Fulltext - ETH E-Collection

Research Collection

Doctoral Thesis

Molecular identification and applied genetics of propionibacteria Author(s): Dasen, Gottfried Heinrich Publication Date: 1998 Permanent Link: https://doi.org/10.3929/ethz-a-002055677

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library

Diss. ETH

«*3

Diss. ETHNr. 12981

Molecular Identification and

Applied

Genetics of

Propionibacteria

A dissertation submitted to the

SWISS FEDERAL INSTITUTE OF TECHNOLOGY ZURICH

(ETHZ)

for the

degree

of

Doctor of Technical Sciences

presented by Gottfried Heinrich Dasen

Dipl. Lm.-Ing.

ETH

born June 25,1966

citizen of Dübendorf ZH and Täuffelen BE

accepted

on

the recommendation of

Prof. Dr. Michael

Prof. Dr. Alexander Dr. Leo

von

Teuber,

examiner

Graevenitz, co-examiner

Meile, co-examiner

Zürich, 1998

Herzlichen Dank

an:

Meine Mutter, fur ihr Verständnis, ihren Glauben Laufe meiner gesamten

er

Vergabe der

Verfügung stellte und mich

Dr. A.

von

zu

jetzigen

er

und

hat, mir seine fachlichen Kompetenz

zur

weiteren Versuchen angeregt hat.

Graevenitz für seine

Korreferates und für seine kritische

Den

Arbeit und die grosse Autonomie die

Stunden, die ich in seinem Büro verbracht habe und in

mich immer wieder motiviert

Herrn Prof.

im

Unterstützung

Verwirklichung gewährt hat.

Herrn Dr. Leo Meile für die vielen

denen

mich und ihre

Ausbildung.

Herrn Prof. Dr. M. Teuber für die

mir bei deren

an

Bearbeitung

ehemaligen „G-Stock

Bereitwilligkeit

zur

Übernahme des

dieser Arbeit.

Leuten" für Ihre

Kollegialität,

fürs tolle

Arbeitsklima, für den gegenseitigen Gedankenaustausch und noch für unzählige andere

Dinge.

Den beiden

Diplomandinnen Jitka Smutny

betreuten Semestranden für ihre

Projektes geleistet

ausgezeichnten Arbeiten, die

von

mir

sie im Rahmen meines

haben.

Frau Dr. Denise Fessier und der

gestellten

und Daniela Guldimann und allen

Forschungsanstalt Liebefeld

flir die mir

zur

Verfügung

Stämme.

Dem schweizerischen Nationalfonds für die

Schwerpunktprogramms Biotechnologie

Finanzierung

2-77-968-94.

meiner Arbeit im Rahmen des

I

Table of Contents

Summary

1

Zusammenfassung

3

1. Introduction 1.1 A short

7

history

of Propionibacteria

1.2 Characteristics and

1.2.1 Classical

occurrence

7

of Propionibacteria

Propionibacteria Propionibacteria

7

1.2.2 "Cutaneous"

1.3

Taxonomy

and its relevance in food

1.4

Taxonomy

of the

8

microbiology

Propionibacteria

1.5.1 "Selective" media 1.5.3

genetics

1.6.1 Plasmids and

11

11

"classical" methods

using Identification by means

1.6 Molecular

9 10

1.5 Isolation and identification of Propionibacteria

1.5.2 Identification

7

of DNA

11

analysis

12

of Propionibacteria

13

phages

13

1.6.2 Transformation of Propionibacteria

14

1.6.3 Genome size and

14

1.6.4 Patent

15

sequenced genes of Propionibacteria applications claiming transformation of Propionibacteria 1.6.5 Mutagenesis of Propionibacterium strains

1.7 Aim of this thesis

16 17

2. Material and methods 2.1 Bacterial strains and

19

plasmids

19

2.2 Strain maintenance and control

20

2.2.1 Growth Conditions

20

2.2.2 Catalase

21

activity

2.2.3 KOH test

21

2.2.4 Nitrate reduction

21

2.2.5

22

Carbohydrate

fermentation patterns

using the API-CH system 2.3 General DNA isolation and analysis procedures 2.3.1 Isolation of total chromosomal and plasmid DNA 2.3.2 Isolation of PCR-grade DNA 2.3.3 Small scale plasmid DNA preparation 2.3.4 2.3.5

2.3.6 2.3.7

scale

DNA isolation from E. coli

Large plasmid Large scale plasmid DNA isolation Agarose gel electrophoresis Quantification of DNA

2.3.8 Restriction endonuclease

2.3.9 Pulsed field 2.4

digests gel electrophoresis

Polymerase chain reactions PCR-primers

from

of DNA

Propionibacteria

22

22 22

22 23

23 23 24 24

24 25

2.4.1

25

2.4.2 Standard PCR conditions

25

2.4.3

26

2.4.4

Multiplex-PCR (MPCR) Long PCR conditions

27

Table of Contents

2.5

Labeling

2.5.1

II

of DNA

Labeling

probes

27

of oligonucleotides with

2.5.2 Random

[y-32P] ATP

27

primed labeling labeling of oligonucleotide-probes with 2.5.4 Labeling of DNA-fragments with DIG-dUTP 2.6 Hybridization techniques

of DNA-fragments with

2.5.3 Terminal

[a-32P]dATP

27

DIG-ddUTP

27 27 28

2.6.1 Southern

28

2.6.2

28

blotting Colony hybridization 2.6.3 Hybridizations with DNA probes 2.7 Analysis of Propionibacteria isolated from dairy products 2.7.1 Isolation of Propionibacteria from cheese 2.7.2 Species determination 2.7.3 Screening for plasmids and plasmid-encoded properties

28 29

29 29 29

2.7.4 Determination of antibiotic resistance patterns 2.7.5 Curing experiments with plasmid containing Propionibacteria

29 30

2.8 DNA

sequencing and analysis techniques Cloning of DNA fragments using pUC18 and E. coli XLl-Blue 2.8.2 Cloning of PCR fragments 2.8.3 Purification of PCR-products for sequencing reactions 2.8.4 Cycle sequencing 2.8.5 Sequence data analysis and phylogenetic calculations 2.9 Conjugation between Propionibacteria and other genera 2.9.1 Selective supplements for the identification of transconjugants 2.9.2 Adaptation of Propionibacteria to high concentrations of antibiotics 2.9.3 Conjugation with the filter-mating technique 2.10 Electrotransformation of Propionibacteria 2.8.1

3. Results

3.2 Identification of Propionibacteria

using molecular methods

3.2.3 Construction of a

31 32

33 33 33 33 34

specific

gene

probe

38 41

for the genus

Propionibacterium Multiplex PCR (MPCR)

43 44

3.2.5 Differentiation and identification of Propionibacteria based

16S-23S rDNA spacer region 3.2.6 Discrimination between the P. freudenreichii

on

the

46

subspecies

51

Genetics'

Analysis

3.3.1

31

38

Sequencing of 16S rRNA genes of Propionibacteria 3.2.2 Classification of isolates from dairy products

3.3

31

37

3.2.1

'Applied

31

37

3.1 Strain maintenance and control

3.2.4

31

of Propionibacterium

Screening

for

plasmids

in

plasmids

Propionibacteria Propionibacteria DNA-DNA hybridization assays within Propionibacterium plasmids Characterization of plasmid pLMElOl Characterization of plasmid pLME108 Electroporation of E. coli XLl-Blue with derivatives of pLME108

55 55

3.3.2 Antibiotics and

56

3.3.3

58

3.3.4 3.3.5 3.3.6

60 63 67

Ill

Table of Contents

3.4 Gene transfer

68

3.4.1 Test of selective 3.4.2 3.4.3

supplements for the isolation of Propionibacteria Conjugation experiments between other genera and Propionibacteria Electroporation of Propionibacteria

4. Discussion

4.2 Identification of Propionibacteria

using

73

molecular methods

4.2.1 Identification to genus level 4.2.2 Identification to species level

75

subspecies

4.2.4 Identification of strains

Multiplex

PCR:

a

approach

to

identify Propioni¬ 77

Phylogenetic analysis

of the genus

Propionibacterium

4.4.3

82

pLME108 Electrotransformation of Propionibacteria Conjugation experiments

pLMElOl

and

4.5 Selectable and transferable markers 4.5.1 Selectable markers for 4.5.2 Transferable markers 4.6 Conclusions

5. References

Curriculum Vitae

80

81

Sequencing strategy

4.4 Genetic system

4.4.2

76 77

fast and reliable

bacteria

4.4.1 Plasmids

74 74

4.2.3 Identification of the P. freudenreichii

4.3.1

70

73

4.1.1 "Selective" media and identification tests

4.3

69

73

4.1 Isolation of Propionibacteria

4.2.5

68

conjugation experiments for conjugation and electroporation

82 84 85

86 86 86 88

89

101

1

Summary

Summary Propionibacteria

have been found

so

far

mainly

in

dairy products

humans and animals. Due to the absence of real selective

consuming.

Propionibacteria are of a certain importance,

dairy

pathogens, is needed

faster identification process is necessary. In

a

as a

basis for further molecular studies with

The taxonomic situation of the genus construction of

a

phylogenetic

When necessary, the

complete

tree based

whether unknown bacteria from

on

belong

Proteinase

(5'-TGCTTTCGATACGGGTTGAC-3') hybridizes only strains

with the 16S rDNA

belonging

isolate when

a

assay. When

the genus

to

were

a

gene

used

probe

in the

bacteria

not

or

the

(two

determine

Briefly, by

DNA

SDS and

Propionibacteria, gdl

for

MPCR assay.

This

probe

(E. coli positions 632-651, V4 variable region) of

Propionibacterium. Propionibacteria

specific 900-bp (basepairs) sized DNA fragment

"foreign"

developed to

was

short method (mediated

specific is

by

determined in this thesis

Propionibacterium

using

A

reevaluated

was

existing database entries were completed.

or

to the genus

K) and amplified by MPCR.

correct identification

Propionibacteria.

multiplex PCR (MPCR)

pure bacterial culture is isolated

a

addition,

potential

as

16S rDNA sequences of all type strains.

on

16S rDNA sequences

rapid molecular method based

the skin of

Because both groups of

starter cultures or

Propionibacterium

classical and four medically relevant species) A

as

on

media, the isolation and

identification of Propionibacteria is difficult and time either

and

(not belonging

to the genus

is

are

present in

an

amplified in the MPCR

Propionibacterium)

occur

in

the sample,

a

control;

proof that the MPCR analysis was performed correctly, but that no detectable

as a

amounts

of

1500-bp

fragment

sized

Propionibacteria

were

Propionibacteria, the

detection limit

The MPCR method

can

samples, a

large

raw

number of

absence) After

e.g. in

or

was

in clinical

habitats

in

present at 2

x

10 a

be

a

fragment

sample.

cfu per ml in

Due to the

is used

For

rapid screening

specimens.

can now

This

a

pure

as an

internal

cultures

of

broth culture.

for

Propionibacteria

in

simple sample preparation,

investigated rapidly

for the presence

(or

of Propionibacteria.

determining the

In this

amplified.

be further used for

milk

new

is

thesis,

a

genus of isolates, the next

molecular

Propionibacteria

was

step is the identification

approach using partial sequencing

selected. Propionibacterium isolates

corresponding species by comparing

to

species

of the 16S rDNA genes of were

classified to their

the obtained 16S rDNA sequences to

database entries (e.g. GenEmbl). The whole analysis

was

level.

performed

within two

existing days.

2

Summary

In addition,

a

method for the identification of P.

freudenreichii, the most relevant classical

species, was developed. By using the PCR-amplified 16S-23S

gion of the P. freudenreichii type

belonging

species

to this

were

strain

as

hybridization probe, all

detected. The

principle

same

the identification of all other strains of the genus but this The P.

cies

freudenreichii species

(freudenreichii

and

has been

rDNA

the 16S

nor

strains of our collection in this thesis for

proposed

was not

separated by certain

shermanii). Neither

is

intergenic spacer re¬

further

investigated.

researchers into two

subspe¬

the 23 S rDNA sequences of the

type strains of both subspecies differed significantly in their base compositions. Therefore the

separation

For the

ofP'.

freudenreichii

into

subspecies

should be abandoned.

genetic analysis of Propionibacteria, plasmids suitable

isolated. These

plasmids

Propionibacteria.

Two

are

needed for the

plasmids, pLMElOl

development pLME108

and

as

of

a

cloning

vectors

were

genetic system

with

were

selected and further

analyzed. From P.

freudenreichii

restriction map were

obtained

identity The P.

using shotgun cloning. For one

identified.

were

sequences.

The rep

plasmid pAPI replication by

a

hypothetical protein

nucleotide sequence of

freudenreichii DF2,

rep,

Orfl

has

no

fragment, an amino acid sequence

Synechocystis

A

with

significant homologies

with

already

also

replicates by

Specific

plasmids

genetic

found. isolated from

known

and

protein

were

and

found

on

pLME108

used in this thesis

the

investigation of

conjugation. Both techniques

were

a

and led to the

the RC mode.

Propionibacteria requires

electroporation

amino acid motifs needed for

Propionibacteria, but in no experiment transformants were obtained. and

was

sp.

putative protein coding regions, or/2

rolling circle (RC) mechanism

genetic system

a

kb) of pLMElOl

plasmid pLME108 (2051 bp),

from Arcanobacterium pyogenes. the

3

isolated and

region shows 42.1% amino acid identity with the Rep protein of

plasmid

like

from

was

(approximately

sequence

determined. Two

was

conclusion that this

methods

plasmid pLMElOl (40-kb sized)

constructed. Partial sequences

was

of 53.2% with

complete

JS53

not suitable

or

complete

modification of Propionibacteria have to be established.

DNA transfer

were

tested

for

Either the conditions

new

approaches

for the

Zusammenfassung

3

Zusammenfassung Propionibakterien wurden vor allem aus Milchprodukten und von der Haut von Menschen und Tieren isoliert. Das Fehlen eines echten Selektivmediums erschwert und verzögert die

Isolierung

und

Identifizierung

von

Propionibakterien.

Von

Bedeutung

bakterien entweder als Starterkulturen in der Milchtechnologie oder als gène. Zu ihrer raschen Identifikation wird eine Methode

als Basis für

molekulargenetische

Arbeiten mit

Konstruktion eines

phylogenetischen

Propioni¬

potentielle

Patho¬

die im weiteren auch

benötigt,

Propionibakterien dienen

Die taxonomische Situation innerhalb des Genus

sind

Propionibacterium

soll.

wurde durch die

Stammbaums reevaluiert, der auf den 16S rDNS

Sequenzen aller Typstämme basiert. Falls notwendig wurden in dieser Arbeit komplette 16S rDNS

Sequenzen

bestimmt

(für

sowie für vier medizinisch relevante Es wurde eine

Spezies

zwei

Spezies)

molekularbiologische

der klassischen

oder bestehende

Propionibakterien,

Sequenzen komplettiert.

Schnellmethode basierend auf

„multiplex

PCR"

(MPCR) entwickelt, die es ermöglicht, die Zugehörigkeit unbekannter Bakterien zum Ge¬ nus

Propionibacterium

spezifische

CGGGTTGAC-3') siert

nur

bestimmen. DNS

(unter Verwendung

methode isoliert vermehrt. Eine

zu

von

Gensonde für

aus

einer Reinkultur wird mit einer Schnell¬

von

MPCR

Propionibakterien, gdl (5'-TGCTTTCGATA-

findet in dieser MPCR Methode

mit der 16S rDNS

K) und mittels

SDS und Proteinase

Stämmen die

zum

Verwendung.

Diese Sonde

hybridi¬

Propionibacterium gehören

Genus

(E. coli Positionen 632-651 der V4-variablen Region). Propionibakterien sind in einer Probe ren

vorhanden,

(Bp)

Genus

wenn

ein

spezifisches DNS-Fragment

mit der MPCR Methode

amplifiziert

Propionibacterium gehören,

mit der

Länge

Methode und

von

1500

einer

wird. Treten

in einer Probe auf,

Länge

nur

Bp gebildet. Dieses Fragment dient der

gilt als Beweis dafür,

900

Basenpaa¬

Fremdkeime, die nicht

wird

so

von

ein

zum

DNS-Fragment

internen Kontrolle der

dass der MPCR-Nachweis korrekt

durchgeführt wur¬

de, aber eben keine Propionibakterien in der Probe vorhanden waren. Die Nachweisgrenze

für Reinkulturen

ml einer

von

Propionibakterien lag bei 2x10

koloniebildenden Einheiten pro

Flüssigkultur.

Im weiteren kann die MPCR-Methode für ein schnelles

werden,

„Probenscreening" eingesetzt

z.B. in Rohmilch oder in klinischen Isolaten. Da die

einfach ist, können viele verschiedene

neue

Habitate

nun

Probenaufarbeitung

auf das Vorhandensein

sehr von

Propionibakterien untersucht werden. Der nächste Schritt nach der

Bestimmung

des Genus eines Isolates ist die Identifikation

Zusammenfassung

bis auf

4

Spezies-Ebene.

gewählt,

das auf der

basiert. Durch

In dieser Arbeit wurde ein

partiellen Sequenzierung der

Vergleichen der gewonnen

molekularbiologisches Vorgehen 16S rDNS Gene

16S rDNS

Spezies zugeordnet.

Diese gesamte

Propionibakterien

Sequenzen mit existierenden Daten¬

bankeinträgen (z.B. GenEmbl) wurden Propionibacterium Isolate den

von

aus¬

den ihnen

entsprechen¬

Analyse kann innerhalb zweier Tage durchgeführt

werden.

Zusätzlich wurde eine Methode schen

Spezies,

P.

freudenreichii,

mehrten 16S-23S rDNS

Hybridisierungs-Sonde

Identifikation des

entwickelt. Durch

„Spacer-Region"

des

wichtigsten

Verwendung

Typstamms

konnten alle Stämme dieser

Spezies

von

Vertreters der klassi¬

der mittels PCR P.

ver¬

freudenreichii

aus unserer

als

Laborstamm¬

korrekt identifiziert werden. Für die Identifikation aller anderen Stämme des

sammlung

Genus kann nicht weiter Die

zur

möglicherweise

das

gleiche Prinzip angewendet werden,

dies wurde aber

verfolgt.

Spezies P. freudenreichii

geteilt (freudenreichii

und

wird

von

shermanii).

verschiedenen Forschern in zwei

Subspezies

Weder die 16S noch die 23S rDNS

auf¬

Sequenzen

der

Typstämme beider Subspezies unterschieden sich wesentlich in ihrer Basenzusammenset¬ zung. Darum wird in dieser Arbeit

chii in zwei

Subspezies

genetisches

Um ein

Plasmide

zur

Entwicklung

Aus P.

eines

gelassen

Arbeiten mit

Konstruktion

und

pLMElOl

fallen

vorgeschlagen,

von

Propionibakterien

Klomerungsvektoren

pLME108 wurden

Trennung

von

P.

freudenrei¬

wird.

genetischen Systems

freudenreichii

dass die

in der

für

zu

ermöglichen,

wurden als Erstes

isoliert. Diese Plasmide werden

Propionibakterien benötigt.

Folge ausgewählt

JS53 wurde das Plasmid

zur

Zwei Plasmide,

und näher untersucht.

pLMElOl (40 kb) isoliert und eine

Restriktionskarte erstellt. Teilsequenzen von pLMEl 01 (ca. 3 kb) wurden durch „Shotgun

Cloning" Identität

gewonnen. Die von

Sequenz

53.2% mit einem

Für das Plasmids

eines

hypothetischen Protein von Synechocystis

pLME108 (2051 bp),

ständige Nukleotidsequenz bestimmt. gionen, orß und bereits bekannte tität mit dem

isoliert

Zwei

rep, wurden identifiziert.

aus

P.

freudenreichii DF2,

möglicherweise

Orfl

eine

sp.

wurde die voll¬

für Proteine kodierende Re¬

weist keine grösseren Ähnlichkeiten mit

Proteinsequenzen auf, während die rep Region 42.1% Aminosäure-Iden¬

Rep Protein des Plasmids pAPl

Spezielle Aminosäuremuster, chanismus

Fragments zeigte auf Aminosäureebene

benötigt werden,

die für die

aus

Arcanobacterium pyogenes aufweist.

Replikation

durch den

konnten innerhalb der rep

„rolling

Region

circle"

des Plasmids

(RC)

Me¬

pLME108

5

identifiziert werden. Dies führte

chanismus für seine Die

Entwicklung

zum

Zusammenfassung

Schluss, dass dieses Plasmid ebenfalls den RC-Me¬

Replikation verwendet.

eines

genetischen Systems

für

Propionibakterien

setzt im Weiteren die

Untersuchung von DNS Transfermethoden wie Elektroporation und Konjugation voraus. Beide Techniken wurden mit keinem

Experiment

Propionibakterien getestet, aber Transformanten konnten in

erzielt werden. Entweder

Plasmide in dieser Arbeit nicht Modifikation

von

geeignet

Propionibakterien

oder

waren

die verwendeten

vollständig

neue

müssen entwickelt werden.

Bedingungen und

Ansätze

zur

genetischen

Leer

-

Vide

-

Empty

Introduction

7

1. Introduction of Propionibacteria

1.1 A short

history

In 1906,

Freudenreich and Orla-Jensen (1906) isolated bacteria from Emmental

von

cheese which These

they

organisms

authors

named Bacterium

for the

responsible

them to be

suspected

by their ability

characterized

were

cheese. The genus Propionibacterium

produce propionic

to

typical

acid and the

eye formation of that kind of

proposed by Orla-Jensen (1909)

first

was

acidi-propionici.

and Bacillus

acidi-propionici

for these

bacteria. The

inclusion

of

Propionibacterium reconfirmed

by

Corynebacterium acnes

first

was

Moore and Cato

1.2 Characteristics and

Propionibacteria

the

Propionibacterium

genus

proposed by Douglas

and Gunter

(1946)

as

and

(1963).

occurrence

Gram-positive

are

into

acnes

and

of Propionibacteria

rodshaped ("coryneform")

microaerophilic to anaerobic growth conditions.

bacteria which

prefer

The G+C content of their genome ranges

from 53 to 67 mol% (Cummins and Johnson, 1986 & Kusano et al., 1997) which places them in the

high

G+C group of bacteria

The most characteristic and

propionic and

acetic acid

the formula: 3 lactate

=

2

common

as

(Olsen

et

al., 1994).

property of the Propionibacteria is the production of

main metabolites from lactic acid

propionate

+

acetate +

C02

+

as

substrate

according

to

H20.

1.2.1 Classical

Propionibacteria

The classical

dairy Propionibacteria are found mainly in raw milk and in dairy products

or

like buttermilk and

ripening

process

Propionibacteria

(Cummins

are

utilized to obtain

responsible for

especially

a

the

of great

in

and

propionic

et

as

typical

eyes that

in olives

are

1992).

are

cheese

modern or

as

Swiss-type

important

starter

cheese.

in the

making, cultures,

They

are

characteristic for that kind of cheese.

Propionibacteria and

In

wild types strains

the flavor of these

(Plastourgos

they

where

full flavored and well textured

and acetic acid.

spoilage organisms

Johnson,

importance

Propionibacteria contribute also to of

Swiss-type cheeses,

were

products, mainly by found in

Vaughn, 1957)

honey

the

production

and maize and

and orange

as

juice (Kusano

al., 1997). The latter contained Propionibacterium cyclohexanicum, an organism that is

8

Introduction

highly

heating (survival

resistant to

Propionibacteria are found (Weinberg

al., 1998.), sekete (Adegoke

et

are

Johnson, 1992; ability

Glatz

of certain

other bacteria

et

al., 1995),

P.

a

préservants

as

red beet

or

juice

Nigerian beverage made

or

production

of vitamin

B12 and propionic acid (Cummins and

1992).

Propionibacteria

to

produce substances inhibitory

1992). The study

one

of the

major research

areas

freudenreichii,

which

Bb B2, B12 and

milk with

multiresistant to antibiotics and

was

C. This fermented milk

antibiotic treatment to children

(age

dysbacteriosis. Reconvalescence

time for children

of 4 to 18

et

growth

of

al., 1991;

of substances like

concerning Propionibacteria.

study, Sidorchuk and Bondarenko (1984) inoculated

vitamins

to the

yeasts and moulds has also been investigated (Al-Zoreky

bacteriocins is currently a

90°C). Further, classical

added sometimes also

Hsieh et al., 1996a, 1996b; Grinstead and Barefoot,

In

at

maize, have been improved or made with Propionibacteria. Propionibacteria are also

discussed for the industrial

The

silage, where they

in

minutes

al., 1995). Vegetable products like "sauerkraut"

et

(Babuchowski from

10

of

was

mutant strain of

a

produced high

administered

amounts of

together

with

an

months)

with symptoms of enteral

supplied

with this milk

was

shorter

than for children of the control group.

1.2.2 "Cutaneous"

Propionibacteria

The first isolates of the cutaneous

They have

also been found in feces, in the mouth and

(Cummins

and Johnson,

1992).

certain human diseases

(for example

acne,

and animals

organisms are

the

are

cause

procedure.

As

moist parts of the skin of humans

on

The cutaneous strains

an

an

infection

or

have been isolated

example, Propionibacterium

106 cfu/cm2 in body

areas

rich with sebaceous

as

acnes

Because these

contaminants during the

they

sampling

has been found at densities of 105-

glands (scalp, face)

and at 102 cfu/cm2

on

(Funke et al., 1997, Jakab

al., 1996) indicate that Propionibacteria are indeed the cause of certain infections. In the

case

of acne, which is the most

ibacteria, et

to be involved in

to determine whether

other parts of the skin (Leyden et al., 1998). Recent publications et

seem

endophthalmitis, endocarditis).

part of the natural skin flora, it is often difficult

of

from the human skin.

Propionibacteria were obtained

P.

acnes

is

clearly

al., 1998). Mainly in

can cause

common

and best known disease connected with

involved but

cases

where

an

Propion¬

only in combination with other factors (Leyden

invasive treatment is carried out,

serious infections in the human

body (Jakab

et

Propionibacteria

al., 1996; Tunney

et

al., 1998).

