Subscriber access provided by NEW YORK UNIV
Article
Mucin-inspired lubrication on hydrophobic surfaces Benjamin Tillmann Käsdorf, Florian Weber, Georgia Petrou, Vaibhav Srivastava, Thomas Crouzier, and Oliver Lieleg Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.7b00605 • Publication Date (Web): 21 Jun 2017 Downloaded from http://pubs.acs.org on June 27, 2017
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.
Biomacromolecules 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 32
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
Biomacromolecules
1
2
Mucin-inspired lubrication on hydrophobic surfaces
3 4 5
Benjamin T. Käsdorf1,, Florian Weber1, Georgia Petrou2, Vaibhav Srivastava2,
6
Thomas Crouzier2, and Oliver Lieleg1,*
7 8 9 10 11 12
1
Department of Mechanical Engineering and Munich School of Bioengineering, Technical University of Munich, Boltzmannstrasse 11, 85748, Garching, Germany 2
Division of Glycoscience School of Biotechnology
Royal Institute of Technology, 106 91 Stockholm, Sweden
13 14 15 16
17
18
ACS Paragon Plus Environment
1
Biomacromolecules
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 2 of 32
19
ABSTRACT
20
In the human body, high-molecular-weight glycoproteins called mucins play a key role in
21
protecting epithelial surfaces against pathogenic attack, controlling the passage of molecules
22
towards the tissue and enabling boundary lubrication with very low friction coefficients.
23
However,
24
biomacromolecules involved in these fundamental processes are fully understood. Thus,
25
identifying the key features that render biomacromolecules such as mucins outstanding boundary
26
lubricants could set the stage for creating versatile artificial superlubricants. We here
27
demonstrate the importance of the hydrophobic terminal peptide domains of porcine gastric
28
mucin (MUC5AC) and human salivary mucin (MUC5B) in the processes of adsorbing to and
29
lubricating a hydrophobic PDMS surface. Tryptic digestion of those mucins results in removal of
30
those terminal domains which is accompanied by a loss of lubricity as well as surface adsorption.
31
We show that this loss can in part be compensated by attaching hydrophobic phenyl groups to
32
the glycosylated central part of the mucin macromolecule. Furthermore, we demonstrate that the
33
simple biopolysaccharide dextran can be functionalized with hydrophobic groups which confers
34
efficient surface adsorption and good lubricity on PDMS to the polysaccharide.
neither
the
molecular
mechanisms
nor
the
chemical
motifs
of
those
35 36 37
Keywords: Biopolymers, Boundary lubrication, Hydrated layer, Hydration lubrication,
38
Amphiphiles
39 40
ACS Paragon Plus Environment
2
Page 3 of 32
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
Biomacromolecules
41
INTRODUCTION
42 43
In the human body, all wet epithelia are covered with mucus, a transparent viscoelastic
44
hydrogel which has two important functions: first, mucus shields the underlying tissue from
45
pathogenic attack 2, 3, and second, it provides mechanical protection for the tissues when they are
46
exposed to shear forces. The macromolecular key component of mucus is mucin, a complex
47
glycoprotein which can have molecular weights up to several MDa 4. Those densely glycosylated
48
proteins can be divided in three distinct groups: membrane-bound epithelial mucins, secreted
49
non-gel-forming mucins and secreted gel-forming mucins with the latter being the major
50
constituent of mucus
51
properties of native mucus, i.e. they reduce viral activity 6, limit biofilm formation 7, and (when
52
used as coatings) reduce cell
53
mucus systems such as saliva
54
artificial and biological surfaces 12-16: in the boundary lubrication regime, the friction coefficient
55
µ measured with mucin-based lubricants can be as small as µ = 0.01 or even less.
3, 5
1
. Reconstituted solutions of manually purified mucins reproduce key
8
and bacterial 9 adhesion to surfaces. Moreover, similar to native 10
and tear fluid 11, mucin solutions reduce friction both on
56
In this boundary regime, hydrodynamic effects are negligible and two opposing surfaces come
57
into direct contact 17. The very low friction coefficients observed with mucin-based lubricants are
58
therefore critically related to the ability of the highly glycosylated mucin molecules to strongly
59
adsorb to a broad range of surfaces
60
assemble into well-hydrated macromolecular layers 9, 22, 23, which prevent two opposing surfaces
61
from getting into direct contact with each other. This process enables mucin or mucin-like
62
molecules to lubricate numerous tissues in the human body: those tissues often comprise a
63
mixture of hydrophilic and hydrophobic parts (e.g. the corneal epithelium, the tongue or the
18-21
. Both on hydrophilic and hydrophobic surfaces, mucins
ACS Paragon Plus Environment
3
Biomacromolecules
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 4 of 32
64
surface of articular cartilage) with the latter being rendered hydrophilic by adsorption of e.g.
65
mucins 24-28.
66
In part, this separation of two counter surfaces during shear is achieved by the macromolecule
67
itself which is adsorbing to the surfaces and forms a hydrated layer. Friction is reduced by
68
shearing off this polymer layer (‘sacrificial layer mechanism’) 29, 30. The second mechanism that
69
is involved in reducing friction between opposing surfaces is based on surface-bound water
70
molecules (‘hydration lubrication’) trapped by the various hydrophilic moieties on the mucin
71
molecule
72
quickly reattach to a surface after it was detached by shear forces during a friction process. The
73
second mechanism involves the exchange of free water molecules from the lubricant solution
74
with water molecules in the hydration layer of surface-adsorbed mucin molecules. This exchange
75
takes place as a consequence of the shear occurring during the friction process and provides a
76
surface-bound water layer that reduces friction.
31, 32
(Figure 1a). To be efficient, the first mechanism requires the mucin molecule to
77 78
ACS Paragon Plus Environment
4
Page 5 of 32
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
Biomacromolecules
79 80 81 82 83 84 85 86 87 88
Figure 1: Schematic overview of the molecular mechanisms responsible for the lubricity of mucin based lubricants and mucin structure. a) Both the formation of a sacrificial surface layer and hydration lubrication contribute to the lubricity of mucin solutions (see main text for details). The amino acid sequence of human MUC5AC (b) and human MUC5B (c) is analyzed in terms of polar/nonpolar side chains and net charge (see methods for details). The red marks indicate the fragments of the mucin polypeptide which were detected by mass spectrometry after tryptic digestion (see Table S1 and S2 in the Supporting Information).
89
to decrease friction by up to two orders of magnitude compared to lubrication with simple buffer
90
12, 33
91
mucin physisorption to surfaces can be mediated by two different types of physical forces:
92
electrostatic interactions between charged moieties of the glycoprotein and oppositely charged
93
groups on the surface of the material on the one hand, and hydrophobic interactions between the
94
glycoprotein and the surface on the other hand. A mathematical analysis of the amino acid
95
sequence of human MUC5B (salivary mucin) and human MUC5AC (gastric mucin) reveals that
96
both types of interactions are indeed possible for either mucin variant studied here. Both mucins
Together, sacrificial layer formation and hydration lubrication enable mucin-based lubricants
. Mucins carry both charged and polar moieties as well as hydrophobic residues 4. Thus,
ACS Paragon Plus Environment
5
Biomacromolecules
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 6 of 32
97
exhibit a mixed distribution of charged amino acids throughout the whole sequence, and the
98
terminal domains of both MUC5AC and MUC5B feature an increased density of hydrophobic
99
amino acids compared to the central (glycosylated) region of the glycoprotein (Figure 1b and c).
