- Email: [email protected]
Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production
Accepted Manuscript Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production Wioletta Adamus-Bialek, Tara L. Vollmerhausen, Katrin Janik PII:
S0882-4010(18)31273-7
DOI:
https://doi.org/10.1016/j.micpath.2018.11.020
Reference:
YMPAT 3257
To appear in:
Microbial Pathogenesis
Received Date: 12 July 2018 Revised Date:
12 November 2018
Accepted Date: 13 November 2018
Please cite this article as: Adamus-Bialek W, Vollmerhausen TL, Janik K, Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production, Microbial Pathogenesis (2018), doi: https://doi.org/10.1016/j.micpath.2018.11.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production
2 3
Authors and Affiliations
4
Wioletta Adamus-Bialek1,2∗, Tara L Vollmerhausen,2, Katrin Janik2
5 1
Jan Kochanowski University, Institute of Medical Sciences, Kielce, Poland
7
2
Department of Microbiology, Tumor and Cell Biology, Karolinska University Hospital & Karolinska
8
Institutet, Stockholm, Sweden
10
∗
Corresponding author: Phone +48 413496431, fax +48 413496307, e-mail: wioletta.adamus-
[email protected], 15 Swietokrzyska St., 25-406 Kielce, Poland
11 12
Key words: UPEC, ROIs, cellulose, bacterial biofilm
14
UTI: Urinary Tract Infection
15
UPEC: Uropathogenic E. coli
16
ROI: Reactive oxygen intermediate
AC C
EP
TE D
17
M AN U
13
SC
9
RI PT
6
1
ACCEPTED MANUSCRIPT Abstract
19
Reactive oxygen intermediates, such as hydrogen peroxide, are toxic molecules produced by immune
20
cells in response to bacterial invasion into the host. Bacteria try to protect themselves against the
21
immune system through specific properties such as biofilm formation. This phenomenon occurs also
22
during urinary tract infections. Cellulose is an important factor of Escherichia coli biofilm and
23
contributes to building a protective shield around bacterial cells upon the host immune response. In
24
this study, we aimed to analyze the effect of hydrogen peroxide on the production of this biofilm
25
component. To achieve this goal, 25 clinical E. coli strains isolated from patients with urinary tract
26
infections were used. These bacterial strains were characterized based on their growth characteristics,
27
their ability to form biofilm and their capacity to produce cellulose upon exposure to sub-lethal
28
concentrations of hydrogen peroxide growth, and the biofilm formation of these strains was analyzed.
29
Our results revealed that the analyzed uropathogenic E. coli strains slightly, but significantly, reduced
30
growth and biofilm production upon hydrogen peroxide treatment. However, when separating these
31
strains regarding their ability to produce cellulose, we found that general biofilm production was
32
reduced but cellulose expression was induced upon peroxide treatment. This finding contributes to a
33
better understanding of how bacterial biofilm formation is triggered and provides interesting insights
34
into how uropathogenic E. coli protect themselves in an inhospitable environment.
AC C
EP
TE D
M AN U
SC
RI PT
18
2
ACCEPTED MANUSCRIPT 35
1. Introduction Uropathogenic Escherichia coli (UPEC) is a specific pathotype phylogenetically distinct from
37
other E. coli groups [1-3]. UPEC strains have developed various mechanisms to defend themselves
38
and acquire nutrients, which helps them to survive and persist in adverse conditions of the urinary tract
39
[4-6]. One of the most important properties is the formation of specific biofilm inside the umbrella
40
cells of the bladder, known from intracellular bacteria communities. The formed biofilm consists of
41
different components [7-9].
RI PT
36
Cellulose is one of the main components of the extracellular matrix of biofilms formed by
43
Enterobacteriaceae [10-12]. Cellulose is mainly produced by plants, but it can also be produced by
44
algae, fungi and bacteria. Biochemically, this polysaccharide possesses β-1,4-glycosidic linkage and
45
forms long chains of anhydroglucose units after condensation polymerization [13]. Cellulose is
46
synthesized by the enzyme cellulose synthase encoded by the bcsA gene and regulation of its
47
expression is complex and connected with different bacterial properties and environmental factors [14-
48
17]. The different structures of the bacterial walls and their molecules gives the bacteria different
49
adhesive properties for different surfaces [18]. The curli - the first described bacterial functional
50
amyloid important for biofilm formation has been also well documented. Curli is highly ordered
51
protein aggregates with an unbranched filamentous morphology. It is worth to add that its'
52
transcriptional regulator CsgD activates also biosynthesis of cellulose [19]. The simultaneous
53
expression of cellulose and curli produce a highly inert, hydrophobic extracellular matrix around the
54
bacteria which increases bacterial resistance to adverse environmental conditions [10,20]. However,
55
the hydrophobic properties of the bacterial cell surface are provided by curli [21], while cellulose
56
increase the hydrophilicity of bacterial surface [14].
TE D
M AN U
SC
42
The development of specific adaptive mechanisms allows many bacteria to modify their cell
58
surfaces towards its hydrophobicity to permit direct hydrophobic-hydrophobic interactions with the
59
substrates [22]. E. coli cells coated with cellulose are protected against a cascade of host immune
60
responses, especially acting on neutrophils and they may be easier to maintain in the urinary tracts
61
[20,23,24]. In early stages of infection, UPEC strains stimulate Toll-like receptor 4 (TLR4), which
62
triggers the influx of polymorphonuclear leukocytes (PMNs) [25,26]. During the infection, the
63
epithelial and immune cells produce a panel of cytokines, chemokines, and inflammatory molecules
64
such as nitric oxide and reactive oxygen species (ROS) [9,26-28]. One of the sources for ROS is
65
hydrogen peroxide and significant concentrations of H2O2 above 100µM have been detected in freshly
66
voided human urine of healthy volunteers. It is anticipated that due to host defense mechanisms, H2O2
67
concentrations are even higher in urine from patients suffering from an UPEC infection [29-31].
AC C
EP
57
68
The aim of this study was to analyze the response of clinical and laboratory E. coli strains to
69
the reactive oxygen intermediate H2O2 as a proxy for ROS. Biofilm formation and cellulose
3
ACCEPTED MANUSCRIPT 70
production among a collection of clinical uropathogenic strains were correlated with the bacterial
71
growth upon treatment with hydrogen peroxide.
AC C
EP
TE D
M AN U
SC
RI PT
72
4
ACCEPTED MANUSCRIPT 73
2. Materials and Methods
74
2.1 Bacterial strain collections A total of 25 clinical uropathogenic E. coli strains were collected from children with sporadic and
76
recurrent urinary tract infections (UTIs) [32]. All strains were stored at -80°C. The expression of
77
cellulose was previously determined by observation of the colony morphology of isolates after 48 h of
78
growth on Congo red and Calcofluor [33].
RI PT
75
The clinical E. coli strain No. 71.8, which was isolated from the urine of a child with UTI, was
80
chosen as the reference strain for this study. This strain expressed cellulose and formed a high level of
81
biofilm in comparison to other clinical strains. To generate cellulose and curli deficient mutants of E.
82
coli No. 71.8, a one-step knockout of bcsA and csgBA was carried out according to the protocol [20,
83
34] with the modifications described previously [35]. The obtained mutants are designated as follows:
84
bcsA::Cm, deficient in cellulose production; csgBA::Cm, deficient in curli production; and
85
csgBA::bscA::Cm, deficient in both cellulose and curli production. The gene expression was
86
confirmed by using the knockout of one of these genes for comparison of the transcription of the
87
second gene between mutant and wild-type in strain E. coli No. 71.8. For optimal growth and biofilm
88
development, E. coli isolates were cultured in the liquid of solid Luria–Bertani (LB) media without
89
salt as previously described [35].
90 91
M AN U
SC
79
TE D
2.2 Bacterial culture with hydrogen peroxide
Approximately 106 CFU/ml were treated with different fresh concentrations of H2O2 (from 0.625
93
mM to 275 mM). Catalase (1000 units/ml, Sigma) was added to stop the reaction after 15 minutes of
94
treatment. Bacteria were serially diluted and grown on blood agar plates overnight at 37°C. The
95
number of viable bacteria were then quantified by counting colony-forming units (CFU) on blood
96
agar. The experiment was performed twice.
EP
92
E. coli strains were cultured in Luria Broth without salt with 0.625 mM of H2O2 at 37°C for 24
98
hours. Based on the growth results, 0.625 mM H2O2 was considered a subinhibitory concentration for
99
bacterial growth. Bacterial growth was determined by optical density measuring absorbance at 630
100 101 102
nm.