Introduction

9

Strains of P.

acnes

longer valid), used

as

or

(often named Corynebacterium

of P. avidum

(e.g. Pottkämper

to achieve

an

et

Nakamura

(1978)

Taxonomy

and

by

an

orally

al., 1997)

are

heat and

were

administered to the test persons

been described for P.

Paul and Booth

important

acnes

by Fujimura and

(1988).

role in food

microbiology

microbiology

because

they

are

involved

productive and destructive processes in food. As spoilage organisms they make

food unsuitable for the even worse

food and

production has

and its relevance in food

Microorganisms play in various

and

a name no

activation of their immune systems.

In addition bacteriocin

An

al., 1996; Randerath

immunostimulants (for animals and humans). In these studies, cells

formaldehyde inactivated, washed, lyophilized

1.3

et

parvum in this context,

scenario is the

endanger the

to preserve

or

consumption and occurrence

are

for substantial financial losses.

responsible

of pathogenic

microorganisms that contaminate

health of consumers. On the other

produce food.

Products like bread,

wine,

side, microorganisms

are

used

beer and cheese would not exist

without them. The clear identification of microorganisms from and in food is necessary to

potential to

for the consumer, to achieve minimal losses

ensure an

optimal product quality. The

use

during

the

production

an

process and

strains, which

important task of food microbiology.

The basis to identification is the taxonomic are

the risk-

of defined and well known strains for fer¬

mentation processes includes also methods for the identification of these

has become

asses

related have

more

common

position

of a

microorganism. Organisms

characteristics than non-related

ones.

that

In the classical

taxonomy, phenotypic features like cell morphology, carbohydrate fermentation patterns and the

production

classification of and

can

an

of

primary

organism.

and

secondary metabolites

Theses features

are

were

used

highly dependent

on

as a

key

for the

the environment

change for example when bacteria are under stress, leading to faulty identification

of isolates. In modern taxonomy, information about the

obtained from the

analysis of conserved DNA regions

of bacteria, is also taken into consideration

(Ludwig

genotype, especially data

like the genes for 16S

and

Schleifer, 1994).

or

23 S rDNA

10

Introduction

Taxonomy

1.4

Propionibacteria

of the form

a

Propionibacteria relatively homogeneous

1991) within the class Actinobacteria which

(1997).

This class consists

mainly

of

bacteria and characterized

mainly by

content of their DNA. The

proposal

rDNA sequences. the

family

was

taxonomic cluster

recently

defined

organisms previously

their

by

mainly

referred to

Propionibacteria are positioned in the

based

suborder

on

Propioniferax

which

coryneform

as

and

a

high

comparisons

G+C

of 16S

Propionibacterineae

of Propionibacteriaceae. Their closest taxonomic relatives

genera Luteococcus, Microlunatus and

Riedel,

Stackebrandt et al.

shape (club shaped, branched)

for this class is

and

(Britz

belong

are

in

members of the

also to the

family

Propionibacteriaceae.

Various

approaches

to

classify Propionibacteria have

species Propionibacterium cyclohexanicum, proposed (Kusano

et

Carvalho et al., 1995)

currently

table 1. The

or were

known and

transferred from other genera

species.

new

et

juice,

of the genus

by

their

and the

Propionibacterium source

medically

of isolation relevant

species P. cyclohexanicum is referred to

Propionibacteria because

Species

Classical

as

it has been isolated from food and

classified

currently

Propionibacteria

in the genus

has been

Cutaneous

acidipropionici cyclohexanicum P. freudenreichii P.jensenii

P.

P. avidum

P.

granulosum lymphophilum

P. thoenii

P.

propionicumx

Correct latin form of the

name.

acnes

are

listed in

(habitat)

into

belonging to the

seems

Propionibacteria

P.

P.

to

(or cutaneous)

Propionibacterium

P.

1

new

(Arachnia propionica

pathogenic.

Table 1:

the

al., 1988).

been divided

dairy (or classical)

In this thesis the

"classical"

accepted species

Propionibacteria have

two main groups: the

isolated from orange

Recently

al., 1997), strains have been reclassified within the genus (de

Propionibacteriumpropionicus; Charfreitag

The

been carried out.

to be

non¬

11

The

Introduction

species P. freudenreichii is further divided by

P.freudenreichii subsp. freudenreichii sometimes P. This

(or 3) subspecies:

certain authors into 2

and P. freudenreichii

subsp.

shermanii and

freudenreichii subsp. globosum (Cummins and Johnson, 1992; Holt, 1984).

separation

is based

mainly

of a strain to

freudenreichii)

produce

ability (subsp. shermanii)

the

on

acid from lactose

(Cummins

and

or

lack

(subsp.

Johnson, 1986).

1.5 Isolation and identification of Propionibacteria 1.5.1 "Selective" media

Several media for the isolation of

proposed,

popular being

the most

Malik et al.

(1968).

selectivity.

Because

will almost

addition

always

U-g/ml) to the medium to

8

commonly used

as

enumeration of

Propionibacteria

starter cultures in

isolates)

PAL is rather

based

or

detection of single

by using

on

its

strains,

Methods based to

on

using "classical"

carbohydrate

source

and lacks

this medium. Drinan and cloxacillin

to 14

Cogan

growth of the lactic acid bacteria

PAL

selectivity

compounds

are

(Standa industries,

in other systems

not available. In

(Beveridge, 1990).

(like

addition,

and the identification of Propionibacteria is This

can

lead to

errors

no

longer possible at high colony

no

selective media

are

since the

counts.

in use, the isolates

are

performed

1996).

methods

identify Propionibacteria (Cummins Propionibacteria

for

(MIC

chemotaxonomy and colony morphology are the most commonly

taken of the fact that

conditions

lactate) proposed by

A selective solid medium for the

classical methods (von Graevenitz and Funke,

procedures

have been

standard media under anaerobic conditions and identification is

1.5.2 Identification

et

on

cheese)

growing organisms (1 day

(pink to yellow).

colony contaminants is

For the isolation of the cutaneous

incubated

plate)

color reaction in the medium

on a

extract

dairy products named

the nature of the selective per

from

antibiotic

inhibit the

dairy products.

from

expensive (sFr. 1.50

on

of the

Rouen, France) is patented and sold, but data clinical

as

slow

grow faster

Hg/ml

of 4

(yeast

lactic acid

on

Propionibacteria are relatively

the

Propionibacteria
12 kb). Plasmid-DNA

McKay (1983) and purified using CsCl2

density gradient centrifugation (Sambrook et al., 1989). Restriction endonuclease digests of the

plasmid-DNA

onto the agarose

hours

gel,

according

System (BioRad)

0.5% TBE

running

mixed with

loading dye (Sambrook

et

al., 1989) and applied

gel.

The CHEF-DR II agarose

were

time.

was

used with the

running buffer, 4°C,

Preparation

200 V, 1-20 seconds

of the apparatus and

to the manufacturers instructions.

following parameters:

1.5% TBE

pulse ramping time,

electrophoresis

were

18

performed

25

2.4

Polymerase

2.4.1

Material and methods

chain reactions

PCR-primers

All

oligonucleotide primers

are

listed in table 5. A stock solution

synthesized by Microsynth (Balgach, Switzerland)

were

(0.1 mM)

distilled water, irradiated with ultraviolet

of each

primer was prepared in

light (302 nm) for 5

Sequence (§'

Name

—>

3')

position

bakllw

AGTTTGATCMTGGCTCAG

ÏÔ51

bak4

AGGAGGTGATCCARCCGCA

1540-1522

for PCR

16S rDNA

Goldenberger,

16S rDNA

Greisen et al., 1994

eubac338ACTCCTACGGGAGGCAGC

338-355

TGCTTTCGATACGGGTTGAC

632-6511

Genus

16S rDNA

'

GTGCCWAGGCATCCACCG

2020-2003

*

lm30

CGTGTCGTGAGATGTTGG

lm38

TTAACCTTCCAGCACCG

ml3rev

CAGGAAACAGCTATGAC

1033-10501 Gram-positive 16S 3835-38191 Gram-positive 23S 465-4814 M13/pUC18

High Propionibacteria 23 S

1997

Amannetal., 1995

Propionibacterium

CGGCGGYGWCCTACTCTCCCA 5102-5072

pramprev AAGGGCCTCGTGATACGCCT

MPCR

reference

gdl gd5srev gd6r

prl08rw GCCCGCCACACCCGAACCGGT prl08rev CAACGTCCGCAGGAAGATCAT prampfw GCAAGCAGCAGATTACGCGC

or

target gene

universal5 universal5 universal5

'

sterile bi¬

minutes and stored at -20°C.

Synthetic oligonucleotides, position and origin used

Table 5:

and

this this

GC 5S rDNA

rDNA this

study study study

rDNA

Meile, 1998

rDNA

Meile, 1998

Pharmacia, Sweden

1845-18652pLMEl08 1771-17512pLME108 1418-14374am^RofpUC18 2681-26624 ampR of pUC18

this

study study this study this study this

TM1

TGATAGCGGGAACAAATA

307-3343

tetM(pAM\20)

Perreten, 1995

TM2

ATACAGGACACAATATCC

2570-25533tefM(pAM120)

Perreten, 1995

vpl vp2

GAYACGCCAGGMCATATGGA

738-7583

tetM(pAMl20)

Perreten, 1995

TYGGRCGRCGGGGYTGGCAA

2392-23733fefM(pAM120)

Perreten, 1995

1 2

3 4 5

E. coli

positions according

Position

according

to

Position

according

to tetM

Position

according

to

to

Brosius

et

al., 1978.

pLME108 (EMBL accession

no

AJ006662).

(EMBL accession no M85225).

pUC18 (EMBL accession

no

L09136).

Universal for most of the known bacterial 16S rDNA sequences.

2.4.2 Standard PCR conditions

Target DNA was amplified in 0.2 ml thin-walled tubes using thermocyclers equipped with heated lids (Personal

Cycler, Biometra; Genius, Techne).

mixture contained 0.2 mM each of

DNA-Polymerase (Pharmacia),

5

dATP, dCTP, dGTP, dTTP (Pharmacia), ul

lOx standard PCR-buffer

bidistilled water, 1 uM each of forward and

quantities

prepared

of probes,

a

A standard 50 ul reaction

reverse

primers

and

mastermix that contained all components

1 Unit

(Pharmacia),

template

Taq

sterile

high

DNA. For

(no template DNA)

was

and distributed to the reaction tubes.

The standard PCR program is shown in

calculated using

a

simple formula (TA

optimized empirically.

For

one

=

kb of

polymerization time. The resulting

figure

[no.

1.

Annealing temperatures (TA)

of GC]x 4°C

amplicon size,

PCR programs

are

+

[no.

one

of AT]x

minute

2°C)

was

shown in table 8.

were

and further

added to the

26


HHHHj^H HHiHHIH

1 ^^H

40

^j^H

amplification cycles:

1

denaturation

95°C,

Initial denaturation

sec.

Final hold

Final

polymerization

primer annealing TA °C, 30 sec.

3 min.

95°C,

15

72°C,

4°C,oo

7 min.

polymerization 72°C, tPo| min.

1 ^^^^^^^l^Hi^^^^H | Figure

1:

PCR

Standard

polymerization times (tp0j ) Table 6: Primer

PCR

The

protocol.

for the different

and

primer pairs

annealing

amplification protocols

Ta

bakllw-bak4

16S rDNA,-1500

bakllw-gd6r bakllw-gd5sr bak4-gd6r

16SrDNA +ITS,-1900 16S

->

tpoi.

44°C 3 min.

Sequencing Sequencing Sequencing Hybridization probe Sequencing Cloning Cloning Cloning Hybridization probe Cloning

56°C 2 min.

MPCR

56°C 2 min.

bp

5S rDNA, -5000

bp bp

TM1-TM2

partial pLME108,1977 bp complete ampR, 1263 bp partial pLME116, 3800 bp partial tetM, 1654 bp complete tetM, 2246 bp

(bakllw, gdl)-bak4

16S rDNA, 900+1500

prl08fw-prl08rev prampfw-pramprev prampfw-ml 3rev vpl-vp2

shown in table 6.

Application

56°C 2 min. 50°C 5 min.

56°C lmin. ITS, -400 bp 16S -> 23S rDNA, -2800 bp 50°C 3 min.

Im30-lm38

are

and

amplicons

Amplicon

pair

(TA)

temperatures

56°C 3 min. 50°C 4 min.

44°C 5 min. 56°C 2 min.

bp

TA: Annealing temperature; tPoj : Polymerization time. ITS: 16S-23S rDNA

2.4.3

intergenic

spacer

region.

Multiplex-PCR (MPCR)

To bind SDS Tween 20

for

(inhibitory

(Merck)

were

added to the 50

primers bakllw, gdl and temperature

was

set to

DNA-polymerases)

1

pM

reverse

used in the

|il standard

sample preparation,

reaction mixture. 1

primer bak4 (table 5)

56°C, with the polymerization steps

were

|iM forward

used.

at 72°C for 2

2%

Annealing

minutes.

27

2.4.4

PCR conditions

Long

To generate PCR

(Stratagene)

2.5

was

products longer

used

according to

of DNA

Labeling

of the

pmol

probe

the manufacturers instructions.

was

respective oligonucleotide

purified using

Prior to use, the

before it

added to the

was

2.5.2 Random

technique

as

NAP-columns

radioactivity.

a

kb, the 'Ta^Plus Long PCR System"

probes

the T4-Kinase

(DuPont) using

In

than 3

Labeling of oligonucleotides with [y-32P]ATP

2.5.1 100


probe was

hybridization

primed labeling

was

[y-32P]ATP

labeled with radioactive

described

by Sambrook

(Pharmacia)

to

remove

et al.

(1989). The

non-incorporated

denatured for 5 minutes at 95°C and

kept on ice

solution.

of DNA-fragments with

DNA-probe

final volume of 25 u,l, the linearized

[oc-32P]dATP labeled with

was

[oc-32P]dATP

(DuPont) using the random priming technique of Feinberg and Vogelstein (1983). remove

surplus radioactivity, the probe

according

2.5.3 Terminal

kept

labeling

Oligonucleotide probes labeling kit (Boehringer)

DNA

Labeling

a

NICK-column

on

probe

was

(Pharmacia)

linearized for 5

ice.

of oligonucleotide-probes with DIG-ddUTP were

as

labeled with

recommended

digoxigenin-ddUTP using

by

the DIG 3'-end

the manufacturer.

of DNA-fragments with DIG-dUTP

fragments

random

purified using

to the manufacturers instructions. Prior to use, the

minutes at 95°C and

2.5.4

was

To

were

labeled

using

the "DIG DNA

primer incorporating method

as

described

labeling

by

mix"

(Boehringer)

the manufacturer.

and the

Material

2.6

28

and methods

Hybridization techniques

2.6.1 Southern

blotting

DNA transfer from agarose

Southern

gels

to Zeta-blot membranes

blotting technique using

0.4 N NaOH

(BioRad) was

transfer buffer and

as

achieved with the a vacuum

blotting

equipment (Biometra).

2.6.2

Colony hybridization

Nylon membranes (Amersham) each

and

side)

placed

toothpicks, single membrane with

was

were

petri-dishes

on

colonies

lysis-solution.

After the

(pH 8.0)

cell-lysis step

placed

growth

(302

nm, 10 minutes

medium.

M

3 MM paper

on a

Using

sterile

30 units/ml was

(Whatman)

mutanolysin placed

on a

were

at

room

fresh 3 MM paper soaked with 1 M Tris

was

soaked with 2x SSC

there for 4 minutes. The membrane

Finally

added

3MM paper

NaOH; 1 M NaCl). After 4 minutes a

soaked

mg/ml lysozyme) and

M sucrose; 5

the membrane

transferred to

fix the released DNA.

at 80°C to

2.6.3

was

(0.5

for 4 minutes. A fresh filter paper

membrane

protected

upwards

side

Propionibacteria,

soaked with denaturation solution

temperature, the membrane

UV-irradiation

with suitable

Tris, pH 7.5; 0.25

mM

incubated for 1 hour at 37°C. For to the

by

transferred to the membrane. After incubation, the

were

placed with the colony

lysis-solution (10

sterilized

was

membranes

(pH 7.0)

and the

air-dried and kept for 2 hours

were

stored

air-tight

and

light-

at 4°C.

Hybridizations

Membranes

were

with DNA

incubated in

probes Micro-4

a

oven

(Hybaid)

at

hybridization temperature

(large DNA-fragments: 65°C; primer gdl : 56°C) with 30 ml prehybridization

solution

(5x

SSC; 5x Denhardts; 0.25 mg/ml sssDNA; 0.05 M sodiumphosphate buffer, pH 6.5). After 3

hours, the prehybridization solution

(5x SSC;

containing

lx

Denhardts;

the labeled

When DIG-labeled DNA-DNA

or

0.5

probe

probes

DNA-RNA

were

0.04 M

sodiumphosphate buffer pH 6.5)

added.

used, washing steps and colorimetric detection of the

hybrids

the "DIG nucleic acid detection kit"

Sambrook et al. (1989).

discarded and 30 ml of hybridization solution

mg/ml sssDNA;

were

Membranes with radioactive

was

were

performed

as

described

by the manufacturer

in

(Boehringer).

probes were washed, detected

and

reprobed as described by

29

2.7


of Propionibacteria isolated from

Analysis

dairy products

2.7.1 Isolation of Propionibacteria from cheese 10 g

samples of each cheese were dissolved in 90 ml YEL broth and grinded for 5

in

stomacher

a

(Colworth). Serial

dilutions

were

prepared

YEL agar. After anaerobic incubation for up to 1 per plate, 200 colonies per

After

hybridization with

the

probe eubac338 (table 5), specific probes isolates

were

isolation) were

further

were

was

propionibacteria-isolates to

2.7.2

developed

were

positive signal

To obtain pure

determined

both with the universal and

cultures, single colonies of the were

Gram-stained (KOH

measured and for strains with uncertain

patterns

the genus

study.

In

were

determined.

Propionibacterium

The

was

identification,

affiliation

reconfirmed

by

addition, for strains of special interest, described in this

using

intergenic

(1997), dairy Propionibacterium species

Rossi et al.

by

spacer

a

simple

regions. spacer

In

restriction

region was used in this study

of

all

the MPCR 16S rRNA

study.

to

classify

and

plasmid-encoded properties

plasmids,

the presence of plasmids

antibiotic resistance patterns

easily rDNA

species

level.

the identification scheme of

plasmids

Propionibacterium-isolates

be

16S-23S

isolates to

used.

In all

can

16S rDNA and the 16S-23S

was

Screening

was

PCR-amplified

freudenreichii subspecies,

2.7.3

that contained

of

(1986)

Cummins and Johnson

for

analysis

addition, partial sequencing of

To differentiate between the P.

these strains

oligonucleotide

determination

As described

rDNA

in this

colony-hybridization.

and the universal

sequenced using the cycle sequencing method

Species

intergenic

investigated.

fermentation

carbohydrate

genes

genus-specific probe gdl a

on

week, "typical" colonies (50 colonies

transferred to fresh agar for 3 times. These strains

test), catalase activity

method

in YEL and streaked out

selected and prepared for

colonies that gave

minutes

was

were

investigated.

For strains

determined and

curing

of

attempted.

2.7.4 Determination of antibiotic resistance patterns

Propionibacteria were precultured in broth culture under optimal conditions and harvested

by centrifugation (5 minutes, corresponding

to the

room

temperature, 6'000 g). A solution of the cells,

McFarland standard 2,

was

prepared

in peptone-water

(0.1%


30

peptone, 0.8% NaCl). Using

plates. Etest-strips (Etest® System,

agar

placed

on

the

at 30°C. The was

sterile swap, this solution

a

plates and reading

performed

The

as

results

were

of the minimal

recommended

following antibiotics

were

by

AB

was

spread

on

MuellerHinton

Biodisk) with different antibiotics

determined after anaerobic incubation for 3-4

inhibitory

(MIC)

concentrations

the manufacturer.

investigated: ampicillin, bacitracin, benzylpenicillin,

gentamicin, kanamycin, methicillin,

2.7.5

and

nalidixic acid, ofloxacin,

of

fusidic acid,

rifampicin, streptomycin,

vancomycin.

Curing experiments with plasmid containing Propionibacterium

Curing

Propionibacteria

was

performed by

incubation

at

strains

suboptimal growth

temperatures (Smutny, 1997). Briefly, the highest possible growth temperature a

strain

was

determined and the strain

serial dilution were

was

selected and

described in further

was

prepared and plated

colony hybridizations

chapter

analysis

days

of each antibiotic

cefotaxim, cephalothin, chloramphenicol, ciprofloxacin, erythromycin,

tetracycline

were

on

incubated at this temperature for 2 weeks. A solid NL-medium, incubated at

with

2.6.2. Colonies that gave

to confirm the

MPCR confirmation for

absence of the

Propionibacteria).

(Tmax) for

plasmid-DNA no

as

probe

were

signal with this probe

Tmax.

Cells

performed

were

as

selected for

original plasmids (plasmid-DNA isolation,

31

2.8 DNA 2.8.1

sequencing

Cloning

Vector DNA

the

of DNA

Pharmacia) from the

gels

fragments using pUC18

fragments

enzymes, if necessary

and

analyzed

and

with

to

and E. coli XLl-Blue

be inserted

gel-electrophoresis.

purified using

the

Quiaex

Sambrook et al. (1989). The origin of the confirmed

2.8.2

Cloning

using hybridization

of PCR

PCR fragments

were

were

digested separately

with

dephosphorylated using alkaline phosphatase (CIP,

II

The desired

fragments

excised

performed according

was

plasmid insertions from

and PCR

were

System (Quiagen). Ligation, electro-

transformation and selection of recombinant E. coli

were

and methods

analysis techniques

and DNA

(pUC18)

appropriate

and

Material

techniques

the

described

as

to

resulting clones

previously.

fragments

cloned in E. coli XL 1 -Blue

using the pGEM®-T Easy Vector system

(Promega).

2.8.3 Purification of PCR-products for 100 ul of each PCR

Nagel)

product

were

Cycle sequencing

PCR

products (200-300 ng)

or

reactions

purified with the Nucleospin-Extract kit (Macherey

and eluated with 50 ul of 10 mM

2.8.4

sequencing

&

Tris, pH 8.5 (Merck).

cloned DNA

fragments

were

amplified with the

"Thermo

Sequenase fluorescent labeled primer cycle sequencing kit with 7-deaza dGTP"

(Amersham)

as

recommended by the manufacturer.

Sequencing primers were synthesized

(Microsynth, Balgach, Switzerland)

and labeled with

listed in table 7. The

program consisted of 25

amplification

95°C, 30 seconds 50 °C and 1 minute solution

(Amersham) was

automatic

analyzed

added.

at 72°C.

Finally

the

cyanine (Cy5, Amersham)

Samples were

products

were

cycles

described in the

following chapter.

are

with 30 seconds at

cooled to 4°C and 5 ul stop

analyzed on

DNA-sequencer (Pharmacia) and the generated DNA

as

and

an

ALF-Express

sequences

were

further

32


Table 7:

Cy5-labeled oligonucleotide primers

Sequence (5'

Name

Primers used for the

—»

E. coli

3') of DNA

sequencing

alfml3uniCGACGTTGTAAAACGACGGCCAGT Primers used for 23S rDNA

position target

cloned in

fragments

alftil3rev CAGGAAACAGCTATGAC

used for automated

sequencing reference

gene

pUC18

-

pUC18

Pharmacia,

-

pUC18

Pharmacia, Sweden

Sweden

sequencing

bO

CCTTTCCCTCACGGTACT

Boesch, 1998

ATAGCTGGTTCTCCCCGAAA

3974-3957, fw 4321-4302, fw

Proteobacteria 23S rDNA

bl

universal 23S rDNA

blrev

TTTCGGGGAGAACCAGCTAT

4302-4321,

universal 23S rDNA

Boesch, Boesch,

b2

GTTGGCTTAGAAGCAGCCATC

4559-4579, fw

universal 23S rDNA

Boesch, 1998

b2rev

GATGGCTGCTTCTAAGCCAAC

4579-4559,

universal 23S rDNA

Boesch, 1998

b3

CCCCTAAGGCGAGGCCGAAAG

4848-4868, fw

universal 23S rDNA

b4

AGAGAATACCAAGGCGCTTG

5130-5194, fw

Proteobacteria 23S rDNA

Boesch, 1998 Boesch, 1998

b5

AAGTTCCGACCTGCACGAAT

5452-5471, fw

universal 23S rDNA

Boesch, 1998

b6

GTAACGGAGGCGCGCGAT

5757-577'4, fw

a-Proteobacteria 23S rDNA

Boesch, 1998

b7

AGAACGTCGTGAGACAGTTC

6027-6080, fw

a-Proteobacteria 23 S rDNA

Boesch, 1998

gd6r

GTGCCWAGGCATCCACCG

2003-2020,

Propionibacteria,

Primers used for 16S rDNA

rev

rev

rev

23S rDNA this

1998 1998

study

sequencing

eubac338 ACTCCTACGGGAGGCAGC

338-355, fw

universal 16S rDNA

Amannetal., 1995

ms350r

CTGCTGCCTCCCGTA

343-357,

rev

universal 16S rDNA

Lane, 1991

uni515

ACCGCGGCTGCTGGCAC

515-531,

rev

universal 16S rDNA

Lane,

uni785

GGMTTAGATACCCTGGTAGTCC

785-806, fw

universal 16S rDNA

Amannetal., 1995

UIÜ1088

GGTTAAGTCCCGCAACGAGC

1088-1107, fw

universal 16S rDNA

Amann et al., 1995

unil392

GTACACACCGCCCGTCA

1392-1408, fw

universal 16S rDNA

Lane, 1991

unil392r

TGACGGGCGGTGTGTAC

1392-1408,

universal 16S rDNA

Lane, 1991

fw=

forward;

rev=

reverse; E coli

positions according

rev

1991

to Brosius et al., 1978.

Universal: universal for all known bacterial sequences.

2.8.5

Sequence

DNA sequences 8.0

data were

analysis

and

phylogenetic

analyzed with

calculations

the programs

provided

in the

GCG-package

version

(Genetics Computer Group, Madison, Wisconsin, USA), the EGCG programs (Peter

Rice, The Sanger Centre, Cambridge, GB) and the PHYLIP software tools (Felsenstein,

1982).