100
In this central part of the molecule, the high amount of serine and threonine (both hydrophilic
101
amino acids to which glycans are typically attached to via O-glycosidic bonds) is responsible for
102
the comparably strong hydrophilic character of the peptide sequence. Together, this unfolded
103
central hydrophilic part and the globular hydrophobic termini result in a daisy-chain-like
104
configuration with amphiphilic character 34, 35. It was already shown that the high glycosylation
105
density on the mucin glycoprotein is crucial for mucin hydration, and that both the hydration
106
state of mucin
107
mucins . This particular molecular architecture suggests that the hydrophilic amino acids in the
108
central region of the polypeptide serve as anchor points for the glycosylation of the protein
109
promoting mucin hydration. In contrast, the hydrophobic terminal regions may be responsible for
110
mediating the attachment of the glycoprotein to hydrophobic surfaces.
22
and the detailed glycosylation pattern
36
are directly related to the lubricity of
111
We here show that the hydrophobic terminal peptide sequences of both human salivary mucin
112
and porcine gastric mucin are crucial for the lubricating abilities of these glycoproteins on
113
hydrophobic surfaces. Enzymatic removal of those peptide sequences not only eliminates mucin
114
lubricity but also reduces the adsorption efficiency of the glycoproteins to PDMS. Vice versa, we
115
demonstrate that the addition of hydrophobic groups to synthetic dextran molecules promotes the
116
dextran adsorption to PDMS and conveys lubricity to the dextran solution if a hydrophobic
117
surface is part of the tribological material pairing. Finally, we present a molecular repair
118
approach that partially restores the lubricating potential of enzymatically treated mucins on
119
hydrophobic surfaces by grafting artificial hydrophobic groups onto the damaged mucins.
ACS Paragon Plus Environment
6
Page 7 of 32
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
120
Biomacromolecules
MATERIALS & METHODS
121 122
Mucin purification
123
The purification process of mucins was described in detail previously 33. In short, mucus was
124
obtained by manual scraping pig stomachs after rinsing them gently with tap water. The mucus
125
was diluted 5-fold in 10 mM sodium phosphate buffer (pH 7.0) containing 170 mM NaCl and
126
stirred overnight at 4 °C. Cellular debris was removed via several centrifugation steps and a final
127
ultracentrifugation step (150000 x g for 1 h at 4 °C). Afterwards, the mucins were separated by
128
size exclusion chromatography using an ÄKTA purifier system (GE Healthcare) and a
129
XK50/100 column packed with Sepharose 6FF. The obtained mucin fractions were pooled,
130
dialyzed against ultrapure water and concentrated by cross-flow dialysis. The concentrate was
131
then lyophilized and stored at -80 °C. For purification of human salivary mucin (MUC5B),
132
unstimulated human whole saliva was collected from healthy, non-smoking, 20-30 year old
133
donors, which refrained from consuming food or beverages other than water for 1 h prior to
134
saliva donation. Saliva samples were stored on ice during collection and purified using the same
135
protocol as used for MUC5AC. Monomeric MUC5AC and MUC5B has a molecular weight of
136
~ 3 MDa including the glycan motifs attached to the protein backbone. As the formation of
137
oligomers is possible for both mucin types, the molecular weight range for oligomeric MUC5AC
138
and MUC5B is expected to be heterogeneous.
139 140
Dextrans
141
Dextrans with a MW of 150 kDa were obtained from TdB Consultancy (Uppsala, Sweden).
142
The polysaccharides were either unmodified or modified with carboxymethyl (CM),
ACS Paragon Plus Environment
7
Biomacromolecules
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 8 of 32
143
diethylaminoethyl (DEAE), or phenyl groups - the latter of which were present on the dextrans at
144
densities of either 0.15 phenyl groups/glucose or 0.40 phenyl groups/glucose, respectively. 40
145
kDa dextran with a phenylation degree of 0.32 to 0.40 phenyl groups/glucose was obtained from
146
Sigma Aldrich (St. Louis, MO).
147 148
Enzymatic digestion of mucin
149
Enzymatic treatment of mucins with trypsin was performed as described in Madsen et al. 37. In
150
brief, mucin was dissolved at 10 mg/mL in a 50 mM ammonium bicarbonate solution, and
151
disulfide bonds were reduced by adding 5 vol% 200 mM DTT (dissolved in 100 mM ammonium
152
bicarbonate) at room temperature for 1 h. 4 vol% of 1 M iodoacetamide dissolved in a 100 mM
153
ammonium bicarbonate solution was added to alkylate the mucin again at room temperature for
154
1 h. The reaction was quenched by adding 20 vol% of DTT. For proteolytic degradation,
155
40 vol% of 1 mg/mL trypsin (from bovine pancreas, Sigma Aldrich, St. Louis, MO) dissolved in
156
a 100 mM ammonium bicarbonate solution was added to the mucin solution and incubated at
157
30 °C for 18 h. Afterwards, the trypsinated mucin was purified and desalted via size exclusion
158
chromatography and cross-flow filtration as described above for native mucin.
159 160
Mucin sequence analysis
161
The mucin peptide sequences used for analysis in this work were obtained from the UniProt
162
protein database (MUC5AC, P98088; MUC5B, Q9HC84). Up to now, no peptide sequence for
163
porcine gastric mucin is available yet. Thus, for our analysis, we used the sequence of human
164
MUC5AC instead. Both the sequences of MUC5B and MUC5AC were first divided into sections
165
of 50 amino acids each. To characterize the hydrophobic character of these sections, the number
ACS Paragon Plus Environment
8
Page 9 of 32
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
Biomacromolecules
166
of amino acids with nonpolar side chains (i.e., Gly, Ala, Val, Met, Leu, Ile, Pro, Trp, Phe) and
167
polar side chains (i.e., Arg, Lys, Asn, Asp, Gln, Glu, His, Tyr, Ser, Thr, Cys) was counted. To
168
analyze the charge distribution along the amino acid backbone, the number of charged amino
169
acids at neutral pH (negatively charged amino acids: Asp, pKa = 3.9; Glu, pKa = 4.1; positively
170
charged amino acids: Arg, pKa = 12.5; Lys, pKa = 10.5) was counted. Each charged amino acid
171
was assigned one elementary charge, and the net charge of each section of 50 consecutive amino
172
acids is displayed as a bar in Figure 1. The glycosylation pattern was estimated based on the
173
distribution of the amino acids serine and threonine. They serve as an anchor for O-linked
174
glycans
175
Each displayed glycan molecule in Figure 1 represents 10 monosaccharides. Cleavage sites of
176
trypsin in the sequences of MUC5AC and MUC5B were analyzed with the tool PeptideCutter
177
from ExPASy Bioformatics Resources Portal.
38
with each O-linked oligosaccharide side chain comprising up to 20 sugar units
3, 15
.
