 

 

 

1

3. w.
:Mmﬂﬂﬁnxi I

i: “mﬁrﬂﬁisﬂium ..
‘28.. 1.2.15.4... .
1.4. It}?! r
,1.“
m n “m
tun:
iii:
(Run

v.30

.3)\‘.C\19) .
..|,.)..Ll:§ ‘
.kla‘AJiifq‘u‘.
.I.\lllnv‘l‘

 

LIBRARY
Michigan State
University

 

 

This is to certify that the
dissertation entitled

INHIBITION OF NEISSERIA
GONORRHOEAE EPITHELIAL CELL
INTERACTIONS BY VAGINAL
LACTOBACILLUS SPECIES

presented by

Rachel R. Spurbeck

has been accepted towards fulﬁllment
of the requirements for the

Ph.D. degree in Genetics

 

 

h six: iéiyﬁ‘mwt \lkwsk 6AM\

‘éJMajor Professor’s Signature
VKLRLE W t ‘31:: \s 0

Date

MSU is an Afﬁnnative Action/Equal Opportunity Employer

 

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE DATE DUE DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

5/08 KjProleocaPreleIRClDateDm.indd

 

INHIBITION OF NEISSERIA GONORRHOEAE EPITHELIAL CELL
INTERACTIONS BY VAGINAL LA C T OBA CILLUS SPECIES

By

Rachel R. Spurbeck

A DISSERTATION

Submitted to
Michigan State University
in partial fulﬁllment of the requirements
for the degree of

DOCTOR OF PHILOSOPHY
Genetics

2010

ABSTRACT

INHIBITION OF NEISSERIA GONORRHOEAE EPITHELIAL CELL
INTERACTIONS BY VAGINAL LA C T OBA CILLUS SPECIES

By
Rachel R. Spurbeck

Urogenital infections, including the sexually transmitted disease gonorrhea, are
one of the main reasons women visit the doctor in the United States. One potential
approach to the treatment or prevention of vaginally acquired infections is through the
use of beneﬁcial microbes known as probiotics. Most probiotic research, in both the
gastrointestinal and vaginal tracts, has focused on species from the genus
Lactobacillus since this genus is a member of the indigenous microbiota that is
associated with health. The presence of Lactobacillus species in the vaginal tract has
been correlated with a reduced risk of acquiring sexually transmitted infections upon
exposure, and therefore has the potential for development as a vaginal probiotic. In
this work Lactobacillusjenseniz‘, one of the most common species of lactobacilli in the
healthy human vaginal tract, was found to inhibit the interaction of Neisseria
gonorrhoeae with epithelial cells at the levels of adherence and invasion. This
inhibition was not due to direct killing of the pathogen or co-aggregation, as has been
suggested in the case of uropathogens. Further characterization of the inhibition of
adherence mediated by L. jensem'i determined that the inhibition was independent of
an epithelial cell response and due to Lactobacillus surface-associated proteins. By
characterizing a cell-free preparation of released surface components from L. jensem'z’,
it was shown that one of the surface associated inhibitory proteins was an enolase. The

surface bound enolase inhibits gonococcal adherence to epithelial cells when produced

recombinantly in E. coli and puriﬁed as a C-terminal HisG-tagged protein, and retains
enolase activity. Finally, it was determined that the substrate-binding site of the
enzyme, but not the enolase activity, is necessary for inhibition of gonococcal
adherence. The discovery of a protein from a commensal organism of the healthy
human vaginal tract that inhibits gonococcal adherence is signiﬁcant, and may lead to
the production of new therapeutics to treat and prevent the sexually transmitted

disease, gonorrhea, as well as other vaginally acquired infections.

ACKNOWLEDGEMENTS
I would like to thank Robert Britton and his lab for assistance, advice in
working with lactobacilli, and critical reading of manuscripts. I also thank Martha
Mulks, Todd Ciche, Helmut Bertrand, and Hilary Phelps for critical reading of this
document. I thank Dr. Lin Tao for his help with transformation of L. jensenii. I also

would like to thank Eric Carter for his help in understanding enzyme kinetics, and

Kannan Raghunathan and Paul Harris for puriﬁcation of His6-Eno. Proteomics data

were provided by the Michigan Proteome Consortium
(www.proteomeconsortium.org).

I also thank the College of Natural Sciences for the funding I have received in
the form of fellowships and Dr. Walt Esselman for providing me with teaching
assistantships throughout my career at MSU. Last but certainly not least, I would like
to thank my advisor, Cindy Arvidson, for allowing me to work in her lab and follow

where the science took us.

iv

TABLE OF CONTENTS

LIST OF FIGURES .................................................................................................... vii
KEY TO SYMBOLS OR ABBREVIATIONS .......................................................... xii
CHAPTER 1 .................................................................................................................. 1
Abstract .............................................................................................................. 2
Introduction ........................................................................................................ 3
Indigenous Microbiota ....................................................................................... 4
Mechanisms by which vaginal lactobacilli inhibit pathogen colonization ........ 7
Direct killing ................................................................................................ 8
Co-aggregation ........................................................................................... 10
Direct competition for host cell receptors .................................................. 11
Potential Vaginal Lactobacillus Mechanisms of Pathogen Inhibition ....... 14
Biosurfactants ............................................................................................ l 5
Mucin production ....................................................................................... 17
Increasing Ban'ier ﬁinction ........................................................................ l7
Immunomodulation ..................................... . ............................................... 1 8
Interference with pathogen gene expression .............................................. 18
Conclusion ....................................................................................................... 19
Signiﬁcance of current work ............................................................................ 20
CHAPTER 2 ................................................................................................................ 23
Abstract ............................................................................................................ 24
Introduction ...................................................................................................... 25
Materials and Methods ..................................................................................... 27
Results .............................................................................................................. 34
Conditions for N. gonorrhoeae and Lactobacillus co-culture ................... 34
Hydrogen peroxide production in cell culture medium. ............................ 35
Adherence of N. gonorrhoeae to epithelial cells pre-colonized with
lactobacilli. . . . . ........................................................................................... 36
Gonococcal invasion of epithelial cells pre-colonized with lactobacilli. ..39
Displacement of adherent gonococci by lactobacilli. ................................ 41
Spatial arrangement of lactobacilli and gonococci adherent to epithelial
cells. ........................................................................................................... 42
Lactobacilli and gonococci do not co-aggregate in co-culture. ................. 42
Lactobacillus pre-conditioned medium does not inhibit gonococcal
adherence. .................................................................................................. 44
Discussion ........................................................................................................ 46
CHAPTER 3 ................................................................................................................ 50
Abstract ............................................................................................................ 51

Introduction ...................................................................................................... 52

Materials and Methods ..................................................................................... 55
Results .............................................................................................................. 62
Lactobacillus inhibition of gonococcal adherence to ﬁxed epithelial cells
.................................................................................................................... 62
Pilus-mediated gonococcal adherence is not targeted by L. jensenz'i. ........ 64
Surface components of L. jensenii inhibit gonococcal adherence ............. 64
Released surface components (RSC) from L. jensem‘i inhibit gonococcal
adherence to epithelial cells in a dose-dependent manner ......................... 65
Inhibition of gonococcal adherence to epithelial cells remains after removal
of RSC ........................................................................................................ 68
Characterization of the inhibitory components of the RSC ....................... 69
Puriﬁcation of the RSC proteins ................................................................ 70
Soluble ﬁbronectin blocks inhibitory activity of the RSC. ........................ 72
Effect of Hisé-Eno on gonococcal adherence to Hec-l-B cells ................. 74
Discussion ........................................................................................................ 76
CHAPTER 4 ................................................................................................................ 81
Abstract ............................................................................................................ 82
Introduction ...................................................................................................... 83
Materials and Methods ..................................................................................... 85
Results .............................................................................................................. 91
DNA sequence analysis of L. jensem’z' ATCC 25258 eno. ......................... 91
Enolase activity of His6—GAP1 in the presence and absence of inhibitors
.................................................................................................................... 93
Effect of enolase inhibitors on His6-GAP1 reduction of gonococcal
adherence ................................................................................................... 94
Construction of an enolase mutant ............................................................. 95
Enolase activity of intact L. jensenii and cytoplasmic fractions ................ 96
Effect of gap] - on gonococcal adherence to epithelial cells ...................... 97
Effect of GAP]- RSC on gonococcal adherence to Hec-l -B cells. ........... 99
Discussion ...................................................................................................... l 01
CHAPTER 5 .............................................................................................................. 105
Signiﬁcance and impact of this work ............................................................. 106
Future experiments ......................................................................................... l 10
APPENDIX ................................................................................................................ 1 1 5
Abstract .......................................................................................................... l 16
Introduction .................................................................................................... 1 17
Materials and Methods ................................................................................... 118
Results and Discussion .................................................................................. 120
Conclusion ..................................................................................................... 121
REFERENCES .......................................................................................................... 123

vi

LIST OF FIGURES

Figure 2.1. Growth of N. gonorrhoeae in the presence or absence of L. jensenii. N.
gonorrhoeae (10 CFU/mL inoculum) was grown in the (El) presence or (O) absence

7 . . . . . .
of L. jensem'z' (10 CFU/mL inoculum). Sena] d11utlons were plated on selectlve media
at 2 h intervals to quantify CFU. Data are the averages of three independent
experiments. Error bars indicate standard deviations .................................................. 35

Figure 2.2. Adherence of N. gonorrhoeae to Hec-l-B cells pre-colonized with
lactobacilli. Epithelial cells were incubated with lactobacilli for 3 h. Non-adherent
lactobacilli were removed prior to the addition of gonococci, and the Hec-l -B cells
were incubated an additional 3 h. Relative adherence is the percentage of adherent
gonococci in wells containing lactobacilli divided by the percentage of adherent
gonococci in the control. The data are averages of three independent experiments
performed in duplicate. Error bars indicate standard deviations. * statistically
signiﬁcant ..................................................................................................................... 37

Figure 2.3. Adherence of N. gonorrhoeae to Hec-l-B cells in the presence of adherent
and non-adherent lactobacilli. Relative adherence is the percentage of adherent
gonococci in wells containing lactobacilli divided by the percentage of adherent
gonococci in the control. The data are the averages of three independent experiments

performed in triplicate. Ratios of gonococci to lactobacilli were: 1:0, 10 gonococcizno

lactobacilli; 1:1, 106 gonococci:106lactobacilli; 1:10, 106 gonococci:107 lactobacilli.
At 3 h pi. the CFU of N. gonorrhoeae doubled, and the total CFU of gonococci in the
control was not signiﬁcantly different from that of the Lactobacillus treated samples
[p-values: 0.324 (1 :1) and 0.536 (1:10)]. Error bars indicate standard deviations.
Student’s t-test p-values were 0.11 for 1:1 and 0.004 for 1:10. * statistically signiﬁcant
...................................................................................................................................... 39

Figure 2.4. Displacement of N. gonorrhoeae adherent to Hec-l-B cells by lactobacilli.
N. gonorrhoeae adherent to Hec-l-B cells after 3 h were then infected with L. gasseri
33323 or L. jensenii at an MOI of 100 for 1 h. Adherence data are presented as a
percentage of adhered gonococci in the well divided by that of the control and are the
averages of four or more experiments performed in triplicate. Error bars indicate
standard deviations. Student’s t—test p—values were 0.019 (L. gasseri 33323) and 0.174
(L. jensenii). *, statistically signiﬁcant ........................................................................ 42

Figure 2.5. Gram stains from an adherence assay performed as for the experiment
shown in Fig. 2.2 with Hec-l-B cells pre-colonized with 107 CFU L. jensem'i. Panel A
depicts gonococci and lactobacilli adherent to epithelial cells. Panel B depicts
gonococci adherent to epithelial cells in the absence of lactobacilli. Panel C depicts the
arrangement of gonococci and lactobacilli in the supernatant. In all panels the white
arrow points to lactobacilli and the black arrow points to gonococci (Olympus model
CHBS microscope, 100X magniﬁcation) .................................................................... 43

vii

Figure 2.6. Adherence of gonococci in medium pre-conditioned with L. jensem'i and
Hec-l -B cells. Percent adherence is the ratio of adherent gonococci to the total number
of gonococci for each condition. Number of lactobacilli used to condition the medium:

Fresh medium; no lactobacilli, fresh DMEM/5% FCS/GC supp II; PC-10, 106 CFU

lactobacilli:105 Hec-l-B cells; PC-100, 107 CFU lactobacilli:105 Hec-l-B cells. Data
are the average of four independent experiments performed in triplicate. Error bars
indicate the standard deviation. PC-lO: p = 0.631, and PC-100: p =0.884 ................. 45

Figure 3.1. The effect of L. jensem'i on gonococcal adherence. A. Adherence of N.
gonorrhoeae to live Hec-l-B cells in the presence or absence of L. jensenii. The data
are averages of 20 independent experiments performed in triplicate. Error bars indicate
standard error. Student’s t-test p-value < 0.001. B. Adherence of N. gonorrhoeae to
ﬁxed Hec-l -B cells in the presence or absence of L. jensem'i. The data are averages of
four independent experiments performed in triplicate. Error bars indicate standard
error. Student’s t-test p-value = 0.036. C. Adherence of non-piliated (MSl 1-307)
gonococci in the presence and absence of L. jensenii. The data are averages of six
independent experiments performed in triplicate. Error bars indicate standard error.
Student’s t-test p-value = 0.008. .................................................................................. 63

Figure 3.2. Adherence of N. gonorrhoeae in the presence of live, MeOH-ﬁxed, and
ProK-treated L. jensenii. Relative adherence is the percentage of adherent gonococci
in the presence of lactobacilli divided by the percentage of adherent gonococci from
the corresponding control. None, no lactobacilli; Live, untreated L. jensem’i; MeOH,
methanol-treated L. jensenii; ProK, proteinase K-treated L. jensenii.The data are
averages of three or more independent experiments performed in triplicate. Error bars
indicate standard error. Student’s t-test p-values were <0.001 (live), 0.001 (MeOH)
and 0.533 (ProK). ......................................................................................................... 63

Figure 3.3. RSC inhibition of gonococcal adherence is dose dependent. Hec-l-B cells
were incubated with either 1.7 mg/ml, 0.42 mg/ml, or 0.17 mg/ml RSC for 3 h before
the addition of gonococci. Gonococcal adherence to epithelial cells was reduced to
43.9 :I: 21.4% (L. jensenii, p= 0.01), 21.5 :h 14.4% (1.7 mg/ml RSC,p= 0.001), 65.1 i
1.7% (0.42 mg/ml RSC, p= <0.001), and 95.6 i 27.9% (0.17 mg/ml RSC, p= 0.81) of
the untreated control. The data are averages of three independent experiments
performed in triplicate. Error bars indicate standard error. ......................................... 67

Figure 3.4. Washing of RSC-treated cells prior to inoculation with gonococci does not
remove inhibitory activity. RSC was incubated with Hec-l-B cells for 3 h prior to
infection with gonococci. Gonococcal adherence to epithelial cells was reduced to
53.3 d: 17.2% (L. jensem',ip= 0.01), 31.2 i 15.0% (no wash, p= 0.001), 34.9 i 18.1%
(washed, p= 0.003). The data are averages of three independent experiments
performed in triplicate. Error bars indicate standard error. Student’s t-test p-value was
0.729 when washed RSC sample was compared to unwashed. ................................... 69

Figure 3.5. Proteinase K treatment of RSC abolishes inhibitory effect on N.
gonorrhoeae adherence to Hec-l-B cells. RSC was incubated for 2 h at 37°C with

viii

proteinase K at 20 mg/ml. Gonococcal adherence to epithelial cells was reduced to
68.2 :I: 2.0% (L. jensem'i, p<0.001), 50.1: 14.6% (RSC, p= 0.004), 96.3 i 18.9%
(ProK treated RSC, p= 0.755). The data are averages of three independent experiments
performed in triplicate. Error bars indicate standard error. ......................................... 70

Figure 3.6. Fractionation of RSC by anion exchange chromatography. A. Elution
proﬁle of samples detected by UV absorbance (2:260 nm) as the protein eluted from
the column. Peaks 1 and 2 are marked P1 and P2. B. SDS PAGE analysis of P1 and
P2. Long arrow points to the band with similarity to the enzyme enolase, and the short
arrow points to the band with similarity to the enzyme GAPDH, as determined by
MS/MS. C. Gonococcal adherence to epithelial cells pretreated with P1 and P2.
Treatment of epithelial cells inhibited gonococcal adherence to 52.9 i 23.9% (L.
jensenii, p=0.027), 51.2 :t 14.0% (P1, p=0.004), and 48.8 i 14.8% (P2, p=0.004). The
data are the averages of three independent experiments performed in duplicate. Error
bars indicate standard error. ......................................................................................... 71

Figure 3.7. Gonococcal adherence to cells treated with RSC that had been pre-treated
with increasing amounts of soluble ﬁbronectin. Gonococcal adherence frequencies are
reported as relative to a no RSC-no Fn control and Student’s t-tests were used to
compare each Fn treatment with no F n treatment. Adherence to epithelial cells was
reduced to 40.4 i 17.0% (0.42 RSC), 54.9 i 8.0% (RSC-2.5 pM Fn, p= 0.374), 70.1 :t
8.7% (RSC-5.0 pM Fn,p= 0.121), 74.6 i 9.9% (RSC-7.5 [1M Fn,p= 0.091) and 95.3
:I: 7.5% (RSC-10 MM Fn, p= 0.019). The data are averages of three independent
experiments performed in duplicate. Error bars indicate standard error. .................... 74

Figure 3.8. Puriﬁcation of His6-Eno and its effect on gonococcal adherence to Hec-l -

B cells. A. An SDS-PAGE gel depicting the protein proﬁle of overloaded His6-Eno to
show impurities (lane 1), a kalidescope standard ladder (lane 2), nickel and anion

exchange column puriﬁed His6-Eno (lane 3), uninduced E. coli BL21)\DE3
pET24azzeno (lane 4), and induced E. coli BL21>\DE3 pET24a::eno (lane 5). The

arrow points to His6-Eno. B. Gonococcal adherence to cells pre-treated with various
concentrations of His6-Eno. The N gonorrhoeae control (no His6-Eno) adhered at a
frequency of 20.1 d: 4.0%. In the presence of His6-Eno the adherence of N.
gonorrhoeae was reduced to 10.4 i 2.8% (0.14 mg/ml His6-Eno, p= 0.028), 6.5 i
1.8% (0.28 mg/ml His6-Eno, p= 0.013), 6.9 :1: 0.7% (0.42 mg/ml His6-Eno, p= 0.004),

6.7 a 1.4% (0.56 mg/ml His6-Eno, p= 0.013), 3.81: 1.0% (0.83 mg/ml His6-Eno, p=
0.006). ......................................................................................................................... 75

Figure 4.1. Insertional mutagenesis of L. jensem'i 25258 gap] . A. Schematic diagram
of plasmid integration into chromosomal locus to disrupt gap] . Primers used to
ampliﬁy regions to conﬁrm construct are indicated. B. PCR ampliﬁcation to verify

insertion. Genomic DNA from ErmRCmS transformants was used as template for PCR
using primers enond and M13rev. Templates: 1 = undiluted mutant genomic DNA,

ix

1:10 = mutant genomic DNA, 1:100 mutant genomic DNA, - = no template control,
and WT = L. jensenii WT genomic DNA. ................................................................... 88

Figure 4.2. GAPl protein characteristics. A. The primary amino acid sequence of L.
jensenii 25258 GAPl. The region shaded in gray corresponds to the TIM 6—barrel
domain common to enolases (16). Amino acids of interest that are speciﬁc to GAPl
are highlighted in bold font. The conserved amino acids are marked below the residue
with # for metal binding, * for substrate binding and A for dimer interface. B. A
phylogenetic tree of enolase protein sequences. % = identity to GAPl. Distances were
calculated using the Grishin model with maximum sequence difference of 0.85 (142)
...................................................................................................................................... 92

Figure 4.3. His6-GAP1 inhibition of gonococcal adherence to glutaraldehyde ﬁxed
Hec-l-B cells. N. gonorrhoeae alone adhered to the ﬁxed cells with a frequency of

12.9 d: 1.6%. When treated with 0.14 mg/ml His6-GAP1, gonococcal adherence was
reduced to 7.9 :I: 1.3% (no inhibitors, p = 0.036), 7.9 a 1.5% (Ca2+, p = 0.043), or 13.4

:I: 0.4% (F-+ Pi, p = 0.746). The data are the averages of 3 independent experiments
performed in duplicate. Error bars depict standard error. ............................................ 95

Figure 4.4. Enolase activity of L. jensenii WT and L. jensenii gap]: A. Enolase
activity of increasing concentrations of intact lactobacilli plotted as a function of PEP
produced. B. Enolase activity of increasing amounts of cytoplasmic extracts from L.
jensenii WT and L. jensenii gap]: Solid line = L. jensenii WT and the dotted line = L.

jensenii gap]: The data are the averages of 3 independent experiments performed in
triplicate. Error bars depict standard error. .................................................................. 97

Figure 4.5. Inhibition of N. gonorrhoeae adherence to epithelial cells by L. jensenii

WT and L. jensenii gapI'. Relative adherence is the percentage of adherent gonococci
in the presence of lactobacilli divided by the percentage of adherent gonococci in the
no Lactobacillus control. The data are the averages of 3 independent experiments
performed in triplicate. * Signiﬁcantly different from no lactobacilli, ** Signiﬁcantly
different from L. jensenii WT and no lactobacilli. Error bars depict standard error. .98

Figure 4.6. Inhibition of gonococcal adherence by released surface components (RSC)

from L. jensenii gap] - and L. jensenii WT. Gonococcal adherence frequencies are
reported as relative to a no RSC control, and p-values are reported as compared to WT
RSC. Gonococcal adherence to epithelial cells was reduced to 31.2 d: 6.8% (1.7 mg/ml

wr RSC), 98.3 :1: 12.0% (0.42 mg/ml GAPI- RSC, p = 0.004), 79.1 :I: 10.5% (0.83
mg/ml GAP]- RSC, p = 0.011), 68.5 i 15.4% (1.7 mg/ml GAPl' RSC, p = 0.048), and

42.8 a: 6.2% (2.2 mg/ml GAPl' RSC, p = 0.206) of the untreated control. The data are
the averages of 3 independent experiments performed in duplicate. Error bars depict
standard error. ............................................................................................................ 100

Figure 5.1. Potential probiotic mechanisms of inhibition of pathogen adherence. A.

Secreted compounds such as (O) lactic acid, ((I) hydrogen peroxide, (A )
biosurfactants, or (O) bacteriocins. B. Direct competition for host cell receptors. C.
Co-aggregation with the pathogen. D. Inhibition of virulence gene expression. E.
Inducing the host epithelial cells to become more resistant to infection. .................. 108

Figure 5.2. Proposed mechanism by which L. jensenii enolase inhibits gonococcal
adherence. L. jensenii deposits its surface bound enolase on epithelial cell receptors.
This blocks the adherence of N. gonorrhoeae to the same receptors. ....................... 109

Figure 5.3. The peptide sequence of L. jensenii GAP]. " only in L. jensenii like group,
# in L. jensenii like group and pathogen group, @ only in Lactobacillus groups, and *
conserved in all groups. In bold are residues of interest for site directed mutagenesis.
The gray shaded region corresponds to the TIM barrel domain. ............................... 112

Figure A.1. Semi-quantitative PCR of L. jensenii 16S rDNA (panel B) and N.
gonorrhoeae pil T (panel A). A. Lane 1: no template control; Lane 2: 1 kb ladder; Lane
3: N. gonorrhoeae genomic DNA; Lanes 4-8: 10 rig/pl, 1 ng/ul, 100 pg/ul, 10 pg/ul,
and 1 pg/ul cDNA from N. gonorrhoeae; Lanes 9-13: 10 ng/ul, 1 ng/ul, 100 pg/ul, 10
pg/ul, and 1 pg/ul cDNA N. gonorrhoeae-mixed with L. jensenii which was subjected
to the subtractive hybridization method to remove the L. jensenii RNA before reverse
transcription; Lane 14: 1 kb ladder. B. Lane 1: no template control; Lane 2: 1 kb
ladder; Lane 3: L. jensenii genomic DNA; Lanes 4-8: 10 ng/ul, 1 ng/ul, 100 pg/ul, 10
pg/ul, and 1 pg/ul cDNA from L. jensenii; Lanes 9-13: 10 ng/ul, 1 ng/ul, 100 pg/ul,
10 pg/ul, and 1 pg/ul cDNA N. gonorrhoeae-mixed with L. jensenii which was
subjected to the subtractive hybridization method to remove the L. jensenii RNA

before reverse transcription, and Lane 14: 1 kb ladder ....................................................
.................................................................................................................................... 122

xi

KEY TO SYMBOLS OR ABBREVIATIONS

BV .................................................................................................... Bacterial Vaginosis
CFU ............................................................................................... Colony Forming Unit
C02 ........................................................................................................ Carbon Dioxide
DMEM .................................................................. Dulbecco's Modiﬁed Eagle Medium
ECM ................................................................................................ Extracellular Matrix
F CS ...................................................................................................... Fetal Calf Serum
F-+Pi ................................................................................... Fluoride And Inorganic Phosphate
Fn .................................................................................................................. Fibronectin
GAPDH ................................................... Glyceraldehyde-3-Phosphate Dehydrogenase
GC supp II ......................................................................................... 12.4 11M Fe(NO3)3
GCV ......................................................................................... GC Agar + Vancomycin
His6-Eno ...................................................................... Epitope Tagged Enolase Protein
HIV ............................................................................. Human Immunodeﬁciency Virus
H202 ................................................................................................ Hydrogen Peroxide
HRP ............................................................................................ Horseradish Peroxidase
LEE .............................................................................. Locus of Enterocyte Effacement
LOS ................................................................................................. Lipooligosaccharide
LTA ..................................................................................................... Lipoteichoic Acid
MeOH .............................................................................................................. Methanol
MRS ................................................................... de Man, Rogosa, And Sharpe Medium
Opa ......................................................................................................... Opacity Protein

xii

+

P ........................................................................................................................ Piliated
P‘ ............................................................................................................... Non-pPiliated
PBS ....................................................................................... Phosphate Buffered Saline
PEP ................................................................................................ Phosphoenolpyruvate
2PGE ................................................................................................ 2-Phosphoglycerate
p.i. ............................................................................................................. Post Infection
PID .................................................................................... Pelvic Inﬂammatory Disease
PCR ..................................................................................... Polymerase Chain Reaction
PAGE ..................................................................... Polyacrylimide Gel Electrophoresis
ProK ............................................................................................................ Proteinase K
RSC ................................................................................. Released Surface Components
RT ................................................................................................ Reverse Transcription
semiQ ................................................................................................. Semi-Quantitative
Slps .............................................................................................. Surface Layer Proteins
STI .................................................................................. Sexually Transmitted Infection
TMB .............................................................................. 3,3’,5,5’-Tetramethylbenzidine
WT ................................................................................................................. Wild Type

xiii

CHAPTER ONE
Spurbeck R. R., and Arvidson, CG. (2010). Lactobacilli at the front line of defense

against vaginal infections.

ABSTRACT

Probiotics, organisms that provide a health beneﬁt to the host, are being
studied for the treatment or prevention of infectious diseases. One of the most rapidly
developing areas in which probiotics are being studied is in the management of
vaginally acquired pathogens. Most of the work has centered on determining which
bacterial species from the genus Lactobacillus have potential as probiotics. Several
species have been found to produce compounds that inhibit the growth of pathogenic
microorganisms, such as hydrogen peroxide, lactic acid, and bacteriocins. Other
lactobacilli demonstrate an ability to reduce the adherence of pathogens to vaginal
epithelial cells, including the bacterial vaginosis associated pathogen, Gardnerella
vaginalz's. However, more research is required to demonstrate what mechanism(s)
these organisms use to defend the vaginal tract against infection in vivo. This review
discusses the known mechanisms by which vaginal lactobacilli inhibit pathogen
colonization in vitro. Other mechanisms that still need to be investigated to conclude
what factors of the indigenous microbiota are necessary and sufﬁcient for prevention

of diseases of the female reproductive tract are also discussed.

INTRODUCTION

Infections of the female reproductive tract, such as vaginitis, pelvic
inﬂammatory disease (PID), vulvovaginal candidiasis, bacterial vaginosis (BV), and
sexually transmitted infections (STIs), are among the most common reasons women
visit the doctor in the United States. In 2008, there were over 1.5 million cases in the
United States of the three reportable STIs [chlamydia (1,201,664), gonorrhea
(332,899), and syphilis (13,637)] (27). Since asymptomatic STIs often go undetected,
the reported numbers are probably far lower than the actual incidence of these
diseases.

If an STI is left untreated, serious and potentially irreversible complications
can result. It is estimated that 30-40% of all female infertility is due to damage caused
by an STI that ascended to the fallopian tubes. In fact, 10-40% of chlamydia]
infections develop into PID, which is an infection of the uterus, fallopian tubes,
ovaries, and the surrounding tissues (118). PID affects approximately 1,000,000
women per year in the United States and it is estimated that 10% of those women will
become sterile as a result (184). Due to the mild or absent symptoms of PID, over two
thirds of all PID cases go undiagnosed until fertility problems arise (184). Women
who have had at least one episode of PID are 6-10 times more likely to have an
ectopic pregnancy, with an estimated 40-50% of all ectopic pregnancies being
attributable to P11) (184), demonstrating the need for better control of vaginal
infections.

While reproductive problems are a major concern when women are infected,

another critical issue is that the transmission of the human immunodeﬁciency virus

(HIV) is facilitated by the presence of an STI. In 2008, 56,300 new HIV infections
were reported in the United States, with 31% of these new infections being attributed
to high risk heterosexual contact (28). Gonorrhea and chlamydia, the two most
common bacterial STIs in the United States, are known to increase the shedding of
HIV leading to an increased transmission to sexual partners (64, 123). Vulvovaginal
candidiasis has also been correlated with increased shedding of HIV (64).

BV is a dysbiosis of the vaginal microbiota that is characterized by a shift from
a bacterial community dominated by the genus Lactobacillus to an overgrowth of
anaerobic Gram-negative rods and Gardnerella vaginalis (99). This shift in the
bacterial community is also correlated with an increase in the vaginal pH (100). BV is
associated with an increased risk of acquiring HIV, gonorrhea, Chlamydia and
vulvovaginal candidiasis (99, 187). Therefore, identiﬁcation of a method to prevent
dysbiosis of the vaginal microbiota could reduce the spread of STIs, HIV, and
decrease the incidence of vulvovaginal candidiasis. This would lessen the amount of
resources needed for treatment of vaginal infections and improve the quality of life for
women at risk for these diseases.

INDIGENOUS MICROBIOTA

Since Elie Metchnikoff discovered microbes in Bulgarian yogurt that seemed
to prolong life in 1907, microorganisms have been shown to impact human health
(71). This discovery led to research of the interactions between commensal organisms
and the host. Several bacterial species indigenous to the mucosal surfaces of the
human body were consequently found to be beneﬁcial to the health of the host. These

good microbes, now called probiotics, are deﬁned as "live microorganisms which

when administered in adequate amounts confer a health beneﬁt on the host (47). While
the majority of probiotic research has been conducted on the gastrointestinal
microbiota, relatively little has been done on inhabitants of the vaginal tract (58).

The vaginal microbiota is a diverse community comprised of at least 40
bacterial genera (146). In healthy women of childbearing age, eight distinct
supergroups of microbial species have been associated with normal indigenous
microbiota, with Lactobacillus being the predominant genus in ﬁve (195). Lactobacilli
are present at concentrations of 107-108 CFU/ml of vagina] ﬂuid (135), and the four
most prevalent species are L. iners, L. crispatus, L. gasseri, and L. jensenii as
determined by culture independent techniques (129, 172, 194-195).

