Proteolytic Degradation of Bovine Submaxillary Mucin (BSM) and Its

Jul 7, 2015 - †Department of Mechanical Engineering and ‡Enzyme and Protein Chemistry, Department of Systems Biology, Technical University of Denm...
6 downloads 14 Views 1MB Size
Download PDF
Article pubs.acs.org/Langmuir

Proteolytic Degradation of Bovine Submaxillary Mucin (BSM) and Its Impact on Adsorption and Lubrication at a Hydrophobic Surface Jan Busk Madsen,† Birte Svensson,‡ Maher Abou Hachem,‡ and Seunghwan Lee*,† †

Department of Mechanical Engineering and ‡Enzyme and Protein Chemistry, Department of Systems Biology, Technical University of Denmark, DK-2800 Kgs. Lyngby, Denmark S Supporting Information *

ABSTRACT: The effects of proteolytic digestion on bovine submaxillary mucin (BSM) were investigated in terms of changes in size, secondary structure, surface adsorption, and lubricating properties. Two proteases with distinctly different cleavage specificities, namely trypsin and pepsin, were employed. SDS-PAGE analysis with staining for proteins and carbohydrate moieties showed that only the unglycosylated terminal regions of BSM were degraded by the proteases. Size exclusion chromatography (SEC) and dynamic light scattering (DLS) analyses indicated that tryptic digestion mainly led to the reduction in size, whereas pepsin digestion rather caused an increase in the size of BSM. Less complete cleavage in terminal peptide regions by pepsin and subsequent aggregation were possibly responsible for the increased size. Far-UV circular dichroism (CD) spectra of the protease-treated BSM showed a slight change in the secondary structure owing to the removal of terminal domains, but the overall random coil conformation adopted by the central glycosylated domain remained dominant and essentially unchanged. Surface adsorption properties as characterized by optical waveguide lightmode spectroscopy (OWLS) showed that tryptic digestion of BSM resulted in a decrease in the adsorbed mass, but pepsin digestion led to a slight increase in the adsorbed mass onto a hydrophobic surface compared to intact BSM. This is in agreement with the partial preservation of peptide segments in the terminal regions after pepsin digestion as confirmed by SEC and DLS studies. Despite a contrast in the adsorbed amount of the protease-treated BSMs onto the surface, both proteases substantially deteriorated the lubricating capabilities of BSM at a hydrophobic interface. The present study supports the notion that the terminal domains of BSM are critical to the adsorption and lubricating properties of BSM at hydrophobic interfaces.

■

INTRODUCTION Mucins are a family of large glycoproteins, either secreted or cell surface tethered, and key components in the mucus gels that coat the epithelial lining and act as a passive barrier to protect the underlying tissue from physical abrasion and bacterial infection.1,2 Secreted mucins are produced either in specialized goblet cells that are incorporated into the epithelial lining or in submucosal glands.3 Mucins are amphiphilic in that a major part consists of one or several central domains that are heavily modified with O-linked oligosaccharides whereas the Nand C-terminal domains are mostly unglycosylated and could be hydrophobic.7,8 Because of the presence of the negatively charged sialic acid moieties and sulfate groups, the central domains carry an overall negative charge under physiological conditions. The N- and C-terminal domains are usually described as being globular with an overall neutral charge due to a high content of uncharged amino acid residues as well as the presence of both negatively and positively charged amino acids. The negative charge of the glycans is responsible for the intramolecular repulsion, rendering mucins to assume extended “bottlebrush”-like conformation. The primary structure of the polypeptide backbone of the central domains is made up of a © 2015 American Chemical Society

variable number of tandem repeats (VNTRs) consisting of proline, serine, and threonine amino acid residues. Because of their composition and omnipresence in all mucins, they have been termed as mucin-like PTS-domains.9−11 Glycans are linked via glycosidic bonds between the glycan residue at the reducing ends and hydroxyl groups of serine or threonine in the VNTRs. Mucins show interesting surface and interfacial properties outside biological systems, too.12−16 When mucins adsorb onto a hydrophobic surface from an aqueous solution, the terminal domains of mucins are believed to interact with the surface while the charged central domains protrude away from the surface toward bulk solution.12 Thus, mucins act as macromolecular surfactants at a water/hydrophobic interface. A related outstanding effect from the amphiphilicity of mucins is their lubricity, in particular at hydrophobic interfaces in aqueous environment;13−16 formation of a mucin layer on hydrophobic surface can render the interface hydrophilic, Received: April 8, 2015 Revised: June 9, 2015 Published: July 7, 2015 8303

