence also the antimicrobial properties of the
strain.
The exceptions were the ME-3 kefir isolates the antagonistic activity of
which was significantly higher. The starter mixture of kefir consisted of seven
species of microbes, representing 4 different genera: in addition to LAB
(Leuconostoc, different lactococci and Lactobacillus) two species of yeasts,
Klyveromyces marxianus and Saccharomyces unisporus were present.
Interestingly, the same lactococci species were present also in the sour cream
starter mix, but pathogen inhibition by the ME-3 isolates from this product type
remained lower. Therefore, the higher number of different microbes compared
with the other starters and probably especially the presence of yeasts in the kefir
starter may have stimulated the antagonistic activity of ME-3 to remain high.
Kefir has been used empirically to relieve gastrointestinal disorders. Though the
high antagonistic activity of ME-3 gives additional food safety value to the kefir
concerning occasional food contaminating pathogens, it remains unknown if the
property persists in also inside the human GI tract, providing putative defence
against enteric infections.

74
 In fermented milk products the interactions between the starter microbes
(and, in a cheese environment, additionally non-starter lactobacilli species) and
the probiotic additive are among main factors influencing the viability and
functional properties of the probiotic strain. It has been shown that probiotic
strains can be more suppressive towards starters than vice versa (Vinderola et
al., 2002), which in turn might have negative impact on the quality of the
product. Our data of the suppressive activity of ME-3 on starter microbes
revealed any negative interactions with the tested starters in vitro. Moreover, all
fermented milk products produced in the pilot plant experiment were of
commercial grade. Lack of antagonistic activity against different starter cultures
seems to be a strain-specific property of ME-3, making it easy to incorporate the
strain as a probiotic additive into different food products.
The antagonistic activity of the ME-3 isolates from capsule against the
tested pathogens was found to be significantly stronger than the respective basic
values of the probiotic culture before freeze-drying and capsulation. Probably
the inoculation of the inactive culture into a nutritious environment abruptly
reactivated the strain, the cells recovered from injuries caused by freeze-drying
and that in turn stimulated the expression of higher probiotic properties.
According to our results, L. fermentum ME-3 withstood the manufacturing
process, while technological handling did not affect negatively the probiotic
properties of the strain.

20. Health effects of ME-3

The stability of strain-specific properties plays an important role in the fate of
the ingested probiotic strain. A probiotic strain is supposed to survive,
transiently persist in the GI tract of the host and reach high numbers of viable
count in the targeted part of the gut. Probiotic strains react differently while
facing challenges in the GI tract.
L. fermentum ME-3 expressed in vitro good characteristics for survival and
performance in the human GI tract: the strain survived in the presence of solely
ox gall or at low pH as well as in simulated gastric and intestinal juices without
remarkable loss in viability. The health effects and possible side effects as well
as the dose responses of a potential probiotic strain are usually first evaluated in
trials with healthy volunteers and only then in clinical trials with patients
suffering from a particular health problem.
Mostly non-transmissible natural antibiotic resistance exists among
lactobacilli (Yazid et al., 2000; Mändar et al., 2001; Danielsen and Wind,
2003). Though plasmid based antibiotic resistance is not very common among
lactobacilli strains, it still can occur. L. fermentum ME-3 possesses intrinsic
resistance metronidazole, ofloxacin, aztreonam, cefoxitin and trimethoprim-

75
sulfmethoxazole. The localization (chromosomal or plasmid origin) of these
antibiotic resistance enconing genes need further investigations
In present study, no possible side effects of ME-3 consumption were noted
in the trial where healthy volunteers consumed capsulated ME-3 in a daily dose
of 3x109 CFU during 10 days. Besides, during the marketing period of the
HELLUS products since year 2003, no reports, concerning any side effects are
documented either being an indirect proof of the safety of L. fermentum
ME-3containing foods for consumers.
Interactions between the probiotic strain and the indigenous microbiota of
the host. Mutual interactions take place between a probiotic strain and the host’s
indigenous microbiota in the small intestine. It is documented that ingestion of a
certain probiotic may cause changes in the fecal flora by increasing the total
number of a particular genus like lactobacilli, bifidobacteria or enterococci
(Sepp et al., 1993; Alander et al., 1999; Brigidi et al., 2001; Cesena et al.,
2001). In the present study, ingestion of ME-3 caused a significant increase in
the total count of fecal LAB in comparison with the initial count. In all three
trials the increase was almost the same (more than ten times) in spite of the
probiotic formulation or daily dose. Competition for nutrients and attachment
sites as well as the influence of metabolites can cause changes in the fecal flora.
Besides, substances secreted by one microorganism could stimulate the growth
of the other microorganisms. It has been shown that even cell-free milk
products fermented with probiotics can enhance the viable number of some
genera of fecal LAB (Romond et al., 1998). The increase in total LAB count in
feces in the present study could be due to some metabolites secreted by ME-3
into the GI tract, which were used as a substrate by other LAB. Though
significant increase in total fecal count of lactobacilli was detected in all trials
with healthy subjects in the present study, there occurred some correlation
between the probiotic ME-3 or/and the delivery vehicle in association with
increased number of certain Lactobacillus species.
To measure the disease risk reduction by probiotic, the main pathogenic
mechanisms of a particular disease have to be known and suitable markers for
measurement selected. In development of atherosclerosis the oxidative stress
and related factors play an important role (Witzum, 1994, Pihl et al., 2003).
Concerning reduction of infectious risk, Truusalu and colleagues proved on the
basis of a mice model, that freeze-dried ME-3, administered with water
suppressed Salmonella infection. This was associated with the increased count
of lactobacilli and reduction of excessive oxidative stress caused by pathogen in
the intestinal mucosa (Truusalu et al., 2004). In our volunteer studies the
increased count of lactobacilli, accompanied by the stability of the antagonistic
activity of ingested ME-3 and non-predominance in the total lactoflora seem
valuable for optimization of the gut microflora balance and provision of
putative defense against enteric infections.
In the present study, two dose response trials were carried out with healthy
persons consuming either capsulated ME-3 (daily dose 1.5x109CFU) or ME-3

76
containing fermented goat milk (daily dose 3x1011) in comparison with goat
milk with the aim to clarify if the daily dose during 3 weeks was enough to
achieve some improvement in selected oxidative stress markers. Six markers
were chosen to evaluate the positive effect of the strain on human health. Two
microbiological markers – fecal recovery of the strain as an indicator of its
survival in the human GI tract and changes in total LAB count affected by
probiotic consumption – were determined. Four biochemical markers were
chosen concerning the possible change in whole human body oxidative stress
markers resulting from probiotic consumption. Urine 8-isoprostanes were
measured as indirect marker for lipid peroxidation. The state of the lipid
fraction of the antioxidative defence system of the human blood was evaluated
by TAA and TAS evaluated the state of the water-soluble fraction of the human
blood. The glutathione redox ratio, as a marker indicating intracellular oxidative
stress, and the balance between the reduced form and the oxidized form of
glutathione were measured from the sera.
The mode of probiotic administration. One important effecter on the
successful performance of the ingested strain is the mode of administration of
the probiotic into the human GI tract. Milk as a delivery vehicle has a dual
effect on the probiotic additive: buffering capacity of milk protects the viability
of the strain against stomachs’ acidic conditions in the stomach. On the other
hand, according to the investigations of Ouwehand et al. (2001) milk, especially
milk with higher fat content, can reduce the adhesive properties of a strain
(although the negative effect seems to be strain-dependent) and remove the
strain from the GI tract with feces. According to literature, the fecal recovery of
probiotic strains is higher when consumed with fermented milk/whey (mostly
90…100%) in comparison with relatively poorer (25…86%) fecal recovery,
when they are administered in the freeze-dried form in capsules (Goldin et al.,
1992; Jacobsen, et al., 1999; Mattila-Sandholm et al., 1999; Brigidi et al., 2001;
Hattaka et al., 2003). Moreover, in addition to the protective effect, which
affects the survival of the ingested probiotic, milk contains natural ”lactogenic”
factors like lactose, minerals, vitamins and other components as bioactive
petides, which enhance the metabolic activity of the ingested probiotic strain in
the GI tract.
Therefore, the high fecal recovery of a probiotic strain after consumption
can reflect not only the viability and good survival of a particular strain in the
GI tract, but also it’s reduced adhesion. The fecal recovery of administered
ME-3 was different when it was administered in fermented goat milk or in
capsules. The strain recovered in the feces of all participants of the fermented
goat milk trial, but it remained below levels detectable by classical
bacteriological methods when it was consumed in the capsulated form.
Probiotic dose. The daily dose is another factor affecting the presence of the
strain in feces. The dose–response study with capsulated Lactobacillus
rhamnosus GG showed colonisation of feces with a daily consumption of
109 CFU (Saxelin et al., 1991). The best recovery of the ingested strain was

77
noted with a daily dose starting from 1010 viable cells, however, there may be
strain-specificity concerning the dose response of different probiotic strains. In
the present study, the highest daily dose of the capsulated formulation of ME-3
was up to 3x109 CFU. The strain was below detection level in the feces of the
participants of both volunteer trials. Yet, its presence in gut was proved by the
positive antioxidative effect on blood but not on urine indices. Probably, if
administered in moderate quantities, as in the case of capsule trials in the
present study, a majority of the ingested cells of the probiotic strain attach to the
upper parts of the GI tract and the strain is undetectable in feces by classical
bacteriological methods. On the other hand, when ingested with higher daily
doses, the surplus of the unadhered strain is probably washed out.
Time of consumption. The time of consumption of the probiotic is important
for the survival of the strain: the probiotic was ingested before, after or during
the meal, during the active part of day or in the evening immediately before
sleep when digestion has slowed down. In the case, the time of contact of the
ingested probiotic with gastric juice is longer, which definitely affects the
viability of the strain. In the present study, the capsules were distributed to all
participants at the beginning of the trial (day 0) and no prescriptions were made
concerning probiotic ingestion time or way (i.e. during or between meals). In
the fermented goat milk trial, the daily dose was given to the participants during
the active part of the day.
Ecological conditions of the digestive tract. Fecal persistence does not
necessarily reflect the fate of the ingested probiotic strain in the small intestine,
as ecological conditions of the upper parts of the digestive tract differ from
those of the lower parts (distal colon, large bowel). It has been demonstrated
that the human cecal flora differs quantitatively and qualitatively from the fecal
flora, LAB count being 3.5 times higher in the cecum (Marteau et al., 2001).
Besides, the microbes present in feces represent mostly the luminal flora, which
probably differs from the mucosal flora. Alander and co-workers (1999)
demonstrated that the ingested probiotic (LGG) persisted in the colonic mucosa
even after its disappearance from fecal samples. Therefore, although ME-3 did
not recover in the feces of the participants of the probiotic capsule trial even
after 3 weeks of consumption, we can still presume that it survived passage
even in the case of a low dose and attached to the intestinal mucosa, which was
confirmed with positive changes in the health markers for antioxidative status.
Metabolic status of the probiotic strain. In the present study, ME-3 was
administered to healthy volunteers in two different formulations: in fermented
goat milk, i.e. in the metabolically active form, and as a lyophilised culture, i.e.
inactive form, in gelatine capsules. As a result of consumption of ME-3 with
either vehicle no adverse side effects were detected.
According to the results of the present investigation, fermented milk
appeared a more suitable carrier for ME-3 in comparison with the food
supplement, enhancing the effect of ME-3 in the gut. Consumption of fermented
goat milk, containing ME-3, lowered significantly the oxidative stress markers

78
for all members of the study group. Besides the strain, the vehicle itself has a
positive effect on human health. On the other hand, the certain strain, the
chemical composition of milk and the probiotic fermentation products in milk
ingested together with the live probiotic may also play important role for in vivo
health effects of the probiotic. The aspect needs further investigation in relation
to ME-3.
Impact of ME-3 on selected antioxidativity markers of the healthy humans.
The present study shows strong significant association between the mode of
formulation of the probiotic and the expression of its functional properties
inside the healthy host. The antioxidative potential of the food supplement
containing ME-3 was excellent, as reisolates of the strain from capsule
expressed significantly higher TAA in comparison with the base values of the
strain in vitro, still the shifts in the TAA markers in blood serum were less
pronounced in comparison with ME-3 fermented goat milk.
Although a significant increase in the total antioxidative status was detected
after longer consumption of capsules, another parameter of antioxidativity,
reduction in 8-isoprostanes in urine, was not so clearly expressed. The
explanation for this may be in the different responses of the subjects to capsule
consumption.
The reduction of the glutathione red-ox ratio was detected after the
consumption of fermented by ME-3 goat milk but not when consumed in the
capsule. More pronounced positive shifts in oxidative stress markers for former
volunteers can be due to the synergistic effect of the probiotic and the substrate.
According to the results of safety and functional efficacy trials with pro-
biotic capsules, the daily dose (1.5x109–3x109 CFU) was obviously moderate:
ME-3 was below detection level in fecal samples by bacteriological methods.
On the other hand, a clear improvement in the laboratory indices of the
antioxidative defense system of a healthy host was documented in all trials.
Additionally, the amount of total fecal lactoflora was increased. All this serves
as an indirect proof for the viability and functioning of lactobacilli in the host
when ingested in either formulation. The daily dose of 109 CFU can be
sufficient for improvement of antioxidative parameters when used during long
term, and through this, for prevention of the range of the diseases related to
oxidative stress.
Surely, in order to achieve clearer health effects to be applied in prevention
and treatment of different oxidative stress induced diseases some new trials
must be carried out. The trials of this study with freeze-dried capsulated
L. fermentum ME-3 should be considered as pilot studies, though helping to
bring some clarity to the dose and health effects of the antimicrobial and
antioxidative probiotic strain.
The effects of L. fermentum ME-3 on selected oxidative stress markers of
humans with other formulations, like with yoghurt or cheese must also be
carried out in comparison with the consumption of the regular product without
the probiotic.