Lactobacilli dominate the vaginal microbiota due to several factors. First, they
are facultative anaerobic or anaerobic, allowing them to survive in the oxygen-limited
environment of the vaginal tract. Second, in the post-menarchaI/pre-menopausa]
vaginal tract, there is an accumulation of glycogen in the vaginal epithelium due to the
elevated level of estrogen; lactobacilli use this glycogen as a substrate for growth
(122, 189). Through the fermentation of glycogen, lactobacilli produce lactic acid
which contributes directly to the low pH of the vaginal tract (pH 4-4.5) (21, 122, 172).

To maintain their presence in the vaginal tract it is thought that lactobacilli
must adhere to the epithelium otherwise they would be removed by the continual ﬂow
of vaginal ﬂuid through the mucosa] layer. However, most Lactobacillus isolates
adhere poorly to vaginal epithelial cells (5). In a study examining the composition of
the microbial community in vaginal biopsy specimens isolated from healthy

premenopausal women, the microbiota was not tightly adhered to the epithelium.

Instead, the lactobacilli and other microbes associated with the healthy state were
found to be primarily in the vaginal ﬂuid and mucosal layers (162). However, in
biopsy specimens obtained from women with BV, Gardnerella vaginalis produced
tightly adherent bioﬁlms on the surface of the epithelium (162). This demonstrates that
in a state of perturbation the community is more closely associated with the vaginal
epithelium than when in a state of health. However, most research focuses on
lactobacilli that are highly adherent to vagina] epithelial cells in vitro (20, 102),
thereby overlooking some species that may have potential as probiotics. Two of the
most common vaginal Lactobacillus species, L. crispatus and L. jensenii, were
determined to most consistently colonize the vaginal tract in a study over a period of
eight months (172), and therefore may be the best candidates for sustainable
probiotics. In addition, epidemiological evidence shows that women colonized with
either L. crispatus or L. jensenii in the vaginal tract and/or rectum were less likely to
have BV than women colonized with other species of lactobacilli (10), supporting the
role of these organisms in protecting the vaginal tract from infection.
Epidemiological evidence suggests that women colonized with high levels of
vaginal lactobacilli are less susceptible to STIs such as gonorrhea and chlamydia

(187). Lactobacilli and their products have been shown to affect the grth of
Neisseria gonorrhoeae in vitro (143). In one study, H202 was found to be the major
inhibiting factor (156); however, gonococci produce catalase which breaks down

H202 into H20 and Oz. The activity of this enzyme increases when gonococci are co-

cultured with lactobacilli at a neutral pH in vitro (193). However, at low pH (~4-5) the

production of catalase decreased, and the inhibition of gonococcal growth was

observed, therefore, H202 and low pH may be necessary for an effect on the growth of

N. gonorrhoeae (193). To further complicate the situation, gonococci utilize lactate as
a carbon source (152), therefore, whether lactobacilli can kill gonococci in vivo has yet
to be determined.

Several studies have shown that lactobacilli can inhibit steps in the infection
process including adherence of pathogens associated with urinary tract infections,
vulvovaginal candidiasis, and BV (13-15, 102, 119-120, 157, 170, 191). Furthermore,
lactobacilli were found to be able to disperse bioﬁhns associated with BV (145). All of
these observations support the premise that lactobacilli act as protective agents against
vaginal infections; however, few studies have been conducted in vivo to verify the
biological relevance of these results.

MECHANISMS BY WHICH VAGINAL LACTOBACILLI INHIBIT
PATHOGEN COLONIZATION
Lactobacilli are thought to utilize several mechanisms to inhibit pathogen

colonization of the vaginal tract. The most studied mechanism is direct killing of the

pathogen by compounds secreted by lactobacilli, such as lactic acid, H202, and

bacteriocins. Only the production of lactic acid has been demonstrated in vivo, and
therefore, is the lone mechanism that has known biological relevance at this time.
Other inhibitory mechanisms studied using in vitro systems, such as cell culture
models, are co-aggregation between the pathogen and the lactobacilli, and competition
for adherence to host cell receptors. However, both co-aggregation and competition
for host receptors have not been demonstrated as of yet in vivo, therefore, the

biological relevance of these phenomena is unknown. To increase the applicability of

probiotics for the treatment and prevention of vaginal infections, it is necessary for
these proposed mechanisms to be examined in vivo.
Direct killing

Lactobacilli produce several compounds that can inhibit the growth of

urogenital pathogens including lactic acid, H202, and bacteriocins (158). Lactic acid

produced by lactobacilli contributes to the low pH (~ 4-6) of the vaginal tract,
inhibiting the growth of bacteria that are not acid tolerant (21). For example, a vaginal
isolate, L. acidophilus CRL 1259, was found to kill uropathogenic Escherichia coli
under conditions that favored lactic acid production (67). Likewise, supematants ﬁom
L. salivarius subspecies salivarius CRL 1259 cultures were shown to inhibit the
growth of E. coli, Klebsiella sp., Staphylococcus aureus, Streptococcus agalactiae,
and G. vaginalis due to the low pH caused by lactic acid production (116) In another
study, it was determined that the amount of inhibition by a given species of
Lactobacillus correlated with the amount of lactic acid produced (105). Therefore,
acidiﬁcation of the vaginal tract by lactobacilli is a major factor in protecting against

pathogen colonization.

H202 generated by lactobacilli is another compound that can kill incoming

pathogens (135, 157, 172, 191). It has been determined that 96-98% of vaginal

Lactobacillus isolates from healthy women are HzOz-producers (46, 134), whereas

only 6-23% of lactobacilli isolated from women with infections associated with

preterm delivery, produced H202 (77). Pre-term delivery is thought to be induced by

infections of the vaginal tract that ascend to the uterus causing inﬂammation of the

lower uterine pole (188). BV is associated with preterm delivery, as early acquisition

during pregnancy correlates with a higher risk of premature birth (188). The presence

of HZOZ-producing Lactobacillus strains, including L. jensenii, L. crispatus, and L.

gasseri, is associated with a reduced risk of bacterial vaginosis, pre-term birth, and

chorioamnionitis, which is one of the main causes of fetal and neonatal death (188).

HZOZ-producing lactobacilli are also correlated with a reduced risk of acquiring

sexually transmitted pathogens, such as N. gonorrhoeae, C. trachomatis, and HIV (1 7,
80, 147, 187). Furthermore, the hydrogen peroxide produced by several Lactobacillus

species has been shown to inhibit the growth of G. vaginalis, S. aureus, and N.

gonorrhoeae (114-115, 156, 167). Therefore, the focus has been on HzOz-producing

Lactobacillus strains for the development of probiotics. However, HzOz-production

by lactobacilli has not been measured in vivo; therefore, the biological relevance of

H202 in the prevention of vaginal diseases has yet to be determined.

Bacteriocin production is also involved in direct killing of microorganisms by
lactobacilli. Bacteriocins are bactericidal proteins produced by bacteria to reduce
competition for shared resources (72). Several bacteriocin-like compounds are
produced by lactobacilli that affect the growth of bacteria from the same genus as well
as a variety of Gram-positive and Gram-negative pathogens (72, 94, 164). A
bacteriocin from the culture supernatant of L. salivarz'us subspecies salivarius CRL
1259 was able to inhibit the growth of Enterococcus faecalis and N. gonorrhoeae
demonstrating that it can inhibit both Gram-positive and Gram-negative bacteria

(116). Another bacteriocin isolated from L. fermentum L23 was also fungicidal (127).

Some of the most prevalent vaginal Lactobacillus species (L. gasseri, L. crispatus, and
L. jensenii) have been found to produce bacteriocins (68, 73). The bacterocins
produced by these Lactobacillus strains had microbicidal activity against the vaginal
pathogens G. vaginalis, C. albicans, and E. coli (68). Even so, a study of 45 human
vaginal isolates of the species L. crispatus, L. jensenii, L. gasseri, and L. plantarum,
determined that none produced bacteriocins (100). This suggests that bacteriocins are
not generally produced by vaginal lactobacilli, and may be a minor factor in the
reduced risk of transmission observed in epidemiological studies. Nevertheless, the
microbicidal activity associated with bacteriocin producing strains suggests there is
potential for their use as a probiotic treatment of BV or vulvovaginal candidiasis,
although ﬁirther research is necessary to demonstrate that bacteriocins are effective in
vivo.
Co-aggregation

Lactobacilli could also protect the vaginal tract from infection by aggregation
with the incoming pathogen, thereby titrating the pathogen away from the surface of
the host epithelium. This would allow it to be removed more effectively by the vaginal
ﬂuid and/or killed by antimicrobial products secreted by the lactobacilli (102). Some
vaginal lactobacilli, including L. gasseri, L. jensenii, and L. crispatus, have been
shown to co-aggregate with uropathogenic E. coli (UPEC). However, there were
differences in the amount of co-aggregation seen, with the most prevalent lactobacilli
co-aggregating less efﬁciently than rarer isolates (45, 79). Lactobacillus sp. HV 142, a
producer of aggregation promoting factor (APF), co-aggregated with a wider spectrum

of UPEC strains than most lactobacilli tested. This may suggest that APF expressing

10

strains could be more effective against UPEC (79). Lactobacilli have also been shown
to co-aggregate with G. vaginalis and C. albicans (102). Again, co-aggregation varied
from no aggregation to complete aggregation, differing between strains as well as

between species (102). The two most highly aggregative, L. salivarz'us FV2 and L.

gasseri MB 335, were also strong H202 producers and had signiﬁcant bactericidal

effects against G. vaginalis, seeming to support the hypothesis that aggregation
promotes killing of the pathogen. However, these strains were not fungicidal,
therefore, the ability to co-aggregate with a pathogen does not correlate directly with
the ability to kill the pathogen (102). Taken together, co-aggregation may enhance the
inhibition of pathogen colonization, but is not a universal mechanism utilized by
vaginal lactobacilli. This mechanism has yet to be tested in vivo, which would
determine if co-aggregating lactobacilli should be examined further for use as
probiotics.
Direct competition for host cell receptors

Another mechanism by which lactobacilli could protect the human vaginal
tract is through competition with pathogenic microbes for host cell receptors. If a
pathogen adheres to epithelial cells via the same receptors already occupied by the
indigenous lactobacilli, then its adherence will be inhibited by exclusion. Several
Lactobacillus species isolated from the vaginal tract have been shown to exclude G.
vaginalis, P. bivz'a C. albicans, S. aureus, S. agalactz'ae, E. coli, K. pneumoniae, and
Pseudomonas aeruginosa from adhering to host epithelial cells (13-15, 68, 119-120,
191). Furthermore, if lactobacilli have a higher afﬁnity than the pathogenic bacteria

for the host receptors, they could displace already adherent pathogens and potentially

11

be used as a treatment for vaginal infections. Some lactobacilli have been shown to
displace G. vaginalis and C. albicans, suggesting that lactobacilli could be used to
treat BV or vulvovaginal candidiasis (20, 102). However, it is not yet clear what
factors lactobacilli use to exclude or displace pathogens from the host epithelium.
Therefore, more in depth studies will be necessary to identify such factors and
determine their mechanism of action.

A cell-free preparation of lipoteichoic acid complexed with peptidoglycan
(LTA-peptidoglycan) from a Lactobacillus strain was shown to exclude the
uropathogens E. coli, K. pneumoniae, and P. aeruginosa from uroepithelial cells (30).
However, the authors hypothesized that the exclusion of pathogen adherence was
likely due to steric hindrance, rather than direct competition for receptors, as the bulky
LTA-peptidoglycan complex could inhibit pathogens from reaching adjacent receptors

(30). L. paracasei CRL 1289, a strain that can inhibit the growth of S. aureus by

production of H202, also reduces pathogen adherence by competing directly for host

cell receptors (19]). Using a mouse model, L. paracasei CRL 1289 was shown to not
only reduce the inﬂammatory response to S. aureus infection but also reduce the
number of the pathogenic bacteria colonizing the vaginal tract when infected with a
relatively low infectious dose. However, the lactobacilli were unable to protect the
mice when the infectious dose was increased (192). Therefore, more work is needed to
improve the probiotic strain before use as a preventative for S. aureus vaginal
infections.

While research on the adherence factors of vaginal Lactobacillus isolates is

rather limited, some have been characterized for intestinal lactobacilli. The one study

12

pertaining to vaginal lactobacilli suggests that L. acidophilus and L. gasseri use
glycoproteins while L. jensenii utilize carbohydrates to adhere to vaginal epithelial
cells (20). In the intestinal tract, it has been shown that lactobacilli utilize proteins (54,
56, 110, 138), carbohydrate components, and/or lipoteichoic acid as adhesins (1-2, 20,
31, 53-54, 56, 113). These adherence factors bind to several different substrates
including colonic mucus, and extracellular matrix (ECM) proteins such as ﬁbronectin,
laminin, and collagen (3, 6, 8, 26, 59, 70, 89, 140, 151).

In L. reuteri, some of the proteinaceous adhesins have been deﬁned to the
molecular level, such as MapA, a mucus adhesion promoting protein, which binds to
human intestinal epithelial cells (110). Also, a collagen binding protein is expressed
by L. reuterz’ (3, 140), and was determined to be part of a biosurfactant preparation
known to inhibit uropathogen adherence (55). Other lactobacillar adhesins involved in
binding to ECM are enzymes normally found in the cytoplasm such as enolase and
glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (7, 26, 144). These enzymes
are surface associated on several lactobacilli, and can be shed by changing the pH of
the medium (8). However, the involvement of these adhesins in pathogen inhibition or
competition for receptors has yet to be characterized.

Surface layer proteins (Slps) have also been implicated in the adherence of
lactobacilli to epithelial cells, and may play a role in the interference with pathogen
colonization. Slps are located outside of the bacterial cell wall and form a protective
paracrystalline layer around the bacterium (63). Slps from the intestinal probiotic
strain L. helevetz'cus, were shown to inhibit the binding of E. coli 0157:H7 to intestinal

epithelial cells (63), thus indicating that these proteins can affect the adherence of

13

pathogens. However, the role of Slps as inhibitors of pathogen adherence in the
vaginal tract has not been demonstrated. Of the Lactobacillus species that dominate
the microbiota of the vaginal tract, only L. crispatus produces a surface layer, and the
Slp was determined to be a collagen binding protein (166). APF, the autoaggregation
promoting factor expressed in some strains of L. gasseri, has also been suggested to be
a Slp (178). However, L. jensenii, and some strains of L. gasserz', do not produce
surface layers (18, 129); therefore, the role of Slps in inhibition of vaginal pathogen
adherence is yet to be determined.

Given that the adherence mechanisms of vaginal lactobacilli are still relatively
uncharacterized, identiﬁcation of the adhesins used by these organisms would greatly
advance the ﬁeld of probiotics by allowing more sustainable probiotics to be
developed. Since several vaginal lactobacilli are thought to directly compete with
pathogens for receptors on epithelial cells, understanding the adherence mechanisms
of lactobacilli, would enable predictions to be made as to which probiotic would
exclude or displace speciﬁc pathogens.

POTENTIAL VAGINAL LA CT OBA CILLUS MECHANISMS OF PATHOGEN
INHIBITION
Several mechanisms have proposed to inhibit pathogen colonization that have yet to
be examined in the vaginal tract, such as biosurfactants, mucin production, increased
barrier function, immunomodulation, and change in virulence gene expression.
However, lactobacilli from the gastrointestinal tract have been shown to utilize some

of these mechanisms. Since there may be niche speciﬁc mechanisms for pathogen

l4

inhibition, the mechanisms that have been elucidated in the intestinal tract should be
studied for relevance for the vaginal tract.
Biosurfactants

Biosurfactants are amphipathic molecules, which contain both hydrophobic
and hydrophilic domains (139). These compounds, which can be protein,
carbohydrate, or glycoproteins, are utilized by microbes for a variety of purposes, such
as emulsiﬁcation of hydrocarbons, quorum sensing, bioﬁlrn regulation, competition by
antimicrobial activity, and regulation of attachment and detachment (139). A
biosurfactant-producing organism can use the surfactant as an adhesin, allowmg the
bacteria to adsorb to the surface through hydrophobic or hydrophilic interactions.
Upon detachment from the substrate, some bacteria leave behind the biosurfactant
molecules, changing the chemistry of the surface. This alteration could interfere with
the adherence of an incoming bacterial species reducing its colonization of that
surface.

More than 15 strains of lactobacilli have been shown to produce biosurfactant
activity. Some of the lactobacillar biosurfactants are secreted, whereas others are
associated with the surface of the bacterium (137). While the effect of biosurfactants
on vaginal pathogens is unknown, there is evidence that biosurfactants produced by
vaginal lactobacilli effect uropathogen adherence. Biosurfactants produced from
Lactobacillusfermentum B54 and Lactobacillus acidophilus RC-l4 [also known as L.
fermentum RC-14 and Lactobacillus reuteri RC-14], isolates from the healthy human
vaginal tract, were tested for effects on the adherence of E. faecalis to glass surfaces.

These were compared to the effect of a biosurfactant produced from L. casei subsp.

15

rhamnosus 36, isolated from a woman with a history of urogenital infections (177).
The strains isolated from the healthy vaginal tract were able to reduce the initial
deposition of E. faecalis onto the glass, whereas L. casei subsp. rhamnosus 36 did not
signiﬁcantly affect the adherence of E. faecalis to glass (177). In a similar study, the
biosurfactant from L. acidophilus RC-14 (surlactin) was able to inhibit the adherence
of E. faecalis to silicone rubber (176). The ability of surlactin to inhibit the adherence
of the uropathogens C. albicans, K. pneumoniae, Providencia stuartii, P. aeruginosa,
E. coli, and Staphylococcus epidermidis to silicone rubber has since been studied
(175). Surlactin could inhibit the adherence of all of these pathogens to varying
degrees. C. albicans was inhibited the least, with only a 12% reduction seen after 4 h
incubation. S. epidermidis was the most affected, with an 82% inhibition observed
within the same time period (175). Since both silicone rubber and glass can be used in
catheters, the ability of Lactobacillus-produced biosurfactants to inhibit the adherence
of uropathogens to these substrates could potentially be utilized to reduce the
incidence of urinary tract infections in catheterized patients. Furthermore, surlactin
inhibited S. aureus adherence to surgical implants and abscess formation by 89% in a
rat model (51), and biosurfactants produced by other lactobacilli have also been shown
to inhibit S. aureus bioﬁlrn formation (183). Therefore, biosurfactants produced by
lactobacilli could play a signiﬁcant role in reducing nosocomial infections, and need to
be examined further to determine the involvement of these molecules in the inhibition

of vagina] pathogen colonization.

l6

Mucin production

The vaginal tract is lined by a thick mucosa] layer that serves as a physical
barrier against microbial colonization. Mucin, the main component of mucus, is a
glycoprotein which forms a gel-like network over the epithelial cells (141). This
matrix is difﬁcult for incoming microorganisms to penetrate. In the gastrointestinal
tract, adherent lactobacilli have been shown to increase the production of mucins by
the epithelium by increasing the expression of mucin genes (4, 76, 92-93, 103). This
increase in mucin production in response to Lactobacillus inhibits the adherence of
both enterohemorrhagic and enteropathogenic E. coli strains to the epithelial cells,
thereby reducing pathogen colonization (76, 93). While studies on the effect of
lactobacilli on the production of mucin in the vaginal tract have yet to be conducted, it
has been demonstrated that adherence of C. albicans is reduced when cervical mucins
are present (36). Thus it is conceivable that increased mucin production could reduce
vaginal pathogen adherence.
Increasing barrier function

Some vaginal pathogens, such as N. gonorrhoeae, are known to invade or
translocate through the epithelium to the subepithelial space (104, 108). Therefore, a
mechanism to reduce bacterial invasion or increase the barrier function of the
epithelium could reduce the progression of vaginal infections. Lactobacilli are known
to inhibit damage caused by E. coli 0157:H7 to the barrier ﬁinction of intestinal
epithelial cells, reducing the severity of the disease (62). During an infection, E. coli
0157:H7 reduces intercellular adhesion between intestinal cells and transmembrane

electrical resistance, however, when pre-treated with Lactobacillus rhamnosus GG,

l7

these effects were prevented (62). Furthermore, several Lactobacillus species have
been found to protect tight junction proteins from damage caused by enteroinvasive E.
coli (133), Shigella dysenteriae (111), and E. coli 01572H7 (62, 132). It is still to be
determined if lactobacilli can protect the vaginal tract against similar damage.
Immunomodulation

Another mechanism probiotic lactobacilli utilize in the gastrointestinal tract to
prevent infectious diseases is immunomodulation of the host (49, 117, 171 , 180).
Lactobacilli can increase the production of secretory immunoglobulin A (sIgA) in the
colon and ileum, which traps pathogens in the mucosa] layer, preventing its
colonization (97-98, 180). Intestinal lactobacilli have also been shown to increase
CD4+ and CD8+ cells in the gastrointestinal tract, which in turn, leads to an increase in
active phagocytes and could reduce the amount of exogenous bacteria available to
colonize the host, thus reducing the risk of pathogen colonization (98, 171, 180).
Furthermore, intestinal lactobacilli can induce anti-inﬂammatory cytokines and reduce
the expression of pro-inﬂammatory cytokines (49, 76, 180), thus decreasing the
damage caused by the host inﬂammatory response to infection.
Interference with pathogen gene expression

Several pathogenic bacteria are known to modify virulence gene expression
upon entering the host (39, 82, 106). One mechanism utilized is autoinducer—mediated
quorum sensing, where the pathogens only express virulence genes when a speciﬁc
concentration, or quorum, of the bacteria has been reached (23, 82, 106). It has
recently been shown that probiotic organisms can inhibit pathogen colonization of the

host by interfering with quorum sensing, thereby inhibiting the expression of virulence

18

factors necessary for persistent colonization. For instance, L. reuteri RC-l4, by an
unknown mechanism, disrupts S. aureus quorum sensing causing a decrease in the
expression of staphylococcal superantigen-like protein and the accessory gene
regulator (agr) locus (85). Similarly, L. acidophilus La-S has been shown to secrete a
molecule that interferes with the quorum sensing regulation of the locus of enterocyte
effacement (LEE) in E. coli 0157:H7 by reducing the concentration of extracellular
autoinducer-2 (106). The reduced expression of the LEE locus correlated with.
decreased colonization of intestinal epithelial cells, demonstrating that lactobacilli can
inhibit pathogen colonization by reducing virulence gene expression (106). This area
of research is relatively new and could lead to exciting new therapeutics such as
autoinducer-Z inhibitors to reduce bacterial pathogen virulence.
CONCLUSION

Lactobacilli are the most prevalent microorganisms in the indigenous
microbiota of the vaginal tract, and the presence of these organisms has been
correlated with a reduced risk of vaginal infections. Therefore the use of these
microbes as probiotics to treat and/or prevent vaginal infections has recently become
an exciting topic of interest. Much has been done to examine which Lactobacillus
species are effective in treating or preventing BV and vulvovaginal candidiasis.
Comparatively little has been done to examine the effects of lactobacilli on the
interactions of sexually transmitted pathogens with the vaginal epithelium. Since
adherence is an essential ﬁrst step in an infection of the vaginal tract, it is important to
discern the effect lactobacilli may have on the two most common bacterial infections,

gonorrhea and Chlamydia. To further the ﬁeld of vaginal probiotics, future study

19

needs to include attention to mechanistic detail as well as differences in Lactobacillus
strains. It is clear that vaginal lactobacilli utilize multiple factors to reduce competition
with exogenous bacteria, and how these factors inhibit pathogens in vivo must be
understood in order to develop effective probiotics. Once it is known what speciﬁc
adhesins are used by each probiotic strain of lactobacilli, better predictions of the
efﬁcacy against particular pathogens can be deduced. Studies of vaginal probiotics
should continue to include the mechanisms of action proposed for intestinal
lactobacilli, as well as considering other possibilities. It is likely that some
mechanisms for reduction of pathogen colonization are niche speciﬁc, necessitating
ﬁirther studies in systems speciﬁc to the female reproductive tract. In vivo experiments
would then be required to determine if the effective mechanisms of inhibition seen in
vitro can translate to clinical purposes.
SIGNIFICANCE OF THE CURRENT WORK

The aim of the current work was to determine the effects of vaginal
Lactobacillus species on the interactions of N gonorrhoeae with host epithelial cells.
N gonorrhoeae (gonococcus), an obligate human pathogen, is the causative agent of
gonorrhea, and infects an estimated 62 million people per year worldwide
(www.who.org). If left untreated, gonorrhea can lead to PID, disseminated
gonococcemia, gonococcal arthritis, sterility or ectopic pregnancy. Asymptomatic
infections are also very common, and it is estimated that 50% of gonorrhea] infections
are undiagnosed (27). Due to similarities with other sexually transmitted infectious
agents, and since it is amenable to laboratory manipulation, N gonorrhoeae can serve

as a model for development of new treatment and prevention methods.

20

The pathogenesis of gonorrhea has been well characterized. In the initial phase
of a gonococcal infection in a female host, N. gonorrhoeae adheres to receptors on
epithelial cells of the genital tract, which include CD46 (42), complement 3 receptor
(42, 44), lutropin receptor (153), glycolipids (126), ﬁbronectin (41, 174), vitronectin
(34), collagens I and IV(41), and several types of proteoglycans (109). Type IV pili
are the primary adhesins (160), although several additional surface structures can
mediate adherence, including lipooligosaccharide (LOS, (43)) porin (P1, (173)) and
opacity proteins (Opa, P11, (95, 182)). The process of adherence is a crucial step in
mounting a successful infection; if the pathogen cannot adhere, it will be washed away
by the ﬂow of vaginal ﬂuid. Once N. gonorrhoeae has adhered to the epithelium,
gonococci invade the host cells and transcytose to the subepithelial space, where they
elicit the acute inﬂammatory response characteristic of gonorrhea. N. gonorrhoeae, as
an incoming pathogen, resists some of the host’s innate defenses by the production of
IgA protease (88), catalase (193), sialyation of lipooligosaccharide (96), phase and
antigenic variation of surface structures (33, 84), and blockage of IgG (65). However,
the effect of lactobacilli on the interactions of gonococci with the host has not been
elucidated.

In 2007, N. gonorrhoeae was classiﬁed as a “superbug” due to the acquisition
of resistance to most antibiotics used to treat gonorrhea. Recent gonococcal isolates
have been shown to be resistant to penicillins, tetracyclines, spectinomycin, and most

recently, ﬂuoroquinolones. This excessive and rapidly developing antibiotic resistance

of gonorrhea has led the CDC to recommend only 3rd generation cephalosporins to

treat all gonococcal infections (www.cdc. gov). However, it is only a matter of time

21

before resistance to these drugs arise. Therefore, it is imperative that new therapies for
the treatment and prevention of gonorrhea be explored. One potential target for
therapeutics is at the critical initial step of adherence to host tissues, and methods to
prevent adherence of gonococci to the host epithelium would be effective in
preventing an infection before it begins. As previously stated, altered states in the
microbiota can increase the risk for transmission of STIs, thus it is likely that the
indigenous microbes of the vagina play a role in preventing pathogen colonization.
Study of the innate mechanisms of the vaginal microbiota have potential to lead to the
development of new therapeutics to treat and/or prevent vaginal infections. In this
dissertation, I ascertained the effect of Lactobacillus jensenii on gonococcal adherence
to and invasion of epithelial cells, and then systematically determined what L. jensenii

factors inhibit pathogen adherence, starting with the mechanisms discussed above.

22

CHAPTER TWO
Spurbeck RR. and CG. Arvidson. (2008) Inhibition of Neisseria gonorrhoeae

Epithelial Cell Interactions by Vaginal Lactobacillus species. Infect. and Immun.
76(7): 3124-30.

23

ABSTRACT

High levels of Lactobacillus, the dominant genus of the healthy human vaginal
microbiota, have been epidemiologically linked to a reduced risk of infection
following exposure to the sexually transmitted pathogen, Neisseria gonorrhoeae. In
this work, a cell culture model of gonococcal infection was adapted to examine the
effects of lactobacilli on gonococcal interactions with endometrial epithelial cells in
vitro. Pre-colonization of epithelial cells with L. jensenii, L. gasseri 33323, or L.
gasseri 9857 reduced gonococcal adherence by nearly 50%. Lactobacilli also inhibited
gonococcal invasion of epithelial cells by more than 60%, which was independent of
the effect on adherence. Furthermore, lactobacilli were able to displace adherent
gonococci from epithelial cells, suggesting that these organisms have potential as a
post-exposure prophylactic. Thus, vaginal lactobacilli have the ability to inhibit
gonococci at two key steps of an infection, which might have a signiﬁcant effect in
determining whether the gonococcus will be able to successfully establish an infection

following exposure in vivo.

24

INTRODUCTION

Neisseria gonorrhoeae (gonococcus), the causative agent of gonorrhea, is an
obligate human pathogen that infects an estimated 62 million people per year
worldwide (www.who.org). Gonorrhea is treatable with antibiotics; however, the
disease is still prevalent due in part to the high frequency of asymptomatic and
subclinical infections in women, providing a signiﬁcant reservoir for transmission.
Untreated gonococcal infections ascend the reproductive tract in up to 45% of infected
women and can lead to disseminated infection, pelvic inﬂammatory disease, ectopic
pregnancy, or sterility. Thus, gonococcal infections place a signiﬁcant health burden
on our society.

In the initial phase of a gonococcal infection in a female host, N. gonorrhoeae
adheres to receptors on epithelial cells of the genital tract. Type IV pili are the primary
adhesins (160), although several additional surface structures can also mediate
adherence, including lipooligosaccharide [LOS; (43)], porin (173), and opacity
proteins [(Opa, P11; (95, 182)]. The process of adherence is a crucial step in mounting
a successful infection; if the pathogen cannot adhere, it could be swept away from the
epithelial cell surface by the ﬂow of vaginal ﬂuid. While the ﬂow of vaginal ﬂuid (on
average 1.55 g/8 h) is not as rapid as that of urine, if the pathogen is non-adherent, it
will eventually be cleared or killed by the antimicrobial agents present in the vaginal
mucus (52, 189).

The indigenous microbiota plays an important role in protecting the host from
colonization by incoming pathogens. Lactobacillus is the predominant genus in the

vaginal (195) and endocervical microbial communities (189) present at concentrations

25

of 107-108 CFU/mL of vaginal ﬂuid in healthy post-menarchal/pre-menopausal

women (135). L. jensenii and L. gasseri are two of the most common species present
as determined by culture-independent techniques (129, 195). Communities dominated
by lactobacilli seemed to be more resilient and less susceptible to altered microbial
states such as bacteria] vaginosis (BV (195)). BV, which is in part deﬁned as a lack of
lactobacilli, has been correlated with an increased risk of sexually transmitted
infections (STIs (187)). This suggests that women who have large numbers of vaginal
lactobacilli are less susceptible to STIs, such as gonorrhea and chlamydia, than women
with BV.

While the epidemiological evidence suggests that the presence of lactobacilli is
correlated with a reduced susceptibility to gonococcal infections, there has been no
direct evidence linking pre-colonization by lactobacilli and effects on N. gonorrhoeae
at any step in the infection process in cell culture or in vivo. Two studies have shown
that lactobacilli and their products, such as hydrogen peroxide, can affect the growth
of N gonorrhoeae in vitro (143, 156); however, these experiments were conducted in
the absence of host cells, so any inhibitory effects lactobacilli might have on
gonococcal interactions with epithelial cells would not have been observed. In vivo,
interactions between lactobacilli and gonococci are likely different due to the presence
of the host. In this work, we hypothesized that lactobacilli affect gonococcal
adherence to and invasion of epithelial cells, and have adapted a tissue culture model

of gonococcal infection to investigate the effect of lactobacilli on these interactions.