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Article

Langmuir

For further analyses stated below, protease-treated BSM (1 mg/ mL) was dialyzed against PBS buffer using a 1 mL Float-a-lyzer dialysis membrane with a 100 kDa cutoff. Size Exclusion Chromatography (SEC). BSM samples were dissolved to a final concentration of 3 mg/mL and clarified by filtration (5 μm hydrophilic poly(ether sulfone) sterile filter; Pall Corporation, Cornwall, UK) to remove aggregates. Six mg of sample was loaded on a HiPrep 26/60 Sephacryl S1000 Superfine gel filtration column (GE Healthcare Life Sciences, Uppsala, Sweden) installed on an Ä kta Avant chromatograph (GE Healthcare) and eluted with PBS buffer. Peptide bond absorbance at 214 nm was used to monitor the elution. Dynamic Light Scattering (DLS). The hydrodynamic diameters, Dh, of BSM molecules were measured using a Malvern Zetasizer Nano ZS two angle particle and molecular size analyzer (Malvern Instruments, Worcestershire, UK). The noninvasive back scatter (NIBS, measurement angle 173°) was used to achieve highest sensitivity. The temperature was set at 25 °C and the He−Ne laser light source to a wavelength of 633 nm. Disposable cuvettes (PMMA, Plastibrand) were employed for DLS measurements. Each sample was measured in triplicates. The Malvern Zetasizer software (Version 7.02) was used to analyze the obtained data, and the intensity distribution of Dh was chosen as the display mode. Circular Dichroism (CD) Spectroscopy. Circular dichroism (CD) spectra of protease-treated BSM (1 mg/mL) were acquired using a rectangular quartz cuvette with 0.5 mm path length (Hellma GmbH & Co. KG, Müllheim, Germany) and Chirascan spectrophotometer (Applied Photophysics Ltd., Surrey, UK). Spectra were recorded in the far-UV range from 260 to 195 nm with a step size of 1 nm, a bandwidth of 0.5 nm, and a time-per-point value of 1.5 s. One measurement was obtained from averaging three traces. All samples were measured thrice and then averaged, and the background from buffer was subtracted. Optical Waveguide Lightmode Spectroscopy (OWLS). OWLS is based on grating-assisted in-coupling of a He−Ne laser into a planar waveguide coating (200 nm thick Si0.25Ti0.75O2 waveguide layer on 1 mm thick AF 45 glass; Microvacuum Ltd., Budapest, Hungary). Adsorption of biomolecules from bulk liquid to the interfacing solid surface is measured by monitoring changes in the refractive index close to the solid−liquid interface. This method is highly sensitive out to a distance of ∼200 nm from the surface of the waveguide. Experiments were carried out using an OWLS 210 Label-free Biosensor system (Microvacuum Ltd., Budapest, Hungary). For consistency between surface adsorption measurements and the tribology measurements described below, the waveguides were coated with a layer of poly(dimethylsiloxane) (PDMS). The waveguides were first spin-coated at 2500 rpm for 15 s with an ultrathin layer (ca. 24.3 ± 3.1 nm, determined by scratch test using atomic force microscopy in tapping mode) of polystyrene (Sigma-Aldrich, Brøndby, Denmark), dissolved in HPLC grade toluene at 6 mg/mL. A subsequent ultrathin layer of PDMS (ca. 16.4 ± 0.17 nm) was applied.23 The base and curing agent of a commercial silicone elastomer (Sylgard 184 elastomer kit, Dow Corning, Midland, MI) were dissolved in hexane at a ratio of 10:3 (0.5% w/w final concentration), and the solution was spin-coated onto the waveguides at 2000 rpm for 25 s, followed by curing in an oven at 70 °C overnight. The PDMS-coated waveguide was exposed to PBS buffer to obtain a stable baseline using a programmable syringe pump (Model 1000NE, New Era Pump Systems, Inc., Farmingdale, NY) to pump buffer solutions through the flow-cell across the OWLS waveguide surface. 100 μL of dialyzed protease treated BSM (1 mg/mL) was then injected via a loading loop. Upon observing surface adsorption, the pump was stopped so that the BSM molecules could adsorb onto the surface under static conditions. After 10 min, the flow cell was rinsed with PBS buffer by restarting pumping. The adsorbed mass density data were calculated according to de Feijter’s equations.24 The experiment was repeated two or three times for each type of protease treated BSM. A refractive index increment (dn/dc) value of 0.150 cm3/ g for BSM was used for the calculation of the adsorption masses.25 Pin-on-Disk (PoD) Tribometry. Macroscale tribological properties of intact and protease-treated BSMs were characterized with a pin-

suppress strong interfacial adhesion, and effectively entrain aqueous solvent (base lubricant) into the interface. In this sense, the unglycosylated terminal domains are pivotal in mucins’ unique lubricating properties as they act as “anchoring feet” to the surface and primarily determine the adsorption and stabilization of mucin layers on the surface, which is essential for effective boundary lubrication. In the present study, we were interested in understanding how proteolytic modification of the terminal domains impacts the adsorption and lubricating properties of bovine submaxillary mucin (BSM) by employing proteases that are acting on the unglycosylated terminal regions, namely trypsin and pepsin. Changes in structure of mucins as well as in their bulk rheological properties after proteolysis by these two proteases were reported earlier.17−20 In order to understand possible changes in the lubricating properties of mucins after proteolysis, it is important to understand not only the structural variation due to proteolysis but also the impact on adsorption and conformation on tribopair surfaces. An array of biomolecular and surface analytical tools and pin-on-disk tribometry were employed for holistic comprehension of the changes in size, structure, conformation, surface adsorption, and lubricating properties of protease-treated BSM molecules.

