Page 1
1
1
2
The Haemophilus ducreyi Fis Protein Is Involved In Controlling 3
4
Expression of the lspB-lspA2 Operon and Other Virulence Factors 5
6
7
8
9
10
Maria Labandeira-Rey1, Dana A. Dodd
1, Chad A. Brautigam
2, Kate R. Fortney
3, 11
Stanley M. Spinola3-6
, and Eric J. Hansen1*
12
13
14
15
Departments of Microbiology1 and Biochemistry
2, University of Texas 16
17
Southwestern Medical Center, Dallas, TX 75390, and Departments of Microbiology and 18
19
Immunology3, Medicine
4, Pathology and Laboratory Medicine
5, and the Center 20
21
for Immunobiology6 , Indiana University, Indianapolis, Indiana 46202 22
23
24
25
26
Running Head: H. ducreyi Fis Protein 27
28
29
30
31
*Corresponding Author: 32
33
Eric J. Hansen, Ph.D. 34
Department of Microbiology 35
University of Texas Southwestern Medical Center 36
5323 Harry Hines Boulevard 37
Dallas, TX 75390-9048 38
39
Telephone: 214-633-1386 40
FAX: 214-648-5905 41
E-mail:[email protected] 42
43
IAI Accepts, published online ahead of print on 26 August 2013Infect. Immun. doi:10.1128/IAI.00714-13Copyright © 2013, American Society for Microbiology. All Rights Reserved.
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 2
2
ABSTRACT 44
45
Expression of the lspB-lspA2 operon encoding a virulence-related two-partner secretion 46
system in Haemophilus ducreyi 35000HP is directly regulated by the CpxRA regulatory system 47
[Labandeira-Rey M, Mock JR, Hansen E. Regulation of expression of the Haemophilus ducreyi 48
LspB and LspA2 proteins by CpxR. Infect. Immun. 77: 3402-3411 (2009)]. In the present 49
study, we show that this secretion system is also regulated by the small nucleoid-associated 50
protein Fis. Inactivation of the H. ducreyi fis gene resulted in a reduction in expression of both 51
the H. ducreyi LspB and LspA2 proteins. DNA microarray experiments showed that a H. 52
ducreyi fis deletion mutant exhibited altered expression of genes encoding other important H. 53
ducreyi virulence factors including DsrA and Flp1, suggesting a possible global role for Fis in 54
control of virulence in this obligate human pathogen. While the H. ducreyi Fis protein has a high 55
degree of sequence and structural similarity to the Fis proteins of other bacteria, its temporal 56
pattern of expression was very different from that of enterobacterial Fis proteins. The use of a 57
lacZ-based transcriptional reporter provided evidence which indicated that the H. ducreyi Fis 58
homolog is a positive regulator of gyrB, a gene that is negatively regulated by Fis in enteric 59
bacteria. Taken together, the Fis protein expression data and the observed regulatory effects of 60
Fis in H. ducreyi suggest that this small DNA binding protein has a regulatory role in H. ducreyi 61
which may differ in substantial ways from that of other Fis proteins. 62
63
64
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 3
3
INTRODUCTION 65
66
Haemophilus ducreyi is the Gram-negative pathogen responsible for the sexually 67
transmitted disease chancroid (1, 2). Information about the pathogenesis of this genital ulcer 68
disease remains fairly limited, despite the fact that chancroid is endemic in some developing 69
countries in Africa, Asia, and South America (2). Studies in sub-Saharan Africa provided 70
evidence that chancroid can be a cofactor for human immunodeficiency virus acquisition and 71
transmission (3, 4). In the United States, H. ducreyi infections are very rare and typically occur 72
only in isolated cases that are commonly associated with the sex trade industry (5, 6). 73
74
Similar to the paucity of knowledge about the pathogenesis of chancroid, the specifics 75
about which H. ducreyi gene products are responsible or essential for the expression of virulence 76
are only partly defined. Nevertheless, the introduction of the human challenge model for 77
experimental chancroid (7) in 1994 made possible the direct testing of H. ducreyi wild-78
type/mutant pairs in a well-controlled manner in a most appropriate system (8). In the 79
subsequent two decades, a significant number of H. ducreyi mutants were found to be fully 80
virulent, partially attenuated, or substantially deficient in virulence in this model system (8-15). 81
Among these, a mutant lacking the ability to express both the LspA1 and LspA2 proteins was 82
found to be very attenuated (16). These two very large H. ducreyi proteins are both secreted by 83
the LspB outer membrane protein, with all three proteins comprising a two-partner secretion 84
system (17). Expression of either LspA1 or LspA2 has been shown to be necessary for H. 85
ducreyi to inhibit phagocytosis by macrophages and other phagocytic cell lines in vitro (18) via a 86
mechanism that involves inhibition of Src family protein tyrosine kinases (19). Interestingly, the 87
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 4
4
LspA proteins themselves contain multiple specific motifs that can be tyrosine-phosphorylated 88
by macrophages (20). 89
90
The genetic basis for control of virulence expression by H. ducreyi remains largely 91
unexplored, with H. ducreyi regulatory systems having received scant attention. In fact, to date 92
only five reports addressed this issue at all. The first involved mutant analysis of genes encoding 93
proteins involved in the utilization of hemoglobin (21), whereas the second used a transcript 94
capture method to identify a large number of H. ducreyi genes expressed in human volunteers 95
after experimental challenge (22). Evidence for in vivo expression of both LspA1 and LspA2 96
was obtained in the latter study. It has also been established that the CpxRA two-component 97
signal transduction system negatively regulates expression of the lspB-lspA2 operon, as well as 98
other ORFs proven to be important in the human challenge model of chancroid (23, 24). 99
Interestingly, deletion of the mtrC gene encoding a protein involved in resistance to 100
antimicrobial peptides resulted in activation of the CpxRA regulon (25). Most recently, it was 101
reported that inactivation of the H. ducreyi carbon storage regulator A (CsrA) gene resulted in 102
multiple changes in gene expression involving virulent determinants (14). 103
104
A recent report indicated that the small nucleoid-associated protein Fis of Pasteurella 105
multocida was involved in controlling the expression of several important virulence factors 106
including a two-partner secretion system composed of LspB_2 and PfhB_2 (26). These two P. 107
multocida proteins have homology to the H. ducreyi LspB and LspA1/LspA2 proteins. In the 108
present study, we investigated the potential involvement of Fis in the regulation of expression of 109
the H. ducreyi LspB-LspA1/LspA2 system, and determined that inactivation of the H. ducreyi fis 110
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 5
5
gene resulted in decreased expression of both LspB and LspA2. We also show that Fis is 111
involved in controlling the expression of other proven H. ducreyi virulence factors. 112
113
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 6
6
MATERIALS AND METHODS 114
115
Bacterial strains, plasmids, and culture conditions. Bacterial strains and plasmids 116
used in this study are listed in Table 1. H. ducreyi strains were routinely grown on chocolate 117
agar (CA) as previously described (23). When kanamycin selection was necessary, H. ducreyi 118
was grown on a GC agar-based medium (27). For broth culture, strains were incubated in a 119
Columbia broth-based (CB) medium at 33°C in a gyratory water bath at 100 rpm as described 120
(23). Escherichia coli XL10-Gold and DH5g were used as hosts for general cloning 121
manipulations and protein expression, and were grown in Luria-Bertani medium supplemented
122
with ampicillin (100 µg/ml), kanamycin (30 µg/ml), or chloramphenicol (30 µg/ml) when 123
appropriate for maintenance of plasmids. Before complementation of H. ducreyi deletion 124
mutants, plasmid constructs were transformed into and isolated from E. coli HB101. 125
126
Tissue culture cells and media. The human foreskin fibroblast cell line Hs27 (CRL-127
1634) was obtained from the American Type Culture Collection (Manassas, VA). Hs27 cells 128
were cultivated in DMEM (Fisher Scientific Co., Pittsburgh, PA) supplemented with 2 mM 129
GlutaMAX (GIBCO-BRL, Rockville, MD), 10 mM HEPES, 1 mM sodium pyruvate, and 10% 130
(vol/vol) heat-inactivated fetal bovine serum at 37°C in a humidified incubator containing an 131
atmosphere of 95% air-5% CO2. 132
133
Purification of a GST-tagged Fis fusion protein and development of a polyclonal Fis 134
antibody. The complete fis ORF was amplified from H. ducreyi 35000HP chromosomal DNA 135
by PCR using primers HD547 and HD548 (Table 2) which added BamHI and SmaI sites to the 5' 136
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 7
7
and 3' ends of the fragment, respectively. The resulting amplicon was BamHI- and SmaI- 137
digested and ligated to pGEX-T4-2 (GE Healthcare, Pittsburgh, PA) cut with the same enzymes 138
to obtain pML165. E. coli XL10-Gold cells containing pML165 were cultured to mid-139
exponential phase (OD600~0.5) and, after the addition of isopropyl く-D-1-thiogalactopyranoside 140
(IPTG) to a final concentration of 0.1 mM, the cultures were further incubated for 4 h. After 141
induction, the cells were harvested by centrifugation, suspended in buffer A [50 mM Tris (pH 142
8.0), 1 mM EDTA, and 25% (wt/vol) sucrose)] containing protease inhibitors (Sigma-Aldrich, 143
St. Louis, MO) and disrupted by sonication. The sonicated mixture was centrifuged (15,000 x g 144
for 30 min) and the resultant pellet was suspended in a solution containing 20 mM Tris (pH 8.0), 145
