Subscriber access provided by DUESSELDORF LIBRARIES
Article
Physiochemical Properties of Caulobacter crescentus Holdfast: A Localized Bacterial Adhesive Cecile Chantal Berne, Xiang Ma, Nicholas A. Licata, Bernardo Ruegger Almeida Neves, Sima Setayeshgar, Yves V. Brun, and Bogdan Dragnea J. Phys. Chem. B, Just Accepted Manuscript • DOI: 10.1021/jp405802e • Publication Date (Web): 07 Aug 2013 Downloaded from http://pubs.acs.org on August 20, 2013
Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.
The Journal of Physical Chemistry B is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.
Page 1 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
1
Physiochemical Properties of Caulobacter crescentus Holdfast: a Localized
2
Bacterial Adhesive
3
Cécile Berne †, Xiang Ma ‡, Nicholas A. Licata §, Bernardo R.A. Neves /, Sima
4
Setayeshgar //, Yves V. Brun* †, Bogdan Dragnea* ‡
5 6
†
Department of Biology, Indiana University, Bloomington, IN 47405, USA
7
‡
Department of Chemistry, Indiana University, Bloomington, IN 47405, USA
8
§
Department of Natural Sciences, University of Michigan-Dearborn, Dearborn, MI 48128,
9
USA
10
/
11
//
Universidade Federal de Minas Gerais, Belo Horizonte, MG 30123-970, Brazil. Department of Physics, Indiana University, Bloomington, IN 47405, USA
12
13
* Address correspondence to:
[email protected] and
[email protected] 14 15
16
KEYWORDS. Caulobacter crescentus, bacterial bioadhesive, atomic force microscopy,
17
dynamic force spectrometry, adhesion kinetics
18 1 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
19
20
Page 2 of 43
ABSTRACT
To colonize surfaces, the bacterium Caulobacter crescentus employs a polar
21
polysaccharide, the holdfast, located at the end of a thin, long stalk protruding from the
22
cell body. Unlike many other bacteria which adhere through an extended extracellular
23
polymeric network, the holdfast footprint area is tens of thousands times smaller than
24
that of the total bacterium cross-sectional surface, making for some very demanding
25
adhesion requirements. At present, the mechanism of holdfast adhesion remains poorly
26
understood. We explore it here along three lines of investigation: a) the impact of
27
environmental conditions on holdfast binding affinity, b) adhesion kinetics by dynamic
28
force spectroscopy, and c) kinetic modeling of the attachment process to interpret the
29
observed time-dependence of the adhesion force at short and long time scales. A picture
30
emerged in which discrete molecular units called adhesins are responsible for initial
31
holdfast adhesion, by acting in a cooperative manner.
32
33 34 35
INTRODUCTION Biological adhesives found in the bacterial world are an abundant, yet mostly
36
untapped, source of adhesives with varied composition and properties. They hold
37
promise for industrial and medical applications, offering impressive performance in their
38
natural context: they enable attachment to a broad variety of surfaces, share desirable
39
properties, such as sustainability, biodegradability and biocompatibily, and yield a much-
40
reduced impact on the environment compared to their synthetic counterparts 1.
2 ACS Paragon Plus Environment
Page 3 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Most bacteria are found attached to surfaces where they form large groups of
41 42
cells, called biofilms. A critical step in biofilm formation is the initial single cell
43
attachment, which proceeds from a reversible stage, often mediated by proteinaceous
44
appendages like flagella and pili, to an irreversible stage mediated by polysaccharide
45
adhesins 2,3. After biofilm maturation, a dense extracellular matrix mainly composed of
46
polysaccharides usually encloses the cells and helps to maintain adhesive properties in
47
a broad range of aqueous environments and on a variety of surfaces 4. Earlier studies
48
attempted to describe the viscoelastic properties of biofilms in terms of
49
phenomenological parameters 5,6 and have shed some light on the mechanisms that
50
mediate the transition from the reversible to the irreversible stage of bacterial adhesion 7-
51
11
52
atomic force microscopy (AFM) tips 13, or quartz crystal microbalances 14 to study the
53
development of bacterial adhesion forces on different surfaces. Additionally, single-
54
molecule force microscopy has enabled direct measurement of the elasticity of single
55
biomacromolecules, including bacterial polysaccharides involved in adhesion 15,16. Most
56
of these studies were done on bacterial species which having a mixture of
57
polysaccharides of different composition and structure and appendages such as pili and
58
flagella distributed around the cell surface. However, a few bacterial species of the
59
Alphaproteobacteria group adhere to surfaces using a discrete, microscopic patch of
60
polysaccharide-based adhesive 17,18.
61
. Other studies have used flow displacement systems 12, bacterial cells immobilized on
Such localized anchor points have received much less attention in the past
62
probably mainly due to the difficulty of adhesion measurements at sub-micron scale.
63
Nevertheless, their study is timely because it is reasonable to believe that they must be
64
coping with mechanical stress differently than extended adhesive films. In this work, we
65
quantitatively explore the physiochemical properties responsible for adhesion in a
66
member of this group, Caulobacter crescentus. 3 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
67
C. crescentus synthesizes its holdfast adhesin during the differentiation of the
68
motile swarmer cell into a sessile stalked cell (Figure 1A). The holdfast patch (~100 nm
69
diameter) responsible for permanent adhesion to surfaces is found at the tip of a thin
70
cylindrical stalk-like extension (~ 1 µm long) of the cell envelope 19-23. The holdfast has
71
mechanical properties characteristic of an elastic gel 24 and outperforms the strongest
72
biological and commercial glues, with a force of adhesion exceeding 68 N/mm2,
73
sufficient to resist to a variety of stresses including fluid flow and capillary forces 25. The
74
holdfast elastic modulus was estimated at approximately 2.5 x 104 Pa 26, comparable to
75
other biological gels such as collagen or gelatin matrices 26,27.
76
Page 4 of 43
Specific binding of wheat germ agglutinin lectin to the holdfast and its sensitivity
77
to lysozyme indicate that holdfast contains ß-1,4 N-acetylglucosamine polymers 24,28.
78
However, its detailed composition and structure remain largely unknown, due to its
79
strong adhesiveness and inherent insolubility. While oligomers of ß-1,4 N-
80
acetylglucosamine confer gel-like properties on holdfasts 24 and may play a major role in
81
holdfast elastic properties, available data strongly suggest the existence of additional
82
adhesive components in the holdfast 24,28.
83
In this study, we report the first analysis of the development of adhesive forces in a
84
microscopic holdfast anchor through measurements of the time-dependence of holdfast
85
rupture forces on a variety of substrates. Having access to a Caulobacter mutant which
86
sheds holdfast free of cellular components 29, we were able to determine that holdfast
87
morphology is dependent on the surface to which it is bound and that holdfast affinity for
88
a substrate is modulated by hydrophobic interactions and depends on buffer ionic
89
strength and pH. In addition, we evaluated the maximum tensile strength of pure
90
holdfast, free of interference from cellular components, in relation to the nature of the
91
substrate to which it was attached. To this end, we employed Dynamic Force
92
Spectroscopy (DFS) to measure rupture forces between the holdfast and the substrate. 4 ACS Paragon Plus Environment
Page 5 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
93
This method provides both the spatial and temporal resolution required for bridging the
94
molecular and mesoscopic scales at which crucial phenomena take place. We found that
95
holdfast adhesion is strongly time-dependent, involving transformations on multiple time
96
scales. We further demonstrate that the observed time-dependence is well described by
97
a kinetic rate model of adhesin-surface interaction coupled to diffusion of molecular
98
adhesins within the bulk of the holdfast. Our DFS results also show that the initial
99
adhesion of holdfast to surfaces is dependent on the substrate hydrophobicity and
100
roughness. Finally, our data suggest that the GlcNac polymers present in the holdfast
101
and the holdfast anchor proteins are not likely to be major players in the adhesion
102
mechanism, and that cooperative contributions from discrete adhesive units within the
103
holdfast are dominantly responsible for initial adhesion. These findings provide a
104
framework for future molecular mechanistic studies and for comparison of bacterial
105
holdfast properties with the more extensively studied case of adhesive extracellular
106
matrices.