AC C
97
2.3 Biofilm formation assay by crystal violet staining
103
The ability to form biofilm was estimated using a microtiter plate method. Bacteria were spread
104
on LB Agar without salt and incubated for 24 hours at 37°C. A few colonies from the incubated plates
105
were resuspended in 5 ml PBS, and centrifuged at 300 × g for 10 minutes at room temperature to
106
remove bacterial aggregates. The bacterial suspension was diluted in LB broth without salt to obtain
107
approximately 103 CFU/mL. Bacteria were incubated in 96-well microtiter plates at 37°C in LB broth
5
ACCEPTED MANUSCRIPT without salt with 0.625 mM hydrogen peroxide and without hydrogen peroxide as a control. The plates
109
were incubated for 24 h at 37°C. Bacterial growth was measured at 630 nm using an ELISA plate
110
reader. To stain the formed biofilm, 0.3% (w/v) crystal violet was added to each well and incubated
111
for 5 minutes. The staining solution was discarded and the wells were washed twice with PBS and
112
after the last washing step, the PBS was removed and excess PBS was removed with a paper towel.
113
Crystal violet that had adhered to the biofilm was dissolved with 20:80 acetone/ethanol solution with
114
shaking at 200 rpm for 10 minutes. The optical density (OD) of the dissolved stain was then measured
115
at 550 nm using an ELISA plate reader. The biofilm measurement was performed at least in duplicates
116
for each condition and each strain.
117 2.4 Calcofluor binding assay for cellulose expression
SC
118
RI PT
108
Bacteria were grown statically at 37°C in 13 ml LB broth without salt, with calcofluor (0.05
120
mg/ml) and 0.625 mM hydrogen peroxide. After 24 h bacterial cells were harvested by centrifugation
121
at 3500 × g for 5 minutes. The pellet was resuspended in 1 mL 4% PFA (paraformaldehyde) in PBS.
122
Thereafter, 200 µl of culture was aliquoted to each well in triplicates. OD was measured at 630 nm
123
using an ELISA plate reader. OD was adjusted to a similar range between 0.5-0.9. The level of
124
calcofluor bound up with cellulose was measured (excitation 355 nm, emission 460 nm) using a
125
Wallac Victor2 1420 multilabel counter. Relative cellulose production was determined based on the
126
ratio between the fluorescence level to the OD at 630 nm. This method of calculation was used to
127
obtain relative cellulose production independent from bacterial growth. The experiment was
128
performed twice.
129 130
TE D
M AN U
119
2.5 Statistical analysis
Relative growth and biofilm formation were determined as the growth or biofilm ratios of H2O2
132
treated versus the respective non-treated control. Differences between the strains regarding relative
133
growth or biofilm formation were analyzed by paired (for the same strains) or un-paired (for non-
134
homogenous groups of strains) T-test, nonparametric. Differences with a P value of <0.05 were
135
considered significant. Statistical analyses were performed using GraphPad Prism, version 6 (San
136
Diego, CA, USA).
AC C
137
EP
131
138
6
ACCEPTED MANUSCRIPT 139
3. Results and discussion A collection of 25 clinical uropathogenic Escherichia coli strains were analyzed including E.
141
coli strain No. 71.8 as a reference strain. During optimization of the assay for the reference strain,
142
0.625 mM of H2O2 was identified as the minimum subinhibitory concentration of hydrogen peroxide
143
for bacterial growth of treatment in comparison to the respective control condition. However, some
144
strains survived extremely high concentrations of hydrogen peroxide (275 mM; 15 min treatment) and
145
over half of the strains were able to survive short term treatment with 137 mM of H2O2 during this
146
experiment (data not shown). This is consistent with previous work which found that the surface
147
adhesion of Antigen 43 increased survival for E. coli after 15 minutes treatment of 275 mM H2O2 [36].
148
Concentrations above 1 mM of H2O2 can be detected in freshly voided human urine of healthy adults
149
[30] and even higher concentrations of H2O2 in the urine of patients during urinary tract infections are
150
assumed in close proximity of H2O2 releasing immune cells [29-31]. However, according to our
151
results, a concentration of 1 mM of H2O2 is not sufficient for bactericidal activity against UPEC –
152
other human inflammatory components are also involved during infection. Although growth was
153
significantly inhibited by H2O2 treatment, a majority of E. coli strains (85% isolates) were able to
154
survive 24 hour exposure of H2O2 (Fig. 1). This tolerance to hydrogen peroxide varied between
155
strains, but it provides evidence that E. coli can survive high hydrogen peroxide concentrations at 275
156
mM. Besides, we could show that E. coli is much more tolerant against the subinhibitory H2O2
157
concentration than the human epithelial bladder cells line T24, which did not survive after 24 hour
158
treatment with 0.625 mM of H2O2, while a lower treatment 0.125 mM of H2O2 was not cytotoxic
159
(XTT test, data not shown). These finding can be correlated with others which have reported that 0.1
160
mM of H2O2 can be detected in freshly voided human urine, which may suggest non-toxicity for
161
human cells [30]. However, the lower values of H2O2 concentrations (0.002-0.005 mM) were
162
cytotoxic for eukaryotic cells [29-31]. The fact that higher concentrations of hydrogen peroxide of up
163
to 1 mM have been detected in voided urine of healthy individuals, despite in vitro results showing
164
lower levels to be cytotoxic to human uroepithelial cells in our cell culture model, may suggest
165
epithelial cells in vivo are better able to withstand hydrogen peroxide.
AC C
EP
TE D
M AN U
SC
RI PT
140
166
The same clinical UPEC strains were also analyzed regarding biofilm formation after 24-hour
167
treatment with 0.625 mM hydrogen peroxide. The different isolates formed varying amounts of
168
biofilm ranging from OD550 crystal violet staining 0.5 to 2.0 under non-treated conditions. A total of
169
44% of the studied strains showed decreased biofilm formation upon treatment with 0.625 mM
170
hydrogen peroxide (Fig. 2). However, when biofilm formation is normalized to bacterial growth, no
171
effect of H2O2 upon biofilm formation can be observed (Fig. 3). This is likely due to hydrogen
172
peroxide treatment inhibiting the growth of strains, which further inhibited biofilm formation. A
173
similar analysis was performed by De Pas et al. [37] which found that rugose biofilm formation by
174
the uropathogenic E. coli strain UTI89 has previously been shown to have increased resistance to
7
ACCEPTED MANUSCRIPT 175
H2O2 toxicity. Our results suggest that subinhibitory concentrations of hydrogen peroxide may slow
176
down, but not inhibit biofilm formation. Among the collection of clinical UTI strains investigated, we observed a high degree of
178
variability between growth and survival in the presence of hydrogen peroxide. Our results suggest a
179
correlation between H2O2 sensitivity of E. coli strains and the presence of certain virulence factors. It
180
has been well established that virulence factors such as fimbriae, curli and cellulose are important for
181
biofilm formation [10,19,38-41]. Our results showed that cellulose production was significantly
182
increased upon hydrogen peroxide treatment among the clinical cellulose positive strains (Fig. 4). This
183
observation was confirmed during the experiment with using the reference strains (E. coli No. 71.8)
184
and its isogenic mutant csgBA::Cm, deficient in curli production with active cellulose production (Fig.
185
5). The downregulation of csgBA had no influence on cellulose production which was consistent with
186
the results presented by Da Re et al. [42]. We also observed that expression of csgA was
187
downregulated upon treatment with H2O2 (data not shown), what can indicates a different pathway of
188
cellulose expression during treatment with H2O2. Despite the upregulation of cellulose, we were
189
unable to show a significant role of cellulose to improve biofilm formation during hydrogen peroxide
190
treatment (data not shown). Our results suggest that increased production of cellulose induced by
191
hydrogen peroxide is likely to protect against toxic ROIs or other components of the inflammatory
192
process [20,23,24]; however, it is unlikely to contribute to stronger biofilm formation. We should take
193
into account that cellulose expression could be connected with other properties of the bacteria
194
(expression of other virulence factors) and environment (structure of surface, pH, temperature etc.)
195
which could be reflected in the resistance to hydrogen peroxide. The cellulose production can modify
196
the adhesion properties of bacterial surface [22,43]. If the cellulose expression is stimulated, as in our
197
case, by the oxygen radicals, then the bacteria can become more hydrophilic and increase affinity for
198
hydrophilic surfaces [14,44]. Recent studies have revealed that the use of hydrophilic materials in
199
catheters may prevent catheter-associated UTIs [45-47]. The affinity of cellulose producing strains to
200
hydrophilic surfaces may contribute to the problem of catheter-associated infections. Certainly,
201
cellulose production and the ability to form biofilm enhance the ability for bacteria to cause urinary
202
tract infections, often, recurrent [48].