79
 CONCLUSIONS
The present study evaluates survival in the GI tract, safety, stability of
functional and technological properties of the probiotic strain Lactobacillus
fermentum ME-3 in vitro and in human volunteer trials in accordance with the
FAO/WHO guidelines with the next main results:

1. Survival of L. fermentum ME-3 in the GI tract was proved on the basis of the
ability to resist acid, bile and digestive enzymes. The recovery of ME-3 from
the human GI tract was confirmed by elaborated simple rapid phenotypic
and molecular identification methods in volunteers consuming goat milk
fermented with ME-3 with the daily dose of 3x1011CFU. However, no
recovery was detected after the consumption of freeze-dried ME-3 (the daily
dose 3x109CFU). Thus, the efficacy of recovery seems to depend on the
daily dose and on the supportive properties of formulations consumed.

2. The good tolerance of L. fermentum ME-3 consumption and its safety were
confirmed by the absence of any side effects in volunteer trial participants.
Moreover, the two-year market period of ME-3 products (HELLUS brand)
without reports on any adverse effects indicates its safe use.

3. High antioxidative and antimicrobial activity values of L. fermentum ME-3

in different carriers (fermented milk products, juices, capsules) prove the
stability of its functional properties in food and food additives.

4. L. fermentum ME-3 is well suited for the technological processing of

fermented milk products and cheese without negatively affecting their
quality and commercial grade. Various useful properties of ME-3 have
granted its adaptation to different carriers: absence of antagonistic activity
between the probiotic strain and commonly used starters of milk products;
ability to reorganize metabolism by deficiency of nutrients; ability to tolerate
harsh environmental conditions as high temperature and acidity; survival at
viable count levels recommended for probiotics in various products.

5. Increased fecal lactobacilli count was detected in all participants of human

volunteer studies irrespective of the type of the probiotic formulation. This
increase, accompanied by the stability of the antagonistic activity of ingested
ME-3 and non-predominance in the total lactoflora seems valuable for
optimization of the gut microflora balance and provision of putative defense
against enteric infections.

6. In healthy volunteers, the antioxidative effect of L. fermentum ME-3

consumption was proved by positive effect the oxidative stress markes in

80
 sera (total antioxidative activity, TAA of lipid fractions, total antioxidative
status, TAS of water fractions, glutathione redox ratio GSSG/GSH) and
urine 8-isoprostanes. However, after 3-week consumption of ME-3 in fer-
mented goat milk, positive changes were detected in all measured oxidative
stress-related indices (8-isoprostanes, GSSG/GSH, TAA and TAS), while in
the capsule trial only blood serum TAA and TAS values got increased. Thus,
for detection of the antioxidative effect of a probiotic, the testing of several
oxidative stress parameters of the healthy humans is necessary.

7. Our experimental and volunteer studies indicate that L. fermentum ME-3 has
the principal characteristics necessary for its application as a probiotic
component for functional food and food supplement in normal population.
The specific health claims include the augmentation of the human defence
system against oxidative stress.

81
 REFERENCES

Alander, M., Satokari, R., Korpela, R., Saxelin, M., Vilpponen-Salmela, T., Mattila-
Sandholm, T. and von Wright, A. Persistence of colonization of human colonic
mucosa by a probiotic strain, Lactobacillus rhamnosus GG, after oral consumption.
Appl. Environ. Microbiol. 65, 351–354 (1999).
Annuk, H. Selection of medicinal plants and intestinal lactobacilli as antimicrobial
components for functional foods. Dissertation, Tartu (2002).
Annuk, H., Vainumäe, I., Aimla, K., Ausmees, K., Kullisaar, T., Kask, S. and Mikelsaar
M. Antagonistic activity of Lactobacilli against urinary Escherichia coli.
Microecol. Therapy. 28, 169–175 (1999).
Annuk, M., Zilmer, M. and Fellström, B. Endothelium-depedent vasodilatation and
oxidative stress in chronic renal failure: impact on cardiovascular disease. Kidney
Int. 84, 50–53 (2003).
Axelsson, L. Lactic Acid Bacteria: Classification and Physiology. In: Lactic Acid
Bacteria (S. Salminen and A. von Wright, eds.). Marcel Dekker, New York, pp.
1–72 (1998).
Björksten, B. Risk factors in early childhood for development of atopic diseases.
Allergy. 49, 400–407 (1994).
Boland, M., MacGibbon, A. and Hill, J. Designer milks for the new millennium.
Livestock Production Science. 72, 99–109 (2001).
Bos, C., Gaudichon, C. and Tomé, D. Nutritional and physiological criteria in the
assessment of milk protein quality for humans. J. Am. Coll. Nutr. 19, 191–205
(2000).
Bottazzi, V. Other fermented dairy products. Food and feed production with micro-
organisms. In: Biotechnology. (G. Reed, ed.) Verlag Chemie, Weinhein, 5, pp.
315−364 (1983).
Brandt, K. and Alatossava, T. Specific identification of certain probiotic Lactobacillus
rhamnosus strain with PCR primers based on phage-related sequences. Int. J. Food
Microbiol. 84, pp. 189–196 (2003).
Brigidi, P., Vitali, B., Swennen, E., Bazzocchi, G. and Matteuzzi, D. Effects of
probiotic administration upon the composition and enzymatic activity of human
fecal microbiota in patients with irritable bowel syndrome or functional diarrhea.
Res. Microbiol. 152, 735–741 (2001).
Cesena, C., Moreli, L., Alander, T., Siljander, T., Tuomola, E., Salminen, S., Mattila-
Sandholm, T., Vipponen-Salmela, T. and von Wright, A. Lactobacillus crispatus
and its nonaggregating mutant in human colonization trials. J. Dairy Sci. 84, 1001–
1010 (2001).
Champomier-Vergès, M.-C., Manguin, E., Mistou, M.-I., Anglade, P. and Chich, J.-F.
Lactic acid bacteria and proteomics: current knowledge and perspectives. J.
Chromatogra. 771, 329–342 (2002).
Christensen, J.E, Dudley, E.G., Pederson, J.A. and Steele, J.L. Peptidases and amino
acid catabolism in lactic acid bacteria. Antonie van Leeuwenhoek. 76, 217−246
(1999).
Chukeatirote, E. Potential use of probiotics. Songklanakarin Sci. Technol. 25, 275–282
(2002).

82
Collins, J.K., Thornton, G. and Sullivan, G.O. Selection of probiotic strains for human
application. Int. Dairy J. 8, 487–490 (1998).
Commission of the European Communities. Proposal for a regulation of the European
Parliament and of the Council on nutrition and health claims made on foods.
Brussels, 2003 Retrieved from: http://europa.eu.int/eur-lex/en/com/pdf/2003/
com2003_0424 en01.pdf
Conway, P.L. Prebiotics and human health: the state-of-the-art and future perspectives.
Scand. J. Nutr. 45, 13–21 (2001).
Corsetti, A., De Angelis, M., Dellaglio, F., Paparella, A., Fox, P.F., Settanni, L. and
Gobbetti, M. Characterization of sourdough lactic acid bacteria on genotypic and
cell-wall protein analysis. J. Appl. Microbiol. 94, 641–654 (2003).
Crow, D. Inulin – a comprehensive scientific review. Retrieved April 28, 2004 from
http://members.shaw.ca/duncancrow/inulin_review.html
Cusick, S.M. and O'Sullivan, D.J. Use of a single, triplicate arbitrarly primed –PCR
procedure for molecular fingerprinting of lactic acid bacteria. Appl. Environ.
Microbiol. 66, 2227–2231 (2000).
Esch, T. and Stefano, G.B. Proinflammation: a common denominator or initiator of
different patopfysiological disease processes. Med. Sci. Monit. 2002. Retrieved
from: http://www.MedSciMonit.com/pub/vol_8/no_5/2686.pdf
FAO/WHO. Guidelines for the evaluation of probiotics in food. Food and Agriculture
Organization of the United Nations and World Health Organization Working Group
Report. London, Ontario, Canada, 2002. Retrieved from:
http://www.fao.org/es/ESN/Probio/probio.htm.
Floch, M.H., Binder, H.J., Filburn, B. and Gershengoren, W. The effect of bile acids on
intestinal microflora. Am. J. Clin. Nutr. 25, 1418–14126 (1972).
Fuller, R. Probiotics in human medicine. Gut 32, 439–442 (1991).
Danielsen, M. and Wind, A. Susceptibility of Lactobacillus spp. to antimicrobial agents.
Int. J. Food Microbiol. 82, 1–11(2003).
De Angelis, M., Corsetti, A., Tosti, N., Rossi, J., Corbo, M.R. and Gobbetti, M.
characterization of non-starter lactic acid bacteria from Italian ewe cheeses on
phenotypic, genotypic, and cell wall protein analyses. Appl. Env. Microbiol. 67,
2011–2020 (2001).
De Vuyst, L. Technology aspects related to the application of functional starter cultures.
Food Technol. Biotech. 38, 105–112 (2000).
Diplock, A.T., Charleux, J.-L., Crozier-Willi, G., Kok, F. J., Rice-Evans, C.,
Roberfroid, M., Stahl, W. and Viňa-Ribes, J. Functional food science and defence
against reactive oxidative species. Br. J. Nutr. 80, 77S–112S (1998).
Eisen, M. Pathogenesis of contact dermatitis: participation of oxidative stress.
Dissertation, Tartu (2002).
Gibson, G.R. Dietary modulation of the human gut microflora using the prebiotics
oligofructose and inulin. J. Nutr. 129, 1438S–1441S (1999).
Gibson, G.R. and Roberfroid, M.B. Dietary modulation of the human colonic
microflora: introducing the concept of prebiotics. J. Nutr. 125, 1401–1412 (1995).
Gibson, G. Prebiotics for improved gut health. Retrieved November 20, 2001 from:
http://www.ifis.org/forum/ /March_2001/prebiotics4health_real.html
Goldin, B.R., Gorbach, S. L., Saxelin, M., Barakat, S., Gualtieri, I. and Salminen, S.
Survival of Lactobacillus species (strain GG) in human gastrointestinal tract. Dig.
Dis. Sci. 37, 121–128 (1992).

83
Gomes, A.M.P. and Malcata, F.X. Bifidobacterium spp and Lactobacillus acidophilus:
biological, biochemical, technological and therapeutical properties relevant for use
as probiotics. Trends Food Sci Technology. 10, 139–157 (1999).
Gould, J.C. Quantity and quality in the diagnosis of urinary tract infections. Brit. J.
Urol. 37, 7–11 (1965).
Griffith, O.W. Determination of glutathione and glutathione disulfide using glutathione
reductase and 2-vinylpyridine. Anal. Biochem. 106, 207–212 (1980).
Hammes, W.P. and Vogel, R.F. The genus Lactobacillus. In: The Genera of Lactic Acid
Bacteria (B. J. B. Wood and W.H. Holzapfel, eds.). Chapman & Hall, London, Vol.
2, pp. 19–54 (1995).
Hammes, W.P., Weiss, N. and Holzapfel, W. The genera Lactobacillus and
Carnobacterium. In: The Procaryotes (A. Balows, H.S. Trüpes, M. Drowkin and K.-
H. Schleifer, eds.). Springer-Verlag, Heidelberg, New York, Vol. 2, pp. 1535–1594
(1992).
Hattaka, K., Martio, J., Korpela, M., Herranen, M., Poussa, T., Laasanen, T., Saxelin,
M. and Vapaatalo, H. Effects of probotic therapy on the activity and activation of
mild rheumatroid arthritis – a pilot study. Scand. J. Rheumatol. 32, 211–215
(2003).
Helander, I.M., von Wright, A. and Mattila-Sandholm, T.M. Potential of lactic acid
bacteria and novel antimicrobials against Gram-negative bacteria. Trends Food Sci.
8, 146–150 (1997).
Heller, K.J. Probiotic bacteria in fermented foods: product characteristics and starter
organisms. Am. J. Clin. Nutr. 73, 374S–379S (2001).
Holdeman, L.V., Cato, E.P., and Moore, W.E.C. Anaerobe Laboratory Manual. Virginia
Polytechnic Institute and State Laboratory 4th edn. Virginia, Blacksburg (1977).
Holzapfel, H.W., Haberer, P., Geisen, R., Björkroth, J. and Schillinger, U. Taxonomy
and important features of probiotic microorganisms in food and nutrition. Am. J.
Clin. Nutr. 73, 365S–373S (2001).
Hütt, P., Lõivukene, K., Kullisaar, T., Shchepetova, J and Mikelsaar, M. Antagonistic
activity of probiotic lactic acid abcteria and bifidobacteria against entero- and
uropathogens. J. Appl. Microbiol. (2005) submitted.
ILSI Europe. Scientific concepts of functional foods in Europe: consensus document.
Br. J. Nutr. 81, 1S–27S (1999).
Jacobsen, C.N., Rosenfeldt Nielsen, V., Hayford, A. E., Møller, P. L., Michaelsen, K.
F., Pærregaard, A., Sandström, B., Tvede, M. and Jakobsen, M. Screening of
probiotic activities of forty−seven strains of lactobacillus ssp by in vitro techniques
and evaluation of the colonization ability of five selected strains in humans. Appl.
Environ. Microbiol. 65, 4949−4956 (1999).
Jan, G., Leverrier, P., Proudy, I. and Roland, N. Survival and beneficial effects of
propionibacteria in the human gut: in vivo and in vitro investigations. Lait. 82,
131–144 (2002).
Isolauri, E., Sütas, Y., Kakaanpää, P., Arvilommi, H. and Salminen, S. Probiotics:
effects of immunity. Am. J. Clin. Nutr. 73, 444S–450S (2001).
Jenkins, J. K., Harper, W. J. and Courtney, P. D. Genetic diversity of Swiss cheese
starter cultures assessed by pulse field gell elecrophoresis. Lett. Appl. Microbiol.
35, 423–427 (2002).

