Mussel Adhesive Proteins

The harsh living conditions of the seashore, the impact of strong waves, and the buoyant force of water turn out to be selective forces in evolution. Only organisms that developed strategies for holding tightly to surfaces survived. The science of the mussel's strong adhesion was the subject of famous work done by Brown in 1952 [90], and later, Waite and Tanzer made a major breakthrough [91]. In their study, they found that MAPs in threads and plaques have extensive amount of posttranslationally modified amino acids. This result immediately suggested that the modified amino acids might be the key players in strong adhesion. Subsequent biochemical analysis confirmed that MAPs contained two modified amino acids, 3,4-dihy-droxy-L-phenylalanine (DOPA) and hydroxyproline (Hyp), both of which were found in decapeptide repeat motifs in the protein named Mytilus edulis foot protein-1 (Mefp-1). So far, five adhesive proteins have been identified (Mefp-1 through Mefp-5). We will discuss their biochemistry, the adhesion chemistry of DOPA, and finally the potential medical applications using either whole MAPs or their functional mimics. Characterization of Mussel Adhesive Proteins

Mefp-1 is perhaps the best-characterized MAP. It has molecular weight ^110,000 Da and is composed of tandemly repeating (75-80 times) decapeptides with the following primary sequence: N-Ala-Lys-Pro-Ser-Tyr-Hyp-Hyp-Thr-DOPA-Lys-C, which was obtained by extensive tryptic digests [92]. Further chemical analysis found positional variations of posttranslational modification among the repetitive decapeptides. For example, some peptide fragments contained a DOPA residue instead of tyrosine at the fifth position and a Hyp instead of a proline at the third position and vice versa [93]. Nearly a decade later, it was discovered that another type of posttranslational modification of proline, trans-2,3-cis-3,4-dihydroxyproline was present at the sixth position in the decapeptide sequence [94]. Therefore, the corrected primary sequence of Mefp-1 decapeptide was revised to N-Ala-Lys-Pro-Ser-Tyr-Hyp-di'Hyp-Thr-DOPA-Lys-C.

Contrary to the initial belief that the biological role of Mefp-1 was as a potent adhesive protein, immunohistochemistry using the antibody raised against recombinant Dreissena polymorpha foot protein-1 (Dpfp-1), a close homolog of Mefp-1, showed unexpected tissue distribution of Mefp-1. The antibody specifically bound to nascent threads, but not to adhesive plaques (Figure 3.7), suggesting that the role of Mefp-1 might be the waterproof, hydrophobic outer coating of byssal thread [95]. This conclusion was further supported by the poor solubility that resulted after an oxidative cross-linking reaction between decapeptides. The catechol moiety of DOPA has a redox potential that causes it to undergo spontaneous oxidative cross-linking at ambient conditions. Two major cross-linking pathways have been suggested: (1) amine addition to the aryl ring of DOPA (Michael addition) and (2) aryl-aryl coupling between DOPA residues. In vitro cross-linking experiments clearly showed aryl coupling reactions (diDOPA) detected in both solid-state 13C NMR and MALDI-TOF mass spectrometry [96,97]. Additionally, mass spectrometry data suggested intermolecular Michael-addition reactions between Lys2 (or Lys10) and DOPA9, although this conclusion awaits confirmation as multiple adjacent peaks interfered with reliable analysis [97]. Two conflicting structural studies, a bent right-handed a-helix [98] and a left-handed type II polyproline helix [99] were reported. Despite the difference, these findings also support the suggested biological role of Mefp-1 as the hydrophobic outer coatings in byssal threads. The reason for this is that the a-helical structure was shared in both structural models and this common denominator strongly suggests that the decapeptide is the basic rigid structural unit, that when repeated 80 times, can achieve a long rodlike structure. In addition, this repeating structure facilitated extensive DOPA-mediated intermolecular cross-linking, resulting in tough waterproof proteinaceous materials. These physical features were confirmed: triggering oxidation of DOPAs in a film made of Mefp-1 by periodate or mushroom tyrosinase drastically increased its stiffness with remarkable volume shrinkage and decrease in solubility [100]. Taken together, these results showed that the functionality of Mefp-1 resided in the stiff coating of byssal threads rather than in adhesive plaques.

Although the role of DOPA in Mefp-1 appears cohesive rather than adhesive, the widespread belief in the adhesive function of DOPA remains due to clear demonstration of strong adhesion in synthetic DOPA-containing molecules (see Section Furthermore indirect evidence was provided by a direct purification of MAP from adhesive plaques. However, technically, this was a challenging task due to the extremely poor solubility of proteins located at the interfaces. This may be because of intrinsic insolubility of the protein, extensive DOPA, or other amino acid-mediated cross-linking, or other unknown, integral solidifying components. A breakthrough was made by a surprisingly simple technique, in which transfer of newly secreted soft plaques to low-temperature seawater (4°C-8°C) maximized the amount of extractable proteins [101]. Waite and colleagues found that the extracted protein gave four distinct bands in acid-urea PAGE gels and they further purified the smallest protein species. Amino acid sequencing showed that it had 48 amino acids (6 kDa) with a high DOPA content (20%) and unexpectedly, they found another posttranslationally modified amino acid, hydroxyarginine [102]. This small protein was named Mefp-3.