0.2 M NaCl, 1% (wt/vol) sodium deoxycholate, and 2 mM ethylene glycol tetra-acetic acid 146
(EGTA). After incubation at room temperature for 30 min, this mixture was centrifuged at 147
10,000 x g for 20 min. The resultant supernatant was applied to glutathione beads (GE 148
Healthcare), the beads were washed with Buffer B [50 mM Tris (pH 8.0), 5 mM EDTA, 50 mM 149
NaCl, and 5 mM く-mercaptoethanol], and the bound protein was eluted with Buffer C [50 mM 150
Tris (pH 8.0), 5 mM EDTA, 150 mM NaCl, 5 mM く-mercaptoethanol, and 10 mM glutathione]. 151
After dialysis against a solution of 50 mM Tris (pH 8.0) and 100 mM NaCl, the fusion protein 152
was stored at -70°C with the addition of glycerol to a final concentration of 20% (wt/vol). The 153
use of a commercial facility for production of mouse antiserum was approved by the Institutional 154
Animal Care and Use Committee at UT Southwestern Medical Center. Polyclonal mouse 155
antibody to this GST-fusion protein was produced by Rockland Immunochemicals (Boyertown, 156
PA). 157
158
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 8
8
Western blot analysis. Whole cell lysate preparations (~5x107 cells/lane) were resolved 159
by SDS-PAGE in 4-20% (wt/vol) polyacrylamide separating gels and transferred onto PVDF 160
membranes (Millipore, Billerica, MA). Membranes were incubated in StartingBlock (PBS) 161
Blocking Buffer (Thermo Scientific, Rockford, IL) containing 5% (vol/vol) normal goat serum 162
for 1 h at room temperature or overnight at 4°C. The membranes were then incubated for 3-4 h 163
at room temperature or overnight at 4°C in primary antibody at the appropriate dilution, followed 164
by a 1 h incubation at room temperature in a 1:20,000 dilution of either goat anti-mouse IgG-165
HRP or goat anti-rabbit IgG-HRP (BioRad, Hercules, CA). The LspA1-specific monoclonal 166
antibody (MAb) 40A4 (28), the LspA2-specific MAb 1H9 (28), the mouse polyclonal LspB 167
antibody (17), the PAL-specific MAb 3B9 (29), the mouse polyclonal DsrA-reactive antibody 168
(30), the rabbit polyclonal Flp1 antibody (31), and the rabbit polyclonal CpxR-reactive antibody 169
(23) have been described. Western blots were developed using the Western Lightning 170
Chemiluminescence Reagent Plus (New England Nuclear, Boston, MA). 171
172
Construction and complementation of H. ducreyi fis deletion mutants. A ~1-kb 173
fragment corresponding to the 5' upstream region of the H. ducreyi fis ORF was amplified from 174
chromosomal DNA with ExTaq DNA polymerase (Takara Bio Inc., Shiga, Japan) and primers 175
HD524 and HD526 (Table 2). Another ~500-bp fragment corresponding to the 3' downstream 176
region of the H. ducreyi fis ORF was amplified with primers
HD527 and HD525. A cat cartridge 177
from pSL33 (32), modified to contain its native promoter (23), was amplified from pML122 178
(Table 1) using primers HD528 and HD529. Primer HD528 shared a 21-nt complementary 179
sequence with primer HD526 (in bold, Table 2), and primer HD529 shared
a 21-nt 180
complementary sequence with primer HD527 (in bold, Table 2). The three PCR fragments were 181
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 9
9
gel-purified and equal amounts were mixed and used as the templates in overlapping extension 182
PCR (33) with primers HD524 and HD525. The resultant ~2.3-kb PCR product was subjected to 183
restriction enzyme digestion with DpnI, gel-purified, and a 100 たg quantity was used to 184
electroporate H. ducreyi 35000HP as previously described (11, 34). A fis deletion mutant 185
(35000HP〉fis) was selected on CA plates containing chloramphenicol (1 µg/ml); nucleotide 186
sequence analysis confirmed the non-polar nature of this construct. 187
188
A H. ducreyi 〉cpxR〉fis mutant was also constructed in which a kan cartridge was used 189
in place of the cat cartridge in fis. A modified kan cartridge containing its native promoter (35) 190
was cloned into pCR2.1 to obtain pML168 (Table 1). This kan construct was amplified using 191
primers HD809 and HD810. Primer HD809 shared a 21-nt complementary sequence with primer 192
HD526 (in bold, Table 2), and primer HD810 shared a 21-nt complementary sequence with 193
primer HD527 (in bold, Table 2). This amplicon, together with the upstream and downstream 194
sequences described above for the original fis mutant, were used as templates in overlapping 195
extension PCR (33) with primers HD524 and HD525. The resultant amplicon was digested with 196
DpnI, gel-purified, and a 100 たg quantity was used to electroporate H. ducreyi 35000HP〉cpxR 197
as previously described (11, 34). A 35000HP 〉cpxR〉fis double mutant was selected on GC 198
plates containing kanamycin (30 µg/ml); nucleotide sequence analysis confirmed the non-polar 199
nature of the insertion in fis. 200
201
The wild-type fis gene from H. ducreyi together with ~300-nt 5' of the ATG translation 202
start codon was amplified from chromosomal DNA using primers HD543 and HD544 (Table 2). 203
The amplicon was digested with XhoI and ligated to XhoI-digested pACYC177 (NEB) to obtain 204
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 10
10
pML164. A 100 ng quantity of pML164 DNA was used to transform 35000HP〉fis (as described 205
above) to obtain the kanamycin- and chloramphenicol-resistant strain 35000HP〉fis(pML164). 206
The 35000HP〉fis mutant was also transformed with 100 ng of pACYC177 to be used as a 207
vector-only control. 208
209
Phagocytosis assay. Opsonization of latex beads with human IgG was accomplished 210
essentially as described (12) except that the antibody coating step was allowed to proceed 211
overnight. The next day, the beads were washed three times with PBS, resuspended in 500 たl 212
PBS, and incubated with 5 たl fluorescent Cy3 donkey anti-human antibody (Jackson 213
ImmunoResearch, West Grove, PA) for 1 h at room temperature with gentle agitation. After 214
several washes with PBS, the opsonized beads were resuspended in 500 たl of DMEM containing 215
10% fetal bovine serum (DMEM-S). The day before the phagocytosis experiment, J774A.1 cells 216
were plated at a density of 1.7 x 105 cells/well in chamber slides. The next day, these 217
monolayers were washed once with DMEM-S and then 500 たl DMEM-S was added, followed 218
by the addition of 200 たl of bacterial cell suspension (OD600 = 0.25-0.5). After centrifugation at 219
200 x g for 5 min at room temperature, the slides were incubated at 33°C with 95% air-5% CO2 220
for 1 h prior to the addition of a 10 たl volume of opsonized beads. After addition of the beads, 221
the slides were subjected to centrifugation at 300 x g for 1 min and then the slides were placed in 222
a 37°C incubator with 90% air -10% CO2 for 5 min to allow for attachment of the beads to the 223
phagocytes. Each chamber was then washed twice with 500 たl DMEM-S. A final 500 たl 224
volume of DMEM-S was added to each chamber and the slides were incubated at 37°C for 10 225
min to allow for ingestion of the beads. Each slide was then placed on ice and washed with cold 226
DMEM-S. Incubation with fluorescent Cy2 donkey anti-human antibody and nuclear staining 227
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 11
11
with Hoechst 33342 was carried out essentially as described (12). Total beads (external and 228
ingested) stained with Cy3 (red) and external beads stained with Cy2 (green) were counted and 229
used to calculate the phagocytosis index via the formula: (total beads-external beads/total 230
number of J774A.1 cells). 231
232
RNA isolation, DNA microarray analysis, and real-time RT-PCR. Total RNA was 233
extracted from broth-grown H. ducreyi bacteria and used for DNA microarray analysis as 234
previously described (23, 34). Briefly, 5 たg of total RNA extracted from 35000HP or 235
35000HPÄfis cells grown in CB medium to mid-exponential phase (~8 h of growth) were used to 236
obtain aminoallyl-cDNA which was labeled post-transcriptionally with Cy3 or Cy5 as described 237
(23, 34). Each experimental replicate was subjected to reverse labeling (i.e., a dye swap) to 238
avoid dye bias. Differential expression was defined as a minimum of a two-fold change in 239
expression in the H. ducreyi fis deletion mutant relative to wild-type H. ducreyi 35000HP. The 240
final results only include expression profiles that had a P ø 0.05 after a one-sample t-test 241
analysis. The raw data from these experiments were deposited at the NCBI Gene Expression 242
Omnibus (GEO) database (http://www.ncbi.nlm.nih.gov/geo/) under accession number 243
GSE44413. From the final DNA microarray results data, 24 genes were randomly selected for 244
further confirmation of their relative transcription levels by two-step real-time RT-PCR. 245
Oligonucleotide primers used in this study are listed in Table 2 and real-time RT-PCR assays 246
were performed as described (23, 34) on three independent biological replicates, using HD0084 247
(ldh) to normalize the amount of cDNA per sample. The fold-change of each gene was 248
calculated using the 2
–〉〉CT method. 249
250
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 12
12
Bactericidal assay. Bactericidal assays were performed with normal human serum 251
(NHS) obtained from a single healthy donor, exactly as described previously (13). In brief, we 252
compared the survival in 50% NHS of plate-grown 35000HP, its isogenic fis, cpxA, and dsrA 253
mutants, 35000HPÄfis(pACYC177), and 35000HPÄfis(pML164). Data were reported as percent 254
survival in active NHS compared to heat-inactivated serum [(geometric mean CFU in active 255
NHS/geometric mean CFU in heat-inactivated NHS) x 100]. Each experiment was repeated five 256
times; the arithmetic mean and standard deviation (SD) of the percent survival were calculated. 257