84
Kaizu, M., Sasaki, M., Nakajama, H., and Y. Suzuki. Effect of antioxidative lactic acid
bacteria on rats fed a diet deficient in vitamin E. J. Dairy Sci. 76, 2493–2499
(1993).
Kandler O. and Weiss N. Regular non-sporing Gram-positive rods.In: Bergey`s Manual
of Systematic Bacteriology. (P.H.A. Sneath, N.S. Mair and M.E. Sharpe eds.),
Williams and Wilkins, Baltimore, Vol. 2, pp. 1209–1234 (1986).
Karelson, E., Bogdanovic, N., Garlind, A., Winbland, B., Zilmer, K., Kullisaar, T.,
Vihalemm, T., Kairane, C. and Zilmer, M. The cerebrocortical areas in normal
brain aging and in Alheimer’s disease: noticeable differences in the lipid
peroxidation level and in antioxidant defence. Neurochem. Res. 26, 353–361
(2001).
Karimi, O. and Peña, A.S. Probiotics: isolated bacteria strain or mixtures of different
strains? Drugs Today. 39, 565–597 (2003).
Kaur, I.P., Chopra, K. and Saini, A. Probiotics: potential pharmaceutical applications.
Eur. J. Pharm. Sci. 15, 1–9 (2002).
Kask, S. Identification and characterization of dairy-related Lactobacillus. Dissertation,
Tallinn, (2003).
Klaenhammer, T.R. Bacteriocins of lactic acid bacteria. Biochimie. 70, 337–379 (1988).
Klaenhammer, T.R. and Kullen, M.J. Selection and design of probiotics. Int. J. Food
Microbiol. 50, 45–57 (1999).
Klein, G., Pack, A., Bonaparte, C. and Reuter, G. Taxonomy and physiology of
probiotic lactic acid bacteria. Int. J. Food Microbiol. 41, 103–125 (1998).
Korhonen, H. Technology options for new nutritional concepts. Int. J. Dairy Technol.
55, 79–88 (2002).
Kullisaar T, Zilmer M, Mikelsaar M, Vihalemm T, Annuk H, Kairane C, Kilk A. Two
antioxidative lactobacilli strains as promising probiotics. Int. J. Food Microbiol. 72,
215–224 (2002).
Laht, T.-M. Influence of microstructure of the curd on enzymatic and microbiological
processes in Swiss-type cheese. Dissertation, Tallinn (2003).
Lee, Y-K. and Wong, S-F. Stability of lactic acid bacteria in fermented milk. In: Lactic
Acid Bacteria. Microbiology and Functional Aspects. (S. Salminen and A. von
Wright, eds.) Marcel Dekker Inc, New York, pp.103−115 (1998).
Lencner, A.A. Lactobacilli of human microflora. Dissertation, Tartu (1973).
Lencner, A.A., Lencner, Ch.P., Mikelsaar, M., Tjuri, M.E., Toom, M.A., Väljaots,
M.E., Šilov, V.M., Liz´ko, N.N., Legenikov, V. I., and Reznikov I. M. Die
quantitative Zusammensetzung der Laktoflora des Verdauungstraktes vor und nach
kosmischen Flügen unterschiedlicher Dauer. Die Nahrung. 28, 607–613 (1984).
Lenzner H.P. and Lenzner A.A. Lysozyme activity and susceptibility of Lactobacilli to
lysozyme of human microflora lactobacilli. Mikrobielle Umwelt und
antimikrobielle Massnahmen. (J. A. Barth, ed.) Verlag, Leipzig, pp. 73–81 (1982).
Lin, M.-Y. and Chang, F.-J. Antioxidative effect of intestinal bacteria Bifidobacterium
longum ATCC 15708 and Lactobacillus acidophilus ATCC 4356. Dig. Dis. Sci. 45,
1617–1622 (2000).
Lin, M.-Y. and Yen, C.-L. Antioxidative ability of lactic acid bacteria. J. Agric. Food
Chem. 47, 1460–1466 (1999a).
Lin, M.-Y. and Yen, C.-L. Inhibition of lipid peroxidation by Lactobacillus acidophilus
and Bifidobacterium longum. J. Agric. Food Chem. 47, 3661–3664 (1999b).

85
Mackay, A.D., Taylor, M.B., Kibbler, C.C. and Hamilton-Miller, J.M.T. Lactobacillus
endocarditis caused by a probiotic organism. Clin. Microbiol. Infect. 5, 290–292
(1999).
Maeno, M., Yamamoto, N. and Takano, T. Identification of anti-hypertensive peptides
from casein hydrolysate produced by a proteinase from Lactobacillus helveticus
CP790. J. Dairy Sci. 73, 1316–1321 (1996).
Marteau, P., Pochart, P., Dore, J., Bera-Maillet, C., Bernalier, A. and Corthier, G.
Comparative study of bacterial groups within the cecal and fecal microbiota. Appl.
Env. Microbiol. 67, 4939–4942 (2001).
Mattila-Sandholm, T., Blum, S., Collins, J.K., Crittenden, R., de Vos, W., Dunne, C. et
al. Probiotics: towards demonstrating efficacy. Trends Food Sci. Technol. 10, 393–
399 (1999).
Matsumiya, Y., Kato, N. and Watanabe, K. Molecular epidemiological study of vertical
transmission of vaginal Lactbacillus species from mothers to newborn infants in
Japanese, by arbitrarily primer polymerase chain reaction. J. Infect. Chemother. 8,
43–49 (2002).
Meisel, H. and Bocklemann, W. Bioactive peptides encrypted in milk proteins:
proteolytic activation and tropho−functional properties. In: Proceedings of the sixth
symposium on lactic acid bacteria: genetics, metabolism and applications. 19−23
September 1999. Veldhoven. (W. N. Konings, O. P. Kuipers, and J. Huis in`t Veld.,
eds.) Kluver Academic Publishers, The Netherlands. pp.207−215 (1999).
Meister, A. and Anderson, M.E. Glutathione. Annu. Rev. Biochem. 52, 711–760
(1983).
Mercenier, A., Pavan, S. and Pot, B. Probiotics as biotherapeutic agents: present
knowledge and future prospects. Curr. Pharm. Des. 8, 99–110 (2002).
Mercenier, A., Hols, P., Roussel, Y., Perez-Martinez, G., Buesa, J., Wilks, M., Pozzi G.,
Remaut, E., Morelli, L., Grangette, C., Monedero, V., Palumbo, E. Foligne, B.
Steidler, L. and Nutten, S. Screening and construction of probiotic strains with
enhanced protective properties against intestinal disorders. Microb. Ecol Health
Dis. 16, 86–95 (2004).
Mikelsaar, M. Lactobacilli in the fecal microflora by some noninfectious diseases of GI
tract. Acad. Dissertation, Tartu (1969) (in Russian).
Mikelsaar, M., Lenzner A and Goljanova L.A. Methods for the estimation of the
quantitative composition of fecal microflora. Laboratornoje delo 1, 41–45 (1972)
(in Russian)
Mikelsaar, M., Türi, M., Lenzner, H., Kolts, K., Kirch, R. and Lenzner, A. Interrelations
between mucosal and luminal microflora of gastrointestine. Die Nahrung. 31, 449–
456 (1987).
Mikelsaar, M., Mändar, R. and Sepp, E. Lactic acid microflora in the human microbial
ecosystem. In Lactic Acid Bacteria. (S. Salminen and A. von Wright, eds.) pp. 279–
342. New York: Marcel Dekker (1998).
Mikelsaar, M., Kullisaar, T, Zilmer, M. Antagonistic and antioxidative activity of
lactobacilli and survival in oxidative milieu. Am. J. Clin. Nutr. 73, 45 (2001).
Mikelsaar, M., Annuk, H., Shchepetova, J., Mändar, R., Sepp, E., and Björkstén, B.
Intestinal Lactobacilli of Estonian and Swedish children. Microb. Ecol. Health Dis.
14, 75–80 (2002).

86
Mikelsaar, M., Mändar, R., Sepp, E., Annuk, H. Human lactic acid microflora and its
role in the welfare of the host. In: Lactic Acid Bacteria. (S. Salminen, von Wright,
A. and A. Ouwehand, eds.) Marcel Dekker Inc., pp. 453–506 (2004).
Miller, R.A. and Britigan, B.E. Role of oxidants in microbial pathophysiology. Clin.
Microecol. Revs. 1–18 (1997).
Mishra, C. and Lambert, J. Production of antimicrobial substances by probiotics. Asia
Pac. J. Clin. Nutr. 5, 20–24 (1996).
Morelli, L., Zonenschain, D., Callegari, M.L., Grossi, E., Maisano, F. and Fusillo, M.
Assessment of a new symbiotic preparation in healthy volunteers: survival,
persistence of probiotic strains and its effect on the indigenous flora. Nutr. J. 2, 11
(2003). Retrieved from: http://www.nutritionj.com/content/2/1/11
Morrow, J.D., Awad, J.A., Boss, H.J., Blair, I.A. and Roberts, L.J. Non-
cyclooxygenase-derived prostanoids (F2-isoprostanes) are formed in situ on
phospholipids. Proc. Natl. Acad. Sci. 89, 10721–10725 (1992).
Moser, S.A. and Savage, D.C. Bile salt hydrolase activity and resistance to toxicity of
conjugated bile salts are unrelated properties in Lactobacilli. Appl. Environ.
Microbiol. 67, 3476–3480 (2001).
Mändar, R., Mändar, H. and Mikelsaar, M. Bioquant−a program for evaluation of fecal
microbiocenosos. 1. Baltic Congress of Laboratory Medicine. Clin. Chem.
Lookout. 56, 46 (1992).
Mändar R, Lõivukene K, Hütt P, Karki T, Mikelsaar M. Antibacterial susceptibility of
intestinal lactobacilli of healthy children. Scand. J. Infect. Dis. 33, 344–349 (2001).
Naaber, P. Clostridium difficile infections and microbial ecology. Dissertation, Tartu
(1997).
Naaber, P. and Mikelsaar, M. Interactions between lactobacilli and antibiotic-associated
diarrhea. Adv. Appl. Microbiol. 54, 231–260 (2004).
Naaber, P., Smidt, I., Štšepetova, J., Brilene, T., Annuk, H. and Mikelsaar, M. Inhibition
of Clostridium difficile strains by intestinal Lactobacillus species. J. Med.
Microbiol. 53, 551–554 (2004).
Orla-Jensen, S. Lactic Acid Bacteria. Host and Son, Copenhagen (1919).
O ̉Sullivan, D. Screening of intestinal microflora for effective probiotic bacteria. J.
Agric. Food Chem. 49, 1751–1760 (2001).
Otles, S. and Cagindi, O. Kefir: a probiotic dairy-composition, nutritional and
therapeutic aspects. Pakistan J. Nutr. 2, 54–59 (2003).
Ouwehand, A. and Salminen, S. The health effects of cultured milk products with viable
and non-viable bacteria. Int. Dairy J. 8, 749–758 (1998).
Ouwehand, A., Kirjavainen, P. V., Schrott, C. and Salminen, S. Probiotics: mechanics
and established effects. Int. Dairy J. 9, 43–52 (1999).
Ouwehand, A., Tölkkö, S., Kulmala, J., Salminen, S. and Salminen, E. Adhesion of
inactivated probiotic strains to intestinal mucus. Lett. Appl. Microbiol. 31, 82–86
(2000).
Ouwehand, A., Tuomola, E. M., Tölkkö, S. and Salminen, S. Assessment of adhesion of
novel probiotic strains to human mucus. Int. J. Food Microbiol. 64, 119–126
(2001).
Pastore, A., Federici, G., Bertini, E. and Piemonte, F. Analysis of glutathione in redox
and detoxification. Clin. Chim. Acta. 333, 19–39 (2003).