■

MATERIALS AND METHODS

Proteins and Chemicals. BSM (M3895-1 G, Type I-S, Lot. Nr. 039K7003V), pepsin (gastric mucosa, P6887), trypsin (bovine pancreas, T1426), and all chemicals were purchased from SigmaAldrich (Brøndby, Denmark). Purification of BSM. As non-mucin proteins are known to be present in commercial BSM, a one-column anion exchange purification step was added.21 Briefly, BSM was dissolved overnight at 4 °C using a nutating mixer in 10 mM Na acetate, 1 mM EDTA, pH 5.0 buffer to a final concentration of 10 mg/mL. The solution was clarified by sterile filtration (5 μm hydrophilic poly(ether sulfone) sterile filter; Pall Corporation, Cornwall, UK) and subsequently fractionated according to charge on a high load 16/26 Q Sepharose high performance anion exchange column (GE Healthcare Life Sciences, Uppsala, Sweden) installed on an Ä kta Avant chromatograph (GE Healthcare, Uppsala, Sweden) equilibrated in buffer (10 mM Na acetate, 1 mM EDTA, 1.2 M NaCl, pH 5.0). BSM was purified as previously reported.21 Briefly, bound proteins were eluted by a multistep gradient in the elution buffer, analyzed by SDS-PAGE, and visualized by Coomassie Brilliant Blue (CBB) staining. BSM containing fractions were pooled, dialyzed against Milli-Q grade water at a ratio of 1:400 (vol:vol), and freezedried. All steps were performed at 4 °C. Proteolytic Digestion of BSM. For hydrolysis by trypsin, BSM was dissolved in 50 mM NH4HCO3 to a final concentration of 10 mg/ mL (100 μL). Disulfide linkages in BSM were reduced by addition of 5 μL of 200 mM DTT in 100 mM NH4HCO3, and the sample was left at room temperature for 1 h. Thereafter, the reduced BSM was alkylated by adding 4 μL of 1 M iodoacetamide in 100 mM NH4HCO3 and incubating 1 h at room temperature. The alkylation reaction was quenched by an additional 20 μL of DTT solution. Trypsin (1 mg/ mL) was added (40 μL) to a final ratio 1:25 (trypsin:BSM, wt/wt), incubated at 30 °C overnight, and added 831 μL of phosphate buffered saline (PBS) corresponding to a final BSM concentration of 1 mg/mL. For hydrolysis by pepsin, BSM was dissolved in 5% formic acid to a final concentration of 10 mg/mL and treated with DTT and iodoacetamide as above. Pepsin in 5% formic acid (1 mg/mL) was added (50 μL) to a final ratio of 1:20 (pepsin:BSM, wt/wt), incubated overnight at 30 °C, and added 821 μL of PBS corresponding to a final BSM concentration of 1 mg/mL. To verify cleavage, trypsin- and pepsin digested BSM (15 μL) were analyzed by SDS-PAGE using CBB staining to visualize BSM peptide fragments and periodic acid/Schiff (PAS) staining22 to specifically detect glycosylated fragments. 8304

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Article

Langmuir on-disk tribometer (CSM, Peseux, Switzerland). In this approach, a loaded pin is allowed to form a contact with a disk, and the sliding friction forces between them are measured at controlled rotation speeds of the disk. The load on the pin is applied with a dead weight (1 N) while sliding speed is controlled by rotating the disk with a motor. Friction forces generated during sliding contacts are monitored by a strain gauge on the arm holding the pin. The friction forces data obtained in this study were measured as a function of number of rotations (20) over a fixed track (radius: 5 mm) while varying the speed of rotation. The speed was varied within 0.25−100 mm/s. The speed was varied from high to low speeds and then to high speeds again. An average coefficient of friction, μ, defined as friction force/ load, for each speed could then be plotted. The tribological contacts started 10 min after the solution was transferred to the tribocup where a set of PDMS pin and disk is placed in order to provide the time for initial adsorption of BSM molecules. In order to avoid crosscontamination, one pair of PDMS pin and disk was used for only one set of experiments and discarded. The PDMS tribopairs used in this study were prepared with the Sylgard 184 elastomer kit described above (Dow Corning, Midland, MI). The base fluid and cross-linker were thoroughly mixed at a ratio of 10:1. Air trapped in the mixture was removed by applying a gentle vacuum. Disks were cast in a machined aluminum plate mold with flat wells with the dimensions of 30 mm in diameter and 5 mm in depth. The surface exposed to air in the course of curing was used for tribological measurements. The hemispherical pins (6 mm in diameter) were cast in a 96 microwell plate (NUNCLON Delta Surface, Roskilde, Denmark). The PDMS mixtures were subsequently cured at 70 °C overnight.

