Investigating mucin interactions with diverse surfaces for biomedical applications

GEORGIA PETROU Licentiate Thesis, 2019 KTH Royal Institute of Technology CBH School of Engineering Sciences in Chemistry, Biotechnology and Health Department of Chemistry Division of Glycoscience Albanova University Center SE+100 44 Stockholm, Sweden

ISBN 978-91-7873-172-5 TRITA-CBH-FOU-2019:24 Cover art by Dalí Salvador. The Persistence of Memory. 1931. The Museum of Modern Art. New York. © 2019 Salvador Dalí, Gala-Salvador Dalí Foundation / Artists Rights Society (ARS), New York

To my family

“It is our choices, Harry, that show what we truly are, far more than our abilities.”

Albus Dumbledore, Harry Potter and the chamber of secrets

Abstract Mucous membranes are covered with mucus, a viscoelastic hydrogel that plays an essential role in their protection from shear and pathogens. The viscoelasticity of mucus is owing to mucins, a group of densely glycosylated proteins. Mucins can interact with a wide range of surfaces; thus, there is big interest in exploring and manipulating such interactions for biomedical applications. This thesis presents investigations of mucin interactions with hydrophobic surfaces in order to identify the key features of mucin lubricity, as well as describes the development of materials that are optimized to interact with mucins. In Paper I we investigated the domains which make mucins outstanding boundary lubricants. The results showed that the hydrophobic terminal domains of mucins play a crucial role in the adsorption and lubrication on hydrophobic surfaces. Specifically, protease digestion of porcine gastric mucins and salivary mucins resulted in the cleavage of these domains and the loss of lubricity and surface adsorption. However, a “rescue” strategy was successfully carried out by grafting hydrophobic phenyl groups to the digested mucins and enhancing their lubricity. This strategy also enhanced the lubricity of polymers which are otherwise bad lubricants. In Paper II we developed mucoadhesive materials based on genetically engineered partial spider silk proteins. The partial spider silk protein 4RepCT was successfully functionalized with six lysines (pLys-4RepCT), or the Human Galectin-3 Carbohydrate Recognition Domain (hGal3-4RepCT). These strategies were aiming to either non-specific electrostatic interactions between the positive lysines and the negative mucins, or specific binding between the hGal3 and the mucin glycans. Coatings, fibers, meshes and foams were prepared from the new silk proteins, and the adsorption of porcine gastric mucins and bovine submaxillary mucins was measured, demonstrating enhanced adsorption. The work presented demonstrates how mucin-material interactions can provide us with valuable information for the development of new biomaterials. Specifically, mucin-based and mucin-inspired lubricants could provide desired lubrication to a wide range of surfaces, while our new silk based materials could be valuable tools for the development of mucosal dressings. Key words mucus, mucin, protein adsorption, biomaterials, lubrication, mucoadhesion, 4RepCT

Sammanfattning Slemhinnor täckts av slem, en viskoelastisk hydrogel som spelar en viktig roll för att skydda mot mekanisk nötning och patogener. Muciner, en grupp av tätt glykosylerade proteiner, spelar en viktig roll i viskoelasticiteten av slem. Eftersom muciner kan interagera med diverse ytor är det av stort intresse att utforska och manipulera sådana interaktioner för biomedicinska tillämpningar. Denna avhandling presenterar undersökningar av mucininteraktioner med hydrofoba ytor för att identifiera de viktigaste egenskaperna hos mucinsmörjning, samt beskriver utveckling av material som optimerades för att interagera med muciner. I Artikel I undersökte vi de domäner som bidrar till mucinernas enastående kapacitet som smörjmedel. Resultaten visade att mucinernas hydrofoba terminaldomäner spelar en avgörande roll vid adsorption och smörjning på hydrofoba ytor. Mer specifikt, proteasklyvning av svinmagemuciner och salivmuciner resulterade i klyvningen av dessa domäner och förlust av smörjning och ytadsorption. Genom att länka hydrofobiska fenylgrupper till de uppbrutna mucinerna, lyckades deras smörjningsegenskaper förbättras. Denna strategi förbättrade också smörjningsegenskaper hos andra polymerer som annars har dåliga smörjningsegenskaper. I Artikel II utvecklade vi mukoadhesiva material baserade på genetiskt modifierade partiella spindelsilkeproteiner. Spindelsilkeproteinet 4RepCT funktionaliserades framgångsrikt med tillsats av sex lysiner (pLys-4RepCT), eller den mänskliga Galectin-3 karbohydrat igenkänningsdomänen (hGal3-4RepCT). Syftet med dessa strategier var antingen att öka ospecifika elektrostatiska interaktioner mellan de positiva lysinerna och de negativa mucinerna, eller den specifika bindningen mellan hGal3 och mucin-glykanerna. Beläggningar, fibrer, nät och skum framställdes från de nya silkeproteinerna. Efter att adsorption av svinmagsmuciner och bovina submaxillära muciner uppmätts, visade de nya silkeproteinerna förbättrad mucin adsorption. Detta arbete visar hur interaktioner mellan mucin-material kan bidra med värdefull information för utvecklingen av nya biomaterial. Mucinbaserade och mucininspirerade smörjmedel kan ge önskad smörjning till ett brett spektrum av ytor, medan vår nya silkesbaserad material kan vara ett värdefullt verktyg för utvecklingen av slemhinneförband. Nyckelord slem, mucin, proteinadsorption, biomaterial, smörjning, mukoadhesion, 4RepCT

List of publications Paper I Käsdorf, B. T., Weber, F., Petrou, G., Srivastava, V., Crouzier, T., & Lieleg, O. (2017). Mucin-inspired lubrication on hydrophobic surfaces. Biomacromolecules, 18(8), 2454-2462. Paper II Petrou, G., Jansson, R., Högqvist, M., Erlandsson, J., Wågberg, L., Hedhammar, M., & Crouzier, T. (2018). Genetically engineered mucoadhesive spider silk. Biomacromolecules, 19(8), 3268-3279.

Contribution to the included publications Paper I Performed the surface plasmon resonance experiments and analyzed the data. Paper II Performed part of the quartz crystal microbalance with dissipation experiments and the bio-layer interferometry experiments. Produced the materials and did the fluorescent labeling of mucins, as well as the fluorescent mucin adsorption experiments together with microscopy. Conducted the related data analysis and contributed to writing the paper to a great extent.

Related work not included in the thesis Petrou, G., & Crouzier, T. (2018). Mucins as multifunctional building blocks of biomaterials. Biomaterials science, 6(9), 2282-2297.

Abbreviations

4RepCT partial spider silk protein

BSA bovine serum albumin

BSM COPD CRD

bovine submaxillary mucins chronic obstructive pulmonary disease carbohydrate recognition domain

GalNAc N-Acetylgalactosamine

hGal3-4RepCT

4RepCT functionalized with Human Galectin-3 Carbohydrate Recognition Domain

MUC5AC mucin 5AC

MUC5B mucin 5B

PDMS polydimethylsiloxane

PEG polyethylene glycol

PGM porcine gastric mucins

pLys-4RepCT 4RepCT functionalized with six lysines

PMMA polymethyl methacrylate

QCM quartz crystal microbalance

QCM-D quartz crystal microbalance with dissipation monitoring

SPR surface plasmon resonance

STP repeats serine, threonine and proline rich repeats

1

Contents 1. Introduction 1

1.1. Aim of the thesis 1

1.2. Mucous membranes 1

1.3. Mucus 2

1.4. Mucins 4

1.5. Mucin adsorption 6

1.5.1. Utilizing the barrier properties of mucins in the form of coatings 8

1.5.2. Utilizing the lubricity of mucins 9

1.6. Mucoadhesive materials 10

1.7. Mucosal diseases 14

1.7.1. Compromised barrier 14

1.7.2. Loss of hydration and lubrication 15

1.7.3. Altered Rheology 16

2. Present investigations 18

2.1. Aim of present investigations 18

2.2. Materials and methods 18

2.2.1. Mucin purification 18

2.2.2. Quartz crystal microbalance with dissipation monitoring (QCM-D) and surface plasmon resonance (SPR); Monitoring adsorption and measuring hydration 19

2.3. Results and Discussion 22

2.3.1. Mucin-inspired lubrication on hydrophobic surfaces (Paper I) 22

2.3.2. Genetically engineered mucoadhesive spider silk (Paper II) 25

2.4. Concluding remarks and future perspectives 30

3. Acknowledgements 32

4. References 34

1

1. Introduction

1.1. Aim of the thesis The mucus gel that covers our mucosa is essential for its protection from shear forces, pathogens, and adverse chemical conditions. However, there is still a lot that is not known about it, in terms of chemical, physical, material and biological properties. Mucins are the major constituents of mucus, but their functions and complex underlying mechanisms behind these functions are still being elucidated. The mucosa is a surface that provides medical engineers with great opportunities, as well as challenges. Various significant phenomena take place at the mucosa, such as nutrient and drug adsorption, microbiome hosting and different pathologies. These can be leveraged to design new therapies to treat mucosa-related diseases. However, our poor understanding of mucus and mucins and of their interaction with the underlying tissue, it is still a challenge to develop optimal materials that can supplement mucus functions such as artificial saliva or tears, or interface with mucus to deliver drugs or cover mucosal wounds. The present thesis addresses the above aspects through the study of mucin interactions with various surfaces, and by exploring the potentials to use these for biomedical applications. Specifically:

❖ First, we aim to shed light on the lubricating properties of mucus by studying the adsorption and lubrication of mucins on hydrophobic surfaces (Paper I).

❖ Second, we engineer silk-based surfaces to interact with mucus, aiming to address the drug delivery and treatment potentials of the mucosa (Paper II).

1.2. Mucous membranes Mucous membranes line over 200 m2 of our body’s wet epithelia, including the gastrointestinal, urogenital and respiratory tracts, as well as the eyes (Figure 1). Mucous membranes are essential for keeping the underlying tissue moist and for protecting the body from adverse environments such as the gastric acid for the stomach lining and the urine for the bladder. Moreover, they protect the underlying tissues from shear forces by lubricating them, as in the case of swallowing food and waste passage (Florey 1955). Last, an extraordinary asset of the mucous membranes is that they act as a barrier to external threats, including pathogens, yet they are still permeable, for example making it possible for the nutrients to absorb along the gastrointestinal tract.

2

Figure 1. Illustration of part of the organs and cavities lined with mucous membranes. Mucous membranes line the organs of: the respiratory tract (oral and nasal cavities, larynx, trachea, lungs), the gastrointestinal tract (esophagus, stomach, small intestine, large intestine), and urinary tract (bladder, urethra). Picture modified with permission from VectorStock®.

1.3. Mucus All mucous membranes are lined by the viscoelastic hydrogel layer called mucus (Lai et al. 2009). Mucus is mostly composed by water (~95%), but also contains various proteins, lipids, and salts (~5%), with mucins being the protein constituent of paramount importance (~3%). Mucus provides to the mucous membranes moisture, lubrication and protection from adverse chemical environments. It also acts as a physical barrier for pathogens (Knowles and Boucher 2002), toxins (Strombeck and Harrold 1974) and foreign particles (Kreyling, Semmler, and Möller 2004). The mucus layer thickness and organization differs between different organs and even between the different areas of the same organ. In the case of the gastrointestinal tract, there are two distinct layers of mucus; an inner layer firmly adherent to the epithelial cells, and a loosely adherent outer layer. The inner layer is dense, thus not allowing bacteria to reach the epithelial surface, while the outer layer is loose, thus accommodating the commensal gut microbiota (Johansson, Holmén Larsson, and Hansson 2011). As we can see in Figure 2, the thickness of the inner and outer layers of mucus differ significantly along the gastrointestinal tract; for example, due to its role as the commensal microbiota habitat,

3

the loose mucus layer of the colon is seven times thicker compared to that of the stomach body (corpus) (Juge 2012). Simply put, the mucus layers all over the body’s mucous membranes are characterized by great variability and versatility. Mucus is secreted by goblet cells and submucosal glands in a sustained fashion (Verdugo 1990) (Birchenough et al. 2015) (Fischer et al. 2019), as depending on the tissue there is a different mucus turnover rate varying from seconds to hours (Lehr et al. 1991) (Schneider et al. 2018). That said, its ability to flow is an important quality of mucus. For example, in the upper part of the respiratory tract, mucus constantly moves with the help of hair-like projections called cilia, this way clearing the lungs from bacteria, viruses and other unwanted particles (Blake 1973) (Fischer et al. 2019). Such dynamic nature of mucus is made possible by its particular rheological properties which can be characterized by its storage modulus and loss modulus. Storage modulus is a measure for the elastic response of mucus through measuring the stored energy in it, while loss modulus is a measure for the viscous response through measuring the dissipated energy as heat. Mucus is considered a viscoelastic gel because it shows both viscous and elastic properties. More specifically, it is a non-Newtonian fluid, meaning that its viscosity can change under stress, with its properties varying between these characterizing a viscous liquid under high shear and an elastic solid under low shear (Lai et al. 2009) (Chhabra 2010). The rheological properties of mucus are crucial for its physiological functions, as any changes in them can affect its function as a lubricant, barrier and the body’s first line of defense against microbes (Randell, Boucher, and University of North Carolina Virtual Lung Group 2006).

Figure 2. Schematic representation of the mucus layers along the gastrointestinal tract of rats. The intestinal epithelium is covered by an inner, firmly adherent mucus layer, and an outer, loosely adherent mucus layer, where most microbiota can be found (represented by black dots). Figure reprinted from (Juge 2012) with permission from Elsevier.

4

1.4. Mucins The major macromolecular constituents of mucus are mucins. Mucins are a family of densely glycosylated proteins that are divided in gel-forming secreted mucins and membrane-tethered mucins. Gel-forming secreted mucins are the mucus constituent that gives mucus its viscoelastic properties. They are secreted by goblet cells (Verdugo 1990) (Birchenough et al. 2015) in the columnar epithelium, as well as by glandular mucous cells of the submucosal glands (Fischer et al. 2019), where prior to secretion they are partially condensed in secretory granules (Juan Perez-Vilar et al. 2005). Secreted mucins are elongated molecules characterized by two distinct regions; the large central hydrophilic region which is extensively glycosylated and gives mucins their characteristic “bottle brush” appearance, and the hydrophobic terminal regions that are very lightly glycosylated. Overall, mucin glycans cover 100 amino acids with 25-30 carbohydrate chains and account for up to 80% of the dry mucin weight, with the protein core making up the remaining 20% (Lamblin et al. 1991). The O-linked glycosylated central region is characterized by serine, threonine and proline rich tandem repeats (STP repeats), while the serine and threonine residues are linked to oligosaccharides typically less than 20 sugar units long via bonds with N-Acetylgalactosamine (GalNAc) (S. Lee et al. 2005). Finally, due to the high sialic acid and sulfate content of mucin sugars, mucins have an overall negative charge which makes them more rigid via charge dependent repulsion (Shogren, Gerken, and Jentoft 1989). In contrast, in the “bare” hydrophobic regions, which include the amino and carboxyl terminals and sometimes between the STP repeats, there is close to no O-linked glycosylation and only minimal N-linked glycosylation is observed. Such scarcity of glycosylation makes these regions sensitive to proteases (Bansil and Turner 2006). The mucin hydrophobic regions are rich in cysteine (>10%) and contain disulfide-rich domains structurally similar to von Willebrand factor C and D domains as well as C-terminal cystine knots. These disulfide-rich domains are of paramount importance because they are essential for the dimerization of the mucin monomers via disulfide bonds, the consecutive dimer polymerization into multimers and the assembly of the mesh-like structure of the mucus gel (J. Perez-Vilar and Hill 1999). The whole hierarchical structure of mucins from domain level to gel level is illustrated in Figure 3.