DNA sequences

and EMBL). The

use

compared

with

of

plasmids

local GenEmbl database copy

drawn

were

(sci.-ed software, Durham, USA)

technologies, USA).

a

of this programs is indicated in the

Graphic representations program

were

or

the

respective parts

(Genbank

of this work.

using the Plasmid Map Enhancer 3

Omiga

software (version 1.1, Rainbow

33

2.9

Conjugation

2.9.1 Selective

separation

mainly

supplements

problem

The main

between

in

Propionibacteria

of donor and

is the selection of the transformants: the

from the real

recipient organisms

transconjugants.

the inhibition of donor cells, the addition of different were

grown

on

following "selection enhancers"

green, 0.02

g/1 medium),

days growth

tested: the

were

the addition of end

supplements to

To achieve

the medium

BHI-agar before inoculation of the

medium. After anaerobic incubation at 30°C for 7 The

and other genera

for the identification of transconjugants

conjugation experiments

tested. All test strains

was


use

products

of

of the

a

was

evaluated

test-

visually.

pH-indicator (bromcresol

"propionibacteria-pathway"

(acetic and propionic acids), of sodium chloride and of the antibiotic cloxacillin.

2.9.2 In

Adaptation

order

to

experiments,

of Propionibacteria to

strains

2.9.3

resistant to 100

Conjugation with

Donor and

were

recipient

supplements.

The cells

-2.0 for donor

the

strains

(Millipore), with a ratio

donor to the

filters and transformants

were

selective medium. The strains

recipient

were as

resistance gene

as

second

phase.

containing

The

resulting

of rifampicin.

a

0.45

based

on

um

were

by

cellulose membrane

optical density (OD60o

the

described

unselective medium, the cells

by Wirsching (1995).

were

incubation of serial dilutions

washed from the on

the

respective

then incubated at 30°C for up to 4 weeks and checked

recipient strains (positive control) was

appropriate

antibiotics. As

plated and incubated

probe.

a

Hg/ml

filter-mating procedure

selected

analyzed by hybridization of the

on

recipient cells of 3:1

on

with the

well

in

High

precultured in broth media with the respective antibiotic

Growth of donor and

supplemented

rifampicin

then concentrated

were

days

media

conjugation

filter-mating technique

were

cells), using

growth daily.

in

strains

the strains in media

by incubating

fusidic acid and 50

u,g/ml

After incubation for 1-3

for

found

concentrations of fusidic acid, and

were

recipient

as

easy selectable marker had to be introduced in these strains.

antibiotic resistant isolates

increasing

concentrations of antibiotics

Propionibacteria suitable

obtain an

high

on

negative controls,

checked

donor and

the selective media. Transformants

clones with the transferred plasmid DNA

or

on

were

the antibiotic

34


2.10 Electrotransformation of Propionibacteria

Competent cells of Propionibacteria were obtained using 3 different methods: the method described for E. coli

by

described

(Sambrook

Gautier

et

al.

et

al., 1989) using 10% glycerol

(1995) and

an

(polyethylenglycole lO'OOO, Merck) (Zirnstein

as

buffer,

a

adapted method using

and

Rehberger, 1991).

procedure

30%

For all

PEG

methods,

Propionibacteria from a 500-ml culture were harvested (mid-logarithmic growth stage) by centrifugation (15 minutes, 4°C, 4'OOOg). buffer system and concentrated volume of 2-3 ml

a

before

achieved.

electroporation. Viability

cells that

study

was

by

were

and the

Table 8:

electroporated

The

pellet

further washes with

Competent cells

was

resuspended

decreasing

were

of the competent cells

amounts of buffer until

stored at -72°C and was

determined

strains

Growth medium1

Buffer2

NLorNLG

10%

P.

cyclohexanicumT

NLorNLG

10%

NL

10%

freudenreichii (shermanii) P. freudenreichii (shermaniiJ) P. freudenreichii ARE1 P. freudenreichii FAM1409 P. freudenreichii FAM1410 P. freudenreichii FAM1411 P. freudenreichii FAM1412 P. freudenreichii FAM1413 P. freudenreichii FAM1414 P. freudenreichii JS53

P.

freudenreichii1

P. "intermedium"

glycerol glycerol glycerol

GS

YELG

NLG

30% PEG

NLorNLG

glycerol glycerol 10% glycerol 10% glycerol or 30% PEG 10% glycerol 10% glycerol or 30% PEG 10% glycerol

NL

30% PEG

NL

or

NLorNLG

NLorNLG NL

or

NLG

or

NLG

NL NL

or

NLG

YELG

GS

NLorNLG

30% PEG 10%

(pituitosum) P. jensenii {raffinosaceum) P. jensenii (technicum) P. jensenii

NLorNLG

10%

NLorNLG

30% PEG

NLorNLG

10%

MRS

10%

P. jensenif

NLorNLG

10%

NLorNLG

30% PEG

NLorNLG

10%

NLorNLG

30% PEG

jensenii (petersonii)

P. jensenii

(wentii)

P. sp. Z/S

thoenif

NL

or

or

10%

or

10%

10%

NLG

P.

glycerol

glycerol glycerol glycerol glycerol glycerol

glycerol

NL-broth; NLG: NL-broth containing 2.5% (w/vol) glycine. MRS: MRS broth; YELG: YEL-broth with 2.5% (w/vol) glycine.

NL:

Polyethylenglycole lO'OOO (Zirnstein and Rehberger, 1991). PEB: PEB-buffer (Luchansky et al., 1988). GS: 10% glycerol and 0.5 M sucrose (Gautier et al., 1995).

PEG:

by plating

Prepared competent Propionibacterium-strains

acidipropionici (arabinosum)

P. sp.

on

is shown in table 8.

P.

P.

kept

ice out

without DNA. A list of the competent cells used in this

respective competence method

"Competent"

P.

in the selected

or

2.5x PEB

35

40

|il


(~1012 cfu) were used for the electroporation experiments.

of competent cells

petent cells and approximately 1 |ig of desalted DNA cuvettes

(Equibio)

Rad: 2.5

kV, 25 uF, 200 Q,

cells

were

kept

0.1 ml of the

cells

was

and

kept on ice 2

for 1 minute.

mm

gap

for 2-3 hours at 30°C in

regenerated

harvested

cells

were

served. As

negative controls (no growth

the incubated for 14

ted without DNA and incubated

propionibactena

inserted vector DNA

putative

probe.

days

the or

a

3-6

should

ms

applied.

medium

on

new

electroporation

pulse (Gene Pulser,

After

Bio-

electroporation,

the

(Gautier et al, 1995).

selective medium. The rest of the

room

temperature, 5'000 g), resuspen-

appropriate until

mixed in

selective medium. The regene¬

growth of the negative

control

was

ob¬

occur), competent cells were electropora-

the selective medium.

were

analyzed by hybridization

For further

of the clones with the

confirmation, plasmid isolation was performed

transformants.

Contamination

developed

as

on

was

regeneration

on

were

on

cuvettes)

by centrifugation (5 minutes,

rated cells

Recombinant

Then,

directly plated

ded in 0.1 ml of NL-broth and plated

were

Com¬

by non-propionibacteria was routinely

in this thesis.

checked

using

the MPCR method

Leer

-

Vide

-

Empty

37

Results

3. Results 3.1 Strain maintenance and control

Propionibacteria are relatively

growing organisms (up to

slow

tain pure cultures and to avoid contaminations, all cultures croscopy for the absence of cocci and examined

production, colony In table terns

color and catalase

9, these characteristics

Table 9:

using simple

(KOH, slime

tests

activity). complete

fermentation pat¬

(data not shown).

determined

were

In order to main¬

routinely checked by mi¬

were

shown. For certain strains,

are

days).

14

characteristics used for

Phenotypic

a

simple

"strain control" of

Propioni¬

bacteria Classical

P.

P. acidi-

P.

propio-

hexani-

freuden-

nici

cum

reichii

KOH1

G+

G+

G+

Slime

d

+

Characteristic

Color

P.

cyclo-

-

Propionibacteria "P. sher-

P.

P.

freuden-

manii"

jensenii

thoenii

reichii

G+

G+

G+

G+

+

d

d

+

cream

JS532

milky

light

light

light

white

brown

brown

brown

++

++

+

+

-

-

-

-

+

+

+

+'

Catalase

-

or

light

red

red

brown ++

+

_4

Lactose

+

nd

L-Arabinose

+

nd

-

-

-

-

-

Esculin

+

+

+

+

+

+

+

Glucose

+

+

+

+

+

+

+

Galactose

+

+

+

+

+

+

+

Maltose

+

nd

-

-

+

+

Nitrate reduction

+

Fermentation of: -

Cutaneous P.

Characteristic

KOH1

P. avidum

acnes

P.

granulosum

G+

Hemolysis

+

+

Catalase

+

+

+

Nitrate reduction

+

-

-

+:

Positive reaction; ++: very strong reaction; nd: no data. G+:

no

slime 2

reaction

-:

negative;

occurs.

1997.

The type strain is characterized 4

Data from Kusano et

al., 1997.

by

P.

lymphophilum G+

P.

j

aropionicum

G+ +

occurred, the strain is Gram-positive.

production Smutny,

Data from

G+

+

-

-

Propionibacteria

G+

Esculin

+

very weak

activity.

weak +

+

-

-

-

d: detected for

some

An

of the strains.

interpretation can be difficult,

when

Results

38

3.2 Identification of Propionibacteria 3.2.1

Sequencing

Due to

for

existing large

16S rDNA

Propionibacterium.

In

Gram-positive bacteria

cycle sequencing strategy, Propionibacterium analyzed.

were

16S

were

The data

from the genome and

of the 16S rDNA

elaborated and were

in

a

good agreement with

P.

applied

A total of 10883 nucleotides

sequences

Ace.

no.

already published.

.

(GenEmbl) rDNA sequences determined in this

avidumT

P. freudenreichii

subsp. shermanif

freudenreichii subsp. globosum

JS53

P.jenseniiDFl

study: AJ003055

1504

bp

Y10819

1510

bp

AJ009989

1505

bp

1502

bp

-

granulosum1 lymphophilum1 P. propionicumT

,,

lePëth

P.

AJ003057

1514

bp

P.

AJ003056

1502

bp

AJ003058

1506

bp

X53221

1351 +134

X53217

1462

X53219

1364+126

X53220

1467

16S rDNA sequences that

were

completed:

acidipropionici1 P. freudenreichii1 P. jensenii1 P.

P.

thoenif

16S rDNA sequences used without

+

+

26

54

bp

bp

bp

bp

change:

P.

acnes1

M61903

1530+

0bp

P.

cyclohexanicumT

D82046

1471

0

Type

PCR

Propionibacteria

.

P.

selected. To test the

was

analyzed (table 10).

Stram

Partial216S

cloning,

the 16S rRNA gene sequences of the type strains of the genus

Table 10:16S rDNA sequences of „.

tree for the

evolutionary

could be confirmed.

rRNA gene

cycle sequencing

and

a correct

sequences

addition, the taxonomic position of Propionibacteria within

avoid isolation of the

amplification

libraries, the determination of corresponding

should allow the construction of

the group of high G+C To

molecular methods

of 16S rRNA genes of Propionibacteria

Propionibacteria

genus

using

+

bp

strain.

1

Complete

2

Only

16S rDNA sequences

sequences

longer than

In addition to the

newly

sequences of the database

medically

relevant

or

1500

missing parts (+) that were sequenced in this study.

bp

are

shown.

determined 16S rDNA sequences, were

updated.

These sequence

Propionibacterium species

where

incomplete

or

poor DNA

updates mainly concerned the

only

sequence

stretches

of

39

approximately P.

350

base

lymphophilum whereas

pairs

Results

known

were

for

avidum,

P.

P.

granulosum and

the known sequence of P. propionicum contained

a

gap of 120

undefined nucleotides. The

analysis

of these sequences

of the genus

was

used to construct the first

complete phylogenetic tree

Propionibacterium containing all type strains belonging

including the medically

relevant

Propionibacteria (figure 2).

P. freudenreichii P. freudenreichii

subsp. shermanii

P. freudenreichii

to this genus,

freudenreichii

subsp, P.

cyclohexanicum

subsp.

globosum JS53

P.

acnes

P.

lymphophilum

P. avidum

P.

propionicum

i^-P. jensenii P. jensenii DF1 P.

P. thoenii

granulosum P.

Figure

2:

Unrooted tree

Propionibacterium. rRNA genes of

revealing

1%

acidipropionici

the

The tree is based

phylogenetic relationships

on an

of 1511

alignment

Propionibacteria (table 10). The alignment

eclustalw program

(GCG). The

tree

topology was

calculated

among the genus

basepairs

of the 16S

was

constructed

using

using

the dnamlk

algorithm

the

(PHYLIP, maximum likelihood method with molecular clock). Terminal branchings marked with

*

indicate that the 16S rDNA sequences

shown distance reveals not the correct

phylogenetic

are

so

closely

related that the

distance between these strains. The

bar indicates 1% base differences. The

phylogenetic tree

contains the information

the selected strains. There is

primer

sequences

only

a

lack of

on

almost the entire 16S rDNA

approximately

(37 nucleotides, 2 ambiguities) used

to

regions

of

the first 8 nucleotides and the

generate the PCR product. A

40

Results

medically

clear distinction between classical and

medically P.

lymphophilum

The a

species

relevant

analysis

is the most

directly

clusters

distantly

of the 16S rDNA

related

values between the

©

1

c

SI

Propionibacterium showed

F

I

1

s

I

•a

-s:

s s:

*

«s

s

•a

:a

8

s:

•"-1

Propionibacteria,

A distance matrix with the determined

i

•a

c

f

genus

JS53 •s

classical

of the 16S rDNA genes of Propionibacteria

tun S

the

is evident. None of the

of the whole genus.

species

species.

species

species is shown in table 11.

Homology values (%)

Table 11:

with

relationships within the

clear distinction between the different

homology

relevant

r

?

a;

a;

a;

a;

a;

a;

a;

a;

a;

a;

a;

a;

a;

1

2

3

4

5

6

7

8

9

10

11

12

13

1

100

97.2

93.6

94.5

91.4

92.5

92.4

95.1

99.4

90.8

94.3

92.4

97.4

2

97.2

100

93.7

94.8

90.9

92.2

92.0

95.8

97.3

90.2

94.5

92.0

96.4

3

93.6

93.7

100

96.4

91.4

92.2

91.9

94.3

94.1

90.1

96.4

91.8

93.7

4

94.5

94.8

96.4

100

91.5

92.7

92.4

94.4

95.1

90.1

99.4

92.4

94.1

5

91.4

90.9

91.4

91.5

100

95.0

95.3

90.2

91.5

90.6

91.3

95.3

91.4

6

92.5

92.2

92.2

92.7

95.0

100

99.5

91.4

92.5

90.3

92.5

99.4

92.5

7

92.4

92.0

91.9

92.4

95.3

99.5

100

91.4

92.3

90.4

92.2

99.9

91.9

8

95.1

95.8

94.3

94.4

90.2

91.4

91.4

100

95.3

91.0

94.4

91.4

94.5

9

99.4

97.3

94.1

95.1

91.5

92.5

92.3

95.3

100

90.9

94.9

92.3

98.0

10

90.8

90.2

90.1

90.1

90.6

90.3

90.4

91.0

90.9

100

90.1

90.4

90.4

11

94.3

94.5

96.4

99.4

91.3

92.5

92.2

94.4

94.9

90.1

100

92.3

94.0

12

92.4

92.0

91.8

92.4

95.3

99.4

99.9

91.4

92.3

90.4

92.3

100

91.9

13

97.4

96.4

93.7

94.1

91.4

92.5

91.9

94.5

98.0

90.4

94.0

91.9

100

The

homologies

were

calculated

listed in table 10. Based program

The

(PHYLIP, method using

highest

(99.4

-

DNA

99.9%), the

P. avidum and P. none

homology two P.

new

(95%

a

the matrix

was

maximum likelihood

a

are

species shows in

a

a

calculated with the dnadist

algorithm).

homology),

a

or

two different

of 99.4% exists. Towards P.

higher homology than

border-

of the sequences

observed for the three P. freudenreichii strains

homology

species P. cyclohexanicum

sequence

alignment (eclustalw, GCG)

alignment,

values

propionicum

of the other

an

jensenii strains (99.4%) and between the

clearly locates this species The

this

on

using

91%

species

lymphophilum

(P. granulosum)

which

outgroup.

is most

closely

fact also observed

by

related to the P. freudenreichii group Kusano et al.

(1997).

41

3.2.2 Classification of isolates from To

investigate

based

(and

on

large

number of

partial sequencing

other

strategy

a

sources)

can

were

dairy products

isolates, the possibility of

of the 16S rRNA genes

investigated.

or

A total of 16698 nucleotides

isolate

Strains identified

Classified as

belonging

"P. arabinosum" DSM

P.

freudenreichii FAM1409 P. freudenreichii FAM1410 P. freudenreichii FAM1411 P. freudenreichii FAM 1412 P. freudenreichii FAM 1413 P. freudenreichii FAM 1414 "P. pentosaceum" DSM "P. petersonii" DSM

P.

P.

"P. rubrum" DSM "P. shermanii" DSM P. sp. DF1 P. sp. DF3 P. sp. DF15 P. sp. DF16

P. sp. JS53 P. sp. JS62 P. sp. JS127

P. sp. ZS

Identification of strains

tested. Isolates of

on

dairy origin

"partial sequencing" certain other Gram-

analyzed.

were

partial 16S

rDNA sequences

Sequence-Identity1

as

to the genus

In accordance

with4

Propionibacterium:

acidipropionici2

92.3%

(765 bp) (373 bp) 96.9% (770 bp) 93.3% (624 bp) 96.9% (779 bp) P. freudenreichii P. freudenreichii 77.2% (136 bp) 94.7% (357 bp) P. freudenreichii P. acidipropionici2 94.9% (783 bp) 92.9% (397 bp) P. thoenii 93.0% (739 bp) P. thoenii 92.1% (607 bp) P. freudenreichii 94.4% (1453 bp) P. jensenii P. jensenii 92.4% (1270 bp) 97.1% (582 bp) P. freudenreichii 97.8% (495 bp) P. freudenreichii 99.6% (1508 bp) P. freudenreichii 95.4% (153 bp) P. freudenreichii 98.6% (645 bp) P. freudenreichii P. jensenii / thoenii 90.3% (627 bp)

freudenreichii P. freudenreichii P. freudenreichii

belonging to

classification of isolates

Propionibacteria (and

Table 12: Identification of bacterial strains based Strain

was

a

As shown in table 12, this

be used for the identification of

positive species).

Results

97.1%

supplier (DSM) supplier (FAM) supplier (FAM) supplier (FAM) supplier (FAM) supplier (FAM) supplier (FAM) supplier (DSM) supplier (DSM) de Carvalho et al., 1995

supplier (DSM) Fessier, 1997 Fessier, 1997

Fessier, 1997 Fessier, 1997

Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy

other genera:

AC1

Lactococcus lactis

97.4%

BU2-60

Lactococcus lactis

97.1%

JS38

Lactobacillus casei

JS63

Lactobacillus casei 97.1%

(449 bp)

JS65


(580 bp)

JS84

Lactobacillus casei 98.0% (646

JS85


JS86

Lactobacillus casei

JS108

Lactobacillus casei

K214

Lactococcus lactis

PA2

Staph, epidermidis

(684 bp) (456 bp) 97.9% (628 bp)

(534 98.1% (698 97.1% (419 96.2% (522 97.8% (980

bp) bp) bp) bp) bp) bp)

Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy Chemotaxonomy

Highest identity value over xy basepairs found using the FASTA algorithm. Best sequence identity found for Eubacterium combesii, an obviously wrong database entry ! Correct would be: P. jensenii (92.6% in 739 bp). Identification by 16S rDNA sequencing confirms the findings of other authors or is in accor¬ dance with chemotaxonomical properties of the respective strains as determined in this thesis.

Results

Database most

42

queries (FASTA algorithm)

closely

related

with the

organism which usually

occurred due to wrong database entries were

obtained (smaller than 400

P. jensenii,

needed).

All identifications based

(supplier, chemotaxonomy of


The

bp).

especially long stretches

identification

newly

some

or

or

was

the correct identification. Errors

when 16S rDNA sequences of limited

For the discrimination of the

species

of the 16S rDNA had to be used

on

the

sequencing data

morphology)

isolates

determined sequences revealed the

based

on

were

length

P. thoenii and

(at least 1200 bp

compared

are

to other data

and could be confirmed. In table 12, the 16S

strains received from other

rDNA

sources

sequences

could be

is

correctly

shown.

All

classified to

level.

analysis

of the 16S rDNA gene turned out to be

identification of unknown isolates to

method developed in this

study

species

level.

a

fast and reliable tool for the

Especially

for

Propionibacteria,

allowed the fast identification of strains within two

the

days.

43

3.2.3 Construction of

Based

on

a

specific

gene

probe

Results

for the genus

the 16S rDNA sequences determined before

probe (gdl,

table

5) specific for the

genus

Table 13: FASTA-comparison of the

Propionibacterium

(chapter 3.2.1),

Propionibacterium

was

an

oligonucleotide

constructed.

Propionibacterium genus-specific probe gdl

with the GenEmbl database

Species Primer

Sequence (16S rDNA,

5'

->

%

3')

acidipropionici

.tgctttcgatacgggttgac.

P.

acnes


P.

acnes


P.

acnes


P. avidum


P.

freudenreichii freudenreichii P. freudenreichii P. granulosum


P.


P.

no.


100%,20bp X53221 100%,20bp M61903 100%, 20 bp Y12288 100%, 20 bp X53218 100%,20bp AJ003055 100%,20bp AJ009989 100%,20bp X53217 100%,20bp Y10819 100%,20bp AJ003057 100%, 20 bp X53219 100%,20bp AJ003056 100%, 20 bp AJ003058 100%, 20 bp X53220 100%,20bp X53228 100%,20bp Z73442 100%,20bp Z69317 100%,20bp M29560 95%, 20 bp AF059764 95%, 20 bp L34614 100%, 19bp Z73456 95%, 20 bp D82046 95%, 20 bp Z73375 95%, 20 bp AB011304 95%, 20 bp AB011303 90%, 20 bp X86620 90%, 20 bp AF005012 90%, 20 bp X53212

.tgctttcgatacgggttgac. .tgctttcgatacgggttgac.

jensenii

P.

lymphophilum


P.

propionicus


P. thoenii


israelii

Unknown bacterial Bacterial

Ace.

tgctttcgatacgggttgac

gdl

P.

Actinomyces

Identity


species

species

.tgctttcgatacgggttgac. .tgctttcgatacgggttgac.

M. chitae


.tgctttcgatacgggttgac. .tgctttcgatacgggttgac. Unknown bacterial species .tgctttcgatacgggttgac. P. cyclohexanicum .tgctttcgatacgggttgac. Actinomycete species .tgctttcgatacgggttgac. Unknown bacterial

species

Eubacterium combesii

Unknown bacterial Unknown bacterial

species species

Nocardioides Nocardioides

.tgctttcgatacgggttgac. .tgctttcgatacgggttgac. .

simplex

Nocardioides luteus

tgctttcgatacggg^gac.

.tgctttcgatacgggtogac. .

tgctttcgatacgggSgac

.

1

Corrected sequence for this

2

Doubtful database entry: a higher homology exists with propionibacteria-sequences than with sequences from the correct genus (up to 20% sequence-differences).

region (determined

In table 13 the best hits of a FASTA

in this

comparison of gdl

study).

versus

3'295'425

(GenEmbl; November 3,1998) are shown. Differences between the with black boxes. The

highest

50

sequences of 16S rRNA genes.

ranking matches

of

gdl

Only propionibactena

DNA-sequences

sequences

are

with the database

marked

were

all

and not further characterized

44

Results

isolates showed

a

propionibacterial one

100%

identity

in all 20 bases of the

only

16S rDNA sequences,

base-mismatch with the

Propionibacteria.

A

and Eubacterium

This indicates either

marked in table 13

combesii).

{Actinomyces israelii,

The whole sequences same

compared

were

genera and from

up to 20% more, with sequences from the

than with the sequences of their "own" genera

was

sequencing, strain purification or identification problems

and thus, their 16S rDNA

organisms

are

higher degree of homology,

Propionibacterium

genus

gdl, they

to other genera showed 100%

belong

with other 16S rDNA sequences from the

(FASTA-algorithm)

cyclohexanicum shows

specific probe gdl.

with the sequence of

Mycobacterium chitae

sequence. From all known

the sequence of P.

Certain 16S rDNA sequences of the database that

identity

primer

sequence-entries

were

obvious.

with these

arbitrarily regarded

as

wrong

and were, therefore, not further taken into consideration in this study.

3.2.4

Multiplex PCR (MPCR)

To reduce time for the detection of the genus

identification of isolates, from other genera based

a

rapid

on a

Using

targeted

this

bak4 (also means

specific primer

the above mentioned

on

915 base

the 16S

an

specific 900-bp

pairs, depending universal 1500 -

the

rDNA), fragments of the

sized

of the

gene

probe gdl,

(table 5).

primers

16S rDNA

were

bakllw and

amplified by

figure 3, Propionibacteria are characterized by

the

fragment (the size of the fragment varies from 889

Propionibacterium species) whereas for other

bp PCR-fragment ("housekeeping" gene)

amplified (bakl 1 w bak4).

verify

Propionibacterium

genus-specific

and two universal bacterial 16S rDNA

of multiplex PCR. As shown in

presence of a

only

on

the 16S rRNA gene of Propionibacteria

on

targeted

method to differentiate the genus

and to

multiplex-PCR (MPCR) approach was developed (Dasen et

al., 1998). The method is based which is

Propionibacterium

is

expected

to

genera to be

Results

45

bak4:

bakllw:

"universal" primer (fw)

"universal

h~ 1

"H

1000 1—i

1

1

1

1

1—i—h

1

h—*

1

2000

1

1

1

gdl: genus-specific primer 889

„specific"fragmentfor Propionibacteria

bp

„

1508

3:

products

universal"fragment

bp

Graphical representation

of

oligonucleotide-positions

and

shown for the 16S rDNA sequence of P. freudenreichii

(GenEmbl

Accession

resulting

a

in

|....23SrDNA

1

1

1

h*~

16S rDNA

Figure

primer" (rev)

889

no

a

shermanii

subsp.