AC C
EP
TE D
M AN U
SC
RI PT
177
203
Our findings suggest that the production of free oxygen radicals during infection may stimulate
204
E. coli to increase cellulose production, which further decreases adhesion to the uroepithelium, leading
205
to easier eradication from the urinary tract. However it should be noted that bacterial response to the
206
ROIs is associated with the expression of many genes including genes encoding for other virulence
207
factors. The contact of uropathogenic E. coli with eukaryotic cells triggers a cascade of very
208
complicated events [24,27,39]. Exposure of E. coli to ROIs activates a diverse set of antioxidant genes
209
that are controlled by the SoxRS regulon, oxidative DNA repairs enzyme endonuclease IV and
210
glucose-6-phosphate dehydrogenase. Equally, H2O2 induces the thiol-containing transcriptional
8
ACCEPTED MANUSCRIPT activator OxyR. It controls the expression of genes for catalase/hydroperoxidase (katG), glutathione
212
reductase (gorA) and alkyl hydroperoxidase reductase subunit C (ahpC) in E. coli [29]. The increased
213
lag phase observed in clinical isolates exposed to H2O2 in this study may represent an adaptation
214
period whereby these pathogens were repairing intracellular damage caused by ROIs and increasing
215
production of catalase to neutralize the damaging H2O2. Pathogenic bacteria are well-equipped with a
216
whole array of mechanisms that enhance resistance to oxidative and nitrosative effector molecules of
217
the host [49-51].
RI PT
211
218 219
4. Conclusions
To conclude, this is an unceasing arms race between pathogenic bacteria and humans. In our
221
study, we show that subinhibitory concentrations of hydrogen peroxide are cytotoxic towards
222
uroepithelial cells, and that these subinhibitory concentrations are able to slow bacterial growth and
223
subsequently biofilm formation, but does not kill all bacterial cell in the population. On the other hand,
224
cellulose production was found to be stimulated by hydrogen peroxide, a factor which may impede
225
bacterial establishment in the bladder, and decrease hydrophobicity of the bacterial surface. However,
226
the study of role of csgD for the regulation of curli and cellulose expression could give an important
227
observation for further explanation of the mechanism expression of these factors during treatment with
228
hydrogen peroxide. It is also interesting how the other urospecific adhesins and toxins (e.g. cnf1,
229
hemolysin) will be expressed in the radical oxygen species environment which result from respiratory
230
burst of polymorphonuclear neutrophils. The strong support and confirmation for the obtained
231
observations would be parallel research in tissue cultures (e.g. T24). Certainly, the obtained research
232
results open a new insight into bacterial pathogenicity mechanisms.
M AN U
TE D
EP AC C
233
SC
220
9
ACCEPTED MANUSCRIPT 234
5. Acknowledgements:
235
We would like to thank Annelie Brauner for participating in this study and the critical reading of the
236
manuscript.
237
This study was funded by the National Science Centre research grant no. NCN 2011/01/D/NZ7/00107
238
and by Jan Kochanowski University Statutory Research no. 615561 in part.
AC C
EP
TE D
M AN U
SC
RI PT
239
10
ACCEPTED MANUSCRIPT 240
6. References
241
1. Dobrindt U. (Patho-)Genomics of Escherichia coli. Int J Med Microbiol 2005;295:357-37.
242
2. Elena SF, Whittam TS, Winkworth CL, Riley MA, Lenski RE. Genomic divergence of
243
Escherichia coli strains: evidence for horizontal transfer and variation in mutation rates. Int
244
Microbiol 2005;8:271-278. 3. Brzuszkiewicz E, Bruggemann H, Liesegang H, Emmerth M, Olschlager T et al. How to
246
become a uropathogen: comparative genomic analysis of extraintestinal pathogenic
247
Escherichia coli strains. Proc Natl Acad Sci U S A 2006;103:12879-12884.
RI PT
245
4. Brauner A, Katouli M, Tullus K, Jacobson S. Production of cytotoxic necrotizing factor,
249
verocytotoxin and haemolysin by pyelonephritogenic Escherichia coli. Eur J Clin Microbiol
250
Infect Dis 1990;9:762–767.
SC
248
5. Silveira WD, Benetti F, Lancellotti M, Ferreira A, Solferini VN et al. Biological and
252
genetic characteristics of uropathogenic Escherichia coli strains. Rev Inst Med Trop
253
2001;43:303-310.
254 255
M AN U
251
6. Yamamoto S. Molecular epidemiology of uropathogenic Escherichia coli. J Infect Chemother 2007;13:68-73.
256
7. Justice SS, Hung C, Theriot JA, Fletcher DA, Anderson GG et al. Differentiation and
257
developmental pathways of uropathogenic Escherichia coli in urinary tract pathogenesis. Proc
258
Natl Acad Sci U S A 2004;101:1333–1338.
8. Anderson GG, Dodson KW, Hooton TM, Hultgren SJ. Intracellular bacterial communities
260
of uropathogenic Escherichia coli in urinary tract pathogenesis. Trends Microbiol
261
2004;12:424–430.
263 264 265
9. Hunstad DA, Justice SS. Intracellular lifestyles and immune evasion strategies of uropathogenic Escherichia coli. Annu Rev Microbiol 2010;64:203-221.
EP
262
TE D
259
10. Barnhart MM, Chapman MR. Curli biogenesis and function. Annu Rev Microbiol 2006;60:131–147.
11. Zogaj X, Nimtz M, Rohde M, Bokranz W, Römling U. The multicellular morphotypes of
267
Salmonella typhimurium and Escherichia coli produce cellulose as the second component of
268 269 270
AC C
266
the extracellular matrix. Mol Microbiol 2001;39(6):1452-63.
12. Solano C, Garcia B, Valle J, Berasain C, Ghigo JM et al. Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Mol Microbiol 2002;43: 793–808.
271
13. Abeer MM, Amin MCIM, Martin C. A review of bacterial cellulose-based drug delivery
272
systems: their biochemistry, current approaches and future prospects. J Pharm Pharmacol
273
2014;66:1047–1061.
274
14. Ma Q, Wood TK. OmpA influences Escherichia coli biofilm formation by repressing
275
cellulose production through the CpxRA two-component system. Environ Microbiol
276
2009;11(10):2735-46.
11
ACCEPTED MANUSCRIPT 277
15. Gualdi L, Tagliabue L, Bertagnoli S, Ieranò T, De Castro C et al. Cellulose modulates
278
biofilm formation by counteracting curli-mediated colonization of solid surfaces in
279
Escherichia coli. Microbiology 2008;154(7):2017-24.
280
16. Monteiro C, Saxena I, Wang X, Kader A, Bokranz W et al. Characterization of cellulose
281
production in Escherichia coli Nissle 1917 and its biological consequences. Environ
282
Microbiol 2009;11:1105–1116.
285 286 287 288
RI PT
284
17. Römling U. Molecular biology of cellulose production in bacteria. Res Microbiol 2002;153:205–212.
18. Houdt RV, Michiels CW. Role of bacterial cell surface structures in Escherichia coli biofilm formation. Res Microbiol 2005;156(5-6):626-33.
19. Gerven VN, Van der Verren SE, Reiter DM, Remaut H. The Role of Functional Amyloids
SC
283
in Bacterial Virulence. J Mol Biol 2018;430(20):3657-3684.
20. Kai-Larsen Y, Luthje P, Chromek M, Peters V, Wang X et al. Uropathogenic Escherichia
290
coli modulates immune responses and its curli fimbriae interact with the antimicrobial peptide
291
LL-37. PLoS Pathog 2010;6:1-16.
M AN U
289
292
21. Patel J, Sharma M, Ravishakar S. Effect of curli expression and hydrophobicity of
293
Escherichia coli O157:H7 on attachment to fresh produce surfaces. J Appl Microbiol
294
2011;110(3):737-45.
296
22. Krasowska A, Sigler K. How microorganisms use hydrophobicity and what does this mean for human needs? Front Cell Infect Microbiol 2014;4:112.
TE D
295 297
23. Zhou G, Mo WJ, Sebbel P, Min G, Neubert TA et al. Uroplakin Ia is the urothelial receptor
298
for uropathogenic Escherichia coli: evidence from in vitro FimH binding. J Cell Sci
299
2001;114(22):4095-103.
24. Backhed F, Soderhall M, Ekman P, Normark S, Richter-Dahlfors A. Induction of innate
301
immune responses by Escherichia coli and purified lipopolysaccharide correlate with organ-
302
and cell-specific expression of Toll-like receptors within the human urinary tract. Cell
303
Microbiol 2001;3:153–158.
305 306 307
AC C
304
EP
300
25. Agace WW. The role of the epithelial cell in Escherichia coli induced neutrophil migration into the urinary tract. Eur Respir J 1996;9:1713–1728.
26. Mazzulli T. Resistance trends in urinary tract pathogens and impact on management. J Urol 2002;168:1720-17222.
308
27. Blomgran R, Zheng L, Stendahl O. Uropathogenic Escherichia coli triggers oxygen-
309
dependent apoptosis in human neutrophils through the cooperative effect of type 1 fimbriae
310
and lipopolysaccharide. Inf and Imm 2004;72:4570–4578.
311 312
28. Bian Z, Brauner A, Li Y, Normark S. Expression of and cytokine activation by Escherichia coli curli fibers in human sepsis. J Infect Dis 2000;181:602–612.