26

MATERIALS and METHODS

Bacterial strains and culture conditions. N gonorrhoeae MS11 [P+ Tr (148)] was

grown at 37°C in a humidiﬁed 5% C02 environment (+ COZ) on GC agar

(Accumedia, Lansing, MI) with supplements (75) and VCNT inhibitor (Becton,

Dickinson, and Company, Sparks, MD). L. jensenii [ATCC 25258, Hzoz+ (156),
human vaginal origin] and L. gasseri [ATCC 33323, H202+ (156), human unknown

tissue of origin, and ATCC 9857, human vaginal origin] were grown at 37°C + C02

on MRS agar (Difco, Sparks, MD).

Hydrogen peroxide production. All Lactobacillus strains were tested for hydrogen
peroxide production utilizing the plate assay developed by Eschenbach et al (46) and
modiﬁed by Felten et al (48). Brieﬂy, MRS agar plates were made containing 0.25
mg/mL 3,3’,5,5’-tetramethylbenzidine (TMB; Sigma-Aldrich, St. Louis, MO), and
0.0] mg/mL horseradish peroxidase (HRP; Sigma-Aldrich). Each strain was streaked
for isolated colonies on MRS-TMB plates and on MRS-no TMB control plates that

contained 0.0] mg/mL HRP and 225 pL EtOH (diluent for TMB). The plates were

incubated for 48 hours at 37°C + C02. Hydrogen peroxide produced by the

lactobacilli reacts with HRP to produce Oz, which then reacts with TMB to form a

dark blue precipitate. Colonies of hydrogen peroxide-producing strains are stained
dark blue.
The production of hydrogen peroxide by L. jensenii, L. gasseri 33323, L.

gasseri 9857, and N gonorrhoeae individually and in co-culture was also measured in

27

cell culture medium using an assay described previously (40, 130-131) with some
modiﬁcations. Brieﬂy, lactobacilli and gonococci were swabbed from fresh plate
cultures into Dulbecco’s Modiﬁed Eagle Medium without phenol red (DMEM,

Invitrogen, Carlsbad, CA) supplemented with 5% fetal calf serum (FCS, Invitrogen),

GC supplement II [12.4 uM Fe(NO3)3], and 110 mM sodium pyruvate. Bacterial
concentrations were determined spectrophotometerically, and diluted to 107 CF U/mL

for lactobacilli and 106 CFU/mL for gonococci. Serial dilutions were plated on

appropriate selective media (lactobacilli, MRS; N gonorrhoeae, GCV) to quantify

inocula. Bacterial cultures were incubated at 37°C + C02 for 3 h, and were then

centrifuged for 10 min at 3428.5 x g (3000 RPM) and ﬁltered through a 0.22 pm

(pore-size) membrane. 0.75 mL of peroxide assay buffer [5.0 mM KZHPO4, 1.0 mM

KH2P04, 140 mM NaCl, 0.5 mM glucose, with phenol red (ﬁnal concentration 0.46

mM) and horseradish peroxidase (ﬁnal concentration 0.046 U/mL) added immediately
prior to the assay] was placed into a borosilicate glass tube to which 0.25 mL of the
sample supematants were added. The samples were incubated for 30 min at 37°C, and
then NaOH (ﬁnal concentration 0.004 N) was added to stop the reaction. The
absorbance at 610 nm was determined spectrophotometrically, and the amount of

hydrogen peroxide produced under each condition was determined by comparing to 3
H202 standard curve generated in the same assay.
Cell culture. The human endometrial epithelial cell line Hec—l-B (ATCC HTB-113)

was grown in DMEM (Invitrogen) supplemented with 5% FCS at 37°C + C02. Cell

28

culture assays were canied out in 24-well cell culture plates with Hec-l-B cells grown
to 50-90% conﬂuency. Adherence and invasion assays were as described previously
(12, 39) with modiﬁcations as follows.

Adherence assay 1. Lactobacilli were swabbed from fresh plate cultures into DMEM
(without phenol red) supplemented with 5% FCS, GC supplement II, and 110 mM
sodium pyruvate (DMEM/5% FCS/GC supp 11). Bacteria] concentrations were

determined spectrophotometerically, and diluted in DMEM/5% FCS/GC supp II to a

concentration of 4 x 107 CFU/mL. Serial dilutions of this inoculum were plated on

MRS agar. Hec-l-B cells were pre-colonized with 250 [1L of the lactobacilli inoculum
for a multiplicity of infection (MOI) of 100, and the infected epithelial cells were
incubated at 37°C + C02 for 3 h. The epithelial cells were then washed 5 times with
phosphate buffered saline (PBS) to remove non-adherent lactobacilli. A mock
infection was conducted as a control in which the epithelial cells were incubated with
250 ILL DMEM/5% F CS/GC supp II for 3 h and then washed 5 times with PBS prior

to the introduction of gonococci. Gonococci were swabbed from fresh plate cultures

into DMEM/5% FCS/GC supp II. Bacterial concentrations were determined

. . . . . . 6
spectrophotometencally, and dlluted in this medlum to a concentratron of 4 x 10

CFU/mL. Serial dilutions of this inoculum were plated on GCV agar. Hec-l-B cells
pre-colonized with adherent lactobacilli, along with the control, were then infected

with 250 pL of the N gonorrhoeae inoculum (MOI of 10) and the epithelial cells were

incubated at 37°C + C02 for 3 h. Following incubation, the medium was removed and

the infected epithelial cells were washed ﬁve times with sterile PBS to remove non-

29

adherent bacteria. The epithelial cells and cell-associated bacteria were lifted with 1
mL PBS containing 5 mM EDTA (PBS/EDTA). Serial dilutions were plated on GCV
agar to determine the cell-associated CFU. The CFU of gonococci adherent to
epithelial cells pre-colonized with lactobacilli was then normalized to the cell-
associated CFU from the control, which lacked lactobacilli.

Adherence assay 2. L. jensenii was swabbed from ﬁ'esh culture plates into

DMEM/5% FCS/GC supp II. Bacterial concentrations were determined and diluted in

this medium to 4 x 106 CFU/mL and 4 x 107 CFU/mL. Serial dilutions of each
inoculum were plated on MRS agar. One row of 6 wells containing 105 Hec-l-B cells
was pre-colonized with 250 [LL of the 4 x 106 CFU/mL lactobacilli inoculum (MOI of
10) and a second row of wells containing 105 Hec-l-B cells was pre-colonized with

250 pL of the 4 x 107 CFU/mL lactobacilli inoculum (MOI of 100). A third row was

used as a mock infection in which the media on the epithelial cells was replaced with

250 pL fresh DMEM/5% FCS/GC supp H. The infected epithelial cells were then
incubated for 1 h at 37°C + C02. Gonococci were swabbed from fresh plate cultures
into DMEM/5% FCS/GC supp II. Bacterial concentrations were determined
spectrophotometerically, and diluted to 4 x 106 CFU/mL. Serial dilutions of this

inoculum were plated on GCV agar. Hec-l-B cells pre-colonized with adherent and
non-adherent lactobacilli, along with the control containing no lactobacilli, were then

infected with 250 [LL of the N gonorrhoeae inoculum (MOI of 10) and the infected

epithelial cells were incubated at 37°C + C02 for 3 h. Following incubation, the

30

infected Hec-l-B cells were split into two sets, one set was to calculate the total CFU
of bacteria, and the other set was to calculate the cell-associated (adherent) CFU. For
the ﬁrst set, the supernatant (0.5 mL) from each well was placed in a sterile tube. The
Hec-l-B cells and cell-associated bacteria were then lifted with 500 11L of PBS/EDTA
and added to the sterile tube containing the unattached bacteria. Serial dilutions were
then plated on appropriate selective media to determine the total CFU/well. For the
second set, the supernatant was removed and the epithelial cells were washed ﬁve
times with sterile PBS. The epithelial cells and cell-associated bacteria were then lifted
with 1 mL PBS/EDTA and placed in a sterile tube. Serial dilutions were plated on
selective media to obtain the cell associated (adherent) CFU/well. The adherence
frequency was calculated by dividing the cell-associated CFU/well by the total

CF U/well. The adherence frequencies were then normalized to the control, which
contained no lactobacilli.

Invasion assays. Pre-colonization of the Hec-l—B cells with L. jensenii was conducted

as described above for adherence assay 2, with the lactobacilli inoculum at a

concentration of 4 x 107 CFU/mL. After incubation for lb, 250 ILL of N. gonorrhoeae

(4 x 106 CFU/mL) was added to the Hec-l-B cells pro-colonized with lactobacilli and

incubated for 7 h. Following incubation, wells were divided into three sets. For the
ﬁrst set (total CFU), the supernatant (0.5 mL) was removed and placed in a sterile
tube. The Hec-l-B cells and cell-associated bacteria were lifted with 500 [LL
PBS/EDTA, and this suspension was added to the tube of supernatant. Serial dilutions
were plated on selective media to determine the total CFU/well. For the second set

(cell-associated CFU), the epithelial cells were washed 5 times with PBS to remove

31

non-adherent bacteria, and the Hec-l -B cells with adherent bacteria were then lifted
with 1 mL PBS/EDTA. Serial dilutions were plated on selective media to determine
the cell-associated CF U/well. The third set of wells (intracellular CF U) was washed

ﬁve times with PBS to remove non—adherent bacteria, treated with 75 mg/L

gentamicin (Gm) in DMEM and then incubated at 37°C + C02 for 1 h to kill

extracellular bacteria. The Hec-l-B cells (with intracellular bacteria) were then

washed 5 times with PBS and lifted with 1 mL PBS/EDTA. Serial dilutions were

plated on selective media to determine the GmR CFU/well. The invasion frequency is

deﬁned as the ratio of GmR CFU/well to cell associated CFU/well.

Microscopy. Hec—l-B cells were seeded into 6 well cell culture plates containing
sterilized glass coverslips (25 mm, Fisher Scientiﬁc). Once the epithelial cells reached
conﬂuence, an adherence assay as described above for adherence assay 2 was canied
out with some modiﬁcations. Brieﬂy, at 3 h post infection (p.i.), non-adherent bacteria
in the supematant were removed and placed on glass microscope slides and Gram
stained. The bacteria were then visualized by light microscopy (magniﬁcation 100X)
and photographed using a digital still camera (Cyber-shot DSC-P50, SONY). The
epithelial cells with adherent bacteria were washed ﬁve times with PBS and then ﬁxed
with 100% methanol for ﬁve minutes. The coverslips were then removed from the
well, placed on microscope slides, and Gram stained. The bacteria adherent to the
Hec-l -B cells were visualized by light microscopy (magniﬁcation 100X) and
photographed using a digital still camera. The number of adherent lactobacilli was

determined for thirty consecutive Hec-l-B cells in each replicate. Three replicate

32

experiments, canied out in duplicate were conducted and the adherent bacterial counts
were then averaged.

Pro-conditioned medium. Hec-l-B cells were infected with L. jensenii at an MOI of

10 or 100, and incubated for l h at 37°C + C02. The supernatant from the infected

epithelial cells was removed, centrifuged at 16,000 x g (14,000 RPM) for 30 sec to
pellet the bacteria, and then ﬁltered through a 0.22 pm membrane. These pre-
conditioned media were then inoculated with 106 CFU gonococci and used to infect

fresh Hec-l -B cells (M01 10). The infected Hec-l-B cells were incubated at 37°C +
C02 and at 3 h p.i., the epithelial cells with adherent bacteria were washed, lifted, and

plated as described above for adherence assays.
Statistical analysis. All data was analyzed by unpaired student t-test. P-values of less
than 0.05 were considered statistically signiﬁcant, and are indicated by a * in each of

the graphs.

33

RESULTS
Conditions for N. gonorrhoeae and Lactobacillus co-culture. L. jensenii and L.
gasseri are two of the most prevalent Lactobacillus species recovered from the vaginal
tract of healthy women (9, 22, 25, 194), and therefore were chosen for this study.
Previous research showed that L. jensenii 25258 and L. gasseri 33323 had the
potential to kill or inhibit the growth of N gonorrhoeae in vitro (156). Thus, it was
necessary to identify conditions in which both species could grow in order to study the
effects on gonococcal interactions with epithelial cells. Each bacterial strain was

grown separately in DMEM/5% FCS/GC supp H and dilutions were plated at intervals

to determine viable CFU/mL. Next, N gonorrhoeae (starting at 106 CFU/mL) was co-

cultured with L. jensenii, L. gasseri 33323, or L. gasseri 9857, at a ratio of 1:10
(gonococci: lactobacilli) and plated for viable CF U at intervals to determine the effect
of co-culture on grth of each bacterial strain. As shown in Figure 2.1, the grth of
N. gonorrhoeae in the presence of L. jensenii was similar to that observed when grown
separately. Similar results were observed when N gonorrhoeae was grown in the
presence of L. gasseri (data not shown). Each Lactobacillus strain also survived
similarly in the presence and absence of gonococci (data not shown). These results
establish that gonococci and lactobacilli survive and grow in co-culture under these

conditions.

34

 

 

 

 

 

1.E+08
i
1.E+07 n
1.E+06 4
1.E+05 r , T I
O 2 4 6 8

Time in hours

Figure 2.1 Growth of N. gonorrhoeae in the presence or absence of L. jensenii. N
gonorrhoeae (106 CFU/mL inoculum) was grown in the (El) presence or (O) absence

of L. jensenii (107 CFU/mL inoculum). Serial dilutions were plated on selective media
at 2 h intervals to quantify CF U. Data are the averages of three independent
experiments. Error bars indicate standard deviations.

Hydrogen peroxide production in cell culture medium. Two of the strains used in
this work, L. jensenii (ATCC 25258) and L. gasserz’ 33323, were previously reported
to kill N gonorrhoeae in an agar overlay assay due to the production of hydrogen
peroxide (61 , 156). Under the conditions of our assays, the lactobacilli did not inhibit
the growth of gonococci (Figure 2.1). Therefore, the Lactobacillus strains used in our
study were tested for hydrogen peroxide production by an MRS-TMB plate assay (46,
48). Using this method, in which hydrogen peroxide production is visualized by the

production of a dark blue precipitate, L. jensenii and L. gasserz’ 33323 were conﬁrmed

to be hydrogen peroxide producers (solid dark blue), while L. gasserz' 9857 produced

35

modest levels of hydrogen peroxide (blue after a prolonged incubation). Next,
hydrogen peroxide production by lactobacilli grown in our cell culture conditions in
both the presence and absence of gonococci was examined. When incubated for 3 h in
DMEM/5% FCS/GC supp 11, L. jensenii, L. gasseri 33323, L. gasseri 9857, and N
gonorrhoeae all produced undetectable levels of hydrogen peroxide (less than 0.1
mM). When incubated in co-culture with N. gonorrhoeae in DMEM/5% FCS/GC
supp II for 3 h, L. jensenii, L. gasseri 33323, and L. gasseri 9857 again produced
undetectable levels of hydrogen peroxide. The lack of detectable levels of hydrogen
peroxide is consistent with our observations that when co-cultured in DMEM/ 5%
FCS/GC supp II lactobacilli do not kill gonococci, and with the ﬁndings of other
researchers that at neutral pH, hydrogen peroxide producing lactobacilli do not inhibit
the growth of gonococci ( 193).

Adherence of N. gonorrhoeae to epithelial cells pre-colonized with lactobacilli.
Epidemiological evidence suggests there is a reduced risk of developing an infection
following exposure to an STI, such as gonorrhea, when lactobacilli are present in large
numbers (187). Therefore, the effect of pre-colonization of epithelial cells with
lactobacilli on the adherence of N gonorrhoeae was examined. Hec-l-B cells, a
human endometrial epithelial cell line, were infected with lactobacilli at an MOI of
100, or DMEM/5% FCS/GC supp II (control). At 3 h p.i., non-adherent lactobacilli
were removed by washing. Epithelial cells pre-colonized with lactobacilli and the
control epithelial cells were then infected with gonococci at an MOI of 10. At 3 h p.i.,
the Hec-l-B cells were again washed to remove non-adherent bacteria, lifted, and

plated on selective media. The adherence of each Lactobacillus strain to the epithelial

36

cells after exposure to N gonorrhoeae was as follows: L. jensenii — 0.73%, L. gasseri
33323 — 0.61%, and L. gasseri 9857 — 0.93% of the inoculum. The average CFU/well
of N gonorrhoeae adherent to Hec-l-B cells pre-colonized with each Lactobacillus
strain was compared to the average CFU/well of gonococci adherent in the control,
and these results are shown in Figure 2.2. When Hec-l -B cells were pre-colonized
with L. jensenii, gonococcal adherence dropped to 63.0 i 21.9 % (p = 0.043) of that
determined for the control. Pro-colonization with L. gasserz' 33323 reduced gonococcal
adherence to 56. 6 i 13.8 % (p = 0.003), and pre-colonization with L. gasseri 9857
reduced gonococcal adherence to 63.0 i 24.6 % (p = 0.059). Thus, in the presence of
two of the adherent strains of Lactobacillus tested, the adherence of N gonorrhoeae

was signiﬁcantly decreased as compared to the control.

 

 

    

 

relative adherence (%)

 

L. gasseri L. gasseri L. jensenii no lactobacilli
9857 33323
Pre-colonizing strain

Figure 2.2. Adherence of N gonorrhoeae to Hec-I-B cells pro-colonized with
lactobacilli. Epithelial cells were incubated with lactobacilli for 3 h. Non-adherent
lactobacilli were removed prior to the addition of gonococci, and the Hec-l -B cells
were incubated an additional 3 h. Relative adherence is the percentage of adherent
gonococci in wells containing lactobacilli divided by the percentage of adherent
gonococci in the control. The data are averages of three independent experiments
performed in duplicate. Error bars indicate standard deviations. * statistically
signiﬁcant.

37

Since the human vaginal tract contains lactobacilli in the ﬂuid and mucosal
layer that are non-adherent to the epithelia, and because adherence is a dynamic
process with bacteria attaching and then detaching from the host cells (162), we next
developed an approach to examine the adherence of N gonorrhoeae in a system more

closely mimicking this natural state. In the experiments described above, non-adherent

lactobacilli were washed off prior to the addition of gonococci, leaving fewer than 105

CFU of lactobacilli adherent to epithelial cells in the well. It has been reported that

there are 107-108 CFU/mL lactobacilli of vaginal ﬂuid of healthy pro-menopausal

women (135). Therefore, gonococcal adherence to Hec-l -B cells pre-colonized with
lactobacilli was next examined without removing the non-adherent lactobacilli.
Hec-l-B cells were inoculated with L. jensenii at an MOI of 10 or 100, and
incubated for l h. Hec-l-B cells were incubated with sterile DMEM/5% FCS/GC supp
II for 1 h as a control. Gonococci were then added at an MOI of 10, and the infected
epithelial cells were incubated an additional 3 h. The relative gonococcal adherence
frequencies are shown graphically in Figure 2.3. At a 1:1 ratio of N gonorrhoeae to L.
jensenii, gonococcal adherence was reduced to 73.9 i 22.1 % (p = 0.11) of the no
lactobacilli control, whereas at the ratio 1:10 N gonorrhoeae to L. jensenii the
adherence of N gonorrhoeae was reduced to 52.5 i 14.2 % (p = 0.004). Thus,
inhibition of gonococcal adherence increased with the number of lactobacilli present.

The adherent CFU per well of L. jensenii increased proportionally with the

number of lactobacilli in the inoculum. At an MOI of 10, 4 x 104 CFU/ well L.

jensenii were adherent to the epithelial cells (an adherence frequency of ~1 % cell-

38

5
associated lactobacilli/total lactobacilli), while at an MOI of 100, 3 x 10 CFU/ well L.

jensenii were adherent (an adherence frequency of ~1 %). When the ﬁrst adherence
assay (Figure 2.2) and the second adherence assay (Figure 2.3) are compared, the CF U
of adherent lactobacilli per well inversely correlates with the gonococcal adherence
frequency, suggesting that as the number of adherent lactobacilli increases, the number

of adherent gonococci decreases correspondingly.

.5

O

O
1

 

(3,/,0)

0')
O
1

relative adherence

 

 

 

1:0 1 :1 1 :10
ratio (20nococcizlactobacilli)

Figure 2.3. Adherence of N gonorrhoeae to Hec-l-B cells in the presence of adherent
and non-adherent lactobacilli. Relative adherence is the percentage of adherent
gonococci in wells containing lactobacilli divided by the percentage of adherent
gonococci in the control. The data are the averages of three independent experiments
performed in triplicate. Ratios of gonococci to lactobacilli were: 1:0, 10 gonococcizno
lactobacilli; 1:1, 106 gonococci:lO lactobacilli; 1:10, 106 gonococci:107 lactobacilli.
At 3 h p.i. the CFU of N gonorrhoeae doubled, and the total CFU of gonococci in the
control was not signiﬁcantly different from that of the Lactobacillus treated samples
[p-values: 0.324 (1:1) and 0.536 (1:10)]. Error bars indicate standard deviations.

Student’s t-testp-values were 0.11 for 1:] and 0.004 for 1:10. * statistically
signiﬁcant.

Gonococcal invasion of epithelial cells pre-colonized with lactobacilli. After

adherence has been attained in a gonococcal infection, a subpopulation of these

39

It? ii

bacteria invades the epithelial cells to transcytose to the subepithelial space. To
determine whether lactobacilli could affect gonococcal invasion, a gentamicin (Gm)
protection assay was used in which Hec-l-B cells were ﬁrst pre-colonized for 1 h with
L. jensenii at M01 100 prior to the addition of gonococci (M01 10). As a control, Hec-

l-B cells were incubated with fresh DMEM/5% FCS/GC supp II for 1 h prior to
inoculation with gonococci. The frequency of invasion was determined as GmR
CFU/well per cell-associated CFU/well at 7 h p.i., thereby segregating the frequency

of adherence from the frequency of invasion. To check for GmR gonococci that are

not intracellular, 4 x 106 CFU/mL gonococci were incubated for 1 h with Gm at the

same concentration used in the invasion assay and then plated dilutions from 101—] 0
. R . . 6

for v1able cell. counts. There were less than one Gm gonococcr 1n 4 x 10 CFU/mL,

. . . . . R
venfyrng that the gonococcal strain used 1n these experiments was not Gm . The

results of the invasion experiments show that gonococcal invasion was reduced to
39.46 i 10.47% (p = 0.001) in the presence of L. jensenii compared to invasion in its
absence. Interestingly, at 7 h post (gonococcal) infection when invasion was
measured, the adherence frequency for the gonococci in co-culture with L. jensenii
was not signiﬁcantly different from that of the control (data not shown). Thus,
although gonococci eventually recover from the initial inhibition of adherence, these

results indicate that lactobacilli affect both steps of gonococcal interactions with Hec-

l-B cells.

40

idhsri
IIOI t
remon
ﬁrst-2n}
hour. \
beaten;-
‘1: nun

Lit COP.

Displacement of adherent gonococci by lactobacilli. To determine whether
lactobacilli had the potential to affect gonococcal interactions with host epithelial cells
post-exposure, we assessed the ability of lactobacilli to displace gonococci already
adherent to epithelial cells. Hec-l -B cells were infected with N gonorrhoeae at an
MOI of 10, as for an adherence assay. At 3 h p.i., non-adherent gonococci were
removed, and DMEM containing L. jensenii, L. gasseri 33323 (MOI of 100), or no
bacteria (control) was added. Infected epithelial cells were incubated for an additional
hour. Non-adherent bacteria were then removed and Hec-l-B cells with adherent
bacteria were lifted and plated on selective media. After addition of L. gasseri 33323,
the number of adherent gonococci was reduced by 28.7% (p = 0.019) as compared to
the control (Figure 2.4). Displacement by L. jensenii was less consistent (p = 0.174),
although treatment with either Lactobacillus species tended to reduce the number of

adherent gonococci.

41

120

 

 

 

relative adherence (%)

 

 

—

No lactobaci

11 L. gasseri L. jensenii
33323

Figure 2.4. Displacement of N gonorrhoeae adherent to Hec-l-B cells by lactobacilli.
N. gonorrhoeae adherent to Hec-l-B cells after 3 h were then infected with L. gasseri
33323 or L. jensenii at an MOI of 100 for l h. Adherence data are presented as a
percentage of adhered gonococci in the well divided by that of the control and are the
averages of four or more experiments performed in triplicate. Error bars indicate
standard deviations. Student’s t-test p-values were 0.019 (L. gasseri 33323) and 0.174
(L. jensenii). *, statistically signiﬁcant.

Spatial arrangement of lactobacilli and gonococci adherent to epithelial cells. To
visualize the spatial arrangement of gonococci and lactobacilli on the epithelial cell
monolayer during a co-infection, adherence assays were conducted in which the
epithelial cells were adherent to a glass coverslip, allowing for microscopic analysis.
Adherent lactobacilli were observed either individually or in short chains (1-3 bacilli),
at an adherence frequency of 0.72 lactobacilli per epithelial cell, consistent with the
adherence frequencies found using the plate count method. Gonococci were
predominantly in microcolonies when adherent to epithelial cells in the presence or

absence of lactobacilli, thus counting of individual gonococci adherent to Hec-l-B

cells was not possible. In all ﬁelds observed, lactobacilli and gonococci were found

42

randomly dispersed on the epithelial cells and generally not in contact with each other

(Figure 2.5A and B).

 

Figure 2.5. Gram stains from an adherence assay performed as for the experiment

shown in Fig. 2.2 with Hec-l -B cells pre-colonized with '10 CF U L. jensenii. Panel A
depicts gonococci and lactobacilli adherent to epithelial cells. Panel B depicts
gonococci adherent to epithelial cells in the absence of lactobacilli. Panel C depicts the
arrangement of gonococci and lactobacilli in the supernatant. In all panels the white
arrow points to lactobacilli and the black arrow points to gonococci (Olympus model
CHBS microscope, 100x magniﬁcation).

43

Lactobacilli and gonococci do not co-aggregate in co-culture. Previous reports have
suggested that lactobacilli produce aggregation factors that could allow lactobacilli to
aggregate with other bacterial species, thus sequestering pathogens and decreasing
adherence (19, 79). To determine if co-aggregation of lactobacilli with gonococci
could be the mechanism of inhibition observed in the adherence assays, the
supematants of three separate adherence assays (performed in duplicate) were Gram
stained and visualized by light microscopy. In all ﬁelds observed, lactobacilli and
gonococci never formed co-aggregates and a representative ﬁeld of such an
experiment is shown in Figure 2.5C.

Lactobacillus pre-conditioned medium does not inhibit gonococcal adherence.
Recently, Lactobacillus acidophilus was found to secrete a soluble compound that can
reduce expression of virulence factors in Escherichia coli OlS7:H7 necessary for
colonization of intestinal epithelial cells (106). To examine the possibility that
lactobacilli secrete a soluble compound that inhibits gonococcal adherence to
epithelial cells, gonococcal adherence assays were performed using Lactobacillus pre-
conditioned media. L. jensem'i at an MOI of 10 (PC-10) and an MOI of 100 (PC-100)
were incubated in cell culture medium with Hec-l-B cells for l h and then removed by
centrifugation and ﬁltration. These media were then inoculated with gonococci at an
MOI of 10 for an adherence assay using fresh Hec-l-B cells and compared to a control
of gonococci inoculated in ﬁ'esh media. The results of these experiments show that the
gonococcal adherence frequency was not statistically different from that of the control

in either PC-lO (p = 0.631), or PC-100 (p =0.884, Figure 2.6).

44

60.00%

 

: 50.00%
40.00%
30.00%

20.00%

adherence frequen

10.00%

0.00%

 

MS11 Control p10 p100

Figure 2.6. Adherence of gonococci in medium pre-conditioned with L. jensenii and
Hec-l—B cells. Percent adherence is the ratio of adherent gonococci to the total number
of gonococci for each condition. Number of lactobacilli used to condition the medium:
Fresh medium; no lactobacilli, fresh DMEM/5% FCS/GC supp II; P-lO, 106 CFU
lactobacilli2105 Hec-l-B cells; P-100, 107 CFU lactobacilli110 Hec-l-B cells. Data
are the average of four independent experiments performed in triplicate. Error bars
indicate the standard deviation. P-10: p = 0.631, and P-100: p =0.884.

45

DISCUSSION

To successﬁilly infect the female genital tract, gonococci must ﬁrst adhere to
the epithelia to avoid clearance by the continual ﬂow of vaginal ﬂuid. A reduction in
the ability to adhere would signiﬁcantly affect the outcome of an infection following
exposure. Although lactobacilli have been implicated in protecting the female vaginal
tract from exogenous pathogens, research on the direct effect of lactobacilli on STIs
has been lacking. Therefore, we developed a co-culture model of gonococcal infection
that includes lactobacilli to examine their effect on gonococcal interactions with
epithelial cells.

In our in vitro model of gonococcal infection, pre-colonization with lactobacilli
reduced gonococcal adherence to epithelial cells by 40-50% (Figure 2.2 and 2.3). This
inhibition of adherence was dependent on the ratio of gonococci to adherent
lactobacilli, with the inhibition becoming more pronounced in the presence of
increased numbers of lactobacilli (Figure 2.3). In a gonococcal infection, following
adherence to epithelial cells, gonococci will invade these epithelial cells and
transcytose to the subepithelial space. Our results show that gonococcal invasion of
epithelial cells pro-colonized with L. jensenii was decreased by 60%. This inhibition
of invasion was distinct from the inhibition of adherence, demonstrating that even the
gonococci that overcome the inhibition of adherence are negatively affected in
invasiveness by the presence of lactobacilli. The inhibition of adherence, along with
the inhibition of invasion, suggests that colonization with lactobacilli could reduce the

risk for infection by N gonorrhoeae by potentially decreasing the amount of

46

L
In: I
i

pry 1
Laid);
5

Oil 511

911‘.
411i: ‘Irl

i in!
.: 'l‘
~L'L\i

"‘~;,l

gonococci that adhere and invade the mucosa] epithelia, thus providing a greater

opportunity for the innate host defenses to prevent the establishment of an infection.
Due to the nature of N gonorrhoeae as an obligate human pathogen, testing

the biological relevance of the 50% reduction in gonococcal adherence caused by

lactobacilli cannot be tested in vivo. However, semen samples from males infected

with gonorrhea on average contain 7 x 106 CFU per ejaculate (60), similar to the
gonococcal inoculum used in these studies (6 x 106 CFU/well). Cervical aspirates of

females with gonorrhea have been reported to contain 104-106 CFU/mL (91, 190). In

our studies the average CFU of gonococci adherent to epithelial cells in the controls

falls within this range (1 x 106 CFU/well), again lending support to the relevance of

our experimental conditions. Thus, the inhibition of gonococcal adherence to and
invasion of epithelial cells pre-colonized with lactobacilli may explain the
epidemiological observations in which women with bacterial vaginosis (reduced
numbers of lactobacilli) have an increased susceptibility for gonorrhea following
exposure (187).

Lactobacilli are now being promoted as probiotic bacteria that have broad
health beneﬁts. Lactobacilli have the ability to displace adherent pathogens associated
with urinary tract infections (137), as well as Gardnerella vaginalis bioﬁlms
associated with BV (145), suggesting potential as a treatment for these diseases. While
it is unlikely that lactobacilli alone could be used as a treatment for gonorrhea, it is
conceivable that lactobacilli could reduce the likelihood of gonococci establishing an

infection when directly administered to the vaginal tract shortly after exposure. Reid et

47

al. (137) suggests a displacement of 20-50% would be clinically relevant for
pathogens infecting the vaginal tract. Our results indicate that gonococcal
displacement by L. gasserz' 33323 is within this range at 28.7% (Fig 2.4). Thus, the
ability of lactobacilli to displace adherent gonococci in a cell culture model of
infection suggests that lactobacilli have potential as a post-exposure prophylactic for
gonococcal infection.