Y10819). Multiplex PCR reaction targeted

(primers bakllw, gdl) and

amplification

on

16S rDNA

1508-bp sized (primers bakllw, bak4)

fragment.

A total of 150

Propionibacteria isolates and reference strains from culture collections were

tested with the MPCR assay and the desired In

900-bp PCR-fragment

always obtained.

addition, various species related to Propionibacteria or involved in dairy fermentations

were

tested

as

negative controls and no

detection limit for

false

Propionibacteria was

at 2

positive signals were detected (figure 4). x

10

cfu

As shown in

figure 4, only Propionibacteria (figure

characterized

by

controls

kb

was

the formation of the

(figure 4b,

fragment)

was

PCR assay. The

confirms that

c, and

(colony forming units).

4a and

specific 900-bp

sized

b;

up to strain

fragment.

no.

For the

d; starting with strain no. 43), only the housekeeping

amplified,

which

The

39)

are

negative

gene

proved the absence of inhibitory substances

(1,5-

in the

negative control (lane -) where no DNA was added to the MPCR reaction

no

contaminating DNA was

introduced into the assay system.

Results

Figure

46

4: Detection of

Propionibacterium

strains

using Multiplex

PCR

(primers

bakllw, gdl and bak4). DNA from Propionibacteria and other organisms

analyzed using on a

the MPCR

2% TBE agarose

protocol,

without

stained with ethidium-bromide and examined

gel. The numbering of the lanes corresponds

in table 3. Lane "kb": kb-ladder

(Gibco). (-)

rDNA spacer

the MPCR method

the identification of As shown

by

regions

developed

possible

can

were

for

in this

Propionibacteria

(1997),

thesis,

the fast and

Propionibacteria.

Propionibacterium

Rossi et al.

Propionibacteria

25

negative

control

(MPCR

In

isolates down to

the

16S-23S rDNA

be used to differentiate between the

amplified

and then

sequenced.

next

a

on

simple

identification of

(figure 5).

step,

species

the 16S-23S

a

level

method was

allowing

investigated.

intergenic

spacer

species. By

means

regions

of

of PCR the

The DNA sequences of the spacers of

strains, including all type strains were determined. Based

tree was constructed

based

region

isolates to genus level is

spacer

stands for the

to the strains listed

DNA)

3.2.5 Differentiation and identification of

Using

was

on

these data,

a

phylogenetic

47

Results

P.freudenreichii FAM1413

P.freudenreichii FAM1410 P.freudenreichii FAM1411 P. freudenreichii

P.freudenreichii FAM1414 "P. shermanii 2" "P. shermanii"

P.freudenreichii JS53 P.

lymphophilum "P. zeae"

P.jensenii

P.jenseniiDFl P.jenseniiJ)¥3 P.jenseniiZS "P. rubrum" P.

acnes

P.

propionic urn

P. avid urn P.

P.

granulosum

cyclohexanicum P. thoenii "P. pentosaceum"

"P. arabinosum"

1%

P.

Figure

5:

Phylogenetic

16S-23S rDNA "

":

old strain

separation).

tree of the genus

intergenic

spacer

Propionibacterium based

regions (eclustalw-alignment

designations which

are

no

longer

correct

The bar indicates 1% base differences. The

the eclustalw program (GCG) and the tree

algorithm (PHYLIP,

topology

acidipropionici on

sequences of

of 265

nucleotides).

(used

to

simplify

strain

alignment was constructed using

was

calculated

maximum likelihood method with molecular

using

clock).

the dnamlk

Results

Based

48

on

the

analysis

correct

species

cluster

or

as

of the

shown for

example

all the isolates

belonging to

figure 2,

species

shown in

the

cluster does not appear

In

intergenic

near

figure 6, the alignment

Propionibacteria,

in

it is

spacer

region

figure

5 for the strains combined in the P.

the P.

freudenreichii group.

grouped differently

are

the P. thoenii I P.

in

assign

to

figure 5,

used for the construction of the

16S-23S

intergenic

e.g. the P.

rDNA

ITS

jensenii

jensenii

of the tree.

regions

spacer

phylogenetic

strains to the

But contrary to the tree

acidipropionici branches

of the 16S-23S rDNA

Start

possible

of selected

of figure 5, is shown.

tree

region

I Start

of

alignment

for

tree

constrution

i

~iL

P.

freud.

FAM1413

.aaggngccttttcgccatHgtgcgtttcgtgggctgtg

TCCTGCG.

p.

freud.

FAM1410

freud.

FAM1411

Bt.AAGGAGCCTTTTCGCCATJGTGCGTTTCGTGGGCTGTG

TCCTGCG.

p. p.

freudenreichii

p.

freud.

FAM1414

"P

shervian

"P

shermami

+

2

i "

K3TGCGNNN

ÇG.CT& SGCÎG

freud.

p.

lympnopn^lum

SGC1

jensenii jensenii

DF1

-m

valid).

1120

a one-

but better

alignment

a

shown in

discrepancies between the four

freudenreichii subsp. globosum

seems

where the sequence of this strain is differentiated

by

a

possible

around

TG insert from

the other three sequences. Because of the close

the 16S rDNA

relationship

alignment

Base shifts may

even

of

between the

figure

occur

8

only

subspecies

of P

freudenreichii shown

in

few differences between the sequences exist.

between the

same

subspecies (e.g. position

580 of the

53

alignment).

investigated.

Due to these

The 23 S

determined in this genus

problems

rDNA

with the

sequence

study (1st complete

Propionibacterium)

and

freudenreichii (not published,

16S rRNA genes, the 23S rDNA

of P

freudenreichu subsp. shermanu

with the sequence of P

kindly provided by

W.

Ludwig,

TU

Munich).

TGACAAGCTACTAAGTGCGATCGGTGGATGCCTAGGCACCAAGAGCCGATGAAGGACGTTGTAACCTGCGATAAGCCCTG

snerm

TGACAAGCTACTAAGTGCGATCGGTGGATGCCTAGGCACCAAGAGCCGATGAAGGACGTTGTAACCTGCGATAAGCCCTG

freud

GGGAGCTGGTAAACGAGCTTTGATCCGGGGATGTCCGAATGGGGAÄACCTCGAAGGTGACCAGTTTAGCTACTGGCGACC

snerm

GGGAGCTGGTAAACGAGCTTTGATCCGGGGATGTCCGAATGGGGAAACCTCGAAGGTGACCAGTTTAGCTACTGGCGACC

freud

GCCGCCTGAATGTATAGGGCGGTTGGAGGGAACGTGGGGAAGTGAAACATCTTAGTACCCACAGGAAGAGAAAACAACCG

snerm

GCCGCCTGAATGTATAGGGCGGTTGGAGGGAACGTGGGGAAGTGAAACATCTTAGTACCCACAGGAAGAGNAAACAACCG

freud

tgattccgtgabtaEtggcgagcgaaagcggaagaggccaaaco

snerm

tgantccgtgabtaBtggcgagcgaaagcggaagaggccaaaco

freua

GGGTTGTGGGAAGCGTTTTGACTGAAGTGCCGTGAGGTCGGAGAGTGATAAAGGATTGATGAAGCAGAAGCGTCTGGGAA

snerm

GGGTTGTGGGAAGCGTTTTGACTGAACTGCCGTGAGGTCGGAGAGTGATAAAGGATTGATGAAGCAGAAGCGTCTGGGAÄ

freua

GGCGCGGCATAGATGGTGATACCCCTGTATGCGTAAGTTGATCTCTCTCTTAATGTTTTCCCAAGTAGTACGGAACCCCT

snerm

GGCGCGGCATAGATGGTGATACCCCTGTATGCGTAAGTTGATCTCTCTCTTAATGTTTTCCCAAGTAGTACGGAACCCCT

fr^jd

GAAATTCCGTACGAATCTGGCGi

GCACCCGTTAAGCCTAAATACTCCTTGGTGA^CGATAGCGGACAAGTACCBGTGA

snerm

GAAATNCCGTACGAATCTGGCGi

.ccacccgttaagcctaaatacnccttggtgaccgatngcggacaagtaccBgtga

GTGTGTGATAGCCGGCAGGTGTTGCATGTGCG

80

160

240

3"0

GTGTGTGATAGCCGGCAGGTGTTGCATGTGCG 400

48C

560

freua

GGGAAAGG

ICCCCGGGAGGGGAGTGAAATAGTACCTGAAACCGATCGCATACAATCCGTCGGAGCCTGC

snerT

GGGAAAGG

CCCGGGAGGGGAGTGAAATAGTNCCTGAAACCGATCGCATACAATCCGTCGGAGCCTGC

freud

CCTTGTGGTGGGTGACGGCGTGCCTTTTG AAGAATGAGCCTGCGAGTTAGTGGTGTGTGGCGAGGTTAACCCGTGTGGG

snerm

CCTTGTGGTGGGTGACGGCGTGCCTTTTGNAAGAATGAGCCTGCGAGTTAGTGGTGTGTGGCGAGGTTAACCCGTGTGGG

freud

GAAGCCGTAGCGAAAGCGAGTCCGAATAGGGCGTTTGAGTCGCATGCTCTAGACCCGAAGCGGTGTGATCTATCCATGGC

sherm

GAAGCCGTAGCGAAAGCGAGTCCGAATAGGGCGTTTGAGTCGCATGCTCTAGACCCGAAGCGGTGTGATCTATCCATGGC

freud

CAGGGTGAAGCGACGGTAAGACGTCGTGGAGGCCCGAACCCACCAGGGTTGCAAACCTGGGGGATGAGCTGTGGATAGGG

snerm

CAGGGTGAAGCGACGGTAAGACGTCGTGGAGGCCCGAACCCACCAGGGTTGCAAACCTGGGGGATGAGCTGTGGATAGGG

freud

GTGAAAGGCCAATCAAACACCGTGATAGCTGGTTCTCCCCGAAATGCATTTAGGTGCAGCGTCATGTGTTTCTTGTCGGA

snerm

GTGAAAGGCCAATCAAACACCGTGATAGCTGGTTCTCCCCGAAATGCATTTAGGTGCAGCGTCATGTGTTTCTTGTCGGA

P

freua

P

snerm

GGTAGAGCACTGGATGGTCTAGGGGGCTTACCAGC'TACCGAAATCAGCCAAACTCCGAATGCCGACAAGTGAGAGCATG 104 0 GGTAGAGCACTGGATGGTCTAGGGGGCTTACCAGCTTACCGAAATCAGCCAAACTCCGAATGCCGACAAGTGAGAGCATG

P

freua

P

snerm

P

freua

P

snerm

P

freua

P

snerm

P

freud

P

snerm

P

freud

P

sherm

P

freud

P

snerm

p p

64 0

"0

800

880

960

GCAGTGAGACGGCGGGGGATAAGCTTCGTCGTCGAGAGGGAAACAGCCCAGATCATCAGCTAAGGCCCCTAAGTGGTGAC GCAGTGAGACGGCGGGGGATAAGCTTCGTCGTCGAGAGGGAAACAGCCCAGATCATCAGCTAAGGCCCCTAAGTGGTGAC

1120

TAAGTGGAAAAGGACGTGGAGTTGCGGAGACAACCAGGAGGTTGGCTTGGAAGCAGCCATCCTTGAAAGAGTGCGTAATA TAAGTGGAAAAGGACGTGGAGTTGCGGAGACAACCAGGAGGTTGGCTTGGAAGCAGCCATCCTTGAAAGAGTGCGTAATA

1200

GCTCACTGGTCAAGTGATTCTGCACCGACAATTTAGCGGGGCTCAAGTCATCCGCCGAAGCTGTGGCATCTACGCGTGTA GCTCACTGGTCAAGTGATTCTGCACCGACAATTTAGCGGGGCTCAAGTCATCCGCCGAAGCTGTGGCATCTACGCGTGTA

1280

TCCGGCATCCTTTGGGGTGTCCAGGTGCGTGGATGGGTAGGGCAGCGTTGTGTGTGCGTTGAAGCGGCGGGGTGACCCGG TCCGGCATCCTTTGGGGTGTCCAGGTGCGTGGATGGGTAGGGGAGCGTTGTGTGTGCGTTGAAGCGGCGGGGTGACCCGG

1360

TCGTGGAGTGCACGCAAGTGAGAATGCAGGCATGAGTAGCGTATGACGGGTGAGAAACCCGTCCGCCGAATATCCAAGGG TCGTGGAGTGCACGCAAGTGAGAATGCAGGCATGAGTAGCGTATGACGGGTGAGAAACCCGTCCGCCGAATATCCAAGGG

1440

TTCCAGGGTCAAGCTAATCTGCCCTGGGTGAGTCGGGTCCTAAGGCGAGGCCGACAGGCGTAGTCGATGGACAACGGGTT TTCCAGGGTCAAGCTAATCTGCCCTGGGTGAGTCGGGTCCTAAGGCGAGGCCGACAGGCGTAGTCGATGGACAACGGGTT

1520

freua

GATüTTCCCGTACCGGCGCGAGAACGATCCTGCCGAG'

GAGTGATGCTAAGCATGCAAGGCGGTCGTGGGGGCTTCGGT

1600

snerm

GATATTCCCGTACCGGCGCGAGAACGATCCTGCCGAGi

GAGTGATGCTAAGCATGCAAGGCGGTCGTGGGGGCTTCGGT

p

freua

tcccthgatcgttgagtctgtaacccgatcttgtagtaggcaagctgcggagggacgcaggaaggtagtctggcaccgta

P

snerm

tccct|gatcgttgagtctgtaacccgatcttgtagtaggcaagctgcggagggacgcaggaaggtagtctggcaccgta

D

freua

°

snerm

P

freud snerm

P

freud

P

snerm

was

freudenreichu subsp.

freud

P

was

23 S DNA sequence in the database of the whole

compared

but

Results

1680

ttggtttgcggtgttaagcctgtagggtgtctggccaggtaaatccggtcggacgtgtgcctgagaggtgatgagtggtg ttggtttgcggtgttaagcctgtagggtgtctggccaggtaaatccggtcggacgtgtgcctgagaggtgatgagtggtg

1760

ccacttttgtggtacgtatccggatgatcctatgctgcctagaaaatcttcgtgagcgagttctcgagctgcccgtaccc ccacttttgtggtacgtatccggatgatcctatgctgcctagaaaatcttcgtgagcgagttctcgagctgcccgtaccc

1840

caaaccgacactggtggataggtagagaataccaaggcgatcgagataatcatggtgaaggaactcggcaaaatcctccc 1920 caaaccgacactggtggataggtagagaataccaaggcgatcgagataatcatggtgaaggaactcggcaaaatcctccc

Results

54

GTAACTTCGGBATAAGGGAGACTGGAGGCGTGACGGCAGTTTACTTGTCGGTGCGTCGATAGTCGCAGAGAATAGGCCCA GTAACTTCGGgATAAGGGAGACTGGAGGCGTGACGGCAGTTTACTTGTCGGTGCGTCGATAGTCGCAGAGAATAGGCCCA

2000

freud.

AGCGACTGCTTACTAAAAGCACAGGTCCGTGCTAAGTCGAAAGACGATGTATACGGACTGACTCCTGCCCGGTGCtB.GA

2080

sherm.

AGCGACTGCTTACTAAMGCACAGGTCCGTGCTAAGTCGAMGACGATGTATACGGACTGACTCCTGCCCGGTGCTgNGA

p.

freud.

AGGTTAAGGGGACGTGTTAGCACTTTTGTGCGAGGCACTGAACTTAAGCCCCAGTAAACGGCGGTGGTAACTATAACCAT 2160

p.

sherm.

AGGTTAAGGGGACGTGTTAGCACTTTTGTGCGAGGCACTGAACTTAAGCCCCAGTAAACGGCGGTGGTAACTATAACCAT

p.

freud.

CCTAAGGTAGCGAAATTCCTTGTCGGGTAAGTTCCGACCTGCACGAATGGAGTAACGACTTGGGCGCTGTCTCCACCATG

p.

snerm.

cctaaggnagcgaaattccttgtcgggnaagttccgncctgnacgaatggagtaacgacttgggcgctgtctccaccatg

p.

freud.

aactcggcgaaattgcattacgagtaaagatgctcgttacgcgcaHHggacggaaagaccccgggacctttactatagt

p.

snern.

aactcggcgaaattgcattacgagtaaagatgctcgttacgcgca^Bggacggaaagnncccgggacctttactatagt

p.

freud.

p.

sherm.

p.

freud.

p.

sherm.

p.

freud.

p.

sherm.

p.

p.

24 00

GTTGAAATACCACTCTGGTCGTTCTGGTTATCTAACCTAGGTCCGTGATCCGGATCAGGGACAGTGCCTGATGGGTAGTT GTTGAAATACCACTCTGGTCGTTCTGGTTATCTAACCTAGGTCCGTGATCCGGATCAGGGäCAGTGCCTGATGGGTAGTT

2 4 80

TGACTGGGGCGGTCGCCTCCCAAAAGGTAACGGAGGCGCCCAAAGGTTCCCTCAGCCTGGTTGGTAATCAGGTGTTGAGT TGACTGGGGCGGTCGCCTCCNAAAAGGTAACGGAGGCGCCCAAAGGTTCCCTCAGCCTGGTTGGTAATCAGGTGTTGAGT

2560

GTAAGTGCACAAGGGAGCTTGACTGTGAGACAGACATGTCGAGCAGGGACGAAAGTCGGGACTAGTGATCTTCTGGTGGA GTAAGTGCACAAGGGAGCTTGACTGTGAGACAGACATGTCGAGCAGGGACGAAAGTCGGGACTAGTGATCNTCTGGTGGA

2 64 0

TTGTGGAATCGCCAGAACTCAACGGATAAAAGGTACCCCGGGGATAACAGGCTGATCTTTCCCGAGCGCTCACAGCGACG TTGTGGAATCGCCAGAACTCAACGGATAAAAGGTACCCCGGGGATAACAGGCTGATCTTTCCCGAGCGCTCACAGCGACG

2720

GAATGGTTTGGCACCTCGATGTCGGCTCGTCGCATCCTGGGGCTGGAGTCGGTCCCAAGGGTTGGGCTGTTCGCCCATTA GAATGGNTTGGCACCTCGATGTCGGCTCGTCGCATCCTGGGGCTGGAGTCGGTCCCAAGGGTTGGGCTGTTCGCCCATTA

2800

p.

freud. sherm.

p.

freud.

p.

sherm.

p.

freud.

p.

sherm.

p.

freud.

p.

snerm.

p.

freud.

p.

sherm.

p.

freud.

GGGCTGTCCTTAGTACGCAAGGACCGGGACGGACCäACCTCTGGTGTGCCAGTTGTTCCACCAGGAGCATGGCTGGTT

p.

sherm.

Jgggctgtccttagtacgcaaggaccgggacggaccaacctctggtgtgccagttgttccaccaggagcatggctggtt

P.

freud.

P.

sherm.

freud.

P.

sherm.

Figure

aagcggcacgcga.gctgggl.taagaacgtcgtgagacagttchggtccctatbccgctgcgcglg^ggaatcttgag 2880

aagcggcacgcgangctgggBntaagaacgtcgtgagäcagtt JggnccctatBccgctgcncgHn^Bgnatcttgag

GGCTACGTTGGGGAGTGATAACCGCTGAAAGCATCTAAGTGGGAAGCACGCTTCAAGATGAGGGTTCCTGCACAGTTAAT GGCTACGTTGGGGAGTGATAACCGCTGAAAGCATCTAAGTGGGAAGCACGCTTCAAGATGAGGGTTCCTGCACAGTTAAT

9:

Alignment kindly

P. freudenreichii

are

of 23S rRNA genes of P. freudenreichii

provided

by

subsp. shermanii

W.

Ludwig,

(this study,

TU

304 0

marked with black boxes

(ambiguities

were

both sequences determined with the FASTA

The evaluation of these would be

sequencing

not

subsp. freudenreichii1

Munich,

GenEmbl

aligned using the eclustalw algorithm (GCG).

probes

2 960

GTGGTAAGGCCCCCGGTAGACCACCGGGTGATAGGTCGGATGTGGAAGCATGGTGACATGTGGAGCTGACCGATACTAAG 3120 GTGGTAAGGCCCCCGGTAGACCACCGGGTGATAGGTCGGATGTGGAAGCATGGTGACATGTGGAGCTGACCGATACTAAG

(sequence

were

2320

TTGGTATTGGTGATCGGTACGACTTGTGTAGGATAGGTGGGAGACTTTGAAGCGGTCACGCTAGTGATTGTGGAGTCATT TTGGTATTGGTGATCGGTACGACTTGTGTAGGATAGGTGGGAGACTTTGAAGCGGTCACGCTAGTGATTGTGGAGTCATT

p.

P.

22 4 0

Germany)

Y10819). The

and

sequences

Differences between the sequences taken into

algorithm

alignments revealed regions

account). The identity of

is 98.5%.

where the construction of

possible (e.g. alignment positions 569-583

data from other strains should be obtained first.

of

figure 9),

specific

but additional

55

'Applied

Genetics'

The construction of a

genetic system for Propionibacteria requires first the construction of

suitable vectors before transformation

performed. presented In

Results

The isolation and

analysis

experiments plasmids

of

with these

from

organisms

Propionibacteria

conjugation or electroporation experiments

in this thesis before

addition, the fast identification methods for Propionibacteria that

previously (chapter 3.2)

are

needed to

ensure

are

are

be

is therefore

shown.

were

that the "correct" strains

can

developed

used for the

experiments.

3.3

Analysis

3.3.1 As

a

our

of Propionibacterium

Screening

for

preliminary to

plasmids

in

find suitable

plasmids

propionibacteria

plasmid vectors all defined Propionibacterium

collection and all other isolates were screened for the presence of plasmids

Three

plasmids

were

selected and further characterized:

strains of

(table 14).

pLMElOl (40 kb; Smutny,

1997), pLME106 (7 kb; Stierli, 1998) and pLME108 (2 kb; this study). Table 14: Plasmids isolated from no source

propionibacteria

of strains

investigated

of strains

plasmids

References

study

200

14

this

cheese

130

15

Smutny, 1997

Culture collections1

27

2

this

cheese

30

8

Pérez-Chaia et al., 1988

50

6

Rehberger and Glatz,

446

1352

50

4

cheese,

raw

milk

no

with

industrial strains "

"

raw

milk

"industrial" strains 1 2

As listed in table

30 strains

were

3,

Fessier, 1997 Gautier and Rouault, 1990

the FAM for further

analysis

in this thesis.

14, plasmids have been found only in small numbers in strains of

propionibacteria.

investigated

1985

species were not investigated for the presence of plasmids.

kindly provided by

As shown in table

classical

cutaneous

study

The

medically

for the presence of plasmids.

relevant

species have

so

far

not

been

56

Results

3.3.2 Antibiotics and

Propionibacteria

Antibiotics

as

were

used

markers in gene transfer

the isolation of

experiments to simplify

transformed cells. In table 15 antibiotic resistance patterns of Propionibacteria are shown.

Table 15: Antibiotic resistance patterns of medically and classical Antibiotic

Cutaneous

Dairy

Propionibacteria

species

granulosm proincum Cutaneous P.

acnes

avidum

P.

P.

Ampicillin

S

s

S

Bacitracin

S

s

Benzylpenicillin

S

s

Cefotaxime

s

Cephalothin Chloramphenicol Ciprofloxacin Clindamycin

P.

P.

Propionibacteria

Propionibacteria2

JS53

acidpro nic

0

P.

P.jens i

s

s

•**

S

a;

freudnich Rfreudnich P.

s

thoeni P.

S

S S

S

S

S

nd

nd

s

s

nd

S

nd

s

nd

nd

nd

nd

nd

nd

nd

s

nd

nd

nd

nd

I

nd

R

nd

S

V

S

S

S

S

S

S

S

nd

V

nd

s

s

S

S

S

S

S

V

S

S

S

V

I

nd

nd

nd

s

nd

nd

nd

nd

nd

R

S

S

S

S

S

s

S

nd

S

nd

S

I

nd

nd

nd

nd

nd

S

S

nd

S

nd

S

Dicloxacillin

S

nd

nd

nd

nd

nd

nd

nd

nd

nd

nd

Erythromycin

S4

S

S

S

S

S

S

S

S

S

S4

Fusidic acid

S

I

I

nd

nd

S

nd

S

S

S

I

R

R

R

Cloxacillin

Gentamicin

V

I

S

nd

nd

I

nd

R

Kanamycin Lincomycin

R

R

R

nd

nd

R

R

R

R

R

R

S

S

S

nd

nd

V

S

I

S

R

V

Methicillin

s

nd

nd

nd

nd

R

R

R

R

R

R

Metronidazole

R

nd

nd

R

R

R

nd

R

R

R

R

Nalidixic acid

nd

nd

nd

nd

nd

R

R

R

R

R

R

Neomycin

R

R

R

nd

nd

V

V

S

R

R

S

s

nd

nd

S

s

S

S

S

S

S

S

Polymyxin Rifampicin Streptomycin

nd

nd

nd

nd

nd

V

V

I

V

S

V

nd

nd

nd

nd

nd

s

nd

s

nd

S

s4

I

I

S

nd

nd

I

I

I

S

R

I

Sulfonamides

nd

nd

nd

nd

nd

R

R

R

R

R

R

Tetracycline Tobramycin Trimethoprim

s4 R3

I

nd

R

R

V

S

s

S

S

S4

R

S

nd

nd

R

nd

R

nd

R

nd

S

S

I

nd

nd

nd

nd

nd

S

nd

nd

Trimet/Sulfa.

S

S

S

nd

nd

I

nd

R

R

R

S

S

s

S

S

S

Penicillin G B

s

Vancomycin nd: not determined 1

2

or no

s

nd

nd

s

S

data available. R: resistant; S: sensitive; I: intermediate; V: variable.

al., 1996; Ednie et al., 1998; Essers, 1982; Hoeffler et al., 1976; Hoellmann et al., 1998; Laforest et al., 1988; Lorian, 1996; Sutter and Finegold, 1973; Tunney et al., 1998; Wang et al., 1977. This study; Guldimann, 1998; Smutny, 1997; Reddy et al., 1973a & 1973c. Combined resistance patterns:

al., 1983; Dubreuil

et

(Hoeffler et al, 1976). study; Ross et al., 1998; Ross

et

Denys

3

Strains with intermediate MIC exist

4

High resistant strains

exist: this

et

al., 1997; Tunney

et

al., 1998.