87
Pathmakanthan, S., Meance, S. and Edwards, C. A. Probiotics: a review of human
studies to date and methodological approaches. Microb. Ecol. Health Dis. 2, 10−30
(2000).
Pihl, E., Zilmer, K., Kullisaar, T., Kairane, C., Pulges, A. and Zilmer, M. High-sensitive
C-reactive protein level and oxidative stress-related status in former athletes in
relation to traditional cardiovascular risk factors. Atherosclerosis 171, 321–326
(2003).
Rapposch, S., Eliskases-Lechner, F. and Ginzinger, W. Growth of facultatively
heterofermentative Lactobacilli on starter cell suspensions. Appl. Environ. Micro-
biol. 65, 5597–5599 (1999)
Rautio, M., Jousimies-Somer, H., Kauma, H., Pietarinen, I., Saxelin, M., Tynkkinen, S.
and Koskela, M. Liver abscess due to a Lactobacillus rhamnosus strain
indistinguishable from L. rhamnosus strain GG. Clin. Infect. Dis. 28, 1159–60
(1999).
Reid, G., Jass, J., Sebulsky, M. T. and McCormick, J. K. Potential uses of probiotics in
clinical practice. Clin. Microbiol. Rev. 16, 658–672 (2003).
Reuter, G., Klein, G. and Goldberg, M. Identification of probiotic cultures in food
samples. Can. J. Microbiol. 34, 344–351 (2002).
Roberfroid, M.B. Prebiotics and probiotics: are they functional foods? Am. J. Clin.
Nutr. 71, 1682S–1687S (2000).
Romond, M.-B., Ais, A., Guillemot, F., Bounouader, R., Cortot, A. and Romond, C.
Cell-free whey from milk fermented with Bifidobacterium breve C50 used to
modify the colonic microflora of healthy subjects. J. Dairy Sci. 81, 1229–1235
(1998).
Ross, S. Functional foods: the Food and Drug Administration perspective. Am. J. Clin.
Nutr. 71, 1735S–1738S (2000).
Saarela, M., Morgensen, G., Fondén, R., Mättö, J. and Mattila-Sandholm, T. Probiotic
bacteria: safety, functional and technological properties. J. Biotechnol. 84, 197–215
(2000).
Salminen, S., Von Wrght, A., Morelli, L., Marteau, P., Brassart, D., de Vos, W.M.,
Fonden, R., Saxelin, M., Collins, K., Mogensen, G., Birkeland, S.-E. and Mattila-
Sandholm, T. Demonstration of safety of probiotics – a review. Int. J. Food
Microbiol. 44, 93–106 (1998).
Salminen, S., Ouwerhand, A., Benno, Y. and Lee, Y. K. Probiotics: how should they be
defined? Trends Food Sci. Technol. 10, 107–110 (1999).
Salminen, S. Human studies on probiotics: aspects of scientific documentation. Scand.
J. Nutr. 45, 8–12 (2001).
Sanders, M. E. Probiotics. Food Technol. 53, 67–77 (1999).
Sanders, M. E. and Huis in`t Veld J. Bringing a probiotic−containing functional food to
the market: microbiological, product, regulatory and labeling issues. In:
Proceedings of the sixth symposium on lactic acid bacteria: genetics, metabolism
and applications, 19−23 September 1999, Veldhoven, (W. N. Konings, O.P.
Kuipers, and J. Huis in`t Veld, eds.) Kluver Academic Publishers, The Netherlands.
pp. 293−316 (1999).
Saxelin, M., Chuang, N.H., Chassy, B., Rautelin, H., Salminen, S. and Gorbach, S.L.
Lactobacilli and bacteremia in Finland, 1989 – 1992. Clin. Infect. Dis. 22, 564–566
(1996).

88
Saxelin, M., Grenov, B., Svensson, U; Fonden, R; Reniero, R. and Mattila-Sandholm,
T. The technology of probiotics. Trends Food Sci. Tehcnol. 10, 387–392 (1999).
Saxelin, M., Elo, S., Salminen, S and Vaapatalo, H. Dose response colonisation of feces
after oral administration of Lactobacillus casei strain GG. Microb. Ecol. Health
Dis. 4, 209–214 (1991).
Shan, X., Aw, T.Y. and Jones, D.P. Glutathione-dependent protection against oxidative
injury. Pharmacol. Ther. 47, 61–71 (1990).
Sepp, E., Mikelsaar, M., and Salminen, S. Effect of administration of Lactobacillus
casei strain GG on the gastrointestinal microbiota of newborns. Microb. Ecol.
Health Dis. 6, 309–314 (1993).
Sepp E. Formation of intestinal microbial ecosystem in children. Dissertation, Tartu,
(1998).
Sepp, E., Julge, K., Vasar, M., Björksten, B. and Mikelsaar M. Intestinal microflora of
Estonian and Swedish infants. Acta Paediatr. 86, 956–961 (1997).
Sepp, E., Naaber, P., Voor, T., Mikelsaar, M. and Björksten, B. Development of
intestinal microflora during the first month of life in Estonian and Swedish infants.
Microb. Ecol. Health Dis. 12, 22–26 (2000).
Seppo, L., Kerojoki, O, Suomalainen, T. and Korpela, R. The effect of a Lactobacillus
helveticus LBK-16H fermented milk on hypertension- a pilot study on humans.
Milchwissenschaft. 57, 124–127 (2002).
Seppo, L., Jauhiainen, T., Poussa, T. and Korpela, R.A fermented milk high in bioactive
peptides has blood pressure-lowering effect in hypertensive subjects. Am. J.Clin.
Nutr. 77, 326–330 (2003).
Schillinger, U. Isolation and identification of lactobacilli from novel-type probiotic and
mild yoghurts and their stability during refrigerated storage. Int. J. Food Microbiol.
47, 79–87 (1999).
Simakachon, N., Pichaipat, V., Rithiopornpaisarn, P., Kongkaew, C., Tongpradit, P. and
Varavithya, W. Clinical evaluation of the addition of lyophilized, heat-killed
Lactobacillus acidophilus LB to oral rehydration therapy in the treatment of acute
diarrhea in children. J. Pediatr. Gastroenterol. Nutr. 30, 68–72 (2000).
Soleman, N., Laferl, H., Kneifel, W., Tucek, G., Budschedl, E., Weber, H., Pichler, H.
and Mayer, H. K. How safe is safe? – A case of Lactobacillus paracasei ssp.
paracasei endocarditis and discussion of the safety of lactic acid bacteria. Scand. J.
Infect. Dis. 35, 759–762 (2003).
Song, Y-L., Kato, N., Liu, C.-X., Matsumiya, Y., Kato, H. and Watanabe, K. Rapid
identification of 11 human intestinal Lactobacillus species by multiplex PCR assays
using group- and species-specific primers derived from 16S-23S rRNA intergenic
spacer region and its flanking 23S rRNA. FEMS Microbiol. Lett. 187, 167–173
(2000).
Stanton, C., Gardiner, G., Meehan, H., Collins, K., Fitzgerald, G., Lynch, P. B. and
Ross, R. P. Market potential for probiotics. Am. J. Clin. Nutr. 73, 476–483 (2001).
Stecchini, M.L., Del Torre, M. and Munari, M. Determination of peroxyl-radical-
scavenging of lactic acid bacteria. Int. J. Food Microbiol. 64, 183–188 (2001).
Stewart, L., Pellegrini, C. A. and Way, L. W. Antibacterial activity of bile acids against
common biliary tract organisms. Sugr. Forum. 37, 157–159 (1986).
Stiles, M.E. and Holzapfel, W.H. Lactic acid bacteria of foods and their current
taxonomy. Int. J. Food Microbiol. 36, 1–29 (1997).

89
Temmerman, R., Pot, B., Huys, G. and Swings, J. Identification and antibiotic suscepti-
bility of bacterial isolates from probiotic products. Int. J. Food Microbiol. 81, 1–10
(2002).
Tenover, F.C., Jorgensen, J.H., and Turnidge, I.D. Antimicrobial agents and
susceptibility testing. In: Manual of clinical microbiology (Murray, P.R., Baron, E.
J., Pfaller, M. A. et al., eds.) ASM Press, Washington DC. pp. 1467–662 (1999).
Tilsala-Timisärvi, A. and Alatossava, T. Strain-specific identification of probiotic
Lactobacillus rhamnosus with randomly amplified polymorphic DNA-derived PCR
primers. Appl. Environ. Microbiol. 64, 816–4819 (1998).
Truusalu, K., Naaber, P., Kullisaar, T., Tamm, H., Mikelsaar, R.-H., Zilmer, K.,
Rehema, A., Zilmer, M. and Mikelsaar, M. The influence of antibacterial and
antioxidative probiotic lactobacilli on gut mucosa in a mouse model of Salmonella
infection. Microb. Ecol. Health Dis. 16, 180–187 (2004).
Tuomola, E., Crittenden, R., Playne, M., Isolauri, E. and Salminen, S. Quality assurance
criteria for probiotic bacteria. Am. J. Clin. Nutr. 73, 393S–398S (2001).
Xanthopoulos, V., Ztaliou, I., Gaier, W., Tzanetakis, N. and Litopoulou-Tzanetaki, E.
Differentiation of Lactobacillus isolates from infant feces by SDS-PAGE and
rRNA-targeted oligonucleotide probes. J. Appl. Microbiol. 87, 743–749 (1999).
Xiao, S. D., Zhang, D.Z., Lu, H., Jiang, S.H., Liu, H.Y., Wang, G.S., Xu, G.M., Zhang,
Z.B., Lin, G.J., Wang, G.L. Multicenter randomised controlled trial of heat-killed
Lactobacillus acidophilus LB in patients with chronic diarrhea. Chin. J. Dig. Dis. 3,
167–171 (2002).
Vergin, F. Anti- und Probiotica. Hippocrates. 25, 16–119 (1954).
Vervaart, P. and Knight, K.R. Oxidative stress and the cell. Clin. Biochem. Rev. 17, 3–
16 (1996).
Vinderola, C.G., Mocchiutti, P. and Reinheimer, J.A. Interactions among lactic acid
starter and probiotic bacteria used for fermented dairy products. J. Dairy Sci. 85,
721–728 (2002).
Voronina, M. Lactobacilli of human stomach microflora. Dissertation, Tartu (1968).
Wetzel, K., Menzel, M. and Heller, K. J. Stress response in Lactococcus lactis and
Streptococcus thermophilus induced by carbon starvation. Kieler Milchwirtschaft
Forsch. Ber. 51, 319–332 (1999).
Witzum, J.L. The oxidation hypothesis of atherosclerosis. Lancet 344, 793–795 (1994).
Yazid, M.A., Ali, A. M., Shuhaimi, M., Kalaivaani, V., Rokiah, M.Y. and Reezal, A.
Antimicrobial susceptibility of bifidobacteria. Lett. Appl. Microbiol. 31, 57–62
(2000).
Yeung, P.S.M., Sanders, M.E., Kitts, C.L., Cano, R. and Tong, P.S. Species-specific
identification of commercial strains. J. Dairy Sci. 85, 1039–1051 (2002).
Zhong, W., Millsap, K., Bialkowska-Hobrzanska, H. and Reid, G. Differentiation of
Lactobacillus species by molecular typing. Appl. Environ. Microbiol. 64, 2418–
2423 (1998).

90
 SUMMARY IN ESTONIAN

Uudse probiootikumi Lactobacillus fermentum ME-3

tehnoloogiliste ja funktsionaalsete omaduste hindamine
Probiootikum on inimeselt pärinev elus mittepatogeenne mikroob, mille
manustamine on tervisele kasulik. Kõige sagedamini on probiootikumideks
laktobatsillid, mis on aastatuhandeid olnud inimtoidu koostisosaks. Kuigi neid
mikroobe peetakse üldiselt tervisele ohututeks, on iga potentsiaalset pro-
biootikumi vaja põhjalikult testida. Vajalik on tagada probiootikumi tarbimise
ohutus, aga ka tervistavate omaduste avaldumine peremeesorganismis. FAO ja
WHO 2002 aastal soovitatud astmeline skeem probiootikumide hindamiseks
hõlmab vastava mikroobi üldist iseloomustamist in vitro, funktsionaalsete
omaduste väljaselgitamist ja ohutuse tuvastamist loommudelil. Järgneb tüve
tervistava toime hindamine ning sobiva doosi väljaselgitamine kliinilistes
katsetustes, esmalt tervetel vabatahtlikel ja hiljem haigetel.
Lisaks eelnevale on oluline hinnata ka potentsiaalse probiootikumi sobivust
tehnoloogiliseks käitlemiseks ja säilimist tootes. Sellele olulisele aspektile ei ole
seni küllaldaselt tähelepanu pööratud. Samuti on vähe andmeid probiootikumi
tervistavate omaduste seosest probiootikumi erinevate kandjatega (piimatooted,
kapseldatud toidulisandid).
Probiootikumi tervistava toime eelduseks on tema eluvõime säilimine peale
seedekulgla läbimist. Samas on vajalik leida kergesti määratavaid inim-
organismi füsioloogilisi, biokeemilisi ja/või immunoloogilisi näitajaid, et
objektiivselt hinnata manustatud probiootikumi tervistava toime olemust.
Eesti ja Rootsi tervete laste laktobatsillide võrdleva uuringu käigus isoleeriti
1995. aastal Lactobacillus fermentum tüvi ME-3 (DSM 14241, algselt tähistatud
822-1-1 ja E-3), mille mõningaid olulisi omadusi on eelnevates publikat-
sioonides kirjeldatud. ME-3 iseloomustab tugev antioksüdatiivne aktiivsus, tüvi
tekitab Mn-superoksiidi dismutaasi, lõhustab hüdroksüülradikaali ja on
võimeline alandama glutatiooni redoks-suhet (GSH/GSSG). Lisaks on ME-3
antimikroobse toimega Gram-positiivsete ja -negatiivsete entero- ja uropato-
geenide ning Helicobacter pylori suhtes.