5

Figure 3. The hierarchical structure of mucins; from domain level to monomer, dimer, multimer and gel level. The constituent domains of the two distinct mucin regions (hydrophilic glycosylated and hydrophobic non-glycosylated) are shown in panel 1. The mucin monomer consists of a linear protein backbone with serine and threonine rich tandem repeats, where oligosaccharides (red hexagons) are O-linked via GalNAc (panel 2). Mucin monomers dimerize via forming C-terminal disulfide bonds of cysteine groups (panel 3). Mucin dimers polymerize further and form multimers with alternating hydrophilic and hydrophobic regions (red and blue respectively), while N-terminal branching can also occur occasionally (panel 4). A mucin gel forms by the crosslinking of the hydrophobic domains of the mucin multimers via disulfide bonds. The crosslinked hydrophobic regions are represented by green instead of blue circles (panel 5). Figure reprinted with permission from (Bansil et al. 2013). Again, in order to self-assemble the mucus gel, mucin fibers are reversibly crosslinked by disulfide bonds, bundled and entangled in a mesh-like structure which “traps” water in its pores. Another way with which the densely glycosylated mucins retain water is by their hydroxyl-rich glycans which can form hydrogen bonds with water at the molecular scale. Mucins are heavy molecules, with molecular weights in the megadalton (MDa) range; mucin monomers can weigh up to 2 MDa, while entangled mucins can result into networks of up to 50 MDa. Size-wise, the elongated molecules can span from 100 nm long for monomers, up to more than 10 μm long for mucin networks (Thornton, Rousseau, and McGuckin 2008), while the monomer diameter ranges between 3 and 10 nm (Handbook of Mucosal Immunology 2012). On a different note, membrane-tethered mucins do not drive mucus viscoelasticity but reportedly act as cell surface receptors, activating intracellular signaling pathways involved in cancer for example. Due to the aberrant expression of membrane-associated mucins in different tumors, they are promising targets for cancer diagnosis (Singh et al. 2006) and therapy (Singh et al. 2004) (Singh, Chaturvedi, and Batra 2007).

1.

2.

3.

4.

5.

6

In this study we used the gel-forming bovine submaxillary mucins (BSM) and porcine gastric mucins (PGM) as models. In general, BSM showcase a simple architecture with short glycan chains and a very high sialic content (≤ 30 %), similar to human salivary mucins (Bhavanandan and Hegarty 1987). In contrast, porcine gastric mucins showcase more complex glycosylation patterns and contain less sialic acid (1-3 %), similar to human gastric mucins (H. Nordman et al. 1997). On one hand, after the investigation of the carbohydrate structures of BSM, three acidic and five neutral core oligosaccharide structures were identified (Tsuji and Osawa 1986). From these, the major neutral core structures were found to be GalNAc-ol with relative abundance 27.2 % of total structures, while β-Gal-(1→3)-GalNAc-0l followed with 12.7% relative abundance (Mårtensson et al. 1998). On the contrary, the identified oligosaccharides in PGM were by far more and characterized by great structural diversity. Remarkably, Karlsson et al. characterized thirty different oligosaccharides with up to six monosaccharide residues. However, it is important to note that even though the oligosaccharides from PGM displayed vast structural diversity, not all of them were abundant; on the contrary, a lot of species were rare and the dominating species were Galβ1-3GalNAcα1- and Galβ1-3(GlcNAcβ1-6)GalNAcα1- structures. Moreover, it is clear that there are different glycosylation patterns observed between the distinct areas of the stomach. For example while oligosaccharides based on Galβ1-3GalNAcα1- and Galβ1-3(GlcNAcβ1-6)GalNAcα1- structures were widely distributed in all the stomach areas, GlcNAcβ1-3GalNAcα1- structures were present only in mucins from the cardiac region and corpus, and GlcNAcβ1-3(GlcNAcβ1-6)GalNAcα1- structures were found only in cardiac region mucins (Karlsson et al. 1997).

1.5. Mucin adsorption Mucus secretions have important protective and lubricating properties, primarily owing to their ability to form a gel layer adherent to the underlying epithelium. Since mucin solutions can recapitulate the mucus properties up to a level, forming mucin coatings on surfaces appears to be a relevant model to study mucus and as a mucus-mimicking biomaterial. Indeed, mucin coatings have long been used as mucus models in the field of microbiology, studying the colonization of the mucosa by different bacterial strains. For example, nasal mucin coatings on polystyrene have been used to study the colonization of nasopharyngeal mucosa by Staphylococcus aureus (Shuter, Hatcher, and Lowy 1996), while porcine gastric mucin coatings on polystyrene have been used to study the colonization of the gastrointestinal mucosa by Lactococcus lactis (Lukić et al. 2012). Mucin coatings have also found interesting applications, which we describe in more detail below. Mucin coatings are easily generated by simply exposing mucin solution to a surface. Indeed, PGM and BSM strongly adsorb and form stable coatings on hydrophobic surfaces, such as polystyrene (L. Shi and Caldwell 2000), polydimethylsiloxane (PDMS) (S. Lee et al. 2005) and hydrophobized mica (Malmsten et al. 1992). Specifically, BSM forms 4-5 nm thick monolayers on polystyrene (L. Shi and Caldwell 2000) and the monitoring of its adsorption with quartz crystal microbalance with dissipation (QCM-D) showed that mucins are adsorbed on the surface in a diffuse, viscoelastic layer (Feiler et al. 2007). The current hypothesis is that on hydrophobic surfaces, mucins extend their

7

glycosylated domains into the aqueous solution, while the hydrophobic non-glycosylated protein domains act as anchor points that interact with the hydrophobic surface. Mucins also adsorb on hydrophilic surfaces such as ThermanoxTM, a trademarked polyester film surface which is modified to be hydrophilic for cell adherence (Sandberg, Carlsson, and Ott 2007), as well as mica (Perez and Proust 1987) and silica (Lindh et al. 2002). The above surfaces have a net negative charge, and so do mucins due to their negatively charged sugars. This indicates that mucin molecules electrostatically interact with these surfaces by positively charged anchor points in their non-glycosylated regions, in a fashion similar to the hydrophobic interactions described before. Thus, the glycosylated regions are extended into the aqueous solution forming the distinctive bottle brush structure, while the non-glycosylated regions get anchored on the surface. Adsorbed mucins can take different conformations resembling either mucin brushes (Figure 4a) or a flatter conformation (Figure 4b).

Figure 4. Model for the conformation of adsorbed mucins on a negatively charged hydrophobic surface (e.g. polystyrene). (a) Mucin brush, (b) flat conformation. Green represents the mucin sugars, while red represents the positively charged protein and blue the negatively charged protein. Reprinted with permission from (Yakubov et al. 2007). Copyright 2007 American Chemical Society. Another hydrophilic surface that mucins adsorb well on is gold. In this case, the main driving force for adhesion is the special affinity of gold and sulfur for one another (Pacchioni 2019); gold-sulfur bonds (Au-S) are thus formed between the direct interactions of mucins with the gold core (Ouellette et al. 2018). Lastly, it is important to note that in addition to surface properties (T. Crouzier et al. 2013) and the mucin type used (Madsen et al. 2016), mucin adsorption depends on coating conditions such as pH, ionic strength (Celli et al. 2007) (S. Lee et al. 2005) and temperature (McColl, Yakubov, and Ramsden 2008). Looking at the versatility of

8

mucins, it is important to stress that no mucin interaction with a surface is just governed by one and only driving force. However, there is often a “dominant” force involved. For example, polystyrene is both hydrophobic and negatively charged, but even if the electrostatic interactions between mucin and polystyrene were shielded by high ionic strength, mucins would still adsorb on the hydrophobic surface through the hydrophobic interactions. In a nutshell, mucin molecules are able to adsorb and form a dense hydrated layer on a wide range of surfaces chemistries. This unique ability to adsorb on such a variety of surfaces is greatly owed to the diverse chemistry and architecture of mucins; their amphiphilic and polyampholytic nature, their thiol groups that can form different bonds, their hydroxyl-rich sugars that can form hydrogen bonds with water, all contribute to the impressive versatility of mucins (Yakubov et al. 2007) and can be seen summarized in Table 1. Along these lines, mucin coatings can be used for a variety of applications on mucosal surfaces where mucins naturally occur, but also on other biomedically interesting surfaces, for instance implant materials such as titanium (Lori and Nok 2004) and hydroxyapatite (Berg et al. 2001).

Interaction types Surface examples References

hydrophobic hydrophobic surfaces; polystyrene, PDMS, hydrophobized mica

(L. Shi and Caldwell 2000), (S. Lee et al. 2005), (Malmsten et al. 1992)

electrostatic negatively charged surfaces; ThermanoxTM, polystyrene, mica, silica

(Sandberg, Carlsson, and Ott 2007) (L. Shi and Caldwell 2000), (Perez and Proust 1987), (Lindh et al. 2002) (Lindh et al. 2002)

gold-sulfur bonds (Au-S) gold (Ouellette et al. 2018)

Table 1. Mucin interactions and mucin-interacting materials. The table provides a sample of the interactions that mucins are capable of, as well as examples of surfaces where these interactions occur.

1.5.1. Utilizing the barrier properties of mucins in the form of coatings

The interactions of mucins with various surfaces have been exploited in order to create functional surfaces that act as barriers towards pathogens and other cells. Such functional surfaces are relevant for the research field of implants and biomimetic materials, where for example a bacterial infection could cause the rejection of the implant. In point of fact, BSM coatings on polymethyl methacrylate (PMMA), silicon, Tecoflex polyurethane and polystyrene surfaces demonstrated increased wettability and anti-adhesive properties against Staphylococcus aureus and Staphylococcus epidermidis (Lei Shi et al. 2000) (Lei Shi et al. 2001). Moreover, BSM and PGM coatings on

9

Thermanox substrates were shown to suppress the adhesion of neutrophils (Sandberg, Carlsson, and Ott 2007) (Sandberg et al. 2009). Overall, the fact that mucin coatings can make a hydrophobic surface hydrophilic, makes them very promising tools for eliminating the unwanted protein and cell/microbial adhesion occurring on implants (Lei Shi 2000).

1.5.2. Utilizing the lubricity of mucins The hydration and lubrication that mucins offer to the epithelia, together with the fact that purified mucins can form coatings which contain approximately 95% water on a broad range of surfaces, make mucins good candidates for the development of biolubricants. In particular, mucin coatings offer boundary lubrication, turning hydrophobic surfaces into hydrophilic surfaces (Coles, Chang, and Zauscher 2010). The applications for which mucins could be used as lubricants, or inspire the development of other mucin-like molecules are plenty and not only limited to where mucin lubrication occurs physiologically. As a matter of fact, qniumucin from Aurelia aurita jellyfish was previously injected in the joints of a rabbit osteoarthritis model and found to inhibit the cartilage degeneration (Ohta et al. 2009). In the same light, PGM solution outperformed hyaluronic acid and lubricin in protecting sheep knee joints from wear (Boettcher et al. 2017). Moreover, a PGM-based saliva substitute was shown to reduce the wear of austenitic stainless steel, which is the most common metal used for dental applications (Mystkowska, Łysik, and Klekotka 2019). Looking at the lubrication of surfaces where mucins occur physiologically, PGM tested for the lubrication of contact lenses resulted to not only be equally good lubricants as hyaluronic acid, but also prevented the damage of corneal tissue explants (Winkeljann et al. 2017). However, although the mucin-containing oral spray “Saliva Orthana” was helpful for xerostomic hospice patients, it did not have any significant advantage compared to a placebo spray (Sweeney et al. 1997) The lubricating properties of mucins can be affected by various environmental factors. In vitro tests have shown that protein degradation or deglycosylation of the mucins can compromise their hydrating and lubricating properties. However, the compromised lubricity and hydration of these mucins was able to be “rescued” by the grafting of polyethylene glycol (Thomas Crouzier et al. 2015). It should also be noted that mucin lubricity has been shown to be dependent on pH and ionic strength (S. Lee et al. 2005), reflecting the physiological conditions where the different mucin types usually lubricate the different tissues. Thus, PGM demonstrates optimal lubricity at acidic pH assimilating the pH of the stomach, while BSM shows enhanced lubricity at neutral pH similar with the salivary conditions (S. Lee 2013). As can be seen summarized in Table 2, mucins have demonstrated lubricating or other beneficial effects when coating biological surfaces, as well as biomedically interesting surfaces. Taking into account that mucin lubricity can be “rescued” by chemical modification, we understand that there are various opportunities in store for mucins in the field of biolubrication.

10

Application Mucin type Surface Observation Reference

Utilizing the barrier

properties of mucins

BSM PMMA, silicon, Tecoflex polyurethane, polystyrene

increased wettability, anti-adhesive properties against Staphylococcus aureus and Staphylococcus epidermidis

(Lei Shi et al. 2000)

BSM, PGM ThermanoxTM suppressed the adhesion of neutrophils

(Sandberg, Carlsson, and Ott 2007)

Utilizing the lubricity of

mucins

Jellyfish mucin rabbit joint inhibited cartilage degeneration

(Ohta et al. 2009)

PGM-based saliva substitute

austenitic stainless steel

reduced wear (Mystkowska, Łysik, and Klekotka 2019)

PGM contact lens (polyvinylpyrrolidone embedded into hydroxyethyl methacrylate/methacrylic acid)

equally good lubrication as hyaluronic acid, prevented the damage of corneal tissue explants

(Winkeljann et al. 2017)

PGM-based saliva substitute “Saliva Orthana”

oral mucosa from xerostomic patients

helpful but not better than placebo

(Sweeney et al. 1997)

Table 2. Types of applications for mucin coatings. A part of the published studies investigating the different applications of mucin coatings is presented shortly.

1.6. Mucoadhesive materials The knowledge gained on mucin chemistry and on mucin interactions with different surfaces can be used to engineer materials that are optimized for interaction with mucins. These are called mucoadhesive materials and are useful for the development of mucosal wound dressings and non-invasive transmucosal drug delivery systems. One of the main challenges that mucoadhesive material research is called to overcome is the mucosal environment itself; the high degree of hydration and lubrication, as well as the dynamic mucus turnover are calling for materials showcasing strong mucoadhesion.

11

Mucoadhesion is defined as the state in which a polymer material and the mucus gel covering a mucous membrane are held together for extended periods of time through adhesive forces. Mucoadhesion occurs due to a range of physicochemical interactions between the material and mucosal environment and gets established through two stages. The first stage includes the initial close contact of the material and the mucosal surface, and the second stage the consolidation period during which the molecular mucoadhesive bonding is getting reinforced (Cook et al. 2017). There are various mucoadhesion theories which present different mechanisms as determinants of mucoadhesion, and the most accepted ones are shortly presented below and in Figure 5:

● Wetting theory typically concerns liquid mucoadhesives and describes the ability of such polymers to spread and swell on the wet mucosa. The better the spreading over the surface, the stronger the mucoadhesion (Yu, Andrews, and Jones 2014).