In this study, L. jensenii and L. gasseri strains were chosen because these
species are two of the most prevalent species recovered from healthy women and had
been previously studied for effects on N gonorrhoeae in the absence of epithelial cells
(9, 22, 25, 194). Each strain inhibited the adherence of N gonorrhoeae to epithelial
cells, but adhered poorly to epithelial cells (average adherence frequency ~1%). When
epithelial cells are pro-colonized lactobacilli at an MOI of 100, there is on average one
adherent Lactobacillus per epithelial cell. Thus, it is unlikely that the mechanism for
exclusion of gonococci is direct competition for receptors on the epithelial cells.
However, the number of adherent lactobacilli is inversely correlated with the
gonococcal adherence frequency: as the number of adherent lactobacilli increases the
gonococcal adherence frequency decreases. This suggests that it is the adherent
lactobacilli that play an important role in the observed inhibition of gonococcal
adherence. Our results also show that the mechanism of inhibition of adherence is not
due to co-aggregation of lactobacilli with gonococci (Figure 2.5C), growth inhibition
(Figure 2.1), or the secretion of a stable soluble compound into the medium by either
the epithelial cells or lactobacilli before inoculation with gonococci (Figure 2.6).

Further study will be necessary to identify the mechanisms of inhibition, which will be

48

fundamental in the development of Lactobacillus-based treatments or preventatives for
gonococcal infection.

In summary, we have adapted a tissue culture model of gonococcal infection to
include a major constituent of the human vaginal tract, lactobacilli. This approach will
be essential in elucidating the role of the indigenous microbiota in the early, and
therefore preventable steps of a gonococcal infection. The observation that lactobacilli
inhibit gonococcal adherence and invasion, and could even displace adherent
gonococci is intriguing; however, the mechanisms by which these events occur are not
yet clear. While co-aggregation, the secretion of soluble inhibitory molecules, and
growth inhibition can be ruled out as possible mechanisms, there are still several
hypotheses to test. Possible mechanisms might include a contact inducible change in
the epithelial cells due to the adherence of lactobacilli that would reduce susceptibility
to gonococci or the secretion of a short-lived inhibitory factor produced by lactobacilli
directly in response to incoming gonococci. This model will be indispensable in ﬁrture
studies to examine the molecular basis for the observed inhibition of gonococcal

interactions with epithelial cells.

49

CHAPTER THREE
Spurbeck RR. and C.G. Arvidson. (2010) Lactobacillusjensenii Surface Associated

Proteins Inhibit Neisseria gonorrhoeae Adherence to Epithelial Cells. Infect. and
Immun. Epub: April 12, 2010.

50

. P
bit

(16,.
.4-..

.f‘

311

ABSTRACT

High numbers of lactobacilli in the vaginal tract have been correlated with a
decreased risk of infection by the sexually transmitted pathogen, Neisseria
gonorrhoeae. We have previously shown that Lactobacillusjensenii, one of the most
prevalent microorganisms in the healthy human vaginal tract, can inhibit gonococcal
adherence to epithelial cells in culture. Here we examined the role of the epithelial
cells and the components of L. jensenii involved in the inhibition of gonococcal
adherence. L. jensenii inhibited the adherence of gonococci to glutaraldehyde-ﬁxed
epithelial cells similarly to live epithelial cells, suggesting that the epithelial cells do
not need to be metabolically active for this inhibition to occur. In addition, methanol-
ﬁxed L. jensenii inhibited gonococcal adherence to live epithelial cells, indicating L.
jensenii uses a constitutive component to inhibit gonococcal interactions with
epithelial cells. Proteinase K treatment of the methanol-ﬁxed lactobacilli abolished the
inhibitory effect, suggesting that the inhibitory component contains protein. Released
surface components (RSC) isolated from L. jensenii were found to contain at least two
inhibitory components, both of which are protease-sensitive. Using anion exchange
and size exclusion chromatography an inhibitory protein was isolated, which has
signiﬁcant similarity to the enzyme enolase. A recombinant His6-tagged version of
this protein was subsequently produced and shown to inhibit gonococcal adherence to

epithelial cells in a dose-dependent manner.

51

INTRODUCTION

Neisseria gonorrhoeae (gonococcus) is an obligate human pathogen that causes
the sexually-transmitted infection, gonorrhea. Gonorrhea is one of the most commonly
reported infectious diseases, second only to chlamydia, with 336,742 cases reported in
2008 in the United States according to the Centers for Disease Control and Prevention.
Gonorrhea is readily treated with antibiotics; however, N gonorrhoeae has developed
resistance to each antibiotic used historically, including penicillins, tetracyclines,
spectinomycin, and most recently, ﬂuoroquinolones. As of 2007, the CDC
recommends only one class of antibiotics to treat all types of gonorrhea, the
cephalosporins (29). Furthermore, women infected with gonorrhea often have
asymptomatic infections, providing a signiﬁcant reservoir for transmission.
Undiagnosed, and therefore untreated gonococcal infections can lead to permanent
damage of the female reproductive system resulting in infertility or ectopic pregnancy.
Gonorrhea has also been reported to increase susceptibility to HIV infections (150).
Therefore, a method to reduce the number of gonococcal infections that does not
involve antibiotics would be beneﬁcial for public health.

Adherence to the host epithelia is the ﬁrst and most critical step in a gonococcal
infection, thus, this step is a target of interest for the development of new therapeutics.
By inhibiting gonococcal adherence to the epithelial cells, the pathogen could be
washed away by the ﬂow of vagina] or menstrual ﬂuid, or killed by the antimicrobial
agents found in the vaginal mucus, rendering the pathogen unable to establish an
infection (189). One potential source for an anti-adherence treatment is probiotic

bacteria. Probiotics are deﬁned as “live microorganisms which, when administered in

52

adequate amounts, confer a health beneﬁt on the host” (47). The bacterial genus most
studied for its probiotic properties is Lactobacillus. Lactobacilli naturally inhabit both
the gastrointestinal tract and the female reproductive tract of healthy humans. In the
female reproductive tract, lactobacilli make up the majority of the indigenous vaginal
and endocervical microbiota (189, 195). Lactobacilli have been shown to play a role in
protecting women from infection by incoming pathogens including HIV (149), and
epidemiological evidence suggests that women with high numbers of vaginal
lactobacilli have a reduced susceptibility to gonorrhea and chlamydia following
exposure (187). Clearly, lactobacilli are a key component of the human defense
against colonization by sexually transmitted pathogens.

Lactobacilli utilize several mechanisms to prevent the colonization of incoming
pathogens. These include direct killing of the organisms by hydrogen peroxide,
bacteriocins, or lowering the pH of the vaginal tract to ~4 by the production of lactic
acid (11, 14, 21). Lactobacilli can inhibit adherence by the production of
biosurfactants (137, 177, 183), receptor competition (20, 30), or co-aggregation with
the pathogen, allowing the pathogen to be swept away by the host’s bodily ﬂuids (19,
79). Lactobacilli can also inhibit pathogen colonization by causing the host cells to
become more resistant to adherence (62, 76, 81, 133) or by suppressing the expression
of virulence factors in the pathogen (106).

Lactobacillusjensenii, one of the most prevalent species of lactobacilli
indigenous to the human vaginal tract, has been shown to inhibit gonococcal
adherence to and invasion of epithelial cells (154). To date, the mechanism used by L.

jensenii to inhibit gonococcal adherence is unknown. However, we have shown in our

53

I--.

 

previous work that L. jensenii does not directly inhibit the growth of gonococci or co-
aggregate with the gonococci either in the medium or at the cell surface. Furthermore,
the hydrogen peroxide produced by L. jensenii is also not effecting the gonococcal
adherence to the epithelial cells (154). In the current work, we test the hypothesis that
L. jensenii utilizes a surface component to inhibit gonococcal adherence and examine
the processes and components of the epithelial cell, gonococci, and lactobacilli that

play a role in this inhibition.

54

MATERIALS AND METHODS

+
Bacterial strains and culture conditions. N gonorrhoeae M81] (P , Tr) (148) and N.
gonorrhoeae MS] 1-307 (ApiIEI::erm; ApilE2, P-, Tr, (107)) were grown at 37°C in a

humidiﬁed 5% C02 environment (+ C02) on GC agar (Acumedia, Lansing, MI) with

supplements (75) and VCNT inhibitor (GCV, Becton, Dickinson, and Company,
Sparks, MD). L. jensenii (ATCC 25258, H202+) was grown at 37°C + C02 on MRS
agar (Becton, Dickinson). Bacillus subtilis strain RB 247 (169), was grown at 37°C on
LB agar (Acumedia, Lansing, MI). Escherichia coli strains DH50L and BL21>\DE3 was

grown on LB agar at 37°C, with kanamycin at 100 mg/l (KnlOO) added for pET24a

and derivatives.
Cell culture. The human endometrial epithelial cell line Hec-l-B (ATCC HTB-113)
was grown in Dulbecco's Modiﬁed Eagle Medium (High glucose with L-glutamate,

DMEM, Invitrogen, Carlsbad, CA) supplemented with 5% fetal calf serum (FCS,

lnvitrogen) at 37°C + C02. Cell culture assays were carried out in 24-well cell culture

plates with Hec-l-B cells grown to 50-90% conﬂuency.

Glutaraldehyde fixation of Hec—l-B cells. Hec-l-B cells were incubated with 1 ml of
glutaraldehyde (2.5% v/v) for 30 min at 37°C + C02. The glutaraldehyde was then
removed and the ﬁxed cells were washed 5 times with phosphate-buffered saline

(PBS), immediately prior to use in adherence assays (153). Gonococcal adherence

assays on glutaraldehyde ﬁxed cells were canied out as shown below with the

55

exception that ﬁxed cells were lifted with saponin (1% w/v in GCB) instead of PBS +
5 mM EDTA.

Methanol and proteinase K treatment of L. jensenii. An L. jensenii inoculum was
divided into three 2 ml aliquots containing 4 x 107 CFU/ml. Aliquot 1 (live
lactobacilli) was held at 37°C. Aliquots 2 and 3 were treated with an equal volume of
ice cold methanol (MeOH) for 10 minutes, and then centrifuged at 14,000 x g for 3
min to pellet the bacteria. The supernatant was removed and replaced with 200 u]
DMEM (Aliquot 2, MeOH-treated) or 160 pl DMEM and 40 pl proteinase K (ProK,
20 mg/ml, Aliquot 3, ProK-treated). All three aliquots were then incubated for 2 h at
37°C. Aliquots 2 and 3 were boiled for 3 min and then centrifuged at 14,000 x g for 3
min to pellet the bacteria. The supematants were removed and the bacteria were
resuspended in 2 ml fresh DMEM. Microscopic visualization of Gram stained samples

conﬁrmed that MeOH treatment and ProK treatment left the lactobacilli intact.

Adherence assays. Adherence assays were performed as previously described (154).

Brieﬂy, wells containing 105 Hec-l-B cells were inoculated with L. jensenii at an MOI
of 100 (1 x 107 CFU/ml), B. subtilis at an MOI of 100 (1 x 107 CFU/ml), or fresh

DMEM supplemented with 5% F CS, 12.4 11M Fe(NO3)3 (GC supplement 1]), and 110

mM sodium pyruvate (DMEM/5% F CS/GC supp II) as a mock infection. Following

incubation at 37°C + C02 for l h, the epithelial cells were infected with N.
gonorrhoeae at an MOI of 10 (1 x106 CFU/ml). Following 3 h of incubation at 37°C +

C02, the infected Hec-l-B cells were divided into two sets: the ﬁrst set was to

56

determine the total number of bacteria in the well, and the second set to determine the
cell-associated (adherent) bacteria. For the ﬁrst set, the supernatant from each well
(0.5 ml) was placed in a sterile tube. The Hec-l-B cells, with the cell-associated
bacteria, were then lifted with 0.5 ml of PBS + 5 mM EDTA (PBS/EDTA) and added
to the supernatant. Serial dilutions were plated on appropriate selective media to
quantify the total CFU/well. For the second set, the supernatant was removed and the
epithelial cells were washed ﬁve times with sterile PBS. The epithelial cells with the
cell-associated bacteria were then lifted with 1 m1 PBS/EDTA and transferred to a
sterile tube. Serial dilutions were plated on selective media to quantify the cell
associated (adherent) CF U/well. The adherence frequency was calculated by dividing
the cell-associated CFU/well by the total CFU/well. Where indicated, the adherence
frequencies were normalized to the mock infection, which contained no lactobacilli.
Production of released surface components (RSC). A modiﬁcation of the method of
Reid et al. (137) was used to isolate released surface-associated proteins and

biosurfactants from L. jensenii. Brieﬂy, a 1 1 culture of L. jensenii in MRS broth was

grown to an A600 of 1.6-1.8, and the bacteria were harvested by centrifugation (10,000

x g for 10 min at 7°C). The bacterial cell pellet was washed twice with sterile H20.

The cell pellet was then resuspended in 134 ml PBS and gently stirred for 2 h at room
temperature. The Lactobacillus suspension was then centriﬁrged for 20 min at 3,000 x
g and the supernatant was ﬁltered (pore size 0.22 pm) to remove remaining bacteria.
This cell-free preparation was concentrated to ~10 ml by Amicon ultraconcentration
using a 10 kDa MWCO ﬁlter membrane. Protein concentration was determined using

the Bradford method (Biorad, Richmond, CA). The products obtained were then

57

assayed for inhibition of gonococcal adherence to Hec-l-B cells by pretreating the
epithelial cells for 3 h prior to infection with N. gonorrhoeae.

Fibronectin blocking assay. To assess the effect of ﬁbronectin (F n) on the inhibitory
activity of the RSC, 0.42 mg/ml RSC was incubated with 50 pt] of soluble Fn (0 — 10
uM) for l h at room temperature. This mixture was used to treat Hec-l-B cell
monolayers (105 cells/well) for 3 h. The suspension was then removed and the cells
washed once with PBS. An adherence assay with N gonorrhoeae was carried out

using these cells as described above.

Trypan blue exclusion assay. Wells containing 105 Hec-l-B cells were inoculated as

for adherence assays above with either sterile DMEM, L. jensenii, N. gonorrhoeae,

RSC, or a combination of N. gonorrhoeae and L. jensenii or N gonorrhoeae and RSC.

Following a 3 h incubation at 37°C + C02, the cells were washed 5 times with PBS,

and then trypan blue (0.2% in PBS) was added. After 3 min, the trypan blue was
removed and the cells washed once with PBS. The number of live and dead cells were
counted visually and compared between all conditions.

Collapsed drop analysis. To assess biosurfactant activity, the RSC was assayed by a
collapsed drop analysis as described by Walencka et al. (183). 2 a] of parafﬁn oil was
allowed to equilibrate over night in the wells of 96-well microtiter plate lid. 5 ,ul of
each sample was dropped onto the oil covered surface and observed for l h. The
samples were scored as positive if the drop ﬂattened (had biosurfactant activity) or
negative if the drop remained beaded on the surface of the oil (no biosurfactant

activity). Water was used as a negative control.

58

Anion

. ‘

hf'l'illn'

LIL LAN]
»

{HO Ill 1

WIT
.11.]1

..l .0
513..st

Prolii

I‘ll F
I. {1‘

TVD'QJ‘
' “$15:

loin: I

Anion exchange column chromatography. After dialysis against H20, the RSC was

brought to 50 mM Tris (pH 7.6) and separated by anion exchange column
chromatography (Econo-Pac Q, Biorad) using a 0-1 M NaCl gradient in a buffer of 50
mM Tris, 5 mM EDTA. Column ﬂow through and a 25 ml 0 M NaCl wash were
collected before the gradient was run. Twenty-ﬁve 2 ml fractions were collected
during the gradient and ﬁve 2 ml fractions were collected from a terminal 1 M NaCl
wash. The protein in fractions was quantiﬁed using the Bradford method and samples
were run on a 12% SDS-PAGE gel. The fractions were then dialyzed into PBS before
being assayed for inhibitory activity against gonococcal adherence to Hec-l-B cells.
Inhibitory fractions were also assayed for biosurfactant activity using the collapsed
drop analysis described above.

Protein identiﬁcation by tandem mass spectrometry. Proteins were separated on a
12% SDS PAGE gel and stained with 1% Coomassie Blue. The protein band of
interest was excised from the gel and sent to the Michigan Proteome Consortium for
analysis. Brieﬂy, the sample was digested with trypsin, concentrated, and then spotted
on a 192-well MALDI target and allowed to dry at room temperature. The sample was
analyzed on a 4800 Proteomics Analyzer (TOF/TOF, Applied Biosystems) and the
mass spectra were acquired in Reﬂector Positive Ion mode for a peptide range of 800-
3500 kDa. MS spectra were summed from 2,000 laser shots from an Nd-YAG laser
operating at 355 nm and 200 Hz. Internal calibration was performed using a minimum
of 3 trypsin autolysis peaks. Database searching was performed using Applied
Biosystems GPS Explorer v. 3.6, with Mascot v. 2.1. Spectra were subjected to a 7-

point Gaussian smooth prior to peak picking. For the PMF search, a maximum of 65

59

ﬂ:‘i’_)’
., has.
1

arm
:Lkluol
‘ﬁ‘l a
5.3...

"f [C [71

ill“

peaks were submitted with a S/N of 30, and a maximum peak density of 50 peaks per
200 Da. Data were searched against all Lactobacillus genome sequences in NCBI
(www.mcbi.nlm.nih.gov/).

Cloning, expression and purification of recombinant L. jensenii enolase. The
sequence of the enolase genes from L. gasseri 33323, L. johnsonii NC533, and all
sequenced L. jensenii strains was compared using ClustalW2. Based on these
sequences, primers were designed to amplify the putative enolase ORF from L.
jensenii 25258. Initial attempts to amplify the entire open reading ﬁame (ORF) were
not successful. When conserved internal primers were used in combination with full
length primers, it was determined that while the 5’ end of the gene was highly
conserved, the 3‘ end was not. To determine the sequence of the 3’ end, a primer
speciﬁc to the gene located 3’ of the enolase gene in other lactobacilli (secG) was
designed and used to PCR amplify the region. The DNA sequence of the product was
determined and used to design a new primer for the 3’ end of the enolase gene,
EnoRev, which incorporated a NotI site. The 5’ primer, Enond, contained a Ndel site
which included the ATG start codon. The ORF was PCR ampliﬁed, puriﬁed using a
PCR puriﬁcation kit (Qiagen, Germantown, MD) and then digested with Notl and
Ndel (New England Biolabs, Ipswich, MA). The insert was ligated into similarly
digested pET24a (Novagen, EMD4 Biosciences, San Diego, CA), and transformed
into E. coli DH50r. Transformants were identiﬁed and the DNA sequence of the entire
insert was determined. An isolate, pET24a-em, was then transformed into E. coli

BL21hDE3 for expression.

60

High level production of the His6-enolase was achieved using the method

described by Studier (159). Brieﬂy, 50 mls of an overnight culture of

BL21hDE3pET24a-eno grown in LB Kn100 at 37°C was used to inoculate 2 l of ZY-

5052 autoinducing media supplemented with Kn100 and incubated overnight at 30°C

with shaking at 250 rpm. The bacteria were harvested by centrifugation and

resuspended in 50 mM NaHzPO4 pH 7.0, 300 mM NaCl. The cells were then lysed by

shearing using a M-l 10P processor (Micro-ﬂuidics Corp) set at 20k psi. The insoluble

material was removed by centrifugation and the crude lysate was loaded onto a

NiNTA resin column (BD Biosciences) that was equilibrated with 50 mM NaHzPO4

pH 7.0, 300 mM NaCl. The column was washed with 50 mM NaHzPO4 pH 7.0, 300

mM NaCl, 10 mM imidazole and then the bound protein was eluted using 50 mM

NaHzPO4 pH 7.0, 300 mM NaCl, 250 mM imidazole. The protein was further puriﬁed

by anion exchange chromatography as described above. Prior to use in cell culture
assays, the protein was dialyzed against PBS and the protein concentration was
determined using the Bradford method. Puriﬁed protein was then applied to epithelial
cells at various concentrations for 3 h before infection with N gonorrhoeae.
Statistical analysis. All data was analyzed by unpaired Student’s t—test. A p-value
below 0.05 was considered statistically signiﬁcant and is indicated by asterisk (*) in

the graphs.

61

RESULTS
Lactobacillus inhibition of gonococcal adherence to fixed epithelial cells. In our
experimental system, there are three players that could be involved in the
Lactobacillus inhibition of gonococcal adherence to Hec-l-B cells: the epithelial cells,
the lactobacilli, and the gonococci. Since intestinal Lactobacillus strains have been
shown to induce epithelial cells to become more resistant to pathogen adherence (62,
76, 81, 133) we utilized glutaraldehyde-ﬁxed Hec-l-B cells to determine the role of
the epithelial cells in the inhibitory interaction. L. jensenii (M01 100) was used to pre-
colonize both glutaraldehyde-ﬁxed and live epithelial cells for 1 h prior to gonococcal
infection as described in the Materials and Methods. The cells were then lifted with
saponin and plated onto selective media (MRS for lactobacilli, GCV for gonococci).
On live epithelial cells, when L. jensenii is present, 8.9 i 3.9% of the total gonococci
present adhered, compared to 20.2 i 8.0% adherence in the absence of lactobacilli.
This is a signiﬁcant reduction in the adherence of gonococci (Figure 3.1A, p < 0.001).
This inhibition is speciﬁc to lactobacilli, as pre-treatrnent of the epithelial cells with
Bacillus subtilis (MOI = 100) had no effect on the adherence frequency of gonococci
when compared to the control (90.7 1- 13.5%, p = 0.117).

On ﬁxed epithelial cells, gonococci adhere at a two-fold lower frequency than to
live epithelial cells. Nonetheless, the presence of L. jensenii is able to reduce
gonococcal adherence on ﬁxed cells from 11.2 i 2.8% to 4.4 i 2.0% (Figure 3.18, p =
0.036). The inhibition of gonococcal adherence to ﬁxed epithelial cells was not

signiﬁcantly different from the level of inhibition seen for live epithelial cells (p =

62

0.433), suggesting that Lactobacillus-mediated inhibition of gonococcal adherence to

epithelial cells does not require the epithelial cells to be metabolically active.

% adherence

CD

% adherence

0/o adherence

 

 

0.5

0.0

 

MS11 alone MS11 +
L. jensenii

 

MS11 alone MS11 +
L. jensenii

 

   

MS11-307 MS11-307 +
alone L. jensenii

63

Figure 3.1. The effect of L.
jensenii on gonococcal adherence.
A. Adherence of N gonorrhoeae
to live Hec-l-B cells in the
presence or absence of L. jensenii.
The data are averages of 20
independent experiments
performed in triplicate. Error bars
indicate standard error. Student’s
t-test p-value < 0.001. B.
Adherence of N gonorrhoeae to
ﬁxed Hec-l-B cells in the
presence or absence of L. jensenii.
The data are averages of four
independent experiments
performed in triplicate. Error bars
indicate standard error. Student’s
t-test p-value = 0.036. C.
Adherence of non-piliated
(MS11-307) gonococci in the
presence and absence of L.
jensenii. The data are averages of
six independent experiments
performed in triplicate. Error bars
indicate standard error. Student’s
t-test p-value = 0.008.

Pilus-mediated gonococcal adherence is not targeted by L. jensenii. The type IV
pilus is the primary adhesin for gonococci and has been shown to be necessary for
infection (75, 161). We hypothesized that lactobacilli inhibit the adherence of N

gonorrhoeae by speciﬁcally targeting pilus-mediated adherence. The effect of L.

jensenii on the non-piliated gonococcal strain MSl 1-307 (P-) was examined to test this

hypothesis. Non—piliated gonococci adhered to Hec-l-B cells at a frequency of 1.2 i

0.4% (Figure 3.1C), which is signiﬁcantly lower than that of the P+ gonococcal strain

(20.2 i 8.0%, Figure 3.1A). Again, when the epithelial cells were pre-colonized with
L. jensenii, the P- gonococcal adherence frequency was reduced signiﬁcantly to 0.43 i

0.13% (p = 0.008, Figure 3.1C). After normalizing the data to the appropriate controls,
gonococcal adherence frequencies were compared between the L. jensenii treated non-
piliated and pilated gonococcal strains. No statistically signiﬁcant difference was
found (p = 0.334), which suggests that L. jensenii does not inhibit gonococcal
adherence by speciﬁcally targeting type IV pilus-mediated adherence to epithelial
cells.

Surface components of L. jensenii inhibit gonococcal adherence. Since we
determined that the epithelial cells do not need to be metabolically active for L.
jensenii to inhibit gonococcal adherence to host cells (Figure 3.1B), we next focused
on determining the components of lactobacilli that are involved in this inhibition by
comparing the effect of treating the lactobacilli with MeOH and/or ProK prior to
infection with gonococci. As shown in Figure 2, MeOH-treated lactobacilli inhibited

gonococcal adherence to epithelial cells 50.2 i- 25.9% (p = 0.001), similar to inhibition

64

observed with live lactobacilli 63.0 i 14.9% (p < 0.001). This result suggests that the
inhibitory factor is a surface component of L. jensenii. Additionally, ProK treatment
abolished Lactobacillus—mediated inhibition of gonococcal adherence (96.9i 12.7%, p
= 0.533). Taken together with the MeOH results, this suggests that a constitutive
protein on the exposed surface of L. jensenii is involved in the inhibition of

gonococcal adherence to epithelial cells.

 

     

None -. Live ‘— MeOH ProK

relative adherence (%)

Figure 3.2. Adherence of N gonorrhoeae in the presence of live, MeOH-ﬁxed, and
ProK—treated L. jensenii. Relative adherence is the percentage of adherent gonococci
in the presence of lactobacilli divided by the percentage of adherent gonococci from
the corresponding control. None, no lactobacilli; Live, untreated L. jensenii; MeOH,
methanol-treated L. jensenii; ProK, proteinase K-treated L. jensenii. The data are
averages of three or more independent experiments performed in triplicate. Error bars
indicate standard error. Student’s t-test p-values were <0.001 (live), 0.001 (MeOH)
and 0.533 (ProK).

Released surface components (RSC) from L. jensenii inhibit gonococcal
adherence to epithelial cells in a dose-dependent manner. Lactobacillus species
have been shown to produce several compounds that can inhibit pathogen adherence,
including hydrogen peroxide, bacteriocins, and biosurfactants (11, 128, 136, 183). A
probiotic Lactobacillus species, L. helveticus, has been shown to utilize constitutive

surface layer proteins to inhibit Escherichia coli OlS7:H7 adherence to epithelial cells

Johnson-Henry (63). Since we previously reported that L. jensenii inhibits gonococcal

65

adherence by a mechanism that does not involve the secretion of inhibitory substances
(154), and we have shown that surface proteins are involved in the inhibition of
gonococcal adherence, we next determined if surface components released from L.
jensenii could inhibit gonococcal adherence to epithelial cells. Using a method
published by Reid et al. (137) to isolate biosurfactants from lactobacilli, we isolated
surface components released from lactobacilli and the biosurfactant activity of each
preparation was tested by collapsed drop analysis. Each drop of RSC collapsed on the
oily surface, demonstrating that RSC has biosurfactant activity, while the water
controls remained beaded up on the oily surface.

The RSC was then tested for inhibitory activity against gonococcal adherence
to epithelial cells by treating Hec-l-B cells for 3 h with varying concentrations of RSC
prior to infection with gonococci in an adherence assay. The results of this experiment
showed that gonococcal adherence was reduced by nearly 5-fold when the epithelial
cells were treated with 1.7 mg/ml RSC (Figure 3.3). When the epithelial cells were
treated with 0.42 mg/ml RSC, the gonococcal adherence frequency was inhibited less
than 2-fold, and 0.17 mg/ml RSC had no effect on gonococcal adherence. The
adherence of gonococci to the treated epithelial cells decreased as the amount of RSC
increased, indicating that gonococcal adherence to epithelial cells is inhibited in a

dose-dependent manner.

66

 

 

relative adherence (%)

0.42 0.17

1.7
control L. jensenii mg/ml mg/ml mg/ml

Figure 3.3. RSC inhibition of gonococcal adherence is dose dependent. Hec-l-B cells
were incubated with either 1.7 mg/ml, 0.42 mg/ml, or 0.17 mg/ml RSC for 3 h before
the addition of gonococci. Gonococcal adherence to epithelial cells was reduced to
43.9 i 21.4% (L. jensenii, p= 0.01), 21.5 i 14.4% (1.7 mg/ml RSC,p= 0.001), 65.1 :t
1.7% (0.42 mg/ml RSC,p= <0.001), and 95.6 i 27.9% (0.17 mg/ml RSC,p= 0.81) of
the untreated control. The data are averages of three independent experiments
performed in triplicate. Error bars indicate stande error.

To rule out the possibility that the apparent reduction of gonococcal adherence
to epithelial cells was caused by cell death or detachment caused by treatment with
either RSC or bacteria, cell number and viability was assessed by trypan blue
exclusion. There was no signiﬁcant difference in the total number of nonviable cells
(those that took up the trypan blue dye) after each treatment when compared to Hec-l -
B cells with fresh medium (85.7 :t 22.2 blue cells per well), RSC (78.0 i 11.3 blue
cells per well, p = 0.704), lactobacilli (99.0 i 16.0 blue cells per well, p = 0.552),
gonococci (118.3 :1: 35.1 blue cells per well, p = 0.368), lactobacilli + gonococci (73.7
i 6.9 blue cells per well,p = 0.535), or RSC + gonococci (89.3 :t 12.2 blue cells per
well, p = 0.857). Therefore, the effect on gonococcal adherence is not due cell death.

When the number of live cells present in the cell monolayer (those that excluded the

trypan blue dye) was compared after each treatment there was again no signiﬁcant

difference in cell counts when compared to Hec-l -B cells in fresh medium (7.83 x 104

67

i 2.89 x 104 cells/well), N. gonorrhoeae (9.0 x 104 i 4.0 x 104 cells/well, p = 0.779),
. .. 4 4 4
L. jensenii (8.67 x 10 i 1.89 x 10 cells/well,p = 0.786), RSC (9.67 x 10 i 1.44 x
4 . . . 4 4
10 cells/well, p = 0.533), lactobacrllr + gonococcr (8.67 x 10 i 1.89 x 10 cells/well,

p = 0.786), or RSC + gonococci (9.33 x 104 i 8.89 x 104 cells/well,p = 0.589). This

suggests that any effects on gonococcal adherence frequency observed is due to an
effect on gonococcal adherence to the epithelial cells, not an effect on the adherence of
the epithelial cells to the cell culture plate.
Inhibition of gonococcal adherence to epithelial cells remains after removal of
RSC. The RSC contains biosurfactant activity, which by deﬁnition are surface active
molecules (139). Therefore, we hypothesized that a surface component of L. jensenii
in the RSC inhibits gonococcal adherence by interacting with the surface of the
epithelium. To test whether a component of the RSC interacts with the epithelial cell
surface to inhibit gonococcal adherence, RSC, at 1.7 mg/ml, was incubated with Hec-
l-B cells and then either washed away before infection with gonococci or left in the
well during the infection. The relative adherence frequencies were then compared.
There was no signiﬁcant difference in the inhibition of gonococcal adherence when
the RSC was removed (Figure 3.4, p =0.729). This suggests that the inhibitory

component(s) of the RSC remain associated with the epithelial cells.