PAS staining in the present study. Both trypsin (“Try-BSM”) and pepsin (“Pep-BSM”) digestions of BSM displayed fragments in the lower molecular mass range on the CBB stained gel (“Try” and “Pep” lanes), demonstrating that proteolytic degradation of BMS had occurred. The fragments observed from Try-BSM migrate as distinct bands of ∼6, ∼4, and ∼3 kDa. This corresponds approximately to the fragment sizes expected from a theoretical cleavage the N-terminal segment of 1589 amino acids residues, where the largest fragment had a mass of 7152 Da when allowing for one missed cleavage.27 The theoretical digestions by both trypsin and pepsin can be found in the Supporting Information. For PepBSM, a weak band at ∼120 kDa is observed in the CBB stained gel, possibly stemming from a partial digestion of the terminal globular domains of BSM. This is consistent with the notion that pepsin preferentially cleaves at hydrophobic, in particular aromatic residues,26 which may result in less complete hydrolysis of the terminal regions of BSM. Additionally, ineffective disulfide bond reduction and alkylation at acidic pH used for pepsin proteolysis can contribute to less complete digestion of BSM terminal regions by pepsin. A distinct band at ∼40 kDa was identified to be pepsin on the CBB stained gel. Additionally, three distinct bands at ∼21, ∼15, and ∼5 kDa can be seen in Figure 1. The largest peptide in the theoretical digest of the sequence had a mass of 12 783 Da and is not exactly corresponding to a band seen on the gel. A possible explanation is that the large peptides observed originate either from several missed cleavages or from the C-terminal part of BSM. PAS staining showed a dense smear for BSM above 200 kDa (Figure 1) which, for both Try-BSM and Pep-BSM, migrated to a lower molecular weight position than intact BSM. The peptides in the low molecular range of the gel did not PAS-stain, indicating that they are not glycosylated and most likely originate from terminal regions of BSM. As expected, either the oligosaccharides shield protease cleavage sites present in the central mucin domains or there are simply no such sites in that part of the protein. Hydrodynamic Size of Protease Digested BSMs. Protease digestion reduces the molecular weight of BSM compared to the intact BSM. But the impact of proteolytic cleavage on the hydrodynamic size is not as straightforward as it is affected by possible intra- and intermolecular interactions before and after proteolysis. SEC and DLS were employed to monitor changes to the hydrodynamic size of BSM upon proteolytic treatments, and the results are presented in Figures 2 and 3, respectively. Figure 2 shows the chromatograms obtained from 6 mg of the different BSM samples. Intact BSM (Figure 2, black) was used as a reference to monitor changes in the digested samples. The intact BSM eluted as a single asymmetric broad peak between 150 and 250 mL and thus over a total volume of approximately 100 mL. Furthermore, DLS (Figure 3) showed a single peak distribution of intact BSM with an average hydrodynamic diameter of 50 nm. The chromatogram for Try-BSM (Figure 2, red) shows two distinct peaks at 275 and 350 mL indicating that trypsin digestion provides molecular entities of smaller hydrodynamic size than intact BSM. The DLS data, by contrast, show multiple peaks for Try-BSM where the largest has a Dh of 295 nm, the second largest is similar to that of intact BSM (50 nm), and the smallest has Dh of 10 nm (Figure 3). The species corresponding to the peak at 295 nm is substantially larger than intact BSM in Dh which cannot be explained by a conformational change of BSM, such as

■

RESULTS AND DISCUSSION Proteolytic Digestion of BSM. BSM was hydrolyzed by trypsin or pepsin to investigate whether any difference would be observed according to cleavage sites. Trypsin specifically hydrolyses peptide bonds N-terminal to either lysine or arginine residues.26 Pepsin, on the other hand, exhibits preferential cleavage for hydrophobic, in particular aromatic residues.26 Figure 1 shows the SDS-PAGE data obtained after protease treatments. CBB staining visualized the position of essentially unglycosylated fragments of the BSM, while PAS staining selectively reacts with carbohydrate moieties. BSM has previously been reported to be visible in the high molecular weight range above 200 kDa21 as reproduced using CBB and

Figure 1. SDS-PAGE gels stained with CBB (left) and PAS (right). Intact BSM is indicated by “B”, trypsin digested BSM by “Try”, and pepsin digested BSM by “Pep”. 8305

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Article

Langmuir

indicate that Pep-BSMs are strongly interacting to form aggregates. It should be noted that, in contrast to Try-BSM, the relative peak intensity of larger species for Pep-BSM is much more enhanced that its number fraction should also be more substantial. More importantly, cleavage sites in the terminal regions of BSM may lead to hydrolysis of the protein in a way that exposes long hydrophobic patches to the solvent. Furthermore, aromatic residues, comprising the preferred cleavage sites for pepsin, are not abundant in BSM; therefore, the possibility of retaining partial hydrophobic domains after the digestion is relatively higher. Hydrophobic regions tend to interact to shield themselves from the aqueous solvent, and thus intermolecular interactions between Pep-BSMs are likely to occur. Lastly, a small additional peak is present in Figure 2 with material eluting at 260 mL. This peak may correspond to nonaggregated Pep-BSM as it elutes after intact BSM. Additionally, the small peak with a Dh of ca. 15 nm in the DLS may correspond to peptides that were not removed by dialysis. Changes in the Secondary Structure of BSM Due to Proteolytic Degradation. BSM shows a far-UV CD spectrum that corresponds to a random coil structure as there are no detectable features of α-helices or β-sheets (Figure 4).28−31

Figure 2. SEC of BSM (black), Try-BSM (red), and Pep-BSM (blue). 6 mg of sample was loaded onto the column, and the elution volume was monitored by absorbance at 214 nm.