12
ACCEPTED MANUSCRIPT 313
29. Long LH, Evans PJ, Halliwell B. Hydrogen Peroxide in human urine: implications for
314
antioxidant defense and redox regulation. Biochem Biophys Res Commun 1999;262:605–609.
315
30. Halliwell B, Clement MV, Ramalingam J, Long LH. Hydrogen peroxide. Ubiquitous in cell
317 318 319 320
culture and in vivo? IUBMB Life 2000;50:251–257. 31. Halliwell B, Clement MV, Long LH. Hydrogen peroxide in the human body. FEBS Lett 2000;486:10-13. 32. Lüthje P, Brauner A. Ag43 promotes persistence of uropathogenic Escherichia coli isolates in the urinary tract. J Clin Microbiol 2010;48(6):2316–2317.
RI PT
316
33. Ramos NL, Dzung DT, Stopsack K, Jankó V, Pourshafie MR et al. Characterization of
322
uropathogenic Escherichia coli from children with urinary tract infection in different
323
countries. Eur J Clin Microbiol Infect Dis 2011;30(12):1587-93.
324 325
SC
321
34. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 2000;97:6640–6645. 35. Bokranz W, Wang X, Tschape H, Romling U. Expression of cellulose and curli fimbriae by
327
Escherichia coli isolated from the gastrointestinal tract. J Med Microbiol 2005;54:1171–1182.
328
36. Schembri MA, Hjerrild L, Gjermansen M, Klemm P. Differential expression of the
329
M AN U
326
Escherichia coli autoaggregation factor antigen 43. J Bacteriol 2003;185(7):2236-42. 37. De Pas WH, Hufnagel DA, Lee JS, Blanco LP, Bernstein HC et al. Iron induces bimodal
331
population development by Escherichia coli. Proc Natl Acad Sci U S A 2013;12;110(7):2629-
332
34.
TE D
330
333
38. Kudinha T, Johnson JR, Andrew SD, Kong F, Anderson P et al. Genotypic and
334
phenotypic characterization of Escherichia coli isolates from children with urinary tract
335
infection and from healthy carriers. Pediatr Infect Dis J 2013;32(5):543-8.
338 339
EP
337
39. Lüthje P, Brauner A. Virulence factors of uropathogenic E. coli and their interaction with the host. Adv Microb Physiol 2014;65:337-72. 40. Lüthje P, Brauner A. Novel strategies in the prevention and treatment of urinary tract infections. Pathogens 2016;27:5(1).
AC C
336
340
41. Kikuchi T, Mizunoe Y, Takade A, Naito S, Yoshida S. Curli fibers are required for
341
development of biofilm architecture in Escherichia coli K-12 and enhance bacterial adherence
342 343 344
to human uroepithelial cells. Microbiol Immunol 2005;49:875–848.
42. Da Re S, Ghigo JM. A CsgD-independent pathway for cellulose production and biofilm formation in Escherichia coli. J of Bact 2006;188:3073–3087.
345
43. Gonçalves S, Rodrigues IP, Padrão J, Silva JP, Sencadas V et al. Acetylated bacterial
346
cellulose coated with urinary bladder matrix as a substrate for retinal pigment epithelium.
347
Colloids Surf B Biointerfaces 2016;139:1-9.
348
44. Adamus-Białek W, Kubiak A, Czerwonka G. Analysis of uropathogenic Escherichia coli
349
biofilm formation under different growth conditions. Acta Biochim Pol 2015;62(4):765-71.
13
ACCEPTED MANUSCRIPT 350
45. De Ridder DJ, Everaert K, Fernández LG, Valero JV, Durán AB et al. Intermittent
351
catheterisation with hydrophilic-coated catheters (SpeediCath) reduces the risk of clinical
352
urinary tract infection in spinal cord injured patients: a prospective randomised parallel
353
comparative trial. Eur Urol 2005;48(6):991–5. 46. Cardenas DD, Hoffman JM. Hydrophilic catheters versus noncoated catheters for reducing
355
the incidence of urinary tract infections: a randomized controlled trial. Arch Phys Med
356
Rehabil 2009;90(10):1668–71.
RI PT
354
47. Rognoni C, Tarricone R. Intermittent catheterisation with hydrophilic and non-hydrophilic
358
urinary catheters: systematic literature review and meta-analyses. BMC Urol 2017;17(1):4.
359
48. Norinder BS, Luthje P, Yadav M, Kadas L, Fang H et al. Cellulose and PapG are
360
important for Escherichia coli causing recurrent urinary tract infection in women. Infection
361
2011;39:571–574.
363 364 365
49. Bogdan C, Röllinghoff M, Diefenbach A. Reactive oxygen and reactive nitrogen intermediates in innate and specific immunity. Curr Opin Immunol 2000;12:64–76.
M AN U
362
SC
357
50. Fux CA, Costerton JW, Stewart PS, Stoodley P. Survival strategies of infectious biofilms. Trends Microbiol 2005;13:34–40.
51. Tukel C, Raffatellu M, Humphries AD, Wilson RP, Andrews-Polymenis HL et al. CsgA
367
is a pathogen-associated molecular pattern of Salmonella enterica serotype Typhimurium that
368
is recognized by Toll-like receptor 2. Mol Microbiol 2005;58: 289–304.
AC C
EP
TE D
366
14
ACCEPTED MANUSCRIPT Figures
370
Fig. 1 Growth of clinical uropathogenic E. coli strains upon treatment with 0.625 mM of
371
H2O2. The growth of E. coli isolates (n=25) collected from urine of patients with urinary tract
372
infections was measured after 24 h culture at 37°C in LB without salt. Individual values and
373
medians are presented, depicted as optical density (OD) at 640 nm. Individual levels and
374
medians are presented. The study was performed in four replications in two independent
375
experiments.
376
nonparametric).
The
difference
is
significant
377
(P<0.0001,
RI PT
369
paired
T-test,
two-tailed,
Fig. 2 Biofilm formation of the clinical uropathogenic E. coli strains upon treatment with
379
0.625 mM of H2O2. The clinical uropathogenic E. coli strains (n=25) were isolated from
380
patients with urinary tract infection. The biofilm was analyzed based on the OD value of the
381
absorbed crystal violet (0.3%) measured at 530 nm (A530). The study was performed in four
382
replications in two independent experiments. The difference is significant (P<0.05, unpaired
383
T-test, two-tailed, nonparametric).
384
M AN U
SC
378
Fig. 3 Normalized biofilm formation of clinical uropathogenic E. coli strains upon treatment
386
with 0.625 mM of H2O2. The clinical uropathogenic E. coli strains (n=25) were isolated from
387
patients with urinary tract infection. Biofilm was analyzed based on the ratio between the
388
absorbed crystal violet (0.3%) measured at OD 530 nm (A530) and OD level of the growth
389
(A630). The study was performed in four replications, in two independent experiments. The
390
difference is significant (P<0.05, unpaired T-test, two-tailed, nonparametric).
EP
391
TE D
385
Fig. 4 Relative cellulose production of E. coli strains upon treatment with 0.625 mM of H2O2
393
compared to the control conditions (without H2O2). Only the cellulose positive E. coli strains
394
(n=11) were analyzed. The cellulose production method was measured using Calcoflour
395
fluorescence. The relative cellulose production was estimated based on the ratio between
396
fluorescence level and OD level of the growth (A630) upon treatment. Values were then
397
compared to the relative cellulose production upon control condition (indicated by the dotted
398
line). The study was performed in four replications in two independent experiments. The
399
difference is significant (P<0.0001, paired T-test, two-tailed, nonparametric).
AC C
392
400 401
Fig. 5 Relative cellulose production of E. coli No. 71.8 wild-type and mutant. Cellulose
402
production was estimated for the wild-type and the isogenic mutant csgBA::Cm, deficient in 15
ACCEPTED MANUSCRIPT curli production of E. coli strain No. 71.8 upon different growth conditions (0.625 mM of an
404
H2O2, control). The relative cellulose production was determined by normalizing the
405
fluorescence level to OD level of the growth (A630) upon treatment with H2O2 and then
406
compared to the relative cellulose production upon control condition (indicated by dotted
407
line). The study was performed in four replications in two independent experiments. The
408
difference is significant (P<0.0001, paired T-test, two-tailed, nonparametric).
AC C
EP
TE D
M AN U
SC
RI PT
403
16
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights
RI PT SC M AN U TE D EP
•
Uropathogenic E. coli strains can survive in a highly toxic environment such as free radicals. The production of free oxygen radicals during infection stimulates E. coli to increase cellulose production. The affinity of cellulose producing strains to hydrophilic surfaces may intensify the problem of catheter-associated infections.