68

 

 

relative adherence (%)

control L. jensenii Unwashed Washed
RSC RS

Figure 3.4. Washing of RSC-treated cells prior to inoculation with gonococci does not
remove inhibitory activity. RSC was incubated with Hec-l-B cells for 3 h prior to
infection with gonococci. Gonococcal adherence to epithelial cells was reduced to
53.3 i 17.2% (L. jenseni,ip= 0.01), 31.2 :h 15.0% (no wash, p= 0.001), 34.9 i 18.1%
(washed, p= 0.003). The data are averages of three independent experiments
performed in triplicate. Error bars indicate standard error. Student’s t-test p-value was
0.729 when washed RSC sample was compared to unwashed.
Characterization of the inhibitory components of the RSC. To determine if the
inhibitory component(s) of the RSC was protein, an aliquot of RSC (1.7 mg/ml) was
treated with ProK and incubated at 37°C for 2 h. A second aliquot was incubated at
37°C for 2 h as a negative control. Both aliquots were heat treated to inactivate the
protease and centrifuged to remove any debris. These samples were then used to pre-
treat Hec-l-B cells for 3 h before infection with gonococci (M01 10). At 3 h post—
gonococcal infection, the cells were lifted and plated to quantify adherence. Compared
to the control, gonococcal adherence to RSC-treated cells was reduced 2-fold (Figure
3 .5, p = 0.004). However, when the epithelial cells were pre—treated with ProK-RSC
the gonococcal adherence frequency was similar to the adherence of the untreated

Control (p = 0.755). These results suggest that the inhibitory activity of the RSC is due

1’0 the presence of a protein(s).

69

 

100
80
60
40
2O . . .

 

 

relative adherence (%)
0

control L. jensenii RSC ProK-RSC

Figure 3.5. Proteinase K treatment of RSC abolishes inhibitory effect on N
gonorrhoeae adherence to Hec-l-B cells. RSC was incubated for 2 h at 37°C with
proteinase K at 20 mg/ml. Gonococcal adherence to epithelial cells was reduced to
68.2 i 2.0% (L.jensenii, p<0.001), 50.1i 14.6% (RSC,p= 0.004), 96.3 :t 18.9%
(ProK treated RSC, p= 0.755). The data are averages of three independent experiments
performed in triplicate. Error bars indicate standard error.

Puriﬁcation of the RSC proteins. We next fractionated the inhibitory protein

components of the RSC by anion exchange chromatography. The column was eluted
with a gradient of 0-1 M NaCl and the proteins detected by A260. A representative

elution proﬁle is shown in Figure 2.6A. Peak 1 (Pl) eluted at 20 mM NaCl, and Peak
2 (P2) eluted at 1 M NaCl. SDS-PAGE analysis of samples from P1 and P2 showed
there were several proteins in each pool (Figure 3.68). The most abundant protein in
P1 and the most abundant protein in P2, indicated in Fig. 3.6B, were excised and
analyzed by tandem mass spectrometry. The protein from P1 had 2 peptides of
observed mass over 1000 with 100% sequence identity to glyceraldehyde-3-phosphate
dehydrogenase (GAPDH) from Lactobacillusjohnsonii NCC 533. The protein from
P2 had 8 peptides of observed mass over 1000 with 100% sequence identity to enolase
from Lactobacillusjohnsonii NCC 533 and L. gasseri 33323. P] and P2 were dialyzed
iI'rto PBS and then assayed for inhibitory activity. P1 (0.42 mg/ml) inhibited

gonococcal adherence to epithelial cells by 2-fold. Similarly, P2 (0.42 mg/ml) also

70

“that
it ' W

l i

In

avg;

A, (a?

A? ”Jo? .

 

inhibited gonococcal adherence by 2-fold (Figure 3.6C). This suggests there are at

least two separable inhibitory components in the RSC.

Figure 3.6. Fractionation of RSC

 

 

 

A 2 by anion exchange
1 i chromatography. A. Elution proﬁle
0.8 of samples detected by UV
8 0-6 absorbance (7.=260 nm) as the
g? 0-4 protein eluted from the column.
< 0'2 Peaks 1 and 2 are marked P1 and
0 T ' r P2. B. SDS PAGE analysis ofPl
64 74 84 :me 22:") 114 124 134 and P2. Long arrow points to the
B band with similarity to the enzyme

enolase, and the short arrow points
to the band with similarity to the
enzyme GAPDH, as determined by
MS/MS. C. Gonococcal adherence
to epithelial cells pretreated with
P1 and P2. Treatment of epithelial
cells inhibited gonococcal
adherence to 52.9 i 23.9% (L.
jensenii, p=0.027), 51.2 i 14.0%
(P1,p=0.004), and 48.8 i 14.8%
(P2, p=0.004). The data are the
averages of three independent
experiments performed in
duplicate. Error bars indicate
standard error.

 

_u
o
o

 

G
O

A
0

relative adherence (%) O
B 8

 

0

control L. jensenii P1 P2

To further separate the proteins in P2, the pool of protein was dialyzed against 50
HIM Tris (pH 7.6), concentrated by ultraﬁltration, and then separated by gel ﬁltration

SiZe exclusion chromatography (Sephacryl S-200HR). Samples from fractions

71

containing protein were visualized by SDS-PAGE and assayed for biosurfactant
activity. Fraction 49 was the only fraction that contained biosurfactant activity, and
when visualized by SDS-PAGE, the sample contained a single band of protein.
However, when assayed for inhibition of gonococcal adherence, this sample did not
contain inhibitory activity. This suggests that the biosurfactant activity of the RSC is
distinct and separable from the inhibitory activity.

Fraction 36, another fraction that contained a single protein band was tested for
inhibitory activity. This fraction also did not inhibit gonococcal adherence to epithelial
cells. Two other fractions assayed for inhibition of gonococcal adherence were
fractions 24 and 29. These fractions were chosen because when visualized by SDS-
PAGE, both fractions appeared to contain the putative enolase seen in Fig. 3.6B.
Fraction 24 contained the putative enolase with an additional protein of 65.8 kdal, and
reduced gonococcal adherence to epithelial cells to 26.5% of the no RSC control.
Fraction 29 contained the putative enolase, the 65.8 kdal protein of fraction 24, and an
additional protein of 30.6 kdal. This fraction reduced gonococcal adherence to 40.4%
of the no RSC control. The level of inhibition appeared to correlate with the amount of
the putative enolase in the fraction as seen by SDS-PAGE, with fraction 24 containing
more of the putative enolase than fraction 29. The band correlating to the putative
enolase was excised from an SDS-PAGE gel and veriﬁed by tandem mass

spectrometry as having homology to enolase (7 peptides with m/z over 1000, with
1 00% sequence identity to enolase from Lactobacillus gasseri ATCC 33323).
S oluble fibronectin blocks inhibitory activity of the RSC. Gonococci are known to

bind to ﬁbronectin (Fn) during the invasion of epithelial cells (174). Since enolase of

72

0'; er .
rails
h.’ r it]
blocli
lo :65
IO :11:
Infect
m.

...OC

h
b

other Lactobacillus species has been shown to bind F n (7, 26, 70) and a putative
enolase is the main component of one of the inhibitory fractions of the RSC, we
hypothesized that the RSC might reduce gonococcal adherence to epithelial cells by
blocking gonococcal binding to extracellular matrix components such as ﬁbronectin.
To test this, RSC was treated with increasing amounts of soluble F 11 prior to addition
to the epithelial cells. After 3 h, the RSC-Fn solution was removed and the cells were
infected with N gonorrhoeae. When no Fn was added, 0.42 mg/ml RSC reduced
gonococcal adherence to 40.4 i 17.0%. When the RSC was treated with either 2.5, 5,
or 7.5 MM Fn, gonococcal adherence was reduced, but as the amount of F n increased,
the inhibitory activity of the RSC was reduced, to 54.9 i 8.0%, 70.1 :1: 8.7%, and
74.6i 9.9%, respectively (Figure 3.7). When the RSC was treated with 10 11M soluble
Fn, the gonococcal adherence frequency was 95.3 i 7.5% relative to the control,
essentially no different from gonococcal adherence in the absence of RSC. This is
consistent with the hypothesis that the mechanism of RSC inhibition of gonococcal

adherence is due to an inhibition of gonococcal interactions with Fn on the cells.

73

 

100
80 '
6O
40 ‘

20

relative adherence (%)

 

0
RSC -
anM) 0 0 2.5 5.0 7.5 10.0

Figure 3.7. Gonococcal adherence to cells treated with RSC that had been pre-treated
with increasing amounts of soluble ﬁbronectin. Gonococcal adherence frequencies are
reported as relative to a no RSC—no Fn control and Student’s t-tests were used to
compare each Fn treatment with no Fn treatment. Adherence to epithelial cells was
reduced to 40.4 i 17.0% (0.42 RSC), 54.9 i 8.0% (RSC—2.5 p.M Fn,p= 0.374), 70.1 a:
8.7% (RSC—5.0 uM Fn, p= 0.121), 74.6 i 9.9% (RSC-7.5 MM Fn,p= 0.091) and 95.3
i 7.5% (RSC-10 11M Fn, p= 0.019). The data are averages of three independent
experiments performed in duplicate. Error bars indicate standard error.

Effect of His6-Eno on gonococcal adherence to Hec-l-B cells. One of the two

proteins of the RSC that co-puriﬁed with the inhibitory activity had signiﬁcant
similarity to an enolase from Lactobacillus gasseri. To determine if this protein is the

inhibitory component, the enolase gene from L. jensenii ATCC 25258 was
recombinantly expressed as a His6-tagged protein in E. coli and puriﬁed by
chromatography (Figure 3.8A). The puriﬁed protein was then used at various
concentrations to treat epithelial cells prior to gonococcal infection. His6-Eno
inhibited gonococcal adherence to epithelial cells at all concentrations tested by 50-
80% (Figure 3.88). As the concentration of Hi56-Eno increased, the adherence of

gonococci to epithelial cells decreased, again indicating a dose-response.

74

kDa

200
122

86

 

 

o
Him-Eno: 0
(mg/ml)

0.14 0.27 0.42 0.56 0.83

75

Figure 3.8. Puriﬁcation of His6-
Eno and its effect on gonococcal
adherence to Hec-l-B cells. A.
SDS-PAGE gel analysis of H156-
Eno. 18 pg puriﬁed Hisb-Eno (lane
1), MW standards (lane 2), 0.72
ug puriﬁed His6-Eno (lane 3),
uninduced E. coli BL21)\DE3
pET24a-em (lane 4), induced E.
coli BL21>\DE3 pET24a-em (lane
5). Arrow indicates His6-Eno. B.
Gonococcal adherence to cells pre-
treated with His6-Eno. Gonococcal
adherence in the absence of His6-
Eno: 20.1 i 4.0%; in the presence
ofO. 14 mg/ml His6-Eno: 10.4 :I:
2.8%, p= 0.028; 0.27 mg/ml His6-
Eno: 6.5 :t 1.8; p= 0.013; 0.42
myml HiS5-Eno: 6.9 i 0.7%, p=
0.004; 0.56 mg/ml His6-Eno: 6.7 i
1.4%, p= 0.013; 0.83 mg/ml HiS5-
Eno: 3.8i 1.0%, p= 0.006. The
data are averages of three
independent experiments
performed in duplicate. Error bars
indicate standard error.

DISCUSSION

Adherence is a critical ﬁrst step in the infection of a new host for many
pathogens, including N gonorrhoeae. As an incoming pathogen, gonococci have to
overcome the host’s defenses and compete with the indigenous microbiota to
effectively colonize the endocervical epithelia. Lactobacillusjensenii, one of the most
commonly isolated bacterial species from the healthy human vaginal tract, has been
shown to inhibit gonococcal adherence to and invasion of epithelial cells in a co-
culture model of infection (154). This inhibition of gonococcal adherence was not
caused by any secreted factor, such as hydrogen peroxide, due to co-aggregation
between the gonococci and the lactobacilli, or competition for receptors. This ability
to inhibit gonococcal adherence, without direct killing of the pathogen, might be
speciﬁc to Lactobacillus species, as we see no inhibition of adherence when cells are
pre-treated with another Gram positive bacterium, Bacillus subtilis. This ﬁnding
correlates with research which showed vaginal Lactobacillus isolates inhibit
gonococcal adherence to the cell line ME180 better than intestinal isolates (179),
suggesting that this inhibitory mechanism may have developed due to competition in
the vagina] ecosystem. In this work, we have examined the components of lactobacilli,
gonococci, and the epithelial cells involved in an effort to elucidate the mechanism(s)
of this inhibition.

In our in vitro model of gonococcal infection, pre-colonization of
glutaraldehyde-ﬁxed epithelial cells with L. jensenii reduced gonococcal adherence by
2-fold (Figure 3.1B), suggesting that L. jensenii inhibits gonococcal adherence by a

mechanism that does not require the epithelial cells to be metabolically active. Since

76

lactobacilli inhibit gonococcal adherence to epithelial cells by a mechanism that does
not require a response from the epithelial cells, this suggests that a component of L.
jensenii is important for the interaction. Our results show that methanol-ﬁxed L.
jensenii inhibited gonococcal adherence to epithelial cells to a similar extent as live
lactobacilli (Figure 3.2). Therefore, it is likely that the components of lactobacilli
necessary to inhibit gonococcal adherence to epithelial cells are surface associated.
Treatment of methanol-ﬁxed L. jensenii with proteinase K abolished the inhibition,
indicating that the inhibitory factor is a protein or contains a protein component.

N gonorrhoeae utilizes a type IV pilus for adherence to epithelial cells in culture
(104, 160) and this adhesin has been shown to be essential for gonococci to colonize
human volunteers (74, 161). Therefore, pilus-mediated adherence would be a likely
target for the observed Lactobacillus inhibition. However, when the effect of L.
jensenii pre-colonization of epithelial cells on the adherence of piliated gonococci and
non-piliated gonococci was examined, both piliated and non-piliated gonococci were
inhibited to a similar extent suggesting that the mechanism of inhibition is not speciﬁc
to the main gonococcal adhesin, type IV pili, but is a more nonspeciﬁc mechanism.

One mechanism that could explain this global inhibition of gonococcal adherence
would be the production of a biosurfactant by the lactobacilli. Biosurfactants are
amphipathic molecules produced by microorganisms that serve a variety of purposes,
including adsorption to surfaces (139). It is possible that L. jensenii produces a
biosurfactant that adheres to the epithelial cell surface, which is left behind when the
bacterium desorbs. The molecule left on the epithelial cell surface could change the

nature of the surface such that it is less able to support gonococcal adherence. Over 15

77

species of lactobacilli have been shown to produce biosurfactants (137). One of the
most well characterized Lactobacillus-produced biosurfactants is produced by L.
reuteri RC-l4, and was found to be effective in preventing the adherence of
Enterococcusfaecalis to hydrophilic glass and silicone rubber surfaces (55, 175, 177).
The biosurfactant from L. reuteri RC-14 was also found to inhibit abscess formation
by Staphylococcus aureus in a surgical implant model in rats, demonstrating that
biosurfactants can inhibit pathogen adherence to biotic surfaces as well as abiotic
surfaces (51). Furthermore, a surface-associated protein was puriﬁed and identiﬁed as
the inhibitory component from this biosurfactant (55). This protein was identiﬁed as a
collagen-binding protein, however, it was not demonstrated if this protein alone had
biosurfactant activity.

Using a method to isolate biosurfactants, L. jensenii was caused to release surface
proteins. The released surface component mixture (RSC) has surfactant activity as
demonstrated by collapsed drop analysis and also inhibited gonococcal adherence in a
dose-dependent manner (Figure 3.3). However, when the RSC was fractionated by
column chromatography, the inhibitory activity was separated from the biosurfactant
activity. A cell-free protein extract from L. helveticus, without biosurfactant activity,
has also been found to inhibit pathogen adherence to epithelial cells (63). This
preparation was an extract of surface layer proteins. Surface layer proteins are a
paracrystalline layer associated with the outer surface of some bacteria, and play a role
in adhesion to surfaces (63). However, we do not consider surface layer proteins as a

potential inhibitory factor, since L. jensenii does not produce surface layer proteins

(129).

78

In the RSC there are at least two inhibitory components, since two distinct
inhibitory fractions were isolated from the RSC following anion exchange
chromatography. The ﬁrst fraction contains several proteins shown by SDS-PAGE
(Figure 3.68), of which the major band has been identiﬁed by MS/MS as a putative
glyceraldehyde-3-phosphate dehydrogenase (GAPDH). The second fraction was
analyzed further and an inhibitory protein was identiﬁed as a putative enolase by

MS/MS. We subsequently cloned the enolase gene from L. jensenii and puriﬁed the

recombinant His6-tagged protein tested it for ability to inhibit gonococcal adherence to

epithelial cells. Since His6-Eno was able to inhibit the adherence of N gonorrhoeae

(Figure 3.88), we can conclude that the putative enolase was one of the inhibitory
factors present in the RSC.

While both GAPDH and enolase are normally thought of as cytosolic enzymes
involved in glycolysis, the enzymes have both been found surface associated in several
Lactobacillus species (8, 26, 78). Furthermore, lactobacillar enolase has been shown to
bind to extracellular matrix components such as ﬁbronectin, collagen, and laminin (7,
70, 144). Our data shows that pre-incubation of the RSC with soluble ﬁbronectin
before use in gonococcal adherence assays abolished the inhibitory effect of the RSC
in a dose dependent manner (Figure 3.7), consistent with the hypothesis that L.
jensenii inhibits gonococcal interactions with epithelial cells by occluding ﬁbronectin
binding sites with its enolase. In the vaginal tract, ﬁbronectin is secreted and coats the
mucosal surface (32), indicating the hypothesis that a surface component of L. jensenii
binds to ﬁbronectin to prevent gonococcal adherence is biologically relevant. Future

experiments will be necessary to determine if the putative enolase is in fact the

79

ﬁbronectin binding protein and whether the enzymatic activity is necessary for the
inhibition.

While a putative GAPDH was identiﬁed as the most abundant protein in P1 from
the RSC (Figure 3.6C), it still has not been identiﬁed as the inhibitory component of
that fraction. Since GAPDH from other Lactobacillus species is known to bind colonic
mucin and ﬁbronectin (144), this protein is a candidate for an inhibitor of gonococcal
adherence to epithelial cells, however, this hypothesis remains to be tested.

In summary, we have shown that Lactobacillusjensenii inhibits gonococcal
adherence to epithelial cells independently of an epithelial cell response, utilizing at
least one surface associated protein. When removed from the surface of lactobacilli,
these proteins can still inhibit gonococcal adherence to epithelial cells, however
inhibition is abolished by treatment with ﬁbronectin, suggesting the inhibitory protein
blocks gonococcal binding to this extracellular matrix component. Now, with
gonococci having achieved “superbug” status it is imperative that new methods of
treatment and prevention be discovered that do not rely solely on antibiotics. The
potential use of Lactobacillus products for prevention of gonorrhea is an exciting

possibility that needs to be further explored.

80

CHAPTER FOUR

Spurbeck RR. and C.G. Arvidson. (2010). Lactobacillusjensenii utilizes an enolase
to inhibit Neisseria gonorrhoeae Adherence to Epithelial Cells.

81

Abstract

Recently a surface-associated protein produced by Lactobacillusjensenii
ATCC 25258 with homology to enolase was found to inhibit the adherence of the
sexually transmitted pathogen, Neisseria gonorrhoeae, to epithelial cells in culture.
Here, we show that protein, now called gonococcal anti-adherence protein 1 or GAPl ,
is an active enolase in vitro. A recombinantly expressed, epitope-tagged version of the
protein, Hisb-GAPI, inhibited gonococcal adherence utilizing the substrate-binding
site, but does not require the activity for inhibition. Furthermore, an L. jensenii gapI'
strain carrying an insertional mutation in the L. jensenii GAP] gene, exhibited reduced

inhibition of gonococcal adherence consistent with this being a multifactorial process.

82

Introduction
Pathogenic and commensal bacteria are known to produce “anchorless”
proteins that are associated with the cell surface, but have no known mechanism for
export or attachment (7). One such protein is the enzyme enolase. Usually considered

to be a cytosolic enzyme, enolase catalyzes the conversion of 2-phosphoglycerate

(2PGA) into phosphoenolpyruvate (PEP) through the removal of H20, a key step in

glycolysis. Enolase can also be thought of as a “moonlighting enzyme,” or a protein
that performs more than one function for the bacterium (5 7). Several pathogens
including Staphylococcus aureus, Streptococcus pyogenes, and Streptococcus
pneumoniae (7), as well as the ﬁingal pathogen Paracoccidiodes brasiliensis, the
protist T richomonas vaginalis, and the ﬁliarial parasite Onchocerca volvulus, utilize
this enzyme to interact with the host (37, 66, 112). Speciﬁcally, these pathogens use
their surface—associated enolases to bind plasminogen, which allows for efﬁcient
cleavage by plasminogen activators to the proteolytic form, plasmin (7, 37, 66, 112).
Once active, plasmin degrades extracellular matrix proteins, allowing the pathogens to
invade deeper into the host tissues (7, 37, 66, 112).

Surface-associated enolases have also been found in several species of
Lactobacillus, usually along with cytoplasmic counterparts. For example, L. johnsonii
was shown to express three distinct enolase proteins (7). The enolases from both L.
johnsonii and L. crispatus bound plasminogen thereby enhancing the activation to
plasmin in vitro (7). However, these organisms are rarely associated with disease, and
are present in the human gastrointestinal and vaginal tracts of healthy individuals. The

reason these organisms have surface-associated enolase able to bind plasminogen has

83

not been elucidated. However, lactobacillar enolases could allow the bacteria to
colonize the mucosa] epithelium, since they are also known to bind to extracellular
matrix proteins, such as laminin, collagen, and ﬁbronectin (7, 26).

We have recently shown that Lactobacillusjensem’z' can inhibit the adherence
of Neisseria gonorrhoeae to epithelial cells by surface associated proteins (154-155).
A preparation of released surface components (RSC) was able to inhibit gonococcal
adherence to epithelial cells, and a protein with homology to enolase 3 (Eno3) of L.
johnsonii correlated to the inhibitory activity. Here, we examine the role of this
protein, GAP], as an inhibitor of gonococcal adherence using insertional mutagenesis
in conjunction with biochemical characterization of a recombinantly expressed,

epitope-tagged enolase protein.

84

Materials and Methods

Bacterial strains and culture conditions. N. gonorrhoeae MS11 [P+, Tr] (148) was

grown at 37°C in a humidiﬁed 5% C02 environment (+ C02) on GC agar (Acumedia,

Lansing, MI) with supplements (75) and VCNT inhibitor (GCV, Becton, Dickinson,

and Company, Sparks, MD). N gonorrhoeae MS11-911 [‘lacZ-EmR] was grown at
37°C + C02 on GC agar with supplements, VCNT inhibitor, and erythromycin at 3

mg/L (Erm3). L. jensenii WT (ATCC 25258, wild type) was grown at 37°C + C02 on
MRS agar (Becton, Dickinson, and Co). L. jensenii-gap1'(ATCC 25258, gap!) was
grown at 37°C + C02 on MRS agar with erythromycin at 5 mg/L (Erms). Both

Escherichia coli strain BL21>\DE3 containing pET24a-em and E. coli EC101 were

maintained on LB agar with kanamycin at 100 mg/L, (KanlOO) at 37°C + C02. E. coli

EC101+ pORIl9 was maintained on LB agar with erythromycin at 300 mg/L

(Erm300) at 37°C + C02. L. jensenii+ pVE6007 was maintained on MRS agar with

chlorarnphenicol at 5 mg/L (Cms) at 35°C + C02.

Cell culture. Hec-l-B cells (ATCC HTB-113), were cultured in Dulbecco’s Modiﬁed

Eagle Medium (DMEM with high glucose and L-glutamate, Invitrogen, Carlsbad, CA)

supplemented with 5% fetal calf serum (FCS, Invitrogen) at 37°C + C02. Adherence

assays were canied out as described previously in 24 well cell culture plates in. which

the Hec-l -B cells were grown to 50-90% conﬂuency (Spurbeck, 2010).

85

Protein expression and purification. His6-GAP1 was expressed in E. coli

BL21)\DE3 and puriﬁed as previously described (155). Puriﬁed protein was dialyzed
against 1x phosphate buffered saline (PBS, pH 7.0) before use in cell culture and
enzyme assays.

Insertional mutagenesis. A mutant was constructed in L. jensenii 25258 using
the method described by Whitehead et a] (185) that has been used to generate
insertional mutations of Lactobacillus reuteri, with some modiﬁcations. The PCR

product from the GAP] gene was digested with EcoRI and a 400 bp fragment was

ligated into pORll9 (EmR), which replicates in E. coli and Lactobacillus only if the

strain contains RepA, and transformed into E. coli ECIOl (repA+). Transformants

were screened on plates containing X-Ga] (40mg/l) for insertion within the lacZ gene
of pORI]9, and conﬁrmed by PCR. The resulting plasmid, pRRSZ-3 7, was then

electroporated (2500 mV, 25 pF, 400 0) into L. jensenii 25258 containing pVE6007

R + . . . . . . .
(Cm , repA ), which rs temperature-sensrtrve for replication. Electroporatron and

preparation of electrocompetent L. jensenii was carried out as described in Papagianni

et al. (125) with the following modiﬁcations. The lactobacilli were grown overnight in
4 x 8 ml/tube of MRS broth at 37°C + C02. A 1:5 dilution was made into pre-warrned
MRS without mixing. The lactobacilli were incubated at 30°C + COz for 2 h until the

OD. reached 0.5. 200 u] penicillin G (10 mg/ml) was added and the culture was

incubated an additional h at 30°C + C02. The cells were then harvested by

centrifugation and the supernatant was removed. The cells were resuspended in 1 ml

86

 

 

LiAc solution (100 mM LiAc; 10 mM DTT; 0.6 M sucrose; 10 mM Tris-HCI, pH 7.5;

ﬁlter-sterilized) and incubated l h at room temperature (RT). The cells were washed
twice with deionized H20, then with 50 mM EDTA, deionized H20, and 0.3 M
sucrose. The cells were then resuspended in 0.3 M sucrose and placed on ice for at

least 30 min before electroporation (protocol provided by Dr. Lin Tao, University of

Chicago, personal communication). Transformants were selected on MRS agar

5 . 5
(Cm Enns) at 35°C + C02. Two transformants were grown in MRS broth (CmSErm )

 

for 18 h without shaking at 35°C + C02. To select for integration of pRRS2-37 into

 

the Lactobacillus chromosome, the culture was diluted 1:200 and grown in MRS Erm5 :-

broth for 8 h at 42°C under anaerobic conditions. Serial dilutions were then plated on

MRS Erm5 agar and incubated at 42°C + C02 for 24-48 h to select for a single cross-
over event with pRRS2-37 inserted into the chromosome at the region of sequence

. . . . R . R
srmrlanty to the gene fragment on the plasmrd. Erm rsolates were screened for Cm
at 35°C + C02 to ensure loss of pVE6007. To conﬁrm the placement of pRRS2-37 on

the L. jensenii chromosome and to show that the gene encoding GAP] was no longer
intact, genomic DNA from the isolates was subjected to PCR using a primer speciﬁc

for the gene (Enond) and another speciﬁc to pORIl9 (M13R, Figure 4.1).

87

enond enoRev

 

enond *

5, _ ‘_ Em I 3‘
Ml3rev SBCG

B

 

1 1:10 1:100 — ‘Ikbladder WT

Figure 4.1. Insertional mutagenesis of L. jensenii 25258 eno. A. Schematic diagram of
plasmid integration into chromosomal locus to disrupt eno. Primers used to ampliﬁy
regions to conﬁrm construct are indicated. B. PCR ampliﬁcation to verify insertion.
Genomic DNA from Erm Cm transformants was used as template for PCR using
primers enond and M13rev. Templates: 1 = undiluted mutant genomic DNA, 1:10 =
mutant genomic DNA, 1:100 mutant genomic DNA, - = no template control, and WT
= L. jensenii WT genomic DNA.

88

Production of released surface components (RSC). RSC was produced from 1 L
overnight cultures of L. jensenii or L. jensenii-gapI' as described previously (155).
The Bradford method was used to determine the protein concentration for each RSC
preparation (Bio-Rad Laboratories) and the RSC was tested for inhibitory activity
against gonococcal adherence to Hec-l-B cells as described below.

Whole cell enolase activity assays. Whole cell enolase activity assays were carried
out as described by Pancholi and Fischetti (124), with some modiﬁcations. Brieﬂy, L.
jensenii WT and L. jensenii gapI' were grown overnight at 37°C. Various amounts of

bacteria were harvested by centrifugation, washed with 100 mM HEPES buffer, pH 7,

and then resuspended in 100 mM HEPES, pH 7, with 10 mM MgC12 and 7.7 mM KC1

with or without substrate (3 mM 2PGA). The bacterial suspensions were incubated 5
min at 37°C and then the bacteria were removed by centrifugation (6000 x g for 5

min). PEP production was determined spectrophotometrically (A = 240 nm) using the

extinction coefﬁcient of PEP of 2.6 x 103 M.

Hi56-GAP1 activity assays. Enolase activity of puriﬁed His6-GAP1 was assayed as
described (124) with some modiﬁcations. 5 pg of protein was incubated with various

concentrations of 2PGA (0.5-3mM) in the reaction buffer described above (100 mM

HEPES buffer, pH 7, 10 mM MgC12, 7.7 mM KCl). All reactions were carried out at

RT and A240 was measured every 5-10 5 to determine enzyme kinetics. Values were

corrected by subtracting out the absorbance recorded for reactions without substrate.

The effects of the known enolase inhibitors, CaC12 (CaZ+) and an equimolar solution

89

of NaF and NaHZPO4, pH7.0 (Fl-+Pi), on the activity of His6-GAP1 was also tested at

various concentrations as described in the text.

PCR amplification of cytosolic enolase. PCR ampliﬁcation of gap] was conducted
using primers enond and enoRev, while ampliﬁcation of eno2 used 0838Fwd and
0838Rev. Genomic DNA was isolated from L. gasseri 33323 and served as a positive
control for the presence of both genes. Genomic DNA from both L. jensenii WT and
L. jensenii gap] ' was isolated and ampliﬁed to determine if both genes were present.
pET24a-em served as a negative control for the ampliﬁcation of enoZ.

Statistical analysis. All data were analyzed by using an unpaired Student’s t-test. A p-
value below 0.05 was considered statistically signiﬁcant and is indicated by an asterisk

(*) in the graphs.

90

Results
DNA sequence analysis of L. jensenii ATCC 25258 gap]. We previously cloned the
gap] gene from L. jensenii 25258 by ampliﬁcation of the ORF and the intergenic 3’
sequence to the neighboring gene, secG (155). This sequence is 99% identical to the
gene number 1305 of L. gasseri 33323. This species of lactobacilli also inhibits
gonococcal adherence to epithelial cells (154). L. jensenii 25258 gap] is also 92%
identical to that of L. johnsonii NCC533 eno3, which has been well characterized (7).
However, it is only 83% identical to the putative enolase gene sequences from either
L. jensenii 1153 or JV-16.