107 108
EXPERIMENTAL SECTION
109
Bacterial strains and growth conditions. The main strain used in this study was
110
Caulobacter crescentus CB15 ∆hfaB (YB4251) 29, a mutant strain from C. crescentus
111
CB15 wild-type (YB135). This mutant has a clean deletion of the hfaB gene and
112
therefore does not synthesize HfaB, one of the holdfast anchor proteins. This strain still
113
produces a holdfast, but is unable to anchor it to the cell envelope. As a consequence,
114
the newly synthesized holdfast is shed in the culture medium and on surfaces 29.
115
Another C. crescentus CB15 mutant, ∆hfsH (YB2198), was used to study the role
116
of deacethylation in adhesion efficiency. Indeed, this mutant is lacking the gene hfsH,
117
encoding a deacetylase that affects both cohesive and adhesive properties of the 5 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
118
holdfast 30. C. crescentus ∆hfsH produces smaller holdfasts compared to the wild-type
119
and the ∆hfaB strains. These fully acetylated holdfasts are not anchored properly to the
120
cell envelope and are shed in the medium 30.
121
C. crescentus strains were grown at 30˚C in minimal M2 medium supplemented
122
with 0.2% glucose (M2G) 31 or in low phosphate HIGG medium 32 containing 120 µM
123
phosphate (for stalk sample preparations).
124
Page 6 of 43
Escherichia coli TRMG (MG1655 csrA::kan), a strain that overproduces the
125
polysaccharide PGA (poly-ß-1,6- N-acetylglucosamine polymer) and releases it in the
126
culture medium 2 was grown in LB medium at 37˚C with constant shaking (150 rpm), to
127
maximize PGA production and release 2.
128 129
PGA purification. PGA was purified from E. coli TRMG stationary phase cultures (24 h
130
at 37˚C), as described previously 2. Cells were harvested by centrifugation and the
131
supernatant (8 ml) was concentrated using MWCO 3,000 Centricon units (Millipore) to
132
500 µl final.
133 134
Glass treatment. A hydrophobic treatment was performed on 12 mm glass coverslips
135
(#26020, Ted Pella Inc.). Coverslips were incubated with a 1:1 3-trimethoxysilyl propyl
136
methacrylate (3-TMSM, Acros Organics): anhydrous dimethylformamide (Acros
137
Organics) mixture for 2 h, rinsed twice using 100% acetone and then air-dried.
138
Hydrophobic coverslips were used within a day of treatment.
139 140
Purified holdfast affinity assay. Purified holdfast affinity assays were performed as
141
described previously 33, with few modifications. C. crescentus ∆hfaB cells were grown to
142
late exponential phase (OD600 of 0.6 – 0.8) and cells were pelleted by centrifugation (30 6 ACS Paragon Plus Environment
Page 7 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
143
min at 4,000 g). The supernatant contains free holdfasts shed by the cells. 100 µl of
144
purified holdfasts in solution were spotted on a 12 mm borosilicate glass coverslip
145
(#26020, Ted Pella Inc.), previously glued to a microscope glass slide, and incubated for
146
4 h at room temperature in a saturated humidity chamber. After incubation, the slides
147
were rinsed with dH2O to remove unbound material. Holdfasts were visualized by
148
labeling using AlexaFluor 488 (AF488) conjugated Wheat Germ Agglutinin (WGA)
149
(Molecular Probes). WGA binds specifically to the N-acetylglucosamine residues of the
150
holdfast 28. AF488-labeled WGA (50 µl at 5 µg/ml) was added to the rinsed coverslips
151
and incubated in the dark for 20 min at room temperature. Slides were then rinsed with
152
dH2O, toped with a large glass coverslip (24 x 50 mm) and sealed with nail polish.
153
Holdfast attachment to the coverslips was visualized by epifluorescence microscopy
154
using a Nikon Eclipse 90i and a Photometrics Cascade 1K EMCCD camera. Fluorescent
155
holdfasts were quantified using ImageJ analysis software 34: microscopy 16-bit pictures
156
were manually thresholded using the B/W default setting and fluorescent particles were
157
automatically analyzed with the ImageJ built in function.
158 159
pH sensitivity assays. Purified holdfast binding assays under different pH conditions
160
were performed in 100 mM citrate-phosphate or sodium-acetate buffers (from pH 2.6 to
161
pH 6), 100 mM phosphate or Tris buffers (from pH 6 to pH 8) and N-cyclohexyl-3-
162
aminopropanesulfonic acid (CAPS) buffers (from pH 8 to pH 12). 50 µl of purified
163
holdfasts in suspension were mixed with 50 µl 100 mM buffer and incubated on glass
164
coverslips for 4 h at room temperature in a humid chamber, as described above. Bound
165
holdfast labeling, imaging and quantification were performed as described above. PGA
166
binding assays were run under the same conditions, but incubated for 24 h instead of 4
167
h, to maximize binding. 7 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
168
Page 8 of 43
To determine if the low binding efficiency of holdfasts at low pH (< 6) or high pH
169
(> 8) was due to a physical or a chemical modification of the holdfasts, 50 µl purified
170
holdfasts were incubated in suspension with 25 µl of 100 mM buffers at different pH for 2
171
h at room temperature (1st incubation). 50 µl of new buffer were added to the samples to
172
modify their pH, and the samples were allowed to bind to coverslips for 2 h, as described
173
above (2nd incubation). Bound holdfast labeling, imaging and quantification were
174
performed as described above.
175 176
Sample preparation for Atomic Force Microscopy analysis. Early exponential phase
177
grown C. crescentus ∆hfaB cells (OD600 of 0.3 – 0.4) were diluted to an OD of 0.1 in
178
M2G and spotted on a 10 x 10 mm piece of freshly cleaved mica. Samples were
179
incubated at room temperature in a humid chamber. After overnight incubation, the mica
180
was thoroughly rinsed with sterile dH2O to remove all cells and debris. A 100 µl aliquot of
181
sterile dH2O was placed on the surface for DFS experiments.
182
Typically, an AFM image was taken from the holdfast-covered mica surface prior
183
to the experiment. To cover the AFM silicon nitrite tip with holdfast, the tip was placed in
184
contact with a holdfast present on the mica surface for 90 s. Using a trigger force of 5 nN
185
insured maximal tip penetration (down to the substrate). This procedure was repeated
186
several times, until a part of the holdfast present on the mica had been transferred to the
187
AFM tip. A second AFM image was then taken to ensure that part of the holdfast was
188
missing from the surface and was therefore attached to the tip. To confirm holdfast
189
attachment to the tip, the holdfast-loaded tip was moved above a clean mica surface
190
while maintained in dH2O, and a force-displacement curve was recorded to ensure a
191
significant force due to the holdfast coating the tip.
8 ACS Paragon Plus Environment
Page 9 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
192 193
The Journal of Physical Chemistry
The same procedure was followed to coat the AFM tip using purified PGA previously immobilized on a clean mica surface.
194 195
Scanning Electron Microscopy (SEM). The typical area occupied by and the thickness
196
of holdfast attached to the AFM tip were determined using a FEG environmental SEM
197
(Quanta 600F, FEI). Samples were first fixed with 2.5% (v/v) electron microscopy grade
198
glutaraldehyde (Ted Pella, Inc.) in 10 mM phosphate buffer pH 7 for 1.5 h. The samples
199
were then treated with a series of ethanol dehydration steps (30%, 50%, 70%, 90% and
200
100% (v/v), 15 minutes each) and dried using a critical point dryer (Blazers CPD 030).
201
Uncoated samples were affixed to a metal stub with double-stick conductive carbon tape
202
(Electron Microscopy Sciences) and then visualized under secondary electron mode.
203 204
AFM analysis. AFM AC mode images and force-displacement curves were obtained
205
using a Cypher AFM (Asylum Research). Measurements were performed in sterile dH2O
206
at room temperature using gold-coated silicon nitride Biolever cantilevers (Frequency f0
207
= 13 kHz, spring constant k = 0.006 N/m, Olympus Inc.). Spring constants were
208
measured from the thermal noise spectrum of the cantilevers. Force measurements
209
were performed using tips previously coated with holdfasts as described above, using a
210
trigger force of 500 pN if not stated otherwise.