Figure 3. DLS measurements of BSM (black), Try-BSM (red), and Pep-BSM (blue) plotted as hydrodynamic size vs intensity. Figure 4. Influence of proteolytic degradation of BSM. Far-UV spectra of BSM (black), Try-BSM (red), and Pep-BSM (blue).

unfolding or extension after tryptic digestion. Instead, it may indicate an aggregation between the BSM fragments. This is inconsistent with the SEC data (Figure 2), where no species eluted before intact BSM. It should be noted though since the Dh distribution in Figure 3 is intensity-weighted, the larger particles tend to scatter light more strongly. In other words, the number fraction of species with 295 nm as maximum Dh is in fact extremely small, and as a consequence a corresponding peak may not be clearly manifested in the SEC chromatogram. The presence of a peak similar to BSM in Dh may indicate incomplete proteolytic degradation of the sample or insignificant size change by tryptic digestion, although the latter is more likely, given the nonspecific nature of trypsin digestion and the consequently insignificant molecular weight changes shown in SDS-PAGE gel data (Figure 1). Lastly, the occurrence of a very small peak, ca. 10 nm, may be originating from digestion of the terminal peptides. The chromatogram for Pep-BSM shows three main peaks that elute at 75, 118, and 155 mL, indicating that the hydrodynamic size of Pep-BSM increased due to proteolytic degradation. This correlates with the DLS data in Figure 3 where the Z-average Dh of the Pep-BSM is 531 nm. This may

However, the structural features are also reminiscent of a poly(Pro) II helical structure although the position of the maxima and minima are shifted slightly compared to a pure poly(Pro) II helix.31 The far-UV CD spectrum of BSM acquired in this study is similar to those reported previously for BSM with a local maximum at ca. 220 nm and a minimum at ca. 203 nm.15,21 Furthermore, as previously reported, no tertiary structural features were detected for BSM.21 Figure 4 shows the far-UV CD spectra recorded for intact BSM, Try-BSM, and Pep-BSM samples. The spectra obtained from Try-BSM and Pep-BSM show a positive maximum similar to that of intact BSM at ca. 220 nm (Figure 4), despite somewhat more enhanced intensity. The minima for both samples, on the other hand, have shifted slightly to ca. 200 nm, and the intensity of the spectra is slightly diminished compared to that of intact BSM. It is not surprising that the overall secondary structure of Try-BSM and Pep-BSM has not changed drastically as the vast majority of BSM is made up of the glycosylated central domains. As the oligosaccharide moieties shield the peptide 8306

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Article

Langmuir

a high propensity to form aggregates in bulk solution. How the structural/conformational changes of Pep-BSM in bulk solution, especially the formation of aggregates, would influence on surface adsorption is not straightforward to predict though. While increased molecular mass/size by aggregation can boost the adsorption via enhanced van der Waals interaction with the surface, it can also lead to less effective packing of macromolecules on the surface because of the bulkiness of aggregates. The slightly increased adsorbed mass of Pep-BSM compared to intact BSM (Figure 5) suggests that the access of newly formed hydrophobic patches in Pep-BSM aggregates to PDMS surface is still possible and that the flexible feature of BSM for optimal packing on the surface is sustained even for Pep-BSM and its aggregates. Figure 6 shows the representative μ vs speed plots obtained from sliding contacts between PDMS surfaces lubricated by aqueous solution of intact BSM and the protease digested BSMs.

bonds from the proteases, the majority of the protein would not have been cleaved. However, as the samples were dialyzed prior to measurement, the reduction in signal intensity may also arise from a reduction in protein concentration, e.g., small peptide fragments from the terminal domains. The slight shift to a shorter wavelength could originate from further disordering in the structure of BSM. The terminal domains of mucins are known to contain many structural motifs such a VWF-domains and CysD domains,7,8 and proteolytic degradation of these motifs and their subsequent removal could very well be the cause of the shift in the position of the local minimum. Surface Adsorption and Tribological Properties of BSM after Proteolytic Digestion. OWLS was employed to characterize the adsorption properties of the protease-digested BSMs onto a hydrophobic PDMS surface. The results are presented in Figure 5.

Figure 5. Surface adsorption masses in ng/cm2 as measured by OWLS for whole BSM and protease-treated BSMs. Figure 6. μ vs speed plots for the sliding contacts of PDMS−PDMS lubricated by PBS buffer (gray) and the aqueous solution of BSM (black), Try-BSM (red), and Pep-BSM (blue).