AC C
• •
S0882-4010(18)31273-7
DOI:
https://doi.org/10.1016/j.micpath.2018.11.020
Reference:
YMPAT 3257
To appear in:
Microbial Pathogenesis
Received Date: 12 July 2018 Revised Date:
12 November 2018
Accepted Date: 13 November 2018
Please cite this article as: Adamus-Bialek W, Vollmerhausen TL, Janik K, Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production, Microbial Pathogenesis (2018), doi: https://doi.org/10.1016/j.micpath.2018.11.020. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT 1
Hydrogen peroxide stimulates uropathogenic Escherichia coli strains to cellulose production
2 3
Authors and Affiliations
4
Wioletta Adamus-Bialek1,2∗, Tara L Vollmerhausen,2, Katrin Janik2
5 1
Jan Kochanowski University, Institute of Medical Sciences, Kielce, Poland
7
2
Department of Microbiology, Tumor and Cell Biology, Karolinska University Hospital & Karolinska
8
Institutet, Stockholm, Sweden
10
∗
Corresponding author: Phone +48 413496431, fax +48 413496307, e-mail: wioletta.adamus-
[email protected], 15 Swietokrzyska St., 25-406 Kielce, Poland
11 12
Key words: UPEC, ROIs, cellulose, bacterial biofilm
14
UTI: Urinary Tract Infection
15
UPEC: Uropathogenic E. coli
16
ROI: Reactive oxygen intermediate
AC C
EP
TE D
17
M AN U
13
SC
9
RI PT
6
1
ACCEPTED MANUSCRIPT Abstract
19
Reactive oxygen intermediates, such as hydrogen peroxide, are toxic molecules produced by immune
20
cells in response to bacterial invasion into the host. Bacteria try to protect themselves against the
21
immune system through specific properties such as biofilm formation. This phenomenon occurs also
22
during urinary tract infections. Cellulose is an important factor of Escherichia coli biofilm and
23
contributes to building a protective shield around bacterial cells upon the host immune response. In
24
this study, we aimed to analyze the effect of hydrogen peroxide on the production of this biofilm
25
component. To achieve this goal, 25 clinical E. coli strains isolated from patients with urinary tract
26
infections were used. These bacterial strains were characterized based on their growth characteristics,
27
their ability to form biofilm and their capacity to produce cellulose upon exposure to sub-lethal
28
concentrations of hydrogen peroxide growth, and the biofilm formation of these strains was analyzed.
29
Our results revealed that the analyzed uropathogenic E. coli strains slightly, but significantly, reduced
30
growth and biofilm production upon hydrogen peroxide treatment. However, when separating these
31
strains regarding their ability to produce cellulose, we found that general biofilm production was
32
reduced but cellulose expression was induced upon peroxide treatment. This finding contributes to a
33
better understanding of how bacterial biofilm formation is triggered and provides interesting insights
34
into how uropathogenic E. coli protect themselves in an inhospitable environment.
AC C
EP
TE D
M AN U
SC
RI PT
18
2
ACCEPTED MANUSCRIPT 35
1. Introduction Uropathogenic Escherichia coli (UPEC) is a specific pathotype phylogenetically distinct from
37
other E. coli groups [1-3]. UPEC strains have developed various mechanisms to defend themselves
38
and acquire nutrients, which helps them to survive and persist in adverse conditions of the urinary tract
39
[4-6]. One of the most important properties is the formation of specific biofilm inside the umbrella
40
cells of the bladder, known from intracellular bacteria communities. The formed biofilm consists of
41
different components [7-9].
RI PT
36
Cellulose is one of the main components of the extracellular matrix of biofilms formed by
43
Enterobacteriaceae [10-12]. Cellulose is mainly produced by plants, but it can also be produced by
44
algae, fungi and bacteria. Biochemically, this polysaccharide possesses β-1,4-glycosidic linkage and
45
forms long chains of anhydroglucose units after condensation polymerization [13]. Cellulose is
46
synthesized by the enzyme cellulose synthase encoded by the bcsA gene and regulation of its
47
expression is complex and connected with different bacterial properties and environmental factors [14-
48
17]. The different structures of the bacterial walls and their molecules gives the bacteria different
49
adhesive properties for different surfaces [18]. The curli - the first described bacterial functional
50
amyloid important for biofilm formation has been also well documented. Curli is highly ordered
51
protein aggregates with an unbranched filamentous morphology. It is worth to add that its'
52
transcriptional regulator CsgD activates also biosynthesis of cellulose [19]. The simultaneous
53
expression of cellulose and curli produce a highly inert, hydrophobic extracellular matrix around the
54
bacteria which increases bacterial resistance to adverse environmental conditions [10,20]. However,
55
the hydrophobic properties of the bacterial cell surface are provided by curli [21], while cellulose
56
increase the hydrophilicity of bacterial surface [14].
TE D
M AN U
SC
42
The development of specific adaptive mechanisms allows many bacteria to modify their cell
58
surfaces towards its hydrophobicity to permit direct hydrophobic-hydrophobic interactions with the
59
substrates [22]. E. coli cells coated with cellulose are protected against a cascade of host immune
60
responses, especially acting on neutrophils and they may be easier to maintain in the urinary tracts
61
[20,23,24]. In early stages of infection, UPEC strains stimulate Toll-like receptor 4 (TLR4), which
62
triggers the influx of polymorphonuclear leukocytes (PMNs) [25,26]. During the infection, the
63
epithelial and immune cells produce a panel of cytokines, chemokines, and inflammatory molecules
64
such as nitric oxide and reactive oxygen species (ROS) [9,26-28]. One of the sources for ROS is
65
hydrogen peroxide and significant concentrations of H2O2 above 100µM have been detected in freshly
66
voided human urine of healthy volunteers. It is anticipated that due to host defense mechanisms, H2O2
67
concentrations are even higher in urine from patients suffering from an UPEC infection [29-31].
AC C
EP
57
68
The aim of this study was to analyze the response of clinical and laboratory E. coli strains to
69
the reactive oxygen intermediate H2O2 as a proxy for ROS. Biofilm formation and cellulose
3
ACCEPTED MANUSCRIPT 70
production among a collection of clinical uropathogenic strains were correlated with the bacterial
71
growth upon treatment with hydrogen peroxide.
AC C
EP
TE D
M AN U
SC
RI PT
72
4
ACCEPTED MANUSCRIPT 73
2. Materials and Methods
74
2.1 Bacterial strain collections A total of 25 clinical uropathogenic E. coli strains were collected from children with sporadic and
76
recurrent urinary tract infections (UTIs) [32]. All strains were stored at -80°C. The expression of
77
cellulose was previously determined by observation of the colony morphology of isolates after 48 h of
78
growth on Congo red and Calcofluor [33].
RI PT
75
The clinical E. coli strain No. 71.8, which was isolated from the urine of a child with UTI, was
80
chosen as the reference strain for this study. This strain expressed cellulose and formed a high level of
81
biofilm in comparison to other clinical strains. To generate cellulose and curli deficient mutants of E.
82
coli No. 71.8, a one-step knockout of bcsA and csgBA was carried out according to the protocol [20,
83
34] with the modifications described previously [35]. The obtained mutants are designated as follows:
84
bcsA::Cm, deficient in cellulose production; csgBA::Cm, deficient in curli production; and
85
csgBA::bscA::Cm, deficient in both cellulose and curli production. The gene expression was
86
confirmed by using the knockout of one of these genes for comparison of the transcription of the
87
second gene between mutant and wild-type in strain E. coli No. 71.8. For optimal growth and biofilm
88
development, E. coli isolates were cultured in the liquid of solid Luria–Bertani (LB) media without
89
salt as previously described [35].
90 91
M AN U
SC
79
TE D
2.2 Bacterial culture with hydrogen peroxide
Approximately 106 CFU/ml were treated with different fresh concentrations of H2O2 (from 0.625
93
mM to 275 mM). Catalase (1000 units/ml, Sigma) was added to stop the reaction after 15 minutes of
94
treatment. Bacteria were serially diluted and grown on blood agar plates overnight at 37°C. The
95
number of viable bacteria were then quantified by counting colony-forming units (CFU) on blood
96
agar. The experiment was performed twice.
EP
92
E. coli strains were cultured in Luria Broth without salt with 0.625 mM of H2O2 at 37°C for 24
98
hours. Based on the growth results, 0.625 mM H2O2 was considered a subinhibitory concentration for
99
bacterial growth. Bacterial growth was determined by optical density measuring absorbance at 630
100 101 102
nm.