211 212
RESULTS AND DISCUSSION
213 214 215 216
Holdfast on polar and non-polar surfaces Previous biophysical studies on holdfast adhesion have been performed using only one type of surface: borosilicate glass 24,25. To determine if capillary forces may
9 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 10 of 43
217
influence significantly initial adhesion, we investigated the morphology of holdfasts
218
bound to two surfaces of different polar character: hydrophilic mica and hydrophobic
219
highly-ordered pyrolitic graphite. We used AFM to image holdfasts at 16 h after
220
attachment (Figure 2A). Height, diameter and contact angles were thus determined for
221
~200 particles (Figure 2). Holdfast height varied from 5 to 100 nm on both surfaces, with
222
a few holdfasts reaching up to 160 nm (Figure 2B). The average height was 30.6 ± 2.4
223
nm and 21.5 ± 0.9 on mica and graphite, respectively. The average holdfast footprint
224
diameter was also substrate dependent (Figure 2C). Holdfasts attached to mica had
225
diameters from 30 to 280 nm, with an average of 90.2 ± 2.7 nm, while holdfasts attached
226
to graphite ranged from 45 to 440 nm, with an average of 119.2 ± 4.1 nm. The average
227
contact angles were 52.6 ± 1.3 ˚ and 38.9 ± 2 ˚ on mica and graphite, respectively
228
(Figure 2 D-F). Since contact angles reflect the relative strength of holdfast-liquid,
229
substrate-liquid, and holdfast-substrate interaction, results in Figure 2 suggest that the
230
graphite-holdfast interaction is stronger than mica-holdfast interaction. Note that both
231
graphite and mica substrate preparations yield atomically-flat surfaces, thus minimizing a
232
possible role played by roughness.
233
Another substrate of interest is glass. Having established that the holdfast
234
contact angle showed marked differences between graphite and mica, we measured the
235
binding affinity of holdfasts to clean and non-polar adsorbate (3-TMSM) coated glass
236
surfaces. In these experiments, purified holdfasts in suspension were allowed to bind to
237
the two types of surfaces, and the amount of surface-deposited holdfasts were quantified
238
using fluorescently-labeled wheat germ agglutinin (WGA), a lectin specific for N-
239
acetylglucosamine residues present in the holdfast 28 (Figure 3A). Figure 3C shows that
240
the binding affinity to hydrophobic 3-TMSM-treated glass seems somewhat smaller (~ 55
241
%) than that of clean glass, which seems to disagree with the finding above that
10 ACS Paragon Plus Environment
Page 11 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
242
attractive capillary forces are stronger on hydrophobic substrates. One explanation that
243
would reconcile these apparently contradictory results is the possibility of a time-
244
dependent curing process, which occurs at slower time scales than those characteristic
245
of capillary interactions. Indeed, we will discuss later in the paper the independent
246
evidence for such processes.
247
To determine if the only identified component, N-acetylglucosamine plays a role
248
in the dependence of adhesion efficiency on substrate polarity, we measured the affinity
249
of PGA (a poly-ß-1,6- N-acetylglucosamine polymer purified from E. coli TRMG 2) for the
250
two types of glass substrates (Figure 3B). Due to very low binding affinity, PGA samples
251
had to be incubated for 16 h to provide measurable coatings (instead of 4 h for holdfast
252
samples). PGA affinity assays exhibited no significant variation as a function of substrate
253
(Figure 3C). These results suggest that the holdfast adhesion mechanism is not
254
dominated by the N-acetylglucosamine adhesive properties and is likely to involve
255
additional components. This hypothesis is also supported by the fact that we had to
256
incubate the PGA samples four times longer than the holdfast ones to obtain
257
measurable coverages.
258
Nevertheless, as we are showing in the following, N-acetylglucosamine plays an
259
important albeit indirect role. Thus, a recent study showed that, in C. crescentus, a
260
mutation in the hfsH gene encoding a deacetylase acting on the holdfast, affects both
261
cohesive and adhesive properties of the holdfast 30. Partial deacetylation of N-
262
acetylglucosamine in holdfast should leave free amine residues in place of acetyl
263
groups, thereby changing the charge of the polysaccharide. Biologically, the free amine
264
group could be useful to covalently link an adhesin or for crosslinking. Indeed, binding
265
affinity of holdfasts produced by the ∆hfsH deacetylase mutant is drastically decreased
266
with around 30-40% of holdfasts attached compared to deacetylated ∆hfaB holdfasts 11 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 12 of 43
267
(Figure 3C). This result is in agreement with previous studies 30 and strongly suggests
268
that N-acetylglucosamine deacetylation is crucial for holdfast adhesive properties. The
269
phenomenon is reminiscent of the effect of deacetylation of chitin, a long chain polymer
270
of N-acetylglucosamine, to produce adhesive chitosan 35,36. Note that binding affinity of
271
∆hfsH holdfasts is not significantly different on the hydrophobic and hydrophilic glass
272
(Figure 3C), similarly to PGA.
273
In summary, these results suggest that (i) N-acetylglucosamine is not solely
274
responsible for holdfast adhesion; rather, active components, here referred to as
275
adhesins, dispersed in the holdfast bulk, generate the stronger adhesion and may be
276
responsible for the observed differential surface response, and (ii) N-acetylglucosamine
277
deacetylation mediated by HfsH is important for establishing a cohesive network, and
278
possibly interconnecting adhesins. Further experiments should focus on elucidating the
279
nature of the putative adhesins described in this work; one possibility would be for the
280
adhesin to be a protein or peptide, comparable to bacterial fimbriae protein subunits 37 ,
281
the mussel Mytilus edulis foot proteins 38, or gingipain adhesin peptides 39.
282 283
284
Ionic strength and pH affect holdfast affinity In order to further identify characteristics of adhesin-surface interaction, we
285
investigated the possible role of electrostatics interactions between substrate and
286
holdfast. Thus, purified holdfast binding to glass at different NaCl concentrations was
287
quantified using fluorescence labeling (Figure 4A). PGA binding affinity was found to be
288
insensitive to added salt (Figure 4B), indicating that the adhesive properties of the N-
289
acetylglucosamine molecules were not affected by ionic strength. Similarly, the adhesive
290
properties of fully acetylated holdfasts from the C. crescentus ∆hfsH mutant holdfasts
291
were not affected by ionic strength (Figure 4C). 12 ACS Paragon Plus Environment
Page 13 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
292
The Journal of Physical Chemistry
In stark contrast to PGA and holdfasts produced by the ∆hfsH mutant, binding
293
affinity on hydrophilic clean glass of deacetylated holdfasts decreased visibly with
294
increasing the NaCl concentration. On 3-TMSM-treated glass, surface coverage was
295
insensitive to salt concentration. This behavior points to the occurrence of attractive
296
interactions between charged or polar groups of deacetylated holdfast and the polar
297
glass surface. Since the principal mechanism by which glass and silica surfaces acquire
298
a charge in contact with water is the dissociation of silanol groups, glass is negatively
299
charged at close to neutral pH. At the same time amines in the deacetylated holdfast are
300
positively charged. Silanols can be gradually deprotonated in aqueous solution by
301
adjusting pH. Thus, to further confirm the origin of the electrostatic interaction, holdfast
302
binding assays were next performed in solutions at different pH (Figure 5A).
303
For clean glass, binding affinity rapidly increased from acidic to neutral pH, with
304
roughly 15% and 40% of surface binding at pH 2 and pH 5 respectively, to reach 100%
305
at pH 6.5 to 7.5. Under the same conditions, the binding affinity for 3-TMSM treated
306
glass remained approximately constant (55 to 75% for pH ranging from 2 to 7.5). This
307
observation supports the hypothesis that ionization of silanol groups, which is at least
308
partly suppressed on the 3-TMSM treated glass surface, is responsible for the observed
309
electrostatic interaction. At the same time, PGA binding is insensitive to ionic strength
310
(Figure 4B), therefore other moieties than N-acetylglucosamine and carrying positive
311
charges must be involved from the holdfast side.
312
At pHs higher than 8 and for both types of surfaces, holdfast binding affinity
313
dropped steeply and at the same rate (Figure 5A). PGA binding is greater at acidic pH
314
and the maximal binding affinity of PGA on clean glass occurred around pH 5-6 (Figure
315
5B), whereas neutral pH (6.5 to 7.5) was optimal for holdfast binding efficiency (Figure
316
5A). At basic pH, PGA binding efficiency decreased drastically, being completely 13 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 14 of 43
317
abolished at pH higher than 8 (Figure 5B). For fully acetylated ∆hfsH holdfasts, maximal
318
binding was achieved at pH 4-5 and decreased steadily with increasing pH (Figure 5C).