FIGURE 3.7 A cartoon showing the structure of bys-sus and the distribution of mussel adhesive proteins (MAPs) of Mytilus edulis. Mefp-1 is found on the surfaces of byssal threads as well as adhesion plaques. Mefp-2 and Mefp-4 are proteins located primarily in the bulk space of adhesion plaques. Finally, Mefp-3 and Mefp-5 are mainly found at the interface of adhesive pads and substrates.

FIGURE 3.7 A cartoon showing the structure of bys-sus and the distribution of mussel adhesive proteins (MAPs) of Mytilus edulis. Mefp-1 is found on the surfaces of byssal threads as well as adhesion plaques. Mefp-2 and Mefp-4 are proteins located primarily in the bulk space of adhesion plaques. Finally, Mefp-3 and Mefp-5 are mainly found at the interface of adhesive pads and substrates.

Although it is speculative, the role of DOPA in Mefp-3 maybe quite different from that in Mefp-1, in which it is an intermolecular cross-linker. In Mefp-3, DOPA residues may be true molecular adhesives (Figure 3.7) but at the same time, may still participate in oxidative cross-linking during the curing process. This plausible hypothesis is clearly supported by increased amount of extractable proteins during cold curing: the low temperature may effectively slow down the cross-linking reactions. However, the biological role of hydroxyarginine is still unknown. The additional hydroxyl moiety potentially generates up to six hydrogen-bond donor groups, which might be useful in enhancing adhesive properties of the protein.

Mefp-5 is a relatively small 74 amino acid protein (9.5 kDa) that is located mostly in adhesive pads, colocalizing with Mefp-3 (Figure 3.7). As expected based on its biodistribution, it also showed a high level of DOPA content (27 mol%), the highest among all MAPs, and contained an additional posttranslation-ally modified amino acid, phosphoserine [103]. Both Mefp-5 and Mefp-3 exhibit interesting primary amino acid sequences in which basic residues are frequently found adjacent to DOPA, i.e., Lys-DOPA or DOPA-Lys. This occurrence might present some functional benefits in which lysines would (1) increase solubility, (2) facilitate DOPA-mediated cross-linking, (3) enhance electrostatic adhesion to surfaces, and (4) provide structural rigidity to maximize surface contact areas. The phosphoserine was reported to have some degree of adhesive functionality especially in calcium-based mineralized substrates, which are the primary material (calcium carbonate) of a mussel shell [104]. Consequently, Mefp-5 has possibly three adhesive amino acid residues, DOPA, Lys, and phosphoserine, but functional mapping of the primary adhesive player as a function of surface variability remains an important goal for the future.

Mefp-2 and Mefp-4 are the least-studied proteins in this family, and it is perhaps the small content of DOPA that makes them less attractive to study: only 3% for Mefp-2 and 4% for Mefp-4. In striking contrast to the small DOPA content, they contribute more than 30% of the total plaque mass, which suggests a structural purpose for Mefp-2 and Mefp-4 in plaques [105].

Mefp-2 is a relatively large 40-kDa protein. Similar to Mefp-1, it has 11 repeats of a 35-40 amino acid consensus sequence. Compared to other Mefp proteins, however, it is unique with regard to its amino acid composition, having three disulfide bonds found in each repeat unit. Cysteine residues were not identified in any other Mefp-family proteins. Upon proteolysis, it was not fully digested by several types of proteases, strongly indicating that Mefp-2 has a well-defined structure due to the three disulfide linkages. However, detailed structural information is still not available [106], although an important structural clue was obtained by Inoue et al. [107]. In this study, they cloned cDNA from Mytilus galloprovincialis foot protein-2 (Mgfp-2), a species closely related to M. edulis. It also encoded a repeating ~40 amino acid peptide (11 repeats) with three disulfide bonds, and which is homologous to epidermal growth factor (EGF)-like domain. Another interesting feature of Mefp-2 is the distribution of six DOPA residues along the primary sequence of Mefp-2. They are located either in the N-terminus (#23, 31, 36, and 43) or C-terminus (#468 and 473), suggesting that DOPA may play a critical role in inter- or intraprotein cross-linking, for example, Mefp-2/Mefp-2, Mefp-2/Mefp-1, or Mefp-2/Mefp-3,5. This is conceivable when considering the tissue distribution of Mefp-2 (Figure 3.7). Chemical Basis of Adhesion