Uurimistöö eesmärgid ja ülesanded

Uurimuse üldiseks eesmärgiks oli hinnata, kas L. fermentum tüvi ME-3 kui
antioksüdatiivsete tervistavate omadustega mikroob on sobiv kasutamiseks
tervetel inimestel probiootikumina kas funktsionaalse toidu komponendina
ja/või toidulisandina.

91
Uurimustöö ülesanded:
1. Hinnata in vitro L. fermentum ME-3 erinevate omaduste sobivust
tehnoloogiliseks käitlemiseks:
• kõrge happesuse toimet tüve elulemusele;
• kõrgete temperatuuride taluvust happelises ja neutraalses keskkonnas;
• elulemust toitainete defitsiidi tingimustes
• toimet piimatoodetes kasutatavate starterkultuuride ja teiste toidust
pärinevate laktobatsillide suhtes
• elulemust erinevates fermenteeritud piimatoodetes, mahlades ja kapslites.

2. Hinnata L. fermentum ME-3 elulemust seedetrakti mudelis pepsiini,

soolhappe, pankreatiini ja sapi manulusel.

3. Selgitada L. fermentum ME-3 funktsionaalsete omaduste püsivust erinevates

fermenteeritud piimatoodetes, mahlades ja kapslites.

4. Hinnata ME-3 manustamise ohutust ja tervistavaid omadusi tervetel

täiskasvanud vabatahtlikel katseisikutel selgitades:
• manustamise talutavust ja kaasneda võivaid kõrvaltoimeid;
• tüve eritumist peale seetrakti läbimist bakterioloogiliste ja molekulaarsete
meetodite abil;
• toimet rooja laktobatsillidele hulgale ja liigilisele koostisele;
• toimet mõningatele vere ja uriini oksüdatiivse stressi markeritele;
• probiootilise tüve optimaalset doosi kahes erinevas vormis
manustamisel.

Materjal ja meetodid

L. fermentum ME-3 samastati liigi tasemel laktobatsillide samastamiskitiga

(API CHL 50) ja molekulaarselt ITS-PCR abil võrdluses tüüptüvedega. Tüve
tasemel samastamine toimus bakterioloogiliselt füsioloogilis-biokeemiliste
omaduste ning molekulaarselt AP-PCR alusel. L. fermentum ME-3 meta-
boliitide hulgad (äädik-, piim-, merivaikhape ja etanool) määrati gaaskromato-
graafilisel meetodil MRS vedelsöötmes. Antibiootikumtundlikkust selgitati
diskdifusiooni ja E-testide meetodil.
ME-3 antimikroobsete omaduste stabiilsuse uurimiseks Esherichia coli,
Shigella sonnei, Staphylococcus aureus'e, Salmonella enteritidis’e tüvede suhtes
ning ME-3 ja juuretise kultuuride/mittestartermikroobide vastastikuse toime
uurimiseks kasutati agardifusiooni ning antagonistliku aktiivsuse määramist
vedelas kasvukeskkonnas (piimasöötmes, lihapeptoonpuljongis). Antioksüdant-
sust in vitro määrati linoleenhappe testi (TAA) abil. Kliinilistes katsetustes
kasutati vereseerumi oksüdatiivse stressi markerite määramiseks Randox kitti,

92
linoleenhappe ja glutatiooni redoks testi ning ELISA testi uriinist 8-
isoprostaanide määramiseks.
L. fermentum ME-3 elulemust piim- ja äädikhappe keskkonnas määrati 24
tunni jooksul erinevatel pH väärtustel. Kõrgete temperatuuride (73oC, 85oC ja
95oC) taluvust hinnati happelises (Tallinna Piimatööstuse, TPT mahlad) ja
neutraalses keskkonnas (lõss), ME-3 elulemust “HELLUS” sarja piimatoodetes,
Vana-Kuuste Piimaühistu pikantses Tervisejuustus ja Võru Juustutööstuse
Atleet juustus ning TPT mahlades (apelsini mahl, multivitamiini nektar ja
troopiline jook porgandiga) toodete säilivusaja jooksul. Tüve elulemust
toitainete defitsiidi tingimustes määrati fosfaatpuhvris, mis sisaldas ainult
surmatud või elusaid juuretise ja mittestartermikroobe.
Seedeensüümide koostoimet ME-3 elulemusele uuriti pepsiini ja soolhapet
ning järgnevalt sappi ja pankreatiini sisaldavas fosfaatpuhvris.
ME-3 ohutust inimesele hinnati 10 päeva jooksul 22 tervel vabatahtlikul
manustades tüve probiootiliste kapslitena päevases annuses 3x109 mikroobi-
rakku. ME-3 antioksüdantset toimet ja probiootikumi optimaalset doosi hinnati
manustades tüve 45 tervele vabatahtlikule probiootiliste kapslite või fermen-
teeritud kitsepiimana erinevates päevastes doosides (vastavalt 109 või 1011
mikroobirakku) kolme nädala jooksul.
Probiootilise tüve elulemust inimese seedetraktis ja laktobatsillide hulga ja
liikide muutusi määrati katse eel ja lõpus erinevate rooja lahjenduste
väljakülvidega MRS söötmele. Isoleeritud laktobatsillid samastati füsioloogilis-
biokeemiliste omaduste alusel. ME-3 taasisolaate roojast samastati moleku-
laarselt kasutades AP-PCR. Oksüdatiivse stressi markeritena määrati vere-
seerumist TAA, TAS ja GSSG/GSH ning uriinist 8-isoprostaanid.

Uurimistöö tulemused ja järeldused

Käesolevas uurimuses iseloomustati L. fermentum ME-3 ohutust, eluvõimelisust

seedetraktis, funktsionaalsete ja tehnoloogiliste omaduste stabiilsust in vitro ja
katsetes tervete vabatahtlikega, toetudes FAO/WHO soovitatud skeemile.

1. ME-3 vastupidavust seedetraktis valitsevatele tingimustele tõestati in vitro

sapi, happestressi ning seedeensüümide koostoimel. Tüve eritus pärast
seedetrakti läbimist leidis kinnitust ka tervete vabatahtlikega läbiviidud
katsetuses fermenteeritud kitsepiimaga päevase doosi 3x1011 mikroobirakku
korral: ME-3 oli katseisikute roojast leitav väljatöötatud fenotüüpiliste
kiirtestide ning molekulaarsete meetodite abil, kuid tüvi ei olnud leitav ME-3
kapsleid manustanud isikute roojast (päevane doos 3x109 mikroobirakku).
Seega sõltub probiootikumi eritus roojaga tõenäoliselt nii sissevõetavast
doosist kui ka selle kandja toetavatest omadustest.

2. L. fermentum ME-3 oli katseisikutele hästi talutav ning selle manustamisega

vabatahtlikele katseisikutele ei kaasnenud ebasoovitavaid kõrvaltoimeid.

93
 ME-3 ohutuse kaudseks kinnituseks on kaubandusvõrgus saadaolevate
Tallinna Piimatööstus AS “HELLUS” sarja toodete kõrvaltoimeteta
tarbimine elanikkonnas kahe aasta jooksul.

3. L. fermentum ME-3 isolaatide kõrged antioksüdatiivsed ja antimikroobsed

väärtused erinevates kandjates (fermenteeritud piimatooted, mahlad, kapslid)
tõestavad mikroobi funktsionaalsete omaduste püsivust toidus ja
toidulisandites.

4. L. fermentum ME-3 on tehnoloogiliseks käitlemiseks sobiv, sest tüvi ei

mõjutanud negatiivselt fermenteeritud piimatoodete ja juustu kvaliteeti ning
kommertsiaalset väärtust. Tüvi ME-3 ei vähenda Eestis enimkasutatavate
juuretiste eluvõimet; on võimeline kohandama oma ainevahetust toitainete
defitsiidi tingimustele ning on suuteline taluma võimalikke tootmisprotsessis
ja toidus esinevaid stressoreid, nagu kõrgendatud temperatuur või happesus.
Toidus säilib L. fermentum ME-3 probiootikumidele ettenähtud hulgas.

5. Sõltumata probiootikumi manustamisvormist täheldati kõikide katseisikute

rooja laktobatsillide hulga suurenemist. See leid koos L. fermentum ME-3
antimikroobsete omadustega ja mittedomineerimine seedetraktis peale
manustamist loob võimaluse seedefloora tasakaalustamiseks ning pakub
võimalikku kaitset seedeinfektsioonide vastu.

6. ME-3 manustamisel saavutati tervetel vabatahtlikel vereseerumis (totaalne

antioksüdatiivne aktiivsus lipiidses fraktsioonis, TAA, totaalne antioksüda-
tiivne staatus vesifraktsioonis, TAS ja glutatiooni redokssuhe GSSG/GSH)
ning uriini (8-isoprostaanid) oksüdatiivse stressi parameetrites positiivseid
nihkeid, mis tõestab ME-3 antioksüdatiivset toimet inimorganismile. ME-3
sisaldava fermenteeritud kitsepiima joomine 3 nädala jooksul parandas
oluliselt kõiki meie poolt mõõdetud oksüdatiivse stressi parameetreid (8-
isoprostaane, GSSG/GSH, TAA ja TAS), seevastu sama ajaperioodi jooksul
ME-3 sissevõtmisel probiootiliste kapslitena muutusid oluliselt vaid TAA ja
TAS. Seega on probiootikumi antioksüdantsete toime hindamiseks oluline
mõõta samaaegselt mitmeid inimorganismi oksüdatiivse stressi para-
meetreid.

7. Meie eksperimentaalsed ja tervete vabatahtlikega läbiviidud uurimused

tõestavad, et L. fermentum ME-3 omab põhilisi tunnuseid, mille alusel ta
sobib kasutamiseks probiootikumina, kas funktsionaalse toidu või toidu-
lisandina. L. fermentum ME-3 tervistavad omadused on suunatud inimorga-
nismi antioksüdantsete kaitsesüsteemide tugevdamisele.

94
 ACKNOWLEDGMENTS
This study was carried out at the Department of Microbiology of University of
Tartu, Estonia.

The study was supported by the grant from the Estonian Science Foundation
(base fundings 0418 and 0411) and Estonian Technology Agency (funding No.
07/2002) and EU QLRT-2001-00135.

I wish to express my sincere gratitude to my supervisor Professor Marika

Mikelsaar, for giving me the opportunity to work in the field of microbiology,
for her advice and also her great support throughout my studies.

My heartest thanks to prof. Mihkel Zilmer for his support, brilliant ideas and
guidance.

I want to thank associated professor Priit Elias for technological advising.

Special thanks to Tiiu Kullisaar for her great optimism and belief in
Lactobacillus fermentum ME-3.

I thank co-authors of my publications for their valuable help and also all my
colleagues who have contributed to the completion of this thesis.

I am grateful for the advice of Professor Raivo Uibo and Professor Enn Seppet
serving as referees of my thesis.

I am sincerely grateful to Mai Laanes for her support and practical advising and
to Enely Alas for the excellent technical assistance.

My heartest thanks to Professor Irja Lutsar, Pirje Hütt, Kai Truusalu, Reet
Mändar for critical reading the manuscript and valuable comments.

Great thanks to Ester Jaigma and Irja Roots for the profound and competent
revision of the manuscript.

I express my gratitude to the whole collective of Starter St OY, especially to

Üllas Jaaska for support and patience throughout all the years.

I wish to thank all my friends for their interest and encouragement

Thanks to my mother for everything...

95
PUBLICATIONS
 Mikelsaar, M., Zilmer, M., Kullisaar, T., Annuk, H. and Songisepp, E.
Strain of microorganism Lactobacillus fermentum ME-3 as novel antimicrobial
and antioxidative probiotic.
International Patent application 2001, WO03002131
(http://ep.espacenet.com).

Annuk, H., Shchepetova, J., Kullisaar, T., Songisepp, E., Zilmer, M., Mikelsaar, M.
Characterization of intestinal lactobacilli as putative probiotic candidates.
Journal Applied Microbiology 94, 403–412 (2003).

Songisepp, E., Kullisaar, T., Hütt, P., Elias, P., Brilene, T., Zilmer, M., Mikelsaar, M.
A New Probiotic Cheese with Antioxidative and Antimicrobial Activity.
Journal of Dairy Science 87, 2017–2023 (2004).

Kullisaar, T., Songisepp, E., Mikelsaar, M., Zilmer, K., Vihalemm, T., Zilmer, M.
Antioxidative probiotic fermented goats milk decreases oxidative stress-mediated
atherogenicity in human subjects.
British Journal of Nutrition 90, 449–456 (2003).