57

The data obtained in this were

agar-

contrary, the or

study were combined with published data.

most recent values

broth-dilution and Etest

of the selected MIC-values

investigated strains)

Results

was

were

selected and

ranked

as

figures obtained by

concentration in

recommended

sensitive to the

published data methods like

higher than disk-test values. Interpretation

(minimal inhibitory

performed

Propionibacteria are usually

were

When

by

following

Lorian

n,g/ml

for 90% of the

(1996).

antibiotics:

ampicillin, bacitracin,

benzylpenicillin, cephalothin, chloramphenicol, clindamycin, cloxacillin, erythromycin, fusidic acid, sensitive to Resistance

lincomycin, penicillin G, rifampicin

trimethoprim against

and

kanamycin,

sulfonamides is observed. Most strains

are

also resistant to

and

are

nalidixic

acid

and

tobramycin, neomycin (most

were

found for cefotaxime,

trimethoprim/sulfamethoxazole.

concerning antibiotic resistance patterns

cyclohexanicum.

metronidazole,

Propionibacteria

ciprofloxacin, polymyxin B, streptomycin

P.

Most strains

and methicillin.

Variable resistance patterns between

No data

vancomycin.

tetracycline.

gentamicin,

aminoglycoside antibiotics)

and

were

obtained for P.

lymphophilum and

Results

58

3.3.3 DNA-DNA

hybridization

A Southern blot with DNA of the

and

pLME108 respective

pRGOl (used

host strains

was

Propionibacterium plasmids

assays within

as

Propionibacterium plasmids pLMElOl, pLME106, reference

plasmid)

hybridized, stripped

and chromosomal DNA of the

reprobed

and

with each

plasmid

DNA

([(X-32P]dATP labeled) as probe. As shown in table

(approximately

size

showed strong not

16, only the plasmids pRGOl and pLME106, both of roughly the

hybridize

results with

kb)

7

but with different restriction endonuclease

cross-hybridization signals with

with chromosomal DNA of the host strains. In

pRGOl

as

probe

are

shown

as an

digestion patterns,

plasmid-DNA probes did

figure

10 the

hybridization

example.

hybridization analysis of Propionibacterium plasmids

Table 16:DNA-DNA

Probe

Target

each other. The

same

pLME106

pLMElOl

pLME108

pRGOl

DNA

Plasmid DNA:

pLMElOl

+

pLME106

+

+

-

pLME108

+

pRGOl

+

+

-

Chromosomal DNA:

freudenreichii JS531

P.

-

Pjensenii DF12 P. P.

-

freudenreichii DF23

-

acidipropionici4

...

-

'host strain of pLMEl 01.

2hostofpLME106. 3hostofpLME108. 4hostofpRG01. +:

positive hybridization signal detected;

The

relationship

between the

-: no

signal detected.

plasmid pLME106

where the selected plasmid DNA is probed with

this probe. In the lanes 10 and 11 of

hybridizes

with itself

fragments (lane 10). respective

as

Neither lane 3

host strains

and

figure 10B,

well with the or

pRGOl

is shown in

figure

10B

[oc-P32]dATP labeled DNA from pRGOl.

plasmid pLME106, digested (lane 4)

DNA from

and

the

undigested (lane 5) hybridizes with

positive

digested (lane 11)

as

control is shown:

with the

pRGOl

undigested

DNA

9, which contain the chromosomal DNA of the

hybridized with pRGOl

as

probe.

59

kb

ccc

1

6

ccc

Results

7

kb

9

10

11

12

ccc

kb

B Figure

10: Southern

agarose

gel

hybridization analysis

stained with ethidium bromide.

bound DNA that

hybridized

Plasmid-DNA of

pLMElOl; lane

with the

pLME106;

lane 5:

[a-P32]dATP labeled plasmid

freudenreichii

//well-digest

of

plasmids. (A)

(B) Autoradiography

2: Plasmid-DNA of

lane 3: Chromosomal DNA from P. 4:

oi Propionibacterium

1%

of membrane-

pRGOl.

Lane 1:

pLMElOl, digested with Kpnl;

JS53

pLME106; lane

(host

strain of pLMElOl). lane

6: Chr. DNA from

P.jensenii

DFl; lane 7: pLME108; lane 8: tfwcll-digest of pLME108; lane 9: Chr. DNA from P.

freudenreichii DF2;

DNA from P.

DNA-ladder

lane 10:

acidipropionici

(2-10

pRGOl;

lane 11:

pRGOl digested with Sail;

DSM 20272. kb: kb-ladder

kb in 1 kb steps,

Promega).

(Gibco);

ccc:

12: Chr.

supercoiled

Results

60

3.3.4 Characterization of

Because "natural"

plasmid pLMElOl

plasmid transfer by conjugation usually requires larger plasmids,

plasmid pLMElOl

was

selected and further characterized in this

restriction endonuclease

plasmid pLMElOl

was

analysis (an example estimated to 40.3 kb

is shown in

by adding

study.

Based

figure 11)

of the

on

the

DNA

the size of the

fragment

sizes of Sstl

digested pLME 101 (Smutny, 1997).

Figure

11: Restriction endonuclease

electrophoresis ladder

of

(BioRad);

analysis

pLMElOl digested

Lanes 2-15:

digests

of

plasmid pLMElOl.

with different enzymes.

of pLMElOl. 2: Hindlll

Pulsed field

Lanes 1:

digest;

3:

gel

8-48 kb

BamHl; 4:

EcoRV; 5: Sstl; 6: Pstl; 7: Hindi; 8: Sphl; 9: Kpnl; 10: Hindlll / Sphl digest; 11: Hindlll I Pstl; 12: Hindlll I

Sphl

I Pstl; 13: BamHl I EcoRV; 14: BamHl I

Hindlll;

15: EcoRY I

Hindlll; 16: undigested pLMElOl; 17: Hindlll digest of X DNA (Gibco); 18: kb-ladder

(Gibco).

Based

on

simple

restriction map of pLMElOl

relatively

data of

single

and double

digests

was

of

pLMElOl

with restriction enzymes,

constructed and is shown in

few sites of often used restriction enzymes between

bp

figure

a

12. Note the

20'000 and 40'000.

Results

61

Sp/7l,1 Sa/1,2000

Kpn 1,6500 Eco RV.7500

Eco RV.9000

H/ndlll,9000

Kpn 1,9900 30000,Sa/l

Eco RV, 10600

Kpn 1,11000

Sam H 1,14500 Pst 1,15000

Sa/1,15500 Bam Hl,17000 Sam H 1,17000 Sam HI Bam HI.19000

Figure

12:

Restriction

P. freudenreichü JS53

of

map

(this study

plasmid pLMElOl, and

a

plasmid

isolated

from

Smutny, 1997).

Shotgun-cloning of pLMElOl inpUC18 (Smutny, 1997; Guldimann, 1998) produced

plasmids with 0.3

to

1.0-kb sized insertions of SM-digested pLMElOl and 86 clones with

0.3 to 3.0-kb insertions of

of 3815

basepairs

With to

one

Sstl

from these

exception,

already

from the

a

digest.

plasmids

single connecting fragment

cover one

the

sequencing

10 clones

were

one

database led to

(hypothetical

70.2 kDa

hypothetical proteins

clone a

were

a

total

these sequences don't

of pLMElOl.

and

analysis

(566 bp,

of cloned DNA revealed

comparison

figure

13.

nor

similarities

stop codon) with the

of relatedness with

protein from Synechocystis sp.).

no

of the amino acid sequence

188 aa, neither start

relatively high degree

is shown in

selected in this thesis and

sequenced. However,

known DNA sequences. The FASTA

plasmid of

Swissprot

18

The sequence

a

known sequence

alignment

of both

Results

62

pLMElOl

14

PTGVATDEYxPQVQFWTSRGFGVLSVNYSGSAGFGRAYRERLRGQWGIAD 1 |:.|.:..

Synechocystis

417

sp.

pLMElOl sp.

pLMElOl sp.

pLMElOl

Figure

13:

pLMElOl

Amino

with

a

1 1 1

:

1 1

.

1

1 1

:

.

1 1

:

1 1

1 1 1 1 1

.

.

63

1 466

4 67

VADCVNAARYLADQGLVDGEQLAISGGSAGGYTTLAALTFHNVFKAGASY 516

114

YGIADLVAMQEGGTHKFEATYNDGLLGPWPQARKVYEERSPIHHLDQLHA 163

517

YGVSDLTAL.ATDTHKFEARYLDGLIGPYPERKDLYERRSPVNHADQLTC

164

PMLILQGLDDAVVPPQQADELG

566

sp.

.

VDDCIDAAESLLSADLADQSKIAIMGGSAGGFTVLAALTRSSVFSAGICR 113

1

Synechocystis

:

64

:

:

11

1

1 1:. 1 I.I:

Synechocystis

I: | I I I | I:

PTAAAGNSLSLKIQYWTSRGFAYVDVNYGGSTGYGRDYRQRLNGQWGIVD

1.1 1

Synechocystis

.:

•

1

1

1

..

..:

1

1

1

:

1

1

•

...

I I I 1 1 I

1

1

:

1

1 1 1

1 1 1 1 1 1

:

1 1

:

|:

:

1

I 1 1 1 1

.

:.:

.

.

1 1 |:: |

1 1

1 1

:

1 1 1

•

565

185

587

alignment (bestfit algorithm)

sequence

1 1

1

PVIFFQGLEDKVVPPNQTEMMV

acid

1 1

hypothetical 70.2 kDa protein from Synechocystis

of

a

part of

sp.

(Swissprot

Q55413).

In

a

171-amino acid

between the

protein

overlap region

similarity

a

sequences of figure 13

was

of 71.9% and

calculated

identity of

an

53.2%

(bestfit algorithm).

Curing experiments with P. freudenreichii JS53, the host strain of pLME 101, followed by

hybridization with pLMElOl of the potential 1997),

8 strains did not

After

plasmid analysis,

P.freudenreichii smaller

hybridize two

with the

of the

AREl harbored

no

cured strains gave

probe

strains

and

were

clear result

(Smutny,

selected for further

analysis.

contained

plasmid whereas

in

no

no

or

smaller

plasmids:

P.freudenreichii

ARE2

a

plasmid was found. Carbohydrate fermentation patterns (API CH50, BioMérieux)

revealed

plasmid). sucrose.

differences between the

no

Differences

were

original

strain JS53

(with pLMElOl) and AREl (no

observed for strain ARE2 for the fermentation of maltose and

Antibiotic resistance patterns of AREl and JS53

ARE2 had

strangely acquired additional

Additional

analysis

of AREl revealed, that

still present in this strain. Due to this

detected for

resistances to

pLMElOl.

pLMElOl

problems,

no

were

the same, whereas strain

clindamycin

was

and

penicillin

G.

under certain circumstances

plasmid-encoded properties

could be

Results

63

3.3.5 Characterization of plasmid Small sized

plasmids

are

experiments. Therefore,

analysis. The

pLME108

commonly used

as

the 2-kb sized

plasmid pLME108

cloning

vectors,

mainly was

plasmid pLME108 (2051 bp)

whole sequence of the

study using shotgun cloning, cycle sequencing

plasmids

of the

(table 17) and an automated DNA sequencer (ALF Express,


was

determined in this

of the

The sequence

PFR6662)

and is the first published

was

resulting

Pharmacia Biotech,

submitted to the GenEmbl database

Sweden).

electroporation

for

clones

Uppsala,

(Accession

no.

sequence of a Propionibacterium

complete nucleotide

plasmid. A physical map of the plasmid pLME108 was constructed and is shown in figure 14.

Table 17:

Cloning

pLME 108 digest Hindi

digest

digest

Pvull

Hindi

Pvttll

digest

digest

Nhel digest PCR 1

product1

PCR

The

Size of inserted DNA 1400

digested pUC 18

bp

Hindi

digested pUC18

500

1200

bp

Hindi

digested pUCl8

800 -1500

bp

Smal

digested pCL 1921

600-1000

bp

Smal

digested

pCL 1921

400

Xbal

digested

pUC 18

2051

bp (complete)

1800

bp

pGEM-T Easy

fragments

-

bp

-

mapping

pLME108

of

pLME108 revealed single

previously. Especially region that

"attractive"

the sites for

is not part of the

with these enzymes

sites for restriction endonucleases that

were

Nhel, Xhol and £co47III

putative

rep gene shown in

successful, indicating that the

correct. The sites for the enzymes used to clone

and Pvull, In

different vectors

amplification of missing part of pLME108 (primers prl08fw prl08rev)

not found

was

using

Ligated in Aval

digest

Aval

of pLME108

were

addition,

no

or

figure

located in

14.

a

Digests of

sequence of pLME108

pUC18, Hindi,

Aval

previous experiments

were

pLME108

in

all found in the sequence.

sites for enzymes that did not cut

found when the sequence BamBl

are

were

Hinàlll

are

not

was

analyzed.

pLME108

Sites for

common

in

enzymes like

present in the DNA of pLME108.

EcoRl, EcoRV,

Results

64

2022,Aat 1943, Pvu\

Orf2

1933,ßflf/l.

H/ncll,5

.

Sc/1,164 1831,Aval

Mme 1,227

Bel 1,363

1682,Eco47lll 1632,/Wal 1632,X/7ol. 1595,Pi/ull...

C-Term

ßsf XI.538

1498,A//?el

>wvl,571 H/ncll,616 Motif III

1380,Ban II 1352,A/arl A va

1317,Sacll

Motif I

14:

Figure

DF2. Two

Physical

map of

putative

open

marked. The

putative the

origin (sso) and I to

IV, C-Term)

binding

sites for the

pLME108 by

The

analysis

means

forms

and

an

be

ribosome

positions

a

frames rep

binding

foreign

plasmid

(223-1434)

are

gene

and

isolated from P.

and

for

rolling

pLME108

prl08rev

used for

was

jensenii

orfl (1642-2013)

(RBS), the putative single

position

zero

primers prl08fw of PCR

sites

a

of characteristic motifs for

shown. The

enzyme that

"sticky"

can

reading

jAafMi918

Motif IV

plasmid pLME108,

circle

strand

are

replication

replication (motif

selected

completing

randomly.

The

the sequence of

also shown.

of pLME108 shown in

introduction of

1682),

are

reP

Motif II

1199,ßgr/l

1,730

figure

seems

produces blunt

14 revealed

possible:

the

only

cutting

ends and for^YTzoI

ends. Both enzymes cut

pLME108 only

two restriction sites where the

site for isco47III

(at position 1632) once,

are

putative

rep gene.

enzyme that

commercially

ligated with products of a digest with other common enzymes.

Eco47Hl and Xhol do cut outside the

an

(at position

In

available

addition, both

65

Table 18: The most

closely

related

putative Rep protein rep of

Length

Aligned1

Results

"plasmidic" Rep protein

sequences

of the

of pLME108

Gaps Similarity Identity

GenEmbl Isolated

from2

pAPl

460

aa

474

aa

14

57.0%

36.3%

U83788 Arcanohacterium pyogenes

pIJlOl

456

aa

398

aa

12

54.8%

31.0%

M21778

Streptomyces

sp.

pJVl

528

aa

476

aa

21

53.6%

26.6%

L06019

Streptomyces

sp.

pSG5

481

aa

361

aa

12

51.3%

28.5%

X80774

Streptomyces

sp.

pAB49

394

aa

440

aa

12

51.2%

27.2%

L77992 Acinetobacter baumannii

pCA2.4

336

aa

394

aa

15

49.1%

22.4%

LI3739

Synechocystis

pTDl

335

aa

413

aa

15

48.6%

19.6%

M87856

Treponema

1 2

GCG-bestfit

alignment

with the

complete putative Rep protein

sp.

denticola

of pLME108

(445 aa).

pAPl (Billington et al., 1998), pIJlOl (Kendall and Cohen, 1988), pJVl (Servin-Gonzâlez et al., 1995), pSG5 (Muth et al, 1995), pCA2.4 (Yang and McFadden, 1993), pTDl (MacDougall et al., 1992), pAB49 (no reference, only accession number is given).

By comparison of 23 possible open reading frames (orf s) of pLME108 with the GenEmbl and

Swissprot databases,

identified. An

putative

a

rep gene

ranging

from bases 223 to 1434

was

overlap of 159 amino acids shows 63.5% similarity and 42.1 % identity with

the putative Rep

protein

of

plasmid pAPl

(GenEmbl

from Arcanohacterium pyogenes

U83788), the localization of this overlap region is shown in figure 14. The alignment of the two

putative Rep proteins is shown in figure

15.

The second open reading frame of pLME108, orf2, shows sequences of the database. as

Orf2 was

selected based

start-codon, ribosome binding site (RBS)

frame

as

the

putative

on

the

no

homology

following criteria: methionine

at its start and

orß

is

processed

with

in the

same

rep.

In table 18 the overall similarities and identities of the closest related

Plasmids

to known

pLME108

are

shown.

Rep proteins of

66

Results

pLME108

89

AADKRKHRFSVRYWLWRHTSLKRVAFCGRVAASAVASVGVRCSDGRAGFA 138

89

AKSRRSERYELRDGLAEISTIESVRKCGRVPVAPLVSLRAKSDGKGAGYG

138

139

GLQSCGSVWACPVCNAKIMARRGLELGAAVETWTKHGGRVAFMTFTVRHS

188

(Rep)

.:I..|:.:I

I

pAPl

(Rep)

pLME108

(Rep)

1 1

(Rep)

pAPl

pLME108

(Rep)

:.

(Rep)

pLME108

(Rep)

Figure

to the

with other motifs

:

:

1.

1 1

:

:

on

1.::1

:

1

::

II

..:

1 .1 1 1 1 1

1..

:

.

.

1

:

1

:

1 |.

1

:

1 1 1

:

:

1 1

KGQGLKHLWDALSTAWNRVTSGRRWIEFKEQFGLVGYVRANEITHGKHGW

239

HVHLHVLVF

247

247

HVHSHVLII

of the

the

of the motifs

alignment (bestfit algorithm)

plasmids pLME108

or

the

uses

suspected

replicons

are

238

III::

of pLME108,

found

-III

:

1 1

:

189

replication, characterized by

alignments

:

and

rolling to

use

of all these

circle the

of

pAPl (GenEmbl

Billington et al. (1998), the plasmid pAPl,

one

:.::.

238

plasmids, known

were

1... 1

:

|

RKDS LTAVWDGVAS GWRRVT S GKGWT S DQLRHGVEG FVRVVEVTHGRNGW

putative Rep proteins

protein

III

:..::.

189

15: Amino acid sequence

As assumed by

1 1 1

| 1 | |

188

239

(Rep)

•

1

GLHTCGSVWACPVCSAKIAARRKTDLQQVVDHAVKHGMTVSMLTLTQRHH

III

pAPl

1 1 1 1 1 1 1 1 1 1

..:..

139

:

pAPl

|

figure

no.

U83788).

which has the most similar

Rep

replication mode. By comparison

same

plasmids.

replication mechanism, typical

The

sso

(single strand origin)

the sequence TAGCGT, is present also in

shown in

parts of the

pLME108.

The

16.

The localization of the motifs I to IV and for the C-term motif identified for pLME108 indicated in

figure

alignment are for the

14. Amino acids conserved for at least 50% of the

positions

plasmids

occurs

The

of the

alignment

are

indicated

before motif III and is indicated in

plasmids originate

Streptomyces

Synechocystis

are

in the

marked with black boxes. The distances in amino acids between the motifs as

well. Almost all motifs appear at

approximately the same distances in all plasmids. Only the C-Term motif of pAPl it

of

sp. sp.

from the

(pIJlOl,

(pCA2.4)

and

figure

16 with

a

differs:

(-).

following species: Arcanobacteriumpyogenes (pAPl),

pJVl,

pSG5),

Acinetobacter

Treponema denticola pTDl.

baumannii

(pAB49),

67

Results

Aaa

pLME108 pAPl pAB49 pCA2.4 pIJlOl

.57

pJVl pSG5

.66

pTDl

.56

.57 .57

.51 .84

.68

Motif

Aaa

III

Aaa

23*

pLME108

.60

297

pAPl

.77

314

.-48

pAB49

.55

242

.50

pCA2.4 pIJlOl

.79

.64

pJVl

.126

.81

pSG5

.78

.62

.52

pTDl

C-Term Motif

pLME108 pAPl pAB4 9 pIJlOl pJVl pSG5

349

jflF^WjgKGSR

2 66

3kJ3k3sF

2 92

FQ^FAISMK

289

g

353 260

;1GRRAI

3AQ»|EgLA

_

YRJ3rJ3fGVPQVJ3kHYR

Figure 16: Alignment of

replication proteins

of

the

rolling

putative Rep protein of pLME108 against other circle

replication plasmids. With

amino acids between the patterns in the

following

distance indicates that the

respective

3.3.6

The

used

are

negative

are

marked with black boxes. The

listed in table 18.

Electroporation

single restriction

of £". coli XLl-Blue with derivatives of pLME108

sites for Nhel and^ol

by mapping of the

sequence of pLME108,

pUC18. Only

Nhel,

for

sequences is shown. A

motif is located upstream of the actual motif. Amino

acids conserved for at least 50% of the sequences

plasmids

Aaa the distance in

the

pLME116andpLME117.

(figure 14, positions

were

electroporation

used to fuse the

was

1498 and

1632)

found

complete pLME108

with

successful, resulting in the plasmids

Results

68

3.4 Gene transfer 3.4.1 Test of selective

The main the

problem

supplements for

in all

separation of donor

conjugation experiments was and

recipient

the selection of the transformants:

cells from the real

transconjugants.

the inhibition of donor cells, the addition of different

mainly

tested. The results

was

the isolation of Propionibacteria

Table 19: Growth of

are

supplements to

To achieve

the medium

shown in table 19.

Propionibacteria

without selective

and other

species

various media with

on

or

supplements

Isolation-medium and

Enterococcus

E. coli

Lactobacillus

"Propioni¬

respective "selective" supplement

faecalis

XLl-Blue

bulgaricus

bacterium

JH2-2

YEL

+

PAL1

+4

shermaniiT" +

+

-

-

YEL

indicator2 + cloxacillin3

YEL

+

0.1%

YEL

+

0.05%

YEL

+

2% acetate

+

+

YEL

+

3% NaCl

+

+

YEL

+

6%NaCl

+

YEL

+

propionate propionate

YEL+ 10% NaCl

+

+

+

+

+

+

+

+

+

BHI

+

1

Pal

4

4

+

-

+

-

+

-

+

-

+

-

+

-

weak

-

-

-

-

-

+

+

+

+

+

Propiobac (Standa industrie, Caën, France).

0.02 3

+

-

-

MRS

+

g/1

u,g/ml

bromcresol green. cloxacillin.

Growth, but

"propionibacteria-typical"

no

As shown in table

19, all tested strains

change

color

grew well

on

of the medium.

YEL-medium, usually the medium

recommended for the isolation of Propionibacteria from

dairy products,

observed. The inhibition of Lactobacillus bulgaricus and E. coli

proved

PAL, the addition of cloxacillin

the

easy:

growth

on

sufficient to inhibit

only the addition

one or

both

species. Enterococcus faecalis

of 10% salt to the medium

also P. shermanii did not grow at this not inhibit E. to

yellow

faecalis,

occurs

when

to the medium

but E. faecalis

high

was

was

or

inhibition was

no

to be

use

of MRS

grew under all

too much for this strain.

relatively were

conditions,

Unfortunately,

salt concentration. The commercial PAL did

detectable because

Propionibacteria grow

on

a

this medium.

color reaction from violet

69

3.4.2

The

Conjugation experiments

ability

resistance genes that

of

acquire foreign

conjugation experiments.

propionibacteria using

were

between other genera and

of propionibacteria to

with various

filter

were

also characterized

transconjugants.

a

used

DNA

as

of features

and results

propionibacteria a

"natural" way

plasmids

All donor

selective markers.

experiments

using

Broad host range

mating technique.

by the presence

The

Results

were

strains

allowing simple

"presented"

to

(propionibacteria)

and

specific selection

shown in table 20. In

are

tested

carried antibiotic

plasmids

Recipient

was

none

of the

experiments transconjugants could be selected.

Table 20: Conjugation

experiments

propionibacteria

recipient

Plasmid

Donor x

as

using

filter

the

xP. thoenii

Negative E.

pFOl1

T

Selection2

Result3

controls:

faecium

YEL-Agar,

Tetl5

YEL-Agar,

growth, pRE39

Enterococcus sp. RE39

pRE39

P.

thoenii

no

TC after 3 weeks

no

MRS-Agar, Rif50,

no

growth growth TC after 3 weeks

EmlO

FuslOO, EmlO

no

MRS-Agar, Rif50,

no

RF

Negative controls: E. faecium RE39

growth

MRS-Agar, Rif50,

faecium RE39

P. thoenii

no

FuslOO, EmlO

FuslOO,

controls:


red cfu

MRS-Agar, Rif50,

P. shermanii

P. shermanii

MRS-Agar, Rif50,

no

FuslOO, EmlO

Negative controls: E. faecium RE39 RF P. freudenreichii

E.

TC after 2 weeks

Tetl5

FOI

P. freudenreichii

Negative

no

red colonies


x

and

strains.

P. thoenii

x

technique

Recipient

E. faecium FOI

x

mating

pRE39

growth growth TC after 3 weeks

FuslOO, EmlO

MRS-Agar, Rif50, FuslOO, EmlO

growth growth

no

Results

70

Table 20 continued.. Donor x

Selection2

Result-

pK214

MRS-Agar, Cm20,

no

TC after 4 weeks

no

growth

no

TC after 4 weeks

no

growth

Recipient

Lactococcus lactis K214 x

Plasmid

P. shermanii

Negative

TetlO, FuslOO, Rif50

MRS-Agar, Cm20,

controls:

L. lactis K214


P. shermanii

Lactococcus lactis K214 x

P.

thoenii^

Negative

MRS-Agar, Cm20,

pK214


controls:

MRS-Agar, Cm20,

L. lactis K214

P. 1


thoenii^F

Containing transposonTnFO/. Transconjugants; RF: strain has been made resistant to Rifampi¬ cin and Fusidic acid (50 resp. 100 ug/ml); Cm: Chloramphenicol; Em: Erythromycin; Fus: Fusidic acid, Rif: Rifampicin; Tet: Tetracycline; numbers after antibiotics indi¬ cate their concentrations in the medium (e.g. TetlO: 10 ug/ml of Tetracycline). Experimental data from Guldimann (1998) were included in this table.