178 179
Adsorption measurements
180
Adsorption measurements were performed with an eCell-T quartz crystal micro-balance
181
(3T-analytik, Tuttlingen, Germany) and a Gamry eQCM 10M data acquisition device
182
(Warminster, Pennsylvania, USA). The quartz crystals used for this study have a gold surface,
183
which was spin-coated with a thin layer of polydimethylsiloxane (PDMS, Sylgard 184,
184
DowCorning, Wiesbaden, Germany). Therefore, PDMS was mixed in a prepolymer/cross-linker
185
ratio of 10:1 and diluted to 1 vol% in n-hexane. 100 µL of this solution was applied to the center
186
of the quartz crystal distributed by operating the spin-coater at 3000 rpm for 60 seconds.
187
Afterwards, the PDMS was cured at 80 °C for 4 h. A profilometric analysis of the coated crystals
188
showed that the thickness of the PDMS layer was ~3 µm (see Supporting Information
ACS Paragon Plus Environment
9
Biomacromolecules
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 32
189
Figure S1). The concentration of biomolecules used for adsorption measurements was
190
100 µg/mL, and each biomolecule type was diluted in filtered 20 mM HEPES buffer (filter
191
threshold: 0.22 µm). For each measurement, a quartz crystal with a fresh PDMS coating was
192
used. Prior to each measurement, the setup was equilibrated with 20 mM HEPES buffer until a
193
stable frequency signal was reached. This procedure ensured that the coated PDMS layer (which
194
can absorb small amounts of water 39) has reached an equilibrated state so that water uptake into
195
this PDMS layer does not affect the measurement. At the beginning of each measurement,
196
2 minutes of HEPES buffer signal was recorded as a baseline. Afterwards, the biomolecules were
197
injected at 100 µL/min for 2 minutes, and the flow rate was set to 10 µL/min.
198 199
Tribology
200
Friction measurements were conducted on a commercial shear rheometer (MCR 302, Anton
201
Paar, Graz, Austria) equipped with a tribology unit (T-PTD 200, Anton Paar). The measurements
202
were performed as described previously
203
geometry. As opposing friction partners, PDMS cylinders (Ø 5.5 mm) and steel spheres
204
(Ø 12.7 mm, Kugel Pompel, Wien, Austria) were used. The PDMS cylinders were prepared by
205
mixing PDMS prepolymer and cross-linker in a 10:1 ratio (Sylgard 184, DowCorning), exposing
206
the mixture to 1 h vacuum and performing a final curing step at 80 °C for 4 h. Before a
207
measurement, three pins were inserted into the sample holder and cleaned with 80 % EtOH and
208
ultrapure H2O. The measurements were performed at room temperature and the PDMS cylinders
209
were fully covered with lubricant. To be consistent with our previous study on the lubricity of
210
mucin solutions
211
~8.1 mm² (Figure S2) and thus a contact pressure of ≈ 0.35 MPa), and the friction coefficient
22
40
. In brief, the setup used was a ball-on-cylinder
, a normal force of FN = 6 N was applied (leading to a contact area of
ACS Paragon Plus Environment
10
Page 11 of 32
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
Biomacromolecules
212
was recorded for sliding speeds from 1000 to 0.01 mm/s (logarithmic speed ramp, 10 measuring
213
points per decade) using a measuring time of 10 s per data point. The PDMS surface topology
214
was analyzed (Figure S3) after lubricating with unmodified dextrans (0.1 % in 20 mM HEPES)
215
to rule out any artefacts generated by wear while sliding a steel sphere over the PDMS surface.
216
All lubricants were used at a concentration of 1 mg/mL diluted in 20 mM HEPES, pH 7.4.
217 218
Hydration measurements
219
The hydration of the mucin and dextran coatings was measured by combining two
220
complementary techniques. First, the hydrated mass of mucin and dextran coatings was assessed
221
using quartz crystal microbalance with dissipation monitoring (QCM-D, E4 system, Q-Sense)
222
using gold coated crystals (QSX 301, Q-sense) that were cleaned prior to use with a mixture of
223
20 % hydrogen peroxide and 80 % ammonium heated at 80 °C for 10 minutes. The mucin and
224
dextran solutions were prepared at 1 mg/mL in 20 mM HEPES pH 7 buffer and injected into the
225
instrument at a rate of 200 µL/min. In QCM-D, the changes in dissipation reflect the viscoelastic
226
properties of the adsorbed coating. The frequency and dissipation shifts were fed into a
227
Voigt-based model which was used to accurately estimate the hydrated mass (Q-tools
228
software) 41. The density of the mucin coating was fixed at 1050 kg/m3 which is between that of
229
pure water (1000 kg/m3) and pure protein (1350 kg/m3) 42.
230
Second, the dry mass was measured using a surface plasmon resonance technique (SPR,
231
Biacore 2000, GE Healthcare). Mucins or dextrans were injected at a concentration of 1 mg/mL,
232
dissolved in 20 mM HEPES, pH 7. The adsorbed mass density was estimated assuming one
233
converting response units corresponded to 0.1 ng/mm2 43. The hydration level of the coatings was
234
deduced from the dry and hydrated mass using the relationship described in Equation 1.
ACS Paragon Plus Environment
11
Biomacromolecules
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 32
235 % =
ℎ − 100 ℎ
236 237
Repair of trypsinated mucin
238
Since chemical modification of the termini of trypsin treated mucins is very challenging,
239
phenylation of trypsin treated MUC5B was performed by carbodiimide coupling of
240
phenylethylamine (Sigma-Aldrich) to the carboxylic groups of sialic acid motifs along the
241
central part of the trypsin-treated mucin molecules. Trypsin treated MUC5B was dissolved in a
242
20 mM MES buffer solution (pH 5.5) together with 58 mM EDC (1-ethyl-3-(3-
243
dimethylaminopropyl)carbodiimide hydrochloride, ThermoFisher) and 27 mM NHS (N-
244
hydroxysulfosuccinimide,
245
phenylethylamine dissolved in 20 mM MES buffer solution (pH 5.5) were added to the mixture,
246
and the pH was adjusted to pH 5.5 by addition of HCl. The solution was left to react at room
247
temperature for 2 hours. The mucin was then desalted and purified by chromatography (PD-10
248
column, GE Healthcare) and lyophilized before further use. The addition of the phenyl
249
functionalities was verified by UV spectra measurement using 1 mg/mL solutions of the mucin
250
variants.
Simga-Aldrich).