AC C
97
2.3 Biofilm formation assay by crystal violet staining
103
The ability to form biofilm was estimated using a microtiter plate method. Bacteria were spread
104
on LB Agar without salt and incubated for 24 hours at 37°C. A few colonies from the incubated plates
105
were resuspended in 5 ml PBS, and centrifuged at 300 × g for 10 minutes at room temperature to
106
remove bacterial aggregates. The bacterial suspension was diluted in LB broth without salt to obtain
107
approximately 103 CFU/mL. Bacteria were incubated in 96-well microtiter plates at 37°C in LB broth
5
ACCEPTED MANUSCRIPT without salt with 0.625 mM hydrogen peroxide and without hydrogen peroxide as a control. The plates
109
were incubated for 24 h at 37°C. Bacterial growth was measured at 630 nm using an ELISA plate
110
reader. To stain the formed biofilm, 0.3% (w/v) crystal violet was added to each well and incubated
111
for 5 minutes. The staining solution was discarded and the wells were washed twice with PBS and
112
after the last washing step, the PBS was removed and excess PBS was removed with a paper towel.
113
Crystal violet that had adhered to the biofilm was dissolved with 20:80 acetone/ethanol solution with
114
shaking at 200 rpm for 10 minutes. The optical density (OD) of the dissolved stain was then measured
115
at 550 nm using an ELISA plate reader. The biofilm measurement was performed at least in duplicates
116
for each condition and each strain.
117 2.4 Calcofluor binding assay for cellulose expression
SC
118
RI PT
108
Bacteria were grown statically at 37°C in 13 ml LB broth without salt, with calcofluor (0.05
120
mg/ml) and 0.625 mM hydrogen peroxide. After 24 h bacterial cells were harvested by centrifugation
121
at 3500 × g for 5 minutes. The pellet was resuspended in 1 mL 4% PFA (paraformaldehyde) in PBS.
122
Thereafter, 200 µl of culture was aliquoted to each well in triplicates. OD was measured at 630 nm
123
using an ELISA plate reader. OD was adjusted to a similar range between 0.5-0.9. The level of
124
calcofluor bound up with cellulose was measured (excitation 355 nm, emission 460 nm) using a
125
Wallac Victor2 1420 multilabel counter. Relative cellulose production was determined based on the
126
ratio between the fluorescence level to the OD at 630 nm. This method of calculation was used to
127
obtain relative cellulose production independent from bacterial growth. The experiment was
128
performed twice.
129 130
TE D
M AN U
119
2.5 Statistical analysis
Relative growth and biofilm formation were determined as the growth or biofilm ratios of H2O2
132
treated versus the respective non-treated control. Differences between the strains regarding relative
133
growth or biofilm formation were analyzed by paired (for the same strains) or un-paired (for non-
134
homogenous groups of strains) T-test, nonparametric. Differences with a P value of <0.05 were
135
considered significant. Statistical analyses were performed using GraphPad Prism, version 6 (San
136
Diego, CA, USA).
AC C
137
EP
131
138
6
ACCEPTED MANUSCRIPT 139
3. Results and discussion A collection of 25 clinical uropathogenic Escherichia coli strains were analyzed including E.
141
coli strain No. 71.8 as a reference strain. During optimization of the assay for the reference strain,
142
0.625 mM of H2O2 was identified as the minimum subinhibitory concentration of hydrogen peroxide
143
for bacterial growth of treatment in comparison to the respective control condition. However, some
144
strains survived extremely high concentrations of hydrogen peroxide (275 mM; 15 min treatment) and
145
over half of the strains were able to survive short term treatment with 137 mM of H2O2 during this
146
experiment (data not shown). This is consistent with previous work which found that the surface
147
adhesion of Antigen 43 increased survival for E. coli after 15 minutes treatment of 275 mM H2O2 [36].
148
Concentrations above 1 mM of H2O2 can be detected in freshly voided human urine of healthy adults
149
[30] and even higher concentrations of H2O2 in the urine of patients during urinary tract infections are
150
assumed in close proximity of H2O2 releasing immune cells [29-31]. However, according to our
151
results, a concentration of 1 mM of H2O2 is not sufficient for bactericidal activity against UPEC –
152
other human inflammatory components are also involved during infection. Although growth was
153
significantly inhibited by H2O2 treatment, a majority of E. coli strains (85% isolates) were able to
154
survive 24 hour exposure of H2O2 (Fig. 1). This tolerance to hydrogen peroxide varied between
155
strains, but it provides evidence that E. coli can survive high hydrogen peroxide concentrations at 275
156
mM. Besides, we could show that E. coli is much more tolerant against the subinhibitory H2O2
157
concentration than the human epithelial bladder cells line T24, which did not survive after 24 hour
158
treatment with 0.625 mM of H2O2, while a lower treatment 0.125 mM of H2O2 was not cytotoxic
159
(XTT test, data not shown). These finding can be correlated with others which have reported that 0.1
160
mM of H2O2 can be detected in freshly voided human urine, which may suggest non-toxicity for
161
human cells [30]. However, the lower values of H2O2 concentrations (0.002-0.005 mM) were
162
cytotoxic for eukaryotic cells [29-31]. The fact that higher concentrations of hydrogen peroxide of up
163
to 1 mM have been detected in voided urine of healthy individuals, despite in vitro results showing
164
lower levels to be cytotoxic to human uroepithelial cells in our cell culture model, may suggest
165
epithelial cells in vivo are better able to withstand hydrogen peroxide.
AC C
EP
TE D
M AN U
SC
RI PT
140
166
The same clinical UPEC strains were also analyzed regarding biofilm formation after 24-hour
167
treatment with 0.625 mM hydrogen peroxide. The different isolates formed varying amounts of
168
biofilm ranging from OD550 crystal violet staining 0.5 to 2.0 under non-treated conditions. A total of
169
44% of the studied strains showed decreased biofilm formation upon treatment with 0.625 mM
170
hydrogen peroxide (Fig. 2). However, when biofilm formation is normalized to bacterial growth, no
171
effect of H2O2 upon biofilm formation can be observed (Fig. 3). This is likely due to hydrogen
172
peroxide treatment inhibiting the growth of strains, which further inhibited biofilm formation. A
173
similar analysis was performed by De Pas et al. [37] which found that rugose biofilm formation by
174
the uropathogenic E. coli strain UTI89 has previously been shown to have increased resistance to
7
ACCEPTED MANUSCRIPT 175
H2O2 toxicity. Our results suggest that subinhibitory concentrations of hydrogen peroxide may slow
176
down, but not inhibit biofilm formation. Among the collection of clinical UTI strains investigated, we observed a high degree of
178
variability between growth and survival in the presence of hydrogen peroxide. Our results suggest a
179
correlation between H2O2 sensitivity of E. coli strains and the presence of certain virulence factors. It
180
has been well established that virulence factors such as fimbriae, curli and cellulose are important for
181
biofilm formation [10,19,38-41]. Our results showed that cellulose production was significantly
182
increased upon hydrogen peroxide treatment among the clinical cellulose positive strains (Fig. 4). This
183
observation was confirmed during the experiment with using the reference strains (E. coli No. 71.8)
184
and its isogenic mutant csgBA::Cm, deficient in curli production with active cellulose production (Fig.
185
5). The downregulation of csgBA had no influence on cellulose production which was consistent with
186
the results presented by Da Re et al. [42]. We also observed that expression of csgA was
187
downregulated upon treatment with H2O2 (data not shown), what can indicates a different pathway of
188
cellulose expression during treatment with H2O2. Despite the upregulation of cellulose, we were
189
unable to show a significant role of cellulose to improve biofilm formation during hydrogen peroxide
190
treatment (data not shown). Our results suggest that increased production of cellulose induced by
191
hydrogen peroxide is likely to protect against toxic ROIs or other components of the inflammatory
192
process [20,23,24]; however, it is unlikely to contribute to stronger biofilm formation. We should take
193
into account that cellulose expression could be connected with other properties of the bacteria
194
(expression of other virulence factors) and environment (structure of surface, pH, temperature etc.)
195
which could be reflected in the resistance to hydrogen peroxide. The cellulose production can modify
196
the adhesion properties of bacterial surface [22,43]. If the cellulose expression is stimulated, as in our
197
case, by the oxygen radicals, then the bacteria can become more hydrophilic and increase affinity for
198
hydrophilic surfaces [14,44]. Recent studies have revealed that the use of hydrophilic materials in
199
catheters may prevent catheter-associated UTIs [45-47]. The affinity of cellulose producing strains to
200
hydrophilic surfaces may contribute to the problem of catheter-associated infections. Certainly,
201
cellulose production and the ability to form biofilm enhance the ability for bacteria to cause urinary
202
tract infections, often, recurrent [48].