Songisepp, E., Kals, J., Kullisaar, T., Hütt, P., Mändar, R., Zilmer, M., Mikelsaar, M.
Evaluation of the functional efficacy of a probiotic in healthy volunteers.
Nutrition Journal, submitted.
 EVALUATION OF THE FUNCTIONAL EFFICACY
OF AN ANTIOXIDATIVE PROBIOTIC
IN HEALTHY VOLUNTEERS

Epp Songisepp1†, Jaak Kals2†, Tiiu Kullisaar2†, Reet Mändar1†, Pirje Hütt1†,
Mihkel Zilmer2†, Marika Mikelsaar1*
1
Department of Microbiology, University of Tartu, 50411 Tartu, Estonia
2
Department of Biochemistry, University of Tartu, 50411 Tartu, Estonia
*
Corresponding author
†
These authors contributed equally to this work

ES: [email protected]
JK: [email protected]
TK: [email protected]
RM: [email protected]
PH: [email protected]
MZ: [email protected]
MM: [email protected]

ABSTRACT

Backround: In persons without clinical symptom it is difficult to assess an

impact of probiotics regarding its effect on health. We evaluated the functional
efficacy of the probiotic Lactobacillus fermentum ME-3 in healthy volunteers
by measuring the influence of two different formulations on intestinal
lactoflora, fecal recovery of the probiotic strain and oxidative stress markers of
blood and urine after 3 weeks consumption.

Methods: Two studies with healthy adults were performed; altogether 45

randomly allocated persons consumed either the ME-3 capsules/placebo or
fermented goat milk/goat milk in a daily of dose 9.2 to 11.8 log CFU respec-
tively for 3 weeks. The fecal lactoflora composition, fecal ME-3 recovery,
effect of the consumption on intestinal lactoflora, and oxidative stress markers
of blood (total antioxidative activity; total antioxidative status and glutathione
red-ox ratio) and urine (8-isoprostanes) was measured.

Results: ME-3 was well tolerated and a significant increase in total fecal
lactobacilli yet no predominance of ME-3 was detected in all study groups.
Fecal recovery was documented by molecular methods only in fermented milk
group, however the significant improvement of blood TAA and TAS indices

1
was seen in case of both formulations, yet glutathione re-ox ratio and urine
isoprostanes values decreased only in case of fermented by ME-3 goat milk.

Conclusions: The functional efficacy of both consumed formulations of an

antioxidative probiotic L. fermentum ME-3 is proved by the increase of the
intestinal lactobacilli counts providing putative defense gainst enteral infections
and by reduction of the oxidative stress indices of blood and urine of healthy
volunteers. In non-diseased host the probiotic health claims can be assessed by
improvement of some measurable laboratory indices of well-established
physiological functions of host, e.g. markers of antioxidative defense system.

BACKGROUND

Probiotics are defined as live microbial food supplements, which beneficially

influence human health [1;2]. Widely accepted probiotics contain different
lactic acid producing bacteria of human origin: bifidobacteria, lactobacilli or
enterococci. Nowadays the concept of functional foods, incl. probiotic food and
dietary supplements implies to their ability to beneficially influence body
functions in order to improve the state of well-being and health and reduce the
risk of disease [2,3]. The important areas of human physiology that are relevant
to functional food science according ILSI and FUFOSE (The European
Commission Concerted Action on Functional Food Science in Europe) are
besides others, the modulation of basic metabolic processes and defense against
high-grade oxidative stress [4, 5].
Human nutrition is clearly associated with oxidative metabolism, which
beside production of energy is involved in a number of vital functions of the
host. For example, under physiological conditions the reactive species
(including peroxyl radicals, nitric oxide radical, superoxide anion) figure a
crucial role in primary immune defense of the human body by phagocytic cells
against harmful microorganisms [6, 7]. On the other hand, a prolonged excess
of reactive species is highly damaging for the host biomolecules and cells,
resulting in dysbalance of the functional antioxidative network of the organism
and leading to substantial escalation of pathological inflammation. Recently
Petrof et al. showed in vitro that some probiotics protect intestinal epithelial
cells against oxidant stress also through inducing heat shock proteins known as
cytoprotectors against inflammatory cell-derived oxidants [8].
By our knowledge, no systematic studies have been performed to approve
the functional efficacy of different formulations of probiotic on the antioxi-
dative defense system of a healthy human. In our previous study Lactobacillus
fermentum ME-3 (DSM 14241) [9–11], expressed strong antimicrobial activity
against Gram-positive and Gram-negative entero- and uropathogens [12, 13].
The cells and cell lysate of L. fermentum ME-3 possessed substantial antioxi-
dative potency [14]. In an animal experiment ME-3 suppressed the excessive

2
oxidative stress reaction caused by Salmonella infection in intestinal mucosa
and thus improved the gut mucosal antioxidative status [15]. The antioxidative
effect of L. fermentum ME-3 on human body oxidative stress markers was
confirmed by our pilot study with fermented goat milk [16].
The aim present study was to evaluate the functional efficacy of the
probiotic strain L fermentum ME-3 in the human GIT of healthy volunteers. The
fecal recovery, effect of two different formulations on total fecal lactoflora and
oxidative stress markers of blood and urine were compared after 3 weeks
consumption.

METHODS

Formulations

The efficacy of two different formulations (experimental fermented goat milk

and probiotic capsules) on the human body oxidative stress markers was
evaluated.
Lactobacillus fermentum ME-3, a probiotic strain of healthy human
intestinal origin (17), has been identified by biochemical and molecular
methods [9]. The patent application has been submitted to the Estonian Patent
Agency (Application No. 0356/01PV) as well as to the International Bureau of
WIPO (Application No. WO03002131) [11]. L. fermentum ME-3 was used as
freeze-dried powder in capsulated form and in fermented milk.
Capsules. Gelatine coated capsules were manufactured by the Tallinn
Pharmaceutical Company. The freshly prepared probiotic capsules contained
9.0 log CFU of L. fermentum ME-3 per capsule in addition to 250 mg of saccha-
rose and microcellulose. Identical placebo capsules contained only saccharose
and microcellulose. All capsules were stored at +4°C.
Survival of ME-3 in capsule. Survival of ME-3 in capsule was monitored
during 12 months at +4°C. The content of one capsule was dissolved aseptically
in 2 ml of 0.9% NaCl solution. The suspension was vortexed, serially diluted
and seeded 0.1 ml on MRS agar medium (OXOID, U.K.) and incubated 48
hours at 37°C microaerobically (10% CO2). The number of colonies was coun-
ted and the viable cell count in capsule was calculated.
Experimental fermented milk. Three different lots of experimental fermented
goat milk was prepared for the 3-week trial with healthy volunteers in order to
establish the health effects of ME-3 consumption. The study group was supplied
with fresh product once a week. Experimental fermented milk was prepared as
described previously [16] by combining the probiotic strain with two supportive
lactobacilli cultures L. plantarum LB-4 and L. buchneri S-15. L. buchneri strain
S1-5 decreased the specific taste of the goat milk. L. plantarum LB-4 was
included as a strong producer of exopolysaccharides, which gives the fermented
milk a cream-like consistence and delightful acidity. The goat milk was

3
inoculated with 2% mixture of Lactobacillus strains and incubated at 37ºC for
24 hours. The product, ready to use, was cooled and stored at 4°C.
Survival of L. fermentum ME-3 in fermented goat milk. To measure the
viable cell count of ME-3 in fermented goat milk, samples were taken at the end
of fermentation (before cooling the product), and after 24h, 32h, 48h and 7 days
from the preparation, when the product was stored at 4°C. The amount of 0.5 ml
of the fermented milk was serially diluted in saline and plated on MRS agar
medium and incubated for 48 h at 37°C in microaerobic conditions.

DESIGN OF HUMAN VOLUNTEER TRIALS

Two healthy volunteer (n 45) trials, particularly open placebo controlled (OPC)
study and double blind randomised placebo controlled (DBRP) study were
carried on to evaluate the functional efficacy of L. fermentum ME-3 in the
human body. The inclusion criteria included the wish to participate, no known
health problems, and no medical conditions requiring drug therapy, no other
yoghurts or no special diets. The subjects with a history of GIT disease, food
allergy and acute infection, use of any antimicrobial agent within the last month
or use of any regular concomitant medication were excluded. The members of
the trial were daily questioned about their general welfare, intestinal function
(general welfare, gut gas production, stool frequency) and putative adverse
effects. The withdrawal criteria from the trials included acute infections during
the study. Reasons for dropout were the unwillingness to proceed with the study
or relocation to new area. The blood samples (6 ml) from the antecubital vein,
fecal and urine samples were collected before and at the end of all clinical trials.
Participants of all trials gave informed consent to the study protocols approved
by the Ethical Committee of Tartu University.
Open placebo controlled fermented goat milk trial. The study participants
were 5 men and 16 women, mean age 50 years (range 35–60). During three
weeks of the trial the study group (3 males and 13 females) consumed daily 150
ml fermented goat milk. The daily dose of probiotic Lactobacillus strain was
11.2 to 11.8 log CFU per person.
The control group (1 male and 4 females) consumed the same dose of fresh
goat milk.
Probiotic capsule trial. A DBRP study was carried out as follows. The study
group consisted of 15 men and 9 women, mean age 52 years (range 40–60)
allocated according to their wish to participate and randomly divided by an
independent person and computer program for two groups. The study group
members (8 males and 4 females) took three probiotic containing capsules (8.4
log CFU per capsule) two times daily (the daily dose 9.2 log CFU) during three
weeks. The placebo group (7 males and 5 females) received identical capsules
without the probiotic strain.

4
 Fecal samples of all participants to assess change in fecal lactoflora and the
persistence of the ingested probiotic strain were collected before and at the end
of trial. Several laboratory indices of blood and urine were measured before and
after the consumption of ME-3. Here we report on changes in human body
oxidative stress markers as total antioxidative activity (TAA), total
antioxidative status (TAS) and glutathione red-ox ratio (GSH/GSSG) from
blood serum and 8-isoprostanes in urine.

Microbiological analyses of feces

The total count of lactobacilli and the count of L. fermentum were evaluated in
fecal samples. The fecal samples were collected at day 0 and 21 in both trials.
Samples were kept at –80°C before analyzed. Serial dilutions (10–2–10–9) of the
weighed fecal samples were prepared with phosphate buffer (pH 7.2) and 0.05
ml of aliquots was seeded onto MRS agar medium [17]. The plates were
incubated at 37°C for 4 days microaerobically in 10% CO2 environment
(incubator IG 150, Jouan, France). The catalase negative colonies were selected
on the basis of typical for LAB colony morphology, cells microscopy and Gram
staining.
The count of Lactobacillus species was expressed in log10 colony forming
units per gram feces (log10 CFU/g) and percentage (relative share) in the total
count of lactobacilli. The detection level of lactobacilli was a 3.0 log CFU/g
feces.
The relative amount of L. fermentum, colonizing the gastrointestinal tract of
persons in the study groups was expressed as a proportion of the total count
(%), using the Bioquant program [18]). The program gives output data for every
microorganism as an absolute count (log10 CFU/g) and their percentage in the
total count with its normal values.

AP-PCR TYPING

The putative ME-3 isolates were typed by arbitrarily primed polymerase chain
reaction (AP-PCR). Genomic DNA was extracted from 24h old cultures,
cultivated on MRS agar microaerobically with the QIAamp DNA Mini Kit 50
(QIAGEN GmbH., Hilden, Germany) according to the manufacturers
instructions. AP-PCR typing was done with two primers: ERIC1R (5'-
ATGTAAGCTCCT GGGGATTCAC-3') and ERIC2 (5'-
AAGTAAGTGACTGGGGTGAGCG -3') (DNA Technology A/S, Aarhus,
Denmark). A 30 µl volume of reaction mixture consisted of 10xPCR buffer
(Fermentas, Vilnius, Lithuania), 2.5 mM MgCl2 (Fermentas, Vilnius,
Lithuania), 200µM deoxynucleoside triphosphate mixture (dATP, dGTP, dTTP
and dCTP, Amersham Pharmacia Biotech, Freiburg, Germany) 0,60µg of each
primer and 2.5U Taq DNA Polymerase (Fermentas, Vilnius, Lithuania,) and 5

5
µl of extracted DNA according to Matsumiya et al. [19]. The PCR mixture was
subjected to thermal cycling 35 cycles of denaturation at 95°C for 1 min,
annealing at 35°C for 1 min, and extension at 74°C for 2 min, with a final
extension at 74°C for 5 min with the PTC-200 thermal cycler (Eppendorf AG,
Hamburg, Germany). The PCR products were separated by electrophoresis in a
horizontal 2% agarose gel containing 0.1 µl/ml ethidium bromide in Tris-acetic
acid–EDTA (TAE) buffer (40mM Tris, 20mM boric acid, 1mM EDTA, pH 8.3)
(Bio-Rad Laboratories, Hercules, USA) at constant voltage of 120V. A 1kb
ladder (GeneRuler, Fermentas, Vilnius, Lithuania) was used as a base pair size
marker. The banding patterns of isolates were visualized with UV light and
compared with that of L. fermentum ME-3 strain.