2

3

Abbreviations: TC:

The main

problem

in the


recipients Propionibacterium

the

transconjugants. Especially

negative controls, possibly It is

when

on

was

the

growth

of

the media used for the selection of the

erythromycin

was

used for the selection,

due to spontaneous mutations,

was

growth

of the

observed.

presumed, that conjugation may have occurred but the number of transconjugants was

significantly

than

smaller

transconjugant

3.4.3 a

the

number

for 500 spontaneous

be isolated under the selected

As

strains

shown in table 20,

Electroporation

of

mutants).

conjugation

spontaneous

Due to this fact

real

one

transconjugants

could

conditions.

foreign

genes into

conditions had to be established. Plasmids with of

no

(e.g.

of propionibacteria

base to the introduction of

containing parts

mutants

pLME108

were

a

propionibacteria, electroporation broad host range and

used to test the

plasmids

DNA-uptake capability

of

propionibacteria. In

a

first step, the

were

optimal electroporation conditions

determined experimentally and

finally selected for all experiments.

are

for the different buffer systems used

shown in table 21. A resistance of 200 Q

was

71

Table 21:

optimal electroporation conditions using

Determination of buffer systems

Resistance

25 uP, 2

(2.5 kV, 10%

mm

gap cuvettes, 40

of each

\û

GS

glycerol

different

buffer)

30% PEG

1000 Q

22.2

ms

22.8

ms

800 Q

18.3

ms

15.5

ms

11.5

ms

600 Q

13.9

ms

14.0

ms

13.0

ms

400 Q

9.4

ms

9.4

ms

9.1ms

200 Q

4.8

ms

4.8

ms

4.6

ms

100 Q

2.4

ms

2.4

ms

2.4

ms

"Short circuit"

Polyethylenglycole lO'OOO (Zirnstein and Rehberger, 1991). 10% glycerol and 0.5 M sucrose (Gautier et al., 1995).

PEG: GS:

In table 22 the conditions and strains for

transformants strains and

and

Results

were

electroporation experiments

observed under the selected conditions and with the

plasmids.

tetracycline

The

resistance gene

ligated with pLME108 (Nhel site).

directly

in

attempt

an

Propionibacteria was selected.

plasmid pLMElOl (40 kb) in E. coli and therefore

a

fused with

was

investigated

amplified using

PCR

Since the TetM resistance mechanism does not

function well in E. coli (Levy et al., 1989), vector

(tetM)

listed. No

are

The

to

introduce the pLME108::tefM

same was

attempted using the whole

plasmid pJIR751. This plasmid would not replicate

problematic direct cloning attempt was tried. Mainly due to

small amounts of DNA present when

no

E. coli passage is

performed

this

very

approach

is

difficult. The

use

of

pLME116

to

ampicillin pLME119

resistance gene is All

erythromycin.

In

integration

none

selectable marker for the

electroporation

Propionibacteria did not lead to

active in

shown

Gram-negative bacteria,

with

"+"

in

table

success.

was

due to the


in

was

Since this

plasmids

ampicillin

surprising.

22, resulted after selection with

plasmid

DNA

was

found.

also not observed. Growth of these strains

acquisition

chapter

of the

this fact is not

of these transformants the transferred

into the chromosome

the selective media

the

in

mainly

"transformants"

DNA

as

3.4.2.

of erythromycin-resistance

as

on

observed in

Results

72

Table 22:Electrotransformation of competent

Plasmid

pAM120

cyclohexanicum cyclohexanicum P. freudenreichiv

Tetl5, 28 days Tetl5, 28 days

"P. shermanii"

FAM1409 P.

jensenii

cyclohexanicum P. cyclohexanicum P. freudenreichiv P.

pAM1804

2

Selection2

P.

pAM120 pAM120 pAM120 pAM120 pAM180

1

Strain1 P.

pAM1204

pAM180 pAM180 pAM180 pJIR750 pJIR750 pJIR750 pJIR751 pJIR751 pJIR751 pLME116 pLME116 pLME116 pLME116 pLME116 pLME116 pLME116 pLME118 pLME119 pSI22 pSI22 pLME101::pJ!R751 pLME101::pJIR751 pLME101::pJIR751 pLME101::pJIR751 pLME101::pJIR751 pLME101::pJIR751 pLME108::fefM pLME108::fe*M pLME108::tefM

Propionibacterium

"P. shermanii" P.

jensenii

P.

freudenreichiv

"P. shermanii"

P.

jensenii

P.

freudenreichiv

Result3 -

-

Tetl5, Tet30, 28 days Tet30, 28 days

Tetl5, Tet30, 21 days Tetl5, Tet30, 21 days Tetl5, 28 Tetl5, 28

days days

Tetl5, Tet30, 28 days Tet30, 28 days Tet30, 28 days CmlO, Cm20, 28 days CmlO, Cm20, 28 days

P.

cyclohexanicum freudenreichiv

Amp50, Amp50, Amp50, Amp50, Amp50, Amp50, Amp50, Amp50, Amp50,

"P. intermedium"

jensenii

P.

jensenii

P.

jensenii

"P. shermanii" P.

P.

freudenreichiv freudenreichiv

"P. shermanii" P.

freudenreichiv

"P. shermanii" P.

freudenreichiv

-

-

-

-

-

+

jensenii

P.

-

+

Eml5, Em40, 28

JS53

-

CmlO, Cm20, 28 days Em5, Eml5, Em40, 28 d Em5, Eml5, Em40, 28 d

FAM1413

-

-

"P. shermanii"

ARE1

-

-

P.

P.

(EcoRV) (EcoRY) (HindUl) (HindUl) (Pstl) (Pstl)

cells

CmlO, CmlO, EmlO, EmlO, EmlO,

EmlO, EmlO,

days

28

days 28 days 28 days 28 days 28 days 28 days 28 days 28 days 28 days 28 days 28 days 10 days 10 days 10 days 10 days 10 days 10 days

-

-

-

-

-

-

-

-

-

-

+

+ + + +

EmlO,

FAM1413

JS53

Tetl5, Tet30, 21 days Tetl5, Tet30, 21 days

"jR intermedium"

Tetl5, Tet30, 21 days

(Xg/ml

-

-

"P. shermanii"

Made competent as listed in table 8. NL medium with selective supplement (concentrations in phenicol; Em: Erythromycin and Tet: Tetracycline).

-

+ -

-

-

of Cm: Chloram¬

growth on selective medium, potential transformants; -: no transformants observed. Parts of the data were included from Guldimann (1998). Treatment of cells for 10 minutes at 48°C prior to electroporation. +:

73

Discussion

4. Discussion 4.1 Isolation of Propionibacteria 4.1.1 "Selective" media and identification tests

The main

problem

Depending

selective media. can

encountered when on

be either isolated relatively

hard cheeses, where

Propionibacteria (1992),

are

the other

relatively

mainly

much faster than the

on

from

raw

milk

when the

media

as

(Giraffa

et

usually

microflora consists of bacteria that grow

and

sensitive to these substances than enterococci

interpretation activity

or

variable"

microscopy.

certain authors

by

slime.

shapes

a

commercially use

exposition

for

dairy products

are

grow

made

inhibitory are

more

staining,

cultured is the often unclear the determination of catalase

Propionibacteria

is

highly dependent been

even

on

the

called "Gram-

of the KOH-test

(Bamarouf et al.,

but strains of the classical

Propionibacteria

be

activity only

other

or

Propionibacteria

since

The

use

interpreted using

short, almost coccoid rods

reaction is observed

The

after

also found in

Propionibacteria have

helpful,

Tests like the catalase

only

in

is

Consequently, they could

positive

of

(Beveridge, 1990).

Gram-staining

and

mainly

morphology

may vary from very

morphology. strains,

test reactions like Gram

The

!). Especially enterococci

(table 19).

Propionibacteria

and the age of the cells.

instead of

produce Cell

simple

of

growth medium

1996)

that arises when

problem

are

possible,

not

lactic

be isolated from cheese.

al., 1997). The addition of antibiotics

substances to the medium is

Another

can

Cogan

growth of the

to inhibit

(YEL-like)

almost all do that

Propionibacteria

flora for the

the method of Drinan and

Propionibacteria

competing

Propionibacteria (and

same

samples, Propionibacteria

competitive

as

occur

Using

easy to process.

with cloxacillin added to the medium

Problems arise

in

easily or with great difficulties. Samples like hard or semi¬

acid bacteria, almost pure cultures of

well

organisms present

lactic acid bacteria

mainly

isolated is the lack of suitable

Propionibacteria are

can

after

also be

this test

to the

Gram-negative.

"typical" coryneform

misleading,

incubating the

as

because for certain

cells for 10 to 30 minutes

of the cells to oxygen.

available medium, PAL

dairy species

and

no

Propiobac (Standa industrie, France)

data

are

unknown Propionibacterium species will grow PAL is

quite expensive (sFr.

routine

analysis.

1.50 per

plate)

is tested

available how cutaneous strains on

this medium. As

or

still

already mentioned,

and therefore not the medium of choice for

74

Discussion

The

use

fermentation patterns characterize

described

as

Propionibacteria.

Propionibacteria in their

was

no

by

on

the determination of

Britz and Riedel

(1991) is still

Commercial identification kits, e.g.

(BioMérieux),

and API20A

Since

mainly based

of numerical taxonomy,

often include

only

carbohydrate

good approach

a

APICoryne (version 2)

information for

medically

relevant

databases.

real selective medium for the isolation of Propionibacteria from different

found,

a

molecular

approach was

Recently, screening techniques

sources

selected for the identification.

4.2 Identification of Propionibacteria based

on

using

molecular methods

molecular methods revealed the presence of

Propionibacteria in other habitats. These techniques, mainly based on the amplification 16S rDNA genes with universal

primers,

followed

allow the detection of small amounts of DNA, molecule in

a

by cloning

theoretically

of

sequencing analysis,

and

down to

one

single

DNA

sample.

closely

16S rDNA sequences rumen

to

related with those of

Propionibacteria

were

found in the

of red deer (Jarvis et al., 1998). These sequences showed 99.6% sequence

similarity

to P.

sequences in

a

acnes.

Boivin-Jahns et al.

(1996)

found

deep-surface clay sediment (97% similarity

similarity with P. acnes).

Propionibacteria with P.

Other sequences, all related to cutaneous

16S rDNA

granulosum

and 98%

Propionibacteria were

amplified by Sekiguchi et al. (1998) from sludge. A new species, P. cyclohexanicum, isolated from

spoilt

orange juice

by

Kusano et al.

was

(1997).

4.2.1 Identification to genus level

Different researchers have

Propionibacteria.

proposed P.

as

The

already described specific primers

primer

RDR514

hybridization probe

granulosum (Greisen

classical strains, and

et

an

(E. coli position 1376-1400)

to detect P.

acnes, P.

of these

rDNA

optimal.

fragments have been used (Riedel

for the genus. The genus in this

Primers for the

study

are

et

not tested with the

relatively long (25 bp), its suitability

amplification

of

Propionibacterium

al., 1994), but these primers

specific probe, gdl (table 5)

further discussed in

was

16S rDNA sequences shows certain

mismatches (Dasen et al., 1998). Since this primer is for PCR may not be

chapter 4.2.5.

tested and

was

avidum, P. lymphophilum and

al., 1994). However, this primer

alignment

for the detection of

The

are

not

and the MPCR strategy use

of gdl

as a

16S

specific

developed

hybridization probe

75

allowed the confirmation of for direct

Propionibacteria

Discussion

isolated from cheese

colony hybridization experiments, gdl

supposed that the target

region for gdl (E.

16S rRNA

satisfactory

no

gave

coli

samples.

position 632)

When used

results. It

exhibits

a

was

strong

secondary structure which prevents hybridization of the probe with the target or that slime produced by Propionibacteria inhibited

4.2.2 Identification to To

species

lysis.

cell

level

identify mainly dairy Propionibacterium species and in some cases species,

relevant

methods like

ribotyping

or

the 16S rDNA, of the 23S rDNA and of the been described

by

several authors

(Riedel

the restriction

intergenic

et

analysis

also the

of PCR

16S-23S rDNA

al., 1994 & 1998; Fessier

medically

products

of

region (ITS) have al., 1998; Rossi

et

étal., 1997).

primer pairs for the

PCR

Rossi et al. (1998) each P.

of the

acnes

have been

reported by Hykin et al. (1994).

proposed specific oligonucleotide primers

for the PCR detection of

dairy Propionibacteria species (P. acidipropionici,

P.

freudenreichii,

jensenii and P. thoenii).

The

hybridization

Propionibacteria P.

detection of P.

as

probes,

has

freudenreichii (figure 7).

species

as

The main

counting of

a

so

for the

advantage

in such

of bacterial colonies.

quantification This direct

a

a

was

only

the

using

ITS

regions

of

for the detection of strains of

in all

of these

species

as

probes.


The ITS

to allow

enough

hybridization. a

hybridization assay is its potential

for a direct

By replica-plating of colonies grown from a dilution series

transfer of the colonies to

for bifidobacteria

approach

regions

ITS

of a method based on

sample, followed by

thesis,

my

But this method should also allow the detection of other

enough (300-400 bp) probe

in

far been tested

well, by simply using the

sequence is short

specificity

developed

method

not

by hybridization tested in this

is

nylon membranes

and cell

possible (Kaufmann

study,

but

as

a

et

lysis,

a

al., 1997).

first step, DNA of

Propionibacterium isolates (200 strains) isolated by using the protocol of Goldenberger et al.

(1995)

was

transferred to

a

nylon

membrane and

probed

with the ITS

region of

P.freudenreichii. All strains of P. freudenreichii tested could be detected with this method (data not shown). Other methods for the identification of

dairy Propionibacterium species based

on

the

comparative analysis of protein profiles (Baer, 1987; Fessier, 1997) have been developed

76

Discussion

with

good results.

But this method

Propionibacterium based

cultures

techniques would

Propionibacteria.

requires, like some of the other methods the use of pure

prior to identification. Only

practical

use

requires

PCR

equipment (Brochure, Perkin Elmer) better in the

freudenreichii

certain strains is

are

and is therefore

on

researchers

This

were

before

were

small: at least 97.4%

addition, the bases

(N)

identity

sequences for the P.

that

analysis of the

were

same

species

subspecies could

(de Carvalho

from each other, The P.

sequence

a

freudenreichii subsp.

first does not ferment

1986). By using

23 S rDNA sequences of strains of P.

variety of

occur.

subspecies.

freudenreichii

strains contain 23 and 26

figure

It is

a

(Pettersson one

a

so

In the

et

copies

clear

over

occur even

of the

case

of the rRNA genes

(Nübel

et

can

on

their

also be different

al., 1996).

intergenic

grouping

same

their whole 16S rDNA

al., 1997), which

rDNA

the

far, these differences between

al, 1998)

et

to three

16S-23S

Especially

8 reveals that differences

exist

ambiguous

known fact that different strains of the

al., 1994 & Ross

of the

most

for the 16S rDNA, 98.5% for the 23S rDNA. In

strains did not result in

received from the FAM

lactose),

The differences found between these sequences

two sequences

harbor et

a

differences between the two

significant

fact discovered for other bacteria

analysis

freudenreichii

(the

P.

freudenreichii subsp. freudenreichii.

only

also

Propionibacteria

chromosome

subspecies

and Johnson,

freudenreichii

may differ in up to 24 bases

sequence.

division of this species in two subspecies

shermanii

16S rDNA sequences shown in

between the two sequences of P.

the

a

production and

not included in the calculation of these identities.

23 S rDNA sequences, where

perform

should

(protein profiling, RAPD, ribotyping, RFLP),

figure 9).

8 and

experience and expensive

starter culture in cheese

into the two

(Cummins

thesis, four 16S rDNA and two

analyzed (figure

as a

subsp.

unable to find other

were

lot of

a

species of

subspecies

lactose,

separation

other features

no

mentioned

methods

In this

use.

hybridization-

proposed in this study.

important

and P. freudenreichii

freudenreichii is based

is very

not able to ferment

still in

currently

or

simpler hybridization technique

the

4.2.3 Identification of the P. freudenreichii

Since P.

of PCR

use

allow the detection of mixed cultures of different

quantitative

Since

the

into 2

spacer

regions of

subspecies.

the

The strains

(Forschungsanstalt für Milchwirtschaft, Liebefeld, Switzerland),

claimed to be members of "P. shermanii" and used

with both of the type strains of the

subspecies.

as

starter cultures for cheese

grouped

Discussion

77

Rehberger and Glatz (1987 utilization in

&

propionibacteria,

1990) reported the an

plasmid-linked

a

observation which renders the

factor obsolete. A differentiation between with

presence of

a

aplasmid encoded lactose-utilization,

P. freudenreichii

said

a

single

lactose

clear distinctive

subsp. freudenreichii strain,

P. freudenreichii

subsp. shermanii strain

(lactose fermentation chromosomally encoded) is no longer possible. Based P.

experimental facts,

these

on

freudenreichii into

two

the

proposal

subspecies (Fessier, 1997)

to

abandon the

separation

of

is also recommended.

4.2.4 Identification of strains

The

of molecular methods allows the differentiation between strains of the

use

species.

PFGE

(pulsed field electrophoresis) is the most popular method.

distribution of rare

recognition

highly reproducible. based

on

(Jones

to compare strains of P.

freudenreichii

only

a

was

Propionibacterium

et

jensenii,

strains. The

5 and

4.2.5

further

Multiplex

Using 900

was not

a

PCR:

RAPD

"internal" method are

analysis

analysis

or

in the

fairly irreproducible

has been

use

of the

of the 16S-23S rDNA

possible approach

intergenic

performed

for

differentiation

intergenic

were

P.

as a

tool to et al.

freudenreichii

as

could be identified. Because not

available, the

use

spacer

differentiate between

reported by Leblond-Bourget

acidipropionici

propionibacteria

to

region

spacer

figure 6, different strains of

P. thoenii and P.

strains of the cutaneous

strains

as an

high

single laboratory study (Fessier, 1997).

identification among bifidobacteria has been

P.

be used

a

al, 1997). PFGE has been used by Rehberger

found to be another

figure

will lead to

stringent primers,

dairy propionibacteria and allowed the

In my thesis, however, the sequence

As shown in

non

freudenreichii.

strains and other

between these strains in

region (ITS)

with

cases

ideas, because the results of these analysis

between different laboratories

P.

the

particular restriction enzyme on a genome and is

of the results. RAPD should

absence of other

on

RAPD is also used for strain identification. In contrast to PFGE, it is

PCR which in the worst

variability

(1993)

sites of a

It is based

same

species (1996). well

as

enough

of this method for such

investigated.

a

fast and reliable

approach

to

identify propionibacteria

standard PCR protocol with primers gdl and bak4 alone,

bp approximately (889-915 bp, dependent

propionibacteria. Template

DNA from other

the

a

single fragment

species)

is

amplified

organisms resulted

in

no

on

of for

detectable

Discussion

78

amplification products. signal

due to

was

second PCR

developed

applied

pathogens

to detect

1996/ The is that the

same

specific

900

showed

only

from other

a

units

In my MPCR

fragment of

"universal"

fragment is

1500

used

developed

run a

(MPCR)

already

is

been

et

al.,

in this thesis

primers.

organisms by amplification and bak4. All other

bp resulting

of

amplification

internal standard. When the

as an

a

simplify

samples (Deng and Fratamico, 1996; Witham

distinguished

could be

PCR

The method has

simultaneously.

bp fragment amplified with primers gdl

to pure

and bak4

in food

necessary to

approach. Multiplex

like

gene, the 16S rDNA, is the target for all

bakl 1 w and bak4. This

forming

loci

or more

was

and bak4. In order to

difference between "classical" MPCR and the method

Propionibacteria

applied

primers bakl 1 w

multiplex-PCR 2

to determine whether the absence of

limited DNA extraction it

or

the universal

a

amplify

used to

possible

not

was

amplification

no

analysis using

the method I

commonly

Since it

cultures, the detection limit for Propionibacteria

was

at 2

a

organisms

with

primers

protocol x

of

was

colony

10

(cfu).

approach using Propionibacterium DNA, the oligonucleotides gdl, bakl lw

(figure 3), theoretically

fragments of approximately

2

900

bp

and 1500

bp

should be formed. Under the selected conditions and DNA-concentrations, only the

specific fragment concentrations

and

Propionibacteria, was no

of 900

bp

annealing temperatures,

but the

longer amplified

protocol in

The

found to

were

fragments

optimized until

comparison

of this

(November 1998) only nucleotides)

with

Propionibacteria, Of the three

7

24

the

occur

for other tested

similarity to similarity

L34614)

one

for

fragment

sequences

is shown in table 13. In

these

Of

homology (in

sequences,

14

sequence of Actinomyces israelii serotype 1

(Accession

other sequences from the acnes

with

false

organisms (figure 4).

showed at least 95%

sequence.

chitae

no

a

with all sequences of the GenEmbl database

showed 100%

of 96.3% to P.

DNA-

amplified

the "universal"

belonged to organisms that were not further identified

remaining,

no.

primer

target

X53228), Mycobacterium

(Accession

were

on

tested, the specific fragment was amplified and

alignment of the primer gdl (E. coli position 632-651)

FASTA

a

further

was

both

Depending

Propionibacteria.

In all strains of Propionibacteria

positive signals

in detectable amounts.

amplified

was

no.

identity

same

only

gdl.

genera than to

84%

belonged

to

to genus level.

(Accession

M29560) and Eubacterium

with

all 20

no.

combesii

These sequences show lower

Propionibacteria: A.

israelii has

similarity to other species from the genus

79

M. chitae

Actinomyces.

is

more

The

Mycobacterium species (87%). P. thoenii

(92.3%)

and shows

closely

related to P.

acnes

than to other

(98.3%)

sequence of E. combesii is related to

questionable

78%

only

Discussion

similarity with

other Eubacterium

species.

That

indicates either the need of reclassification and re-evaluation of these strains

sequencing problems. When

cells of M chitae

typical Propionibacterium banding pattern suspicion that

were

was

submitted to the MPCR

amplified

for this

protocol

species, hardening

(Accession no. M29560)

the 16S rDNA sequence entry

or

for this

no

the

species

is

wrong.

The 16S rDNA of the our

classified strain P.

new

sequence. Since this strain is

primer

cyclohexanicum relatively

new

has not been

shows

mismatch with

one

(Kusano

al., 1997) its

et

occurrence

in other habitats than orange

be detected

using the MPCR technique by lowering the annealing-temperature from 60°C

to 56°C

resulting

formation of P.

a

in

a

slightly

third

cyclohexanicum)

due to

annealing temperature. primers

with

a

of

tube, great

care

approximately

unspecific binding problem

mixture of A and C at

theoretically

Since PCR is reaction

This

specificity for the

lower

fragment

juice

sensitive

2 kb

of the

to detect

The strain could

and, in certain

(figure 4,

cases, the

P. avidum

oligonucleotides

be solved

18 of the

position

enough

detection

possibly

could

investigated.

at the

and

lower

by using degenerated

oligonucleotide gdl.

one

has to be taken in every PCR

single

molecule of DNA in

amplification

to avoid

even

a

the

smallest amounts of contaminating DNA. The combination of the MPCR method with the DNA isolation method of

Goldenberger

et al.

(1995) using only

SDS and Proteinase K

allowed to minimize the number of handling steps needed to obtain "PCR Probes could be treated in

one

method was successful with all would not be necessary

single

cap, with

species tested.

only minimal pipetting

The

use

grade"

DNA.

necessary. This

of this method for Propionibacteria

(the simple addition of Propionibacteria cells to the

MPCR would

guarantee suitable results) but especially the lysis of lactobacilli is otherwise difficult. The method

was

Without any detect the

tested with milk

purification,

the milk

was

Propionibacteria

submitted to the

have been cultivated.

protocol and it

was

possible

to

Propionibacteria.

In all the PCR

exactly

in which

samples

experiments, a negative control that contained no DNA but that was treated

like DNA

containing samples,

allowed the detection of contaminants.

80

Discussion

4.3

The 16S rDNA sequences of all type strains

possible.

This

necessary to obtain

was

optimal alignment (Ludwig genus

likelihood

method) based

could be

P.

the the

on

separate into

thoenii form

forms

a

and

(1997)

cluster

a

completed

many nucleotides

as

Schleifer, 1994).

A

alignment from E.

data

coli

containing P.

possible tree

species

P.

(1989).

avidum and P.

separate cluster between the classical and

lymphophilum

Propionibacteria

form

of the known

renamed

as

proposed by collection

a

a

and Stackebrandt

very

homogeneous

phylogenetic trees

(1989) and group, with

Charfreitag P.

et al.

dairy products.

(1988) proposed

propionicum. Recently

Kusano et al.

(1997).

strains like "P.

(ATCC)

so

far not been

the first two

were

In the

not

the

a

species

granulosum

rRNA

no

other

species

or

(1991)

the

intermixed.

Corynebacterium

new

species

that Arachnia

species

new

investigated habitats

was

included into the

propionica should be

P.

cyclohexanicum

fact that

impossible

that

occurrence

of

was

new

clearly

"P. wentii" and "P. animalis"

classified within the genus.

Propionibacteria,

as

belong

confirmed in my

Using

"P. animalis"

study by

methods. The MPCR method

the

analysis

listed,

the MPCR

was

not tested.

already existing

of the ITS

region.

et

It is

will be discovered in the future when

in other habitats is

developed in this

are

jensenii group (de Carvalho

to the P.


Propionibacteria

was

catalogue of strains of the American type culture

intermedium",

identified

A

complete

a

"P. rubrum" has been shown to

al., 1995),

P.