AC C
EP
TE D
M AN U
SC
RI PT
177
203
Our findings suggest that the production of free oxygen radicals during infection may stimulate
204
E. coli to increase cellulose production, which further decreases adhesion to the uroepithelium, leading
205
to easier eradication from the urinary tract. However it should be noted that bacterial response to the
206
ROIs is associated with the expression of many genes including genes encoding for other virulence
207
factors. The contact of uropathogenic E. coli with eukaryotic cells triggers a cascade of very
208
complicated events [24,27,39]. Exposure of E. coli to ROIs activates a diverse set of antioxidant genes
209
that are controlled by the SoxRS regulon, oxidative DNA repairs enzyme endonuclease IV and
210
glucose-6-phosphate dehydrogenase. Equally, H2O2 induces the thiol-containing transcriptional
8
ACCEPTED MANUSCRIPT activator OxyR. It controls the expression of genes for catalase/hydroperoxidase (katG), glutathione
212
reductase (gorA) and alkyl hydroperoxidase reductase subunit C (ahpC) in E. coli [29]. The increased
213
lag phase observed in clinical isolates exposed to H2O2 in this study may represent an adaptation
214
period whereby these pathogens were repairing intracellular damage caused by ROIs and increasing
215
production of catalase to neutralize the damaging H2O2. Pathogenic bacteria are well-equipped with a
216
whole array of mechanisms that enhance resistance to oxidative and nitrosative effector molecules of
217
the host [49-51].
RI PT
211
218 219
4. Conclusions
To conclude, this is an unceasing arms race between pathogenic bacteria and humans. In our
221
study, we show that subinhibitory concentrations of hydrogen peroxide are cytotoxic towards
222
uroepithelial cells, and that these subinhibitory concentrations are able to slow bacterial growth and
223
subsequently biofilm formation, but does not kill all bacterial cell in the population. On the other hand,
224
cellulose production was found to be stimulated by hydrogen peroxide, a factor which may impede
225
bacterial establishment in the bladder, and decrease hydrophobicity of the bacterial surface. However,
226
the study of role of csgD for the regulation of curli and cellulose expression could give an important
227
observation for further explanation of the mechanism expression of these factors during treatment with
228
hydrogen peroxide. It is also interesting how the other urospecific adhesins and toxins (e.g. cnf1,
229
hemolysin) will be expressed in the radical oxygen species environment which result from respiratory
230
burst of polymorphonuclear neutrophils. The strong support and confirmation for the obtained
231
observations would be parallel research in tissue cultures (e.g. T24). Certainly, the obtained research
232
results open a new insight into bacterial pathogenicity mechanisms.
M AN U
TE D
EP AC C
233
SC
220
9
ACCEPTED MANUSCRIPT 234
5. Acknowledgements:
235
We would like to thank Annelie Brauner for participating in this study and the critical reading of the
236
manuscript.
237
This study was funded by the National Science Centre research grant no. NCN 2011/01/D/NZ7/00107
238
and by Jan Kochanowski University Statutory Research no. 615561 in part.
AC C
EP
TE D
M AN U
SC
RI PT
239
10
ACCEPTED MANUSCRIPT 240
6. References
241
1. Dobrindt U. (Patho-)Genomics of Escherichia coli. Int J Med Microbiol 2005;295:357-37.
242
2. Elena SF, Whittam TS, Winkworth CL, Riley MA, Lenski RE. Genomic divergence of
243
Escherichia coli strains: evidence for horizontal transfer and variation in mutation rates. Int
244
Microbiol 2005;8:271-278. 3. Brzuszkiewicz E, Bruggemann H, Liesegang H, Emmerth M, Olschlager T et al. How to
246
become a uropathogen: comparative genomic analysis of extraintestinal pathogenic
247
Escherichia coli strains. Proc Natl Acad Sci U S A 2006;103:12879-12884.
RI PT
245
4. Brauner A, Katouli M, Tullus K, Jacobson S. Production of cytotoxic necrotizing factor,
249
verocytotoxin and haemolysin by pyelonephritogenic Escherichia coli. Eur J Clin Microbiol
250
Infect Dis 1990;9:762–767.
SC
248
5. Silveira WD, Benetti F, Lancellotti M, Ferreira A, Solferini VN et al. Biological and
252
genetic characteristics of uropathogenic Escherichia coli strains. Rev Inst Med Trop
253
2001;43:303-310.
254 255
M AN U
251
6. Yamamoto S. Molecular epidemiology of uropathogenic Escherichia coli. J Infect Chemother 2007;13:68-73.
256
7. Justice SS, Hung C, Theriot JA, Fletcher DA, Anderson GG et al. Differentiation and
257
developmental pathways of uropathogenic Escherichia coli in urinary tract pathogenesis. Proc
258
Natl Acad Sci U S A 2004;101:1333–1338.
8. Anderson GG, Dodson KW, Hooton TM, Hultgren SJ. Intracellular bacterial communities
260
of uropathogenic Escherichia coli in urinary tract pathogenesis. Trends Microbiol
261
2004;12:424–430.
263 264 265
9. Hunstad DA, Justice SS. Intracellular lifestyles and immune evasion strategies of uropathogenic Escherichia coli. Annu Rev Microbiol 2010;64:203-221.
EP
262
TE D
259
10. Barnhart MM, Chapman MR. Curli biogenesis and function. Annu Rev Microbiol 2006;60:131–147.
11. Zogaj X, Nimtz M, Rohde M, Bokranz W, Römling U. The multicellular morphotypes of
267
Salmonella typhimurium and Escherichia coli produce cellulose as the second component of
268 269 270
AC C
266
the extracellular matrix. Mol Microbiol 2001;39(6):1452-63.
12. Solano C, Garcia B, Valle J, Berasain C, Ghigo JM et al. Genetic analysis of Salmonella enteritidis biofilm formation: critical role of cellulose. Mol Microbiol 2002;43: 793–808.
271
13. Abeer MM, Amin MCIM, Martin C. A review of bacterial cellulose-based drug delivery
272
systems: their biochemistry, current approaches and future prospects. J Pharm Pharmacol
273
2014;66:1047–1061.
274
14. Ma Q, Wood TK. OmpA influences Escherichia coli biofilm formation by repressing
275
cellulose production through the CpxRA two-component system. Environ Microbiol
276
2009;11(10):2735-46.
11
ACCEPTED MANUSCRIPT 277
15. Gualdi L, Tagliabue L, Bertagnoli S, Ieranò T, De Castro C et al. Cellulose modulates
278
biofilm formation by counteracting curli-mediated colonization of solid surfaces in
279
Escherichia coli. Microbiology 2008;154(7):2017-24.
280
16. Monteiro C, Saxena I, Wang X, Kader A, Bokranz W et al. Characterization of cellulose
281
production in Escherichia coli Nissle 1917 and its biological consequences. Environ
282
Microbiol 2009;11:1105–1116.
285 286 287 288
RI PT
284
17. Römling U. Molecular biology of cellulose production in bacteria. Res Microbiol 2002;153:205–212.
18. Houdt RV, Michiels CW. Role of bacterial cell surface structures in Escherichia coli biofilm formation. Res Microbiol 2005;156(5-6):626-33.
19. Gerven VN, Van der Verren SE, Reiter DM, Remaut H. The Role of Functional Amyloids
SC
283
in Bacterial Virulence. J Mol Biol 2018;430(20):3657-3684.
20. Kai-Larsen Y, Luthje P, Chromek M, Peters V, Wang X et al. Uropathogenic Escherichia
290
coli modulates immune responses and its curli fimbriae interact with the antimicrobial peptide
291
LL-37. PLoS Pathog 2010;6:1-16.
M AN U
289
292
21. Patel J, Sharma M, Ravishakar S. Effect of curli expression and hydrophobicity of
293
Escherichia coli O157:H7 on attachment to fresh produce surfaces. J Appl Microbiol
294
2011;110(3):737-45.
296
22. Krasowska A, Sigler K. How microorganisms use hydrophobicity and what does this mean for human needs? Front Cell Infect Microbiol 2014;4:112.
TE D
295 297
23. Zhou G, Mo WJ, Sebbel P, Min G, Neubert TA et al. Uroplakin Ia is the urothelial receptor
298
for uropathogenic Escherichia coli: evidence from in vitro FimH binding. J Cell Sci
299
2001;114(22):4095-103.
24. Backhed F, Soderhall M, Ekman P, Normark S, Richter-Dahlfors A. Induction of innate
301
immune responses by Escherichia coli and purified lipopolysaccharide correlate with organ-
302
and cell-specific expression of Toll-like receptors within the human urinary tract. Cell
303
Microbiol 2001;3:153–158.
305 306 307
AC C
304
EP
300
25. Agace WW. The role of the epithelial cell in Escherichia coli induced neutrophil migration into the urinary tract. Eur Respir J 1996;9:1713–1728.
26. Mazzulli T. Resistance trends in urinary tract pathogens and impact on management. J Urol 2002;168:1720-17222.
308
27. Blomgran R, Zheng L, Stendahl O. Uropathogenic Escherichia coli triggers oxygen-
309
dependent apoptosis in human neutrophils through the cooperative effect of type 1 fimbriae
310
and lipopolysaccharide. Inf and Imm 2004;72:4570–4578.
311 312
28. Bian Z, Brauner A, Li Y, Normark S. Expression of and cytokine activation by Escherichia coli curli fibers in human sepsis. J Infect Dis 2000;181:602–612.