319
When purified holdfasts were incubated on clean glass at pH 5 for 2 h and subsequently
320
adjusted the pH at 7 for an additional 2 h, binding affinity was partially restored (75%,
321
compared to 40% if the total incubation was performed at pH 5, Figure 5A). In contrast,
322
the effects of basic pH were irreversible (Figure 5D). Since the drop of affinity at basic
323
pH occurs for both holdfast and PGA on both surfaces and is irreversible (Figure 5D) we
324
hypothesize that the decrease in affinity at basic pH may involve degradation of the N-
325
acetylglucosamine matrix, likely through base hydrolysis 40.
326 327 328
Time-dependent adhesion studies by dynamic force spectroscopy The smaller angle of contact on hydrophobic surfaces (Figure 2 D-E) suggested
329
stronger holdfast/surface interactions on this type of substrate and therefore the
330
possibility of hydrophobic interactions between adhesins and substrate. However, affinity
331
results indicated more frequent binding to clean glass than to TMSM-coated glass. A
332
hypothesis that could reconcile these facts is the existence of a curing process that may
333
occur after adsorption. This hypothesis is supported by previous work, which indicated
334
that the individual holdfast footprint on the surface increases with time as it is
335
synthesized after initial surface contact 7, suggesting that the holdfast is initially in a fluid
336
state and stops spreading after reaching a 60-200 nm footprint 41. Moreover, surface-
337
holdfast bonds are extremely strong for samples incubated overnight (~ 68 N nm-2) 25,
338
but possibly much weaker initially, thus allowing the organism to explore its environment
339
before binding irreversibly.
14 ACS Paragon Plus Environment
Page 15 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Since prior to this work it was not known what the initial adhesion forces may be,
340 341
we measured rupture forces after initial holdfast adhesion, taken within seconds of
342
contact by gradually increasing incubation (dwell) times by liquid-cell dynamic force
343
spectrometry (DFS) 42. In these experiments, AFM tips were coated with a layer of
344
holdfast, as described in the experimental section (Figure 6A). The surface area of the
345
AFM tip covered with holdfast was analyzed by SEM (Figure 6B) and estimated at ~10-8
346
mm2.
347
The types of surfaces studied were different in terms of both hydrophobic
348
character and microscopic roughness: 1) mica (hydrophilic, atomically smooth,
349
homogeneous surface chemistry), 2) non treated clean glass (hydrophilic,
350
microscopically rough, heterogeneous surface chemistry), 3) 3-TMSM-treated
351
borosilicate glass (hydrophobic, microscopically rough, heterogeneous surface
352
chemistry) and 4) graphite (hydrophobic, atomically smooth, homogeneous surface
353
chemistry). Figures 6C and 6D represent typical force-displacement curves recorded by
354
DFS. Retraction curves exhibit a negative deflection dip, resulting from the adhesion
355
interaction as the tip is pulled back. The lowest point on the retraction curve corresponds
356
to the rupture force. Two kinds of curves were observed: curves with a single adhesion
357
event (Figure 6C) and curves with numerous local minima corresponding to multiple
358
partial rupture events (Figure 6D). In each case, the area enclosed between the negative
359
deflection curve and abscissa represents the work of adhesion. Retraction and extension
360
curves (red and blue respectively, Figure 6C-D) overlap completely between the trigger
361
and the contact point. Thus, holdfast behaved as an elastic medium for the force loading
362
rate (~ 1µm s-1) and magnitude range (0.1 - 1 nN) used in this study.
363
It is important to note that separation at rupture occurs at the contact interface
364
between holdfast and the substrate. Several lines of evidence support this idea: First, 15 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 16 of 43
365
direct SEM inspection of the AFM tip after DFS experiments show no visible loss of
366
holdfast material. Second, the cantilever resonance (in air, where quality factor is high)
367
did not change significantly before and after adhesion, indicating that the total mass
368
remained constant within the measurement error (~ 10-15 g) 43. Third, as shown in the
369
following section, the bond strength is much smaller initially than that after long contact
370
times (as in the case of tip/holdfast interface) making it much more likely that rupture will
371
occur at the substrate/holdfast interface. Finally, no significant loss of adhesion could be
372
detected after subsequent measurements using the same coated tip under similar
373
conditions.
374
The work of adhesion corresponding to the initial phases of interaction for the
375
four tested substrates is presented in Figure 7A. Clearly, the work of adhesion increased
376
with the hydrophobic character of the substrate. Graphite stands out with almost two
377
orders of magnitude greater work of adhesion than the other substrates. On graphite, the
378
time to onset of the rapidly increasing phase is shorter than the minimum measurable
379
time of 0.01 s. Long dwelling time adhesion to 3-TMSM treated glass is also significantly
380
stronger than adhesion to untreated glass. However, the time to onset of the rapidly
381
increasing phase is longer on 3-TMSM-treated glass than on graphite. Substrate
382
roughness does not seem to play a major role in initial adhesion. Mica, which is
383
hydrophilic and flat (0.2 nm rms), has a similar work of adhesion with glass, which is also
384
hydrophilic but microscopically rough (4.0 nm rms).
385
If we compare the work of adhesion, using the entire data set of force-
386
displacement curves (single and multi peak curves), with the maximal rupture force data
387
(Figure 7B), we observe the same trend for strength of adhesion as a function of
388
different surfaces. This trend indicates that adhesion strength increases with time on all
389
surfaces but the kinetics are different. Table 1 shows the maximum adhesion force per 16 ACS Paragon Plus Environment
Page 17 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
390
unit area on various surfaces. To find these estimates we have used the maximal force
391
determined at 90 seconds of dwell time by DFS (Figure 7B) and an average contact area
392
between the holdfast-covered AFM tip and the surface of 10-8 mm2 (Figure 6B). As for
393
the work of adhesion, the maximal adhesion force depended on the surface: the more
394
hydrophobic the substrate, the higher the adhesion force.
395
Contact area can be varied in principle by adjusting the maximal compression
396
force (the trigger point) acting on the tip/holdfast complex (Figure 7C). However, for all
397
surfaces and for trigger forces between 250 pN and 5 nN, the work of adhesion and
398
maximal force measurements remained constant within experimental error, which means
399
contact area was constant, the tip likely being in contact with the substrate. However, on
400
both hydrophobic surfaces, the work of adhesion increased with the trigger point force
401
above a threshold value of about 50 pN and then remained relatively unchanged. For
402
hydrophilic surfaces (clean glass and mica) the work of adhesion no such trigger force
403
threshold was observed. Note that the existence of a threshold force may prompt a
404
cooperative interaction between hypothetical adhesins since if the adhesins were
405
interacting with surface sites in a non-correlated manner we would expect in all cases a
406
gradual increase of the work of adhesion as a function of trigger force (due to contact
407
area expansion).
408
Qualitative examination of the force-extension curves revealed that roughly 60%
409
of them contained multiple rupture events. The existence of both single and multiple
410
rupture events highlights the underlying complexity of adhesion through multiple surface
411
bonds. A statistical analysis of the magnitudes of rupture forces and extensions in the
412
DFS force displacement curves was performed (Figure 8). Figure 8A shows the
413
distribution of rupture events by visual identification, while the histogram in Figure 8B
414
was derived from algorithms designed to extract these automatically (Supporting 17 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 18 of 43
415
Information). The distribution was fitted with a function consisting of six Gaussian peaks
416
each with a mean corresponding to a particular integer (n = 1-6) multiple of a
417
characteristic rupture force, and with identical widths, plus a constant “background", thus
418
3 fit parameters. The data are well-described by the fit function for a characteristic
419
rupture force parameter of 29.7 ± 0.6 pN. The first three peaks are present at high
420
significance while the others are present at roughly 1.5 - 2 standard deviation level.
421
Therefore, assuming that the high significance peaks in Figure 8A-B are associated with
422
one, two, and three adhesins, we deduce that the initial adhesion occurs through
423
discrete interactions, each carrying approximately 30 pN force. These values are within
424
the range of those found for some small proteins, like ankyrin 44 or dystrophin 45 for
425
example, or polymers, such as polystyrene 46 or polyethylene oxide 47. In contrast,
426
overall single bond rupture force measurements performed on various polysaccharides
427
are an order of magnitude higher than the value obtained here 48,49.