Based on the tissue distribution and DOPA content, the two MAPs likely responsible for mediating adhesion are Mefp-3 and Mefp-5. Fundamental questions at this point are, what is the essential requirement for such a strong adhesiveness? Do we need the entire sequence or just a specific functional element? Deming and colleagues provided some guidance by incorporating DOPA into a synthetic polypeptide chain. The random synthetic copolymers of lysine and DOPA showed that the adhesive strength was directly proportional to the DOPA content: the copolymer containing 20% DOPA was almost 10 times stronger than pure poly-L-lysine [108]. The strong binding of the DOPA-containing polymer was not substrate specific but rather a versatile property. Recent single molecule atomic force microscopy measurements showed that a force of approximately 800 pN was necessary to pull off DOPA from a TiO2 surface, which is roughly four times stronger than biotin-avidin binding ( ~ 200 pN) [137].

In-depth studies regarding DOPA(catechol)-surface interactions can be found in the environmental sciences literature. Titanium (Ti) has long been known to be a material that effectively removes toxic organic pollutants that frequently include a catechol moiety (the side chain of DOPA). The interaction between the pollutants and Ti is relatively easy to study because a Ti-coated surface can be obtained using a simple e-beam evaporation technique. Ti nanoparticles are also commercially available due to the demands from the fields of medicine and chemical catalysis.

A catechol-TiO2 binding model derived from IR and equilibrium adsorption isotherm experiments showed a bidentate-binuclear configuration [109], and was found to be different from the bidentate-mononuclear binding model derived by Rajh [110]. This conflict may come from the two different geometric extremes: the bidentate-mononuclear structure was determined in nanospheres (20 nm nanoparticles) whereas the bidentate-binuclear structure was determined in flat surfaces. Small nano-particles have unusual surface coordination geometries that may lead to a different binding configuration. Clearly, further detailed studies must be conducted to fully understand the role of DOPA in biological adhesion. Applications of MAP Mimetic Bioadhesives Molecular Adhesives

Bioadhesives have many potential uses in medical and dental applications. Adhesives for mineralized tissues can be used for orthopedic as well as dental surgery, and those that can be formulated as injectable adhesive hydrogels can be used for drug delivery systems in the oral cavity, respiratory, gastrointestinal, and reproductive tracts. Among these applications, several attempts have been made to develop mucoadhesive polymers such as poly(acrylic acids) (PAA) [111]. Limitations of a short residence time, a slow drug diffusion rate, and pH-dependent adhesion of PAA have triggered active efforts to find alternatives. For this reason, Messersmith and coworkers developed environmentally sensitive smart hydrogels with mucoadhesive functionality by incorporating DOPA. This rationale was based on the strong interaction of Mefp-1 with pig mucin glycoproteins [112,113]. Thus, we designed a smart hydrogel exhibiting temperature-induced gelation from a liquid precursor. The thermosensitive function was derived from poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) (PEO-PPO-PEO, also called Pluronic) to which DOPA was conjugated at both ends by activating the terminal hydroxyl groups [36]. The synthesized hydrogel showed significant adhesive interactions with mucin, demonstrating that this system can be potentially useful for mucoadhesive drug delivery.

Other adhesive hydrogels have been developed by exploiting the redox potential of DOPA-functio-nalized polymers to cross-link rapidly into hydrogels [114]. In addition, we have also developed DOPA-containing monomers that can be cross-linked by UV irradiation [115]. Finally, enzymatic cross-linking of rationally designed peptide-functionalized polymers has been used to develop DOPA-containing hydrogels [116]. In this study, DOPA-containing peptide substrates for transglutaminase (TGase) were optimized for the rapid TGase-mediated cross-linking so that the resulting hydrogels were readily formed in about 1 min. Such hydrogels are potentially useful for surgical tissue adhesives [117], drug delivery systems, and tissue engineering. Antifouling Surfaces

Perhaps one of the most promising, yet ironic, medical applications of MAP mimetic polymers is the prevention of implant surface fouling by proteins and cells. For example, artificial blood vessels are easily occluded by rapid deposition of proteins followed by subsequent molecular clotting cascades. Cardiovascular stents as well as metal-ceramic implants are important medical devices that have encountered this problem.

Ironically, the use of DOPA, a key component of a prolific fouler, as an antifouling material can be an excellent solution. The key idea here is the chemical conjugation of DOPA to an inert synthetic polymer such as PEG [118,119]. In this work, DOPA functioned as an anchor on gold, TiO2, and other inorganic surfaces and PEG repelled approaching biomacromolecules (mostly proteins). The fouling resistance was also extended to cells in which fibroblasts were cultured on PEG-DOPA-modified surfaces. Such experiments demonstrated more than 95% decrease in cell attachment to surfaces for periods of up to 2 weeks.

0 0

Post a comment