Mucin-Inspired Lubrication on Hydrophobic Surfaces - ACS Publications


Mucin-Inspired Lubrication on Hydrophobic Surfaces - ACS Publicationspubs.acs.org/doi/10.1021/acs.biomac.7b00605Jun 21,...

0 downloads 69 Views 2MB Size

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