428
In addition, the distribution of extension values corresponding to the rupture
429
events shown above is illustrated in Figure 8C. The most probable extension between
430
rupture events is observed to be approximately 2 nm. This is well above the z-noise of
431
the AFM (~ 0.3 nm) within detection bandwidth. Based upon these data, we suggest that
432
main initial adhesion is likely to occur through adhesin/surface interactions, each contact
433
being capable of 2 nm extension before rupturing. It is worth noting here that in DFS
434
experiments, rupture occurs via thermally assisted escape across an activation barrier
435
that diminishes with applied force. Hence, measured forces are not a sole property of the
436
bound complex but also depend on the loading rate 50. Here, however, the loading rate
437
was held fixed, and we expect that the distribution of forces will vary somewhat for
438
different pulling velocities.
439 440
The smallest average number of rupture events per force-displacement curve is approximately 1 and occurs on atomically flat graphite (Figure 9). The largest average 18 ACS Paragon Plus Environment
Page 19 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
441
number of rupture events per force-displacement curve occurs for glass (both clean and
442
hydrophobic). One possible explanation is that cooperativity of adhesion postulated for
443
graphite in relation with the results of Figure 7C may be manifesting here as well. Thus,
444
in a cooperative bonding scenario, rupture of one adhesion-surface bond is quickly
445
followed by neighboring adhesins as in a zipper. On a morphologically heterogeneous
446
surface such as glass, other interactions (such as rapid spatial variations of interfacial
447
tension) may interrupt adhesin linkage and cooperativity. In this case, the result would
448
be a multiplication of rupture events per contact area.
449
We now return to the question of why in Figure 7 the non-polar character of 3-
450
TMSM glass manifests itself ~10 s after contact with the substrate while it is practically
451
instantaneous for graphite? A possible explanation that correlates well with the idea of
452
relatively sparse, mobile adhesins that organize in cooperative units is illustrated in
453
Figure 10. For a given initial distribution of nonpolar adhesins on the surface of the
454
holdfast proximal to the substrate, these adhesin molecules are able to bind immediately
455
to the homogeneous non-polar graphite. In contrast, a period of time may be required for
456
their rearrangement into domains, allowing effective binding to corresponding patches
457
on a heterogeneous glass surface. Enhancement of overlapping between glass (fixed)
458
and holdfast adhesin (mobile) non-polar patches would occur by diffusion and refolding
459
of the latter. Future experiments performed on surfaces with controlled heterogeneity
460
would be able to further substantiate this description. At this point, we provide a coupled
461
reaction-diffusion model of adhesin diffusion within the holdfast matrix and its reaction
462
with the substrate (see below), which reproduces well the observations.
463
The maximum adhesion force per unit area reported in this study (Table 1) was
464
three orders of magnitude lower than the force reported in previous work 25. We
465
hypothesize that the main reason for this difference is due to difference in the amount of
466
time the holdfast has adhered to the surface. In the Tsang et al. work, the holdfast was 19 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 20 of 43
467
in contact with the substrate for days before the pulling measurements were made 25. In
468
our DFS experiments, the holdfast spent approximately one hour on the tip after
469
application compared to an instrument-limited maximum measurement time of 90 s on
470
the surface. Indeed, examination of the force dependence on the dwell time (Figure 7)
471
clearly indicates that holdfast adhesion is strongly time-dependent.
472
A parsimonious reaction-diffusion model given by coupled diffusion of adhesin
473
within the holdfast matrix and its multi-step surface attachment kinetics (Supplemental
474
Information) can indeed describe the dependence of the adhesion force on dwell time in
475
the current DFS experiments as well as reconcile these results with those of Tsang et al.
476
25
477
of the rupture force is determined by surface-adsorbed adhesins, which have not yet
478
undergone the irreversible transition to the surface bound form. The time to onset of
479
weak adhesion is determined by the rates of diffusion of adhesin within the holdfast
480
mass and its reversible association with the substrate, in contrast with the time to onset
481
of the strong adhesion at later times determined by the relatively slower rate of
482
irreversible association with the surface.
(Figure 11A). Within the framework of this model, for short dwell times, the magnitude
483
Following previous works 51,52, we model the rupture geometry as shown in
484
Figure 11B, with parallel surface bonds, coupled through the holdfast to the AFM tip. The
485
dependence of the rupture force on the number of surface-associated adhesin species is
486
assumed to be linear in the scaling regime of loading rate relevant for the experiments
487
reported here 51,52. Within the general framework of this model, for short dwell times, the
488
magnitude of the rupture force is determined by the number of surface-adsorbed
489
adhesins that have not yet undergone an irreversible transition to the surface bound
490
form. The time to onset of weak adhesion is determined by the rates of diffusion of
491
adhesin within the holdfast mass and its reversible association with the substrate, in
20 ACS Paragon Plus Environment
Page 21 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
492
contrast with the longer time to onset of strong adhesion determined by the relatively
493
slower rate of irreversible association with the surface.
494
While this model represents a possible biophysical mechanism with plausible
495
parameter values leading to the separation of time scales for weak and strong adhesion,
496
we additionally consider two related, alternative mechanisms (Supporting Information). i)
497
First, we have quantitatively analyzed the slow diffusion limit of the current reaction-
498
diffusion model with a modified, single-step surface kinetic scheme, where the short and
499
long time scales for adhesion are given by the rates of surface adsorption and bulk
500
diffusion, respectively. A small rate of diffusion of the adhesin could result from rescaling
501
of the bare diffusion constant due to strong adhesin binding to the holdfast
502
polysaccharide matrix. ii) Second, we considered the possibility of cross-linking of the
503
holdfast matrix over time, either by the putative adhesin or side chains of the N-
504
acetylglucosamine polymer matrix itself. This is shown schematically in Figure 11C,
505
leading to stiffening of the holdfast and resulting in a more uniform distribution of an
506
externally applied load on surface bonds. Given the load dependence of the dissociation
507
constant for adhesin-surface binding 53,54, where the unbinding rate increases
508
exponentially with applied load, we hypothesize that when the holdfast is less stiff, an
509
applied load is more likely to be concentrated on a few bonds leading to a greater
510
probability of their rupture. Consequently, a larger load is distributed among the
511
remaining surface bonds, resulting in a cascade of multiple ruptures with shorter rupture
512
time and therefore smaller rupture force. In contrast, with a stiffer holdfast, the applied
513
load transferred to each surface bond and hence the load-dependent dissociation rate is
514
smaller. The rate of holdfast stiffening in the natural environment is expected to be slow
515
to provide sufficient time for the cell to detach from an inhospitable surface, consistent
516
with the onset of strong adhesion at longer times. While these models suggest different
517
mechanisms by which holdfast adhesion strength could evolve from its initial values to 21 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 22 of 43
518
considerably higher values over time, they all rely on multiple step kinetics, in absence of
519
which data could not be fit. Further experiments seeking to identify the underlying
520
mechanisms for the observed kinetics will be the scope of future studies.
521 522 523
CONCLUSION In conclusion, an expanded ensemble of biophysical characteristics of bonding
524
development in an isolated microscopic bioadhesive was investigated. We found that: (i)
525
holdfast binding affinity is modified by environmental conditions and the nature of the
526
substrate; (ii) holdfast adhesion varies on multiple time scales; and (iii) a kinetic model
527
can describe the observation of time-dependence of the adhesion force on short and
528
long time scales.
529
Together, our results suggest adhesion is initiated through discrete, cooperative
530
events, with a magnitude of force suggestive of single molecules. The number of these
531
initial surface interactions is enhanced on non-polar substrates. We propose that that the
532
initial adhesive properties of the C. crescentus holdfast are dominated by a yet to be
533
identified adhesin molecule acting in concert and present within a polysaccharide matrix
534
composed of N-acetylglucosamine multimers.
535
Biologically, being able to modulate the strength and the timing of the adhesion
536
process as a function of environmental cues is vital for bacteria. Indeed it has been
537
suggested that permanent adhesion of newborn cells is prevented when environmental
538
conditions deteriorate, allowing their dispersion and the formation of a new colony where
539
the growth conditions are more favorable 33. Uncovering the chemical nature of adhesins
540
as well as mechanisms underlying interactions of the adhesins within the holdfast matrix
541
in response to environment will be critical not only for understanding the remarkable
542
biology of adhesion, but also to modulate its properties for various applications. Indeed,
543
Caulobacter holdfast has all the desired properties for a valuable bioadhesive: it adheres 22 ACS Paragon Plus Environment
Page 23 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
544
to a wide variety of surfaces under aqueous conditions and its adhesion is time-
545
dependent, probably involving a natural curing mechanism, leading to a currently
546
unequalled adhesion strength of 68 N /mm2.