Adsorbed mass of intact BSM was measured as a reference and determined to be 43.1 ± 0.1 ng/cm2. In comparison, TryBSM resulted in a substantially lower adsorbed mass with only 12.8 ± 0.6 ng/cm2. Pep-BSM showed inverse behavior as the adsorbed mass, 53.3 ± 0.1 ng/cm2, was found to be slightly higher than that of intact BSM. As both proteases are expected to cleave the exposed peptide in the terminal domains of the protein, the altered surface adsorption properties of BSM are thought to be related to the frequency and position of the cleavage sites present in the sequence and subsequent interactions between the protease-treated BSMs. The theoretical digestions (in the Supporting Information) of the sequence of the N-terminal part of the protein show that more cleavage products are generated by trypsin than pepsin. With the higher frequency of cleavage sites for trypsin, there is a higher chance that hydrolysis occurs in close proximity to the central highly glycosylated domains in the primary structure. Thus, the remaining unglycosylated regions of the protein would be severely reduced in size, causing the overall surface adsorption to decrease as the anchoring sites of the protein are diminished. This clearly confirms that adsorption of BSM onto a hydrophobic surface such as PDMS is achieved via the interaction of the terminal domains with the hydrophobic substrate.12 Pep-BSM, on the other hand, tends to retain substantial parts of its terminal domains after proteolytic cleavage due to the lower frequency of unique proteolytic cleavage sites in BSM, which can explain its facile adsorption onto the PDMS surface (Figure 5). Because of the presence of parts of terminal domains, Pep-BSM may sustain ability to adsorb onto the PDMS surface to a certain extent. Furthermore, as shown in by both SEC and DLS data (Figures 2 and 3), Pep-BSM displayed

The experiments were duplicated for each BSM sample by employing separate PDMS−PDMS pairs, and the results were highly reproducible. As a reference, the μ vs speed plots in PBS buffer were also obtained (Figure 6), showing μ in the range of ca. 1−2 due to adhesive contacts between PDMS surfaces. Intact BSM shows a low μ value of ∼0.02, over the entire speed range, consistent with previous studies.15,16 The two proteasedigested BSM samples showed drastic changes in μ as a function of speed (Figure 6). The speed dependence for TryBSM and Pep-BSM was, however, independent of the order of measurements as both increasing and decreasing speed displayed the same trends. Try-BSM (Figure 6, red) shows fairly effective lubricating properties in the high-speed regime but loses the lubricity rapidly with decreasing speed. Deteriorated lubricating capabilities of Try-BSM compared to intact BSM are directly related to the low adsorbed mass as presented in Figure 5. The low coverage of the PDMS surfaces by Try-BSM may induce partial direct contacts between bare PDMS surfaces and cause higher interfacial friction forces. Low friction forces observed in the high-speed regime is indebted from more feasible entrainment of base lubricant, water, into the contacting zone rather than any changes in the adsorbed mass of Try-BSM in the high-speed regime.13 Similar results were obtained for Pep-BSM (Figure 6, blue). The lubricating effect of Pep-BSM is comparable to intact BSM only in the high-speed regime but is clearly inferior in the low-speed regime. As the adsorbed mass of Pep-BSM onto PDMS is even slightly higher than that of intact BSM (Figure 6), low surface 8307

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Langmuir

■

coverage, as with the case of Try-BSM, cannot account for its behavior. However, it should be noted that a high adsorbed mass is a necessary, but not a sufficient, condition for effective lubrication by surface adsorbates. The efficacy of boundary lubrication is ultimately determined by the binding strength and stability of lubricating films formed on the surface,32 and similar behavior with that of Pep-BSM, i.e., sufficiently high surface adsorption, yet insufficient lubricity mostly due to weak surface adsorption, is often observed.14,33,34 Compared to intact BSM, the adsorption of Pep-BSM and its aggregates onto PDMS surface may be achieved in less stable and optimal manner as their hydrophobic anchoring domains were altered from those of intact BSM. Nevertheless, the friction coefficients for PepBSM are slightly, yet distinctively lower than those of Try-BSM, and the onset of deteriorating lubricity starts at somewhat lower speed than Try-BSM. This observation suggests that the origin of the deteriorating lubricating properties of Try-BSM and PepBSM is somewhat different: Try-BSM due to poor adsorption (nearly complete removal of terminal regions) vs Pep-BSM due to weak adsorption (altered structure of hydrophobic patches). Overall, the adsorption and lubricating properties collectively suggest that the terminal domains of BSM are vital for efficient adsorption and aqueous lubrication of hydrophobic interfaces and that intact BSM has a superior structure/composition compared to the protease-treated BSMs for these properties.

AUTHOR INFORMATION

Corresponding Author

*E-mail [email protected] (S.L.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS European Research Council (ERC) is acknowledged for their financial support (Funding Scheme: ERC Starting Grant, 2010, Project Number 261152). The authors appreciate Dr. Petr Efler for his assistance in both experiments and discussion of the results.