● Dehydration theory describes the contact of a water-absorbing material that is capable of gelling with the wet mucosa. Once in contact with the wet mucosa, the material dehydrates the mucosa, this way adhering onto it (S. A. Mortazavi and Smart 1993).

● Diffusion theory describes the entanglement of the mucoadhesive polymer and the mucin fibers through interpenetration, which in turn allow more bonds to form to strengthen the adhesion (Jabbari, Wisniewski, and Peppas 1993).

● Adsorption theory concerns the covalent and non-covalent interactions between the mucosa and the polymer. The covalent interactions include Van der Waals forces, hydrogen bonds, and hydrophobic interactions (Mikos and Peppas 1989), while covalent bonding is for example possible through disulfide bonds between thiolated polymers and the cysteines of mucins (Bernkop-Schnürch 2005).

● Electronic theory describes the electron transfer between the mucoadhesive polymer and the mucus, which results in a charged double layer at the mucin-polymer interface (Derjaguin, Aleinikova, and Toporov 1994).

● Mechanical theory concerns the effect that the contact area has on the polymer-mucosa interactions. A good example for that is the architecture of the tongue, where the saliva layer is thin and due to the papillae there is a large contact area for the mucoadhesive polymers (Yu, Andrews, and Jones 2014).

Practically, mucoadhesion is characterized by a combination of these theories, even if there is usually a dominant force governing the adhesion.

12

Figure 5. The main mechanisms of mucoadhesion: adsorption, dehydration, diffusion, electronic, mechanical, wetting. Adsorption includes hydrogen bonds and disulfide bonds, dipoles which interact with the negatively charged mucins, and hydrophobic interactions between a colloid and the mucosa (from left to right). Dehydration shows the water uptake of a gel from the mucus and consequent adherence. Diffusion demonstrates the interpenetration of the mucoadhesive polymer chains with the mucin fibers. The mechanical mechanism shows how architectural irregularities increase the contact area and interactions between a mucoadhesive and the mucosa. Wetting demonstrates the importance of the contact angle in the consecutive spreading of liquid mucoadhesives. Finally, the electronic mechanism shows the formation of a charged double layer. Figure reprinted with permission from (Cook et al. 2017). Mucoadhesive materials can be classified as first generation or second generation mucoadhesives. First generation mucoadhesives are natural or synthetic hydrophilic molecules that adhere non-specifically with the mucosal surfaces, relying on their intrinsic properties. Such examples are the polysaccharides chitosan and alginate, both extracted from natural sources (shellfish and brown algae respectively). The positively charged chitosan develops electrostatic interactions with the negatively charged sugar moieties of mucins like sialic acid (Lehr et al. 1992), while negatively charged alginate appears to display non-covalent rather than electrostatic interactions with the mucin molecules (Fuongfuchat et al. 1996). Since first generation mucoadhesives provide non-specific interactions and in order to increase the specificity of mucoadhesion, second generation mucoadhesives have been developed. These polymers bind on specific chemical structures of the mucins or the epithelial cells underneath. Such examples are plant lectins, which non-covalently bind on glycosylated components of the cells as well as mucins (Woodley 2001), or thiolated polymers, which form disulfide bridges with the cysteine rich domains of mucins. For example, thiolated chitosan has shown enhanced mucoadhesive properties compared to unmodified chitosan (Bravo-Osuna et al. 2007). The downside of all these molecules is

13

that they have to rely on bound therapeutic molecules to provide extra functionalities. Therefore, developing multifunctional mucoadhesive materials which can still strongly interact with mucous membranes and have present extra functionalities such as pro-wound healing properties is of particular interest. The methods that can be used for the evaluation of mucoadhesive polymers are mainly separated in two groups. On one hand, there are the direct methods which focus on the force needed to break the adhesion, such as texture analyzers (Hägerström and Edsman 2001) and atomic force microscopy (AFM) (Sudhakar, Kuotsu, and Bandyopadhyay 2006). On the other hand, we have the indirect methods which focus on the interactions between the polymer and mucus/mucins, such as quartz crystal microbalance with dissipation (QCM-D) (Oh et al. 2015). and surface plasmon resonance (SPR) (Thongborisute and Takeuchi 2008). Finally, mucoadhesive materials come in all shapes and sizes, from nano- (Kawashima et al. 2000) and micro-particles (Choy, Park, and Prausnitz 2008), to gels (Xu et al. 2017), capsules (Gupta et al. 2013), tablets (Sogias, Williams, and Khutoryanskiy 2012), and patches (Perioli et al. 2004), and can aim for the treatment of any mucosal surface. In Table 3 we can see selected examples of publications dealing with different mucoadhesive material forms and their applications.

Material form

Material Target surface

Applications

Observations References

nano- particles

chitosan coated DL-lactide/ glycolide copolymer spheres

intestinal mucosa (rat in vivo)

delivery of elcatonin (lowers blood calcium)

chitosan coated spheres reduced significantly the blood calcium levels compared with elcatonin solution and uncoated spheres

(Kawashima et al. 2000)

micro- particles

poly(ethylene glycol) coated poly(lactic-co-glycolic acid) spheres

ocular mucosa (rabbit in vivo)

drug delivery micro-particles stayed 30 minutes on the ocular surface of rabbits

(Choy, Park, and Prausnitz 2008)

gel catechol- modified chitosan

rectal mucosal (mouse in vitro)

delivery of sulfasalazine (treats ulcerative colitis)

more effective than oral sulfasalazine

(Xu et al. 2017)

capsule ethyl cellulose coated polymeric matrix

intestinal mucosa (porcine in vitro)

delivery of salmon calcitonin (lowers blood calcium)

enhanced oral bioavailability

(Gupta et al. 2013)

14

tablet chitosan and half-acetylated chitosan

oral mucosa (porcine gastric in vitro)

delivery of ibuprofen (anti-inflammatory)

higher drug release for the acetylated chitosan

(Sogias, Williams, and Khutoryanskiy 2012)

patch carboxymethylcellulose sodium salt

oral mucosa (porcine buccal in vitro and human buccal in vivo)

delivery of ibuprofen (anti-inflammatory)

ibuprofen was present in saliva for up to 5 hours

(Perioli et al. 2004)

Table 3. Types of mucoadhesive material formulations. A part of the published studies investigating different mucoadhesive material formulations for various drug delivery applications are presented shortly.

1.7. Mucosal diseases Knowledge around mucin-material interactions and knowledge around mucin-interacting materials are important since they can help in understanding and treating a number of mucus related diseases. For such diseases, targeted treatments for the mucus or underlying epithelium are needed; therefore the developments in mucoadhesive systems presented before are a useful tool to accomplish that.

1.7.1. Compromised barrier One group of mucosal conditions are mucosal wounds and ulcers, during which the mucus barrier gets disrupted and the underlying epithelium is left susceptible to wound formation. The most common example of mucosal ulcers are mouth sores (Politis et al. 2016), a condition that is recurrent for some, or a comorbidity for immunocompromised people due to chemotherapy (Lalla, Sonis, and Peterson 2008), the human immunodeficiency virus, or autoimmune diseases (Rodsaward et al. 2017). Other common mucosal ulcers are stomach ulcers, caused by Helicobacter pylori infection (Kuipers, Thijs, and Festen 1995), or ulcerative colitis ulcers, caused by inflammation in the gut (Collins and Rhodes 2006). Like for any other wound, there needs to be a very finely orchestrated balance between pro- and anti-inflammatory mechanisms for a mucosal wound to heal (Engeland, Marucha, and Manos 2013). However, the fact that the mucosal epithelium is exposed to a wet and many times adverse environment (e.g. stomach acid, or food in the mouth), makes mucosal wounds harder to heal and more prone to infection than skin wounds. Therefore, in addition to the mechanical protection and coverage that a mucosal wound dressing would offer, it should also preferably promote cell proliferation and wound

15

healing for the epithelium to be restored. Moreover, antimicrobial properties might also be necessary to lower the risk of infections. So far, there has been some attempts to accelerate oral wound healing by using tissue engineering. Fibrin sheets pre-vascularized with fibroblasts, keratinocytes, and endothelial cells demonstrated accelerated buccal wound healing in rats (J. Lee et al. 2017). A similar model tested on rat tongue wounds showed early re-epithelialization, increased wound healing and minimal fibrosis (Roh et al. 2017). The tissue engineering approach where materials are implanted in combination with cells to help regenerate the mucosal tissue, is a promising approach for healing mucosal wounds. However, its short “shelf life” and complicated manufacturing due to the presence of cells would be an obstacle for wide commercial development. Thus, mucoadhesive materials loaded with antioxidants, pharmaceuticals or growth factors are an interesting alternative. Such examples are a phenytoin containing mucoadhesive paste that increased the wound healing of patients after oral biopsy (H. Mortazavi et al. 2014), or an anthocyanin complex mucoadhesive gel that had similar effects in rats and orthodontic wound patients (Limsitthichaikoon et al. 2018). However, mucoadhesive films loaded with epidermal growth factor on human buccal tissue models did not accelerate wound healing but instead resulted in hyperkeratosis, raising questions on how appropriate these commercial in vitro models are for studying wound healing (Ramineni et al. 2015).

1.7.2. Loss of hydration and lubrication Another group of mucosal conditions has mucous membranes with an insufficiently hydrating and lubricating layer of mucus as a common denominator. Such conditions are oral and ocular dryness, which can also be in syndromic form as Sjögren’s syndrome (Mavragani and Moutsopoulos 2014) (Błochowiak et al. 2016), as well as vaginal dryness (McClelland, Holland, and Griggs 2015) (Leiblum et al. 2009). Apart from the discomfort and tissue irritation caused, these conditions also increase the risk of viral and bacterial infections given that mucus is the body’s first line of defense. Although there are some topical treatments for ocular, oral (van der Reijden et al. 1999) and vaginal dryness, they are not satisfactory because of the short duration of the hydration and lubrication relief they bring, or because of their potential side effects. For instance, in the case of ocular dryness, the commonly used preservative-free artificial tears require frequent re-application (Whitcher 2004). For treating menopause-related vaginal dryness oral or topical hormone replacement therapy with estrogen is practiced, although accompanied with side-effects such as increased association with specific cancer types (Tavani and La Vecchia 1999). For that reason, it is recommended that non-hormonal solutions are used. Such an example is a new vaginal moisturizing cream containing distillate from Hamamelis virginiana which showed long-lasting moisturizing effects in a clinical trial (Henneicke-von Zepelin et al. 2017). Finally, there have been a lot of attempts to develop artificial saliva for treating oral dryness, without great success. For example, the PGM-based “Saliva Orthana” did not have any better results compared to placebo spray (Sweeney et al. 1997), while “Saliva Natura”, which contains Eriodictyon crassifolium extract, showed a demineralizing effect on dentin (Tschoppe et al. 2007).

16

To address the challenge of the re-hydration and re-lubrication of the dried out mucous membranes, strategies that mimic or restore the hydrating and lubricating properties of mucins are needed. A very interesting approach described the fusion of the hydrated compound polyethylene glycol (PEG) with mucin-binding lectins, and the subsequent immobilization of the PEG-lectin conjugates on adsorbed mucins. This resulted in the PEG-lectin conjugates restoring the hydration of mucous layers in vitro (Thomas Crouzier et al. 2015).

1.7.3. Altered Rheology The most well-known and well-studied diseases that are characterized by altered mucus rheology are the lung diseases cystic fibrosis and chronic obstructive pulmonary disease (COPD). Cystic fibrosis is an inherited disease caused by a mutation if the gene of cystic fibrosis transmembrane conductance regulator (CFTR), which encodes for a chloride channel expressed in various epithelial membranes (Gregory et al. 1990). The different mutations result in defective ion transport which leads to increased airway mucus viscosity. In turn, this results in impaired mucociliary clearance and high risk for lung infection (Verkman, Song, and Thiagarajah 2003) (Hill et al. 2018). There is no cure for cystic fibrosis, just treatments ameliorating the symptoms. Some approaches are mucus thinners which aim to hydrate the mucus, such as hypertonic saline (Tildy and Rogers 2015), antibiotics to treat the infections and bronchodilators to ease the breathing. There is a lot of research on gene therapy for cystic fibrosis, but even though over 20 approaches reached the clinical trial stage, the outcome was not enough to reach clinical benefit (Burney and Davies 2012). Chronic obstructive pulmonary disease (COPD) on the other hand, is a chronic inflammatory lung disease that is characterized by progressive airflow obstruction. COPD is mainly caused by smoking, but genetic and other environmental factors such as pollution contribute to the disease (Decramer, Janssens, and Miravitlles 2012). As in the case of cystic fibrosis (Henderson et al. 2014), the sputum of COPD patients is characterized by increased mucin concentration (Kesimer et al. 2017). Similarly with cystic fibrosis, there is no cure for COPD, but smoking cessation and treatments like steroids, bronchodilators and antibiotics to treat infections are used to ease the symptoms and slow down the disease progress. To summarize, given that the existing solutions for treating cystic fibrosis and COPD are only ameliorative, approaches for changing the pathological mucus rheology into its physiological state are desired. Such an approach that has been investigated is using guluronate oligomers to lower the mucus density. Guluronates are oligosaccharides from alginate which have shown the ability to disrupt the network cross-links of purified mucins as well as cystic fibrosis sputum, resulting in a weakened and less viscous mucin network (Nordgård and Draget 2011). In addition, the guluronic acid oligomers have enabled improved nanoparticle mobility through the mucus and uptake to the cells, making them also good candidates for the mucosal delivery of nanomedicine (Nordgård et al. 2014). This and many other works are an example of how developing our

17

understanding of polymer and material-mucin interactions can lead to potential new treatments for mucus-related diseases.

18

2. Present investigations

2.1. Aim of present investigations

Specific aim of Paper I The aim of Paper I was to investigate the key features that make mucins exceptionally good surface lubricants. Specific aim of Paper II The aim of Paper II was to functionalize the partial spider silk protein 4RepCT with mucoadhesive properties using genetic engineering.