547 548
549
ACKNOWLEDGMENTS We thank members of the Brun laboratory for comments on the manuscript. We
550
thank Prof. T. Romeo (Department of Microbiology and Cell Science, University of
551
Florida) for providing the E. coli TRMG strain and C. Dufort who participated in the
552
preliminary stages of this study. This work was supported by National Institutes of Health
553
Grant GM102841 and by a grant from the Indiana METACyt Initiative of Indiana
554
University (funded in part through a major grant from the Lilly Endowment, Inc.) to YVB
555
and BD, and by a National Science Foundation Grant PHY-0645652 to SS. BRAN
556
acknowledges financial support from CNPq, Brazil.
557 558 559
SUPPORTING INFORMATION Theoretical Methods; Supplementary data about (i) Quantitative Analysis of the
560
Dependence of the Rupture Force on Dwell Time: Reaction-diffusion model of adhesin-
561
surface association and (ii) Analysis of Rupture Event Force Distributions;
562
Supplementary Tables S1 and S2; Supplementary Figures S1 to S7; Supplementary
563
References. This information is available free of charge via the Internet at
564
http://pubs.acs.org
565 566
REFERENCES
567 568
(1) Rehm, B. H. Bacterial Polymers: Biosynthesis, Modifications and Applications. Nat Rev Microbiol 2010. 8, 578-592.
23 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619
Page 24 of 43
(2) Itoh, Y.; Rice, J. D.; Goller, C.; Pannuri, A.; Taylor, J.; Meisner, J.; Beveridge, T. J.; Preston, J. F., 3rd; Romeo, T. Roles of Pgaabcd Genes in Synthesis, Modification, and Export of the Escherichia Coli Biofilm Adhesin Poly-Beta-1,6-N-Acetyl-DGlucosamine. J Bacteriol 2008. 190, 3670-3680. (3) Beloin, C.; Roux, A.; Ghigo, J. M. Escherichia Coli Biofilms. Curr Top Microbiol Immunol 2008. 322, 249-289. (4) Flemming, H. C.; Wingender, J. The Biofilm Matrix. Nat Rev Microbiol 2010. 8, 623-633. (5) Galy, O.; Latour-Lambert, P.; Zrelli, K.; Ghigo, J. M.; Beloin, C.; Henry, N. Mapping of Bacterial Biofilm Local Mechanics by Magnetic Microparticle Actuation. Biophys J 2012. 103, 1400-1408. (6) Shaw, T.; Winston, M.; Rupp, C. J.; Klapper, I.; Stoodley, P. Commonality of Elastic Relaxation Times in Biofilms. Phys Rev Lett 2004. 93, 098102. (7) Li, G.; Brown, P. J. B.; Tang, J. X.; Xu, J.; Quardokus, E. M.; Fuqua, C.; Brun, Y. V. Surface Contact Stimulates the Just-in-Time Deployment of Bacterial Adhesins. Mol Microbiol 2012. 83, 41-51. (8) O'Toole, G. A.; Kolter, R. Flagellar and Twitching Motility Are Necessary for Pseudomonas Aeruginosa Biofilm Development. Mol Microbiol 1998. 30, 295-304. (9) Pratt, L. A.; Kolter, R. Genetic Analysis of Escherichia Coli Biofilm Formation: Roles of Flagella, Motility, Chemotaxis and Type I Pili. Mol Microbiol 1998. 30, 285-293. (10) Watnick, P. I.; Kolter, R. Steps in the Development of a Vibrio Cholerae El Tor Biofilm. Mol Microbiol 1999. 34, 586-595. (11) Yildiz, F. H.; Visick, K. L. Vibrio Biofilms: So Much the Same yet So Different. Trends Microbiol 2009. 17, 109-118. (12) Vadillo-Rodriguez, V.; Busscher, H. J.; Norde, W.; de Vries, J.; van der Mei, H. C. Atomic Force Microscopic Corroboration of Bond Aging for Adhesion of Streptococcus Thermophilus to Solid Substrata. J Colloid Interface Sci 2004. 278, 251-254. (13) Boks, N. P.; Busscher, H. J.; van der Mei, H. C.; Norde, W. Bond-Strengthening in Staphylococcal Adhesion to Hydrophilic and Hydrophobic Surfaces Using Atomic Force Microscopy. Langmuir 2008. 24, 12990-12994. (14) Olsson, A. L.; van der Mei, H. C.; Busscher, H. J.; Sharma, P. K. Novel Analysis of Bacterium-Substratum Bond Maturation Measured Using a Quartz Crystal Microbalance. Langmuir 2010. 26, 11113-11117. (15) Francius, G.; Alsteens, D.; Dupres, V.; Lebeer, S.; De Keersmaecker, S.; Vanderleyden, J.; Gruber, H. J.; Dufrene, Y. F. Stretching Polysaccharides on Live Cells Using Single Molecule Force Spectroscopy. Nat Protoc 2009. 4, 939-946. (16) Marszalek, P. E.; Dufrene, Y. F. Stretching Single Polysaccharides and Proteins Using Atomic Force Microscopy. Chem Soc Rev 2012. 41, 3523-3534. (17) Brown, P. J.; Hardy, G. G.; Trimble, M. J.; Brun, Y. V. Complex Regulatory Pathways Coordinate Cell-Cycle Progression and Development in Caulobacter Crescentus. Adv Microb Physiol 2009. 54, 1-101. (18) Tomlinson, A. D.; Fuqua, C. Mechanisms and Regulation of Polar Surface Attachment in Agrobacterium Tumefaciens. Curr Opin Microbiol 2009. 12, 708-714. (19) Poindexter, J. S. The Role of Calcium in Stalk Development and in Phosphate Acquisition in Caulobacter Crescentus. Arch Microbiol 1984. 138, 140-152. (20) Gonin, M.; Quardokus, E. M.; O'Donnol, D.; Maddock, J.; Brun, Y. V. Regulation of Stalk Elongation by Phosphate in Caulobacter Crescentus. J Bacteriol 2000. 182, 337-347. (21) Wagner, J. K.; Setayeshgar, S.; Sharon, L. A.; Reilly, J. P.; Brun, Y. V. A Nutrient Uptake Role for Bacterial Cell Envelope Extensions. Proc Natl Acad Sci U S A 2006. 103, 11772-11777. 24 ACS Paragon Plus Environment
Page 25 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669
The Journal of Physical Chemistry
(22) Bodenmiller, D.; Toh, E.; Brun, Y. V. Development of Surface Adhesion in Caulobacter Crescentus. J Bacteriol 2004. 186, 1438-1447. (23) Entcheva-Dimitrov, P.; Spormann, A. M. Dynamics and Control of Biofilms of the Oligotrophic Bacterium Caulobacter Crescentus. J Bacteriol 2004. 186, 8254-8266. (24) Li, G.; Smith, C. S.; Brun, Y. V.; Tang, J. X. The Elastic Properties of the Caulobacter Crescentus Adhesive Holdfast Are Dependent on Oligomers of NAcetylglucosamine. J Bacteriol 2005. 187, 257-265. (25) Tsang, P. H.; Li, G.; Brun, Y. V.; Freund, L. B.; Tang, J. X. Adhesion of Single Bacterial Cells in the Micronewton Range. Proc Natl Acad Sci U S A 2006. 103, 57645768. (26) Alipour-Assiabi, E.; Li, G.; Powers, T. R.; Tang, J. X. Fluctuation Analysis of Caulobacter Crescentus Adhesion. Biophys J 2006. 90, 2206-2212. (27) Sato, H.; Kataoka, N.; Kajiya, F.; Katano, M.; Takigawa, T.; Masuda, T. Kinetic Study on the Elastic Change of Vascular Endothelial Cells on Collagen Matrices by Atomic Force Microscopy. Colloids Surf B Biointerfaces 2004. 34, 141-146. (28) Merker, R. I.; Smit, J. Characterization of the Adhesive Holdfast of Marine and Freshwater Caulobacters. Appl Environ Microbiol 1988. 54, 2078-2085. (29) Hardy, G. G.; Allen, R. C.; Toh, E.; Long, M.; Brown, P. J.; Cole-Tobian, J. L.; Brun, Y. V. A Localized Multimeric Anchor Attaches the Caulobacter Holdfast to the Cell Pole. Mol Microbiol 2010. 76, 409-427. (30) Wan, Z.; Brown, P. J.; Elliott, E. N.; Brun, Y. V. The Adhesive and Cohesive Properties of a Bacterial Polysaccharide Adhesin Are Modulated by a Deacetylase. Mol Microbiol 2013. (31) Johnson, R. C.; Ely, B. Isolation of Spontaneously Derived Mutants of Caulobacter Crescentus. Genetics 1977. 