12
ACCEPTED MANUSCRIPT 313
29. Long LH, Evans PJ, Halliwell B. Hydrogen Peroxide in human urine: implications for
314
antioxidant defense and redox regulation. Biochem Biophys Res Commun 1999;262:605–609.
315
30. Halliwell B, Clement MV, Ramalingam J, Long LH. Hydrogen peroxide. Ubiquitous in cell
317 318 319 320
culture and in vivo? IUBMB Life 2000;50:251–257. 31. Halliwell B, Clement MV, Long LH. Hydrogen peroxide in the human body. FEBS Lett 2000;486:10-13. 32. Lüthje P, Brauner A. Ag43 promotes persistence of uropathogenic Escherichia coli isolates in the urinary tract. J Clin Microbiol 2010;48(6):2316–2317.
RI PT
316
33. Ramos NL, Dzung DT, Stopsack K, Jankó V, Pourshafie MR et al. Characterization of
322
uropathogenic Escherichia coli from children with urinary tract infection in different
323
countries. Eur J Clin Microbiol Infect Dis 2011;30(12):1587-93.
324 325
SC
321
34. Datsenko KA, Wanner BL. One-step inactivation of chromosomal genes in Escherichia coli K-12 using PCR products. Proc Natl Acad Sci U S A 2000;97:6640–6645. 35. Bokranz W, Wang X, Tschape H, Romling U. Expression of cellulose and curli fimbriae by
327
Escherichia coli isolated from the gastrointestinal tract. J Med Microbiol 2005;54:1171–1182.
328
36. Schembri MA, Hjerrild L, Gjermansen M, Klemm P. Differential expression of the
329
M AN U
326
Escherichia coli autoaggregation factor antigen 43. J Bacteriol 2003;185(7):2236-42. 37. De Pas WH, Hufnagel DA, Lee JS, Blanco LP, Bernstein HC et al. Iron induces bimodal
331
population development by Escherichia coli. Proc Natl Acad Sci U S A 2013;12;110(7):2629-
332
34.
TE D
330
333
38. Kudinha T, Johnson JR, Andrew SD, Kong F, Anderson P et al. Genotypic and
334
phenotypic characterization of Escherichia coli isolates from children with urinary tract
335
infection and from healthy carriers. Pediatr Infect Dis J 2013;32(5):543-8.
338 339
EP
337
39. Lüthje P, Brauner A. Virulence factors of uropathogenic E. coli and their interaction with the host. Adv Microb Physiol 2014;65:337-72. 40. Lüthje P, Brauner A. Novel strategies in the prevention and treatment of urinary tract infections. Pathogens 2016;27:5(1).
AC C
336
340
41. Kikuchi T, Mizunoe Y, Takade A, Naito S, Yoshida S. Curli fibers are required for
341
development of biofilm architecture in Escherichia coli K-12 and enhance bacterial adherence
342 343 344
to human uroepithelial cells. Microbiol Immunol 2005;49:875–848.
42. Da Re S, Ghigo JM. A CsgD-independent pathway for cellulose production and biofilm formation in Escherichia coli. J of Bact 2006;188:3073–3087.
345
43. Gonçalves S, Rodrigues IP, Padrão J, Silva JP, Sencadas V et al. Acetylated bacterial
346
cellulose coated with urinary bladder matrix as a substrate for retinal pigment epithelium.
347
Colloids Surf B Biointerfaces 2016;139:1-9.
348
44. Adamus-Białek W, Kubiak A, Czerwonka G. Analysis of uropathogenic Escherichia coli
349
biofilm formation under different growth conditions. Acta Biochim Pol 2015;62(4):765-71.
13
ACCEPTED MANUSCRIPT 350
45. De Ridder DJ, Everaert K, Fernández LG, Valero JV, Durán AB et al. Intermittent
351
catheterisation with hydrophilic-coated catheters (SpeediCath) reduces the risk of clinical
352
urinary tract infection in spinal cord injured patients: a prospective randomised parallel
353
comparative trial. Eur Urol 2005;48(6):991–5. 46. Cardenas DD, Hoffman JM. Hydrophilic catheters versus noncoated catheters for reducing
355
the incidence of urinary tract infections: a randomized controlled trial. Arch Phys Med
356
Rehabil 2009;90(10):1668–71.
RI PT
354
47. Rognoni C, Tarricone R. Intermittent catheterisation with hydrophilic and non-hydrophilic
358
urinary catheters: systematic literature review and meta-analyses. BMC Urol 2017;17(1):4.
359
48. Norinder BS, Luthje P, Yadav M, Kadas L, Fang H et al. Cellulose and PapG are
360
important for Escherichia coli causing recurrent urinary tract infection in women. Infection
361
2011;39:571–574.
363 364 365
49. Bogdan C, Röllinghoff M, Diefenbach A. Reactive oxygen and reactive nitrogen intermediates in innate and specific immunity. Curr Opin Immunol 2000;12:64–76.
M AN U
362
SC
357
50. Fux CA, Costerton JW, Stewart PS, Stoodley P. Survival strategies of infectious biofilms. Trends Microbiol 2005;13:34–40.
51. Tukel C, Raffatellu M, Humphries AD, Wilson RP, Andrews-Polymenis HL et al. CsgA
367
is a pathogen-associated molecular pattern of Salmonella enterica serotype Typhimurium that
368
is recognized by Toll-like receptor 2. Mol Microbiol 2005;58: 289–304.
AC C
EP
TE D
366
14
ACCEPTED MANUSCRIPT Figures
370
Fig. 1 Growth of clinical uropathogenic E. coli strains upon treatment with 0.625 mM of
371
H2O2. The growth of E. coli isolates (n=25) collected from urine of patients with urinary tract
372
infections was measured after 24 h culture at 37°C in LB without salt. Individual values and
373
medians are presented, depicted as optical density (OD) at 640 nm. Individual levels and
374
medians are presented. The study was performed in four replications in two independent
375
experiments.
376
nonparametric).
The
difference
is
significant
377
(P<0.0001,
RI PT
369
paired
T-test,
two-tailed,
Fig. 2 Biofilm formation of the clinical uropathogenic E. coli strains upon treatment with
379
0.625 mM of H2O2. The clinical uropathogenic E. coli strains (n=25) were isolated from
380
patients with urinary tract infection. The biofilm was analyzed based on the OD value of the
381
absorbed crystal violet (0.3%) measured at 530 nm (A530). The study was performed in four
382
replications in two independent experiments. The difference is significant (P<0.05, unpaired
383
T-test, two-tailed, nonparametric).
384
M AN U
SC
378
Fig. 3 Normalized biofilm formation of clinical uropathogenic E. coli strains upon treatment
386
with 0.625 mM of H2O2. The clinical uropathogenic E. coli strains (n=25) were isolated from
387
patients with urinary tract infection. Biofilm was analyzed based on the ratio between the
388
absorbed crystal violet (0.3%) measured at OD 530 nm (A530) and OD level of the growth
389
(A630). The study was performed in four replications, in two independent experiments. The
390
difference is significant (P<0.05, unpaired T-test, two-tailed, nonparametric).
EP
391
TE D
385
Fig. 4 Relative cellulose production of E. coli strains upon treatment with 0.625 mM of H2O2
393
compared to the control conditions (without H2O2). Only the cellulose positive E. coli strains
394
(n=11) were analyzed. The cellulose production method was measured using Calcoflour
395
fluorescence. The relative cellulose production was estimated based on the ratio between
396
fluorescence level and OD level of the growth (A630) upon treatment. Values were then
397
compared to the relative cellulose production upon control condition (indicated by the dotted
398
line). The study was performed in four replications in two independent experiments. The
399
difference is significant (P<0.0001, paired T-test, two-tailed, nonparametric).
AC C
392
400 401
Fig. 5 Relative cellulose production of E. coli No. 71.8 wild-type and mutant. Cellulose
402
production was estimated for the wild-type and the isogenic mutant csgBA::Cm, deficient in 15
ACCEPTED MANUSCRIPT curli production of E. coli strain No. 71.8 upon different growth conditions (0.625 mM of an
404
H2O2, control). The relative cellulose production was determined by normalizing the
405
fluorescence level to OD level of the growth (A630) upon treatment with H2O2 and then
406
compared to the relative cellulose production upon control condition (indicated by dotted
407
line). The study was performed in four replications in two independent experiments. The
408
difference is significant (P<0.0001, paired T-test, two-tailed, nonparametric).
AC C
EP
TE D
M AN U
SC
RI PT
403
16
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT Highlights
RI PT SC M AN U TE D EP
•
Uropathogenic E. coli strains can survive in a highly toxic environment such as free radicals. The production of free oxygen radicals during infection stimulates E. coli to increase cellulose production. The affinity of cellulose producing strains to hydrophilic surfaces may intensify the problem of catheter-associated infections.
AC C
• •