At the amino acid level, L. jensenii 25258 GAP] contains all of the conserved
amino acids necessary for metal binding (4/4), substrate binding (7/7), and the dimer
interface present in all characterized enolases (37/37) (Figure 4.2A). The protein is
also predicted to contain the characteristic TIM B-barrel domain which is found in all
of the enzymes in the enolase superfamily (shaded in gray in Figure 4.2A). A
phylogenetic tree depicting the homology to known enolase protein sequences in other
lactobacilli and Gram positive pathogens is shown in Figure 4.2B. L. jensenii 25258
GAP], L. gasseri 33323 gene 1305 product, and Eno3 ﬁom L. johnsonii form a
divergent group from other lactobacillar enolases, and tend to be more similar to the
enolases known to be surface-associated in L. plantarum and in Gram-positive
pathogens. Within the TIM B—barrel domain, there are 19 residues found only in the
group of GAPl-Iike sequences (highlighted in bold, Figure 4.2A). These amino acid
changes could be signiﬁcant in substrate binding or the conformation of the active site;

however, this has yet to be addressed experimentally.

91

61

121

181

241

301

361

421

I}

MSVITDIHAR
GQGVLTAVEN
AAADELGLPL
ncaerranK
EDCAASEFYN
FiERLGbKVQ
YTAVVSHRSG

*‘k'k

GIHSFYNLHK

EVLDSRGNPT
VNGEIAKAVI
YEYLGGPNAH

GLLQERGEST

.KDTKKYVTVA

IVCDDLFVTN
#
ETEDTTIADL

AAA

OF 432

AEAEVYTELG
GLDVTDQRLI
VLPTPMMNvt
AVGDEGGFAP

A
AA *

DGREYTAEEW

TSYLEKGIKM,

VVATNAGQIK

*

L. jensenii 25258 Gapl

L. gasseri Eno 1305 99%.

L. johnsonii Eno] 56%° '

:0
L. gasseri Eno 0838 56%

GFGRAIVPSG

DQTMIDLDGT

‘NGGKHADNNV

**AAA

NLKNNEEPFE

-TSLJEDLYDK

.GVANSILIKL

*

TGSMSRTDRI

AAA A

o L. johnsonii Eno3 97%

ASTGEHEAVE
#

PNKGRLGANA

'DIQEFMIMPV

ILVEAIQRAG

”YPVISVEDPL.

#
NQIGTLTETF

AKYNQLMRIE

A A AA

0 L. plantarum Eno 78%

' E. hirae Eno 70%
c S. pvogenes Eno 68%

LRDGDKSRFG
ILSVSLASAR
GAKSLHEAYR
YKPGQDLAIA
DENDWEGWKn
EAIEMAKEAG

EALGSTAQYK

_ .1 S. pneumoniae Eno 68%

S. aureus Eno 66%

L. johnsonii En02 53%

L. crispatus or Eno 57%

Figure 4.2. GAP] protein characteristics. A. The primary amino acid sequence of L.
jensenii 25258 GAP]. The region shaded in gray corresponds to the TIM B—barre]
domain common to enolases (16). Amino acids of interest that are speciﬁc to GAP]
are highlighted in bold font. The conserved amino acids are marked below the residue
with # for metal binding, * for substrate binding and A for dimer interface. B. A
phylogenetic tree of enolase protein sequences. % = identity to GAP]. Distances were
calculated using the Grishin model with maximum sequence difference of 0.85 (142).

92

 

Another region of interest is the C-terminus of the protein, where 7 of the last 14
amino acids are found only in the enolases in the GAPl-like group and the pathogen
group and 4 of the 14 are found only in the GAPl-like group. Recently, two C-
tennina] lysine residues of streptococcal surface enolase were shown to have a role in
the binding of plasminogen, suggesting that this region might be important for binding

of altemative substrates (35).

Enolase activity of His6—GAP1 in the presence and absence of inhibitors. Enolase

activity assays were carried out using the puriﬁed recombinant protein, HIS6—GAPI , to

 

determine if the tagged protein was enzymatically active. His6—GAP1 converted

2PGA to PEP, and the activity correlated with the amount of enzyme, suggesting that

the reaction is a ﬁrst order reaction. Furthermore, when the substrate concentration

was varied, the reaction followed Michaelis-Menten kinetics. Km for 2PGA was 92.9
i 0.82 uM and the maximum velocity was determined to be 791.6 i 297.5 11M min-1

. -1 - . . . . .
(mg protern) . Fl +P; rs known to complex wrth the Mg2+ rons present 1n the actrve
site of the enolase, causing a conformational change that closes the active site (87).
When added to the reaction at 5 mM, Fl-+Pi reduced the enolase activity of His6—

GAP] by 87.2%. Ca2+, another inhibitor of enolase activity, binds with higher afﬁnity

2+ . . . . . . . . . .
than Mg to the metal binding srtes 1n the actrve srte, leavrng the actrve srte accessrble

to substrate, but blocking transformation of the substrate (86). When added to the

. . . . 2+ .
reaction 1n rncreasrng amounts, Ca reduced 2PGA transforrnatron. At 2.26 fold more

93

Ca2+ than Mg2+ in the reaction, the enolase activity of His6—GAP1 was reduced by
87.8%.

Effect of enolase inhibitors on His6—GAP1 reduction of gonococcal adherence.

We previously showed that His6—GAP1 concentrations as low as 0.14 mg/ml inhibited

gonococcal adherence to epithelial cells by ~50% (155). This inhibition is speciﬁc to ?

His6—GAP1, since incubation of the epithelial cells with BSA (1.76 mg/ml) prior to

gonococcal inoculation did not affect the adherence of gonococci (BSA: 19.3 :t 6.7%,

 

control: 14.7 :1: 8.5%, p = 0.607). However, it was not clear whether enzymatic activity

of the enolase was required. The enolase inhibitors Fl‘+Pi and Ca2+ were examined

for their effect on His6—GAP1 inhibition of gonococcal adherence to epithelial cells.

Glutaraldehyde ﬁxation was done to ensure that the results observed were due solely
to effects of the inhibitors on the enzyme, and not due to any change in the epithelial

cells. To make certain that there were no direct effects of the inhibitors or enzyme on

N. gonorrhoeae, His6—GAP1 and inhibitors were removed and the ﬁxed cells were

washed with PBS prior to introduction of gonococci. Surprisingly, Fl-+Pi abolished

the inhibition of gonococcal adherence to epithelial cells, such that gonococci adhered

essentially the same as in the absence of His6—GAP1. However, addition of Ca2+ did
not affect the reduction of gonococcal adherence by His5—GAP1 (Figure 4.3).

. . . . 2+
Incubation of the enzyme wrth addrtronal Mg , the metal cofactor necessary for

enolase enzymatic activity, also had no effect on the inhibition of gonococcal

94

adherence (data not shown). Since Ca2+ blocks enolase activity by replacing the
catalytic Mg2+ ions without occluding the substrate binding site, and Fl-+Pi

. 2+ . . . . .
complexes With the structural Mg rons clos1ng the substrate brndrng srte, these data

suggest that the enolase activity is not necessary but the substrate binding site is

essential for inhibition of gonococcal adherence to epithelial cells to occur.

 

% Adherence

 

His6- Eno ‘ + + +
Ca ' - + -
F- + P] ' - - +

Figure 4.3 His6-Eno inhibition of gonococcal adherence to glutaraldehyde ﬁxed Hec-
l-B cells. N gonorrhoeae alone adhered to the ﬁxed cells with a frequency of 12.9 i

1.6%. When treated with 0.14 mg/ml His6-Eno, gonococcal adherence was reduced to

7.9 i 1.3% (no inhibitors,p = 0.036), 7.9 i 1.5% (Ca2+,p = 0.043), or 13.4 i 0.4% (F-
+ Pi, p = 0.746). The data are the averages of 3 independent experiments performed in

duplicate. Error bars depict standard error.

Construction of an enolase mutant. To determine if GAP] is one of the inhibitory
components involved in the inhibition of gonococcal adherence to epithelial cells by
live L. jensenii, the encoding gene was insertionally inactivated on the chromosome of
L. jensenii 25258. Growth of L. jensenii gap] ' was not signiﬁcantly different than that
of L. jensenii WT (data not shown). By sequencing pRRS2-37, it was determined that

the plasmid inserted at bp 1252, disrupting the last 42 bases in the gap] gene sequence

95

(Figure 4.1A). The insertion occurred between amino acid 417A and 418Q in the
protein sequence, 15 residues from the C-terminus.

Enolase activity of intact L. jensenii and cytoplasmic fractions. To determine the
effect of the gap!" mutation on cellular enolase activity, we used an activity assay
(124) using whole, intact lactobacilli to catalyze the reaction. As seen in Figure 4.4A,
intact L. jensenii 25258 have enolase activity that increases as the cell concentration
increases. However, when gap] has been inactivated, as in L. jensenii gapl', enolase
activity is undetectable. From this result, we conclude that the gene that was

inactivated in L. jensenii gapI' encodes an enolase present on the surface of the

 

lactobacilli. Since bacteria typically have cytoplasmic enolases, we also assayed
cytoplasmic extracts of both L. jensenii WT and L. jensenii gapI'. Accordingly,
lactobacilli were sonicated to disrupt the cells, and then centrifuged to remove the
insoluble fraction and any unbroken cells (including the surface-associated GAPl).
The protein content of the soluble fraction was determined and then assayed for
enolase activity at various protein concentrations. The results of these experiments
showed that cytoplasmic extracts of both L. jensenii WT and L. jensenii gapl' have
enolase activity that increases with protein concentration, suggesting that there is
another enolase that is strictly intracellular (Figure 4.48). However, the enolase
activity in the cyt0plasm of L. jensenii gap]' is signiﬁcantly lower than that of L.
jensenii WT (50 pg: p = 0.013, 75 pg: p = 0.015, and 100 pg: p = 0.002) which
suggests that surface-associated enolase is either active intracellularly or some of the

GAP] on the surface of L. jensenii WT is still present in the wildtype samples.

96

Due to the similarity between the L. jensenii gap] and L. gasseri 33323 gene
1305, we hypothesized that if a second enolase gene was present in L. jensenii, that it
would be similar in sequence to that of L. gasseri 33323 gene 0838. Therefore, we
constructed primers speciﬁc to L. gasseri 33323 gene 0838 that do not cross react with
gap]. By PCR, we conﬁrmed the presence of a second enolase, and that the primers
do not amplify GAP] (data not shown). This result suggests that L. jensenii has
evolved to have two enolases, one that works intracellularly, and a second that has 1

been adapted to serve an alternate function on the bacterial cell surface.

 

2b
III

pM PEP min“

uMPEPmin"
o '5 o 5 B 1:; a is

   

2 4 6 8 10
CFU/ml 3:10” p mum

Figure 4.4. Enolase activity of L. jensenii WT and L. jensenii gap] '. A. Enolase
activity of increasing concentrations of intact lactobacilli plotted as a function of PEP
produced. B. Enolase activity of increasing amounts of cytoplasmic extracts from L.

jensenii WT and L. jensenii gap] -. Solid line = L. jensenii WT and the dotted line = L.

jensenii gap] '. The data are the averages of 3 independent experiments performed in
triplicate. Error bars depict standard error.

Effect of gap]' on gonococcal adherence to epithelial cells. To examine the effect of
L. jensenii gap] 'on gonococcal adherence, Hec-l-B cells were colonized with either L.

jensenii WT, or L. jensenii gap] ' for 1 h prior to infection with N gonorrhoeae strain
R . . .
MS11-911 (Em ). Both L. jensenii WT and L. jensenii gap] ' were able to rnhrbrt

gonococcal adherence to epithelial cells; however, the magnitude of inhibition differed

97

signiﬁcantly (p = 0.027, Figure 4.5). L. jensenii WT reduced gonococcal adherence to
59.1 i 7.0% of the no Lactobacillus control (p = 0.002, Figure 4.5), similar to
previous reports (Spurbeck 2008; Spurbeck 2010). However, L. jensenii gapl' only
inhibited gonococcal adherence to 82.0 i 4.4% of the control (p = 0.008 compared to
no Lactobacillus control (Figure 4.5). These results suggest that L. jensenii inhibits
gonococcal adherence to epithelial cells using the surface associated GAP], but this
factor is not solely responsible for the inhibition. Adherence of lactobacilli was also
measured in these experiments, and the adherence of L. jensenii gap] ' (0.57 i 0.18%),
was not signiﬁcantly different from L. jensenii WT (0.40 i 0.14%, p = 0.229). This
suggests that although GAP] is involved in the inhibition of gonococcal adherence, it
does not contribute signiﬁcantly to the adherence of the lactobacilli to the epithelial

cells.

 

 

.5
O
I

Relative adherence (%)
N

 

 

 

O
I

No lactobacilli L. jensenii WT L. jensenii gap] '

Figure 4.5. Inhibition of N gonorrhoeae adherence to epithelial cells by L. jensenii

WT and L. jensenii gap]: Relative adherence is the percentage of adherent gonococci
in the presence of lactobacilli divided by the percentage of adherent gonococci in the
no Lactobacillus control. The data are the averages of 3 independent experiments
performed in triplicate. * Signiﬁcantly different from no lactobacilli, ** Signiﬁcantly
different from L. jensenii WT and no lactobacilli. Error bars depict standard error.

98

Effect of GAP]- RSC on gonococcal adherence to Hec-l-B cells. Since the GAP]

was ﬁrst discovered as a part of the RSC that could inhibit gonococcal adherence
(155), RSC was prepared from L. jensenii gap]- (GAPI- RSC). Varying

concentrations of GAPl' RSC were tested for inhibitory activity against gonococcal
adherence to Hec-l-B cells. GAPl' RSC reduced the adherence of N gonorrhoeae to
epithelial cells in a dose dependent manner, similar to what was observed for RSC
from L. jensenii WT (155). However, 30% more protein was necessary to reach the
level of inhibition seen by RSC preparations derived from wild type (Figure 4.6).
These data support the hypothesis that while GAP] is an inhibitory component of the
RSC, there is at least one other inhibitory component present in the released surface

components of L. jensenii 25258.

99

 

Relative adherence 1%]

 

mg/ml RSCWT: - 1.7 - - _ _
mg/ml RSCgapl: - - 0.42 0.84 1.7 2.24

Figure 4.6. Inhibition of gonococcal adherence by released surface components (RSC)

from L. jensenii gap] - and L. jensenii WT. Gonococcal adherence frequencies are
reported as relative to a no RSC control, and p—values are reported as compared to WT
RSC. Gonococcal adherence to epithelial cells was reduced to 31.2 :h 6.8% (1.7 mg/ml

wr RSC), 98.3 a 12.0% (0.42 mg/ml GAPI' RSC, p = 0.004), 79.1 a 10.5% (0.83
mg/ml GAPl- RSC, p = 0.011), 68.5 a 15.4% (1.7 mg/ml GAPI' RSC, p = 0.048), and

42.8 2+: 6.2% (2.2 mg/ml GAPI- RSC,p = 0.206) ofthe untreated control. The data are
the averages of 3 independent experiments performed in duplicate. Error bars depict
standard error.

100

Discussion

The primary role of the enzyme enolase is the conversion of 2-
phosphoglycerate into phosphoenolpyruvate in glycolysis (16). This reaction is
essential to the survival of cells, as this is a crucial step in the production of ATP.
However, several cells, including Gram-positive bacteria, eukaryotic parasites, and
skeletal myoblasts express enolase on the cell surface, and utilize enolase as a receptor
for various substrates (7, 26, 37, 66, 90, 112). Surface-associated enolases from
pathogens, parasites, and myoblasts have been shown to enhance the activation of
host-derived plasminogen to the proteolytic enzyme plasmin, which then degrades the
extracellular matrix of the host (90, 124). This allows pathogens and parasites to
invade deep into the host tissues. Myoblasts use this degrading enzyme during tissue
repair. However, why lactobacilli have these surface associated enolases is not as
clear. Unlike invasive pathogens, which are known to utilize these proteolytic
enzymes, lactobacilli do not invade the host, but usually reside in the mucosa] layer of
either the intestine or the vaginal tract (189). Several Lactobacillus species have been
found to have surface associated enolases that bind to extracellular matrix proteins
such as collagen, ﬁbronectin, and laminin (7, 26, 144); thus it has been proposed that
these commensal bacteria utilize surface-associated enolases to colonize the host.

One difference between Gram-positive pathogens and lactobacilli is the
number of enolase genes in the genome. Gram-positive pathogens have only one
enolase that is present both in the cytosol and on the cell surface, whereas most
lactobacilli tested have at least two enolase genes (7, 26). Here we have shown L.

jensenii ATCC 25258 has two enolase genes in its genome, gap] and en02. GAP] is a

10]

 

surface associated enolase, while En02 is a cytosolic enolase. Other closely related
lactobacilli, L. gasseri ATCC 33323 and L. johnsonii NCC533 have two enolases with
high sequence identity to those found in L. jensenii, however the cellular localization
of the enzymes has not been studied. L. plantarum, also has two surface-associated
enolases. The enzyme with the highest identity to GAP] binds ﬁbronectin, which
allows L. plantarum to colonize colonic epithelial cells in vitro (26). Therefore, it
appears that several Lactobacillus species have evolved to have two proteins to
perform separate functions, one surface-associated which is involved in interactions
with the host, and the second a cytosolic enzyme functioning in glycolysis.

While GAP] catalyzes the reaction of 2PGA to PEP in vitro, it is likely that it
is inactive on the bacterial cell surface in vivo, as an inhibitor of enolase activity, Ca2+,
is present in the vaginal ﬂuid at 1-5 mM/kg ﬂuid, whereas Mg2+, the cofactor
necessary for catalytic activity, has not been detected (18]). In fact, the formula for
vaginal ﬂuid simulant includes Ca(OH)2 (0.222 g/l) and no magnesium, again
demonstrating that GAP] is most likely inactive in the vaginal tract (12]).
Furthermore, in the cell culture medium (DMEM, Invitrogen) used to initially identify
GAP] as an inhibitor of gonococcal adherence to epithelial cells, there is 2.26-fold
more calcium than magnesium, which would render the enzyme inactive in the cell
culture assays. This supports the ﬁnding that the enzymatic activity is not important
for inhibition of gonococcal adherence in this in vitro system (Figure 3). In addition,
we demonstrated that the substrate-binding site is necessary for the inhibition of
gonococcal adherence, suggesting that the structure of GAP] is essential for the

inhibitory activity. Therefore, we propose that although this protein has retained the

102

 

ability to act as an enolase, that it is not functioning as an enolase when present on the
surface of L. jensenii.

Bacteria for alternative functions have adapted several enzymes known to be
active in central metabolism. These "moonlighting enzymes", while retaining
signiﬁcant homology to known enzymes, can often act as receptors on the cell surface.
In this group of moonlighting proteins are several glycolytic enzymes, including
ﬁ'uctose-1,6-bisphosphate aldolase (168), enolase (8, 26), phosphoglycerate kinase
(24) and glyceraldehyde-3-phosphate dehydrogenase (GAPDH, (8, 78, 165)). All of
these enzymes have been shown to mediate pathogen interactions with the host, either
through adherence to host cells (7-8, 26, 78, 168), evasion of the immune system
(165), or by blocking the invasion of host tissues (24). Therefore, surface-associated
glycolytic enzymes may play a central role in bacterial-host interactions, and deserve
further study.

We propose that L. jensenii produces a surface-associated enolase, GAP], to
reduce competition with other bacteria in the vaginal tract. Other lactobacillar enolases
are known to bind to extracellular matrix proteins such as collagen, ﬁbronectin, and
laminin (7, 26, 144), and we have previously shown the inhibition of gonococcal
adherence by released surface components from L. jensenii can be removed by
incubation with soluble ﬁbronectin (155). Therefore, it is plausible that an interaction
between GAP] and host cell receptors or extracellular matrix proteins, such as
ﬁbronectin, reduces gonococcal adherence to epithelial cells. The receptors gonococci
adhere to on the host cells are bound by the active site of GAP] , thus reducing the

available substrate for gonococci to colonize.

103

 

 

Treatment and prevention of STIs, such as gonorrhea, is a signiﬁcant public
health challenge, particularly considering the rapid development of antibiotic
resistance. The discovery of GAP] from L. jensenii could lead to new therapeutic

breakthroughs for one of the most common infectious diseases in the United States.

104

 

 

 

CHAPTER FIVE

 

105

SIGNIFICANCE AND IMPACT OF THIS WORK

Vaginal infections are a common problem for women worldwide. Several
infections, including vulvovaginal candidiasis and bacterial vaginosis, can develop
into recurrent infections, even in women who are treated with antimicrobial
compounds. The ineffectiveness of these treatments necessitates the search for
alternative treatments or prevention methods to treat these persistent infections.
Women can also be re-infected with N gonorrhoeae, one of the most common
sexually transmitted pathogens in the United States. This disease is treatable with
antibiotics; however, the infectious agent, Neisseria gonorrhoeae, is now classiﬁed as
a multidrug resistant "superbug", and thus is becoming more difﬁcult to treat. To date,
the CDC only recommends one class of antibiotics to be used to treat gonorrhea, third
generation cephalosporins (www.cdc. gov). The high incidence of the disease, coupled
with the bacterium's increasing resistance to antibiotics, produces a need to study
potential alternative methods to treat or prevent gonorrhea.

One potential alternative for the treatment or prevention of vaginally acquired
infections is the use of probiotics, deﬁned as "live microorganisms, which when
administered in adequate amounts, confer a health beneﬁt upon the host". Another
alternative treatment or preventative could be products derived from these beneﬁcial
microbes. Ideally, probiotics and their products should be safe for continuous use in
order to reduce the risk of recurrent infections. One of the most studied genera for use
as probiotics is Lactobacillus. Lactobacilli are thought to protect the vaginal tract

against incoming pathogens, including STIs such as gonorrhea. However, before this

106

 

 

work, there was no experimental evidence that lactobacilli could inhibit gonococcal
interactions with the host epithelial cells.

In this dissertation Lactobacillusjensenii, one of the four main species found
in the healthy human vaginal tract, was shown to inhibit gonococcal adherence and
invasion of epithelial cells in culture (Chapter 2, (154)). The process of adherence and
invasion are the initial steps in the gonococcal infection: if the bacteria cannot adhere
to the epithelial cells in the endocervix, the pathogen cannot produce a sustainable
infection. Therefore, the ability of L. jensenii to inhibit these steps implies that this
organism, or its products, may have potential to be used as a method for reducing the
transmission of gonorrhea. However, L. jensenii was unable to signiﬁcantly displace
gonococci already adherent to epithelial cells, which suggests that these microbes are
limited to the prevention of gonorrhea. I have found that another species of
Lactobacillus, L. gasseri ATCC 33323, was able to displace adherent gonococci and,
therefore, does have potential to provide an alternative treatment. However, L. gasseri
only displaced ~30% of the adherent gonococci, and may need to be genetically
modiﬁed to increase the displacement for effective treatment to be achieved (Chapter
2, (154)).

The mechanism(s) probiotic organisms utilize to prevent pathogen adherence is
an area that has not been thoroughly pursued, hampering the deve10pment of
probiotics as treatments or prophylactics. Therefore, I sought to determine the
molecular mechanism(s) of the inhibition of gonococcal adherence to epithelial cells

by L. jensenii (Figure 5.1).

107

 

Figure 5.1. Potential probiotic mechanisms of inhibition of pathogen adherence. A.
Secreted compounds such as (O) lactic acid, (() hydrogen peroxide, ( A )

biosurfactants, or (O) bacteriocins. B. Direct competition for host cell receptors. C.
Co-aggregation with the pathogen. D. Inhibition of virulence gene expression. E.
Inducing the host epithelial cells to become more resistant to infection.

After eliminating several proposed mechanisms of inhibition, including direct killing,
co-aggregation, and secreted inhibitory factors, I determined that L. jensenii inhibits
gonococcal adherence using at least two surface associated protein components, one of
which is an enolase (Chapter 3). This enolase, GAP 1, is a glycolytic enzyme that
converts 2-phosphoglycerate to the higher energy compound phoshoenolpyruvate (16).
This enzyme inhibits gonococcal adherence by utilizing the substrate-binding site, but
the dehydratase activity is not required for this inhibition (Chapter 4). Therefore, I
propose a model by which L. jensenii inhibits gonococcal adherence by actively
depositing GAP] on the surface of the epithelial cells or on the extracellular matrix.
This enolase binds to an as of yet unknown epithelial cell receptor, thereby inhibiting

the adherence of N gonorrhoeae to the same receptor, or by preventing gonococci

from reaching a different receptor in close proximity to where GAP] binds (Figure

108

 

5.2). One potential receptor for GAP] is ﬁbronectin, since the addition of soluble

ﬁbronectin to the RSC abolished the inhibition of gonococcal adherence (Chapter 3).

AA

u Lactobacillus u

 

Figure 5.2. Proposed mechanism by which L. jensenii GAP] inhibits gonococcal
adherence. L. jensenii deposits its surface bound GAP] on epithelial cell receptors.
This blocks the adherence of N gonorrhoeae to the same or neighboring receptors.
The discovery of a protein produced by L. jensenii that can inhibit gonococcal
adherence is signiﬁcant, in that this protein should be safe for use as a prophylactic
against gonorrhea. Further studies need to be conducted to determine the efﬁcacy of
preventing other vaginal infections, as well as how effective this protein is in vivo.
Once it is known whether L. jensenii or GAP] can be used to prevent gonorrhea, or
other vaginal infections, the organism or its product could be marketed and used to

reduce the incidence of gonorrhea. This would greatly relieve the burden on our health

care system incurred by the treatment of these diseases and their sequelae.

109

 

My work demonstrates that the conventional way pathogenesis research is
conducted, where researchers study pathogen interactions with the host in isolation,
needs to be broadened to include the indigenous microbiota to yield realistic results.
Rarely is it the case that a pathogen solely interacts with a site of the host not already
colonized by commensals. As shown in this work, bacteria that are indigenous to the
human body can drastically affect the interactions of a pathogen with the host. In the
ﬁeld of probiotics, researchers need to pay particular attention to the mechanism by
which the beneﬁcial microorganisms inhibit pathogen interactions with the host. By
understanding how a probiotic works, better predictions can be made about the
efﬁcacy of treatment for different diseases.

FUTURE EXPERIMENTS

While this work has made signiﬁcant contributions to understanding how L.
jensenii effects gonococcal interactions with the host epithelial cells, there are still
many unanswered questions that need to be addressed in future experiments. First,
ﬂuorescently labeled antibodies speciﬁc to the GAP] can be used to conﬁrm that the
GAP] from L. jensenii does bind the host epithelial cells by visualizing the bound
protein through confocal microscopy. Next, the epithelial cell receptor potentially
blocked by the binding of GAP] should be determined. Gonococci are known to bind
several host cell receptors, however, the cell line used to examine the effects of L.

jensenii on gonococcal adherence, Hec-l-B, only express a subset. On Hec-l-B cells,
gonococci bind to extracellular matrix proteins (34, 41, 174), glycolipids (126),
heparin sulfate proteoglycans (50, 109), lutropin receptors (153), and CD46 (69).

Taking into consideration the substrate speciﬁcity of the active site of the GAP] ,

110

 

which is involved in the inhibition of gonococcal adherence to Hec-l-B cells, the
ligand found on the cell surface should be negatively charged, or contain a phosphate
group and/or a carboxyl group (16). To determine the cell receptor, one could utilize
antibodies to each receptor for gonococcal adhesins to try to block the GAP] from
binding. This could be visualized by confocal microscopy, and then the afﬁnity to
each receptor could be quantiﬁed by an ELISA and western blot. An alternative
method would be to allow the GAP] to bind to Hec-l-B cell lysates and cross-link the
protein with bound ligand. A western blot could be used to identify proteins bound by
the GAP]. Such proteins can then be excised from an SDS-PAGE gel and sent for
tandem mass spectrometry to determine their molecular characteristics and biological
properties.

Once the ligand is determined on the epithelial cells, the residues necessary for
binding should be examined. The active site of the GAP] is located on the C terminal
end of the TIM [ti-barrel domain, therefore, the speciﬁc amino acids necessary for
binding the substrate will be found in the Ba loops on this end, as well as the C
terminal ends of the )6 sheets (186). To determine which residues are necessary for the
inhibitory action against gonococcal adherence, site-directed mutagenesis should be
performed to systematically substitute the charged polar amino acids in these regions
with nonpolar amino acids. By aligning the GAP] sequence from L. jensenii 25258
with several surface associated and intracellular enolases, 19 candidate residues have

been identiﬁed (Figure 5.3).

111

 

 

1 MSVITDIHAR EVLDSRGNPT AEAEVYTELG GFGRAIVPSG ASTGEHEAVE LRDGDKSRFG 60
* e # *# * ******* A*‘# ###A* A * A**** ***** **** ****## *@#

61 GQGVLTAVEN VNGEIAKAVI GLDVTDQRLI DQTMIDLDGT PNKGRLGANA ILSVSLASAR 120
****A***#* ** @*#A # @#*A t *A@**@**** **** ***** ** ** *A*

121

181

241

301

361

421

AAADELGLPL
***#@ **
MGAETFHTLR
## #‘#* *

FDCAASEFYN
#* *@*#**@

FTERLGDKVQ
*# ##* *

YTAVVSHRSG

* # *‘k'k‘k'k

GIHSFYNLHK
# ‘#####‘

YBYLGGPNAH

*A****# AA

GLLQERGEST

* @#*AA*

KDTKKYVTVA

@* A A

IVGDDLFVTN

*#** #**

ETEDTTIADL

**#* #‘k *

OF 432

AA

VLPTPMMNVI NGGKHADNNV'

***A* *

* # #*

"AVGDEGGFAPHNLKNNEEBFEMILVEAIQRAG"YKPGdDIAIA """" 240

******** # @ *

D--GREYTAEEW TSLIEDLVDK 'YPVISVEDPL DENDWEGWKT 300

AAA AA

*@*A* *

VVATNAGQIK TGSMSRTDRI AKYNQLMRIE EALGSTAQYK 420

**#*#*i * **# #*

#

*

IIIII

*##***

*

DIQEFMIMPV‘

*****#**

GAKSLHEAVR 180

##A A #

# #*A#** * **A

##**

**@# **** * #

***** ***

#***#*

#‘k‘k

#*# # #*

A*#A #A#

 

Figure 5.3. The peptide sequence of L. jensenii GAP]. A only in L. jensenii like group,
# in L. jensenii like group and pathogen group, @ only in Lactobacillus groups, and *
conserved in all groups. In bold are residues of interest for site directed mutagenesis.
The gray shaded region corresponds to the TIM barrel domain.

These amino acids are found in the closely related enolase sequences from L. jensenii
25258, L. gasseri 33323, and L. johnsonii, and therefore, these could potentially be
involved in modifying the substrate speciﬁcity to include the eukaryotic ligand. A
cytoplasmic enolase could also be tested for the ability to inhibit gonococcal
adherence to epithelial cells. If the cytoplasmic enolase does not inhibit gonococcal
adherence, this protein could undergo site directed mutagenesis to change speciﬁc
residues to those found in GAP] from L. jensenii 25258 to see which are sufﬁcient for
inhibition of gonococcal adherence. If the cytoplasmic enolase is found to inhibit
gonococcal adherence, then its sequence should be compared with that of GAP] to

determine if there is a signal sequence or other motif that determines the cellular

localization of each enzyme.

112

One region of interest is the last 14 residues at the C-terminus. This could be a
potential localization sequence, as this sequence is highly conserved between most
surface-associated enolases, or may be necessary for the binding of an epithelial cell
receptor (ﬁgure 5.3). In Streptococcus pyogenes, this region of a surface bound
enolase was found to contain two lysine residues necessary for the binding of
plasminogen (35). To determine if the 14 amino acids of the C-terminus are necessary
for localization or inhibition of gonococcal adherence, a deletion of this region could
be made and then expressed in L. jensenii gap] ' on a plasmid. Enolase activity of
whole cells could then be compared between the insertional knockout and the deletion
mutant. If the deletion mutant can complement the knockout, then the 14 amino acids
of the C-terminus is not involved in localization to the cell surface. If, however, the
deletion mutant does not complement the surface associated activity, then this region
should be further analyzed by a series of truncations or single amino acid substitutions
to determine the consensus sequence that is necessary and sufﬁcient for cell surface
localization. The deletion mutant should also be examined for the ability to inhibit
gonococcal adherence, since the result would shed light on the role of this region in
that phenomenon.