2.2. Materials and methods

2.2.1. Mucin purification Mucins can be extracted from various animal sources, such as chicken eggs (ovomucin), jellyfish (qniumucin), snails, cows, and pigs, with bovine submaxillary mucins (BSM) and porcine gastric mucins (PGM) being the most relevant mucin models to human. Mucins present a highly diverse set of chemical functionalities and functional properties; their protein sequence, size and glycosylation patterns differ among organisms, tissues, and tissue areas (Karlsson et al. 1997) (Henrik Nordman, Davies, and Carlstedt 1998), as well as according to the disease state of the underlying epithelium (Larsson et al. 2011), and the microbiota (Arike, Holmén-Larsson, and Hansson 2017) (Padra et al. 2018). Therefore, when purifying mucins it is common to have batch to batch variations; in-house purified PGM can vary in its dissolution time and rheology (Schömig et al. 2016), while BSM from Sigma can vary in the amount of insoluble particles or associated protein contamination. All complex biological materials are susceptible to such batch to batch variations and that can be explained mostly by the variations between animals. These variations can be minimized by the pooling of as many tissues/mucin extracts as possible. Mucins can be extracted either by collecting the mucus itself, as in the case of PGM when mucus is scraped out of the stomach interior (Corfield 2000), or by homogenizing whole tissues, as in the case of BSM when the salivary glands of cows are homogenized (Tettamanti and Pigman 1968) (Schoemig et al. 2017). Purification of mucins is achieved after the dissolution of the crude material and following a series of centrifugations, size exclusion chromatography, and filtrations, during which mucin-bound impurities, lipids and smaller proteins are being removed (Schömig et al. 2016) (Allen et al. 1989). Preserving the structure and therefore the natural properties of native mucus - like its gel-forming properties - is desired for purified mucins. Currently, commercially available mucins (e.g. Sigma M3895 BSM and Sigma PGM M1778) do not retain their gelling properties, probably due to the harsh treatments which the mucins undergo, such as

19

protease digestion, which cleaves the non-glycosylated mucin regions. Moreover, it is often that commercial mucins still contain impurities and they need extra purification steps before use. For example, bovine serum albumin (BSA) has been found to be the main contaminant of Sigma M3895 BSM, containing approximately 9% w/w BSA in one of its batches (Lundin et al. 2009), while in other batches which we have used the BSA contamination is of insignificant amount. On the contrary, PGM purified in-house maintain their gelling properties and show low levels of protein impurities, but the purification procedure is rather tedious. In conclusion, there is a need for better quality control and standardized purification protocols of commercial mucins in order to achieve optimal reproducibility while preserving the natural properties of mucins. Such improvement could assist the progress of the field.

2.2.2. Quartz crystal microbalance with dissipation monitoring (QCM-D) and surface plasmon resonance (SPR); Monitoring adsorption and measuring hydration

For investigating mucin coatings we needed specialized techniques that are particularly sensitive to surface events, with limits of detection in the ng/cm2 range. Quartz crystal microbalance with dissipation monitoring (QCM-D) and surface plasmon resonance (SPR) are such sensitive methods which enable us to rapidly monitor the adsorption of materials on a variety of surface matrices in real time. A typical QCM-D sensor consists of a circular quartz crystal sandwiched between two gold electrodes. Given that quartz is piezoelectric, the application of alternating electric voltage on the sensor results in its excitation and alternating expansion and contraction (oscillation). In 1959 Sauerbrey established the quartz crystal microbalance (QCM) method as a mass balance method by proving a relationship between the frequency change (Δf) of the oscillating quartz crystal and the adsorbed mass (Δm) on its surface. The equation that describes this relationship is the following:

𝛥𝛥𝛥𝛥 = �𝐶𝐶𝑛𝑛� 𝛥𝛥𝑓𝑓 (1)

where n stands for the harmonic number and

𝐶𝐶 = 𝑡𝑡𝑞𝑞 �𝜌𝜌𝑞𝑞𝑓𝑓0� (2)

with tq being the thickness of the quartz crystal, and ρq the density of quartz, which is approximately -17.7 Hz ng/cm2 for the 5 MHz crystal used in our work (Sauerbrey 1959). In addition to the information about the total adsorbed hydrated mass that QCM provides already through the frequency response, QCM-D also provides us with information about the viscoelastic properties of the adsorbed material through the measurement of dissipation. Dissipation is the summation of all the energy loss in the

20

system per oscillation cycle. In particular, the dissipation measurements provide us with information about changes in the conformation and the softness/rigidity of the adsorbed molecules, allowing us to delve deeper in the material properties and on how the molecule adsorbs on the surface. For example, a soft and rich in water film gets deformed during oscillation, resulting in high dissipation, while a rigid film with less water does not get deformed and results in low dissipation (Dixon 2008). The QCM-D raw data are displayed with the frequency changes (Δf) on the left y axis and the dissipation changes (ΔD) on the right y axis, both plotted against time at the x axis. As can be seen in Figure 6, the decrease in frequency corresponds to the increase of deposited mass on the sensor, while a higher increase in dissipation corresponds to a softer and more diffuse layer being deposited (c).

Figure 6. Example of QCM-D raw data. The frequency changes are in blue (Δf) on the left y axis, while the dissipation changes (ΔD) are in red on the right y axis, and both are plotted against time at the x axis. This example shows the adsorption of human serum albumin (a) on a gold-coated quartz crystal sensor, followed by wash with buffer (b) and then a specific antibody binding human serum albumin (c), followed by a last washing step with buffer (d). The frequency change and minimal dissipation change at step a indicates the rigidity of the human albumin adsorbed on the sensor surface, while the high frequency and dissipation change at step d shows the higher mass of the antibody binding, as well as the soft and diffuse character of the bound antibodies. Figure reprinted with permission from (Dixon 2008). Liedberg et al. first demonstrated the use of surface plasmon resonance (SPR) for biosensing purposes in 1983 (Bo Liedberg, Nylander, and Lunström 1983), while the first BIACORE SPR instrument was sold in 1990 by Pharmacia (B. Liedberg, Nylander, and Lundström 1995). Like QCM-D, SPR measures mass adsorption on the sensor surface, but the measurement is achieved by using a different physical principle; while QCM-D is an acoustic technique, SPR is an optical technique. Hence, the SPR phenomenon occurs when polarized light shed through a prism hits the gold sensor surface at the interface of media with different refractive indices and the reflected light is then collected and analyzed (Tang, Zeng, and Liang 2010). The binding of molecules on the sensor surface results in refractive index change, subsequently inducing the shift of the SPR angle. The result of the detection is then displayed as a

21

sensorgram where the resonance signal on the y axis is plotted against time at the x axis (Figure 7). In 1990, the method for quantifying the protein surface concentration from SPR response was developed. Radiolabeled proteins were adsorbed via electrostatic interactions to a hydrogel matrix on an SPR sensor surface, and the resulting change in the SPR angle was correlated to the absolute protein amount as detected from radioactive measurements. The correlation between SPR angle shift and material mass was then found to be linear for the detection limits of the method (Stenberg et al. 1991), meaning that the obtained shift in the resonance angle is proportional to the mass of material bound on the surface. In summary, since the estimation of the mass adsorbed on the SPR sensor is based on the refractive index difference between the adsorbed molecules and the displaced water, water is not included in the measurement but the dry molar mass is measured instead.

Figure 7. Surface plasmon resonance (SPR) principle. The adsorption of molecules to the gold sensor surface increases the refractive index, and subsequently results in the shift of the SPR angle. The result of the detection is displayed as a sensorgram where the resonance on the y axis is plotted against time at the x axis. Figure reprinted from (Cooper 2002) with permission from Springer Nature. In conclusion, both QCM-D and SPR are sensitive, label-free detection methods, valuable for monitoring molecule adsorption on surfaces in real time. Even though SPR is intrinsically more sensitive with a 21 times lower mass detection limit compared to QCM-D (Su, Wu, and Knoll 2005), the use of SPR is more relevant for monitoring the formation of compact dense layers. For water-rich layers QCM-D is favored, allowing the better measurement of thicker films. In addition, QCM-D provides the additional dissipation information which is very relevant for diffuse layers. To summarize, one should be aware of each method’s limitations and choose according to the type of molecule studied. Either way, the combination of the methods can provide valuable information. For example, knowing the hydrated mass of the adsorbed molecules from

22

QCM-D as well as their dry mass from SPR, we are able to estimate the hydration level of the adsorbed coatings by using the following formula:

ℎ𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑡𝑡𝑦𝑦𝑦𝑦𝑦𝑦 (%) = ℎ𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦 𝑚𝑚𝑦𝑦𝑚𝑚𝑚𝑚−𝑦𝑦𝑦𝑦𝑦𝑦 𝑚𝑚𝑦𝑦𝑚𝑚𝑚𝑚ℎ𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦𝑦 𝑚𝑚𝑦𝑦𝑚𝑚𝑚𝑚

× 100 (3)

2.3. Results and Discussion

2.3.1. Mucin-inspired lubrication on hydrophobic surfaces (Paper I)

The mucus gel which covers our wet epithelia has various important functions, among which is the mechanical protection of the underlying epithelia from shear forces. Some forms of applied shear forces on the mucosa are for example eye blinking, as well as the travel of food through the gastrointestinal tract. The key components of mucus are mucins, a family of densely glycosylated proteins which provide mucus its shear-protecting and lubricating properties. Mucin molecules are comprised from non-glycosylated hydrophobic ends and a highly glycosylated hydrophilic central domain, with the latter being responsible for the retention of large amounts of water. Purified mucins from animal sources, when reconstituted into solution, can reduce friction between both artificial and biological surfaces, in a fashion similar to native mucus. However, the mechanisms and key features responsible for the lubricating properties of mucins are not known yet. In this work, we aimed to identify the key features of mucins which make them outstanding boundary lubricants. The mucins that we used for the investigation are porcine gastric mucins (PGM), as well as human salivary mucins, both purified in-house. Throughout Paper I and the present section, the porcine gastric mucins are abbreviated as MUC5AC (Mucin 5Ac) and the human salivary mucins as MUC5B (Mucin 5B), representing respectively the most abundant mucin populations in the two mucin types. Our findings demonstrate that the hydrophobic terminal domains of mucins are crucial for its adsorption and lubricity on hydrophobic polydimethylsiloxane (PDMS) surfaces. The mucin adsorption on PDMS was monitored in real time using quartz crystal microbalance with dissipation monitoring (QCM-D) on PDMS coated quartz crystals, while the lubricity was measured using a shear rheometer equipped with a tribology unit (ball-on-cylinder setup). To investigate the role of the terminal hydrophobic domains in mucin lubricity, tryptic digestion of the mucins was performed to cleave off the hydrophobic domains. After the digestion, the mucin fragments were separated with size exclusion chromatography, resulting in a main peak corresponding to the glycosylated mucin fragments, and a smaller peak corresponding to the peptide fragments and the trypsin (Paper I, Figure 2a,b). The glycosylated mucin fragments were then tested for their lubricity, compared to non-treated mucins and a buffer control. Interestingly, the digestion had a large effect on the mucin rheology of both MUC5AC and MUC5B, showing an almost complete loss of lubricity for the treated mucins (Paper I, Figure 2c,d).

23

It should be noted that this dramatic decrease in the lubricity of treated mucins was not due to a change in hydration, as there was no significant reduction in the amount of water bound by the treated mucins when adsorbed on the surfaces (Paper I, Figure 3). The hydration level of the coatings was determined by combining the information about the hydrated mucin mass, obtained by QCM-D, and the dry mucin mass obtained from surface plasmon resonance (SPR) measurements (Paper I, Equation 1). We thus understand that mucin hydration is not the cornerstone of mucin lubrication, and that apparently the hydrophobic terminal domains do play a crucial role in that. An explanation for the dramatic loss of lubricity of the mucins in the boundary lubrication regime could be the reduced adsorption of the enzymatically treated mucins, which was investigated by QCM-D. Indeed, the treated mucins showed a greatly reduced adsorption on hydrophobic PDMS surfaces (Paper I, Figure 4a,b). These results suggest that mucin molecules adsorb on hydrophobic surfaces via their terminal hydrophobic domains and that mucin adsorption itself is also crucial for efficient lubrication. To verify the hypothesis that hydrophobic domains are crucial for the adsorption and consequently for the lubricity of mucins and other molecules, we used a bottom-up approach; dextrans, which are strongly hydrated complex branched glucans that normally do not adsorb on PDMS, were modified by the grafting of hydrophobic phenyl groups on them (0.15 density) and tested for their adsorption on PDMS. The modified dextrans showed increased adsorption on PDMS compared to their unmodified counterparts, but not as high as the untreated mucins. That observation resulted in the further investigation of how important the density of hydrophobic moieties is for efficient adsorption. Therefore, dextrans were modified with a higher number of hydrophobic groups per molecule (0.4 density) and tested for their adsorption on PDMS. Indeed, a higher phenylation density caused higher adsorption of dextrans on PDMS, in a magnitude similar with the untreated mucins (Paper I, Figure 4c). Moreover, we showed that higher phenylation density also results in better lubricating properties of the dextrans, especially in the mixed lubrication regime. In addition, we observed that regarding lubricity, it is the density of hydrophobic groups that matters and not the overall number of hydrophobic groups. That was shown by comparing the lubricating ability of 40 kDa and 150 kDa dextrans with equal phenylation density, and obtaining identical results for both sizes (Paper I, Figure 5a). Overall, our results indicate the two features that a polymer needs in order to be a good boundary lubricant; first it should be a hydrated polymer, but most importantly it needs to adsorb on the surface too. Finally, since tryptic digestion not only removes the hydrophobic domains of mucins, but also charged amino acids, we wanted to ensure that hydrophobic domains are the major contributors to the adsorption of mucins on PDMS. The approach we used to address that was based on comparing the adsorption of positively and negatively charged dextran variants with unmodified dextrans. As expected, both dextrans modified with positive diethylaminoethyl (DEAE) groups or with negative carboxymethyl (CM) groups, showed low adsorption on PDMS, similarly to unmodified dextrans (Paper I, Figure 4d). As for the unmodified dextrans, the poor adsorption led in turn to poor lubricity, which was similar to that of simple buffer (Paper I, Figure 5b).