Measurement of human body oxidative stress status

In urine the changes of the oxidative stress marker 8-isoprostanes concent-
rations (ng/ml) were assessed by a competitive enzyme-linked immunoassay
(ELISA) (BIOXYTECH 8-Isoprostane Assay, Cat No 21019) as described
previously [16].
Blood serum was analysed for total antioxidative activity TAA, total
antioxidative status TAS and GSSG/GSH. TAA of the serum was assessed by
the linolenic acid test (LA-test) described previously [16]. This test evaluates
the ability of the sample to inhibit lipid peroxidation. TAS of the serum was
measured with a commercially available kit (TAS, Randox Laboratories Ltd.
Ardmore, UK) as described elsewhere [16], water-soluble vitamin E (Trolox)
serving as a standard. This method is based on the inhibition of the absorbance
of the ferrylmyoglobin radicals of 2,2’-azinobis-ethylbenzothiazoline 6-
sulfonate (ABTS+) generated by activation of metmyoglobin peroxidase with
H2O2.
The cellular oxidative stress markers as total glutathione and oxidized
glutathione were measured using the method of Griffith [20] as described
elsewhere [16]. The glutathione content was calculated on the basis of a
standard curve generated with known concentration of glutathione. Amount of
GSH (µg/ml) was calculated as a difference between the total glutathione and
GSSG (total glutathione – GSSG). The glutathione red/ox ratio was expressed
as GSH/GSSG.

STATISTICAL ANALYSIS

The computer program Sigma Stat for Windows 2.0 (Jandel Coprporation,
USA) was applied. The counts of fecal lactoflora were compared by using
Student’s t-test and Mann-Whitney rank sum test. Changes in oxidative stress
markers of blood sera (TAA, TAS and glutathione red-ox ratio) and urine
(8-isoprostanes) were evaluated by Student’s t-test, paired t-test and Mann-

6
Whitney rank sum test. The choice of tests was made automatically according to
the distribution of the data. Both microbial and biochemical markers were given
as mean and standard deviation.
One-way ANOVA test was performed to compare the effect of different
formulation on TAA, TAS and fecal lactoflora parameters.
Differences were considered statistically significant if the value was p <0.05.

RESULTS

Survival of ME-3 in formulations

In capsule after approximately 1-log drop after one week from the production of
the capsules, the viable count of the probiotic strain remained stable at the level
of 8.4 log CFU per capsule. Additional results have shown at +4°C the stability
of the freeze-dried capsulated culture at least 17 months from the production.
In fermented goat-milk the cell count of the probiotic strain varied insigni-
ficantly from 9.0 to 9.7 log CFU/ml from one preparation to the other. The
viable count of ME-3 in the fermented goat milk was found to remain stable at
least during 7 days of storage at 4°C.

Human volunteer trials

No dropouts were registered during volunteer trials, yet one participant was
withdrawn from the probiotic capsule trial due to acute respiratory viral
infection. Besides, no adverse affects in general welfare or changes in GI
functionality were assessed during the trial.
Changes in total LAB count. The consumption of both ME-3 fermented milk
and ME-3 capsule significantly increased the total count of lactobacilli in feces
as compared to the initial levels (Fig. 1). In opposite, in the group of volunteers
consuming non-fermented goat-milk there was even a decrease in total LAB
counts during the 3-week trial and no changes were found in capsule placebo
group. Additional increase of lactobacilli counts was found only in persons
consuming fermented goat milk.
Recovery of the probiotic strain. In goat milk group L. fermentum as a
species appeared in fecal samples of all individuals (n=16) after consumption of
fermented goat milk (Table 1). The AP-PCR confirmed the recovery of ME-3 in
the feces of all study group members (Fig. 2). However, different trials the
strain did not perform the predominant Lactobacillus species in total lactobacilli
count in participants in (Table 1), though there was a tendency for increase in
L. fermentum counts. In the probiotic capsule trial the strain ME-3 was not
detectable between L. fermentum isolates by AP-PCR.

7
 Antioxidative health effect of ME-3. The consumption of ME-3 in both
formulations had a positive effect on the blood oxidative stress markers as TAA
and TAS (Fig. 3). Consumption of goat milk and fermented goat milk enhanced
TAA and TAS in the study and control group. There was a significant
additional increase (6% and 9% respectively) in both indices in the fermented
goat milk group. Significant increase in TAA and TAS values occurred during
the consumption of the probiotic strain in capsulated form. No changes were
detected in the placebo group. Additional effect of probiotic consumption in
capsulated form was 4% for TAA and 2.5 % for TAS.
The effect of goat-milk consumption on the TAA and TAS values was
significantly higher (p<0.001) than by the consumption of the capsulated
probiotic (Fig. 2).
The decrease of the glutathione red-ox ratio was significant in both groups:
the study group (from 0.15±0.01 to 0.11±0.04 µg/ml, p<0.01) and control (from
0.14±0.03 to 0.11±0.02 µg/ml, p<0.01) in the goat milk trial (Kullisaar et al.
2003). The fermented goat milk containing L. fermentum ME-3 had no
statistically significant additional effect. When the probiotic was consumed in
capsulated form, no significant decrease was noticed in the glutathione red-ox
ratio. The additional effect of ME-3 fermented goat milk consumption was 6%.
Compared with the baseline values, the consumption of ME-3 in goat milk
reduced urine 8-isoprostanes concentrations from 5.5±0.4 before to 5.0±0.5
ng/ml after the treatment (p<0.01) [16]. No changes in urine 8-isoprostanes
concentrations were seen between baseline and end of the capsule trial (from
3.59±1.3 to 3.0±1.6 ng/ml).

DISCUSSION

We have assessed the functional efficacy of the antimicrobial and antioxidative

probiotic L. fermentum ME-3 for healthy host. First of all the safety of the
L. fermentum strain ME-3 was confirmed as no adverse side effects were
registered in volunteers. Even relatively high (>1011 CFU) doses of consumed
ME-3 had no negative impact on the hosts’ general well-being. Lactobacillus
fermentum as species, used in various food applications, has a well-established
history of safe use and is evaluated as GRAS according to the Food and Drug
Administration of the USA [21].
Second, a clear improvement of laboratory indices of antioxidative defense
system of a healthy host was documented, using both formulations as fermented
by L. fermentum ME-3 goat-milk and probiotic capsules. This effect was
simultaneous with the increase of intestinal lactoflora of healthy volunteers
even without necessity for fecal recovery of the strain. In the human population,
persons without clinical symptoms have still a quite different health status,
including stability, capacity and potency of antioxidative defence to counteract
sufficiently to oxidative stress-caused adverse effects [7]. If a probiotic is able

8
to exhibit a positive functionality on oxidative stress-related indices, it helps
both to stabilize and promote the potency of the whole body antioxidative
defence system in subclinical situations without disease symptoms. That in turn
may have an impact for lowering the risk of atherosclerotic damage of blood
vessels associated with several cardiovascular and neurodegenerative diseases
[22–24].
In our study of healthy volunteers for validation of the antioxidative
functionality of probiotic, four well-known oxidative stress markers of blood
and urine were chosen. Urine 8-isoprostanes reflect the whole body oxidative
stress-load, normally present at low concentrations in human body fluids [25].
We found that both formulations of ME-3 lowered their concentration as
compared to individual baseline values in our clinical settings. This may be
understood as indirect evidence for suppression of LDL oxidation [26, 27]. The
state of the lipid fraction (including also LDL) in the antioxidative defence
system of the blood is evaluated by TAA. TAS on the other hand reflects more
the antioxidativity of the water-soluble fraction of the human blood. Among the
measured blood sera markers both the TAA and TAS values were also reduced
in the two different study groups. However, there was found a significantly
lower improvement of TAA and TAS values in cases of capsule than fermented
goat-milk where the recovery o the strains was assessed by AP-PCR. The
AP-PCR is an easily performed technique very effective for identification of
potential ME-3 fecal isolates.
Similarly, the reduction of the glutathione red-ox ratio was detected after the
consumption of fermented by ME-3 goat-milk but not with the capsule. The
crucial non-enzymatic cellular antioxidant is GSH [28] present in the millimolar
range mainly in the red blood cells, liver, pancreas, kidneys, spleen, eyes, lungs
and intestinal cells [29]. The oxidized form of glutathione becomes even at low
concentrations toxic, and therefore in the cells the glutathione red-ox ratio is
kept as low as possible. In the case of inflammation this balance is shifted
towards the oxidized form, indicating non-physiological intracellular oxidative
stress. Thus, our study shows that there is a good association between the mode
of formulation of probiotic and expression of its functional properties inside the
healthy host. Particularly, the explanation for more expressed positive shifts in
oxidative stress markers of former volunteers could be due to the synergistic
effect of the probiotic and the substrate. Milk is not just a carrier for the
probiotic Lactobacillus strain, but contains natural “lactogenic” factors like
lactose, minerals, vitamins and other components that enhance the metabolic
activity of ingested probiotic strain in GIT. Besides, a variety of bioactive
peptides (e.g. casomorphins, lactorphins, casokinins, etc.) revealed in milk [30–
32] may have the antioxidative potency. This was proved by some antioxidative
effect also in persons consuming non-fermented goat milk. The composition of
goat milk differs from cow milk, containing several biomolecules, which by
consumption may also contribute to some additional antioxidative effect, as
discussed elsewhere [16]. Therefore, the provisional FAO regulations [33]

9
suggesting the need for health claims by specified formulations of probiotic
seem to be of the utmost importance.
Additionally, in our study with experimental fermented milk the average
daily dose of L. fermentum ME-3 being 11.5 logs CFU was clearly higher than
that of capsule (max 9.5 log CFU). It is possible that the dose excesses the
amount of bacteria necessary for interacting with intestinal mucosa and the
unattached lactobacilli are excreted with faeces. The finding of Saxelin and
colleagues confirmed that the fecal recovery of the probiotic strain started from
the consumption of more than 9.0 log CFU daily doses of capsulated LGG [34].
To our surprise, in the present study the similar dose did not result in faecal
recovery of the strain.
It is possible that the ME-3 strain germinated mainly in some upper parts of
intestinal tract where the advantageous conditions for survival and metabolic
activity of probiotic lactobacilli were present. Using molecular tools, Marteau et
al. showed that lactobacilli figuring only 7% of fecal microflora performed up
to 30% of microbial communities in human colon [35]. If administered in lower
quantities as in case of capsule trial, ME-3 did not reach the detectable level in
fecal samples. Yet, its presence in gut was proved by the positive antioxidative
health effect in blood but not in urine. Therefore it is understandable that the
higher load of metabolically active probiotic bacteria in goat-milk resulted also
in their fecal recovery and the highest impact on the oxidative stress indices.
Moreover, in our study the positive impact of ME-3 consumption on the host
lactoflora was proved by the increase of fecal lactobacilli counts in all
participants of human volunteer studies. In experimental settings the high
counts of intestinal lactobacilli have been shown as an important defensive
factor against enteral infections [36, 37]. Though up to now the period of
consumption of probiotics has not been defined, the 3-week ingestion of
fermented goat-milk and capsule seemed enough for reaching the aims.
It is important to mention that after consumption of ME-3, a strain with high
antagonistic activity, neither the species nor the strain predominated among
total lactoflora. This shows a well-granted microbial balance inside the gut,
which cannot be disturbed by high load of probiotic bacteria. Apparently, the
interconnected advanced metabolism of large gut microbiota keeps the
proportions of different species quite stable. Some other investigators have
obtained similar results showing the proportional increase of different microbial
populations (bifidobacteria, coliforms) after administration of Lactobacillus sp.
probiotic [38, 39].
Thus, the functional efficacy of different formulations of antiinfectious and
antioxidative probiotic L. fermentum ME-3 were proved both by the increase of
the lactobacilli counts providing putative defense against infectious agents in
gut and by reduction of the oxidative stress indices of blood and urine of
healthy volunteers. Further, Phase III studies evaluating the efficacy of ME-3 as
adjunct to conventional therapy in patients with atherosclerotic damages and a
high-grade oxidative stress are ongoing.

10
CONCLUSIONS

In non-diseased host, the probiotic health claims can be assessed by

improvement of some measurable laboratory indices of well-established
physiological functions of organism. In our case, the possibility for
augmentation of the antioxidative defence system by the probiotic L. fermentum
ME-3 in normal population can be proposed.

COMPETING INTERESTS

Marika Mikelsaar, Mihkel Zilmer, Tiiu Kullisaar, Heidi Annuk (Hynes) and
Epp Songisepp are sharing the Estonian patent application: no. EE 2001 00356
29.06.01 and International Patent application: no. WO03002131.

AUTHORS’ CONTRIBUTIONS

Epp Songisepp, and Pirje Hütt have been in charge of the microbiological
analysis. The former has been also in charge of analysing the results and writing
of the manuscript. Jaak Kals was responsible for performance and management
of the volunteer trials. Tiiu Kullisaar has been in charge of the biochemical
analysis and writing the manuscript. Mihkel Zilmer has conducted the
biochemical estimations and writing the manuscript. Reet Mändar has been in
charge of the molecular analysis and revising the manuscript. Marika Mikelsaar
is the main conductor of the L. fermentum ME-3 research; for this paper she has
been in charge of the clinical trial design and writing the manuscript.

ACKNOWLEDGEMENTS

The study was supported by a grant from the Estonian Science Foundation (base
funding 0418 and 5327) and Estonian Technology Agency funding 01103 and
EU QLRT-2001-00135.
We are sincerely grateful to Dr. Irja Lutsar for critical reading of the
manuscript, Dr. Heidi Hynes, Mrs. Eha-Maie Laanes and Miss Enely Alas for
excellent technical assistance.