L. jensenii was found not only to inhibit N gonorrhoeae adherence to
epithelial cells, but to inhibit gonococcal invasion of the host cells as well. However,
the effect of either the RSC or GAP] from L. jensenii on gonococcal invasion has not
been examined. Therefore, invasion assays should be performed to determine if GAP]
or other inhibitory components found in the RSC mediate the inhibition of gonococcal

invasion. The prevention of gonococcal invasion of host epithelial cells could reduce

113

 

the severity of the disease; therefore knowing what lactobacillar component mediates
this inhibitory effect would be beneﬁcial for the development of new therapeutics.
Finally, identiﬁcation of the other inhibitory protein(s) present in the RSC
should be completed. One candidate is the glyceraldehyde 3-phosphate dehydrogenase
(GAPDH) present in the ﬁrst inhibitory fraction. Although this enzyme is normally
thought to be cytosolic, GAPDH has been found on the cell surface of other
lactobacilli as well as pathogenic bacteria (8, 35, 78). Furthermore, GAPDH has been n1
shown to mediate the binding of some Lactobacillus species to colonic mucin, and

extracellular matrix proteins such as ﬁbronectin (78, 144). To determine if GAPDH is

 

involved, anion exchange chromatography followed by size exclusion chromatography J
could be conducted to further isolate the protein from the RSC. Then the protein could
be assayed for inhibitory activity against gonococcal adherence. The gene for GAPDH

in L. jensenii should also be identiﬁed and cloned into an expression vector to obtain

puriﬁed His6-tagged protein. The puriﬁed enzyme could then be tested for inhibition

of gonococcal adherence to epithelial cells. Once the other inhibitory protein(s)
present in the RSC is determined, the mechanism of inhibition can then be examined.
Once this research has been completed, therapeutics may be designed to treat or
prevent gonorrhea and possibly other vaginal infections, thus improving women’s

health in our society.

114

APPENDIX
Spurbeck, R. R. and Arvidson, C. G. (2010). Subtractive hybridization method to

isolate organism-speciﬁc RNA from a mixed bacterial sample.

115

ABSTRACT
Microarray analysis is an extremely powerful technique to examine whole
transcriptome changes in response to different conditions. Quantitative PCR is
typically used to corroborate or expand upon microarray data. These techniques
require puriﬁed RNA, which limits their application to single organism systems. Since
we are interested in the response of Neisseria gonorrhoeae to the vaginal microbiota,
we developed a subtractive hybridization technique to obtain highly puriﬁed

gonococcal RNA from samples also containing Lactobacillusjensenii.

116

 

INTRODUCTION

Microarray analysis of gene expression is a widely used method to examine
global changes in gene expression in response to various experimental conditions, and
is a powerful tool for the study of bacterial pathogenesis. Approaches to the study of
gene expression in pathogens that involve analysis of transcript levels are powerful
because changes in response to environmental cues can indicate potential therapeutic
targets. Most studies of bacterial pathogenesis utilize an idealized system containing
only the pathogen in question and host cells. However, in vivo, pathogens are in a
complex environment containing the indigenous microbiota with which they often
interact (83). Our laboratory studies the sexually-transmitted bacterial pathogen,
Neisseria gonorrhoeae. We have developed and used a DNA microarray to examine
global gene expression in gonococci in a tissue culture model of infection (3 8-39) and
have also begun to examine the effects of commensal lactobacilli on gonococcal
interactions with host cells (154-155). In order to overcome the problem of isolating
gonococcal RNA from samples containing lactobacilli, it was necessary for us to
develop a technique to subtract L. jensenii RNA out of a mixed sample, leaving behind

puriﬁed RNA of N gonorrhoeae that can be used for gene expression studies.

117

 

MATERIALS AND METHODS
Bacterial strains and culture conditions. Neisseria gonorrhoeae MSll [P+ Tr (148)]

was grown on GC agar (Accumedia, Lansing, MI) with supplements (75) and VCNT

inhibitor (Becton, Dickinson, and. Company, Sparks, MD) at 37°C with 5% C02 (+
C02) in a humidiﬁed atmosphere. L. jensenii [ATCC 25258, H202+ (156)], was

grown on MRS agar at 37°C + C02 in a humidiﬁed atmosphere (Difco, Sparks, MD).

Co-culture conditions and RNA isolation. Lactobacilli and gonococci were swabbed
from overnight plate cultures into Dulbecco’s modiﬁed Eagle medium without phenol
red (DMEM, Invitrogen, Carlsbad, CA) supplemented with 5% fetal calf serum

(Invitrogen), 110 mM sodium pyruvate, and GC supplement 1] [12.4 11M Fe(NOg)3].

Bacterial concentrations were determined spectrophotometrically (A600). In a 13 x
100 mm cell culture plate, 4 ml of DMEM was inoculated with 109 CFU/ml L.
jensenii and incubated for 1 h at 37°C + C02. The culture was then inoculated with

109 CFU/ml N gonorrhoeae and incubated 3 h at 37°C + C02. The bacterial

suspension was then transferred to a tube containing an equal volume of ice cold
MeOH and incubated on ice for 5 min before centrifugation to harvest the bacteria.
Controls in which only N gonorrhoeae or L. jensenii were present in the culture were

treated similarly. The bacteria were resuspended in 200 p] lysosyme (10 mg/ml in

H20) to lyse the gonococci. RNA was isolated from each sample using the Qiagen

RNeasy kit according to the manufacturer’s protocol (Qiagen, Valencia, CA) and the

concentration was quantiﬁed spectrophotmetrically (A260).

118

 

 

Genomic DNA isolation. L. jensenii was grown overnight in 10 ml MRS broth, with
shaking at 37°C. The bacteria were harvested by centrifugation and then resuspended
in 1 m1 STE buffer (6.7% sucrose, 50 mM Tris-HCl [pH 8.0], 1 mM EDTA)
containing 0.25 U/ u] mutanolysin (Si grrra-Aldrich, St. Louis, MO), and incubated for
30 min at 37°C. 300 [1.1 of 20% SDS was added, and the sample was mixed and
incubated at 42°C for 30 min. DNA was then extracted by phenol:chloroform:isoamyl
alcohol (25:24:1) three times. The sample was treated with RNase, and then extracted

with chloroform. DNA was then ethanol precipitated, resuspended in TE buffer, and

quantiﬁed spectrophotmetrically (A260).

Subtractive hybridization. To remove L. jensenii RNA from RNA samples, an equal
amount of L. jensenii genomic DNA was added to each RNA sample, and brought to
30 u] with TE. The sample was gently mixed by inversion, and incubated at 42°C for
25 min. The RNAzDNA mixture was chilled on ice and 5 a] of RNase H (10 U/pl,
Ambion Inc., Austin, TX) was added. The sample was then incubated at 37°C for 45
min, followed by 5 min incubation at 65°C to inactivate the enzyme. The sample was
then treated with DNase I (Ambion) according to the manufacturer’s directions to
remove the remaining L. jensenii DNA from the sample.

The RNA samples were then reverse transcribed using SuperScript II Reverse

Transcriptase according to the manufacturer’s directions (Invitrogen). The
concentration of cDNA was determined spectrophotmetrically (A260) and then diluted
to 10 ng/ul, 1 ng/ul, 100 pg/pl, 10 pg/ul, and 1 pg/ul. The cDNA was then ampliﬁed

by semi-quantitative PCR (30 cycles) using primers speciﬁc to the 16S rDNA gene of

L. jensenii and pi] T from N gonorrhoeae. Since this was not an end point PCR,

119

differences in the amount of product can be visualized and correspond to the amount
of template present.
RESULTS AND DISCUSSION

To date, very few researchers have attempted gene expression studies of co-
cultured organisms (83), limiting the scope of these experiments to organisms which
can be lysed differentially (101). Since we are interested in the response of N
gonorrhoeae to the host enviromnent, which includes indigenous lactobacilli, we have
added to the differential lysis approach a subtractive hybridization step to remove the

Lactobacillus RNA from a mixed sample leaving only pure gonococcal RNA for

 

transcription studies.
Due to the differences in the cell walls of L. jensenii, a Gram-positive
organism, and N gonorrhoeae, a fragile Gram-negative organism, samples containing
gonococci and lactobacilli were gently lysed with lysozyme (10 mg/ml), which
efﬁciently breaks open the gonococci leaving the lactobacilli largely intact. The lysate
was centrifuged to remove unbroken lactobacilli and RNA puriﬁed from the
supernatant using a Qiagen RNeasy kit. However, this puriﬁed RNA was found to
contain a signiﬁcant amount of Lactobacillus RNA, presumably due to a small amount
of lysis in the original sample. To remove the Lactobacillus RNA the mixed RNA
sample was incubated with an excess of genomic DNA ﬁom L. jensenii allowing
RNAzDNA hybrids to form. The sample was then treated with RNase H, which
selectively degrades RNAzDNA hybrids by speciﬁcally hydrolyzing the I

phosphodiester bonds of the RNA hybridized to the DNA (163). A subsequent DNase

120

I step was done to remove the remaining L. jensenii DNA, leaving only pure RNA
from N. gonorrhoeae.

To validate this method, we utilized semi-quantitative, reverse transcription
PCR (semiQ RT-PCR) to determine the amounts of L. jensenii and N gonorrhoeae
RNA present before and after treatments. As seen in Figure A], row B, the RNase H
treatment completely removed the L. jensenii RNA, leaving no template cDNA for
PCR ampliﬁcation. The N gonorrhoeae RNA remained after treatment as
demonstrated by semiQ RT-PCR of the pi] T gene (Figure A], row A). These results
demonstrate that this method can remove unwanted bacterial RNA from a mixed
sample to undetectable levels, yet leaving other RNA in the sample intact.
Applications for this subtractive hybridization method would be to purify bacterial
RNA for microarray or real-time RT PCR analysis of transcriptional changes in one
microbe grown in the presence of another microbe (or microbes), such as pathogens
grown in the presence of organisms of the commensal microbiota.

CONCLUSIONS

A method of subtractive hybridization has been developed to remove the RNA
of one microorganism from a mixed sample leaving puriﬁed RNA from a single
bacterial species. The technique of hybridizing the commensal RNA with genomic
DNA from the same species, followed by selective degradation of the RNA:DNA
hybrid is effective and speciﬁc. This will allow us to expand the applicability of
techniques for analyzing gene expression to include research into bacteria]

interactions.

12]

 

 

A 1234567891011121314

 

 

Figure A.l. Semi-quantitative PCR of L. jensenii 16S rDNA (panel B) and N.
gonorrhoeae pilT (panel A). A. Lane 1: no template control; Lane 2: 1 kb ladder; Lane
3: N gonorrhoeae genomic DNA; Lanes 4-8: 10 ng/ul, 1 ng/ul, 100 pg/ul, 10 pg/ul,
and 1 pg/pl cDNA from N gonorrhoeae; Lanes 9-13: 10 ng/u], 1 ng/pl, 100 pg/ul, 10
pg/ul, and 1 pg/ul cDNA N gonorrhoeae-mixed with L. jensenii which was subjected
to the subtractive hybridization method to remove the L. jensenii RNA before reverse
transcription; Lane 14: 1 kb ladder. B. Lane 1: no template control; Lane 2: 1 kb
ladder; Lane 3: L. jensenii genomic DNA; Lanes 4-8: 10 ng/ul, 1 ng/ul, 100 pg/ul, 10
pg/ul, and 1 pg/ul cDNA from L. jensenii; Lanes 9-13: 10 ng/ul, 1 ng/u], 100 pg/ul,
10 pg/ul, and l pg/ul cDNA N gonorrhoeae-mixed with L. jensenii which was
subjected to the subtractive hybridization method to remove the L. jensenii RNA
before reverse transcription, and Lane 14: 1 kb ladder.

122

REFERENCES

 

123

 

REFERENCES

Adlerberth, 1., S. Ahrne, M. L. Johansson, G. Molin, L. A. Hanson, and A.
E. Wold. 1996. A mannose-speciﬁc adherence mechanism in Lactobacillus
plantarum conferring binding to the human colonic cell line HT-29. Appl

Environ Microbiol 62:2244-2251.

 

Ahrne, S., S. Nobaek, B. Jeppsson, I. Adlerberth, A. E. Wold, and G.
Molin. 1998. The normal Lactobacillus ﬂora of healthy human rectal and oral
mucosa. J Appl Microbiol 85:88-94.

Aleljung, P., W. Shen, B. Rozalska, U. Hellman, A. Ljungh, and T. r-r-i
Wadstrom. 1994. Puriﬁcation of collagen-binding proteins of Lactobacillus
reuteri NCIB 1 1951. Curr Microbiol 28:231-236.

Amit-Romach, E., Z. Uni, and R. Reifen. 2009. Multistep mechanism of
probiotic bacterium, the effect on innate immune system. Mo] Nutr Food Res
54:277-284.

 

Andreu, A., A. E. Stapleton, C. L. Fennel], S. L. Hillier, and W. E. Stamm.
1995. Hemagglutination, adherence, and surface properties of vagina]
Lactobacillus species. J. Inf. Dis. 171:1237-1243.

Antikainen, J ., L. Anton, J. Sillanpaa, and T. K. Korhonen. 2002. Domains
in the S-layer protein CbsA of Lactobacillus crispatus involved in adherence to

collagens, laminin and lipoteichoic acids and in self-assembly. Mol Microbiol
46:381-394.

Antikainen, J., V. Kuparinen, K. Lahteenmaki, and T. K. Korhonen. 2007.
Enolases from Gram-positive bacterial pathogens and commensal lactobacilli

share ﬁinctional similarity in virulence-associated traits. FEMS Irnmunol Med
Microbiol 51:526-534.

Antikainen, J., V. Kuparinen, K. Lahteenmaki, and T. K. Korhonen. 2007.
pH-dependent association of enolase and glyceraldehyde-3-phosphate

dehydrogenase of Lactobacillus crispatus with the cell wall and lipoteichoic
acids. J Bacteriol 189:4539-4543.

Antonio, M. A., S. E. Hawes, and S. L. Hillier. 1999. The identiﬁcation of
vaginal Lactobacillus species and the demographic and microbiologic

characteristics of women colonized by these species. J. Inf. Dis. 180:1950-
1956.

124

10.

ll.

12.

13.

14.

15.

16.

17.

18.

19.

Antonio, M. A., L. K. Rabe, and S. L. Hillier. 2005. Colonization of the
rectum by Lactobacillus species and decreased risk of bacteria] vaginosis. J.
Infect. Dis. 192:394-398.

Aroutcheva, A., D. Gariti, M. Simon, S. Shott, J. Faro, J. A. Simoes, A.
Gurguis, and S. Faro. 2001. Defense factors of vaginal lactobacilli. Am. J.
Obstet. Gynecol. 185:375-379.

Arvidson, C. G., R. Kirkpatrick, M. T. Witkamp, J. A. Larson, C. A.
Schipper, L. S. Waldbeser, P. O'Gaora, M. Cooper, and M. So. 1999.
Neisseria gonorrhoeae mutants altered in toxicity to human fallopian tubes and

molecular characterization of the genetic locus involved. Infect. Immun.
67 :643-652.

Atassi, F., D. Brassart, P. Grob, F. Graf, and A. L. Servin. 2006. In vitro
antibacterial activity of Lactobacillus helveticus strain KS300 against
diarrhoeagenic, uropathogenic and vaginosis-associated bacteria. J. Appl.
Microbiol. 101 :647-654.

 

Atassi, F., D. Brassart, P. Grob, F. Graf, and A. L. Servin. 2006.
Lactobacillus strains isolated from the vagina] microbiota of healthy women

inhibit Prevotella bivia and Gardnerella vaginalis in coculture and cell culture.
FEMS Irnmunol. Med. Microbiol. 48:424-432.

Atassi, F., D. Brassart, P. Grob, F. Graf, and A. L. Servin. 2006. Vaginal
Lactobacillus isolates inhibit uropathogenic Escherichia coli. FEMS
Microbiol. Lett. 257: 132-138.

Babbitt, P. C., M. S. Hasson, J. E. Wedekind, D. R. Palmer, W. C. Barrett,
G. H. Reed, 1. Rayment, D. Ringe, G. L. Kenyon, and J. A. Gerlt. 1996.
The enolase superfarnily: a general strategy for enzyme-catalyzed abstraction
of the alpha-protons of carboxylic acids. Biochemistry 35:16489-1650].

Baeten, J. M., W. M. Hassan, V. Chohan, B. A. Richardson, K. Mandaliya,
J. O. Ndinya-Achola, W. Jaoko, and R. S. McClelland. 2009. Prospective
study of correlates of vaginal Lactobacillus colonisation among high-risk HIV-
] seronegative women. Sex Transm Infect 85:348-353.

Boot, H. J., C. P. Kolen, B. Pot, K. Kersters, and P. H. Pouwels. 1996. The
presence of two S-layer-protein-encoding genes is conserved among species
related to Lactobacillus acidophilus. Microbiology 142 ( Pt 9):2375-2384.

Boris, S., J. E. Suarez, and C. Barbes. 1997. Characterization of the

aggregation promoting factor from Lactobacillus gasseri, a vaginal isolate. J.
Appl. Microbiol. 83:413-420.

125

 

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

3].

Boris, S., J. E. Suarez, F. Vazquez, and C. Barbes. 1998. Adherence of
human vaginal lactobacilli to vaginal epithelial cells and interaction with
uropathogens. Infect. Immun. 66:1985-1989.

Boskey, E. R., K. M. Telsch, K. J. Whaley, T. R. Moench, and R. A. Cone.
1999. Acid production by vaginal ﬂora in vitro is consistent with the rate and
extent of vaginal acidiﬁcation. Infect. Immun. 67:5170-5175.

Boyd, M. A., M. A. Antonio, and S. L. Hillier. 2005. Comparison of AP] 50
CH strips to whole-chromosomal DNA probes for identiﬁcation of
Lactobacillus species. J. Clin. Microbiol. 43:5309-5311.

Bronner, S., H. Monteil, and G. Prevost. 2004. Regulation of virulence
determinants in Staphylococcus aureus: complexity and applications. FEMS
Microbiol Rev 28: 1 83-200.

Burnham, C. A., S. E. Shokoples, and G. J. Tyrrell. 2005. Phosphoglycerate
kinase inhibits epithelial cell invasion by group B streptococci. Microb Pathog
38: 1 89-200.

Burton, J. P., P. A. Cadieux, and G. Reid. 2003. Improved understanding of
the bacterial vaginal microbiota of women before and after probiotic
instillation. Appl. Environ. Microbiol. 69:97-101.

Castaldo, C., V. Vastano, R. A. Siciliano, M. Candela, M. Vici, L.
Muscariello, R. Marasco, and M. Sacco. 2009. Surface displaced alfa-

enolase of Lactobacillus plantarum is a ﬁbronectin binding protein. Microb
Cell Fact 8:14.

CDC. 2009. Sexually Transmitted Disease Surveillance, 2008, p. 1-154. In U.
S. D. o. H. a. H. Services (ed.), MMWR. Atlanta, GA.

CDC. 2008. Subpopulation Estimates from the HIV Incidence Surveillance
System - United States, 2006., p. 985-989. In U. S. D. o. H. a. H. Services
(ed.), vol. 57. CDC, Atlanta, GA.

CDC. 2007. Update to CDC's Sexually transmitted diseases treatment
guidelines, 2006: Fluoroquinolones no longer recommended for treatment of
gonococcal infections. MMWR 56:332-336.

Chan, R. C., G. Reid, R. T. Irvin, A. W. Bruce, and J. W. Costerton. 1985.
Competitive exclusion of uropathogens from human uroepithelial cells by

Lactobacillus whole cells and cell wall fragments. Infect Immun 47 :84-89.

Coconnier, M. H., T. R. Klaenhammer, S. Kerneis, M. F. Bernet, and A.
L. Servin. 1992. Protein-mediated adhesion of Lactobacillus acidophilus

126

 

 

 

32.

33.

34.

35.

36.

37.

38.

39.

40.

41.

BG2FO4 on human enterocyte and mucus-secreting cell lines in culture. Appl
Environ Microbiol 58:2034-2039.

Cohen, M. S., J. R. Black, R. A. Proctor, and P. F. Sparling. 1984. Host
defences and the vaginal mucosa. A re-evaluation. . Scand. J. Urol. Nephrol.
86:13-22.

Criss, A. K., K. A. Kline, and H. S. Seifert. 2005. The frequency and rate of
pilin antigenic variation in Neisseria gonorrhoeae. Mol. Microbiol. 58:510-
559.

Dehio, M., O. G. Gomez-Duarte, C. Dehio, and T. F. Meyer. 1998.
Vitronectin-dependent invasion of epithelial cells by Neisseria gonorrhoeae
involves alpha(v) integrin receptors. FEBS Lett 424:84-88.

Derbise, A., Y. P. Song, S. Parikh, V. A. Fischetti, and V. Pancholi. 2004.
Role of the C-tenninal lysine residues of streptococcal surface enolase in Glu-

and Lys-plasminogen-binding activities of group A streptococci. Infect Immun
72:94-105.

Domino, S. E., E. A. Hurd, K. A. Thomsson, D. M. Karnak, J. M. Holmen
Larsson, E. Thomsson, M. Backstrom, and G. C. Hansson. 2009. Cervical
mucins carry alpha(1,2)fucosylated glycans that partly protect from
experimental vaginal candidiasis. Glycoconj J 26:] 125-1 134.

Donofrio, F. C., A. C. Calil, E. T. Miranda, A. M. Almeida, G. Benard, C.
P. Soares, S. N. Veloso, C. M. Soares, and M. J. Mendes Giannini. 2009.
Enolase from Paracoccidioides brasiliensis: isolation and identiﬁcation as a
ﬁbronectin-binding protein. J Med Microbiol 58:706-713.

Du, Y., and C. G. Arvidson. 2006. prH mediates the expression of some,
but not all, genes induced in Neisseria gonorrhoeae adherent to epithelial cells.
Infect. Immun. 74:2767-2776.

Du, Y., J. Lenz, and C. G. Arvidson. 2005. Global gene expression and the
role of sigma factors in Neisseria gonorrhoeae in interactions with epithelial
cells. Infect. Immun. 73:4834-4845.

Duane, P. G., J. B. Rubins, H. R. Weisel, and E. N. Janoff. 1993.
Identiﬁcation of hydrogen peroxide as a Streptococcus pneumoniae toxin for
rat alveolar epithelial cells. Infect. Immun. 61:4392-4397.

Eberhard, T., R. Virkola, T. Korhonen, G. Kronvall, and M. Ullberg.

1998. Binding to human extracellular matrix by Neisseria meningitidis. Infect
Immun 66: 1791-1794.

127

 

 

 

42.

43.

44.

45.

46.

47.

48.

49.

50.

51.

52.

Edwards, J. L., and M. A. Apicella. 2005. I-domain-containing integrins
serve as pilus receptors for Neisseria gonorrhoeae adherence to human
epithelial cells. Cell Microbiol 7:1197-1211.

Edwards, J. L., and M. A. Apicella. 2002. The role of lipooligosaccharide in
Neisseria gonorrhoeae pathogenesis of cervical epithelia: lipid A serves as a
C3 acceptor molecule. Cell. Microbiol. 4:585-598.

Edwards, J. L., E. J. Brown, K. A. Ault, and M. A. Apicella. 2001. The role
of complement receptor 3 (CR3) in Neisseria gonorrhoeae infection of human
cervical epithelia. Cell Microbiol 3:611-622.

Ekmekci, H., B. Aslim, and S. Ozturk. 2009. Characterization of vaginal
lactobacilli coaggregation ability with Escherichia coli. Microbiol Irnmunol
53:59-65.

Eschenbach, D. A., P. R. Davick, B. L. Williams, S. J. Klebanoff, K.
Young-Smith, C. M. Critchlow, and K. K. Holmes. 1989. Prevalence of
hydrogen peroxide-producing Lactobacillus species in normal women and
women with bacterial vaginosis. J Clin Microbiol 27:251-256.

FAO/WHO. 2001. Evaluation of health and nutritional properties of probiotics
in food including powder milk with live lactic acid bacteria. Report of a Joint
FAO/WHO Expert Consultation.

Felten, A., C. Barreau, C. Bizet, P. H. Lagrange, and A. Phillipon. 1999.
Lactobacillus species identiﬁcation, H202 production, and antibiotic resistance
and correlation with clinical status. J. Clin. Microbiol. 37:729-733.

Foligne, B., S. Nutten, C. Grangette, V. Dennin, D. Goudercourt, S. Poiret,
J. Dewulf, D. Brassart, A. Mercenier, and B. Pot. 2007. Correlation between

in vitro and in vivo immunomodulatory properties of lactic acid bacteria.
World J Gastroenterol 13:236-243.

Freissler, E., A. Meyer auf der Heyde, G. David, T. F. Meyer, and C.
Dehio. 2000. Syndecan-l and syndecan-4 can mediate the invasion of

OpaHSPG-expressing Neisseria gonorrhoeae into epithelial cells. Cell
Microbiol 2:69-82.

Gan, B. S., J. Kim, G. Reid, P. A. Cadieux, and J. Howard. 2002.
Lactobacillusfermentum RC-14 Inhibits Staphylococcus aureus Infection of
Surgical Implants in Rats. J. Infect. Dis. 185:1369-1372.

Godley, M. J. 1985. Quantitation of vaginal discharge in healthy volunteers.
Br J Obstet Gynaecol 92:739—742.

128

 

 

 

53.

54.

55.

56.

57.

58.

59.

60.

61.

62.

63.

Granato, D., F. Perotti, I. Masserey, M. Rouvet, M. Golliard, A. Servin,
and D. Brassart. 1999. Cell surface-associated lipoteichoic acid acts as an
adhesion factor for attachment of Lactobacillusjohnsonii La] to human
enterocyte-like Caco-2 cells. Appl Environ Microbiol 65: 1071-1077.

Greene, J. D., and T. R. Klaenhammer. 1994. Factors involved in adherence
of lactobacilli to human Caco-2 cells. Appl Environ Microbiol 60:4487-4494.

Heinemann, C., J. E. van Hylckama Vlieg, D. B. J anssen, H. J. Busscher,
H. C. van der Mei, and G. Reid. 2000. Purification and characterization of a
surface-binding protein from Lactobacillusfermentum RC-14 that inhibits
adhesion of Enterococcusfaecalis 1131. FEMS Microbiol. Lett. 190:177-180.

Henriksson, A., R. Szewzyk, and P. L. Conway. 1991. Characteristics of the
adhesive determinants of Lactobacillusfermentum 104. Appl Environ
Microbiol 57:499-502.

Huberts, D. H., and I. J. van der Klei. 2010. Moonlighting proteins: An
intriguing mode of multitasking. Biochim Biophys Acta 1803:520-525.

Huffnagle, G. B., and W. S. 2007. The Probiotics Revolution. Bantam Dell,
New York, New York.

Hynonen, U., B. Westerlund-Wikstrom, A. Palva, and T. K. Korhonen.
2002. Identiﬁcation by ﬂagellum display of an epithelial cell- and ﬁbronectin-
binding function in the SlpA surface protein of Lactobacillus brevis. J
Bacteriol 184:3360-3367.

Isbey, S. F., T. M. Alcorn, R. H. Davis, J. Haizlip, P. A. Leone, and M. S.
Cohen. 1997. Characterisation of Neisseria gonorrhoeae in semen during
urethral infection in men. Genitourin. Med. 73:378-3 82.

Jerse, A. E., E. T. Crow, A. N. Bordner, I. Rahman, C. N. Cornelissen, T.
R. Moench, and K. Mehrazar. 2002. Growth of Neisseria gonorrhoeae in the
female mouse genital tract does not require the gonococcal transfenin or

hemoglobin receptors and may be enhanced by commensal lactobacilli. Infect.
Immun. 70:2549-2558.

Johnson-Henry, K. C., K. A. Donato, G. Shen—Tu, M. Gordanpour, and P.
M. Sherman. 2008. Lactobacillus rhamnosus strain GG prevents

enterohemorrhagic Escherichia coli 01 57:H7-induced changes in epithelial
barrier function. Infect Immun 76:1340-1348.

Johnson-Henry, K. C., K. E. Hagen, M. Gordanpour, T. A. Tompkins, and
P. M. Sherman. 2007. Surface-layer protein extracts from Lactobacillus

129

 

 

 

64.

65.

66.

67.

68.

69.

70.

71.

72.

73.

helveticus inhibit enterohaemorrhagic Escherichia coli OlS7:H7 adhesion to
epithelial cells. Cell Microbiol 9:356-367.

Johnson, L. F., and D. A. Lewis. 2008. The effect of genital tract infections
on HIV-1 shedding in the genital tract: a systematic review and meta-analysis.
Sex Transm Dis 35:946-959.

Joiner, K. A., R. Scales, K. A. Warren, M. M. Frank, and P. A. Rice. 1985.
Mechanism of action of blocking immunoglobulin G for Neisseria
gonorrhoeae. Mol. Microbiol. 13:403-416.

Jolodar, A., P. Fischer, S. Bergmann, D. W. Buttner, S. Hammerschmidt,
and N. W. Brattig. 2003. Molecular cloning of an alpha-enolase from the
human ﬁlarial parasite Onchocerca volvulus that binds human plasminogen.

Biochim Biophys Acta 1627:111-120.

 

Juarez Tomas, M. S., V. S. Ocana, B. Wiese, and M. E. Nader-Macias.
2003. Growth and lactic acid production by vaginal Lactobacillus acidophilus
CRL 1259, and inhibition of uropathogenic Escherichia coli. J Med Microbiol
52:1117-1124.

Kaewsrichan, J., K. Peeyananjarassri, and J. Kongprasertkit. 2006.
Selection and identiﬁcation of anaerobic lactobacilli producing inhibitory
compounds against vaginal pathogens. FEMS Immun. Med. Microbiol. 48:75-
83.

Kallstrom, H., D. Blackmer Gill, B. Albiger, M. K. Liszewski, J. P.
Atkinson, and A. B. Jonsson. 2001. Attachment of Neisseria gonorrhoeae to

the cellular pilus receptor CD46: identiﬁcation of domains important for
bacterial adherence. Cell Microbiol 3:133-143.

Kapczynski, D. R., R. J. Meinersmann, and M. D. Lee. 2000. Adherence of
Lactobacillus to intestinal 407 cells in culture correlates with ﬁbronectin
binding. Curr Microbiol 41:136-141.

 

Kaufmann, S. H. 2008. Elie Metchnikoff‘s and Paul Ehrlich's impact on
infection biology. Microbes Infect 10:1417-1419.

Kawai, Y., Y. Ishii, K. Arakawa, K. Uemura, B. Saitoh, J. Nishimura, H.
Kitazawa, Y. Yamazaki, Y. Tateno, T. Itoh, and T. Saito. 2004. Structural

and functional differences in two cyclic bacteriocins with the same sequences
produced by lactobacilli. Appl. Environ. Microbiol. 70:2906-291 1.

Kawai, Y., J. Kusnadi, R. Kemperman, J. Kok, Y. Ito, M. Endo, K.

Arakawa, H. Uchida, J. Nishimura, H. Kitazawa, and T. Saito. 2009. DNA
sequencing and homologous expression of a small peptide confening

130

74.