24

After demonstrating the importance of hydrophobic groups for the adhesion and lubricity of macromolecules such as mucins and dextrans on PDMS surfaces, the last step was attempting to “rescue” the trypsin treated mucins and investigating if it would be possible for them to reach their initial native lubricity level. Phenyl groups were grafted onto EDC/NHS activated carboxyl groups found in sialic acid residues of the MUC5B sialic acid-rich mucin (Paper I, Figure 6a). The phenylated mucins showed increased adsorption on PDMS (Paper I, Figure 6c), as well as enhanced lubricity, with a friction coefficient lowered by approximately 3 times (Paper I, Figure 6d). However, the “rescued” mucins did not show total recovery, with friction coefficients still over an order of magnitude higher than untreated mucins. This could be attributed to the limited amount of phenyl groups that could get grafted onto the sialic acid carboxyl groups, but also to differences in the configuration of the adsorbed mucins, either bound to the surface via their terminal protein domains, or via the phenyl group along the mucin molecule. In conclusion, our findings indicate that the hydrophobic terminal domains of mucins are crucial for the mucin adsorption on PDMS as well as lubricity, and that mucins which have lost these domains and thus their lubricity can be partly “rescued” by grafting hydrophobic groups on the molecules. These results, together with the finding that grafting hydrophobic groups on otherwise poor lubricants -such as dextrans- can greatly enhance their lubricity, might help to pave the way towards the development of high performance lubricants for biomedical or other industrial surface applications. With our results highlighting the importance of adsorption in lubrication, we could utilize this knowledge to engineer polymer-based lubricants that exhibit enhanced surface adsorption. As it has been mentioned before, a good lubricant is characterized by good hydration and most importantly good adsorption on the surface. Therefore, a very well-hydrated molecule that cannot anchor itself on the surface, cannot lubricate it either. Such example of hydrated polymers that do not provide satisfactory lubrication are artificial tears and saliva products, which do not anchor to the surface, leading to poor lubrication and requiring frequent re-application. There have also been attempts for the commercial use of mucins for lubrication; “Saliva Orthana”, a PGM-based oral spray, has been tested on xerostomic patients, but did not show better results compared to placebo (Sweeney et al. 1997). This is probably due to the harsh purification processes such as protease digestion, which cleaves off the mucins’ protein terminals, and again affects their interaction with the mucosal surfaces, as well as the way that mucins assemble into gel networks. In order to develop high performance lubricants for mucosal surfaces, we could therefore try to mimic native mucins and the way they interact with the mucosa, for example as we did by grafting hydrophobic groups on digested mucins or dextrans. Regarding the lubrication of mucosal surfaces, although the information we gained is valuable for the adsorption and lubrication of mucins on hydrophobic surfaces, our approach also has limitations. This is due to the fact that we have focused our study on one particular aspect of mucin-surface interactions, which relates to hydrophobic interactions with surfaces and the resulting lubrication of such surfaces. It is then clear that with this setup we are not claiming to correctly mimic the mucus layer covering the epithelia, where mucin adsorption is governed by multiple interactions and there are more mucins than a

25

monolayer, but we are exploring part of this vast interaction repertoire which are hydrophobic interactions. Mucins and mucin-inspired molecules are also relevant as potential lubricants on body surfaces where mucins do not occur naturally, but lubrication is still desired. Such is the case of the joints, where the synovial fluid that normally lubricates them might not provide efficient lubrication due to pathologies, injuries or surgery. So far, hyaluronic acid injections have been shown to provide relief to some osteoarthritis patients (Bowman et al. 2018), but mucins’ exceptional lubricity could also prove useful. Other surfaces, for which high performance lubrication is relevant, include tracheostomy tubes and catheters, which need lubrication for a more comfortable insertion and removal. We hypothesize that the coating or grafting of mucins on such surfaces could provide such lubrication. In summary, the concept of functionalizing a hydrated molecule so that it absorbs on specific surfaces, could be utilized for the development of high performance boundary lubricants for any type of surface, biological or not, biomedically relevant or not.

Figure 8. Graphical abstract for Paper I. Mucin lubrication of hydrophobic surfaces depends on mucin adsorption, which in turn is driven by hydrophobic interactions between the surface and the mucin terminal peptide domains. Indeed, the enzymatic treatment of mucins resulted in the cleavage of these domains and the subsequent inability of mucins to adsorb on and lubricate the hydrophobic surfaces. However, by grafting hydrophobic phenyl groups on the mucin glycans, we managed to restore the ability of mucins to adsorb on hydrophobic surfaces and lubricate. Reprinted with permission from (Käsdorf et al. 2017). Copyright 2017 American Chemical Society.

2.3.2. Genetically engineered mucoadhesive spider silk (Paper II)

Our mucosa provides a vast surface area of mucus-covered epithelial membranes which present us with great opportunities for biomedical applications, but also challenges. For example, the increased blood perfusion of the mucosa makes it an ideal site for delivering drugs straight into the blood flow, while bypassing the extensive hepatic metabolism that orally administered drugs need to go through. On the other hand, the different pathologies that can affect the mucosa, such as mucosal wounds, need efficient solutions to be addressed, but are challenging to face because of the high hydration and dynamic mucus turnover of the mucosal surfaces. In order to approach the above, the

26

development of polymers which are optimized to interact with mucosal surfaces is essential. Such polymers are called mucoadhesive polymers and can be for example the building blocks of transmucosal drug delivery systems and mucosal dressings. Mucoadhesive polymers can adhere onto mucous membranes through various types of interactions, spanning from non-specific electrostatic interactions to specific binding with mucins, the main macromolecular components of mucus. These polymers can be assembled into different materials, such as gel slabs, patches and capsules. Although there are well-established mucoadhesive polymers that showcase strong mucoadhesive properties, with the most studied one being chitosan, these polymers merely rely on their natural mucoadhesive properties, without providing any additional functionality. Therefore, in order to address the need for multifunctional mucoadhesive materials, we functionalized the partial spider silk protein 4RepCT (Paper II, Figure 1A) with mucoadhesive properties by using genetic engineering, and prepared different materials from the new protein variants. These new silk variants combined with other biofunctional silk variants could constitute building blocks for multifunctional mucoadhesive materials. One variant, pLys-4RepCT, was functionalized with six positively charged lysines, aiming for non-specific electrostatic interactions with the negatively charged mucins (Paper II, Figure 1B). The other variant, hGal3-4RepCT, was functionalized with the carbohydrate recognition domain (CRD) of the human galectin-3 (hGal-3), which specifically binds the glycans Galβ1-3GlcNAc and Galβ1-4GlcNAc that are found on mucins (Paper II, Figure 1C). This specific binding approach reflects the recognized potential of using galectins to promote localized binding to specific matrices (Farhadi et al. 2018). The asset of using galectins for such approaches is the fact that they target glycans, which can be found everywhere in the body and at the same time are characterized by vast diversity. In the case of mucosal surfaces, the mucin glycans may vary according to tissue, as well according to disease state. Therefore, mucoadhesive materials functionalized with galectins which bind to specific glycans, could be a great way to specifically target healthy and/or diseased tissues. For example, the oral delivery of mucoadhesive capsules functionalized with galectins which bind the glycans of intestinal mucins (MUC2) could be used for targeted delivery in the gut. The recombinant silk proteins pLys-4RepCT and hGal3-4RepCT were successfully expressed in E. coli and purified using immobilized metal affinity chromatography (IMAC) (Paper II, Figure 1D). Then, using QCM-D on gold-coated quartz crystals, we observed that when in their soluble form, the newly produced recombinant proteins spontaneously form rigid coatings on gold and are stable over time (Paper II, Figure 2C,E). The above observation indicates that the functionalized proteins maintain the auto-assembling properties of their non-functionalized counterpart 4RepCT (Paper II, Figure 2A). More specifically, 4RepCT has been reported to form coatings on a variety of surfaces such as gold, titanium, stainless steel, hydroxyapatite and polystyrene (Nilebäck, Hedin, et al. 2017), presenting the same adsorption profile with an initial rapid adsorption and then a slower but continuous silk accumulation. Some of these surfaces are biomedically relevant as they are widely used in implant and dental technology; such examples are titanium, hydroxyapatite and stainless steel. Thus, the use

27

of silk coatings containing antimicrobial or anti-inflammatory properties could help prevent infections and immune-mediated rejection of such implants. Similarly, we could potentially confer mucoadhesive properties to virtually any materials by coating them with the mucoadhesive silk proteins we have developed. In order to further characterize the coatings of our newly produced proteins, we also used atomic force microscopy (AFM) to visualize them; pLys-4RepCT formed fibrils similar to 4RepCT (Paper II, Figure 2B,D), while for hGal3-4RepCT no fibrils were observed, but a rather homogeneous silk layer (Paper II, Figure 1F). This difference in the structure of hGal3-4RepCT coatings might be due the large size of the CRD, which could be interfering with the self-assembly of the silk protein. After successfully coating the gold surfaces with silk, the mucoadhesive properties of the coatings were measured by the real time monitoring of the mucin adsorption on the silk coatings, as well as the stability of the adsorption, using QCM-D. The mucins used were porcine gastric mucins (PGM) purified in-house from pig stomachs, and commercial bovine submaxillary mucins (BSM) from the salivary glands of cows. While none of the mucins adsorbed on 4RepCT coatings (Paper II, Figure 3A,D), both BSM and PGM adsorbed on the pLys-4REpCT and hGal3-4RepCT coatings, and remained adsorbed after the subsequent washes. Specifically, hGal3-4RepCT demonstrated the highest frequency responses when exposed to mucins (Paper II, Figure 3B,C,E,F). An interesting observation was the variations we saw between experiments for the mucin adsorption, which were likely due to differences in the underlying silk coatings. The potential cause of these variations in silk coatings is batch-to-batch variation, which is eliminated intra-experimentally. Finally, the dissipation/frequency ratios suggested that BSM molecules adopt a more extended conformation when adsorbing on pLys-4RepCT, while PGM adsorbed in a similar way on both pLys-4RepCT and hGal3-4RepCT coatings (Paper II, SI Figure 4). Although the mucoadhesive properties of the new silk variant coatings were demonstrated, we wanted to confirm the mechanisms through which mucins adsorbed on the silk coatings. As it has already been mentioned, we based the design of the new silk variants on the hypotheses that there would be electrostatic interactions between the mucins and the lysines of pLys-4RepCT, and specific interactions between the mucin glycans and the carbohydrate recognition domain (CRD) of hGal3-4RepCT. Thus, in order to investigate the role of electrostatic interactions in the adsorption of mucins on pLys-4RepCT coatings, we repeated the PGM adsorption in a higher NaCl concentration (150 mM instead of 10 mM) and observed a reduced mucin adsorption, compared to the low NaCl condition (Paper II, Figure 4). These results may be explained by the screening of the electric charges between the mucins and silk, and suggest that the charge of pLys-4RepCT is indeed essential for the mucins to bind on it. However, other non-specific interactions are possible and cannot be excluded. As it has been discussed in the introduction, mucoadhesion is a complex process which is simultaneously governed by different interactions between material and the mucosa. For instance, in the case of the positively charged chitosan, it is thought to mainly interact with mucins through electrostatic interactions, but urea and ethanol also had an effect on their interaction, suggesting that other interactions can be influencing the adsorption too.

28

Moreover, in order to investigate if the adsorption of mucins on hGal3-4RepCT coatings is indeed due to the binding of mucin glycans with the CRD of the human galectin-3, we tested the functionality of the CRD. That was done by monitoring the adsorption of asialofetuin, a well-established ligand and inhibitor of the human galectin-3, on hGal3-4RepCT coatings and by comparing it with the adsorption of asialofetuin on 4RepCT coatings. The QCM-D results presented a high response for asialofetuin on hGal3-4RepCT coatings, while the corresponding response for 4RepCT coatings was limited to less than half, attributed to non-specific binding (Paper II, Figure 5). Application-wise, mucoadhesive silk coatings are relevant for the coating of other materials, as they could provide them with new functionalities. However, the potential of creating bigger silk materials from our new silk variants enables them to stand alone and thus provides more opportunities in the mucosal treatment field. Consequently, we attempted to prepare a variety of silk materials based on the new mucoadhesive variants. It has been previously demonstrated that 4RepCT can self-assemble into silk fibers at the air-water interface. Thus, we first investigated if pLys-4RepCT and hGal3-4RepCT could maintain that property. Both new silk variants successfully formed silk fibers, with pLys-4RepCT demonstrating an elongated and thin structure very similar to 4RepCT fibers, while hGal3-4RepCT fibers had a less distinct structure resembling cotton (Paper II, Figure 6A-C). It is interesting to note that hGal3-4RepCT fibers took significantly longer time in order to form compared to 4RepCT and hGal3-4RepCT. As it has been brought up previously regarding the microscopic structure of hGal3-4RepCT coatings, the difference in the structure and formation of hGal3-4RepCT fibers could be explained due to the big size of the domain of the human galectin-3. In fact, delayed fiber formation has been previously observed for 4RepCT functionalized with a xylanase domain. The mucoadhesive properties of the fibers were then assessed by the binding of fluorescently labeled mucins on the fibers, followed by fluorescence microscopy and quantification. The fibers that showed the best mucin binding properties for both BSM and PGM were made of hGal3-4RepCT, while pLys-4RepCT fibers showed medium binding of PGM and very low, comparable to the unspecific 4RepCT binding of BSM (Paper II, Figure 6D-K). Overall, our results showed that the mucoadhesive properties of the silk proteins were not altered by the fiber formation. The last step of the study was to prepare bigger materials from the new silk variants, which could physically represent mucosal dressings. For this last step, we combined both mucoadhesive silk variants in order to make use of both mucoadhesion mechanisms. First, we prepared silk fiber meshes; after assembling fiber pieces of both pLys-4RepCT and hGal3-4RepCT into a mesh-like formation, we dried them and saw that they maintained their structure before and after rehydration (Paper II, Figure 7A,B). Then, we incubated the rehydrated fiber meshes with the fluorescently labeled mucins and observed that even after drying they retained their mucin binding properties as seen before (Paper II, Figure 6C-H). Finally, we fabricated mixed pLys-4RepCT and hGal3-4RepCT foams by frothing the soluble proteins (Paper II, Figure 8A,B). Again, the new silk variant foams demonstrated an increased binding for both fluorescently labeled mucin types but especially PGM, compared to control foams (Paper II, Figure 68C-H). These material forms are very interesting for wound healing purposes because apart from providing a large surface area for mucoadhesion, they could also be used as tissue engineering scaffolds offering either a flat or a porous surface for the cells to grow. In

29

fact, 4RepCT meshes and foams have been previously shown to support very well cultures of human fibroblasts (Widhe et al. 2010). Moreover, by mixing the mucoadhesive silk variants with other biofunctional silk variants, such as 4RepCT functionalized with antimicrobial or cell adhesion promoting motifs, we could easily prepare materials that could potentially stimulate wound healing and protect the wound from infections. In conclusion, our mucoadhesive silk protein variants can form coatings, fibers, meshes and foams and thus constitute useful building blocks for new materials aimed for mucosal treatments. Reflecting on the choice of recombinant spider silk for building mucoadhesive materials, silk is well known for its mechanical properties, being impressively strong and lightweight. Most importantly, silk-based materials are appropriate for biomedical applications because they are biocompatible, biodegradable and non-toxic. A great advantage of using recombinant spider silk proteins instead of other silk proteins is that they can be recombinantly produced and self-assembled in stable structures using mild conditions throughout the process; we can thus functionalize them by adding different bioactive domains, and without interfering with the bioactivity of the domains. In our case, the above asset is crucial for hGal3-4RepCT, where we want the carbohydrate recognition domain to stay functional. The biggest advantage of using recombinant silk proteins in mucoadhesive materials is their multifunctionality. In fact, different silk protein variants can be easily mixed and assemble into materials with various functionalities. For example, our mucoadhesive silk variants could be combined with previously developed silk variants which have antimicrobial properties or promote epithelialization in order to make wound dressings that could promote wound healing. Although the use of recombinant silk proteins in mucoadhesive materials seems promising, there are also limitations, as well as room for improvement. For example, the fact that mucins adsorb on our materials does not guarantee mucoadhesion in a real setting. The interactions of mucins with the materials are encouraging, but given that mucoadhesion is a complicated process with multiple stages, the materials should be tested on real mucosa. Moreover, when working with natural polymers, batch to batch variations that could affect the reproducibility are expected, and for recombinant silk proteins we also cannot have total control of the material self-assembly. Finally, the production cost of recombinant silk proteins is rather high, especially if we think of the relatively low yields. However, that could be addressed by combining our silk proteins with cheaper materials to reduce the cost; such examples that have been already combined with other recombinant silk variants are silkworm fibroin (Nilebäck, Chouhan, et al. 2017) and cellulose nanofibrils (Mittal et al. 2017). To summarize, we here present two novel mucoadhesive silk protein variants, which alone or combined with other biofunctional proteins, could constitute promising building blocks for multifunctional silk-based materials aimed for mucosal treatments.