Britz and Riedel

Most other strains like "P. rubrum" have been shown to be variants of the

species,

and P.

fact, the 16S

fact that may be due to the well known and

the skin and

species:

these strains have

approach,

P.

al., 1997) the taxonomic situation of the Propionibacteria has changed little

the last 30 years,

genus when

species

similarity (91.1%) with Luteococcusjaponicus

more

Unlike other genera of the Actinobacteria like Arthrobacter

over

(maximum

Propionibacteria (90.7% with P. freudenreichii).

by Charfreitag

et

an

groups whereas P.

cutaneous

sequence of P.

(Kollöffel

as

of the

acidipropionici,

propionicum.

is the most distant relative of the whole genus. In

As shown

(figure 2)

The cutaneous

lymphophilum

than with other

far

compute

28 to 1521. The

in accordance with the

are

acnes, P.

shows

to

calculated

was

positions

and Stackebrandt

Charfreitag

as

as

freudenreichii (both subspecies)

and P.

one

resequenced

or

phylogenetic

relevant strains

medically

cyclohexanicum another cluster. These of Kusano et al.

were

into different clusters: The classical

grouped

jensenii and

and

including

complete

Propionibacterium

of the genus

Phylogenetic analysis

thesis is

a

investigated using promising tool

that

molecular

can

be used

81

for

screening experiments to reveal

Discussion

the presence of Propionibacteria in environmental

or

clinical samples.

4.3.1

Sequencing strategy

sequencing strategy using PCR-products instead

The

amplification of parts of the days

organism

an

can

16S rDNA of unknown

be identified

according

organisms in a short time.

to the

known organisms of the database. This method is cumbersome

cloning

As shown in this

of either PCR

study,

almost the

products complete

or

similarity of now

corresponding to the E.

universal

primer bakl 1 w binds behind this region.

its 16S rRNA with

DNA

fragments.

coli

rDNA,

not included in the submitted sequences because both

ambiguous bases

templates

for the

showed

good agreement

polymerases

like

not

slightly

the

beginning and the end

sequencing

Taq polymerase, different

reactions.

Control-sequencing

of the data I obtained. It is also

PCR

of the 16S

primers

products

contain

different. A

observed, but

of the rRNA operon

possible to

preference

for

one or

(de

were

used

as

of already known sequences

Pfu (Stratagene) to minimize amplification errors.

Propionibacteria posses two copies may be

5), forming

primers

in their sequences.

To avoid mistakes due to the

a

since the

The sequences of the universal

(bakl lw

were

organisms

positions 1-8,

used for PCR

and bak4, table

Within two

16S rRNA gene of the selected

sequenced, except the

the

often used than the

more

"gel-excised"

could be

bases

fragments allows

of cloned

use

"proofreading"

As mentioned

Carvalho et al.,

the other operon

before,

1994), which

by the polymerase was

cannot be excluded. The 16S rDNA sequence of P.

jensenii, already

published in the database, showed differences when resequenced for the region where the specific primer gdl In my

study, the

binds

same

freudenreichii subsp.

(table 13).

strategy

was

applied

shermanii. A PCR

to

obtain

a

23 S rDNA sequence of P.

product using primers gdl

from the start of the 16S rDNA to the middle of the 5S rDNA 5 and table

6). Sequencing primers

rRNA. The obtained sequence shows

were

a

was

used

selected from conserved

high similarity with the

and as

gd5sr ranging

template (table

regions

of the 23S

23 S rRNA sequence of P.

freudenreichii subsp. freudenreichii (W. Ludwig, personal communication).

Discussion

82

4.4 Genetic In several

of

system

studies, the development of a system allowing genetic analysis and engineering

Propionibacteria

attempted.

was

The

genetic engineering

detection of vector systems (e.g. phages

"foreign"

DNA into

was

have been found in

DNA transport vehicles

and Glatz,

Selectable markers,

Propionibacteria.

last

study concerning

medically

performed by Reddy

might

et al.

were

and

by

3 kb. For

sequence

susceptibilities

strains either isolated

analysis:

a

al., 1988;

et

published.

far not been described for

so

Propionibacteria,

but data

al., 1998). The

et

dairy Propionibacteria

of

during this study

was

and

a

or

received from

plasmids (table 14). Finally,

plasmid

small

isolated from

of 2.051

large plasmid

restriction endonuclease

fragment,

an

P.freudenreichii

two

kb, pLME108, that

of 40

kb, pLMElOl,

amino acid sequence was

JS53

(Smutny, 1997),

mapping and partial sequencing

found. In

identity

of

was

roughly

of 53.2% with

a

curing experiments with the

hypo¬

host of

plasmid free strain was obtained, but no differences concerning carbohydrate

fermentations and antibiotic resistance patterns

were

found between the two strains.

Plasmid pLME108, isolated from P. freudenreichii DF2,

study

the main

pLME108

thetical protein from Synechocystis sp.

this

been

as

DNA


characterized

a

previously

screened for the presence of

large plasmid pLMElOl,

pLMElOl,

al., 1995a) and their

species (Lorian, 1996; Tunney


intended to be used in

needed.

al., 1995b). Plasmids

resistance genes, have

relevant


be used for

one

et

a

(1973).

Propionibacteria

were

sequences have

the antibiotic

pLMElOl

culture collections

plasmids

no

are

Antibiotic resistances have been found for

for the

All strains of

but

et

the

of

recognition

Propionibacteria (Pérez-Chaia

found in

mainly antibiotic

mainly

4.4.1 Plasmids

were

1990)

exist

The

dairy Propionibacteria (Gautier experiments (Gautier

requires

that could be used to transport

functional in the host cells

are

used in electrotransfection

Rehberger

plasmids)

host cell. In addition, selectable markers for the

a

successful DNA transfer that

Phages

or

of bacteria

and

a

putative replicase

relatively high identity (42%) Arcanobacterium pyogenes.

mechanism for its

to

gene

the

pAPl is

was

rep a

replication (Billington

completely sequenced

in

identified {rep). This rep gene shows

gene

plasmid et

was

of that

plasmid pAPl, isolated uses

the

rolling

circle

al., 1998). Briefly, this mechanism

from

(RC) uses a

83

Discussion

single stranded intermediate form of the plasmid for its replication and is characterized by a

two

step "copy" mechanism for the plasmid: first

replicated, followed by DNA-synthesis

at the

1997). Plasmids belonging

to the

family

of

supposedly pLME108,

characterized

by

proteins. the

and

plasmids:

loops

found

was

et

the

on

pLME108

contained the

pLME108

in its

sso

sso

loop region.

site

rep gene. This

with the

were

The dso

clearly the

sso

in

G

identified

as

(Billington

et

it has also been observed for other

Plasmid pAPl has been shown to be able to

1998): by adding

a

kanamycin

introduced in E. coli by In

first attempt, the

a

family

al., 1994).

et

none

dso

region

plasmids (Yang

and at

and

correctly usually

one

of the

pLME108,

For

of these structures

the

for

in

sso

"nick-site" for the occurrence

pLME108

and McFadden,

of two

was

not

1993).

in Escherichia coli

(Billington et al.,

pAPl, the plasmid

was

replicate

resistance gene into

by

of

of RC

origin (sso)

region included

found, but

typical

a

plasmids

is in addition

region usually contains the

al., 1998). Such

Rep

relatedness of

figure 14)

region

break of the DNA double strand. This nick-site is characterized

following

in their

with other

characterize this

(shown

is

pAPl and

The sequence characteristic for the

(Fernandez-Gonzalez

structures around the

occurring

stranded replication

sequence

al., 1993) of the putative

that form this structures

secondary

single

like

plasmids,

RC

plasmid

al., 1998; Khan,

Solar et

motifs

common

special regions

by high secondary structures,

characterized

5

replication (dso).

for double stranded

(Zaman

pIJ101/pJVl

of these motifs for

conserved DNA sequence at the

(TAGCGT) outside

alignment

In addition two

pLME108.

a

origin

the

an

strand of the

lagging strand (del

is shown. The conserved amino acids show

family

same

pAPl

figure 16,

In

are

leading

a

successfully

electroporation. A fact that is promising for the case of pLMEl 08. resistance gene of

ampicillin

pUC18

was

amplified by

means

of

PCR, ligated into the Nhel site of pLME108 and electroporated into E. coli XLl-Blue. No transformants

were

found in this

GCTAGC. which is also part this

plasmid.

When the

pUC18 only the

Nhel

Both Xhol and Nhel rep

region.

cloned is

plasmid.

possibly Even

(underlined)

of the

because Nhel cuts at the sequence

sequence needed for

sso

complete pLME108, digested with Nhel

digest

were

The fact that

experiment, possibly

was

successfully

selected because

cloned into

they

cut

a

competition

pLMEl 16 did

not

ligated

to

was

once

and not in the

could not be added to

between two functional rep genes

replicate

pLME108 had perhaps maintained part

Xhol

of

pUC18 (plasmid pLME116).

pLME108 only

pLME108 digested with^TjoI

due to

or

replication

well in E. coli,

of its

activity.

pUC18 and

on

indicating that the

the

same

rep gene of

Discussion

84

4.4.2 Electrotransformation of Propionibacteria

Electroporation

of

plasmid pE194 P.

from

freudenreichii.

pE194

DNA

Luchansky

was

et al.

Selection with

found in these presumptive an

as

including Propionibacteria. They

cells, PEB, "was used in were

not

reproduce; a

erythromycin transfer of

or

plasmid pC194

from

no

were

for several

Gram-positive

plasmid

transferred to

was

transformed at

erythromycin

rate of 3.2

a

x

101

poorly documented and the protocol described

as an

example, the buffer used to obtain competent

presence of the

chloramphenicol

protoplasts of

to

which contains

concentration of 0.5 to 2.5x". In

analyzed for the

the

recipients.

selectable markers. This

are

(1987),

used to obtain transformants, but

plasmid pGK12 isolated from lactococci,

is difficult to

study

was

and Glatz

transferred

was

electroporation protocol

transformants per |ig DNA. The results in that

aureus

erythromycin

chloramphenicol resistances

several genera,

reported first by Pai

has been

Staphylococcus

(1988) proposed

bacteria using the and

Propionibacteria

plasmid

or

resistance. Zirnstein and

Staphylococcus

aureus

addition, the resulting cells

of the gene

responsible

for

Rehberger (1991) reported the to

Propionibacteria using

also

erythromycin resistance as marker gene. No plasmids were found in the resistant cells, but the gene

was

detected in the genome of these strains

findings of Pai as

and Glatz

poster-presentations

In this

study,

several

"production"

tested. In

Only

no

when

with

no

further

plasmids

following publication.

from different hosts

of competent cells

the

as

well

were

of Gautier et al.

as

used in

(1995b).

electroporation

Other methods for

different electroporation conditions

experiment (table 22) plasmid transfer into the 23 tested strains

erythromycin was

Propionibacteria was observed nor

erythromycin

used for the selection of transformants after transformation, but in all

resistance gene

was

when incubated for

a

longer

cases

grew

on

was

growth

neither

found in these strains. In

electroporated without DNA as negative controls,

erythromycin

sequences. Since in most other studies

Since the

use

erythromycin

is used

as a

detected.

plasmid DNA

addition,

the selective media

time. As discussed

were

of resistant

cells

containing

later, in chapter 4.5,

Propionibacteria can acquire erythromycin resistance by modification of their

phenomenon

Both the

(1987) and Zirnstein and Rehberger (1991) were published only

experiments, mainly according to the strategy the

by hybridization analysis.

23 S rRNA

selectable marker, this

may have occurred before.

of

"foreign" plasmids

may have been another

reason

electroporation experiments, plasmids pLME108 and pLMElOl

for the failure of the

were

used

as

potential

85

vectors. In both

cases

plasmids

fused with the

were

either resistance genes

the

observed. Since the tetM gene

was

transfer

introduced at the Nhel site of pLME108 the

Briefly

following

used

or a

problem described in chapter 4.4.1

facts may have inhibited the

electroporation

investigated Propionibacteria,

Lack of competence of the

•

was

natural absence of this

•

No suitable

•

No suitable selectable marker

•

A restriction system in the

ability

of Propionibacteria:

either due to the methods

in the strains.

plasmid was tested. was

used.

recipient

"destroys" foreign DNA.

strains

Since the traditional methods have been shown in my thesis not to work,

transformation

4.4.3

may have

for this result.

reason

the

(beta-lactamase: ampicillin or tetM) or other

Propionibacteria plasmids and electroporated into

putative competent Propionibacteria. No

been the

Discussion

strategies

and

protocols

must be

completely new

developed.

Conjugation experiments

Plasmid DNA transfer been observed also for

Especially

enterococci

1997) and

were

K214,

was

selection

by

as

are

important

between

as a

species

reservoir for such

donor strains in my

as

donor strain.

of transformed

were

study.

In

Using enterococci,

Propionibacteria

Enterococcus strains. Several

tested, but enterococci

conjugation

of different genera has

species of dairy origin (Wirsching, 1995;

selected

used

of

means

with

a

plasmids (Perreten

addition,

the main total

1998).

Perreten et al.,

resistant

as

as were

the

al.,

Lactococcus lactis

problem

inhibition

became the

of the

donor

media, media supplements and growth conditions

at least

et

were

Propionibacteria (table 19).

Only when Propionibacteria were adapted to high concentrations of rifampicin and fusidic acid, In

a

sufficient

none

of the

erythromycin controls and

selectivity

experiments,

was

no

achieved.

was

a

used, resistant

transferred DNA

DNA transfer mutants of was

by conjugation

Since

tetracycline resistance,

found in the

Propionibacteria

between donor and

plasmids

from

can

produce

no

in the

putative transconjugants. which

can

negative

Even with

integrate chromosomally

DNA transfer was observed.

extracellular slime

(Reddy

recipient cells could be inhibited by this

Propionibacteria with selectable markers to

for the failure of the

observed. When

Propionibacteria appeared

plasmids like pFOl, harboring the transposon TnFOl and commits

was


et

al., 1973) the

contact

slime. The lack of indigenous

be tested may also be

a reason

86

Discussion

4.5 Selectable and transferable markers Both

electroporation


and

be used for the selection of Propionibacteria. In into

functional in these

Propionibacteria and is

4.5.1 Selectable markers for A selectable marker used in

of transformants. Such

strains

acid. All donor strains

Naturally occurring

a

organisms

marker that

can

be transferred

has to be used.

conjugation experiments must allow the simple identification

strain, for example the ability

Propionibacteria

addition,

cam


usually encoded

marker is

a

revealed the need for markers that

were

were

to

lactose

use

adapted

inhibited

resistance of

to

by

as

high

on

the chromosome of the

carbon

In this

source.

concentrations of

recipient

thesis, recipient

rifampicin

and fusidic

these antibiotics.

Propionibacteria (against gentamicin, kanamycin,

metronidazole, methicillin, nalidixic acid, neomycin, sulfonamides and tobramycin) could be used for the selection of recipient cells for genera. These antibiotics could also be used in

a


with other

selective medium for the isolation of

Propionibacteria.

Conjugation experiments resistance markers for

within the

recipient

genus

cells that

resistance. The best solution would be the

strain

specific,

use

of

for

example streptomycin

strains of adapted to

high concentrations of

are

use

Propionibacterium require the

antibiotics.

4.5.2 Transferable markers for To

detect

a

successful Ideal

Propionibacteria.

for

Propionibacteria, antibiotic

resistance

Propionibacteria:

gene

conjugation transfer be

would

example

a

the

ciprofloxacin, tetracycline.

markers

therefore

a

used

marker

bacteriocin resistance

be

must

which

gene.

The

for

acnes were

varying

fully

resistance patterns,

functional. The

chloramphenicol,

found are

from

development

medically acne

in

of

relevant

vulgaris,

(Eady, 1998).

antibiotics for which

indicating

Propionibacterium

that resistance mechanisms

following antibiotics are erythromycin,

functional

originates

after intake of antibiotics used for the treatment of

Good candidates for transferable markers

available and

electroporation

patterns has been observed mainly

antibiotic resistant strains of P.

strains show

and

such candidates:

polymyxin

B,

are

cefotaxime,

streptomycin

and

87

The uselessness of erythromycin rate

was

marker because of a to

shown in this thesis. Ross et al.

rDNA led to resistance to occurs

as a

for

Discussion

high spontaneous

(1997) showed that

a

single

mutation in 23S

erythromycin in cutaneous Propionibacteria. The

tetracycline, where

mutation

same

mutations in the 16S rDNA caused resistance in cutaneous

Propionibacteria (Ross et al., 1998). The mutation rate for tetracycline resistance be lower than for the

experiments

erythromycin

since

growth

no

of the

negative

controls

seems

to have

a

good potential

as

Resistance genes for

without

due to

E. coli shuttle vector

in

success

an

chloramphenicol

are

used in vector systems, for -

to

observed in

was

resistance marker: resistance is

observed for P. thoenii and P. jensenii strains whereas the other

perfringens

seems

of this thesis.

Chloramphenicol

commonly

problem

described for

example

species

a

in this

thesis, but this is

not

unsuitable resistance gene, other factors could have inhibited

and

Clostridium

(Bannam and Rood, 1993). This plasmid


sensitive.

Gram-positive organisms

plasmid pJIR750,

the

are

a

was

used

necessarily successful

transfer.

Propionibacteria

are

sensitive to

ampicillin

observed in

A resistance is

mainly

excreted in the

periplasmatic

and other antibiotics of the

Gram-negative organisms,

gap which exists

only

in these

since the

transformants

attempted.

were

found in this thesis when

an

group.

ß-lactamase

is

organisms (Lorian, 1996).

This classical mechanism does not function in Propionibacteria and no

penicillin

ampicillin

be the

reason

why

resistance transfer

was

can

88

Discussion

4.6 Conclusions

Using

the

multiplex

PCR method

developed

study

in this

simple,

a

reliable and fast

detection of strains belonging to the genus Propionibacterium is possible. The method can also be used for

rapid screening

Species specific amplified example

gene

probes

16S-23S rDNA

with

for

and

such

hybridization spacer

provide

freudenreichii,

Propionibacterium

a

intergenic

digoxigenin

be tested for P.

of various habitats for the presence of Propionibacteria. assay

can

regions. This

stable

for all 10

probes.

be obtained

ITS

regions

by using

can

probes should be possible.

be labeled for

Even if the system has

species currently classified The

probes

PCR

so

far

only

in the genus

could also be used for the

quantitative detection of Propionibacterium species by hybridization. and

Both for

conjugation

plasmids

must be tested.

preferably originating Prior to suitable

electroporation experiments,

Especially

from the classical

conditions, strains and

suitable selectable marker has to be found,

Propionibacteria

or

other

dairy

relevant genera.

conjugation experiments, electroporation conditions have to be established plasmid

must be selected and introduced into

could then further be used in

pLME108

should be further

marker

into

plasmid

Preferentially

this

Propionibacteria.

This

and

a

plasmid


Plasmid

E. coli.

a

more

is

investigated

necessary

for

and the introduction of

further

this selectable marker should "work" in

a

selectable

electroporation experiments. Propionibacteria

as

well

as

in

References

89

5. References Adegoke GO. Nwaigwe R.N., Oguntimein cal

changes during the production

GB. 1995.

of sekete

-

a

Microbiological

fermented

and biochemi¬

made from maize.

beverage

J. Food Sei. Technol. 32: 516-518.

Al-Zoreky N., Ayres J.W.,

Sandine W.E. 1991. Antimicrobial

against food spoilage and pathogenic microorganisms. Amann

J.

activity

Dairy

of

Microgardto

Sei. 74: 758-763.

R.I., Ludwig W., Schleifer K.-H. 1995. Phylogenetic identification and

in situ

detection of individual microbial cells without cultivation. Microbiol. Rev. 59: 143169. Anderson

D.G, McKay

DNA from lactic

Simple and rapid method

L.L. 1983.

streptococci. Appl.

Environ. Microbiol. 46: 549-552.

Babuchowski A., Laniewska-Moroz L.,

ria in fermented milk and

pionibacteria. Cork.

isolating large plasmid

for

vegetables.

Warminska-Radyko Lecture.

2nd

1.1998.

international

symposium

proteins.

Baer A.,

Pro¬

electrophoresis

of

Milchwissenschaft 42: 431-433.

Ryba 1.1991. Identification of Propionibacteria and streptococci by immuno-

blotting. Milchwissenschaft 46: Baer A.,

on

Ireland.

Baer A. 1987. Identification and differentiation of Propionibacteria by

their

Propionibacte¬

292-294.

Ryba 1.1992. Serological identification of Propionibacteria in milk and cheese

samples. Bamarouf

Int.

Dairy

J. 2: 299-310.

A., Eley A., Winstanley T. 1996. Evaluation of methods for distinguishing

Gram-positive

from

Gram-negative anaerobic

Bannam T.L., Rood J.I. 1993.

bacteria. Anaerobe 2: 163-168.

Sequenced Clostridium perfringens-Escherichia

coli

shuttle vectors that carry single antibiotic resistance determinants. Plasmid 229: 233235. Bertram J., Duerre P. 1989. posons in Clostridium

Beveridge

Conjugal

transfer and

expression

acetobutylicum. Arch. Microbiol. 151:

T.J. 1990. Mechanism of Gram

variability

of

streptococcal

trans¬

551-557.

in select bacteria. J. Bacteriol.

172: 1609-1620.

Billington S.J., Jost B.H., Songer genes

tion

plasmid pAPl

plasmids.

is

a

J.G 1998. The Arcanobacterium

member of the

pIJ101/pJVl family

J. Bacteriol. 180: 3233-3236.

(Actinomyces) pyo¬

of rolling circle

replica¬

REFERENCES

90

Boesch C. 1998. Nucleotide sequences

tions of Acetobacter

as a

species isolated

basis to genetic and taxonomic

from industrial fermentations. Thesis

investiga¬ 12647.

no

ETH Zurich. Boivin-Jahns V.,

diversity

in

a

Ruimy R.,

Bianchi A., Daumas

deep-subsurface clay

environment.

S., Christen

Appl.

R. 1996. Bacterial

Environ. Microbiol. 62: 3405-

3412. Britz

T.J., Riedel K.-H.J. 1991. A numerical taxonomic study of Propionibacterium

strains from

dairy

sources.

Brosius J., Palmer M.L.,

J.

Appl.

Bacterid. 71: 407-416.

Kennedy P.J.,

Noller H.F. 1978.

Complete

nucleotide

sequence of a 16S ribosomal RNA gene from Escherichia coli. Proc. Natl. Acad. Sei. USA 75: 4801-4805.

Charfreitag O., propionica

as

Collins M.D., Stackebrandt E. 1988. Reclassification of Arachnia

Propionibacterium propionicus comb.

nov.

Int. J.

Syst. Bacteriol. 38:

354-357.

Charfreitag O., genus

Stackebrandt E. 1989. Inter- and

Propionibacterium

as

determined

by

intrageneric relationships

of the

16S rRNA sequences. J. Gen. Microbiol.

135: 2065-2070.

Clewell D.B.,

Yagi Y., Dunny G,

deoxyribonucleic of a

Schulz S. 1974. Characterization of three

acid molecules in

a

strain of Streptococcus faecalis: identification

plasmid determining erythromycin resistance.

Cummins C.S., Johnson J.L. 1986. Genus I. 1346-1353. In:

logy. Williams Cummins

Krieg N.R.,

Holt J.G.

J. Bacteriol. 117: 283-289.

Propionibacterium

(eds.). Bergey's

J.L. 1992. The genus

Dasen

manual of

systematic

Propionibacterium.

Trüper H.G., Dworkin M., Harder W., Schleifer New

Orla-Jensen 1909. pp. bacterio¬

& Wilkins. Baltimore. USA.

C.S., Johnson

Springer.

plasmid

K.-H.

In: Balows

A.,

(eds.). The Prokaryotes.

York, USA. pp. 834-849.

G, Smutny J.,

Propionibacteria

Teuber M., Meile L. 1998. Classification and identification of

based

on

ribosomal RNA genes and PCR. System. Appl. Microbiol.

21:251-259. de Carvalho rium

A.E, Gautier M., Grimont

species by

de Carvalho

F. 1994. Identification of dairy

Propionibacte¬

rRNA gene restriction patterns. Res. Microbiol. 145: 667-676.

A.F., Guezenec S., Gautier M., Grimont P.A.D. 1995. Reclassification of

"Propionibacterium rubrum"

as

P. jensenii. Res. Microbiol. 146: 51-58.

91

Del Solar

References

G, Giraldo R., Ruiz-Echevarria M.J., Espinosa M., Diaz-Orejas

Replication and

control of circular bacterial

plasmids.

R. 1998.

Microbiol. Mol. Biol. Rev. 62:

434-464.

Deng M.Y., like

Fratamico P.M. 1996. A

multiplex

toxin-producing Escherichia coli

PCR for

rapid identification

of

shiga-

0157:H7 isolated from foods. J. Food Prot. 59:

570-576.

Denys G.A., Jerris R.C., Swenson J.M., Thornsberry

pionibacterium

acnes

Susceptibility

of Pro¬

clinical isolates to 22 antimicrobial agents. Antimicrob.

Agents

C. 1983.

Chemother. 23: 335-337. Gunter S.E. 1946. The taxonomic

Douglas H.C.,

position

of

Corynebacterium

acnes.

J. Bacteriol. 52: 15-23.

Drinan

F.D., Cogan T.M. 1992. Detection of propionic acid bacteria in cheese. J. Dairy

Res. 59: 65-69.

Y., Rossignol

Dubreuil L., Houcke I., Mouton

activities of nitazoxanide and tizoxanide Antimicrob.

Eady

Agents Chemother.

to those of

seven

acne.

Appelbaum

other agents

against

anaerobes and aerobic

organisms.

40: 2266-2270.

E.A. 1998. Bacterial resistance in

Ednie L.M., Jacobs M.R.,

J.F. 1996. In vitro evaluation of

Dermatology

196: 59-66.

P.C. 1998. Activities of gatifloxacin

against

anaerobic

organisms.

compared

Antimicrob.

Agents

Chemother. 42: 2459-2462.

Engelke G,

Gutowski Eckel

K.D. 1994.

Appl.

Regulation

Z., Kiesau P., Siegers K., Hammelmann M., Entian

of nisin

biosynthesis

and

immunity

in Lactococcus lactis 6F3.


Essers L. 1982.

antibiotic

Simple

identification of anaerobic bacteria to genus level

susceptibility patterns.