After
complete
dissolution,
173 mM
ACS Paragon Plus Environment
12
Page 13 of 32
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
Biomacromolecules
251
RESULTS AND DISCUSSION
252
Previous studies revealed that commercially available mucins (e.g. porcine gastric mucin from
253
Sigma Aldrich) lack key properties observed for native mucin systems 6, 44. Prominent examples
254
are that solutions of industrial gastric mucins lack the ability to form oligomers, they do not
255
exhibit a sol-gel transition at acidic pH, and they are not able to reduce friction in the boundary
256
lubrication regime at neutral pH
257
gastric mucin following a previously described protocol 33, which preserves its native functional
258
properties. To investigate the contribution of the terminal hydrophobic peptide domains of
259
mucins to the lubrication potential of mucin solutions, we performed an enzymatic digestion of
260
both MUC5AC and MUC5B using trypsin. This enzyme usually cleaves a broad range of peptide
261
sequences. For this enzyme, an analysis of the sequences of human MUC5AC and MUC5B
262
predicts ~300 cutting sites for either mucin (see Methods), and those cutting sites are distributed
263
equally throughout the peptide sequence. However, both mucins are densely glycosylated, and
264
protein glycosylation is generally known to protect the polypeptide against proteolytic activity 38,
265
45, 46
266
peptidases 47, and accordingly, we expect only partial degradation of MUC5AC and MUC5B by
267
trypsin.
268
(Figure 2a and b) indicated that the treated mucins in fact retained a very large molecular
269
weight and that the cleaved groups were much smaller than the remaining glycoprotein.
13, 19, 33
. As a consequence, we here manually purify porcine
. Indeed, it has been demonstrated that porcine gastric mucin is partially resistant to
A
chromatographic separation
after enzymatic treatment
of
either mucin
ACS Paragon Plus Environment
13
Biomacromolecules
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 32
270 271 272 273 274 275 276 277 278
Figure 2: Lubricity of native and enzymatically treated mucins. Chromatograms of size exclusion chromatography of MUC5AC and MUC5B after tryptic digestion are shown in (a) and (b), respectively. Tribological measurements were performed with a steel/PDMS pairing using 0.1 mg/mL mucin dissolved in HEPES buffer. The measurements show the lubricating abilities of either native or tryptic digested MUC5AC (c) or MUC5B (d), respectively. HEPES buffer is included as a reference. The error bars denote the standard deviation as obtained from three independent measurements.
279
Mass spectrometry confirmed that the peptide fragments obtained after trypsin treatment of
280
human MUC5B indeed originate from the terminal region of the glycoprotein (Figure 1c and
281
Table S2). We interpret this result such that the central part of human MUC5B, which is densely
282
glycosylated, is shielded against proteolytic degradation. The unprotected terminal parts,
283
however, are accessible for the enzyme and seem to be broken down into small fragments of
284
similar size. For porcine MUC5AC, a detailed peptide sequence is not available yet in the
285
literature. However, the chromatographic profiles obtained for both trypsin treated mucins are
286
very similar and only exhibit one second peak at later fractions in addition to the mucin main
287
peak, which occurs at similar fractions as untreated mucin. Additionally the analysis of the
288
peptide fragments of trypsinated porcine MUC5AC via mass spectroscopy show a coverage with
289
the human MUC5AC sequence (Table S2). The overall coverage is lower than for MUC5B but
ACS Paragon Plus Environment
14
Page 15 of 32
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
Biomacromolecules
290
the pattern is similar, as fragments only of the terminal mucin domains can be identified. This
291
suggests that, also for the porcine MUC5AC used here, trypsin treatment did mostly generate
292
small peptide fragments which are likely to originate from the terminal, unglycosilated region of
293
the glycoprotein. After separating them from the proteolytic fragments, the remaining mucin
294
glycoproteins were tested for their lubricity. When the treated mucins were used as a 0.1 % (w/v)
295
lubricant in a steel/PDMS tribology setup, we observed an almost complete loss of their
296
lubricating abilities, especially in the boundary lubrication regime (Figure 2c and d).
297
To efficiently reduce friction by enabling hydration lubrication, macromolecular lubricants
298
such as mucins have to form well-hydrated surface coatings. Thus, the inability of trypsin-
299
digested mucins to lubricate in a hydrophobic PDMS/steel pairing could be due to a reduced
300
hydration of such glycoprotein coatings compared to untreated mucins. Since the strong
301
hydration of mucins is established by the high density of hydrophilic glycans in the central
302
region of the protein, the amount of mucin-bound water should not be strongly influenced by our
303
proteolytic degradation procedure. Indeed, hydration measurements of trypsin-digested
304
MUC5AC and MUC5B showed no significant reduction in the amount of mucin-bound water
305
(Figure 3, for detailed information see Table S3 in the Supporting Information).
306 307 308
ACS Paragon Plus Environment
15
Biomacromolecules
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 32
309 310 311 312 313 314 315 316
Figure 3: Hydration measurements of surface adsorbed layers of different macromolecules. The hydration is calculated for surface coatings of native MUC5AC and MUC5B, trypsin treated MUC5AC and MUC5B, unmodified dextrans and phenylated dextrans (phenylation degree 0.15 and 0.40, respectively), as well as trypsin treated MUC5B with attached phenyl groups. The error bars denote the standard deviation as obtained from three independent measurements.
317
An alternative explanation for the almost complete loss of mucin lubricity in the boundary
318
lubrication regime could be that the adsorption efficiency of enzymatically treated mucins to
319
surfaces is weakened compared to native mucins. It has been put forward that the terminal
320
(hydrophobic) regions of mucins are involved in the adsorption process of the glycoproteins to
321
hydrophobic surfaces whereas the glycosylated (hydrophilic) central region of mucins are
322
relevant for mucin adsorption to hydrophilic surfaces
323
hydrophobic moieties of the mucin glycoprotein might also be necessary for conveying lubricity
324
on apolar surfaces. Thus, in a next step, we analyzed the adsorption behavior of native and
325
trypsin treated mucins to a hydrophobic PDMS surface. Indeed, in contrast to native MUC5AC
326
and MUC5B, the trypsin treated mucins showed a strongly reduced adsorption to the
327
hydrophobic PDMS as indicated by the drastically reduced shift in resonance frequency reported
328
by QCM (Figure 4a and b).
18
. This model suggests that the
ACS Paragon Plus Environment
16
Page 17 of 32
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
Biomacromolecules
329 330 331 332 333 334
Figure 4: Adsorption kinetics of different mucin and dextran variants onto PDMS. The adsorption of native and trypsin treated mucins (MUC5AC (a) and MUC5B (b), respectively) and different dextrans variants (phenylated (c) and charged (d)) to PDMS-coated QCM sensors is shown. The error bars denote the standard deviation as obtained from three independent measurements.
335
This result motivates that, whereas mucin hydration is still high after trypsin treatment, the
336
efficiency of hydration lubrication will be drastically reduced as this mechanism requires surface
337
adsorbed mucin molecules to take effect. Also, the formation of a sacrificial layer, i.e. a dynamic
338
shear-off and readsorption cycle, will be hampered if mucin adsorption is reduced. Together,
339
these findings explain the observed loss in mucin lubricity very well and suggest that the
340
hydrophobic terminal domains of MUC5AC and MUC5B are crucial for promoting mucin
341
adsorption to PDMS as required for a good boundary lubricant on hydrophobic surfaces.