86, 25-32. (32) Poindexter, J. S.; Hagenzieker, J. G. Novel Peptidoglycans in Caulobacter and Asticcacaulis Spp. Journal of bacteriology 1982. 150, 332-347. (33) Berne, C.; Kysela, D. T.; Brun, Y. V. A Bacterial Extracellular DNA Inhibits Settling of Motile Progeny Cells within a Biofilm. Mol Microbiol 2010. (34) Abramoff, M. D.; Magelhaes, P. J.; Ram, S. J. Image Processing with ImageJ. In Biophotonics International, 1994; Vol. 11; pp 36-42. (35) Singla, A. K.; Chawla, M. Chitosan: Some Pharmaceutical and Biological Aspects - an Update. J Pharm Pharmacol 2001. 53, 1047-1067. (36) Sorlier, P.; Denuziere, A.; Viton, C.; Domard, A. Relation between the Degree of Acetylation and the Electrostatic Properties of Chitin and Chitosan. Biomacromolecules 2001. 2, 765-772. (37) Kline, K. A.; Fälker, S.; Dahlberg, S.; Normark, S.; Henriques-Normark, B. Bacterial Adhesins in Host-Microbe Interactions. Cell Host & Microbe 2009. 5, 580-592. (38) Silverman, H. G.; Roberto, F. F. Understanding Marine Mussel Adhesion. Mar Biotechnol (NY) 2007. 9, 661-681. (39) Boisvert, H.; Duncan, M. J. Translocation of Porphyromonas Gingivalis Gingipain Adhesin Peptide A44 to Host Mitochondria Prevents Apoptosis. Infect Immun 2010. 78, 3616-3624. (40) Chebotok, E. N.; Novikov, V. Y.; Konovalova, I. N. Depolymerization of Chitin and Chitosan in the Course of Base Deacetylation. Russian Journal of Applied Chemistry 2006. 79, 1162-1166. (41) Li, G.; Brun, Y. V.; Tang, J. X. Holdfast Spreading and Thickening During Caulobacter Crescentus Attachment to Surfaces. BMC Microbiol 2013. 13, 139. (42) Hinterdorfer, P.; Garcia-Parajo, M. F.; Dufrene, Y. F. Single-Molecule Imaging of Cell Surfaces Using near-Field Nanoscopy. Acc Chem Res 2012. 45, 327-336.
25 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701
Page 26 of 43
(43) Ilic, B.; Craighead, H. G.; Krylov, S.; Senaratne, W.; Ober, C.; Neuzil, P. Attogram Detection Using Nanoelectromechanical Oscillators. Journal of Applied Physics 2004. 95, 3694-3703. (44) Lee, G.; Abdi, K.; Jiang, Y.; Michaely, P.; Bennett, V.; Marszalek, P. E. Nanospring Behaviour of Ankyrin Repeats. Nature 2006. 440, 246-249. (45) Bhasin, N.; Law, R.; Liao, G.; Safer, D.; Ellmer, J.; Discher, B. M.; Sweeney, H. L.; Discher, D. E. Molecular Extensibility of Mini-Dystrophins and a Dystrophin Rod Construct. J Mol Biol 2005. 352, 795-806. (46) Li, I. T.; Walker, G. C. Single Polymer Studies of Hydrophobic Hydration. Acc Chem Res 2012. (47) Liu, K.; Song, Y.; Feng, W.; Liu, N.; Zhang, W.; Zhang, X. Extracting a Single Polyethylene Oxide Chain from a Single Crystal by a Combination of Atomic Force Microscopy Imaging and Single-Molecule Force Spectroscopy: Toward the Investigation of Molecular Interactions in Their Condensed States. J Am Chem Soc 2011. 133, 32263229. (48) Rief, M.; Oesterhelt, F.; Heymann, B.; Gaub, H. E. Single Molecule Force Spectroscopy on Polysaccharides by Atomic Force Microscopy. Science 1997. 275, 1295-1297. (49) Marszalek, P. E.; Oberhauser, A. F.; Pang, Y. P.; Fernandez, J. M. Polysaccharide Elasticity Governed by Chair-Boat Transitions of the Glucopyranose Ring. Nature 1998. 396, 661-664. (50) Dudko, O. K.; Filippov, A. E.; Klafter, J.; Urbakh, M. Beyond the Conventional Description of Dynamic Force Spectroscopy of Adhesion Bonds. Proc Natl Acad Sci U S A 2003. 100, 11378-11381. (51) Seifert, U. Rupture of Multiple Parallel Molecular Bonds under Dynamic Loading. Physical Review Letters 2000. 84, 2750-2753. (52) Seifert, U. Dynamic Strength of Adhesion Molecules: Role of Rebinding and SelfConsistent Rates. Europhys. Lett. 2002. 58, 792. (53) Bell, G. I. Models for the Specific Adhesion of Cells to Cells. Science 1978. 200, 618-627. (54) Evans, E.; Ritchie, K. Dynamic Strength of Molecular Adhesion Bonds. Biophys J 1997. 72, 1541-1555.
702 703 704
FIGURE LEGEND
705 706
Figure 1. (A) Caulobacter crescentus cell cycle. This dimorphic bacterium starts its life
707
as a motile swarmer cell with a single polar flagellum and pili at the same pole. The
708
swarmer cell is unable to initiate DNA replication and eventually differentiates into a
709
replication-competent stalked cell by shedding its flagellum, retracting pili, and
710
synthesizing the holdfast followed by a stalk at the same pole. The stalked cell initiates
26 ACS Paragon Plus Environment
Page 27 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
711
DNA replication, and divides to give rise to a new swarmer cell. (B) AFM picture of a C.
712
crescentus stalked cell. (C) AFM picture of shed holdfasts attached to a mica surface.
713 714
Figure 2. (A-C) Holdfast size distribution. (A) AFM picture of purified holdfasts attached on a
715
mica surface. Height (B) and diameter (C) distribution of holdfasts. on mica (white bars) and
716
graphite (black bars). n = 191 and 212 for mica and graphite samples respectively. (D-F)
717
Contact angle of holdfast on mica (D) and graphite (E). (F) Relative contact angle
718
distributions on mica (white bars) and graphite (black bars) as described in the experimental
719
section. n = 96 for mica and 72 for graphite.
720 721
Figure 3: B/W thresholded image from AF488-WGA labeled WT holdfasts (A), and PGA (B).
722
(C) Surface attachment to clean (white bars) and 3-TMSM treated (black bars) glass.
723
Results are expressed as percent of WT holdfasts attached to clean glass (for all holdfasts
724
experiments) or as percent of surface coverage of PGA on clean glass. The bars represent
725
the average of at least 10 fields of view for four indepedent replicates and the error bars
726
represent the standard error.
727 728
Figure 4: Surface attachment to clean (white symbols, dashed lines) and 3-TMSM-treated
729
(black symbols, solid lines) glass, using different concentrations of NaCl, as described in the
730
experimental section. (A) Purified holdfasts, (B) PGA and (C) ∆hfsH holdfasts.
731 732
Figure 5: Surface attachment to clean (white symbols, dashed lines) and 3-TMSM-treated
733
(black symbols, solid lines) glass, using different pH. (A) WT holdfasts, (B) PGA and (C)
734
∆hfsH holdfasts. (D) Surface attachment of purified holdfasts to clean (white bars) and 3-
27 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
735
TMSM-treated (black bars) glass, using sequential different pH, as described in the
736
experimental section.
Page 28 of 43
737 738
Figure 6: (A) Schematics for holdfast adhesion to the AFM tip as described in the main text.
739
(B) Scanning electron micrographs of a pristine tip and the tip after holdfast attachment (top
740
row). Close up of the pristine tip (bottom left) and the difference image between a tip before
741
and after holdfast loading (bottom right). The holdfast-covered area is in white (bottom right).