11
REFERENCES

1. McFarland L: A review of evidences of health claims for biotherapeutic agents.

Microb Ecol Health Dis 2000, 12: 65–76.
2. Salminen S: Human studies on probiotics: aspects of scientific documentation.
Scand J Nutr 2001, 45: 8–12.
3. Hardy G: Nutraceuticals and functional foods: introduction and meaning.
Nutrition 2000, 16: 688–697.
4. ILSI Europe: Scientific concepts of functional foods in Europe: consensus
document. Br J Nutr 1999, 81: 1–27.
5. Ashwell M: Concepts of functional foods. In ILSI Europe Concise monograph
series, ILSI Press; 2002, 40 S: 7894–10.
6. Diplock AT, Charleux J-L, Crozier-Willi G, Kok FJ, Rice-Evans C, Roberfroid M,
Stahl W and Viňa-Ribes J: Functional food science and defence against reactive
oxidative species. Br J Nutr 1998, 80: 77–112.
7. Halliwell B and Gutteridge JMC: Free radicals in biology and medicine. New York:
Oxford University Press; 1999.
8. Petrof EO, Kojima K, Ropeleski MJ, Musch MW, Tao Y, De Simone C and Chang
EB: Probiotics inhibit nuclear factor kappa-B and induce heat shock proteins in
colonic epithelial cells through proteasome inhibition. Gastroenterology 2004,
127: 1474–1487.
9. Annuk H, Vainumäe I, Aimla K, Ausmees K, Kullisaar T, Kask S and Mikelsaar M:
Antagonistic activity of Lactobacilli against urinary Escherichia coli. Microecol
Therapy 1999, 28: 169–175.
10. Mikelsaar M, Annuk H, Shchepetova J, Mändar R, Sepp E and Björkstén B:
Intestinal Lactobacilli of Estonian and Swedish children. Microb Ecol Health
Dis 2002, 14: 75–80.
11. Mikelsaar M, Zilmer M, Kullisaar T, Annuk H and Songisepp E: Strain of
microorganism Lactobacillus fermentum ME-3 as novel antimicrobial and
antioxidative probiotic. International Patent application 2001, WO03002131
[http://ep.espacenet.com]
12. Annuk H: Selection of medicinal plants and intestinal lactobacilli as
antimicrobial components for functional foods. PhD thesis. Tartu University,
Microbiology Department; 2002.
13. Songisepp E, Kullisaar T, Hütt P, Elias P, Brilene T, Zilmer M and Mikelsaar M: A
New Probiotic Cheese with Antioxidative and Antimicrobial Activity. J Dairy
Sci 2004, 87: 2017–2023.
14. Kullisaar T, Zilmer M, Mikelsaar M, Vihalemm T, Annuk H, Kairane C and Kilk A:
Two antioxidative lactobacilli strains as promising probiotics. Int J Food
Microbiol 2002, 72: 215–224.
15. Truusalu K, Naaber P, Kullisaar T, Tamm H, Mikelsaar R-H, Zilmer K, Rehema A,
Zilmer M and Mikelsaar M: The influence of antibacterial and antioxidative
probiotic lactobacilli on gut mucosa in a mouse model of Salmonella infection.
Micr Ecol Health Dis 2004, 16: 180–187.
16. Kullisaar T, Songisepp E, Mikelsaar M, Zilmer K, Vihalemm T and Zilmer M:
Antioxidative probiotic fermented goats milk decreases oxidative stress-
mediated atherogenicity in human subjects. Br J Nutr 2003, 90: 449–456.

12
17. Mikelsaar M, Lenzner A and Goljanova LA: Methods for the estimation of the
quantitative composition of fecal microflora. Laboratornoje delo 1972, 1: 41–45.
18. Mändar R, Mändar H and Mikelsaar M: Bioquant−a program for evaluation of
fecal microbiocenosos. Clin Chem Lookout 1992, 56: 46.
19. Matsumiya Y, Kato N and Watanabe K: Molecular epidemiological study of
vertical transmission of vaginal Lactobacillus species from mothers to newborn
infants in Japanese, by arbitrarily primer polymerase chain reaction. J Infect
Chemother 2002, 8: 43–49.
20. Griffith OW: Determination of glutathione and glutathione disulfide using
glutathione reductase and 2-vinylpyridine. Anal Biochem 1980, 106: 207–212.
21. Food and Drug Administration: Rules and regulations.
[http://a257.g.akamaitech.net/7/257/2422/15nov20040800/edocket.access.gpo.gov/2
004/pdf/04-25243.pdf].
22. Witzum JL: The oxidation hypothesis of atherosclerosis. Lancet 1994, 344: 793–
795.
23. Bogdanovic N, Zilmer M, Zilmer K, Rehema A and Karelson E: The Swedish
APP670/671 Alzheimer’s disease mutation: The first evidence for strikingly
increased oxidative injury in the temporal inferior cortex. Dementia Geriatric
Cognitive Disorders 2001, 12: 364–370.
24. Pihl E, Zilmer K, Kullisaar T, Kairane C, Pulges A and Zilmer M: High-sensitive
C-reactive protein level and oxidative stress-related status in former athletes in
relation to traditional cardiovascular risk factors. Atherosclerosis 2003, 171:
321–326.
25. Hata AN and Breyer RM: Pharmacology and signaling of prostaglandin
receptors: Multiple roles in inflammation and immune modulation. Pharmacol
Ther 2004, 103: 147–66.
26. O’Brien RC, Luo M, Balazs N and Mecuri J: In vitro and vivo antioxidant
properties of gliclazide. J Diabetes Complic 2000, 14: 201–206.
27. Sanchez-Moreno C, Dashe JF, Scott T, Thaler D, Folstein MF and Martin A:
Decreased levels of plasma vitamin C and increased concentration of
inflammatory and oxidative stress markers after sroke. Stroke 2003, 35: 163.
28. Pastore A, Federici G, Bertini E and Piemonte F: Analysis of glutathione:
implication in redox and detoxification. Clinica Chimica Acta 2003, 333: 19–39.
29. Meister A and Anderson ME: Glutathione. Ann Rev Biochem 1983, 52: 711–760.
30. Meisel H and Bockelmann W: Bioactive peptides encrypted in milk proteins:
proteolytic activation and tropho−functional properties. In Proceedings of the
sixth symposium on lactic acid bacteria: genetics, metabolism and applications.
Edited by Konings, WN, Kuipers OP and Huis in`t Veld J. Veldhoven: Kluver
Academic Publishers; 1999: 207−215.
31. Bos C, Gaudichon C and Tomé D: Nutritional and physiological criteria in the
assessment of milk protein quality for humans. J Am Colledge Nutr 2000, 19:
191–205.
32. Boland M, MacGibbon A and Hill J: Designer milks for the new millennium.
Livestock Prod Sci 2001, 72: 99–109.
33. Food Agriculture Organization: Guidelines for the evaluation of probiotics in
food. [http://www.who.int/foodsafety/fs_management/en/probiotic_guidelines.pdf].

13
34. Saxelin M, Elo S, Salminen S and Vaapatalo H: Dose response colonisation of
faces after oral administration of Lactobacillus casei strain GG. Micr Ecol
Health Dis 1991, 4: 209–214.
35. Marteau P, Pochart P, Dore J, Bera-Maillet C, Bernalier A and Corthier G:
Comparative study of bacterial groups within the cecal and fecal microbiota.
Appl Env Microb 2001, 67: 4939–4942.
36. Kelleher SL, Casas I, Carbajal N and Lonnerdal B: Supplementation of infant
formula with the probiotic Lactobacillus reuteri and zinc: impact on
entericinfection and nutrition in infant rhesus monkeys. J. Pediatr Gastr Nutr
2002, 35: 162–168.
37. Cruchet S, Obregon MC, Salazar G, Diaz E and Gotteland M: Effect of the
ingestion of dietary product containing Lactobacillus johnsonii La1 on
Helicobacter pylori colonization in children. Nutrition 2003, 19: 716–721.
38. Sepp E, Mikelsaar M and Salminen S: Effect of administration of Lactobacillus
casei strain GG on the gastrointestinal microbiota of newborns. Micr Ecol
Health Dis 1993, 6: 309–314.
39. Morelli L, Zonenschain D, Callegari ML, Grossi E, Maisano F and Fusillo M:
Assessment of a new symbiotic preparation in healthy volunteers: survival,
persistence of probiotic strains and its effect on the indigenous flora. Nutr J
2003, 2:11 [http://www.nutritionj.com/content/2/1/11].

14
Table 1. Changes in fecal recovery of L. fermentum during healthy human volunteer
trials
L. fermentum
Groups * † ‡
Prevalence (%) Count (log10) Proportion (%)
Day 0 Day 21 Day 0 Day 21 Day 0 Day 21
Goat milk trial, 25 (4/16) 100 (16/16) 7.0±0.7 7.3±1.4** 21 13
ME-3 (n=16)
Control (n=5) – 20 (1/5) – 3.6 – 28
Capsule trial, 16.7(2/12) 33.3 (2/12) 4.3±0.5 5.8±1.6 4 9
ME-3 (n=11)
Placebo (n=12) 25 (3/12) 16.7 (2/12) 6.3±2.5 8.0±1.6 11 19
*
Percentage of subjects with fecal L. fermentum inside the group
** Significantly different from the pre-treatment values (paired t-test): p<0.001
†
Median value± SD
‡
Proportion of L. fermentum among fecal LAB

15
 Study group control study group placebo

12
‡

10
* * *

8
CFU log 10

1 2 1 2 1 2 1 2
0
Goat milk trial Probiotic capsule trial

Figure 1. Increase of total fecal counts of lactobacilli in healthy volunteers consuming

of ME-3 in fermented goat milk and probiotic capsule.
1 – Day 0, 2 – Day 21
Significantly different from pre-treatment values (Student’s t-test): * p<0.05;
Significantly different from control (Student’s t-test): ‡ p=0.01

M 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

Figure 2. Confirmation for the survival of L. fermentum ME-3 in GIT in subjects

receiving ME-3 fermented goat milk by AP-PCR in a horizontal 2% agarose gel.
From the left: M – molecular weight marker, Line 2 – ME-3, Line 3…17 – ME-3 like profiles
from feces of goat milk trial study group participants

16
 Study group control study group placebo
a)
60 ‡

** ** **
50
TAA (%)

40

30
1 2 1 2 1 2 1 2

Goat milk trial Probiotic capsule trial

2 Study group control study group placebo

b)

1.5
** ** *
TAS (mmol/l)

0.5

1 2 1 2 1 2 1 2

0
Goat milk trial Probiotic capsule trial

Figure 3. Effect of ME-3 consumption in fermented goat milk and capsules on human
blood oxidative stress markers a) TAA (%) and b) TAS (mmol/l)

1 – Day 0, 2 – Day 21
Significantly different from pre-treatment values: *p<0.05 (paired t-test); **p≤0.01 (Student’s
t-test and paired t-test);
ME-3 goat milk effect different from the effect of the ME-3 in capsule-form (ANOVA): ‡p≤0.001

17
 CURRICULUM VITAE
Epp Songisepp

Citizenship: Estonian
Born: November 17, 1965, Võru
Address: Ravila 19, 50411 Tartu, Estonia
Phone: +372 374 175
Fax: +372 374 172
E-mail: [email protected]

Education

1973–1984 Võru Secondary School No 1.

1984–1989 University of Tartu, Faculty of Biology and Geography, Biology
1999–2001 University of Tartu, Medical Faculty, Department of Micro-
biology M. Sci student of biomedicine

Special courses

2001 4 months in Japan, Osaka Municipal Technical Research Institute

Professional employment

1989–1990 Estonian Biocentre, technician

1990–1994 Ldt Estar Strip, microbiologist
1994– Starter ST OÜ, microbiologist and production manager
2002–2005 University of Tartu, Medical Faculty, Department of Micro-
biology, extraordinary research fellow
2005– R/D Centre of Healthy Milk Products Ltd, research fellow

Scientific work

The main subject of the research work has concerned starter microbes and
probiotic lactobacilli, their technological and functional properties. 6 scientific
publications and 15 presentations at the congresses

171
 CURRICULUM VITAE
Epp Songisepp

Kodakondsus: Eesti
Sünniaeg ja koht: November 17, 1965, Võru
Aadress : Ravila 19, 50411 Tartu, Eesti
Tel: +372 374 175
Fax: +372 374 172
E-mail: [email protected]

Haridus

1973–1984 Võru I keskkool

1984–1989 Tartu Ülikool, bioloogia-geograafia teaduskond, bioloogia
1999–2001 Tartu Ülikool, arstiteaduskond, mikrobioloogia instituut,
magistrant

Erialane enesetäiendus

2002 4 kuud Jaapan, Osaka Munitsipaalne Tehnikaülikool

Erialane teenistuskäik

1989–1990 Eesti Biokeskus, laborant

1990–1994 AS Estar Strip, mikrobioloog
1994– Starter ST OÜ, mikrobioloog, tootmisjuht
2002–2005 Tartu Ülikool, Arstiteaduskond, mikrobioloogia instituut, era-
korraline teadur
2005– OÜ Tervisliku Piima Biotehnoloogiate Arenduskeskus, teadur

Teadustöö

Peamiseks uurimisvaldkonnaks on juuretised ja probiootilised laktobatsillid

ning nende tehnoloogilised ja funktsionaalsed omadused
6 teaduspublikatsiooni, 15 konverentsiteesi

172