Feinberg A.P., Vogelstein endonuclease

J.

Appl.

B. 1983. A

using typical

Bacteriol. 52: 319-323.

technique

fragments to high specific activity.

Felsenstein J. 1982. Numerical methods for

for

radiolabelling

DNA restriction

Anal. Biochem. 132: 6-13.

inferring evolutionary

trees. The

Quarterly

R.F., Noirot-Gros M.F., Martin J.F.,

Gil J.A.

Review of Biology 57: 379-404. Fernandez-Gonzalez C, Cadenas 1994. Characterization of tum

involved in

a

region

of plasmid

pBLl

of Brevibacterium

replication via the rolling circle model.

lactofermen-

J. Bacteriol. 176: 3154-3161.

REFERENCES

92

Fessier D. 1997. Characterization of Propionibacteria in Swiss

and

molecular-biological

Fessier D., rium

Casey M.G,

methods. Thesis

Puhan Z. 1998.

no

raw

milk

by

biochemical

12328. ETH Zürich.

Rapid identification

species by restriction analysis of the insertion within

of

dairy Propionibacte¬

the 23 S rRNA gene. Lait

78: 203-216. Frère J. 1994.

Lett.

Simple

method for

Appl. Microbiol.

(Acnein)

DNA from lactic acid bacteria.

18: 227-229. T. 1978. Purification and

properties

Propionibacterium

Antimicrob.

Fujimura S., Nakamura stance

extracting plasmid

of oral

acnes.

of

a

bacteriocin-like sub¬

Agents Chemother.

14:

893-898. Funke

G,

von

Clarridge

Graevenitz A.,

biology of coryneform bacteria. Gautier M., de Carvalho

bacteria strains

Clin. Microbiol. Rev. 10: 125-159.

A., Rouault A. 1996. DNA fingerprinting of dairy Propioni¬

by pulsed-field gel electrophoresis.

Gautier M., Mouchel N., Rouault of four

III J.E., Bernard K.A. 1997. Clinical micro¬

A., Sanséau

Curr. Microbiol. 32: 17-24.

P. 1992. Determination of genome size

Propionibacterium species by pulsed-field gel electrophoresis.

Lait 72: 421-

426. Gautier M., Rouault A. 1990.

Screening

of

plasmids

Brief Communications of the XXIII International 345. Brussels.

Belgium;

International

Gautier M., Rouault A., Sommer terium freuJenrcichii

Propionibacterium.

Dairy Congress, Montreal.

Vol. IL:

Dairy Federation.

P., Briandet

bacteriophages

in strains of

R. 1995a. Occurrence of Propionibac¬

in Swiss cheese.

Appl. Environ. Microbiol.

61:

2572-2576. Gautier M., Rouault A., Lemée R. 1995b. Electrotransfection of

freudenreichii TL 110.

Lett.

Appl.

Microbiol. 20: 125-128.

Gawron-Burke C, Clewell D.B. 1984. tococcal DNA

Strategy

for

Regeneration of insertionally

fragments after excision of transposon

targeting

and

cloning

Propionibacterium

of genes from

inactivated strep¬

TnPitf in Escherichia coli:

Gram-positive

bacteria. J. Bacteriol.

159:214-221. Gerhardt P.,

Phillips

Murray R.GE.,

Costilow

R.N., Nester E.W., Wood W.A., Krieg N.R.,

GB. 1981. Manual of methods for

Washington.

general bacteriology.

ASM Press.

93

Giraffa

G, Carminati D., Neviani E. 1997. Enterococci isolated from dairy products:

review of risks and

potential technological

Glatz B.A. 1992. The classical

trial

References

organisms.

ASM

news

J. Food Prot. 60: 732-738.

use.

Propionibacteria:

a

Their past, present and future

indus¬

as

58:197-201.

Glatz B.A., Anderson K.1.1988. Isolation and characterization of mutants of dairy Pro¬

pionibacterium Goldenberger D. range PCR no

strains. J.

Dairy

Sei. 71: 1769-1776.

1997. Detection of bacterial

amplification and

direct

pathogens

sequencing

in clinical

specimens by

broad-

of part of the 16S rRNA gene. Thesis

12219. ETH Zürich.

Goldenberger D.,

Perschil

DNA extraction

amplification.

I.,

Ritzier

procedure using

PCR Methods and

M., Altwegg M. 1995.

SDS and Proteinase K is

Applications

bacterium

mass

acnes

Sub-species discrimination,

spectrometry and self-organising neural networks, of Propioni¬

isolated from normal human skin. Zbl. Bakt. 284: 501-515.

Rapid identification using pyrolysis of Propionibacterium

acnes

mass

isolated from

16S rRNA gene of most

species

of

dogs.

J.

Appl.

1994.

Bacteriol. 76: 124-34.

primers

pathogenic bacteria, including

and

probes

for the

bacteria found in

fluid. J. Clin. Microbiol 32: 335-351.

Grinstead D.A., Barefoot S.F. 1992. Jenseniin G,

Propionibacterium jensenii

P126.

Appl.

Guldimann D. 1998. Gentransfer bei

Lebensmittelmikrobiologie,

a

heat-stable bacteriocin

produced by


Propionibakterien. Diplomarbeit

im Labor für

ETH Zürich.

Hoeffler U., Ko K.L., Pulverer G 1976. Antimicrobial acnes

Harvey R.G

spectrometry and artificial neural networks

Greisen K., Loeffelholz M., Purohit A., Leon D. 1994. PCR

rium

"universal"

compatible with direct PCR

Goodacre R., Neal M.J., Kell D.B., Greenham L.W., Noble W.C.,

cerebrospinal

simple

4: 368-370.

Goodacre R., Howell S.A., Noble W.C., Neal M.J. 1996.

using pyrolysis

A

and related microbial

susceptibility

of Propionibacte¬

species. Antimicrob. Agents Chemother. 10: 387-

394. Hoellman D.B.,

of cefminox

Spangler S.K., Jacobs M.R., Appelbaum P.C. against

anaerobic bacteria

pounds. Antimicrob. Agents Chemother. Hofherr L.A., Glatz

bacterium to

compared

1998. In vitro activities

with those of nine other

com¬

42: 495-501.

B.A., Hammond E.G 1983. Mutagenesis of strains of Propioni¬

produce cold-sensitive mutants.

J.

Dairy

Sei. 66: 2482-2487.

94

REFERENCES

Holmes

D.S., Quigley

M. 1981. A

Plasmids. Anal. Biochem. Holt J.G

rapid boiling method

for the

preparation

of bacterial

114: 193-197.

(ed.) 1984. The shorter Bergey's manual of determinative bacteriology.

Williams & Wilkins. Baltimore. USA. pp. 233-238. Hsieh H.Y., Glatz B.A. 1996.

Long

term

storage stability of the bacteriocin Propionicin

produced by Propionibacterium thoenii and potential

PLG-1

as a

food preservative.

J. Food Protection 59: 481-486.

Hsieh H.Y., Paik H.D., Glatz B.A. 1996.

Propionicin PLG-1,

bacteriocin

a

Improvement of detection and production of

produced by Propionibacterium

thoenii. J. Food

Protection 59: 734-738.

Hykin P.G,

Tobal

1994. The

K., Mclntyre G, Matheson M.M., Towler H.M.A., Lightman S.L.

diagnosis

of

delayed post-operative endophthalmitis by polymerase

reaction of bacterial DNA in vitreous

samples.

Jakab E., Zbinden R., Gubler J., Ruef Severe infections caused

late

tion of

Appl. Jones

Strompl C,

J. Med. Microbiol. 40: 408-415.

von

by Propionibacterium

postoperative infections.

Jarvis GN.,

C,

chain

Graevenitz A., Krause M. 1996.

acnes: an

underestimated pathogen in

Yale J. Biol. Med. 69: 477-482.

Moore E.R.B., Thiele J.H. 1998. Isolation and characteriza¬

obligately anaerobic, lipolytic

bacteria from the

rumen

of red deer.

System.

Microbiol. 21:135-143.

C.J., Edwards K.J., Castaglione S., Winfield M.O., Sala F.,

Bredemeijer G,

Vosman

van

de Wiel

C,

B., Matthes M., Daly A., Brettschneider R., Bettini P.,

Buiatti M., Maestri E., Malcevschi A., Marmiroli N., Aert R., Volckaert G, Rueda J., Linacero R.,

Vazquez A., Karp

RAPD, AFLP and SSR markers in plants by

Breeding

a

A. 1997.

Reproducibility testing

of

network of European laboratories. Mol.

3: 381-390.

Kaufmann P., Pfefferkorn A., Teuber M., Meile L. 1997. Identification and cation of Bifidobacterium

species

isolated form food with

targeted probes by colony hybridization

and PCR.

genus-specific

Appl.

quantifi¬

16S rRNA-

Environ. Microbiol. 63:

1268-1273. Kendall

idans

K.J., Cohen S.N. 1988. Complete nucleotide

plasmid pIJlOl

sequence of the

and correlation of the sequence with

teriol. 170:4634-4651.

Streptomyces

genetic properties.

liv-

J. Bac-

95

Khan S.A. 1997.

Rolling-circle replication

References

of bacterial

plasmids.

Microbiol. Mol. Biol.

Rev. 61: 442-455.

Kollöffel B., Burri S., Meile L., Teuber M. 1997.

cleotide

probes

Development

dairy

starter cultures.

System. Appl. Microbiol. 20: 409-417.

Kusano K., Yamada H., Niwa M., Yamasato K. 1997. a new

sp. nov.,

rium isolated from

Laforest

oligonu¬

for Brevibacterium, Micrococcus/Arthrobacter and Microbacterium/

Aureobacterium used in

cum

of 16S rRNA

acid-tolerant

spoiled

Propionibacterium cyclohexani-

©-cyclohexyl fatty acid-containing Propionibacte¬

orange juice. Int. J.

Syst. Bacteriol.

47: 825-831.

H., Fourgeaud M., Richet H., Lagrange P.H. 1988. Comparative

activities of

pristinamycin.

Its component and other antimicrobial agents

anaerobic bacteria. Antimicrob. Lane D.J. 1991.16S/23S rRNA

techniques

Nucleic acid

in vitro

Agents

Chemother. 32: 1094-1096.

sequencing.

in bacterial

against

In: Stackebrandt E., Goodfellow M.

systematics.

John

Wiley

(eds.).

& Sons. New

York,

USA. pp. 115-176.

Leblond-Bourget N., Philippe H., Mangin I., 23 S internal transcribed spacer sequence

dobacterium

phylogeny.

Int. J.

analyses

reveal inter- and

intraspecific Bifi¬

Syst. Bacteriol. 46:102-111.

Leenhouts K.J., Kok J., Venema G 1989.

plasmid

Decaris B. 1996. 16S rRNA and 16S to

Campbell-like integration

DNA into the chromosome of Lactococcus lactis

subsp.

of

lactis.

heterologous

Appl.

environ.

Microbiol. 55: 394-400.

CG, Inouye M. 1990. Low copy number plasmids for regulated low-level

Lerner

expression bility.

of cloned genes in Escherichia coli with blue/white insert

screening

capa¬

Nucleic Acids Res. 18: 4631.

Levy S.B., McMurry L.M., Burdett V., Courvalin P., Hillen W., Roberts M.C., Taylor

D.E.

1989.

Nomenclature

for

tetracycline

resistance

determinants.

Antimicrob. Agents Chemother. 33: 1373-1374.

Leyden J.J., McGinley K.J., Vowels acne

and

Lorian V. more.

nonacne.

Dermatology

B. 1998.

Propionibacterium

acnes

colonization in

196: 55-58.

(ed.) 1996. Antibiotics in laboratory medicine. Williams & Wilkins. Balti¬

USA.

REFERENCES

96

Luchansky J.B., Muriana P.M., Klaenhammer T.R.

1988.

Application of electropora-

tion for transfer of plasmid DNA to Lactobacillus, Lactococcus, Leuconostoc, Liste¬

Staphylococcus,

ria, Pediococcus, Bacillus,

Enterococcus and

Propionibacterium.

Mol. Microbiol. 2: 637-646. Schleifer K.H. 1994. Bacterial

Ludwig W., sequence

analysis.

phylogeny based

on

16S and 23 S rRNA

FEMS Microbiol. Rev. 15: 155-173.

MacDougall J., Margarita D., spirochete Treponema

Saint Girons 1.1992.

denticola with the

Homology

single-stranded

DNA

of a

plasmid

plasmids.

from the

J. Bacterid.

174: 2724-2728. Malik

A.C., Reinbold GW., Vedamuthu

molekularbiologischer Verfahren für Problemlösungen

Meile L. 1998. Das Potential

Lebensmittelmikrobiologie. Habilitation,

Moore

glas Muth

W.E.C., Cato

E.P. 1963.

and Gunter comb.

nov.

Validity

pSG5: Nucleotide

analysis

replication

Nübel U.,

in group N

Engelen B.,

W. 1995.

(Gilchrist)

acnes

Dou¬

Streptomyces ghanaensis plas¬

of the self-transmissible minimal

replicon

mode. Plasmid 33: 113-126.

transfer and characterization of bacterio-

(lactic acid) streptococci.

Felske

Backhaus H. 1996.

submitted to ETH Zürich.

Propionibacterium

Conjugal

Neve H., Geis A., Teuber M. 1984.

plasmids

in

J. Bacteriol. 85: 870-874.

sequence

and characterization of the

cin

of

G, Farr M., Hartmann V., Wohlleben

mid

taxonomy of

Can. J. Microbiol. 14: 1185-1191.

Propionibacterium.

der

E.R. 1968. An evaluation of the

J. Bacteriol. 157: 833-838.

A., Snaidr J., Wieshuber A., Amann R.I., Ludwig W.,

Sequence heterogeneities of

Paenibacillus polymyxa detected

genes

encoding

16S rRNAs in

by temperature gradient gel electrophoresis.

J. Bac¬

teriol. 178: 5636-5643. Olsen GJ., Woese C.R., Overbeek R. 1994. The winds of

Breathing

new

life into

microbiology.

Orla-Jensen S. 1909. Die

Hauptlinien

(evolutionary) change:

J. Bacteriol. 176: 1-6.

des natürlichen

Bakteriensystems.

Zbl. Bakt.

Abt. II 22: 305-346. Pai

S.L., Glatz

of

B.A. 1987.

Production, regeneration, and transformation of protoplasts

Propionibacterium strains Propionibacterium freudenreichii transformation

Staphylococcus

aureus

and

Streptococcus faecalis plasmid.

J.

Dairy

Sei. 70: 80.

with

97

Paul

References

GE., Booth S.J. 1988. Properties and characteristics of

stance

produced by Propionibacterium

bacteriocin-like sub¬

a

isolated from dental

acnes

plaque.

Can. J.

Microbiol. 34: 1344-1347. Pérez-Chaia A., Sesma F., Pesce de Ruiz in strains of

Plasmids

Swiss type cheeses. Perreten V. 1995.

Holgado A.,

Propionibacterium

Microbiologie-Aliments-Nutrition 6:

lated from food. Thesis Perreten V., Schwarz

genes in

171-174.

of

and lactic acid bacteria iso¬

11400. ETH Zurich.

no

F., Cresta L., Boeglin M.,

spread in

Perreten V.,

staphylococci

of

lactobacilli isolated from

Distribution, molecular characterization and genetic mobilization

antibiotic resistance genes in enterococci,

resistance

mesophilic

and

Screening

Oliver G 1988.

Dasen

G, Teuber M. 1997. Antibiotic

food. Nature 389: 801-802. Schuler-Schmid U., Teuber M. 1998. Antibiotic resistance

Giampa N.,

coagulase-negative staphylococci isolated

from food.

System. Appl.

Micro¬

biol. 21: 113-120. Pettersson B., Bölske G, Thiaucourt F., Uhlén M.N., Johansson K.E. 1998. Molecu¬ lar evolution of

polymorphisms

Mycoplasma capricolum subsp. capripneumoniae strains,

R.H. 1957.

zapatera spoilage of olives.

terium aviJum

on

in the 16s rRNA genes. J. Bacterid. 180: 2350-2358.

Plastourgos S., Vaughn

Pottkämper M.,

based

Randerath

Kp

40

on

Appl.

Species

of

Propionibacterium associated

with

Microbiol 5: 267-271.

O., Beuth J., Pulverer G 1996. Influence of Propionibac¬

the cellular immune system after intensive sport

activity.

Zbl.

Bakt. 284:367-371. Randerath O., Herba

Pottkämper M.,

abrotani-Tee

und

Beuth J., Pulverer G 1997. Immunmodulation mit

Propionibacterium

avidum

KP-40

bei

professionellen

Eishockeyspielern, http://hepatitis-c.de/abrotanu.htm. Reddy M.S.,

Reinbold

GW., Wiliams F.D. 1973a. Inhibition of Propionibacteria by

antibiotic and antimicrobial agents. J. Milk Food Technol. 36: 564-569.

Reddy M.S.,

Washam

C.J., Reinbold GW., Vedamuthu E.R. 1973b. Immunogenic

properties of slime from Propionibacteria.

Reddy M.S.,

J. Milk Food Technol. 36: 200-201.

Wiliams F.D., Reinbold GW. 1973c. Sulfonamide resistance of Propioni¬

bacteria: nutrition and transport. Antimicrob.

Rehberger J.L., low

Ford J., Glatz B.A. 1994.

Agents Chemother.

4: 254-258.

Response of Propionibacteria

to

pH: Acid adaption and loss of pigment. Abstr. Gen. Meet. ASM: 364.

moderately

REFERENCES

98

Rehberger T.G field

1993. Genome

analysis of Propionibacterium freudenreichii by pulsed-

gel electrophoresis. Curr. Microbiol. 27: 21-25.

Rehberger T.G,

using

rapid

a

Glatz B.A. 1985. Isolation of

and

simple method.

plasmid

DNA in

Propionibacterium

Abstr. Annu. Meet. Am. Soc. Microbiol. 85 Meet.:

254.

Rehberger T.G,

Glatz B.A. 1987. Characterization of plasmid DNA in

rium freudenreichii

tion. J.

Dairy

P93: evidence for

plasmid-linked

lactose utiliza¬

Sei. 70: 80.

Rehberger T.G, Appl.

subsp. globosum

Propionibacte¬

Glatz B.A. 1990. Characterization of

Propionibacterium plasmids.

Env. Microbiol. 56: 864-871.

Riedel K.-H.J., Britz-T.J. 1996. Justification of the classical


concept by ribotyping. System. Appl. Microbiol. 19: 370-380. Riedel

K.-H.J.,

Wingfield

Propionibacterium

B.D.,

species

Britz

using

T.J.

16S

rDNA

polymorphisms. System. Appl. Microbiol. Riedel K.-H.J.,

onibacterium

Wingfield B.D.,

restriction

-

of

fragment

G+C content

species concept by

are

characterized

length

21: 419-428.

restriction

analysis

Propi¬

of the 16S ribosomal RNA

17: 536-542.

C, Ludwig W., Schleifer K.H. 1992. Gram-positive bacteria with

Roller

classical

Britz T.J. 1994. Justification of the "classical"

System. Appl. Microbiol.

genes.

Identification

1998.

by

a common

a

high

DNA

insertion within their 23S rRNA genes.

J. Gen. Microbiol. 138: 1167-1175.

Ross J.I., with

Eady E.A., Cove J.H., Cunliffe W.J. 1998.

Tetracycline resistance

in

a

16S rRNA mutation associated

Gram-positive bacterium. Antimicrob. Agents

Chemother. 72: 1702-1705. Ross J.I.,

Eady E.A., Cove J.H.,

Jones C.E.,

Raryal A.H.,

Cunliffe W.J. 1997. Clinical resistance to neous

erythromycin

Propionibacteria isolated from acne patients

rRNA. Antimicrob.

Agents Chemother.

Miller Y.W., and

clindamycin

41: 1162-1165.

dairy Propionibacteria. Biotechnol. Techniques

Rossi F., Torriani S., for

typing

national

Zapparoli G, Dellaglio

and identification of

in cuta¬

is associated with mutations in 23 S

Rossi F., Sammartino M., Torriani S. 1997. 16S-23S ribosomal spacer in

Vyakrnam S.,

11: 159-161.

F. 1998. Use of different

dairy Propionibacteria.

symposium on Propionibacteria.

polymorphism

Cork. Ireland.

Poster

genetic methods

presentation.

2

inter¬

99

References

Sambrook J., Fritsch E.F., Maniatis T. 1989. Molecular

2nd edition.

Spring

Cold

Harbor

Phylogenetic diversity

mined

by

16S rRNA gene

Servin-Gonzälez L.,

of

manual.

Ohashi A., Harada H., Nakamura K.

mesophilic

and

thermophilic granular sludges

analysis. Microbiology

Sampieri A.I.,

laboratory

a

Press. New York.

Laboratory

Sekiguchi Y., Kamagata Y, Syutsubo K., 1998.

cloning:

deter¬

144: 2655-2665.

Cabello J., Galvän L., Juarez V., Castro C. 1995.

Sequence and functional analysis of the Streptomyces phaeochromogenes plasmid

pJVl reveals rolling

a

modular

organization

Microbiology

circle.

Streptomyces plasmids

of

enteral

shermanii and the

dysbacteriosis. 1997.

Smutny J.

bakterien.

J.

possibility

biologically

of its

Hygiene Epidemiol. Microbiol.

use

in

active mutant of

complex therapy

of

Immunol. 28: 331-338.

Diplomarbeit im Labor für Lebensmittelmikrobiologie, ETH Zürich. Rainey F.A., Ward-Rainey N.L. 1997. Proposal for

classification system Actinobacteria classis Stierli M.P. 1998.

Charakterisierung

Propionibacterium-lsolat

telmikrobiologie, Sutter V.L.,

a

Anwendung von molekularen Techniken zur Typisierung von Propioni-

Stackebrandt E.,

renden

replicate by

141: 2499-2510.

Sidorchuk I.I., Bondarenko V.M. 1984. Selection of

Propionibacterium

that

nov.

Int. J.

eines Plasmides

aus

Milch.

a new

hierarchic

Syst. Bacterid. 47: 479-491.

aus

einem

Diplomarbeit

Bacteriocin-produzie-

im Labor für Lebensmit¬

ETH Zürich.

Finegold S.M. 1976. Susceptibility of anaerobic bacteria

to 23 antimicro¬

bial agents. Antimicrob. Agents Chemother. 10: 735-752.

Tunney M.M., Ramage G, Patrick S., Nixon J.R., Murphy P.G, Antimicrobial revision

hip

susceptibility

of bacteria isolated from

surgery. Antimicrob.

Agents Chemother.

Von Freudenreich E., Orla-Jensen S. 1906.

Propionisäuregärung.

orthopedic implants following

42: 3002-3005.

Über die im Emmentalerkäse stattfindende

Zbl. Bakt. Abt. 2 17: 529-546.

Von Graevenitz A., Funke G 1996. An identification scheme for

cally growing Gram-positive

bacterium 173.

Dairy

Sei. 70: 79.

Everett E.D., Johnson M., Dean E. 1977.

acnes

and aerobi-

Generation, isolation and characterization of auxotro¬

mutants of Propionibacterium. J.

Wang W.L.L.,

rapidly

rods. Zbl. Bakt. 284: 246-254.

Voorhees K.I., Glatz B.A. 1987.

phic

Gorman S.P. 1998.

to seventeen

Susceptibility

of

Propioni¬

antibiotics. Antimicrob. Agents Chemother. 11: 171-

100

REFERENCES

Weinberg Z.G, Ashbell G, terial inoculant

silages.

applied

Hen

at

Y., Azrieli A. 1995. The effect of a propionic acid bac¬

ensiling

the aerobic

on

stability

of wheat and

sorghum

J. Indust. Microbiol. 15: 493-497.

Witham P.K., Yamashiro CT., Livak

detection of Escherichia coli

K.J.,

Shiga-like

Batt CA. 1996. A PCR-based assay for the

toxin genes in

ground

beef.

Appl.

Environ.

Microbiol. 62: 1347-1353. F. 1995.

Wirsching

Molekularbiologische Charakterisierung

Resistenzplasmids pRE39

aus

einem

aus

des

konjugativen

MLS-

Fleisch isolierten Enterococcus-Stamm.

Thesis 11189. ETH Zürich. Woskow

S.A., Glatz B.A. 1990. Improved method for production of protoplasts and

polyethylene glycol induced

transformation of

Propionibacterium freudenreichii.

Abstr. Annu. Meet. Am. Soc. Microbiol. 90 Meet.: 268.

Yang X.,

McFadden B.A. 1993. A small

Synechocystis

plasmid, pCA2.4,

sp. strain PCC 6803 encodes

a

from the

cyanobacterium

Rep protein and replicates by

a

rolling

circle mechanism. J. Bacteriol. 175: 3981-3991. Yanisch-Perron

C, Vieria J., Messing

J. 1985.

and host strains: nucleotide sequences of the

Improved

M13mpl8

Ml3

and

phage cloning

pUC18

vectors

vectors. Gene 33:

103-119. Zaman S.,

Radnedge L.,

second-strand initiation

Richards H., Ward J.M. 1993.

during replication

of the

Analysis

of the site for

Streptomyces plasmid pIJlOl.

J. Gen. Microbiol. 139: 669-676. Zirnstein GW., into the

156.

Rehberger T.G 1991. Electroporation mediated gene transfer of pC194

Propionibacterium genome.

Abstr. Gen. Meet. Am. Soc. Microbiol. 91 Meet.:

101

Curriculum Vitae

Curriculum Vitae

Born June

25,1966 in Stuttgart (Germany)

1973-1981

Primary education

1981-1986

Secondary

in Dübendorf (ZH)

eduction in Dübendorf

(Kantonsschule

Zürcher

Oberland,

Filialabteilung Glattal, KZO)

1986

Matura

1989-1995

Study

Type C (mathematics and natural sciences)

of Food

Engineering

at the Swiss Federal Institute of Technology in

Zurich 1995

Diploma in Food Engineering in Zurich

1995-1998

Scientific

at the Swiss Federal Institute of Technology

(Dipl. Lm-Ing. ETH) collaborator

Microbiology Ph. D. thesis

and

at the Swiss

assistant

the

Laboratory

Federal Institute of

Technology

at

of in

Food

Zurich;