ACS Paragon Plus Environment
17
Biomacromolecules
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 32
342
To verify our hypothesis that hydrophobic groups are required for the adsorption and
343
subsequent lubricity of well-hydrated macromolecules on hydrophobic surfaces, we make use of
344
a bottom up approach. The idea is to test whether a macromolecule that is not an efficient
345
lubricant on hydrophobic PDMS can be turned into a good lubricant when its adsorption to
346
PDMS is improved. Of course, the trypsin treatment performed on the two mucin variants did
347
not only remove hydrophobic amino acids from the terminal region of the polypeptide but also
348
cleaved positively and negatively charged amino acids. As a molecular platform to
349
systematically probe the influence of charged and hydrophobic modifications on the adsorption
350
efficiency and lubricity of the molecule, we chose dextrans. Dextrans are strongly hydrated
351
molecules 48, and their good hydration is due to the high density of hydrophilic hydroxyl groups
352
along the polysaccharide 49. However, solutions containing unmodified dextrans hardly adsorb to
353
PDMS surfaces at all (Figure 4c). Consistently, the Stribeck curves obtained with solutions
354
containing such dextrans are very similar to those obtained with buffer lacking any
355
macromolecules (Figure 5a).
356 357 358 359
ACS Paragon Plus Environment
18
Page 19 of 32
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
Biomacromolecules
360 361 362 363 364 365 366 367
Figure 5: Rotational tribology with dextran solutions as lubricants. Tribological measurements between a steel/PDMS pairing with 0.1 mg/mL dextrans solution. Measured were the lubricating abilities of either hydrophobic (phenyl) dextrans (a) or charged dextrans (b). The dotted line represent the Stribeck curve for HEPES buffer as a reference, the dashed line the Stribeck curve of native MUC5AC. The error bars denote the standard deviation as obtained from three independent measurements.
368
In a next step, we test a dextran variant which was modified to carry hydrophobic moieties.
369
The idea is that the addition of hydrophobic groups to dextrans should improve the adsorption of
370
the molecules to hydrophobic PDMS and thus provide lubricity to the dextran solution. Indeed,
371
phenylated dextrans with a phenyl content of 0.15 phenyl substituents per glucose molecule
372
adsorb to a hydrophobic PDMS surface, although the recorded shift in crystal resonance
373
frequency is smaller than that observed for adsorption of either MUC5AC or MUC5B (Figure
374
4c). A possible explanation for the comparably lower adsorption efficiency of phenyl-dextran
375
might be that the amount of hydrophobic groups present on the dextran molecule is smaller than
376
the corresponding number of hydrophobic moieties on the mucin glycoprotein. This idea would
377
be consistent with the biochemical structure of the mucins which comprises large areas with
378
numerous hydrophobic amino acids. To test this hypothesis, we repeated the adsorption
379
measurements with dextrans carrying an increased density of phenyl groups (0.40 phenyl groups
380
per glucose molecule), i.e. a higher number of hydrophobic groups per dextran molecule. Indeed,
ACS Paragon Plus Environment
19
Biomacromolecules
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 32
381
for this dextran variant, we observe a stronger shift in the crystal resonance frequency
382
corresponding to more efficient adsorption than for the 0.15 phenyl-dextran variant. For the 0.40
383
phenyl-dextran, this frequency shift is now also similar in magnitude as that observed for the
384
adsorption of native mucins (Figure 4c). These findings suggest that the increased degree of
385
dextran phenylation leads to stronger interactions with the hydrophobic PDMS surface and
386
therefore should aid in maintaining a hydrated polymer film on the surface – provided that the
387
phenylation procedure did not interfere with dextran hydration. QCM measurements, however,
388
show that the high degree of hydration observed for native dextrans is maintained after
389
introducing phenyl groups to these polymers (Figure 3). As a consequence, we expect that both
390
phenylated dextrans should show improved lubricity on PDMS compared to unmodified
391
dextrans, but the dextran variant with the higher phenylation degree should exhibit better
392
lubricating abilities than the 0.15 phenyl-dextran.
393
Tribological measurements with the two phenylated dextran variants on hydrophobic PDMS
394
indeed agree with this expectation: both phenyl-dextran variants significantly reduce the friction
395
coefficient by up to two decades, especially in the mixed lubrication regime, i.e. for sliding
396
speeds between 1 mm/s and 100 mm/s (Figure 5a). In the boundary lubrication regime, i.e. at
397
low sliding speeds below 1 mm/s, the 0.40 phenyl-dextran is more efficient: we here measure
398
constantly low friction coefficients on the order of µ ~ 0.03 which is a bit higher than the value
399
obtained for mucins but more than an order of magnitude lower than the friction coefficient
400
measured for the 0.15 phenyl-dextran in this regime. This observation is consistent with our
401
notion that a stronger adsorption of hydrated molecules, i.e. a more stable hydrated polymer film
402
on the PDMS surface, will lead to better lubrication.
ACS Paragon Plus Environment
20
Page 21 of 32
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
Biomacromolecules
403
At this point, it is important to recall that the enzymatic treatment of the mucin glycoproteins
404
has not only removed hydrophobic groups from the macromolecule, but also positively and
405
negatively charge amino acids. However, we do not expect that those charged groups contribute
406
significantly to the adsorption of mucins to hydrophobic PDMS surfaces. To verify this, we next
407
test two dextran variants which were modified with anionic carboxymethyl (CM) and cationic
408
diethylaminoethyl (DEAE) groups, respectively. For those two charged dextran variants, we
409
observe similarly low adsorption to PDMS surfaces as for the unmodified dextrans (Figure 4d).
410
This underscores our notion that there are no strong binding interactions between the charged
411
CM or DEAE groups and hydrophobic PDMS, and it suggests that solutions containing those
412
macromolecules should be poor boundary lubricants. Indeed, the lubricity of those
413
macromolecular solutions on PDMS is low: the Stribeck curves measured in lubrication tests
414
performed with these charged dextran variants resemble the results obtained with either simple
415
buffer or unmodified dextrans (Figure 5b). Together, these findings are in agreement with our
416
hypothesis that it is the hydrophobic character of the terminal peptide sequences of mucins and
417
not the charged groups in this region of the glycoproteins that confers adsorption and lubricity on
418
hydrophobic surfaces such as PDMS to mucins.
419
The results presented so far highlight the importance of hydrophobic moieties for the
420
lubrication process on hydrophobic surfaces. Furthermore the extent of these hydrophobic
421
modifications on dextran molecules seems to be linked to the lubricating potential of those
422
polymers. However, a higher phenylation degree not only increases the number of phenyl groups
423
on the dextran molecule but also the density of those hydrophobic groups. Thus, we now ask,
424
whether the overall number of hydrophobic groups on dextrans is important for conveying
425
lubricity or if a certain density of hydrophobic motifs, i.e. spatial proximity of phenyl groups, is
ACS Paragon Plus Environment
21
Biomacromolecules
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 32
426
required for dextran molecules to act as an effective boundary lubricant. To tackle this question,
427
we probe the lubricating abilities of another dextran variant having a lower molecular weight (i.e.
428
40 kDa) than the dextrans tested so far (150 kDa) but a similar phenylation degree of ~0.40
429
phenyl groups/glucose. Interestingly, for the small dextran molecules, we obtain Stribeck curves
430
which are virtually identical to those obtained with the larger dextrans (Figure 5a). Since the
431
phenylation degree is 0.40 in both cases, this may indicate, that it is indeed the density of
432
hydrophobic groups on a hydrated macromolecule rather than the total number of hydrophobic
433
motifs which is relevant for providing low friction in the boundary lubrication regime.