30

Figure 8. Graphical abstract for Paper II. In order to genetically engineer mucoadhesive spider silk, the partial spider silk protein 4RepCT was functionalized with; six positively charged lysines, aiming for non-specific electrostatic interactions with the negatively charged mucins (pLys-4RepCT), or the carbohydrate recognition domain of the human galectin-3 (hGal3-4RepCT) aiming for specific interactions with particular mucin glycans. From the newly produced silk proteins we made coatings, fibers, meshes and foams (from top to bottom, and from left to right). Reprinted with permission from (Jansson, Högqvist, and Hedhammar 2018). Copyright 2017 American Chemical Society.

2.4. Concluding remarks and future perspectives The work presented in this thesis explored the interactions of mucins with different material surfaces, as well as the potentials to use the gained knowledge for biomedical applications. The first part of the thesis is based on studies of the interactions of mucins with hydrophobic surfaces, while the second part presented investigations on the interactions of mucins with novel silk materials engineered to be mucoadhesive. In the first part of the thesis, we studied the mechanisms of mucin adsorption and lubrication on hydrophobic surfaces, in order to identify the key features that make mucins such outstanding boundary lubricants. Our results showed that the hydrophobic terminal domains of mucins play a paramount role in the adsorption of mucins on hydrophobic surfaces, and consequently their lubricating properties. Moreover, by grafting hydrophobic phenyl groups on damaged mucins or dextrans, which are hydrated molecules but otherwise poor lubricants, we achieved to make the molecules adsorb on the hydrophobic surfaces and lubricate them effectively. Overall, these results helped us conclude that what makes a good boundary lubricant are not only good hydration properties, but also its good adsorption on the surface. Therefore, our findings can be particularly helpful in the development of high performance lubricants which would be characterized by improved interactions with the surfaces, for biomedical or other applications.

31

This thesis informs on potential next steps towards the development of such lubricants. First, it would be necessary to find suitable molecules to apply the grafting strategy on. A good candidate is polyethylene glycol (PEG), a highly hydrated molecule which already has good lubricating properties and is non-toxic, water-soluble and available in various molecular weights (Spencer 2014). Interestingly, linking PEG with mucins has been previously reported to restore mucin hydration (Thomas Crouzier et al. 2015). Finally, another essential step for testing the performance of the lubricant would be to test it on actual tissues in vitro or in vivo. These tissues could be mucosal tissues, but also other tissues where lubrication is relevant, such as joints. After investigating the intrinsic interactions of mucins with hydrophobic surfaces, in the second part of the thesis we developed novel mucoadhesive silk materials which are optimized for mucin interaction by genetic engineering. We thus functionalized the partial spider silk protein 4RepCT with either six positively charged lysine residues (pLys-4RepCT), or the Human Galectin-3 Carbohydrate Recognition Domain (hGal3-4RepCT). These two strategies were aiming to either non-specific electrostatic interactions between the positively charged lysines and the negatively charged mucins, or specific binding between the galectin’s carbohydrate recognition domain and the specific mucin glycans, respectively. A variety of materials was then successfully prepared from the newly produced soluble silk proteins, including coatings, fibers, meshes and foams. Finally, the mucoadhesive properties of the materials were verified by observing the enhanced adsorption of mucins on them. We thus concluded that our novel mucoadhesive silk variants, alone or combined with other biofunctional proteins, constitute promising building blocks for multifunctional materials used for mucosal treatments, such as mucosal wound dressings and transmucosal drug delivery systems. Regarding next steps in this work, it would be necessary to first test the mucoadhesive properties of the materials on animal mucosal surfaces in vitro and in vivo. A good way to approach that in vitro is by using a texture analyzer or atomic force microscopy (AFM), in order to measure the force needed to break the adhesion. Moreover, the aspect of combining the mucoadhesive silk variants with other biofunctional silk variants or cheaper materials should be further investigated. Specifically, the bioactivity of the mixed materials should be tested in vitro and in vivo, as well as the possibility to be used as mucosal tissue engineering matrices. In summary, the work presented in this thesis suggests that mucin-material interactions can provide us with valuable information for the development of a variety of biomaterials, in our case lubricants and mucoadhesive materials. Specifically, our study on mucin adsorption and lubricity could set the scene for developing high performance mucin-based and mucin-inspired lubricants that would provide desired lubrication to a wide range of biological surfaces and more. Finally, our new mucoadhesive silk proteins could potentially be the basis for the development of multifunctional mucoadhesive materials that could treat mucosal pathologies.

32

3. Acknowledgements First I would like to thank my supervisor Thomas Crouzier for assisting me during these two years. What a ride it has been! I learnt a lot and grew even more during this time, so thank you for that. I will be a bit more excited than normal when I hear the word mucus, and that’s a great souvenir I am getting from the lab. Then I would like to thank My Hedhammar for the wonderful collaboration. It has been a blast working with you and thank you for welcoming me in the spider silk family. Special thanks to Ronnie Jansson for his always positive attitude and for patiently introducing me in the spider silk world and always being there to help and make myself feel at home at plan 3. I want to also thank Mark Högqvist for starting the silk project, it would not be possible without your kick-start. Thank you also to Johan Erlandsson for showing me around in Fiber and Polymer Technology and for the help. I also want to thank all the rest of the co-authors of our papers; thank you for the wonderful work and the smooth collaboration! Of course thanks to all the funding sources that made these projects happen; FORMAS, Vetenskapsrådet, Swedish Foundation for Strategic Research. I want to also thank Kristina Divne for her help with the Licentiate procedure and for doing her best for the doctoral programme. Thank you to Kristina Jansson too for her administrative help with courses and different forms; you are great and I always enjoyed my visits to your office! Also, thank you Francisco Vilaplana for playing the eISP game with me for two whole years! Thank you also to Stefan Gaunitz for correcting my Swedish abstract. I want to also thank the Crouzier group members, current and older ones, good luck to you! Of course, thank you to all the Glycoscience and plan 2 people for the lovely co-habitation! If you just said hi to me, gave me a smile, helped me use an instrument or put the dishwasher on for the first time, I am very grateful! It is always the small acts of kindness that matter. Then I really want to thank the family I made in Plan 2. Because, as cliché as it sounds, friends are the family we choose. You all made my stay better.

33

Amparo you are the real star here, thank you so much for the support and love and for always being there for me! You deserve the whole world! Jonatan, thanks for the great company and for always bringing cookies! I will truly miss our meriendas with Amparo and Mamen. Kun, Secil, Reskandi, you guys are great and I wish you the best of luck with the rest of your studies. And finally there is the other part of my Stockholm family. Gonzalo, thank you so much for always being there and for the never-ending memes! I wouldn’t make it without my daily Charlie! Ioanna, thanks for all the Zumba and eating sessions, and for being your awesome self and sharing that with me! Haizea, you might have left our Stockholm gang but you will always be my pitxitxi! David, thank you for the dinners, even if it always took us 1 month to plan! Devy, thanks for the lunch-breaks and for fixing my abstract! Sara, thank you for re-fixing my abstract! Then I want to thank my girl squad from Greece (and Sweden); Maria (you beautiful, rule-breaking moth), Elena, Sonia, Katerina, Alba. Distance ain’t got nothing on us! Thank you for sticking around. Last but not least, comes my beautiful family. I might have not chosen you but I would always choose you before any other shiny family. The biggest thank you to my mom and dad for always being supportive to whatever crazy thing I wanted to do. Μαμά, μπαμπά, το μεγαλύτερο ευχαριστώ για την υποστήριξη οοοοόλα τα χρόνια! All the love to my sister, brother in law, little niece and brand new nephew for being the absolute greatest! Σας αγαπώ! I think that I should be forgetting people, because I honestly believe you can never be grateful enough. To you that made even the slightest impact on my life or work or just read this thesis, thank you!

34

4. References Allen, A., J. P. Pearson, D. A. Hutton, A. H. Mall, R. M. Coan, and L. A. Sellers. 1989.

“The Dependence of Size and Functional Properties of Pig Gastric Mucus on the Isolation Method Used.” Symposia of the Society for Experimental Biology 43: 241–48.

Arike, Liisa, Jessica Holmén-Larsson, and Gunnar C. Hansson. 2017. “Intestinal Muc2 Mucin O-Glycosylation Is Affected by Microbiota and Regulated by Differential Expression of Glycosyltranferases.” Glycobiology 27 (4): 318–28.

Bansil, Rama, Jonathan P. Celli, Joseph M. Hardcastle, and Bradley S. Turner. 2013. “The Influence of Mucus Microstructure and Rheology in Helicobacter Pylori Infection.” Frontiers in Immunology 4 (October): 310.

Bansil, Rama, and Bradley S. Turner. 2006. “Mucin Structure, Aggregation, Physiological Functions and Biomedical Applications.” Current Opinion in Colloid & Interface Science 11 (2): 164–70.

Berg, I. Cecilia Hahn, Ulla M. Elofsson, Andrew Joiner, Martin Malmsten, and Thomas Arnebrant. 2001. “Salivary Protein Adsorption onto Hydroxyapatite and Sds‐mediated Elution Studied by in Situ Ellipsometry.” Biofouling 17 (3): 173–87.

Bernkop-Schnürch, Andreas. 2005. “Thiomers: A New Generation of Mucoadhesive Polymers.” Advanced Drug Delivery Reviews 57 (11): 1569–82.

Bhavanandan, V. P., and J. D. Hegarty. 1987. “Identification of the Mucin Core Protein by Cell-Free Translation of Messenger RNA from Bovine Submaxillary Glands.” The Journal of Biological Chemistry 262 (12): 5913–17.

Birchenough, G. M. H., M. E. V. Johansson, J. K. Gustafsson, J. H. Bergström, and G. C. Hansson. 2015. “New Developments in Goblet Cell Mucus Secretion and Function.” Mucosal Immunology 8 (4): 712–19.

Blake, John. 1973. “Mucus Flows.” Mathematical Biosciences 17 (3): 301–13. Błochowiak, Katarzyna, Anna Olewicz-Gawlik, Adriana Polańska, Michalina Nowak-

Gabryel, Jarosław Kocięcki, Henryk Witmanowski, and Jerzy Sokalski. 2016. “Oral Mucosal Manifestations in Primary and Secondary Sjögren Syndrome and Dry Mouth Syndrome.” Postepy Dermatologii I Alergologii 33 (1): 23–27.

Boettcher, Kathrin, Benjamin Winkeljann, Tannin A. Schmidt, and Oliver Lieleg. 2017. “Quantification of Cartilage Wear Morphologies in Unidirectional Sliding Experiments: Influence of Different Macromolecular Lubricants.” Biotribology 12 (December): 43–51.

Bowman, Steven, Mohamed E. Awad, Mark W. Hamrick, Monte Hunter, and Sadanand Fulzele. 2018. “Recent Advances in Hyaluronic Acid Based Therapy for Osteoarthritis.” Clinical and Translational Medicine 7 (1): 6.

Bravo-Osuna, Irene, Christine Vauthier, Alessandra Farabollini, Giovanni Filippo Palmieri, and Gilles Ponchel. 2007. “Mucoadhesion Mechanism of Chitosan and Thiolated Chitosan-Poly(isobutyl Cyanoacrylate) Core-Shell Nanoparticles.” Biomaterials. https://doi.org/10.1016/j.biomaterials.2007.01.005.

Burney, Tabinda J., and Jane C. Davies. 2012. “Gene Therapy for the Treatment of Cystic Fibrosis.” The Application of Clinical Genetics 5 (May): 29–36.

Celli, Jonathan P., Bradley S. Turner, Nezam H. Afdhal, Randy H. Ewoldt, Gareth H. McKinley, Rama Bansil, and Shyamsunder Erramilli. 2007. “Rheology of Gastric Mucin Exhibits a pH-Dependent Sol−Gel Transition.” Biomacromolecules 8 (5): 1580–86.

Chhabra, Rajendra P. 2010. “Non-Newtonian Fluids: An Introduction.” In Rheology of Complex Fluids, edited by J. Murali Krishnan, Abhijit P. Deshpande, and P. B. Sunil Kumar, 3–34. New York, NY: Springer New York.

Choy, Young Bin, Jung-Hwan Park, and Mark R. Prausnitz. 2008. “Mucoadhesive Microparticles Engineered for Ophthalmic Drug Delivery.” The Journal of Physics and Chemistry of Solids 69 (5): 1533–36.

35

Coles, Jeffrey M., Debby P. Chang, and Stefan Zauscher. 2010. “Molecular Mechanisms of Aqueous Boundary Lubrication by Mucinous Glycoproteins.” Current Opinion in Colloid & Interface Science 15 (6): 406–16.

Collins, Paul, and Jonathan Rhodes. 2006. “Ulcerative Colitis: Diagnosis and Management.” BMJ 333 (7563): 340–43.

Cook, Sarah L., Stephanie P. Bull, Lisa Methven, Jane K. Parker, and Vitaliy V. Khutoryanskiy. 2017. “Mucoadhesion: A Food Perspective.” Food Hydrocolloids 72 (November): 281–96.

Cooper, Matthew A. 2002. “Optical Biosensors in Drug Discovery.” Nature Reviews. Drug Discovery 1 (7): 515–28.

Corfield, Anthony P. 2000. Glycoprotein Methods and Protocols: The Mucins. Springer Science & Business Media.

Crouzier, Thomas, Kathrin Boettcher, Anthony R. Geonnotti, Nicole L. Kavanaugh, Julie B. Hirsch, Katharina Ribbeck, and Oliver Lieleg. 2015. “Modulating Mucin Hydration and Lubrication by Deglycosylation and Polyethylene Glycol Binding.” Advanced Materials Interfaces 2 (18): 1500308.

Crouzier, T., H. Jang, J. Ahn, R. Stocker, and K. Ribbeck. 2013. “Cell Patterning with Mucin Biopolymers.” Biomacromolecules 14 (9): 3010–16.

Decramer, M., W. Janssens, and M. Miravitlles. 2012. “Chronic Obstructive Pulmonary Disease. Lancet [Internet].” Elsevier Ltd.

Derjaguin, B. V., I. N. Aleinikova, and Yu P. Toporov. 1994. “On the Role of Electrostatic Forces in the Adhesion of Polymer Particles to Solid Surfaces.” Progress in Surface Science. https://doi.org/10.1016/0079-6816(94)90042-6.

Dixon, Matthew C. 2008. “Quartz Crystal Microbalance with Dissipation Monitoring: Enabling Real-Time Characterization of Biological Materials and Their Interactions.” Journal of Biomolecular Techniques: JBT 19 (3): 151.

Engeland, C. G., P. T. Marucha, and M. P. Manos. 2013. “131. Oral Mucosal Wound Healing: Sex Differences and Inflammation.” Brain, Behavior, and Immunity 32 (September): e38.

Farhadi, Shaheen A., Evelyn Bracho-Sanchez, Margaret M. Fettis, Dillon T. Seroski, Sabrina L. Freeman, Antonietta Restuccia, Benjamin G. Keselowsky, and Gregory A. Hudalla. 2018. “Locally Anchoring Enzymes to Tissues via Extracellular Glycan Recognition.” Nature Communications. https://doi.org/10.1038/s41467-018-07129-6.