742
(C-D) Representative force-displacement curves collected from mica at 0.5 nN trigger force
743
and 2 s dwell time (red: extension. blue: retraction curve). Single-peak retraction curve (C),
744
(multi-peak retraction curve (D). The light gray area represents the work of adhesion.
745
T=trigger point, C = contact point, R = rupture point.
746 747
Figure 7: Experimental data for work of adhesion (A) and maximal force (B) for four
748
different surfaces as a function of dwell time (fixed trigger point = 500 pN). (C) Work of
749
adhesion as a function of the trigger point (fixed dwell time = 2 s). Data are expressed as an
750
average of 4-16 independent replicates and the error bars represent the standard error. Mica
751
is shown in red, graphite in black, clean glass in blue and 3-TMSM treated glass in green.
752 753
Figure 8: (A-B) Rupture event histograms analysis. The lower panel of each image
754
displays the histogram of rupture-event forces, corresponding to the visual (A) and
755
algorithmic (20 pN threshold) (B) identification of rupture events in the AFM force-
756
displacement data. The corresponding best-fit function comprising six Gaussians plus
757
uniform background is overlaid. The fits were performed over the range 25 to 250 pN.
758
The upper panel of each image displays the normalized residuals, computed as
28 ACS Paragon Plus Environment
Page 29 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
759
(data−fit)/fit. (C) Distribution of the extension between rupture events for the AFM force-
760
extension curves containing multiple rupture events. The multiple broad peaks are due
761
to the finite step size during tip retraction.
762 763
Figure 9: Frequency of number of single rupture events in force displacement curves
764
exhibiting multiple rupture events. n = 24 for all surfaces (mica in red, graphite in black,
765
clean glass in blue and 3-TMSM treated glass in green).
766 767
Figure 10: Proposed mechanism relying on the diffusion of sparse adhesion molecules
768
towards the contact interface and their binding to surface hydrophobic domains. On
769
graphite, the entire substrate is hydrophobic. In this case, reaching the interface is
770
sufficient to create a bond. On chemically heterogeneous substrates such as 3-TMSM-
771
treated glass, the additional step of interfacial diffusion of adhesins towards surface-
772
bound hydrophobic domains adds a lag time to adhesion.
773 774
Figure 11. (A) The rupture force on mica as a function of dwell time, as measured by
775
DFS (solid squares), and in the model (line). The parameter values of the fit are: h = 60
776
nm, k = 7.2 × 10−6 nm2 s−1, ka = 120 nm s−1, kd = 2.0 × 10−4 s−1, S0 = 1.0 nm−2, P0 = 4.6 ×
777
10−3 nm−3, = 5000 s , δf1 = 7 pN , δf2 = 134 pN. The mean force as measured by
778
Tsang et al. 25 is shown as a red triangle (reduced by a factor of 4, which corresponds to
779
/ 2. (B-C). Schematic diagram of holdfast surface attachment. (B) In modeling the
780
total adhesion force, attachment occurs via multiple parallel bonds, each assumed to be
781
a Hookean spring with spring constant, . The number of initial bonds with the surface,
782
, depends on the dwell time, and the load is assumed to be distributed (approximately) 29 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 30 of 43
783
uniformly among them. The holdfast is modeled as an elastic element with spring
784
constant , which is coupled to the AFM cantilever arm with spring constant , pulled
785
at constant speed . (C) Cross-linking of the holdfast matrix, achieved by the putative
786
adhesin or through chemical modification of N-acetylglucosamine, leads to stiffening of
787
the holdfast and increase in the rupture force.
788 789
30 ACS Paragon Plus Environment
Page 31 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
790
The Journal of Physical Chemistry
Table 1: Maximal adhesion forces per unit area on the surfaces
791 Surface
Maximal calculated adhesion force (N/mm2)
Mica
0.05
Clean glass
0.08
3-TSMS treated glass
0.13
graphite
0.66
792
31 ACS Paragon Plus Environment
The Journal of Physical Chemistry
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Figure 1 A Swarmer&cell& Pili&
Flagellum&
9 Mo
) tyle s e if Se le)l
) Predivisional&cell& yle t s fe e)li l i s s Stalked&cell&
Stalk&
Holdfast)
Reversible)a2achment) B
Irreversible)a2achment) C
ACS Paragon Plus Environment
Page 32 of 43
Page 33 of 43
Figure 2 A
B
C 30
80
Frequency (%)
60 Frequency (%)
40
20
10
20
Diameter (nm)
0 42
0 34
0 26
0 18
10
20
0
0 16
0
E
20
F
15 Frequency (%)
10
5
Contact angle (°)
ACS Paragon Plus Environment
80
70
60
50
40
30
20
0 10
D
14
0
12
80
10
60
40
0
Height (nm)
0
0
0
20
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
The Journal of Physical Chemistry
Figure 3 C 100
75
*
50
25
ld
fa
st
s
PG A Δh f
sH
ho
ol
df
as
ts
0
th
B
w
A
Holdfast attached (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 34 of 43
ACS Paragon Plus Environment
Page 35 of 43
Figure 4 B 100 Surface coverage (%)
100
75
50
25
0 0
C Holdfast attached (% of WT holdfasts)
A
Holdfast attached (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
5
10
15
20
NaCl concentration (mM)
25
75
50
25
0 0
5
10
15
20
25
NaCl concentration (mM)
ACS Paragon Plus Environment
100
75
50
25
0 0
5
10
15
20
NaCl concentration (mM)
25
The Journal of Physical Chemistry
Figure 5 A
B
250
Surface coverage (%)
75
50
25
0 0
200 150 100 50
2
4
6
8
10
0 0
12
2
4
6
8
10
12
pH
pH
C
D 100
Holdfast attached (%)
100
75
50
25
50
2nd incubation (on glass)
0
1st incubation (in suspension)
pH 7
ACS Paragon Plus Environment
pH 4
pH 10.5
pH 10
12
pH 7
10
pH 5
pH
8
pH 10
6
pH 7
4
pH 10
2
pH 7
25
pH 5
0 0
75
pH 5
Holdfast attached (%)
100
Holdfast attached (% of WT holdfasts)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 36 of 43
Page 37 of 43
The Journal of Physical Chemistry
Figure 6
1 2 A 3 AFM tip 4 5 6 7 Holdfast 8 9 10 11 Strong contact between the AFM tip and the Figure S2bound to the surface 12 holdfast 13 14 a 15 B 16 1500 nm! 17 18 19 20 21 22 23 125 nm! 24 25 26 27 28 29 30 C D 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Transfer of holdfast from the surface to the AFM tip
b
Force measurement using the holdfast-coated AFM tip on a new surface
1500 nm!
r 125 nm!
ACS Paragon Plus Environment
R
The Journal of Physical Chemistry
Figure 7 A
B 10-15
10-14
10-17
10-18
Mica Graphite Clean glass TMSM-treated glass
0.1
1 Dwelling time (s)
10
100
Work of adhesion (N.m)
10-16
10-19 0.01
C 104
Maximal force (pN)
Work of adhesion (N.m)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
Page 38 of 43
10-16
103
10-18
102 0.01
0.1
1
10
100
Dwell time (s)
ACS Paragon Plus Environment
10-20
101
102
103
Trigger point (pN)
104
Page 39 of 43
Figure 8
Residual
0
-2
40
40
0 0
50
100 150 200 Rupture force (pN)
250
80
0
-2
20
C
2
60 Count
B 2
Events
Residual
A
Events
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
20
0 0
40
20
50
100 150 200 Rupture force (pN)
250
ACS Paragon Plus Environment
0 0
2 4 6 8 10 Extension between rupture events (nm)
The Journal of Physical Chemistry
Figure 9
20
Mica Graphite Clean glass 3-TMSM-treated glass
15 Frequency (%)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
10
5
0 0
2
4
6
8
10
Number of rupture events
ACS Paragon Plus Environment
Page 40 of 43
Page 41 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
The Journal of Physical Chemistry
Figure 10
Homogenous graphite
Heterogenous patchy glass
ACS Paragon Plus Environment
The Journal of Physical Chemistry
Figure 11 A 104
Maximal force (nN)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60
102
100
10-2 10-2
100
102
104
106
Dwell time (s)
B
C
ACS Paragon Plus Environment
Page 42 of 43
Page 43 of 43
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48
The Journal of Physical Chemistry
Caulobacter Crescentus
ACS Paragon Plus Environment