75.

76.

77.

78.

79.

80.

81.

82.

83.

immunity to gassericin A, a circular bacteriocin produced by Lactobacillus
gasseri LA39. Appl Environ Microbiol 75: 1324-1 330.

Kellogg, D., 1. Cohen, L. Norins, A. Schroeter, and G. Reising. 1968.
Neisseria gonorrhoeae. II. Colonial variation and pathogenicity during 35
months in vitro. J. Bacteriol. 96:596-605.

Kellogg, D. 8., Jr., W. L. Peacock, Jr., W. E. Deacon, L. Brown, and D. I.
Pirkle. 1963. Neisseria gonorrhoeae. I. Virulence genetically linked to clonal
variation. J. Bacteriol. 85:1274-1279.

Kim, Y., S. H. Kim, K. Y. Whang, Y. J. Kim, and S. Oh. 2008. Inhibition of
Escherichia coli 01 57:H7 attachment by interactions between lactic acid
bacteria and intestinal epithelial cells. J Microbiol Biotechnol 18:1278-1285.

Kim, Y. H., C. H. Kim, M. K. Cho, J. H. Na, T. B. Song, and J. S. Oh.
2006. Hydrogen peroxide—producing Lactobacilli in the vaginal ﬂora of

pregnant women with preterm labor with intact membranes. Int J Gynaecol
Obstet 93:22-27.

Kinoshita, H., H. Uchida, Y. Kawai, T. Kawasaki, N. Wakahara, H.
Matsuo, M. Watanabe, H. Kitazawa, S. Ohnuma, K. Miura, A. Horii, and
T. Saito. 2008. Cell surface Lactobacillus plantarum LA 318 glyceraldehyde-
3-phosphate dehydrogenase (GAPDH) adheres to human colonic mucin. J
Appl Microbiol 104:1667-1674.

Kmet, V., and F. Lucchini. 1997. Aggregation-promoting factor in human
vaginal Lactobacillus strains. FEMS Immun. Med. Microbiol. 19:111-114.

Knezevic, A., S. Stepanovic, M. Cupic, D. Jevtovic, J. Ranin, and T.
Jovanovic. 2005. Reduced quantity and hydrogen-peroxide production of
vaginal lactobacilli in HIV positive women. Biomed Pharmacother 59:52]-
523.

Ko, J. S., H. R. Yang, J. Y. Chang, and J. K. Seo. 2007. Lactobacillus
plantarum inhibits epithelia] barrier dysfunction and interleukin-8 secretion
induced by tumor necrosis factor-alpha. . World J. Gastroenterol. 13:1962-
1965.

Kohler, T., A. Buckling, and C. van Delden. 2009. Cooperation and
virulence of clinical Pseudomonas aeruginosa populations. Proc Natl Acad Sci
U S A 106:6339-6344.

Kolenbrander, P. E., R. J. Palmer, J r., A. H. Rickard, N. S. Jakubovics, N.

I. Chalmers, and P. I. Diaz. 2006. Bacterial interactions and successions
during plaque development. Periodontol 2000 42:47-79.

13]

 

 

 

84.

85.

86.

87.

88.

89.

90.

9].

92.

93.

Kupsch, E. M., B. Knepper, T. Kuroki, I. Heuer, and T. F. Meyer. 1993.
Variable opacity (Opa) outer membrane proteins account for the cell tropisms

displayed by Neisseria gonorrhoeae for human leukocytes and epithelial cells.
EMBO J. 12:641-650.

Laughton, J. M., E. Devillard, D. E. Heinrichs, G. Reid, and J. K.
McCormick. 2006. Inhibition of expression of a staphylococcal superantigen-

like protein by a soluble factor from Lactobacillus reuteri. Microbiology
152:1155-1167.

Lebioda, L., B. Stec, J. M. Brewer, and E. Tykarska. 1991. Inhibition of
enolase: the crystal structures of enolase-Ca2(+)- 2-phosphoglycerate and

enolase-Zn2(+)-phosphoglycolate complexes at 2.2-A resolution. Biochemistry
30:2823-2827.

Lebioda, L., E. Zhang, K. Lewinski, and J. M. Brewer. 1993. Fluoride
inhibition of yeast enolase: crystal structure of the enolase-Mg(2+)-F(-)-Pi
complex at 2.6 A resolution. Proteins 16:219-225.

Lin, L., P. Ayala, J. Larson, M. Mulks, M. Fukuda, S. R. Carlsson, C.
Enns, and M. So. 1997. The Neisseria type 2 IgA] protease cleaves LAMP]

and promotes survival of bacteria within epithelial cells. Mol. Microbiol.
24:1083-1094.

Lindgren, S. E., H. E. Swaisgood, V. G. Janolino, L. T. Axelsson, C. S.
Richter, J. M. Mackenzie, and W. J. Dobrogosz. 1992. Binding of

Lactobacillus reuteri to ﬁbronectin immobilized on glass beads. Zentralbl
Bakteriol 277:519-528.

Lopez-Alemany, R., M. Suelves, A. Diaz-Ramos, B. Vidal, and P. Munoz-
Canoves. 2005. Alpha-enolase plasminogen receptor in myogenesis. Front
Biosci 10:30-36.

Lowe, T. L., and S. J. Kraus. 1976. Quantitation of Neisseria gonorrhoeae
from women with gonorrhea. J. Infect. Dis. 133:621-626.

Mack, D. R., S. Ahrne, L. Hyde, S. Wei, and M. A. Hollingsworth. 2003.
Extracellular MUC3 mucin secretion follows adherence of Lactobacillus
strains to intestinal epithelial cells in vitro. Gut 52:827-833.

Mack, D. R., S. Michail, S. Wei, L. McDougal], and M. A. Hollingsworth.

1999. Probiotics inhibit enteropathogenic E. coli adherence in vitro by
inducing intestinal mucin gene expression. Am J Physiol 276:G941-950.

132

 

 

94.

95.

96.

97.

98.

99.

100.

101.

102.

103.

Majhenic, A. C., K. Venema, G. E. Allison, B. B. Matijasi, I. Rogelj, and
T. R. Klaenhammer. 2004. DNA analysis of the genes encoding acidocin
LF221 A and acidocin LF221 B, two bacteriocins produced by Lactobacillus
gasseri LF221. Appl. Microbiol. Biotechnol. 63:705-714.

Makino, S., J. P. van Putten, and T. F. Meyer. 1991. Phase variation of the
opacity outer membrane protein controls invasion by Neisseria gonorrhoeae
into human epithelial cells. EMBO J. 10:1307-1315.

Mandrell, R. E., and M. A. Apicella. 1993. Lipo-oligosaccharides (LOS) of
mucosal pathogens: molecular mimicry and host-modiﬁcation of LOS.
Immunobiology. 187:382-402.

Mao, Y., B. Kasravi, S. Nobaek, L. Q. Wang, D. Adawi, G. Roos, U.
Stenram, G. Molin, S. Bengmark, and B. Jeppsson. 1996. Pectin-
supplemented enteral diet reduces the severity of methotrexate induced
enterocolitis in rats. Scand J Gastroenterol 31:558-567.

Mao, Y., S. Nobaek, B. Kasravi, D. Adawi, U. Stenram, G. Molin, and B.
Jeppsson. 1996. The effects of Lactobacillus strains and oat ﬁber on
methotrexate-induced enterocolitis in rats. Gastroenterology 111:334-344.

Martin, H. L., B. A. Richardson, P. M. Nyange, L. Lavreys, S. L. Hillier,
B. Chohan, K. Mandaliya, J. O. Ndinya-Achola, J. Bwayo, and J. Kreiss.
1999. Vaginal lactobacilli, microbial ﬂora, and risk of human

immunodeﬁciency virus type 1 and sexually transmitted disease acquisition. J.
Infect. Dis. 180:1863-1868.

Martin, R., N. Soberon, M. Vaneechoutte, A. B. Florez, F. Vazquez, and J.
E. Suarez. 2008. Characterization of indigenous vaginal lactobacilli from
healthy women as probiotic candidates. Int Microbiol 11:261-266.

Mashburn, L. M., A. M. Jett, D. R. Akins, and M. Whiteley. 2005 .
Staphylococcus aureus serves as an iron source for Pseudomonas aeruginosa
during in vivo coculture. J Bacteriol 187:554-566.

Mastromarino, P., P. Brigidi, S. Macchia, L. Maggi, F. Pirovano, V.
Trinchieri, U. Conte, and D. Matteuzzi. 2002. Characterization and selection
of vaginal Lactobacillus strains for the preparation of vaginal tablets. J. Appl.
Microbiol. 93:884-893.

Mattar, A. F., D. H. Teitelbaum, R. A. Drongowski, F. Yongyi, C. M.

Harmon, and A. G. Coran. 2002. Probiotics up-regulate MUC-2 mucin gene
expression in a Caco-2 cell-culture model. Pediatr Surg Int 18:586-590.

133

 

 

104.

105.

106.

107.

108.

109.

110.

11].

112.

113.

114.

McGee, Z. A., A. P. Johnson, and D. Taylor-Robinson. 1981. Pathogenic
mechanisms of Neisseria gonorrhoeae: observations on damage to human
Fallopian tubes in organ culture by gonococci of colony type 1 or type 4. J.
Infect. Dis. 143:413-422.

McLean, N. W., and I. J. Rosenstein. 2000. Characterisation and selection of
a Lactobacillus species to re-colonise the vagina of women with recurrent
bacterial vaginosis. J. Med. Microbiol. 49:543-552.

Medellin-Pena, M. J., H. Wang, R. Johnson, S. Anand, and M. W.
Grifﬁths. 2007. Probiotics affect virulence-related gene expression in
Escherichia coli OlS7:H7. Appl. Environ. Microbiol. 73:4259-4267.

Merz, A. J., D. B. Rifenbery, C. G. Arvidson, and M. So. 1996. Traversal of
a polarized epithelium by pathogenic Neisseriae: facilitation by type IV pili
and maintenance of epithelial barrier function. Mol. Med. 2:745-754.

Merz, A. J., and M. So. 2000. Interactions of pathogenic Neisseriae with
epithelial cell membranes. Annu. Rev. Cell. Dev. Biol. 16:423-457.

Meyer, T. F. 1999. Pathogenic neisseriae: complexity of pathogen-host cell
interplay. Clin Infect Dis 28:433-441.

Miyoshi, Y., S. Okada, T. Uchimura, and E. Satoh. 2006. A mucus adhesion
promoting protein, MapA, mediates the adhesion of Lactobacillus reuteri to

Caco-2 human intestinal epithelial cells. Biosci Biotechnol Biochem 70:1622-
1628.

Moorthy, G., M. R. Murali, and S. N. Devaraj. 2009. Lactobacilli facilitate
maintenance of intestinal membrane integrity during Shigella dysenteriae 1
infection in rats. Nutrition 25:350-358.

Mundodi, V., A. S. Kucknoor, and J. F. Alderete. 2008. Irnmunogenic and
plasminogen-binding surface-associated alpha-enolase of T richomonas
vaginalis. Infect Immun 76:523-531.

Neeser, J. R., D. Granato, M. Rouvet, A. Servin, S. Teneberg, and K. A.
Karlsson. 2000. Lactobacillusjohnsonii La] shares carbohydrate-binding

speciﬁcities with several enteropathogenic bacteria. Glycobiology 10:1193-
1199.

Ocana, V. S., A. A. de Ruiz Holgado, and M. E. Nader-Macias. 1999.
Growth inhibition of Staphylococcus aureus by HZOZ-producing Lactobacillus

paracasei subsp. paracasei isolated from the human vagina. FEMS Irnmunol
Med Microbiol 23:87-92.

134

115.

116.

117.

118.

119.

120.

121.

122.

123.

124.

125.

126.

Ocana, V. S., A. A. Pesce de Ruiz Holgado, and M. E. Nader-Macias. 1999.
Selection of vaginal HzOz-generating Lactobacillus species for probiotic use.
Curr Microbiol 38:279-284.

Ocana, V. S., A. A. Pesce De Ruiz Holgado, and M. E. Nader-Macias.
1999. Characterization of a bacteriocin-like substance produced by a vaginal
Lactobacillus salivarius strain. Appl. Environ. Microbiol. 65:5631-5635.

Ohashi, Y., and K. Ushida. 2009. Health-beneﬁcial effects of probiotics: Its
mode of action. Anim Sci J 80:361-371.

Organization, W. H. 2007. Sexually transmitted infections. In W. H.
Organization (ed.), www.who.int.

Osset, J., R. M. Bartolome, E. Garcia, and A. Andreu. 2001. Assessment of
the Capacity of Lactobacillus to Inhibit the Growth of Uropathogens and Block
Their Adhesion to Vagina] Epithelial Cells. J. Infect. Dis. 183:485-491.

Osset, J., E. Garcia, R. M. Bartolome, and A. Andreu. 2001. [Role of
Lactobacillus as protector against vaginal candidiasis]. Med Clin (Barc)
117:285-288.

Owen, D. H., and D. F. Katz. 1999. A vaginal ﬂuid simulant. Contraception
59:91-95.

Paavonen, J. 1983. Physiology and ecology of the vagina. Scand J Infect Dis
Suppl 40:31-35.

Page, K. R., R. D. Moore, B. Wilgus, R. Gindi, and E. J. Erbelding. 2008.
Neisseria gonorrhoeae and Chlamydia trachomatis among human
immunodeﬁciency virus-infected women. Sex Transm Dis 35:859-861.

Pancholi, V., and V. A. Fischetti. 1998. alpha-enolase, a novel strong
plasmin(ogen) binding protein on the surface of pathogenic streptococci. J Biol
Chem 273:14503-14515.

Papagianni, M., N. Avramidis, and G. Filioussis. 2007. High efﬁciency
electrotransforrnation of Lactococcus lactis spp. lactis cells pretreated with
lithium acetate and dithiothreitol. BMC Biotechnol 7:15.

Paruchuri, D. K., H. S. Seifert, R. S. Ajioka, K. A. Karlsson, and M. So.

1990. Identiﬁcation and characterization of a Neisseria gonorrhoeae gene
encoding a glycolipid-binding adhesin. Proc Natl Acad Sci U S A 87:333-337.

135

127.

128.

129.

130.

131.

132.

133.

134.

135.

136.

137.

Pascual, L. M., M. B. Daniele, W. Giordano, M. C. Pajaro, and I. L.
Barberis. 2008. Puriﬁcation and partial characterization of novel bacteriocin
L23 produced by Lactobacillusfermentum L23. Curr Microbiol 56:397-402.

Pascual, L. M., M. B. Daniele, C. Pajaro, and L. Barberis. 2006.
Lactobacillus species isolated from the vagina: identiﬁcation, hydrogen
peroxide production and nonoxynol-9 resistance. Contraception. 73:78-81.

Pavlova, S. I., A. O. Kilic, S. S. Kilic, J. S. So, M. E. Nader-Macias, J. A.
Simoes, and L. Tao. 2002. Genetic diversity of vaginal lactobacilli from

women in different countries based on 16S rRNA gene sequences. J. Appl.
Microbiol. 92:451-459.

Pericone, C. D., K. Overweg, P. W. M. Hermans, and J. N. Weiser. 2000.
Inhibitory and bactericidal effects of hydrogen peroxide production by

Streptococcus pneumoniae on other inhabitants of the upper respiratory tract.
Infect. Immun. 68:3990-3997.

Pick, E., and Y. Keisari. 1980. A simple colorimetric method for the
measurement of hydrogen peroxide produced by cells in culture. J. Irnmunol.
Methods. 38:161-170.

Putaala, H., T. Salusjarvi, M. Nordstronr, M. Saarinen, A. C. Ouwehand,
E. Bech Hansen, and N. Rautonen. 2008. Effect of four probiotic strains and
Escherichia coli OlS7:H7 on tight junction integrity and cyclo-oxygenase
expression. Res Microbiol 159:692-698.

Qin, H., Z. Zhang, X. Hang, and Y. Jiang. 2009. L. plantarum prevents
enteroinvasive Escherichia coli-induced tight junction proteins changes in
intestinal epithelial cells. BMC Microbiol 9:63.

Rabe, L. K., and S. L. Hillier. 2003. Optimization of media for detection of
hydrogen peroxide production by Lactobacillus species. J Clin Microbiol
41:3260-3264.

Redondo-Lopez, V., R. L. Cook, and J. D. Sobel. 1990. Emerging role of
lactobacilli in the control and maintenance of the vaginal bacterial microﬂora.
Rev. Infect. Dis. 12:856-872.

Reid, G. 2001. Probiotic agents to protect the urogenital tract against infection.
Am. J. Clin. Nutr. 73:437S-443S.

Reid, G., C. Heinemann, M. Velraeds, H. C. van der Mei, and H. J.

Busscher. 1999. Biosurfactants produced by Lactobacillus. Methods Enzymol.
310:426-433.

136

138.

139.

140.

141.

142.

143.

144.

145.

146.

147.

148.

149.

Reid, G., A. L. Servin, A. W. Bruce, and H. J. Busscher. 1993. Adhesion of
three Lactobacillus strains to human urinary and intestinal epithelial cells.
Microbios 75:57-65.

Ron, E. Z., and E. Rosenberg. 2001. Natural Roles of Biosurfactants.
Environ. Microbiol. 3:229-236.

Roos, S., P. Aleljung, N. Robert, B. Lee, T. Wadstrom, M. Lindberg, and
H. Jonsson. 1996. A collagen binding protein ﬁom Lactobacillus reuteri is
part of an ABC transporter system? FEMS Microbiol Lett 144:33-3 8.

Rose, M. C. 1992. Mucins: structure, function, and role in pulmonary diseases.
Am J Physiol 263:1.413-429.

Sadreyev, R. 1., S. Shi, D. Baker, and N. V. Grishin. 2009. Structure
similarity measure with penalty for close non-equivalent residues.
Bioinformatics 25: 1259-1263.

Saigh, J. H., C. C. Sanders, and J. Sanders, W. E. 1978. Inhibition of
Neisseria gonorrhoeae by aerobic and facultatively anaerobic components of
the endocervical ﬂora: evidence for a protective effect against infection. Infect.
Immun. 19:704-710.

Sanchez, B., J. M. Schmitter, and M. C. Urdaci. 2009. Identiﬁcation of
novel proteins secreted by Lactobacillus plantarum that bind to mucin and
ﬁbronectin. J Mol Microbiol Biotechnol 17:15 8-162.

Saunders, S., A. Bocking, J. Challis, and G. Reid. 2007. Effect of
Lactobacillus challenge on Gardnerella vaginalis bioﬁlms. Colloids Surf. B.
Biointerfaces. 55:13 8-142.

Sautter, R. L., and W. J. Brown. 1980. Sequential vaginal cultures from
normal young women. J. Clin. Microbiol. 11:479-484.

Schwebke, J. R. 2001. Role of Vaginal Flora As a Barrier to HIV Acquisition.
Curr Infect Dis Rep 3:152-155.

Sega], E., E. Billyard, M. So, S. Storzbach, and T. F. Meyer. 1985. Role of
chromosomal rearrangement in N gonorrhoeae pilus phase variation. Cell.
40:293-300.

Sha, B. E., M. R. Zariffard, Q. J. Wang, H. Y. Chen, J. Bremer, M. H.
Cohen, and G. T. Spear. 2005. Female genital-tract HIV load correlates
inversely with Lactobacillus species but positively with bacterial vaginosis and
Mycoplasma hominis. J. Infect. Dis. 191:25-32.

137

 

 

150.

15].

152.

153.

154.

155.

156.

157.

158.

159.

160.

Sheung, A., A. Rebbapragada, L. Y. Shin, W. Dobson-Belaire, J. Kimani,
E. Ngugi, K. S. MacDonald, J. J. Bwayo, S. Moses, S. Gray-Owen, R.
Kaul, and T. K. H. S. Group. 2008. Mucosal Neisseria gonorrhoeae
coinfection during HIV acquisition is associated with enhanced systemic HIV-
speciﬁc CD8 T-cell responses. . AIDS 22:1729-1737.

Sillanpaa, J., B. Martinez, J. Antikainen, T. Toba, N. Kalkkinen, S.
Tankka, K. Lounatmaa, J. Keranen, M. Hook, B. Westerlund-Wikstrom,
P. H. Pouwels, and T. K. Korhonen. 2000. Characterization of the collagen-
binding S-layer protein CbsA of Lactobacillus crispatus. J Bacteriol 182:6440-
6450.

Smith, H., E. A. Yates, J. A. Cole, and N. J. Parsons. 2001. Lactate
stimulation of gonococcal metabolism in media containing glucose:

mechanism, impact on pathogenicity, and wider implications for other
pathogens. Infect. Immun. 69:6565-6572.

Spence, J. M., J. C. Chen, and V. L. Clark. 1997. A proposed role for the
lutropin receptor in contact-inducible gonococcal invasion of HeclB cells.
Infect. Immun. 65:3736-3742.

Spurbeck, R. R., and C. G. Arvidson. 2008. Inhibition of Neisseria
gonorrhoeae epithelial cell interactions by vaginal Lactobacillus species.
Infect Immun. 76:3124-3130.

Spurbeck, R. R., and C. G. Arvidson. 2010. Lactobacillusjensenii Surface
Associated Proteins Inhibit Neisseria gonorrhoeae Adherence to Epithelial
Cells. Infection & Immunity.

St Amant, D. C., I. E. Valentin-Bon, and A. E. Jerse. 2002. Inhibition of
Neisseria gonorrhoeae by Lactobacillus species that are commonly isolated
from the female genital tract. Infect. Immun. 70:7169-7171.

Strus, M., A. Kucharska, G. Kukla, M. Brzychczy-Wloch, K. Maresz, and
P. B. Heczko. 2005. The in vitro activity of vaginal Lactobacillus with
probiotic properties against Candida. Infect. Dis. Obstet. Gynecol. 13:69-75.

Strus, M., M. Malinowska, and P. B. Heczko. 2002. In vitro antagonistic
effect of Lactobacillus on organisms associated with bacterial vaginosis. J
Reprod Med 47:41-46.

Studier, F. W. 2005. Protein production by auto-induction in high density
shaking cultures. Protein Expr Purif41:207-234.

Swanson, J. 1973. Studies on gonococcus infection. IV. Pili: their role in
attachment of gonococci to tissue culture cells. J. Exp. Med. 137:571-589.

138

161.

162.

163.

164.

165.

166.

167.

168.

169.

170.

Swanson, J., K. Robbins, O. Barrera, D. Corwin, J. Boslego, J. Ciak, M.
Blake, and J. M. Koomey. 1987. Gonococcal pilin variants in experimental
gonorrhea. J. Exp. Med. 165:1344-1357.

Swidsinski, A., W. Mendling, V. Loening-Baucke, A. Ladhoff, S.
Swidsinski, L. P. Hale, and H. Lochs. 2005. Adherent bioﬁlms in bacterial
vaginosis. Obstet. Gynecol. 106:1013-1023.

Tadokoro, T., and S. Kanaya. 2009. Ribonuclease H: molecular diversities,
substrate binding domains, and catalytic mechanism of the prokaryotic
enzymes. FEBS J 276:1482-1493.

Tahara, T., S. Yoshioka, R. Utsumi, and K. Kanatani. 1997. Isolation and
partial characterization of bacteriocins produced by Lactobacillus gasseri J CM
2124. FEMS Microbiol. Lett. 148:97-100.

Terao, Y., M. Yamaguchi, S. Hamada, and S. Kawabata. 2006.
Multifunctional glyceraldehyde-3-phosphate dehydrogenase of Streptococcus

pyogenes is essential for evasion from neutrophils. J Biol Chem 281:14215-
14223.

Toba, T., R. Virkola, B. Westerlund, Y. Bjorkman, J. Sillanpaa, T. Vartio,
N. Kalkkinen, and T. K. Korhonen. 1995. A Collagen-Binding S-Layer
Protein in Lactobacillus crispatus. Appl Environ Microbiol 61:2467-2471.

Tomas, M. S., E. Bru, and M. E. Nader-Macias. 2003. Comparison of the
growth and hydrogen peroxide production by vaginal probiotic lactobacilli
under different culture conditions. Am J Obstet Gynecol 188:35-44.

Tunio, S. A., N. J. Oldﬁeld, A. Berry, D. A. Ala'aldeen, K. G. Wooldridge,
and D. P. Turner. 2010. The moonlighting protein ﬁ'uctose-l, 6-bisphosphate
aldolase of Neisseria meningitidis: surface localization and role in host cell
adhesion. Mol Microbiol.

Uicker, W. C., L. Schaefer, and R. A. Britton. 2005. The essential GTPase
RbgA (quF) is required for SOS ribosome assembly in Bacillus subtilis. Mol.
Microbiol. 59:528-540.

Valdez, J. C., M. C. Peral, M. Rachid, M. Santana, and P. G. 2005.
Interference of Lactobacillus plantarum with Pseudomonas aeruginosa in vitro
and in infected burns: the potential use of probiotics in wound treatment. Clin.
Microbiol. Infect. 11:472-479.

139

17].

172.

173.

174.

175.

176.

177.

178.

179.

180.

Valeur, N., P. Engel, N. Carbajal, E. Connolly, and K. Ladefoged. 2004.
Colonization and immunomodulation by Lactobacillus reuteri ATCC 55730 in
the hrunan gastrointestinal tract. Appl Environ Microbiol 70:] 176-1181.

Vallor, A. C., M. A. Antonio, S. E. Hawes, and S. L. Hillier. 2001. Factors
associated with acquisition of, or persistent colonization by, vaginal
lactobacilli: role of hydrogen peroxide production. J. Infect. Dis. 184: 1431-
1436.

van Putten, J. P., T. D. Duensing, and J. Carlson. 1998. Gonococcal
invasion of epithelial cells driven by P.IA, a bacterial ion channel with GTP
binding properties. J. Exp. Med. 188:941—952.

van Putten, J. P., T. D. Duensing, and R. L. Cole. 1998. Entry of OpaA+
gonococci into HEp-2 cells requires concerted action of glycosaminoglycans,
ﬁbronectin and integrin receptors. Mol Microbiol 29:369-379.

Velraeds, M. M., B. van de Belt-Gritter, H. J. Busscher, G. Reid, and H.
C. van der Mei. 2000. Inhibition of uropathogenic bioﬁlm grth on silicone

rubber in human urine by lactobacilli--a teleologic approach. World J. Urol.
18:422-426.

Velraeds, M. M., H. C. van der’ Mei, G. Reid, and H. J. Busscher. 1997.
Inhibition of initial adhesion of uropathogenic Enterococcusfaecalis to solid

substrata by an adsorbed biosurfactant layer from Lactobacillus acidophilus.
Urology. 49:790-794.

Velraeds, M. M. C., H. C. van der Mei, G. Reid, and H. J. Busscher. 1996.
Inhibition of Initial Adhesion of Uropathogenic Enterococcusfaecalis by

Biosurfactants from Lactobacillus Isolates. App]. Env. Microbiol. 62: 1958-
1 963.

Ventura, M., I. J ankovic, D. C. Walker, R. D. Pridmore, and R. Zink.
2002. Identiﬁcation and characterization of novel surface proteins in
Lactobacillusjohnsonii and Lactobacillus gasseri. Appl. Environ. Microbiol.

68:6172-6181.

Vielfort, K., H. Sjolinder, S. Roos, H. Jonsson, and H. Am. 2008.
Adherence of clinically isolated lactobacilli to human cervical cells in
competition with Neisseria gonorrhoeae. Microbes Infect. 10:1325-1334.

Vinderola, C. G., M. Medici, and G. Perdigon. 2004. Relationship between
interaction sites in the gut, hydrophobicity, mucosa] immunomodulating

capacities and cell wall protein proﬁles in indigenous and exogenous bacteria.
J Appl Microbiol 96:230-243.

140

181.

182.

183.

184.

185.

186.

187.

188.

189.

I90.

19].

Wagner, G., and R. J. Levin. 1980. Electrolytes in vaginal ﬂuid during the
menstrual cycle of coitally active and inactive women. J Reprod Fertil 60:17-
27.

Waldbeser, L. S., R. S. Ajioka, A. J. Merz, D. Puaoi, L. Lin, M. Thomas,
and M. So. 1994. The opaH locus of Neisseria gonorrhoeae MS] 1A is
involved in epithelial cell invasion. Mol. Microbiol. 13:919-928.

Walencka, E., S. Rozalska, B. Sadowska, and B. Rozalska. 2008. The
inﬂuence of Lactobacillus acidophilus-derived surfactants on staphylococcal
adhesion and bioﬁlm formation. F olia Microbiol (Praha) 53:61-66.

Westrom, L., and D. Eschenbach. 1999. Pelvic Inﬂammatory Disease, p.
789-809. In K. K. Holmes, Sparling, P. F., Mardh, P., Lemon, S. M., Stamm,
W.E., Piot, P., and Wasserheit, J. N. (ed.), Sexually Transmitted Diseases, 3rd
ed. McGraw-Hill, New York.

Whitehead, K., J. Versalovic, S. Roos, and R. A. Britton. 2008. Genomic
and genetic characterization of the bile stress response of probiotic
Lactobacillus reuteri ATCC 55730. Appl Environ Microbiol 74: 1 812-1 8 19.

Wierenga, R. K. 2001. The TIM-barrel fold: a versatile framework for
efﬁcient enzymes. FEBS Lett 492:193-198.

Wiesenfeld, H. C., S. L. Hillier, M. A. Krohn, D. V. Landers, and R. L.
Sweet. 2003. Bacterial vaginosis is a strong predictor of Neisseria

gonorrhoeae and Chlamydia trachomatis infection. Clin. Infect. Dis. 36:663-
668.

Wilks, M., R. Wiggins, A. Whiley, E. Hennessy, S. Warwick, H. Porter, A.
Corﬁeld, and M. Millar. 2004. Identiﬁcation and H202 production of vaginal

lactobacilli from pregnant women at high risk of preterm birth and relation
with outcome. J Clin Microbiol 42:713-717.

Wilson, M. 2005. Microbial Inhabitants of Humans: Their Ecology and Role
in Health and Disease, lst ed. The Press Syndicate of the University of
Cambridge, Cambridge.

Young, H., S. K. Saraﬁan, A. B. Harris, and A. McMillan. 1983. Non-
cultural detection of Neisseria gonorrhoeae in cervical and vaginal washings.
J. Med. Microbiol. 16:183-191.

Zarate, G., and M. E. Nader-Macias. 2006. Inﬂuence of probiotic vaginal

lactobacilli on in vitro adhesion of urogenital pathogens to vaginal epithelial
cells. Lett. Appl. Microbiol. 43:174-180.

14]

192.

193.

194.

195.

Zarate, G., V. Santos, and M. E. Nader-Macias. 2009. Protective effect of
vaginal Lactobacillus paracasei CRL 1289 against urogenital infection

produced by Staphylococcus aureus in a mouse animal model. Infect Dis
Obstet Gynecol 2009:48358.

Zheng, H. Y., T. M. Alcorn, and M. S. Cohen. 1994. Effects of H202-
producing lactobacilli on Neisseria gonorrhoeae growth and catalase activity.
J. Infect. Dis. 170:1209-1215.

Zhou, X., S. J. Bent, M. G. Schneider, C. C. Davis, M. R. Islam, and L. J.
Forney. 2004. Characterization of vaginal microbial communities in adult

healthy women using cultivation-independent methods. Microbiology.
150:2565-2573.

Zhou, X., C. Brown, J., Z. Abdo, C. C. Davis, M. A. Hansmann, P. Joyce,
J. A. Foster, and L. Forney. 2007. Differences in the composition of vaginal

microbial communities found in healthy Caucasian and black women. ISME J.

1:121-133.

142

.I