Comparison of the strains was performed using a mixed model ANOVA with experiment as the 258
random effect. P values for pairwise comparisons were calculated using the Tukey method of 259
adjustment; an adjusted P < 0.05 was considered significant for these assays. 260
261
Microcolony formation assay. Microcolony assays were performed as previously 262
described (31). Briefly, 24-well tissue culture plates (Costar, Corning, N.Y.) were seeded with 263
105 Hs27 human foreskin fibroblasts per well and incubated for 3 days until they achieved 264
confluency. H. ducreyi cells grown overnight on CA plates were suspended in tissue culture 265
medium to OD600 = 0.1. Portions (5 たl) of the bacterial suspension were added in triplicate to 266
individual wells and the bacterial cells were centrifuged onto the confluent monolayers for 5 min 267
at 1000 x g at room temperature, after which the plates were incubated for 24 h at 33°C in 95% 268
air-5% CO2. After this incubation, each well was washed three times with PBS (pH 7.4) and 269
stained with crystal violet. Images were recorded using a FSX100 BioImager Navigator 270
(Olympus, Center Valley, PA) at 14X magnification. 271
272
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 13
13
Construction of a LacZ-based transcriptional reporter plasmid for H. ducreyi. The 273
LacZ-based transcriptional reporter plasmid pASE222 (36) was digested using restriction 274
enzyme BamHI. The resultant 4.1 kb fragment was filled in using the Klenow fragment (New 275
England Biolabs, Ipswich, MA), and ligated to a 3.2 kb BamHI-ScaI fragment from pACYC177 276
(New England Biolabs) containing the origin of replication and the kan gene; the resultant 277
plasmid was designated pML303. A set of 600-bp fragments including 500-bp upstream and 278
100-bp downstream of the translation start codon of lspB (HD_1155), dsrA (HD_0769), gyrB 279
(HD_1643), flp1 (HD_1312), and ompP2B (HD_1435) were PCR-amplified with NotI and SalI 280
sites (see Table 2 for primer sequences), and subcloned into pGEM-T (Promega, Madison, WI). 281
Promoter fragments were excised from pGEM-T using NotI and SalI, and individually cloned 282
into NotI- and SalI-digested pML303, resulting in plasmids pML306, pML308, pML309, 283
pML312, and pML314, respectively (Table 1). All constructs were sequenced prior to their 284
introduction by electroporation into H. ducreyi strains. 285
286
LacZ-based transcriptional reporter assay. く-galactosidase assays were performed as 287
previously described (37) with minor adjustments. Briefly, H. ducreyi strains carrying the 288
reporter constructs were grown overnight on CA plates. A 5 ml volume of CB medium was 289
inoculated from the fresh CA plates and the cultures were incubated at 33°C with gentle shaking 290
(100 rpm). After overnight growth, cells from 1 ml of the culture were collected and re-291
suspended in 1 ml of a modified Z-buffer [500 mM Na2HPO4, 4 M KCl, 1M MgSO4, 1% 292
(wt/vol) cetyltrimethylammonium bromide (CTAB), and 1% (wt/vol) sodium deoxycholate]. 293
Because H. ducreyi strains tend to auto-aggregate, the protein content of the bacterial 294
suspensions was used to standardize the assay. Briefly, 5 たl of the bacterial suspensions in 295
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 14
14
modified Z-buffer were added to the DCTM
Protein Assay reagents (Bio-Rad) per the 296
manufacturer’s protocol, the OD750 of the sample was determined using a microplate 297
spectrophotometer (Biotek, Winooski, VT), and the values were used to standardize the く-298
galactosidase assay. Prior to the addition of o-nitrophenyl-く-D-galactoside (ONPG), く-299
mercaptoethanol (5.4 たl/ml) was added to the bacterial suspensions. The time of ONPG addition 300
was recorded, the cells were incubated at 37°C for 15 min and the reactions stopped by the 301
addition of 0.5 ml of 1 M Na2CO3. The OD420 and OD550 were recorded and the Miller units 302
were calculated as described (37) except that total protein content was used in place of the 303
OD600 values. 304
305
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 15
15
RESULTS 306
307
Construction and characterization of a H. ducreyi Fis mutant. Recent work with P. 308
multocida (26) revealed that the small nucleoid-associated protein Fis was involved in 309
controlling the expression of several important virulence factors including a two-partner 310
secretion system composed of LspB_2 and PfhB_2. This secretion system has homology to the 311
H. ducreyi LspB-LspA1/LspA2 system, previously shown to be an important H. ducreyi 312
virulence factor (16, 19, 38). A non-polar fis deletion mutant (Fig. 1A) was constructed (as 313
described in Materials and Methods) to allow determination of whether expression of the Lsp 314
protein system in H. ducreyi involved Fis. 315
316
Similar to results obtained with other bacteria (39, 40), deletion of fis in wild-type H. 317
ducreyi 35000HP resulted in a significant growth defect that was apparent during mid-318
exponential growth (Fig. 1B, p<0.001). In addition to this broth growth defect, 35000HPÄfis 319
colonies were smaller than those of the wild-type parent strain (Fig. 1C). Both the small colony 320
phenotype and the growth defect could be complemented in trans (data not shown) using a wild-321
type copy of the H. ducreyi fis gene in the relatively low copy number vector pACYC177 (see 322
Materials and Methods). The total protein profile of 35000HPÄfis (Fig. 1D, lane 2) was 323
different from that of the wild-type parent strain (Fig. 1D, lane 1), and this difference could be 324
eliminated when a wild-type copy of fis was introduced on a plasmid (Fig. 1D, lane 3). Fis has 325
been shown to be involved in several processes including transcription, replication, and 326
recombination, and its expression typically peaks in early-exponential phase (39, 41), suggesting 327
that Fis is most active when bacterial cells are rapidly dividing. In contrast, Fis expression in H. 328
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 16
16
ducreyi remained relatively constant through the growth phase (Fig. 1E), suggesting that the 329
regulatory activities of Fis in H. ducreyi might be different from those observed in other bacterial 330
species. 331
332
Deletion of fis in H. ducreyi results in decreased expression of the LspB/LspA2-two-333
partner secretion system. To ascertain whether H. ducreyi Fis was involved in the regulation 334
of expression of the LspB, LspA2, and LspA1 proteins, whole-cell lysates of wild-type, mutant, 335
and complemented mutant strains were subjected to Western blot analysis. Deletion of the fis 336
gene in H. ducreyi 35000HP resulted in decreased expression of both LspB and LspA2 (Fig. 2A, 337
lane 2), suggesting that Fis could modulate the expression of these proteins. It should be noted 338
here that lspB and lspA2 were previously shown to comprise a bicistronic operon in H. ducreyi 339
(17). In contrast to LspB/LspA2, expression of LspA1 from an unlinked locus remained 340
relatively unchanged in the fis deletion mutant (Fig. 2A, lane 2). Complementation of the fis 341
mutation in trans in 35000HP〉fis(pML164) (Fig. 2A, lane 3) resulted in an increase in LspB and 342
LspA2 expression to levels similar to those observed in wild-type cells (Fig. 2A, lane1). 343
344
The H. ducreyi LspB-LspA2/LspA1 two-partner secretion system has been previously 345
shown to be necessary for the inhibition of phagocytosis in macrophages (18); therefore, the 346
ability of 35000HPÄfis to inhibit the uptake of opsonized secondary targets by J774A.1 347
macrophages was tested. In this assay, the level of phagocytosis inhibition exerted by 348
35000HPÄfis (Fig. 2B, column 3) was similar (p=ns) to that of the wild-type strain (Fig. 2B, 349
column 1), with a phagocytosis index significantly lower than that of the lspA1 lspA2 double 350
mutant 35000HPっ12 (Fig. 2B, column 2; p<0.001). The ability of this fis mutant to effectively 351
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 17
17
inhibit phagocytosis can be explained by the facts that expression of LspB, albeit reduced, was 352
not abolished, and expression of LspA1 was not affected. Either LspA1 or LspA2 is sufficient to 353
effectively inhibit phagocytosis (18). 354
355
H. ducreyi Fis: structural considerations. To glean information regarding the structure 356
of the H. ducreyi Fis protein, we used its amino-acid sequence to query the sequence databases 357
(42). Nearly every protein returned from the search was annotated as a Fis homolog. Fis is a 358
dimeric DNA-binding protein that causes B-form DNA to bend, and thus it has important 359
functions in DNA rearrangements, replication, transcription, and other activities (43). The 360
current Fis consensus sequence is 5´-GXXXXXXXXXXXXXC-3´, where the central portion (5-361
7 bases) is enriched for A and T (44). Many of the returned proteins are very similar to the H. 362
ducreyi Fis protein; for example, the E. coli homologs (several strains were identified) have 363
amino-acid sequence identities to H. ducreyi Fis of about 70%. Using a hidden Markov model 364
search method (45, 46), the Protein Data Bank (PDB) was queried for likely structural homologs. 365
The R71L mutant of E. coli Fis (PDB accession code 1ETO) (47) was returned as the most 366
likely, with a probability of 99.8%. 367
368
Given the high amino-acid sequence identity and the high probability of a structural 369
match, it was deemed that using E. coli Fis as a template for the construction of a homology 370
model of an H. ducreyi Fis monomer was feasible. The MODELLER program (48) was used for 371
this task; the result is shown in Supplemental Figure 1A. For this model, the untemplated parts 372