■

REFERENCES

(1) Thornton, D. J.; Sheehan, J. K. From Mucins to Mucus. Proc. Am. Thorac. Soc. 2004, 1, 54−61. (2) Linden, S. K.; Sutton, P.; Karlsson, N. G.; Korolik, V.; McGuckin, M. A. Mucins in the Mucosal Barrier to Infection. Mucosal Immunol. 2008, 1, 183−197. (3) Wickström, C.; Davies, J. R.; Eriksen, G. V.; Veerman, E. C.; Carlstedt, I. MUC5B is a Major Gel-forming, Oligomeric Mucin from Human Salivary Gland, Respiratory Tract and Endocervix: Identification of Glycoforms and C-terminal Cleavage. Biochem. J. 1998, 334, 685−693. (4) Hovenberg, H. W.; Davies, J. R.; Carlstedt, I. Different Mucins are Produced by the Surface Epithelium and the Submucosa in Human Trachea: Identification of MUC5AC as a Major Mucin from the Goblet Cells. Biochem. J. 1996, 318, 319−324. (5) Sheehan, J. K.; Kirkham, S.; Howard, M.; Woodman, P.; Kutay, S.; Brazeau, C.; Buckley, J.; Thornton, D. J. Identification of Molecular Intermediates in the Assembly Pathway of the MUC5AC Mucin. J. Biol. Chem. 2004, 279, 15698−15705. (6) Kesimer, M.; Makhov, A. M.; Griffith, J. D.; Verdugo, P.; Sheehan, J. K. Unpacking a Gel-Forming Mucin: A View of MUC5B Organization after Granular Release. Am. J. Physiol. Lung Cell. Mol. Physiol. 2010, 298, L15−L22. (7) Thornton, D. J.; Rousseau, K.; McGuckin, M. A. Structure and Function of the Polymeric Mucins in Airways Mucus. Annu. Rev. Physiol. 2008, 70, 459−486. (8) Perez-Vilar, J.; Hill, R. L. The Structure and Assembly of Secreted Mucins. J. Biol. Chem. 1999, 274, 31751−31754. (9) Escande, F.; Aubert, J. P.; Porchet, N.; Buisine, M. P. Human Mucin Gene MUC5AC: Organization of Its 5′-region and Central Repetitive Region. Biochem. J. 2001, 358, 763−772. (10) Toribara, N. W.; Gum, J. R., Jr.; Culhane, P. J.; Lagace, R. E.; Hicks, J. W.; Petersen, G. M.; Kim, Y. S. MUC-2 Human Small Intestinal Mucin Gene Structure. Repeated Arrays and Polymorphism. J. Clin. Invest. 1991, 88, 1005−1013. (11) Desseyn, J.-L.; Tetaert, D.; Gouyer, V. Architecture of the Large Membrane-Bound Mucins. Gene 2008, 410, 215−222. (12) Shi, L.; Caldwell, K. D. Mucin Adsorption to Hydrophobic surfaces. J. Colloid Interface Sci. 2000, 224, 372−381. (13) Cassin, G.; Heinrich, E.; Spikes, H. A. The Influence of Surface Roughness on the Lubrication Properties of Adsorbing and Nonadsorbing Biopolymers. Tribol. Lett. 2001, 11, 95−102. (14) Lee, S.; Müller, 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, 8344−8353. (15) Madsen, J. B.; Pakkanen, K. I.; Lee, S. Thermostability of Bovine Submaxillary Mucin (BSM) in Bulk Solution and at a Sliding Interface. J. Colloid Interface Sci. 2014, 424, 113−119. (16) Nikogeorgos, N.; Madsen, J. B.; Lee, S. Lubricating Properties of Bovine Submaxillary Mucin (BSM) Films on a Hydrophobic Surface: Influence of Impurities and Contact Scale. Colloids Surf., B 2014, 122, 760−766.

■

CONCLUSIONS In the present study, we have investigated the effect of proteolytic degradation using trypsin and pepsin on the structure and conformation of BSM in bulk solution and its adsorption and lubricating properties at an elastomeric hydrophobic surface of PDMS. The SDS-PAGE by CBB and PAS staining showed that the proteolytic cleavage of BSM occurred in the terminal, un-glycosylated peptide regions. Subsequent analysis by SEC indicated that the size of the molecular species of Pep-BSM and Try-BSM increased and decreased, respectively. The abnormal increase in hydrodynamic size of Pep-BSM was ascribed to the formation of aggregates between Pep-BSMs due to an incomplete digestion and partial availability of hydrophobic patches. DLS has revealed that a small portion of Try-BSM also aggregated. The sustained random coil secondary structure of BSM as monitored by far-UV CD spectroscopy, despite the changes in hydrodynamic size after the protease treatments, further validates that the central tandem repeat domains of BSM were not degraded. Try-BSM and Pep-BSM adsorbed onto PDMS surface in a smaller and larger amount compared to intact BSM, respectively. This contrast was correlated with the availability of partial hydrophobic patches on the terminal regions for Pep-BSM and more pronounced aggregation. Nevertheless, the lubrication properties of both Try-BSM and Pep-BSM were found to deteriorate rapidly upon reaching slower speeds. This study provides a strong support for a notion that the hydrophobic terminal domains of BSM are pivotal for effective adsorption and aqueous lubrication on soft hydrophobic surfaces.