Feiler, Adam A., Anna Sahlholm, Tomas Sandberg, and Karin D. Caldwell. 2007. “Adsorption and Viscoelastic Properties of Fractionated Mucin (BSM) and Bovine Serum Albumin (BSA) Studied with Quartz Crystal Microbalance (QCM-D).” Journal of Colloid and Interface Science 315 (2): 475–81.

Fischer, Anthony J., Maria I. Pino-Argumedo, Brieanna M. Hilkin, Cullen R. Shanrock, Nicholas D. Gansemer, Anna L. Chaly, Keyan Zarei, et al. 2019. “Mucus Strands from Submucosal Glands Initiate Mucociliary Transport of Large Particles.” JCI Insight 4 (1). https://doi.org/10.1172/jci.insight.124863.

Florey, Howard Walter. 1955. “The Croonian Lecture: Mucin and the Protection of the Body.” Proceedings of the Royal Society of London. Series B - Biological Sciences, January. http://rspb.royalsocietypublishing.org/content/143/911/147.short.

Fuongfuchat, A., A. M. Jamieson, J. Blackwell, and T. A. Gerken. 1996. “Rheological Studies of the Interaction of Mucins with Alginate and Polyacrylate.” Carbohydrate Research 284 (1): 85–99.

Gregory, R. J., S. H. Cheng, D. P. Rich, J. Marshall, S. Paul, K. Hehir, L. Ostedgaard, K. W. Klinger, M. J. Welsh, and A. E. Smith. 1990. “Expression and Characterization of the Cystic Fibrosis Transmembrane Conductance Regulator.” Nature 347 (6291): 382–86.

Gupta, Vivek, Byeong Hee Hwang, Joohee Lee, Aaron C. Anselmo, Nishit Doshi, and Samir Mitragotri. 2013. “Mucoadhesive Intestinal Devices for Oral Delivery of Salmon Calcitonin.” Journal of Controlled Release: Official Journal of the

36

Controlled Release Society 172 (3): 753–62. Hägerström, H., and K. Edsman. 2001. “Interpretation of Mucoadhesive Properties of

Polymer Gel Preparations Using a Tensile Strength Method.” The Journal of Pharmacy and Pharmacology 53 (12): 1589–99.

Handbook of Mucosal Immunology. 2012. Academic Press. Henderson, Ashley G., Camille Ehre, Brian Button, Lubna H. Abdullah, Li-Heng Cai,

Margaret W. Leigh, Genevieve C. DeMaria, et al. 2014. “Cystic Fibrosis Airway Secretions Exhibit Mucin Hyperconcentration and Increased Osmotic Pressure.” The Journal of Clinical Investigation 124 (7): 3047–60.

Henneicke-von Zepelin, Hans-Heinrich, Ragna Williams, Wiebke Havemeister, Walter Wigger-Alberti, and Klaus-Ulrich Nolte. 2017. “Clinical Trial Shows Lasting Function of a New Moisturizing Cream against Vaginal Dryness.” Wiener Medizinische Wochenschrift 167 (7-8): 189–95.

Hill, David B., Robert F. Long, William J. Kissner, Eyad Atieh, Ian C. Garbarine, Matthew R. Markovetz, Nicholas C. Fontana, et al. 2018. “Pathological Mucus and Impaired Mucus Clearance in Cystic Fibrosis Patients Result from Increased Concentration, Not Altered pH.” The European Respiratory Journal: Official Journal of the European Society for Clinical Respiratory Physiology 52 (6). https://doi.org/10.1183/13993003.01297-2018.

Jabbari, Esmaiel, Natalie Wisniewski, and Nikolaos A. Peppas. 1993. “Evidence of Mucoadhesion by Chain Interpenetration at a Poly (acrylic Acid)/mucin Interface Using ATR-FTIR Spectroscopy.” Journal of Controlled Release: Official Journal of the Controlled Release Society 26 (2): 99–108.

Jansson, R., M. Högqvist, and M. Hedhammar. 2018. “Engineering Mucoadhesive Silk.” Abstract of Papers of. http://www.diva-portal.org/smash/record.jsf?pid=diva2:1277709.

Johansson, Malin E. V., Jessica M. Holmén Larsson, and Gunnar C. Hansson. 2011. “The Two Mucus Layers of Colon Are Organized by the MUC2 Mucin, Whereas the Outer Layer Is a Legislator of Host–microbial Interactions.” Proceedings of the National Academy of Sciences of the United States of America 108 (Supplement 1): 4659–65.

Juge, Nathalie. 2012. “Microbial Adhesins to Gastrointestinal Mucus.” Trends in Microbiology 20 (1): 30–39.

Karlsson, N. G., H. Nordman, H. Karlsson, I. Carlstedt, and G. C. Hansson. 1997. “Glycosylation Differences between Pig Gastric Mucin Populations: A Comparative Study of the Neutral Oligosaccharides Using Mass Spectrometry.” Biochemical Journal 326 ( Pt 3) (September): 911–17.

Käsdorf, Benjamin T., Florian Weber, Georgia Petrou, Vaibhav Srivastava, Thomas Crouzier, and Oliver Lieleg. 2017. “Mucin-Inspired Lubrication on Hydrophobic Surfaces.” Biomacromolecules 18 (8): 2454–62.

Kawashima, Y., H. Yamamoto, H. Takeuchi, and Y. Kuno. 2000. “Mucoadhesive DL-Lactide/glycolide Copolymer Nanospheres Coated with Chitosan to Improve Oral Delivery of Elcatonin.” Pharmaceutical Development and Technology 5 (1): 77–85.

Kesimer, Mehmet, Amina A. Ford, Agathe Ceppe, Giorgia Radicioni, Rui Cao, C. William Davis, Claire M. Doerschuk, et al. 2017. “Airway Mucin Concentration as a Marker of Chronic Bronchitis.” The New England Journal of Medicine 377 (10): 911–22.

Knowles, Michael R., and Richard C. Boucher. 2002. “Mucus Clearance as a Primary Innate Defense Mechanism for Mammalian Airways.” The Journal of Clinical Investigation 109 (5): 571–77.

Kreyling, Wolfgang G., Manuela Semmler, and Winfried Möller. 2004. “Dosimetry and Toxicology of Ultrafine Particles.” Journal of Aerosol Medicine: The Official Journal of the International Society for Aerosols in Medicine 17 (2): 140–52.

Kuipers, E. J., J. C. Thijs, and H. P. Festen. 1995. “The Prevalence of Helicobacter Pylori in Peptic Ulcer Disease.” Alimentary Pharmacology & Therapeutics 9 Suppl 2: 59–69.

Lai, Samuel K., Ying-Ying Wang, Denis Wirtz, and Justin Hanes. 2009. “Micro- and

37

Macrorheology of Mucus.” Advanced Drug Delivery Reviews 61 (2): 86–100. Lalla, Rajesh V., Stephen T. Sonis, and Douglas E. Peterson. 2008. “Management of Oral

Mucositis in Patients Who Have Cancer.” Dental Clinics of North America 52 (1): 61–77.

Lamblin, Genevieve, Michel Lhermitte, André Klein, Nicole Houdret, Andree Scharfman, Reuben Ramphal, and Philippe Roussel. 1991. “The Carbohydrate Diversity of Human Respiratory Mucins: A Protection of the Underlying Mucosa.” The American Review of Respiratory Disease 144 (3 Pt 2): S19–24.

Larsson, Jessica M. Holmn, Hasse Karlsson, Jessica Gråberg Crespo, Malin E. V. Johansson, Lisbeth Eklund, Henrik Sjövall, and Gunnar C. Hansson. 2011. “Altered O-Glycosylation Profile of MUC2 Mucin Occurs in Active Ulcerative Colitis and Is Associated with Increased Inflammation.” Inflammatory Bowel Diseases 17 (11): 2299–2307.

Lee, Jaewang, Eun Hye Kim, Daiha Shin, and Jong-Lyel Roh. 2017. “Accelerated Oral Wound Healing Using a Pre-Vascularized Mucosal Cell Sheet.” Scientific Reports 7 (1): 10667.

Lee, Seunghwan. 2013. “Characterization of Lubricity of Mucins at Polymeric Surfaces for Biomedical Applications.” In Proceedings of World Academy of Science, Engineering and Technology, 424. World Academy of Science, Engineering and Technology (WASET).

Lee, Seunghwan, Markus Müller, Kurosch Rezwan, and Nicholas D. Spencer. 2005. “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: The ACS Journal of Surfaces and Colloids 21 (18): 8344–53.

Lehr, Claus-Michael, Joke A. Bouwstra, Etienne H. Schacht, and Hans E. Junginger. 1992. “In Vitro Evaluation of Mucoadhesive Properties of Chitosan and Some Other Natural Polymers.” International Journal of Pharmaceutics 78 (1): 43–48.

Lehr, Claus-Michael, Fred G. J. Poelma, Hans E. Junginger, and Josef J. Tukker. 1991. “An Estimate of Turnover Time of Intestinal Mucus Gel Layer in the Rat in Situ Loop.” International Journal of Pharmaceutics 70 (3): 235–40.

Leiblum, Sandra R., Richard D. Hayes, Rita A. Wanser, and Jennifer S. Nelson. 2009. “Vaginal Dryness: A Comparison of Prevalence and Interventions in 11 Countries.” The Journal of Sexual Medicine 6 (9): 2425–33.

Liedberg, B., C. Nylander, and I. Lundström. 1995. “Biosensing with Surface Plasmon Resonance--How It All Started.” Biosensors & Bioelectronics 10 (8): i – ix.

Liedberg, Bo, Claes Nylander, and Ingemar Lunström. 1983. “Surface Plasmon Resonance for Gas Detection and Biosensing.” Sensors and Actuators 4 (January): 299–304.

Limsitthichaikoon, Sucharat, Bhattaranitch Khampaenjiraroch, Teerasak Damrongrungruang, Wanwisa Limphirat, Suthasinee Thapphasaraphong, and Aroonsri Priprem. 2018. “Topical Oral Wound Healing Potential of Anthocyanin Complex: Animal and Clinical Studies.” Therapeutic Delivery 9 (5): 359–74.

Lindh, Liselott, Per-Olof Glantz, Ingemar Carlstedt, Claes Wickström, and Thomas Arnebrant. 2002. “Adsorption of MUC5B and the Role of Mucins in Early Salivary Film Formation.” Colloids and Surfaces. B, Biointerfaces 25 (2): 139–46.

Lori, J. A., and A. J. Nok. 2004. “Mechanism of Adsorption of Mucin to Titanium in Vitro.” Bio-Medical Materials and Engineering 14 (4): 557–63.

Lukić, Jovanka, Ivana Strahinić, Branko Jovčić, Brankica Filipić, Ljubiša Topisirović, Milan Kojić, and Jelena Begović. 2012. “Different Roles for Lactococcal Aggregation Factor and Mucin Binding Protein in Adhesion to Gastrointestinal Mucosa.” Applied and Environmental Microbiology 78 (22): 7993–8000.

Lundin, Maria, Tomas Sandberg, Karin D. Caldwell, and Eva Blomberg. 2009. “Comparison of the Adsorption Kinetics and Surface Arrangement of ‘as Received’ and Purified Bovine Submaxillary Gland Mucin (BSM) on Hydrophilic Surfaces.”

38

Journal of Colloid and Interface Science 336 (1): 30–39. Madsen, Jan Busk, Javier Sotres, Kirsi I. Pakkanen, Petr Efler, Birte Svensson, Maher

Abou Hachem, Thomas Arnebrant, and Seunghwan Lee. 2016. “Structural and Mechanical Properties of Thin Films of Bovine Submaxillary Mucin versus Porcine Gastric Mucin on a Hydrophobic Surface in Aqueous Solutions.” Langmuir: The ACS Journal of Surfaces and Colloids 32 (38): 9687–96.

Malmsten, Martin, Eva Blomberg, Per Claesson, Ingemar Carlstedt, and Ingela Ljusegren. 1992. “Mucin Layers on Hydrophobic Surfaces Studied with Ellipsometry and Surface Force Measurements.” Journal of Colloid and Interface Science 151 (2): 579–90.

Mårtensson, S., S. B. Levery, T. T. Fang, and B. Bendiak. 1998. “Neutral Core Oligosaccharides of Bovine Submaxillary Mucin--Use of Lead Tetraacetate in the Cold for Establishing Branch Positions.” European Journal of Biochemistry / FEBS 258 (2): 603–22.

Mavragani, Clio P., and Haralampos M. Moutsopoulos. 2014. “Sjögren Syndrome.” CMAJ: Canadian Medical Association Journal = Journal de l’Association Medicale Canadienne 186 (15): E579–86.

McClelland, Sara I., Kathryn J. Holland, and Jennifer J. Griggs. 2015. “Vaginal Dryness and Beyond: The Sexual Health Needs of Women Diagnosed With Metastatic Breast Cancer.” Journal of Sex Research 52 (6): 604–16.

McColl, James, Gleb E. Yakubov, and Jeremy J. Ramsden. 2008. “Temperature Dependence of Mucin Adsorption.” Langmuir: The ACS Journal of Surfaces and Colloids 24 (3): 902–5.

Mikos, Antonios G., and Nikolaos A. Peppas. 1989. “Measurement of the Surface Tension of Mucin Solutions.” International Journal of Pharmaceutics 53 (1): 1–5.

Mittal, Nitesh, Ronnie Jansson, Mona Widhe, Tobias Benselfelt, Karl M. O. Håkansson, Fredrik Lundell, My Hedhammar, and L. Daniel Söderberg. 2017. “Ultrastrong and Bioactive Nanostructured Bio-Based Composites.” ACS Nano 11 (5): 5148–59.

Mortazavi, H., M. Yaseri, M. Baharvand, and A. Mortazavi. 2014. “Re-Evaluation of the First Phenytoin Paste Healing Effects on Oral Biopsy Ulcers.” Annals of Medical and Health Sciences Research. https://doi.org/10.4103/2141-9248.144877.

Mortazavi, S. Alireza, and John D. Smart. 1993. “An Investigation into the Role of Water Movement and Mucus Gel Dehydration in Mucoadhesion.” Journal of Controlled Release: Official Journal of the Controlled Release Society 25 (3): 197–203.

Mystkowska, Joanna, Dawid Łysik, and Marcin Klekotka. 2019. “Effect of Saliva and Mucin-Based Saliva Substitutes on Fretting Processes of 316 Austenitic Stainless Steel.” Metals 9 (2): 178.

Nilebäck, Linnea, Dimple Chouhan, Ronnie Jansson, Mona Widhe, Biman B. Mandal, and My Hedhammar. 2017. “Silk–Silk Interactions between Silkworm Fibroin and Recombinant Spider Silk Fusion Proteins Enable the Construction of Bioactive Materials.” ACS Applied Materials & Interfaces. https://doi.org/10.1021/acsami.7b10874.

Nilebäck, Linnea, Jesper Hedin, Mona Widhe, Lotta S. Floderus, Annika Krona, Helena Bysell, and My Hedhammar. 2017. “Self-Assembly of Recombinant Silk as a Strategy for Chemical-Free Formation of Bioactive Coatings: A Real-Time Study.” Biomacromolecules 18 (3): 846–54.