(resulting from unmodeled residues in E. coli Fis) were eliminated, as was a く-hairpin near to the 373
N-terminus whose structure is known to adopt many positions with respect to the remainder of 374
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 18
18
the protein (47). The H. ducreyi Fis model comprises residues L27 through G98. Overall, there 375
is a single a-helix at the N-terminus that packs against a helix-turn-helix motif that is responsible 376
for the DNA-binding activity (49). Details of E. coli Fis and other Fis structures are 377
recapitulated in this model. For example, basic and polar residues known to contact DNA 378
phosphates are solvent-exposed in the H. ducreyi Fis model (Supplemental Figure 1B). 379
380
DNA microarray analysis of the H. ducreyi fis deletion mutant. Fis has been shown to 381
be involved in the regulation of expression of many genes, including virulence factors, in 382
different bacterial systems (50, 51). We used DNA microarrays to compare the global 383
expression profile of the wild-type strain with that of the fis mutant in an attempt to determine 384
the extent of Fis involvement in gene expression in H. ducreyi. In the absence of Fis, 9.89% of 385
the predicted ORFs in the H. ducreyi genome were differentially expressed at least two-fold. Of 386
these 181 genes, 100 genes were up-regulated and 81 down-regulated (Supplemental Table 1, 387
p<0.05). From this list, a subset of genes was used to validate the DNA microarray data using 388
real-time RT-PCR (Fig. 3, correlation coefficient R2=0.788). A list of the 15 most up- and 389
down-regulated genes in the absence of Fis, not including those ORFs annotated as encoding 390
hypothetical proteins, is shown in Table 3. Among the most down-regulated genes were 391
ompP2A, ompA2, dsrA, and components of the flp operon. Both DsrA and Flp have been shown 392
to be H. ducreyi virulence factors (52, 53). Consistent with the protein expression data presented 393
above (Fig. 2), the level of lspA2 transcripts was reduced approximately two-fold in this analysis 394
(Supplemental Table 1). 395
396
Effect of the fis mutation on other proven H. ducreyi virulence factors. To further 397
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 19
19
validate these DNA microarray data at the protein level, whole cell lysates of wild-type 398
35000HP, 35000HPÄfis, 35000HPÄfis(pML164), and 35000HPÄfis(pACYC177) were first 399
analyzed by Western blotting for both DsrA and Flp protein expression (Fig. 4A and Fig. 5A, 400
respectively). DsrA is responsible for serum resistance (30) and inactivation of dsrA results in 401
attenuation of H. ducreyi in the human challenge model (52). Deletion of fis resulted in a 402
significant reduction in DsrA levels (Fig. 4A, lane 2), which could be restored to wild-type levels 403
by complementation in trans (Fig. 4A, lane 3). Next, the ability of a 35000HPÄfis mutant to 404
resist killing by normal human serum (NHS) was tested in a bactericidal assay (Fig. 4B). We 405
compared the survival of 35000HP, 35000HPÄfis, 35000HPÄfis(pML164), 406
35000HPÄfis(pACYC177), and a dsrA mutant (FX517) in 50% NHS. The resistance level of 407
the fis mutant (Fig. 4B, column 2) was less than that of wild-type (Fig. 4B, column 1) but greater 408
than that of the serum-sensitive dsrA mutant (Fig. 4B, column 5). Complementation of the fis 409
mutation (Fig. 4B, column 3) resulted in serum resistance equivalent to that of the wild-type 410
parent strain (Fig 4B, column 1). 411
412
Previous work has shown that H. ducreyi can form microcolonies when cultured with 413
human cells in vitro and that this phenotype requires expression of the protein products of the Flp 414
operon (31). The DNA microarray data indicated that, in the absence of Fis, the first genes in the 415
Flp operon (i.e., flp1, flp2, and flp3; see Table 1) are all down-regulated, a result which suggested 416
that 35000HPÄfis should have a diminished ability to form microcolonies. Western blot analysis 417
of whole cell lysates validated our DNA microarray data and showed decreased expression of 418
Flp1 in the absence of Fis (Fig. 5A, lane 2) relative to the wild-type parent strain (Fig. 5A, lane 419
1). This expression deficiency was corrected by complementation with the wild-type fis gene in 420
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 20
20
trans (Fig. 5A, lane 3). This deficiency in Flp1 expression was also apparent in a microcolony 421
formation assay where very few microcolonies were seen after incubation of 35000HPÄfis with 422
Hs27 human foreskin fibroblast cells (Fig. 5B). This microcolony deficiency phenotype could 423
be restored by complementation (Fig. 5B). As a negative control, a H. ducreyi tadA mutant (31) 424
unable to secrete Flp proteins was used (Fig. 5B). 425
426
Fis has a positive effect on lspB expression. Using the known E. coli Fis consensus 427
binding sequence (44) as a reference, a putative consensus-binding motif was located in the lspB 428
promoter region (Supplemental Fig. 2). Attempts to show a specific interaction of purified 429
recombinant H. ducreyi Fis with the promoter region of lspB were unsuccessful; there was no 430
statistical difference between the binding of Fis to the promoter region of lspB and the binding to 431
a fragment from within the lspB ORF (data not shown). This non-specific binding of Fis was 432
also observed with internal fragments of ten different ORFs selected from the DNA microarray 433
results (data not shown). A lacZ-based transcriptional reporter (Fig. 6A) was constructed and 434
used to test promoter activation by Fis in different H. ducreyi strain backgrounds. Although the 435
lspB promoter in pML306 (Table 1) was not very active in the 35000HP wild-type background 436
(Fig. 6B), in the absence of Fis (in the 35000HPÄfis mutant), promoter activity was significantly 437
lower (Fig. 6B) which was consistent with protein expression data (Fig. 2A). The promoter 438
regions of genes encoding other proven H. ducreyi virulence factors whose expression was 439
shown to be affected by the absence of Fis in Western blot analysis (Fig. 4 and 5) were also 440
introduced into this transcriptional reporter. All these promoter regions were screened for the 441
presence of putative Fis binding motifs, and the results are shown in Supplemental Fig. 2. く-442
galactosidase activity assays showed reduced promoter activity for both the dsrA and flp1 443
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 21
21
reporter constructs in the 35000HPÄfis mutant (Fig. 6B), indicating a positive effect of Fis on the 444
transcription of these genes. The promoter region of ompP2B was also tested because the DNA 445
microarray data (Table 3) indicated that expression of this ORF was up-regulated in the absence 446
of Fis. The transcriptional reporter assay using pML314 (Table 1) corroborated these DNA 447
microarray data and confirmed that Fis was involved in the repression of transcription of 448
ompP2B (Fig. 6B). 449
450
In other bacteria, Fis has been show to negatively regulate the gyrB gene (54, 55). To test 451
whether the H. ducreyi Fis homologue would have the same effect on the expression of gyrB, the 452
activity of the gyrB promoter region was tested using this reporter. When compared to the wild-453
type parent strain (Fig. 6B), the activity of the gyrB promoter region in pML309 in the fis 454
deletion mutant was significantly lower (Fig. 6B), indicating a positive effect of Fis on gyrB 455
transcription in H. ducreyi. It is important to note that analysis of the gyrB promoter region 456
showed the presence of two putative Fis binding motifs (Supplemental Fig. 2). These data 457
provide further evidence that the regulatory activities of Fis in H. ducreyi might be different than 458
in other pathogens. 459
460
CpxR and Fis both control LspB expression. The protein expression pattern of 461
35000HPÄfis described above was reminiscent of that observed with a H. ducreyi recombinant 462
strain possessing an over-expressed CpxR protein (34). In fact, when the DNA microarray 463
results from the latter strain were compared to those obtained with the fis deletion mutant, a 464
correlation coefficient of 0.877 was obtained. To examine the effect of a fis mutation in the 465
absence of CpxR, a 35000HPÄcpxR〉fis double mutant was constructed and the levels of 466
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 22
22
expression of LspB were determined by Western blot analysis. CpxR has been previously shown 467
to be a negative regulator of LspB expression (23), and its absence results in increased 468
expression of LspB (Fig. 6C, lane 3) relative to the wild-type strain (Fig. 6C, lane 1). In contrast, 469
the absence of only Fis resulted in decreased LspB expression (Fig. 6C, lane 2), suggesting a 470
positive regulatory effect of Fis on LspB expression. In the 35000HPÄcpxR 〉fis mutant lacking 471
both CpxR and Fis (Fig. 6C, lane 4), the levels of expression of LspB were higher than those of 472
35000HPÄfis (Fig. 6C, lane 2), lower than the LspB expression levels of 35000HPÄcpxR (Fig. 473
6C, lane 3), and very similar to those of the wild-type strain (Fig. 6C, lane 1). These data 474
suggest that removing both the positive and negative regulatory molecules (i.e., Fis and CpxR, 475
respectively) restores LspB expression to essentially wild-type levels. 476
477
478
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 23
23
DISCUSSION 479
480
The nucleoid-associated protein Fis is one of at least twelve nucleoid-associated proteins 481
expressed by E. coli (43). Among these, perhaps the most is known about Fis functionality in the 482