■

Article

ASSOCIATED CONTENT

* Supporting Information S

Theoretical digestion of the partial N-terminal sequence of BSM. The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.langmuir.5b01281. 8308

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309

Article

Langmuir (17) Hashimoto, Y.; Tsuiki, S.; Nisizawa, K.; Pigman, W. Action of Proteolytic Enzymes on Purified Bovine Submaxillary Mucin. Ann. N. Y. Acad. Sci. 1963, 106, 233−246. (18) Hill, H. D., Jr.; Schwyzer, M.; Steinman, H. M.; Hill, R. L. Ovine Submaxillary Mucin. Primary Structure and Peptide Substrates of UDP-N-acetylgalactosamine:mucin transferase. J. Biol. Chem. 1977, 252, 3799−3804. (19) Sarosiek, J.; Mizuta, K.; Slomiany, A.; Slomiany, B. L. Effect of Acetylsalicylic Acid on Gastric Mucin Viscosity, Permeability to Hydrogen Ion, and Susceptibility to Pepsin. Biochem. Pharmacol. 1986, 35, 4291−4295. (20) Lee, S. P.; Nicholls, J. F.; Roberton, A. M.; Park, H. Z. Effect of Pepsin on Partially Purified Pig Gastric Mucus and Purified Mucin. Biochem. Cell Biol. 1988, 66, 367−373. (21) Madsen, J. B.; Pakkanen, K. I.; Duelund, L.; Svensson, B.; Abou Hachem, M.; Lee, S. A Simplified Chromatographic Approach to Purify Commercially Available Bovine Submaxillary Mucins (BSM). Prep. Biochem. Biotechnol. 2015, 45, 84−99. (22) Hotchkiss, R. D. A Microchemical Reaction Resulting in the Staining of Polysaccharide Structures in Fixed Tissue Preparations. Arch. Biochem. 1948, 16, 131−141. (23) Røn, T.; Javakhishvili, I.; Jankova, K.; Hvilsted, S.; Lee, S. Adsorption and Aqueous Lubricating Properties of Charged and Neutral Amphiphilic Diblock Copolymers at a Compliant, Hydrophobic Interface. Langmuir 2013, 29, 7782−7792. (24) Ramsden, J. J. Review of New Experimental Techniques for Investigating Random Sequential Adsorption. J. Stat. Phys. 1993, 73, 853−877. (25) Bettelheim, F. A.; Dey, S. K. Molecular Parameters of Submaxillary Mucins. Arch. Biochem. Biophys. 1965, 109, 259−265. (26) Sweeney, P. J.; Walker, J. M. Proteolytic Enzymes for Peptide Production. Enzymes of Molecular Biology; Burrell, M. M., Ed.; Humana Press Inc.: Clifton, NJ, 1993; Chapter 15, pp 277−303. (27) Jiang, W.; Woitach, J. T.; Gupta, D.; Bhavanandan, V. P. Sequence of a Second Gene Encoding Bovine Submaxillary Mucin: Implication for Mucin Heterogeneity and Cloning. Biochem. Biophys. Res. Commun. 1998, 251, 550−556. (28) Yang, J. T.; Wu, C. S. C.; Böhm, G. Circular Dichroism (CD) Parameters and Secondary Structure Estimates of Proteins. Structural and Physical Data I; Hinz, H. J., Ed.; Springer: Berlin, 2003; Chapter 7, pp 7001−7018. (29) Pigman, W.; Moschera, J.; Weiss, M.; Tettamanti, G. The Occurrence of Repetitive Glycopeptide Sequences in Bovine Submaxillary Glycoprotein. Eur. J. Biochem. 1973, 32, 148−154. (30) Dekker, J.; Rossen, J. W.; Büller, H. A.; Einerhand, A. W. The MUC Family: An Obituary. Trends Biochem. Sci. 2002, 27, 126−131. (31) Kelly, S. M.; Jess, T. J.; Price, N. C. How to Study Proteins by Circular Dichroism. Biochim. Biophys. Acta, Proteins Proteomics 2005, 1751, 119−139. (32) Hardy, W. B.; Doubleday, I. Boundary Lubrication − The Paraffin Series. Proc. R. Soc. London, Ser. A 1922, 100, 550−574. (33) Lee, S.; Iten, R.; Müller, M.; Spencer, N. D. Influence of Molecular Architecture on the Adsorption of Poly(ethylene oxide)Poly(propylene oxide)-Poly(ethylene oxide) (PEO-PPO-PEO) on PDMS Surfaces and Implications for Aqueous Lubrication. Macromolecules 2004, 37, 8349−8356. (34) Lee, S.; Spencer, N. D. Adsorption properties of Poly(L-lysine)graft-Poly(ethylene glycol) (PLL-g-PEG) at a Hydrophobic Interface: Influence of Tribological Stress, pH, Salt Concentration, and Polymer Molecular Weight. Langmuir 2008, 24, 9479−9488.

8309

DOI: 10.1021/acs.langmuir.5b01281 Langmuir 2015, 31, 8303−8309