Nordgård, Catherine Taylor, and Kurt I. Draget. 2011. “Oligosaccharides as Modulators of Rheology in Complex Mucous Systems.” Biomacromolecules 12 (8): 3084–90.

Nordgård, Catherine Taylor, Unni Nonstad, Magnus Ø. Olderøy, Terje Espevik, and Kurt I. Draget. 2014. “Alterations in Mucus Barrier Function and Matrix Structure Induced by Guluronate Oligomers.” Biomacromolecules 15 (6): 2294–2300.

Nordman, H., J. R. Davies, A. Herrmann, N. G. Karlsson, G. C. Hansson, and I. Carlstedt. 1997. “Mucus Glycoproteins from Pig Gastric Mucosa: Identification Ofdifferent Mucin Populations from the Surface Epithelium.” Biochemical Journal 326 ( Pt 3) (September): 903–10.

39

Nordman, Henrik, Julia R. Davies, and Ingemar Carlstedt. 1998. “Mucus Glycoproteins from Pig Gastric Mucosa: Different Mucins Are Produced by the Surface Epithelium and the Glands.” Biochemical Journal. https://doi.org/10.1042/bj3310687.

Oh, Sejin, Matthew Wilcox, Jeffrey P. Pearson, and Salvador Borrós. 2015. “Optimal Design for Studying Mucoadhesive Polymers Interaction with Gastric Mucin Using a Quartz Crystal Microbalance with Dissipation (QCM-D): Comparison of Two Different Mucin Origins.” European Journal of Pharmaceutics and Biopharmaceutics: Official Journal of Arbeitsgemeinschaft Fur Pharmazeutische Verfahrenstechnik e.V 96: 477–83.

Ohta, Naoshi, Masato Sato, Kiminori Ushida, Mami Kokubo, Takayuki Baba, Kayoko Taniguchi, Makoto Urai, Koji Kihira, and Joji Mochida. 2009. “Jellyfish Mucin May Have Potential Disease-Modifying Effects on Osteoarthritis.” BMC Biotechnology 9 (December): 98.

Ouellette, Mathieu, Florence Masse, Mathilde Lefebvre-Demers, Quentin Maestracci, Philippe Grenier, Robert Millar, Nicolas Bertrand, Manuel Prieto, and Élodie Boisselier. 2018. “Insights into Gold Nanoparticles as a Mucoadhesive System.” Scientific Reports 8 (1): 14357.

Pacchioni, Giulia. 2019. “A Not-so-Strong Bond.” Nature Reviews Materials. https://doi.org/10.1038/s41578-019-0094-3.

Padra, Médea, Barbara Adamczyk, John Benktander, Bram Flahou, Emma C. Skoog, János Tamás Padra, Annemieke Smet, et al. 2018. “Helicobacter Suis Binding to Carbohydrates on Human and Porcine Gastric Mucins and Glycolipids Occurs via Two Modes.” Virulence 9 (1): 898–918.

Perez, Eric, and Jacques Emile Proust. 1987. “Forces between Mica Surfaces Covered with Adsorbed Mucin across Aqueous Solution.” Journal of Colloid and Interface Science 118 (1): 182–91.

Perez-Vilar, J., and R. L. Hill. 1999. “The Structure and Assembly of Secreted Mucins.” The Journal of Biological Chemistry 274 (45): 31751–54.

Perez-Vilar, Juan, John C. Olsen, Michael Chua, and Richard C. Boucher. 2005. “pH-Dependent Intraluminal Organization of Mucin Granules in Live Human Mucous/goblet Cells.” The Journal of Biological Chemistry 280 (17): 16868–81.

Perioli, Luana, Valeria Ambrogi, Fausta Angelici, Maurizio Ricci, Stefano Giovagnoli, Marinella Capuccella, and Carlo Rossi. 2004. “Development of Mucoadhesive Patches for Buccal Administration of Ibuprofen.” Journal of Controlled Release: Official Journal of the Controlled Release Society 99 (1): 73–82.

Politis, Constantinus, Joseph Schoenaers, Reinhilde Jacobs, and Jimoh O. Agbaje. 2016. “Wound Healing Problems in the Mouth.” Frontiers in Physiology 7 (November): 507.

Ramineni, Sandeep K., Craig B. Fowler, Paul D. Fisher, Larry L. Cunningham, and David A. Puleo. 2015. “Effects of Epidermal Growth Factor-Loaded Mucoadhesive Films on Wounded Oral Tissue Rafts.” Biomedical Materials 10 (1): 015026.

Randell, Scott H., Richard C. Boucher, and University of North Carolina Virtual Lung Group. 2006. “Effective Mucus Clearance Is Essential for Respiratory Health.” American Journal of Respiratory Cell and Molecular Biology 35 (1): 20–28.

Reijden, W. A. van der, A. Vissink, E. C. Veerman, and A. V. Amerongen. 1999. “Treatment of Oral Dryness Related Complaints (xerostomia) in Sjögren’s Syndrome.” Annals of the Rheumatic Diseases 58 (8): 465–74.

Rodsaward, Pongsawat, Titipong Prueksrisakul, Tawatchai Deekajorndech, Steven W. Edwards, Michael W. Beresford, and Direkrit Chiewchengchol. 2017. “Oral Ulcers in Juvenile-Onset Systemic Lupus Erythematosus: A Review of the Literature.” American Journal of Clinical Dermatology 18 (6): 755–62.

Roh, Jong-Lyel, Hyejin Jang, Jaewang Lee, Eun Hye Kim, and Daiha Shin. 2017. “Promotion of Oral Surgical Wound Healing Using Autologous Mucosal Cell Sheets.” Oral Oncology 69 (June): 84–91.

Sandberg, Tomas, Jan Carlsson, and Marjam Karlsson Ott. 2007. “Mucin Coatings

40

Suppress Neutrophil Adhesion to a Polymeric Model Biomaterial.” Microscopy Research and Technique 70 (10): 864–68.

Sandberg, Tomas, Marjam Karlsson Ott, Jan Carlsson, Adam Feiler, and Karin D. Caldwell. 2009. “Potential Use of Mucins as Biomaterial Coatings. II. Mucin Coatings Affect the Conformation and Neutrophil-Activating Properties of Adsorbed Host Proteins--toward a Mucosal Mimic.” Journal of Biomedical Materials Research. Part A 91 (3): 773–85.

Sauerbrey, Günter. 1959. “Verwendung von Schwingquarzen Zur Wägung Dünner Schichten Und Zur Mikrowägung.” Zeitschrift Für Physik 155 (2): 206–22.

Schneider, Hannah, Thaher Pelaseyed, Frida Svensson, and Malin E. V. Johansson. 2018. “Study of Mucin Turnover in the Small Intestine by in Vivo Labeling.” Scientific Reports 8 (1): 5760.

Schoemig, Veronika, Eda Isik, Lea Martin, and Sonja Berensmeier. 2017. “Solid Liquid Liquid Extraction of Porcine Gastric Mucins from Homogenized Animal Material.” RSC Advances 7 (63): 39708–17.

Schömig, Veronika J., Benjamin T. Käsdorf, Christoph Scholz, Konstantinia Bidmon, Oliver Lieleg, and Sonja Berensmeier. 2016. “An Optimized Purification Process for Porcine Gastric Mucin with Preservation of Its Native Functional Properties.” RSC Advances 6 (50): 44932–43.

Shi, L., and K. D. Caldwell. 2000. “Mucin Adsorption to Hydrophobic Surfaces.” Journal of Colloid and Interface Science 224 (2): 372–81.

Shi, Lei. 2000. “Biomimetic Surfaces of Biomaterials Using Mucin-Type Glycoproteins.” Trends in Glycoscience and Glycotechnology: TIGG 12 (66): 229–39.

Shi, Lei, Reza Ardehali, Karin D. Caldwell, and Paul Valint. 2000. “Mucin Coating on Polymeric Material Surfaces to Suppress Bacterial Adhesion.” Colloids and Surfaces. B, Biointerfaces 17 (4): 229–39.

Shi, Lei, Reza Ardehali, Paul Valint, and Karin D. Caldwell. 2001. “Bacterial Adhesion to a Model Surface with Self-Generated Protection Coating of Mucin via Jacalin.” Biotechnology Letters 23 (6): 437–41.

Shogren, R., T. A. Gerken, and N. Jentoft. 1989. “Role of Glycosylation on the Conformation and Chain Dimensions of O-Linked Glycoproteins: Light-Scattering Studies of Ovine Submaxillary Mucin.” Biochemistry 28 (13): 5525–36.

Shuter, J., V. B. Hatcher, and F. D. Lowy. 1996. “Staphylococcus Aureus Binding to Human Nasal Mucin.” Infection and Immunity 64 (1): 310–18.

Singh, Ajay P., Pallavi Chaturvedi, and Surinder K. Batra. 2007. “Emerging Roles of MUC4 in Cancer: A Novel Target for Diagnosis and Therapy.” Cancer Research 67 (2): 433–36.

Singh, Ajay P., Subhash C. Chauhan, Sangeeta Bafna, Sonny L. Johansson, Lynette M. Smith, Nicolas Moniaux, Ming-Fong Lin, and Surinder K. Batra. 2006. “Aberrant Expression of Transmembrane Mucins, MUC1 and MUC4, in Human Prostate Carcinomas.” The Prostate 66 (4): 421–29.

Singh, Ajay P., Nicolas Moniaux, Subhash C. Chauhan, Jane L. Meza, and Surinder K. Batra. 2004. “Inhibition of MUC4 Expression Suppresses Pancreatic Tumor Cell Growth and Metastasis.” Cancer Research 64 (2): 622–30.

Sogias, Ioannis A., Adrian C. Williams, and Vitaliy V. Khutoryanskiy. 2012. “Chitosan-Based Mucoadhesive Tablets for Oral Delivery of Ibuprofen.” International Journal of Pharmaceutics. https://doi.org/10.1016/j.ijpharm.2012.07.007.

Spencer, Nicholas D. 2014. “Aqueous Lubrication with Poly(Ethylene Glycol) Brushes.” Tribology Online 9 (4): 143–53.

Stenberg, Esa, Björn Persson, Håkan Roos, and Csaba Urbaniczky. 1991. “Quantitative Determination of Surface Concentration of Protein with Surface Plasmon Resonance Using Radiolabeled Proteins.” Journal of Colloid and Interface Science 143 (2): 513–26.

Strombeck, D. R., and D. Harrold. 1974. “Binding of Cholera Toxin to Mucins and Inhibition by Gastric Mucin.” Infection and Immunity 10 (6): 1266–72.

41

Sudhakar, Yajaman, Ketousetuo Kuotsu, and A. K. Bandyopadhyay. 2006. “Buccal Bioadhesive Drug Delivery — A Promising Option for Orally Less Efficient Drugs.” Journal of Controlled Release. https://doi.org/10.1016/j.jconrel.2006.04.012.

Su, Xiaodi, Ying-Ju Wu, and Wolfgang Knoll. 2005. “Comparison of Surface Plasmon Resonance Spectroscopy and Quartz Crystal Microbalance Techniques for Studying DNA Assembly and Hybridization.” Biosensors & Bioelectronics 21 (5): 719–26.

Sweeney, M. P., J. Bagg, W. P. Baxter, and T. C. Aitchison. 1997. “Clinical Trial of a Mucin-Containing Oral Spray for Treatment of Xerostomia in Hospice Patients.” Palliative Medicine 11 (3): 225–32.

Tang, Yijun, Xiangqun Zeng, and Jennifer Liang. 2010. “Surface Plasmon Resonance: An Introduction to a Surface Spectroscopy Technique.” Journal of Chemical Education 87 (7): 742–46.

Tavani, A., and C. La Vecchia. 1999. “The Adverse Effects of Hormone Replacement Therapy.” Drugs & Aging 14 (5): 347–57.

Tettamanti, G., and W. Pigman. 1968. “Purification and Characterization of Bovine and Ovine Submaxillary Mucins.” Archives of Biochemistry and Biophysics 124 (1): 41–50.

Thongborisute, Jringjai, and Hirofumi Takeuchi. 2008. “Evaluation of Mucoadhesiveness of Polymers by BIACORE Method and Mucin-Particle Method.” International Journal of Pharmaceutics 354 (1-2): 204–9.

Thornton, David J., Karine Rousseau, and Michael A. McGuckin. 2008. “Structure and Function of the Polymeric Mucins in Airways Mucus.” Annual Review of Physiology 70: 459–86.

Tildy, Bernadett E., and Duncan F. Rogers. 2015. “Therapeutic Options for Hydrating Airway Mucus in Cystic Fibrosis.” Pharmacology 95 (3-4): 117–32.

Tschoppe, P., H. Meyer-Lueckel, R. Toll, and A. M. Kielbassa. 2007. “In Vitro Analysis of an New Saliva Substitute (Saliva Natura) on Enamel and Dentin.” Laryngo-Rhino-Otologie 86 (10): 723–27.

Tsuji, T., and T. Osawa. 1986. “Carbohydrate Structures of Bovine Submaxillary Mucin.” Carbohydrate Research 151 (August): 391–402.

Verdugo, P. 1990. “Goblet Cells Secretion and Mucogenesis.” Annual Review of Physiology 52: 157–76.

Verkman, A. S., Yuanlin Song, and Jay R. Thiagarajah. 2003. “Role of Airway Surface Liquid and Submucosal Glands in Cystic Fibrosis Lung Disease.” American Journal of Physiology. Cell Physiology 284 (1): C2–15.

Whitcher, J. P. 2004. “The Treatment of Dry Eyes.” The British Journal of Ophthalmology 88 (5): 603–4.

Widhe, Mona, Helena Bysell, Sara Nystedt, Ingrid Schenning, Martin Malmsten, Jan Johansson, Anna Rising, and My Hedhammar. 2010. “Recombinant Spider Silk as Matrices for Cell Culture.” Biomaterials 31 (36): 9575–85.

Winkeljann, Benjamin, Kathrin Boettcher, Bizan N. Balzer, and Oliver Lieleg. 2017. “Mucin Coatings Prevent Tissue Damage at the Cornea--Contact Lens Interface.” Advanced Materials Interfaces 4 (19): 1700186.

Woodley, John. 2001. “Bioadhesion.” Clinical Pharmacokinetics 40 (2): 77–84. Xu, Jinke, Mifong Tam, Sepideh Samaei, Sophie Lerouge, Jake Barralet, Mary M.

Stevenson, and Marta Cerruti. 2017. “Mucoadhesive Chitosan Hydrogels as Rectal Drug Delivery Vessels to Treat Ulcerative Colitis.” Acta Biomaterialia 48 (January): 247–57.

Yakubov, Gleb E., Aristeidis Papagiannopoulos, Elodie Rat, and Thomas A. Waigh. 2007. “Charge and Interfacial Behavior of Short Side-Chain Heavily Glycosylated Porcine Stomach Mucin.” Biomacromolecules 8 (12): 3791–99.

Yu, Tao, Gavin P. Andrews, and David S. Jones. 2014. “Mucoadhesion and Characterization of Mucoadhesive Properties.” In Mucosal Delivery of Biopharmaceuticals: Biology, Challenges and Strategies, edited by José das Neves and Bruno Sarmento, 35–58. Boston, MA: Springer US.