bacterial cell with respect to both physiology and virulence expression. Fis displays a high 483
degree of sequence and function conservation among enteric pathogens (56). However, while 484
there is a high degree of sequence and structure homology between the E. coli and H. ducreyi Fis 485
proteins, our mutant analysis-derived data suggest that their functions might be different. In 486
enteric bacteria like E. coli and Salmonella typhimurium, Fis has been shown to have a very 487
specific pattern of expression. Fis protein levels are highest at early stages of the growth phase 488
as the organism responds to increased levels of nutrients, which support the increasing demands 489
of transcription and translation as the cells grow rapidly. After this quick surge, the de novo 490
production of Fis ends and the amounts of Fis protein per cell decline as the bacterial cells divide 491
rapidly, until Fis is barely detectable upon entry into stationary phase (39, 55, 57). In contrast, 492
our data indicate that the levels of Fis protein in H. ducreyi remain relatively constant throughout 493
the first 16 h of growth (Fig. 1E) into early stationary phase (11), suggesting that Fis might have 494
a different role(s) in affecting gene expression in this obligate human pathogen. 495
496
In other bacterial systems, Fis indirectly regulates expression of genes through the 497
repression of gyrB (54, 55). As the levels of GyrB decline, DNA negative supercoiling 498
decreases, leading to a relaxed DNA state which either allows for transcription factors to access 499
promoter regions or prevents the binding of factors that require a specific DNA topology (55, 500
58, 59). Using a LacZ-based reporter construct, our data show that in H. ducreyi 35000HP Fis 501
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 24
24
has a positive effect on gyrB transcription (Fig. 6B), potentially increasing the levels of DNA 502
supercoiling throughout the entire growth phase. Further work is needed to establish whether, in 503
H. ducreyi, the levels of supercoiling remain constant throughout the growth phase or whether 504
other topoisomerases are involved in this process to allow for differential gene expression during 505
the different stages of growth. 506
507
Fis has been shown to be involved in regulation of expression of virulence factors in a 508
number of pathogens (26, 50, 60-62). In P. multocida, a close relative of H. ducreyi, Fis has 509
been shown to be required for expression of the two-partner secretion system composed of 510
PfhB_2 and LspB_2 (26). This particular P. multocida system has homology to the 511
LspB/LspA2-LspA1 two-partner secretion system of H. ducreyi. In this study, we showed that 512
the H. ducreyi Fis homolog is involved in the regulation of expression of the lspB-lspA2 operon, 513
which encodes for the secretion component (LspB) and one of the two secreted products (LspA2) 514
of this H. ducreyi two-partner secretion system (17). In the absence of Fis, the expression of 515
LspB and LspA2 decreases significantly, while the expression of the LspA1 protein, encoded at a 516
different locus, remains unchanged (Fig. 2A, compare lane 1 with lane 2). The LspA proteins 517
(LspA2 and LspA1) have been shown to be responsible for the ability of H. ducreyi to inhibit 518
phagocytosis by macrophages in vitro (18), with either protein being sufficient to cause 519
inhibition (18). Our new data indicate that a H. ducreyi fis mutant is able to inhibit phagocytosis 520
at a level similar to wild-type (Fig. 2B, compare column 1 with column 3). It should be noted 521
that deletion of fis reduced the expression of LspB, but it does not completely abolish it (Fig. 2A, 522
lane 2). Consequently, enough LspB is produced to allow secretion of LspA1, the expression of 523
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 25
25
which is not affected by inactivation of fis, with subsequent inhibition of phagocytosis (Fig. 2B, 524
column 3). 525
526
While the original intent of the present study was to determine whether H. ducreyi Fis 527
was involved in controlling expression of the LspA proteins, the wide range of both direct and 528
indirect regulatory activities attributed to enteric Fis proteins (50, 62, 63) prompted us to widen 529
the scope of our investigation. Using DNA microarrays to test the extent of the involvement of 530
Fis on gene expression in H. ducreyi, we determined that deletion of fis affected expression of 531
~10% of the genome. Whereas several functional categories are affected by the absence of Fis 532
(Supplemental Fig. 3), our data also indicate that a number of genes shown to be important for 533
virulence of H. ducreyi also are Fis-dependent (Table 3, Fig. 2, Fig. 4, and Fig. 5). These 534
include the genes encoding the DsrA serum resistance protein (30) and the Flp1 protein involved 535
in microcolony formation (31), in addition to the lspB-lspA2 operon. Intensive efforts to show a 536
direct interaction of Fis with the promoter regions of dsrA, flp1, and lspB were unsuccessful due 537
to the non-specific binding of H. ducreyi Fis to DNA (i.e., recombinant H. ducreyi Fis readily 538
bound to internal fragments of ORFs) (data not shown). However, the use of lacZ-based 539
transcriptional fusions showed decreased activity of the promoters for these three ORFs in the 540
absence of Fis (Fig. 6B), suggesting a positive influence of Fis on their expression. 541
542
We recently established that CpxR is a direct repressor of the lspB-lspA2 operon in H. 543
ducreyi 35000HP (23, 34). In the present study, we provide evidence that Fis positively affects 544
the expression of this same operon (Fig. 2A and Fig. 6C). In the absence of Fis, expression of 545
the CpxR protein remained essentially constant (Fig. 2A and Fig. 6C, lane 2) whereas that of 546
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 26
26
LspB (Fig. 2A and Fig. 6C, lane 2) and LspA2 (Fig. 2A, lane 2) decreased dramatically. There 547
are at least two ways in which Fis may affect expression of this operon. Fis may alter promoter 548
DNA topology in the presence of CpxR to allow effective transcription from the lspB-lspA2 549
promoter. It is also possible that the level of phosphorylation of CpxR increases in the absence 550
of Fis, a result that would likely increase the ability of CpxR to decrease expression of the lspB-551
lspA2 operon. Consistent with this latter possibility, the DNA microarray data indicated that 552
there was an increase in expression of phosphotransacetylase (Pta) (Supplemental Table 1), 553
which was confirmed by real-time RT-PCR (data not shown). Phosphotransacetylase is involved 554
in the synthesis of acetyl phosphate, a small phosphodonor that has been shown in E. coli to 555
phosphorylate CpxR independent of CpxA (64). It is possible that an increase in acetyl 556
phosphate levels in the 35000HPÄfis mutant could increase the phosphorylation of CpxR without 557
affecting its level of expression, resulting in a reduction in expression of the virulence factors 558
negatively controlled by CpxR. In this regard, it is interesting to note that a comparison of the 559
transcriptional profiles of the 35000HPÄfis mutant and a 35000HPÄcpxA mutant (34), which is 560
likely to have a highly phosphorylated CpxR (24, 34, 64), showed a correlation coefficient of R2
561
=0.937. 562
563
It should be noted that deletion of both cpxR and fis resulted in levels of expression of 564
LspB that were similar to wild-type (Fig. 6C, lane 4), suggesting that there may exist another 565
mechanism of positive regulation that may work in combination with the normal CpxR 566
repression and Fis activation of this operon. In S. enterica, Fis has been shown to be involved in 567
the control of important virulence factors, but most if not all of the Fis-dependent genes have 568
other levels of regulation (62, 65). Similarly, in enteroaggregative E. coli, Fis is required for full 569
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 27
27
expression of the Pet autotransporter toxin where it functions in concert with CRP (66). It is 570
possible that these opposing actions of CpxR and Fis on the lspB-lspA2 operon are critical in the 571
pathogenesis of H. ducreyi disease. Investigation of the relative importance and potential 572
interaction of CpxR and Fis at the promoter region of lspB are currently underway. 573
574
One limitation to the present study is that it utilized 35000HP, a class I strain of H. 575
ducreyi. Two apparently clonal populations of H. ducreyi (class I and class II) have been 576
described (67) and confirmed by both proteomic methods (68) and additional genetic analyses 577
(69). Whether a fis mutant of a class II strain of H. ducreyi would have a transcriptome profile 578
different from that of 35000HPÄfis remains to be determined, but would seem unlikely in view 579
of the conservation of Fis effects among closely related bacteria (i.e., enteric organisms). 580
Another caveat is that there are no protein expression data available to indicate how the Fis 581
protein is regulated in the infected human host. Perhaps RNA-seq-based analysis of samples 582
from lesions formed in the human challenge model for chancroid could address this issue in the 583
future. 584
585
586
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 28
28
ACKNOWLEDGEMENTS 587
588
This study was supported by U.S. Public Health Service grant AI032011 and ARRA 589
Supplement AI032011-18S1 to E.J.H. and by U.S. Public Health Service grant AI27863 to 590
S.M.S. 591
592
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 29
29
LITERATURE CITED 593