434
Together, these results demonstrate that the presence of hydrophobic moieties is crucial for
435
biopolymers such as mucin or dextran to act as a boundary lubricant on hydrophobic surfaces.
436
Since the trypsin treatment of mucins entailed a nearly complete loss of mucin lubricity on
437
hydrophobic surfaces, we now, in a last step, try to “repair” the enzymatically treated mucin by
438
grafting artificial hydrophobic groups to the “damaged” mucins. For this approach, we choose to
439
covalently add phenyl groups to the carboxylic groups of trypsin treated mucin (Figure 6a).
440
ACS Paragon Plus Environment
22
Page 23 of 32
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
Biomacromolecules
441 442 443 444 445 446 447 448 449 450
Figure 6: Adsorption properties and lubricity of phenylated trypsin-treated MUC5B. A schematic representation of the “repair approach” (= phenylation) of trypsin treated MUC5B is depicted in (a). (b) Spectroscopic comparison of MUC5B-Trypsin and MUC5B-Trypsin with covalently attached phenyl groups. (c) Adsorption kinetics of native, trypsin treated and “repaired” MUC5B onto PDMS. (d) Lubricity of solutions of phenylated MUC5B as probed on a steel/PDMS pairing. The Stribeck curves obtained for native MUC5B (solid line) and trypsin treated MUC5B (dashed line) are shown for comparison. The error bars denote the standard deviation as obtained from three independent measurements.
451
Since the terminal peptide domains of the degraded mucin are no longer present and the central
452
part of the polypeptide sequence is shielded by glycans, we aim to graft those phenyl groups to
453
the carboxyl group of sialic acid, a charged monosaccharide, present among the mucin glycans 50,
454
51
455
which has a higher sialic acid content than MUC5AC
456
indicated a successful phenylation of “damaged” (= trypsin treated) MUC5B (Figure 6b), we
. For optimal efficiency of this grafting process, we perform these experiments with MUC5B 52-54
. Since UV-absorbance measurements
ACS Paragon Plus Environment
23
Biomacromolecules
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 24 of 32
457
next analyze if the addition of those hydrophobic groups would promote the adsorption of the
458
“repaired” mucin onto a hydrophobic PDMS surface. Indeed, we observe a significant increase
459
of adsorbed mucin molecules on a PDMS-coated quartz crystal (Figure 6c). However, the
460
amount of adsorbed molecules seems not to be as high as for native MUC5B. Thus, in a next
461
step, we ask if the “repaired” molecule can achieve similarly good lubricity as native mucin.
462
When performing friction tests with the “repaired” mucins (Figure 6d), we observe a strong
463
decrease of friction coefficient in the mixed lubrication regime (i.e. for sliding speeds between
464
1 mm/s and 100 mm/s) compared to trypsin treated MUC5B. This decrease in the friction
465
coefficient is, however, less pronounced in the boundary lubrication regime where we measure a
466
friction coefficient of ~0.2 - 0.3. This value is still ~3fold lower than the corresponding value
467
obtained for “damaged” mucin but more than an order of magnitude larger than the friction
468
coefficient obtained for native mucins in this regime. It is likely that the incomplete recovery of
469
both the adsorption kinetics as well as the lubricating abilities of this phenylated mucin
470
compared to native mucin can be attributed to the limited amount of phenyl groups grafted onto
471
the damaged mucin molecules. As discussed before (Figure 5a), the density of phenyl groups on
472
dextrans seems to be directly linked to both the adsorption kinetics of the macromolecule and its
473
lubricating ability. Therefore, the amount of sialic acid groups which can be targeted with our
474
phenylation procedure might be too low to achieve a phenylation density that is high enough to
475
achieve adsorption kinetics and lubricity on a level comparable to that of native mucin. This
476
assumption is in good agreement with previous work where it was shown that the adsorption
477
strength and lubricating potential of boundary lubricants are linked 55, 56. However, those findings
478
indicate that MUC5B digested with trypsin can indeed be ‘repaired’: covalently attaching phenyl
479
groups seems to compensate for the lost hydrophobic peptide termini, at least to a certain degree.
ACS Paragon Plus Environment
24
Page 25 of 32
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
Biomacromolecules
480
CONCLUSION
481 482
We here demonstrated, that the hydrophobic domains of mucins are crucial to adsorb and
483
further lubricate hydrophobic surfaces such as PDMS. Since several tissue surfaces in the human
484
body exhibit a hydrophobic character
485
means of hydrophobic interactions is essential to provide boundary lubrication in vivo. When the
486
mucin molecules are deprived of their hydrophobic domains, this lubrication on hydrophobic
487
surfaces cannot take place anymore. However, we were able to compensate for the loss of these
488
hydrophobic domains by grafting phenyl groups onto the damaged mucins, which in part
489
recovered their ability to adsorb and lubricate hydrophobic surfaces. This approach can also be
490
transferred to other biomolecules: Dextrans, highly hydrophilic molecules, can function as good
491
boundary lubricants on hydrophobic PDMS when equipped with phenyl groups. The density of
492
attached hydrophobic groups during this bottom-up approach determines the adsorption kinetics
493
and lubricity of this mucin-inspired macromolecule. Of course, we here study macromolecular
494
lubricity on simple PDMS surfaces which are hydrophobic but uncharged. On more complex
495
surfaces which combine hydrophobic and charged characteristics, a phenyl-modified
496
polyelectrolytic dextran variant might provide even better lubricity than the uncharged
497
phenylated dextran molecules presented here.
25-28
, the interaction of mucins with these surfaces by
498
The results presented here may pave the way towards the rational design of macromolecular
499
superlubricants which provide ultra-low friction on a broad range of biological and technical
500
surfaces. A suitable polymer for such an artificial lubricant could be polyethylene glycol (PEG)
501
which is non-toxic, water soluble, and available with different molecular weights. PEG is already
502
used in numerous fields ranging from industrial to pharmaceutical applications. PEG is highly
503
hydrated and a good lubricant when utilized as surface attached polymer brushes 57. Moreover, it
ACS Paragon Plus Environment
25
Biomacromolecules
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 26 of 32
504
has been demonstrated that linking PEG polymer chains to deglycosylated mucin can serve as a
505
replacement for the hydrated glycans and restore lubricity 22. Combining those existing strategies
506
with the results shown here could lead to an artificial brush-like macromolecule with great
507
potential for providing ultra-low friction on various surfaces.
508 509 510
ACS Paragon Plus Environment
26
Page 27 of 32
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
511
Biomacromolecules
AUTHOR INFORMATION
512
Corresponding author
513
*Prof. Dr. Oliver Lieleg, e-mail:
[email protected], phone: +49 89 289 10952, fax: + 49 89
514
289 10801
515
Present Address
516
Department of Mechanical Engineering and Munich School of Bioengineering
517
Boltzmannstraße 11, 85748 Garching, Germany
518 519
AUTHOR CONTRIBUTIONS
520
BK, OL and TC proposed the experiments. BK, FW, GP and VS performed the experiments
521
and analyzed the data. The manuscript was written by BK and OL.