594
595
1. Bong CT, Bauer ME, Spinola SM. 2002. Haemophilus ducreyi: clinical features, 596
epidemiology, and prospects for disease control. Microbes.Infect. 4:1141-1148. 597
598
2. Spinola SM, Bauer ME, Munson RS, Jr. 2002. Immunopathogenesis of Haemophilus 599
ducreyi infection (chancroid). Infect.Immun. 70:1667-1676. 600
601
3. Fleming DT, Wasserheit JN. 1999. From epidemiological synergy to public health 602
policy and practice: the contribution of other sexually transmitted diseases to sexual 603
transmission of HIV infection. Sex.Transm.Inf. 75:3-17. 604
605
4. Royce RA, Sena A, Cates W, Jr., Cohen MS. 1997. Sexual transmission of HIV. 606
N.Engl.J.Med. 336:1072-1078. 607
608
5. Haydock AK, Martin DH, Morse SA, Cammarata C, Mertz KJ, Totten PA. 1999. 609
Molecular characterization of Haemophilus ducreyi strains from Jackson, Mississippi and 610
New Orleans, Louisiana. J.Infect.Dis. 179:1423-1432. 611
612
6. Mertz KJ, Trees DL, Levine Wc, Lewis JS, Litchfield B, Pettus KS, Morse SA, ME 613
SL, Weiss JB, Schwebke J, Dickes J, Kee R, Reynolds J, Hutcheson D, Green D, 614
Dyer I, Richwald GA, Novotny J, Goldberg M, O'Donnell JA, Knaup R. 1998. 615
Etiology of genital ulcers and prevalence of human immunodeficiency virus coinfection 616
in 10 US cities. J.Infect.Dis. 178:1795-1798. 617
618
7. Spinola SM, Wild LM, Apicella MA, Gaspari AA, Campagnari AA. 1994. 619
Experimental human infection with Haemophilus ducreyi. J.Infect.Dis. 169:1146-1150. 620
621
8. Janowicz DM, Ofner S, Katz BP, Spinola SM. 2009. Experimental Infection of Human 622
Volunteers with Haemophilus ducreyi: Fifteen Years of Clinical Data and Experience. 623
J.Infect.Dis. 199:1671-1679. 624
625
9. Rinker SD, Gu X, Fortney KR, Zwickl BW, Katz BP, Janowicz DM, Spinola SM, 626
Bauer ME. 2012. Permeases of the sap transporter are required for cathelicidin resistance 627
and virulence of Haemophilus ducreyi in humans. J.Infect.Dis. 206:1407-1414. 628
629
10. Janowicz DM, Cooney SA, Walsh J, Baker B, Katz BP, Fortney KR, Zwickl BW, 630
Ellinger S, Munson RS, Jr. 2011. Expression of the Flp proteins by Haemophilus 631
ducreyi is necessary for virulence in human volunteers. BMC.Microbiol. 11:208. 632
633
11. Labandeira-Rey M, Janowicz DM, Blick RJ, Fortney KR, Zwickl B, Katz BP, 634
Spinola SM, Hansen EJ. 2009. Inactivation of the Haemophilus ducreyi luxS gene 635
affects the virulence of this pathogen in human subjects. J.Infect.Dis. 200:409-416. 636
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 30
30
12. Labandeira-Rey M, Dodd D, Fortney KR, Zwickl B, Katz BP, Janowicz DM, 637
Spinola SM, Hansen EJ. 2011. A Haemophilus ducreyi CpxR Deletion Mutant Is 638
Virulent in Human Volunteers. J.Infect.Dis. 203:1859-1865. 639
640
13. Spinola SM, Fortney KR, Baker B, Janowicz DM, Zwickl B, Katz BP, Blick RJ, 641
Munson RS, Jr. 2010. Activation of the CpxRA System by Deletion of cpxA Impairs the 642
Ability of Haemophilus ducreyi To Infect Humans. Infect.Immun. 78:3898-3904. 643
644
14. Gangaiah D, Li W, Fortney KR, Janowicz DM, Ellinger S, Zwickl B, Katz BP, 645
Spinola SM. 2013. Carbon storage regulator A contributes to the virulence of 646
Haemophilus ducreyi in humans by multiple mechanisms. Infect.Immun. 81:608-617. 647
648
15. Spinola SM, Li W, Fortney KR, Janowicz DM, Zwickl B, Katz BP, Munson RS, Jr. 649
2012. Sialylation of lipooligosaccharides is dispensable for the virulence of Haemophilus 650
ducreyi in humans. Infect.Immun. 80:679-687. 651
652
16. Janowicz DM, Fortney KR, Katz BP, Latimer JL, Deng K, Hansen EJ, Spinola SM. 653
2004. Expression of the LspA1 and LspA2 proteins by Haemophilus ducreyi is required 654
for virulence in human volunteers. Infect.Immun. 72:4528-4533. 655
656
17. Ward CK, Mock JR, Hansen EJ. 2004. The LspB Protein Is Involved in the Secretion 657
of the LspA1 and LspA2 Proteins by Haemophilus ducreyi. Infect.Immun. 72:1874-1884. 658
659
18. Vakevainen M, Greenberg S, Hansen EJ. 2003. Inhibition of phagocytosis by 660
Haemophilus ducreyi requires expression of the LspA1 and LspA2 proteins. 661
Infect.Immun. 71:5994-6003. 662
663
19. Mock JR, Vakevainen M, Deng K, Latimer JL, Young JA, van Oers NS, Greenberg 664
S, Hansen EJ. 2005. Haemophilus ducreyi targets Src family protein tyrosine kinases to 665
inhibit phagocytic signaling. Infect.Immun. 73:7808-7816. 666
667
20. Deng K, Mock JR, Greenberg S, van Oers NS, Hansen EJ. 2008. Haemophilus 668
ducreyi LspA Proteins Are Tyrosine Phosphorylated by Macrophage-Encoded Protein 669
Tyrosine Kinases. Infect.Immun. 76:4692-4702. 670
671
21. Elkins C, Chen C-J, Thomas CE. 1995. Characterization of the hgbA locus encoding a 672
hemoglobin receptor from Haemophilus ducreyi. Infect.Immun. 63:2194-2200. 673
674
22. Bauer ME, Fortney KR, Harrison A, Janowicz DM, Munson RS, Jr., Spinola SM. 675
2008. Identification of Haemophilus ducreyi genes expressed during human infection. 676
Microbiology 154:1152-1160. 677
678
23. Labandeira-Rey M, Mock JR, Hansen EJ. 2009. Regulation of expression of the 679
Haemophilus ducreyi LspB and LspA2 proteins by CpxR. Infect.Immun. 77:3402-3411. 680
681
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 31
31
24. Gangaiah D, Zhang X, Fortney KR, Baker B, Liu Y, Munson RS, Jr., Spinola SM. 682
2013. Activation of CpxRA in Haemophilus ducreyi primarily inhibits the expression of 683
its targets including major virulence determinants. J Bacteriol. 195:3486-3502. 684
685
25. Rinker SD, Trombley MP, Gu X, Fortney KR, Bauer ME. 2011. Deletion of mtrC in 686
Haemophilus ducreyi Increases Sensitivity to Human Antimicrobial Peptides and 687
Activates the CpxRA Regulon. Infect.Immun. 79:2324-2334. 688
689
26. Steen JA, Steen JA, Harrison P, Seemann T, Wilkie I, Harper M, Adler B, Boyce 690
JD. 2010. Fis is essential for capsule production in Pasteurella multocida and regulates 691
expression of other important virulence factors. PLoS.Pathog. 6:e1000750. 692
693
27. Hansen EJ, Latimer JL, Thomas SE, Helminen M, Albritton WL, Radolf JD. 1992. 694
Use of electroporation to construct isogenic mutants of Haemophilus ducreyi. J.Bacteriol. 695
174:5442-5449. 696
697
28. Ward CK, Lumbley SR, Latimer JL, Cope LD, Hansen EJ. 1998. Haemophilus 698
ducreyi secretes a filamentous hemagglutinin-like protein. J.Bacteriol. 180:6013-6022. 699
700
29. Spinola SM, Hiltke TJ, Fortney KR, Shanks KL. 1996. The conserved 18,000-701
molecular-weight outer membrane protein of Haemophilus ducreyi has homology to 702
PAL. Infect.Immun. 64:1950-1955. 703
704
30. Elkins C, Morrow KJ, Jr., Olsen B. 2000. Serum resistance in Haemophilus ducreyi 705
requires outer membrane protein DsrA. Infect.Immun. 68:1608-1619. 706
707
31. Nika JR, Latimer JL, Ward CK, Blick RJ, Wagner NJ, Cope LD, Mahairas GG, 708
Munson RS, Jr., Hansen EJ. 2002. Haemophilus ducreyi requires the flp gene cluster 709
for microcolony formation in vitro. Infect.Immun. 70:2965-2975. 710
711
32. Lukomski S, Hull RA, Hull SI. 1996. Identification of the O antigen polymerase (rfc) 712
gene in Escherichia coli O4 by insertional mutagenesis using a nonpolar 713
chloramphenicol resistance cassette. J.Bacteriol. 178:240-247. 714
715
33. Horton RM, Cai Z, Ho SN, Pease LR. 1990. Gene splicing by overlap extension: tailor-716
made genes using the polymerase chain reaction. Biotechniques 8:528-535. 717
718
34. Labandeira-Rey M, Brautigam CA, Hansen EJ. 2010. Characterization of the CpxRA 719
regulon in Haemophilus ducreyi. Infect.Immun. 78:4779-4791. 720
721
35. Kiss K, Liu W, Huntley JF, Norgard MV, Hansen EJ. 2008. Characterization of fig 722
operon mutants of Francisella novicida U112. FEMS Microbiol.Lett. 285:270-277. 723
724
36. Evans AS, Pybus C, Hansen EJ. 2012. Development of a LacZ-based transcriptional 725
reporter system for use with Moraxella catarrhalis. Plasmid. 726
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 32
32
37. Miller JH. 1972. Experiments in molecular genetics. Cold Spring Harbor 727
Laboratory,Cold Spring Harbor,N.Y. 728
729
38. Ward CK, Latimer JL, Nika JR, Vakevainen M, Mock JR, Deng K, Blick RJ, 730
Hansen EH. 2003. Mutations in the lspA1 and lspA2 genes of Haemophilus ducreyi 731
affect the virulence of this pathogen in an animal model system. Infect.Immun. 71:2478-732
2486. 733
734
39. Osuna R, Lienau D, Hughes KT, Johnson RC. 1995. Sequence, regulation, and 735
functions of fis in Salmonella typhimurium. J.Bacteriol. 177:2021-2032. 736
737
40. Bradley MD, Beach MB, de Koning AP, Pratt TS, Osuna R. 2007. Effects of Fis on 738
Escherichia coli gene expression during different growth stages. Microbiology 153:2922-739
2940. 740
741
41. Ali AT, Iwata A, Nishimura A, Ueda S, Ishihama A. 1999. Growth phase-dependent 742
variation in protein composition of the Escherichia coli nucleoid. J.Bacteriol. 181:6361-743
6370. 744
745
42. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment 746
search tool. J.Mol.Biol. 215:403-410. 747
748
43. Dorman CJ. 2009. Nucleoid-associated proteins and bacterial physiology. 749
Adv.Appl.Microbiol. 67:47-64. 750
751
44. Stella S, Cascio D, Johnson RC. 2010. The shape of the DNA minor groove directs 752
binding by the DNA-bending protein Fis. Genes Dev. 24:814-826. 753
754
45. Soding J. 2005. Protein homology detection by HMM-HMM comparison. 755
Bioinformatics. 21:951-960. 756
757
46. Soding J, Biegert A, Lupas AN. 2005. The HHpred interactive server for protein 758
homology detection and structure prediction. Nucleic Acids Res. 33:W244-W248. 759
760
47. Cheng YS, Yang WZ, Johnson RC, Yuan HS. 2000. Structural analysis of the 761
transcriptional activation region on Fis: crystal structures of six Fis mutants with different 762
activation properties. J.Mol.Biol. 302:1139-1151. 763
764
48. Sali A, Blundell TL. 1993. Comparative protein modelling by satisfaction of spatial 765
restraints. J.Mol.Biol. 234:779-815. 766
767
49. Koch C, Ninnemann O, Fuss H, Kahmann R. 1991. The N-terminal part of the E.coli 768
DNA binding protein FIS is essential for stimulating site-specific DNA inversion but is 769
not required for specific DNA binding. Nucleic Acids Res. 19:5915-5922. 770
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 33
33
50. Goldberg MD, Johnson M, Hinton JC, Williams PH. 2001. Role of the nucleoid-771
associated protein Fis in the regulation of virulence properties of enteropathogenic 772
Escherichia coli. Mol.Microbiol. 41:549-559. 773
774
51. Falconi M, Prosseda G, Giangrossi M, Beghetto E, Colonna B. 2001. Involvement of 775
FIS in the H-NS-mediated regulation of virF gene of Shigella and enteroinvasive 776
Escherichia coli. Mol.Microbiol. 42:439-452. 777
778
52. Bong CT, Throm RE, Fortney KR, Katz BP, Hood AF, Elkins C, Spinola SM. 2001. 779
DsrA-deficient mutant of Haemophilus ducreyi is impaired in its ability to infect human 780
volunteers. Infect.Immun. 69:1488-1491. 781
782
53. Spinola SM, Fortney KR, Katz BP, Latimer JL, Mock JR, Vakevainen M, Hansen 783
EJ. 2003. Haemophilus ducreyi requires an intact flp gene cluster for virulence in 784
humans. Infect.Immun. 71:7178-7182. 785
786
54. Dorman CJ, Deighan P. 2003. Regulation of gene expression by histone-like proteins in 787
bacteria. Curr.Opin.Genet.Dev. 13:179-184. 788
789
55. Keane OM, Dorman CJ. 2003. The gyr genes of Salmonella enterica serovar 790
Typhimurium are repressed by the factor for inversion stimulation, Fis. 791
Mol.Genet.Genomics 270:56-65. 792
793
56. Beach MB, Osuna R. 1998. Identification and characterization of the fis operon in 794
enteric bacteria. J.Bacteriol. 180:5932-5946. 795
796
57. Ball CA, Osuna R, Ferguson KC, Johnson RC. 1992. Dramatic changes in Fis levels 797
upon nutrient upshift in Escherichia coli. J.Bacteriol. 174:8043-8056. 798
799
58. Schneider R, Travers A, Kutateladze T, Muskhelishvili G. 1999. A DNA architectural 800
protein couples cellular physiology and DNA topology in Escherichia coli. 801
Mol.Microbiol. 34:953-964. 802
803
59. Travers A, Schneider R, Muskhelishvili G. 2001. DNA supercoiling and transcription 804
in Escherichia coli: The FIS connection. Biochimie 83:213-217. 805
806
60. Lim S, Kim B, Choi HS, Lee Y, Ryu S. 2006. Fis is required for proper regulation of 807
ssaG expression in Salmonella enterica serovar Typhimurium. Microb.Pathog. 41:33-42. 808
809
61. Lautier T, Nasser W. 2007. The DNA nucleoid-associated protein Fis co-ordinates the 810
expression of the main virulence genes in the phytopathogenic bacterium Erwinia 811
chrysanthemi. Mol.Microbiol. 66:1474-1490. 812
813
62. Kelly A, Goldberg MD, Carroll RK, Danino V, Hinton JC, Dorman CJ. 2004. A 814
global role for Fis in the transcriptional control of metabolism and type III secretion in 815
Salmonella enterica serovar Typhimurium. Microbiology 150:2037-2053. 816
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 34
34
63. Kahramanoglou C, Seshasayee AS, Prieto AI, Ibberson D, Schmidt S, Zimmermann 817
J, Benes V, Fraser GM, Luscombe NM. 2011. Direct and indirect effects of H-NS and 818
Fis on global gene expression control in Escherichia coli. Nucleic Acids Res. 39:2073-819
2091. 820
821
64. Lima BP, Thanh Huyen TT, Basell K, Becher D, Antelmann H, Wolfe AJ. 2012. 822
Inhibition of acetyl phosphate-dependent transcription by an acetylatable lysine on RNA 823
polymerase. The Journal of biological chemistry 287:32147-32160. 824
825
65. Dorman CJ. 2013. Co-operative roles for DNA supercoiling and nucleoid-associated 826
proteins in the regulation of bacterial transcription. Biochem.Soc.Trans. 41:542-547. 827
828
66. Rossiter AE, Browning DF, Leyton DL, Johnson MD, Godfrey RE, Wardius CA, 829
Desvaux M, Cunningham AF, Ruiz-Perez F, Nataro JP, Busby SJ, Henderson IR. 830
2011. Transcription of the plasmid-encoded toxin gene from enteroaggregative 831
Escherichia coli is regulated by a novel co-activation mechanism involving CRP and Fis. 832
Mol.Microbiol. 81:179-191. 833
834
67. White CD, Leduc I, Olsen B, Jeter C, Harris C, Elkins C. 2005. Haemophilus ducreyi 835
Outer membrane determinants, including DsrA, define two clonal populations. 836
Infect.Immun. 73:2387-2399. 837
838
68. Post DM, Gibson BW. 2007. Proposed second class of Haemophilus ducreyi strains 839
show altered protein and lipooligosaccharide profiles. Proteomics. 7:3131-3142. 840
841
69. Ricotta EE, Wang N, Cutler R, Lawrence JG, Humphreys TL. 2011. Rapid 842
Divergence of Two Classes of Haemophilus ducreyi. J.Bacteriol. 193:2941-2947. 843
844
70. Al-Tawfiq JA, Thornton AC, Katz BP, Fortney KR, Todd KD, Hood AF, Spinola 845
SM. 1998. Standardization of the experimental model of Haemophilus ducreyi infection 846
in human subjects. J.Infect.Dis. 178:1684-1687. 847
848
71. Spinola SM, Fortney KR, Baker B, Janowicz DM, Zwickl B, Katz BP, Blick RJ, 849
Munson RS, Jr. 2010. Activation of the CpxRA system by deletion of cpxA impairs the 850
ability of Haemophilus ducreyi to infect humans. Infect Immun 78:3898-3904. 851
852
72. Elkins C, Morrow KJ, Olsen B. 2000. Serum resistance in Haemophilus ducreyi 853
requires outer membrane protein DsrA. Infect. Immun. 68:1608-1619. 854
855
73. Sambrook J, Fritsch EF, Maniatis T. 1989. Molecular cloning - a laboratory manual, 856
2nd Edition, 2 ed. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, N.Y. 857
858
859
860
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 35
35
Table 1. Bacterial strains and plasmids used in this study. 861
862
Name Description Reference or Source
Bacterial strains
H. ducreyi
35000HP Human-passaged variant of strain 35000 (70)
35000HPÄfis 35000HPÄfis::cat This study
35000HPÄcpxR 35000HPÄcpxR::cat (23)
35000HPÄcpxR〉fis 35000HPÄcpxR::cat Äfis::kan This study
35000HPÄcpxA 35000HP with unmarked, in-frame cpxA deletion (71)
FX517 35000 dsrA::cat insertion mutant (72)
35000HP tadA 35000HP.400; 35000HP tadA::cat (31)
E. coli
DH5g Host strain for cloning (73)
HB101 Host strain used to propagate pACYC177-based plasmids
prior to transformation into H. ducreyi (73)
XL-10 Gold Host strain for protein expression Stratagene
Plasmids
pGEX-T4-2 N-terminal GST fusion vector with an IPTG-inducible tac
promoter and a LacIq repressor, Ap
r GE HealthCare
pML165 pGEX-T4-2 carrying the 35000HP fis gene This study
pACYC177 Cloning vector, Kmr New England Biolabs
pML164 pACYC177 carrying the wild-type fis gene, Kmr This study
pCR2.1 Cloning vector, Kmr Invitrogen
pML122 pCR2.1 carrying the cat gene from pSL33 together with
the native cat promoter from pACYC184 This study
pML168 pCR2.1 carrying the kan gene from pUC18K3 together
with the native kan promoter from pUCK4 This study
pML303 lacZ-based transcriptional reporter in pACYC177 This study
pML306
pML303 carrying a 600-bp fragment of the lspB promoter
region (500-bp upstream and 100-bp downstream from
the ATG translational start codon)
This study
pML308
pML303 carrying a 600-bp fragment of the dsrA
promoter region (500-bp upstream and 100-bp
downstream from the ATG translational start codon)
This study
pML309
pML303 carrying a 600-bp fragment of the gyrB
promoter region (500-bp upstream and 100-bp
downstream from the ATG translational start codon)
This study
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 36
36
pML312
pML303 carrying a 600-bp fragment of the flp1 promoter
region (500-bp upstream and 100-bp downstream from
the ATG translational start codon)
This study
pML314
pML303 carrying a 600-bp fragment of the ompP2B
promoter region (500-bp upstream and 100-bp
downstream from the ATG translational start codon)
This study
863
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 37
37
Table 2. Oligonucleotide primers used in this study. 864
865
Sequence (5'-3') a,b
HD524 AGAGGTAATATGCGAATCGGG
HD525 CACAAGCTATTTATAAAGGCT
HD526 TAACATTGTCCTCTATCCAAC
HD527 GGCTAAATGCTTAACAAGGTT
HD528 GTTGGATAGAGGACAATGTTAGTTGATACCGGGAAGCCCTGG
HD529 AACCTTGTTAAGCATTTAGCCCATTATTCCCTCCAAAAATTA
HD543 ACGCCTCGAGTGGTTGTTTCAAGAACTG
HD809 GTTGGATAGAGGACAATGTTACCCCGGATCCGTCGACCTGCA
HD810 AACCTTGTTAAGCATTTAGCCGGGTCGCATTATTCCCTCCAG
HD544 ACGCCTCGAGAATTCAATAATCCTAAAT
HD547 ACGCGGATCCATGTTAGAACAACAACCT
HD548 ACGCCCCGGGTTAGCCCATACCGTATTT
Real-Time RT-PCRc
RT-HD0084 (F) TTGGGCGTGGGACGTTGGT
RT-HD0084 (R) CGGGCCGTCCCAGAAATCA
RT-carA (F) TCTCCAAACAACACCGACAT
RT-carA (R) CGGGCCATTAGAAAGAAAGA
RT-HD1895(F) TTATTCGTCGCCATGTTGTT
RT-HD1895(R) CGAGCAGCAATCATTGAAGT
RT-waaA (F) AACATCTCGGCTATGGGAAC
RT-waaA (R) TGTGATCGGTAATGCTGGAT
RT-HD1443(F) TGGCTTTGGTGGCTATAACA
RT-HD1443(R) TACCCTTTCTTCCACCTTCC
RT-ftnA(F) ACTGAGCCATGCTGATGAAG
RT-ftnA(R) CGGAAGTGGCTTCCACTAAT
RT-recD(F) TGAGCAAATACCACCGGTTA
RT-recD(R) TGTGGCGAGATCTTGATTTG
RT-HD1309(F) CGATATTCGCTCTCGCATTA
RT-HD1309(R) GCTCCGATTACGCCTAACAT
RT-glpF(F) CGTGGTGAAAGTGTTGGTTT
RT-glpF(R) TATTGGGCCTTTCGGTAATC
RT-HD1985(F) CGGTGTGCTTGATAGTGGAT
RT-HD1985(R) TCCATTATTGTGCCGATCTT
RT-ner(F) CTGATTGGCATCGTGAAGAC
RT-ner(R) CCCAAATAGTTTCGGCTGAT
RT-hflX(F) CATTACGGCGTATGCAAATC
RT-hflX(R) TTCATCGGCTATATCCACCA
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 38
38
Reporter Constructs
lspB promoter (F) ACGCGTCGACAGTAAATTTCTTCAAAAATGT
lspB promoter (R) ACGCGCGGCCGCTGAAAGCATAAATAAATAAGA
ompP2B promoter (R) ACGCGTCGACATTAAAGAAGTATTTGAT
ompP2B promoter (F) ACGCGCGGCCGCTCCAAATCAATTTTAGTT
gyrB promoter (R) ACGCGTCGACTTAATACTCGAAGAATCATAA
gyrB promoter (F) ACGCGCGGCCGCGCTTGAATATTGCAAAGATTC
dsrA promoter (F) ACGCGTCGACTGAATTGGAGTGGACCAGGAC
dsrA promoter (R) ACGCGCGGCCGCATAGTAGAACAAGCTAATCCC
flp1 promoter (F) ACGCGTCGACGCCAAACCATTTCGTAGCATC
flp1 promoter (R) ACGCGCGGCCGCAAATATAAAATAAATTATGTT