522 523
ACKNOWLEDGEMENTS
524
The authors thank Christine Braig for assistance with the mucin preparation.
525 526
SUPPORTING INFORMATION
527
•
Analysis of PDMS-layer thickness on QCM-chips
528 529
•
Detailed information on the mass spectrometry analysis of trypsin digested MUC5AC and MUC5B.
530
•
Detailed information on the hydration measurements
531
•
Surface topology analysis of the PDMS pin surface
ACS Paragon Plus Environment
27
Biomacromolecules
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
532
Page 28 of 32
REFERENCES
533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574
1. Lai, S. K.; Wang, Y. Y.; Wirtz, D.; Hanes, J., Micro- and macrorheology of mucus. Adv. Drug Delivery Rev. 2009, 61, (2), 86-100. 2. McGuckin, M. A.; Linden, S. K.; Sutton, P.; Florin, T. H., Mucin dynamics and enteric pathogens. Nat. Rev. Microbiol. 2011, 9, (4), 265-78. 3. Linden, S. K.; Sutton, P.; Karlsson, N. G.; Korolik, V.; McGuckin, M. A., Mucins in the mucosal barrier to infection. Mucosal immunology 2008, 1, (3), 183-97. 4. Bansil, R.; Turner, B. S., Mucin structure, aggregation, physiological functions and biomedical applications. Curr. Opin. Colloid Interface Sci. 2006, 11, (2-3), 164-170. 5. Dekker, J.; Rossen, J. W.; Buller, H. A.; Einerhand, A. W., The MUC family: an obituary. Trends Biochem. Sci. 2002, 27, (3), 126-31. 6. Lieleg, O.; Lieleg, C.; Bloom, J.; Buck, C. B.; Ribbeck, K., Mucin biopolymers as broadspectrum antiviral agents. Biomacromolecules 2012, 13, (6), 1724-32. 7. Caldara, M.; Friedlander, R. S.; Kavanaugh, N. L.; Aizenberg, J.; Foster, K. R.; Ribbeck, K., Mucin biopolymers prevent bacterial aggregation by retaining cells in the free-swimming state. Curr. Biol. 2012, 22, (24), 2325-30. 8. Crouzier, T.; Jang, H.; Ahn, J.; Stocker, R.; Ribbeck, K., Cell Patterning with Mucin Biopolymers. Biomacromolecules 2013, 14, (9), 3010-3016. 9. Co, J. Y.; Crouzier, T.; Ribbeck, K., Probing the Role of Mucin-Bound Glycans in Bacterial Repulsion by Mucin Coatings. Adv. Mater. Interfaces 2015, 2, (17), 1500179(1-6). 10. Bongaerts, J. H. H.; Rossetti, D.; Stokes, J. R., The Lubricating Properties of Human Whole Saliva. Tribol. Lett. 2007, 27, (3), 277-287. 11. Ablamowicz, A. F.; Nichols, J. J., Ocular Surface Membrane-Associated Mucins. Ocul. Surf. 2016, 14, (3), 331-41. 12. Biegler, M.; Delius, J.; Käsdorf, B. T.; Hofmann, T.; Lieleg, O., Cationic astringents alter the tribological and rheological properties of human saliva and salivary mucin solutions. Biotribology 2016, 6, 12-20. 13. Lee, S., Characterization of Lubricity of Mucins at Polymeric Surfaces for Biomedical Applications. World Academy of Science, Engineering and Technology, International Journal of Medical, Health, Biomedical, Bioengineering and Pharmaceutical Engineering 2013, 7, (3), 145-150. 14. Mystkowska, J.; Karalus, W.; Sidorenko, J.; Dąbrowski, J. R.; Kalska-Szostko, B., Biotribological properties of dentures lubricated with artificial saliva. Journal of Friction and Wear 2016, 37, (6), 544-551. 15. Wang, X.; Du, M.; Han, H.; Song, Y.; Zheng, Q., Boundary lubrication by associative mucin. Langmuir 2015, 31, (16), 4733-40. 16. Yakubov, G. E.; McColl, J.; Bongaerts, J. H.; Ramsden, J. J., Viscous boundary lubrication of hydrophobic surfaces by mucin. Langmuir 2009, 25, (4), 2313-21. 17. Hamrock, B. J.; Schmid, S. R.; Jacobson, B. O., Fundamentals of fluid film lubrication. CRC press: 2004. 18. Coles, J. M.; Chang, D. P.; Zauscher, S., Molecular mechanisms of aqueous boundary lubrication by mucinous glycoproteins. Curr. Opin. Colloid Interface Sci. 2010, 15, (6), 406-416.
ACS Paragon Plus Environment
28
Page 29 of 32
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
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
Biomacromolecules
19. Lee, S.; Muller, M.; Rezwan, K.; Spencer, N. D., Porcine gastric mucin (PGM) at the water/poly(dimethylsiloxane) (PDMS) interface: influence of pH and ionic strength on its conformation, adsorption, and aqueous lubrication properties. Langmuir 2005, 21, (18), 8344-53. 20. Berry, M.; McMaster, T. J.; Corfield, A. P.; Miles, M. J., Exploring the Molecular Adhesion of Ocular Mucins. Biomacromolecules 2001, 2, (2), 498-503. 21. An, J.; Dedinaite, A.; Nilsson, A.; Holgersson, J.; Claesson, P. M., Comparison of a brush-with-anchor and a train-of-brushes mucin on poly(methyl methacrylate) surfaces: adsorption, surface forces, and friction. Biomacromolecules 2014, 15, (4), 1515-25. 22. Crouzier, T.; Boettcher, K.; Geonnotti, A. R.; Kavanaugh, N. L.; Hirsch, J. B.; Ribbeck, K.; Lieleg, O., Modulating Mucin Hydration and Lubrication by Deglycosylation and Polyethylene Glycol Binding. Adv. Mater. Interfaces 2015, 2, (18), 1500308-1500315. 23. Feiler, A. A.; Sahlholm, A.; Sandberg, T.; Caldwell, K. D., Adsorption and viscoelastic properties of fractionated mucin (BSM) and bovine serum albumin (BSA) studied with quartz crystal microbalance (QCM-D). J. Colloid Interface Sci. 2007, 315, (2), 475-481. 24. Shi, L.; Caldwell, K. D., Mucin Adsorption to Hydrophobic Surfaces. J. Colloid Interface Sci. 2000, 224, (2), 372-381. 25. Hills, B. A., Boundary lubrication in vivo. Proc. Inst. Mech. Eng., Part H 2000, 214, (1), 83-94. 26. Ranc, H.; Elkhyat, A.; Servais, C.; Mac-Mary, S.; Launay, B.; Humbert, P., Friction coefficient and wettability of oral mucosal tissue: Changes induced by a salivary layer. Colloids Surf., A 2006, 276, (1–3), 155-161. 27. Hills, B. A., Lubrication of visceral movement and gastric motility by peritoneal surfactant. J. Gastroenterol. Hepatol. 1996, 11, (9), 797-803. 28. Davidson, H. J.; Kuonen, V. J., The tear film and ocular mucins. Vet. Ophthalmol. 2004, 7, (2), 71-77. 29. Hsu, S. M., Boundary lubrication: current understanding. Tribol. Lett. 1997, 3, (1), 1-11. 30. Chan, S. M. T.; Neu, C. P.; DuRaine, G.; Komvopoulos, K.; Reddi, A. H., Tribological altruism: A sacrificial layer mechanism of synovial joint lubrication in articular cartilage. J. Biomech. 2012, 45, (14), 2426-2431. 31. Ma, L.; Gaisinskaya-Kipnis, A.; Kampf, N.; Klein, J., Origins of hydration lubrication. Nat. Commun. 2015, 6, 6060. 32. Jahn, S.; Klein, J., Hydration Lubrication: The Macromolecular Domain. Macromolecules 2015, 48, (15), 5059-5075. 33. Schömig, V. J.; Käsdorf, B. T.; Scholz, C.; Bidmon, K.; Lieleg, O.; Berensmeier, S., An optimized purification process for porcine gastric mucin with preservation of its native functional properties. RSC Adv. 2016, 6, (50), 44932-44943. 34. Yakubov, G. E.; Papagiannopoulos, A.; Rat, E.; Easton, R. L.; Waigh, T. A., Molecular Structure and Rheological Properties of Short-Side-Chain Heavily Glycosylated Porcine Stomach Mucin. Biomacromolecules 2007, 8, (11), 3467-3477. 35. Yakubov, G. E.; Papagiannopoulos, A.; Rat, E.; Waigh, T. A., Charge and interfacial behavior of short side-chain heavily glycosylated porcine stomach mucin. Biomacromolecules 2007, 8, (12), 3791-9. 36. An, J.; Jin, C.; Dedinaite, A.; Holgersson, J.; Karlsson, N. G.; Claesson, P. M., Influence of Glycosylation on Interfacial Properties of Recombinant Mucins: Adsorption, Surface Forces, and Friction. Langmuir 2017, 33, (18), 4386-4395.