866 aBold text indicates complementary sequence for use in overlapping extension PCR. 867
868 bUnderlining indicates restriction site as described in Materials and Methods. 869
870 cThe primer sets for the following ORFs were described previously (34); aspA, fimA, lspA1, 871
ompP2A, ompP2B, HD1094, adk, artI, cpxR, flp1, dsrA, and ompA2. 872
873
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 39
39
Table 3. Genes whose expression was most affected by the absence of Fis as measured by 874
DNA microarray analysisa. 875
876
ORF Gene Description of gene product
Median log2 ratio
of expression
levelsb
SD
HD0281 fimA possible fimbrial major pilin protein 2.52 0.43
HD0282 fimB possible fimbrial structural subunit 2.35 0.48
HD0564 aspA aspartate ammonia-lyase 2.32 0.13
HD1528 Eha protein 2.26 0.24
HD0765 manA mannose-6-phosphate isomerase 1.81 0.20
HD1146 glpF glycerol uptake facilitator protein 1.79 0.21
HD1084 HesB family protein 1.74 0.24
HD1525 gam mu-phage host-nuclease inhibitor protein 1.72 0.18
HD1435 ompP2B outer membrane protein P2 homolog 1.57 0.06
HD1895 putative adhesin HmwC-like protein 1.56 0.05
HD1163 ribAB riboflavin biosynthesis protein RibA 1.55 0.26
HD0283 fimC possible fimbrial outer membrane usher 1.54 0.30
HD1094 possible outer membrane serine protease 1.53 0.07
HD1109 putative oxalate/formate antiporter 1.52 0.14
HD0454 waaA 3-deoxy-D-manno-octulosonic-acid
transferase
1.52 0.17
HD1926 rpmJ1 50S ribosomal protein L36 -1.69 0.61
HD0344 nrfA nitrate reductase, cytochrome c552 -1.75 0.07
HD1308 flpC flp operon protein C -1.75 0.12
HD1280 possible serine protease homolog -1.77 0.07
HD0769 dsrA serum resistance protein DsrA -1.88 0.16
HD0095 mu phage DNA transposition protein B -1.98 0.58
HD1309 flpB flp operon protein B -2.21 0.10
HD0090 ner possible DNA-binding protein -2.35 0.32
HD1312 flp1 flp operon protein Flp1 -2.40 0.16
HD1278 possible serine protease -2.67 0.07
HD0046 ompA2 major outer membrane protein homolog -2.80 0.14
HD1310 flp3 flp operon protein Flp3 -2.80 0.15
HD1311 flp2 flp operon protein Flp2 -2.86 0.19
HD1433 ompP2A outer membrane protein P2 homolog -3.15 0.31
HD1985 possible DNA transformation protein -3.56 0.27
877 a The table does not include ORFs described as encoding hypothetical or conserved hypothetical 878
proteins. 879
880 b
Median log2 ratio of expression levels from comparisons of 35000HP∆fis to 35000HP in three 881
independent experiments (P< 0.05). 882
883
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 40
40
FIGURE LEGENDS 884
885
886
Figure 1. Characterization of the H. ducreyi fis deletion mutant. (A) Schematic 887
representation of the fis locus in the wild-type strain 35000HP and 35000HP〉fis (〉fis). (B) 888
Growth of the wild-type parent strain 35000HP (WT, closed circles) and 35000HP〉fis (〉fis, 889
open circles) in broth (***p<0.001). (C) Colony size differences between 35000HP (WT) (top) 890
and 35000HP〉fis (bottom). (D) Total cell protein profiles for 35000HP (lane 1), 35000HP〉fis 891
(lane 2), 35000HP〉fis(pML164) (lane 3), and 35000HP〉fis(pACYC177) (lane 4) as determined 892
by resolving proteins by SDS-PAGE and staining with Coomassie blue. Cells were sampled at 893
the 8 h time point. The two black stars indicate bands present in the wild-type and missing or 894
markedly reduced in the mutant. The lower panel represents Western blot analysis with a mouse 895
polyclonal Fis antiserum. (E) Western blot analysis of whole cell lysates from 35000HP and 896
35000HP〉fis probed with the Fis antiserum (upper panel). Cells were sampled at 4, 8 and 16 h. 897
The PAL MAb 3B9 (lower panel) was used to confirm equivalent loading among lanes. 898
899
Figure 2. Analysis of protein expression and phagocytosis activity in wild-type and 900
mutant H. ducreyi strains. (A) Western blot analysis of whole cell lysates from 35000HP (lane 901
1), 35000HP〉fis (lane 2), 35000HP〉fis(pML164) (lane 3), and 35000HP〉fis(pACYC177) (lane 902
4) probed with a LspB polyclonal antibody, the LspA2 MAb 1H9, the LspA1 MAb 40A4, or a 903
CpxR polyclonal antibody. Cells were harvested at 8 h, and the same set of four whole cell 904
lysates was loaded onto multiple different gels, one set per primary antibody probe. The PAL 905
MAb 3B9 was used to confirm equivalent loading among lanes. It should be noted that the 906
LspA1 and LspA2 proteins do not exhibit discrete banding patterns in Western blot analysis but 907
instead form smears (17, 38). Fis protein expression by these same four strains is depicted in 908
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 41
41
Fig. 1D. (B) Phagocytosis assay. The ability of 35000HP (column 1), 35000HPっ12 (column 2), 909
35000HP〉fis (column 3), 35000HP〉fis(pML164) (column 4), and 35000HP〉fis(pACYC177) 910
(column 5) to inhibit the phagocytic activity of murine J774A.1 macrophages, as measured by 911
the uptake of opsonized latex beads was tested. A representative experiment is shown. The 912
multiplicity of infection (MOI) used for each strain is listed at the top of each column. 913
***p<0.001 914
915
Figure 3. Relative expression levels of selected H. ducreyi genes in 35000HP〉fis. 916
Expression of 23 selected genes in 35000HP〉fis compared to wild-type 35000HP cells was 917
measured by DNA microarrays (black bars) or real-time RT-PCR analysis (white bars) as 918
described in Materials and Methods. These data are the means of results from three independent 919
experiments. 920
921
Figure 4. A H. ducreyi fis deletion mutant is sensitive to serum killing. (A) Western 922
blot analysis of whole cell lysates of 35000HP (lane 1), 35000HP〉fis (lane 2), 923
35000HP〉fis(pML164) (lane 3), and 35000HP〉fis(pACYC177) (lane 4) probed with a DsrA 924
polyclonal antibody. Cells were harvested at 8 h. The PAL MAb 3B9 was used to confirm 925
equivalent loading among lanes. (B) Serum bactericidal activity assays. The percentage survival 926
of 35000HP (column 1), 35000HP〉fis (column 2), 35000HP〉fis(pML164) (column 3), 927
35000HP〉fis(pACYC177) (column 4), and the dsrA mutant FX517 (column 5) in 50% normal 928
human serum (NHS) was calculated as (geometric mean colony-forming units [CFUs] in 929
NHS/geometric mean CFUs in heat-inactivated NHS) × 100. Values represent means ± standard 930
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 42
42
deviations of 5 independent experiments. All strains were compared to 35000HP in column 1. 931
****p <0.0001, **p<0.01. 932
933
Figure 5. A H. ducreyi fis deletion mutant is deficient in microcolony formation. (A) 934
Western blot analysis of the same whole cell lysates used in Figure 2A, probed with a Flp1 935
polyclonal antibody and the PAL MAb 3B9. The panel depicting results obtained with MAb 936
3B9 is the same as that in Figure 2A. (B) Microcolony formation assay. The ability of 937
35000HP, 35000HP〉fis, 35000HP〉fis(pML164), 35000HP〉fis(pACYC177), and 35000HP 938
tadA to form microcolonies upon incubation with Hs27 human fibroblasts was tested. Cells were 939
harvested after 16 h growth in CB. A representative experiment is shown. Arrows indicate the 940
position of the microcolonies. 941
942
Figure 6. Fis and CpxR are involved in the regulation of LspB expression. (A) 943
Schematic map of the H. ducreyi LacZ-based reporter construct pML303. The location of the 944
kanamycin gene (kan) and the ori (both originally derived from pACYC177), the multicloning 945
site (MCS), transcriptional terminators (»), and the promoterless lacZ gene (originally derived 946
from pRS551) are indicated. The NotI and SalI sites used for directional cloning of H. ducreyi 947
promoter regions are shown. (B) Use of a く-galactosidase assay with pML303-derived constructs 948
to measure promoter activity. These include pML306 carrying the lspB promoter region, 949
pML308 carrying the dsrA promoter region, pML309 carrying the gyrB promoter region, 950
pML312 carrying the flp1 promoter region, and pML314 carrying the ompP2B promoter region. 951
The data are from a representative experiment and error bars represent standard deviation. Black 952
bars show promoter activity in a H. ducreyi 35000HP wild-type background whereas the white 953
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 43
43
bars show promoter activity in a 35000HP〉fis background. *p=0.0145, **p<0.0001. (C) 954
Western blot analysis of whole cell lysates from 35000HP (lane1), 35000HP〉fis (lane 2), 955
35000HP〉cpxR (lane 3), and 35000HP〉cpxR 〉fis (lane 4). Bacterial cells were harvested at 8 h. 956
Blots were probed with LspB, Fis, and CpxR polyclonal antibodies and with the PAL MAb. 957
958
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 44
fis amiBHD0448
WT
fis
A B
amiBHD0448 cat
1 kb
fis
1 2 3 4DCWT
75
1 2 3 4
E
4 8 16 4 8 16
WT Äfis
WT
*
*
37
4 8 16 4 8 16
Fis
PAL
fis
Fis
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 45
A
B
LspB
CpxR
LspA2
LspA1
PAL
1 2 3 4
1 2 3 4 50
2
4
6
Ph
ag
oc
yto
sis
ind
ex
***ns
***
MOI 360 440 580 370 610
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 46
0
0.5
1
on
Lo
g10
-1 5
-1
-0.5
0
ativ
e Q
ua
ntit
ati
aspA
HD1895
glpF
fimA
waaA
HD1094
lspA1
ompP2B
adk
artI
recD
ftnA
HD1443
flp
dsrA ner
HD1985
HD1278
hflX
HD1309
ompA2
ompP2A
carA
-2
-1.5
Re
la on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 47
1 2 3 4 50
20
40
60
80
100
% S
urv
iva
l
DsrA
PAL
A
B
****
****
**
1 2 3 4
ns on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 48
A
Flp
PAL
1 2 3 4
BWT 〉fis
〉fis(pML164) 〉fis(pACYC177)
〉tadA
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from
Page 49
A BNotI
SalI
35000HP
35000HP∆fis
800
its
p15A
lacZ
35000HP∆fis
400
600
ve
Mille
rU
ni
*
kan
0
200
Rela
tiv
ns *
**
*
CLspB
1 2 3 4
Fis
Promoters
CpxR
PAL
on April 3, 2018 by guest
http://iai.asm.org/
Dow
nloaded from