ACS Paragon Plus Environment
29
Biomacromolecules
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
Page 30 of 32
37. Madsen, J. B.; Svensson, B.; Abou Hachem, M.; Lee, S., Proteolytic Degradation of Bovine Submaxillary Mucin (BSM) and Its Impact on Adsorption and Lubrication at a Hydrophobic Surface. Langmuir 2015, 31, (30), 8303-8309. 38. Jentoft, N., Why Are Proteins O-Glycosylated. Trends Biochem. Sci. 1990, 15, (8), 291294. 39. Verneuil, E.; Buguin, A.; Silberzan, P., Permeation-induced flows: Consequences for silicone-based microfluidics. Europhys. Lett. 2004, 68, (3), 412-418. 40. Boettcher, K.; Grumbein, S.; Winkler, U.; Nachtsheim, J.; Lieleg, O., Adapting a commercial shear rheometer for applications in cartilage research. Rev. Sci. Instrum. 2014, 85, (9), 093903. 41. Voinova, M. V.; Rodahl, M.; Jonson, M.; Kasemo, B., Viscoelastic Acoustic Response of Layered Polymer Films at Fluid-Solid Interfaces: Continuum Mechanics Approach. Phys. Scr. 1999, 59, (5), 391. 42. Weber, N.; Wendel, H. P.; Kohn, J., Formation of viscoelastic protein layers on polymeric surfaces relevant to platelet adhesion. J. Biomed. Mater. Res., Part A 2005, 72A, (4), 420-427. 43. Stenberg, E.; Persson, B.; Roos, H.; Urbaniczky, C., Quantitative determination of surface concentration of protein with surface plasmon resonance using radiolabeled proteins. J. Colloid Interface Sci. 1991, 143, (2), 513-526. 44. Kočevar-Nared, J.; Kristl, J.; Šmid-Korbar, J., Comparative rheological investigation of crude gastric mucin and natural gastric mucus. Biomaterials 1997, 18, (9), 677-681. 45. Jensen, P. H.; Kolarich, D.; Packer, N. H., Mucin-type O-glycosylation – putting the pieces together. FEBS J. 2010, 277, (1), 81-94. 46. Russell, D.; Oldham, N. J.; Davis, B. G., Site-selective chemical protein glycosylation protects from autolysis and proteolytic degradation. Carbohydr. Res. 2009, 344, (12), 1508-1514. 47. Scawen, M.; Allen, A., Action of Proteolytic-Enzymes on Glycoprotein from Pig Gastric Mucus. Biochem. J. 1977, 163, (2), 363-368. 48. Balasubramanian, D.; Raman, B.; Sundari, C. S., Polysaccharides as amphiphiles. J. Am. Chem. Soc. 1993, 115, (1), 74-77. 49. Hunger, J.; Bernecker, A.; Bakker, Huib J.; Bonn, M.; Richter, Ralf P., Hydration Dynamics of Hyaluronan and Dextran. Biophys. J. 2012, 103, (1), L10-L12. 50. Perez-Vilar, J.; Hill, R. L., The structure and assembly of secreted mucins. J Biol Chem 1999, 274, (45), 31751-31754. 51. Authimoolam, S.; Dziubla, T., Biopolymeric Mucin and Synthetic Polymer Analogs: Their Structure, Function and Role in Biomedical Applications. Polymers 2016, 8, (3), 71. 52. Bhaskar, K. R.; Gong, D.; Bansil, R.; Pajevic, S.; Hamilton, J. A.; Turner, B. S.; Lamont, J. T., Profound Increase in Viscosity and Aggregation of Pig Gastric Mucin at Low Ph. Am J Physiol 1991, 261, (5), G827-G833. 53. Snary, D.; Allen, A., Studies on gastric mucoproteins. The isolation and characterization of the mucoprotein of the water-soluble mucus from pig cardiac gastric mucosa. Biochem. J. 1971, 123, (5), 845-53. 54. Thomsson, K. A.; Prakobphol, A.; Leffler, H.; Reddy, M. S.; Levine, M. J.; Fisher, S. J.; Hansson, G. C., The salivary mucin MG1 (MUC5B) carries a repertoire of unique oligosaccharides that is large and diverse. Glycobiology 2002, 12, (1), 1-14.
ACS Paragon Plus Environment
30
Page 31 of 32
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
664 665 666 667 668 669 670
Biomacromolecules
55. Zappone, B.; Ruths, M.; Greene, G. W.; Jay, G. D.; Israelachvili, J. N., Adsorption, Lubrication, and Wear of Lubricin on Model Surfaces: Polymer Brush-Like Behavior of a Glycoprotein. Biophys. J. 2007, 92, (5), 1693-1708. 56. Jahanmir, S.; Beltzer, M., An Adsorption Model for Friction in Boundary Lubrication. ASLE Trans. 1986, 29, (3), 423-430. 57. Spencer, N. D., Aqueous Lubrication with Poly(Ethylene Glycol) Brushes. Tribology Online 2014, 9, (4), 143-153.
671 672
ACS Paragon Plus Environment
31
Biomacromolecules
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
673
Page 32 of 32
TABLE OF CONTENTS GRAPHIC
674 675
ACS Paragon Plus Environment
32