Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Langmuir. Author manuscript; available in PMC 2009 Sep 17.
Published in final edited form as:
PMCID: PMC2746015

Metals and the Integrity of a Biological Coating

The Cuticle of Mussel Byssus


The cuticle of mussel byssal threads is a robust natural coating that combines high extensibility with high stiffness and hardness. In this study, fluorescence microscopy and elemental analysis were exploited to show that the 3,4-dihydroxyphenyl-l-alanine (dopa) residues of mussel foot protein-1 colocalize with Fe and Ca distributions in the cuticle of Mytilus galloprovincialis mussel byssal threads. Chelated removal of Fe and Ca from the cuticle of intact threads resulted in a 50% reduction in cuticle hardness, and thin sections subjected to the same treatment showed a disruption of cuticle integrity. Dopa-metal complexes may provide significant interactions for the integrity of composite cuticles deformed under tension.


Conventional hard coatings for protection against abrasion and wear are of little use on flexible substrates for electronics and medical applications.1,2 High substrate compliance leads to premature cracking of the coatings under even moderate bending and/or tensile loads.3-5 This deficiency is a consequence of the inherent tradeoff between the hardness and extensibility of most engineering materials. That is, an increase in hardness is offset by a reduction in the tensile breaking strain.6,7 We previously reported on a natural organic coating material-the protective cuticle of mussel holdfast threads-that evidently remains intact even when the threads are subjected to tensile strains up to 70%.8,9 Here we probe the cuticle composition with the goal of better understanding the operative strengthening mechanisms.

The mussel holdfast better known as the byssus consists of a bundle of threads evolved to enhance mussel survival in the wave-swept seashore.10 Each has a fibrous core covered by a thin protective cuticle and self-assembles from an ensemble of precursor proteins in the groove of the mussel foot in a process that resembles reaction injection molding.11 The core is composed of three different collagen-like proteins with silk- and elastin-like blocks arranged in such a way as to resemble the compliance of nylon at the distal end and a rubbery elastomer at the proximal end of each thread.12 In resisting mussel dislodgement by waves, the threads experience repeated high strain (e.g. 40% or more10). Consistent with its function to protect the fibers, the cuticle covering the fibrous core exhibits hardness and stiffness that are about an order of magnitude greater than those of the thread interior yet remains surprisingly extensible, with breaking strains approaching 70%.9 Microstructural examinations revealed that cuticle extensibility is enabled in part by deformable microphase-separated granules. The granules serve to arrest the growth of matrix microcracks that form during tensile strain, thereby keeping the coating over underlying fibers of the thread core intact.9

Given the evident contribution of microarchitecture to the mechanical properties of the cuticle, it is also necessary to understand the interrelationship between mechanics and chemistry. We have previously discussed how the microcracks themselves could result from the breaking of reversible noncovalent bonds.9 The high content of 3,4-dihydroxyphenyl-l-alanine (dopa) in the cuticle protein, mussel foot protein 1 (mfp-1), coupled with its affinity for metal coordination led us to propose dopa-metal complexes as candidates of such reversible cross-links in the thread cuticle.8,13 In the present study, we demonstrate that dopa and metals are colocated within the cuticle of M. galloprovincialis mussel threads and that metal removal significantly reduces cuticle hardness and integrity.


The autofluorescence of dopa (λem = 400-500 nm) has previously been used to verify its presence in biological materials.14-18 Consistent with the high dopa content in the cuticle protein, mfp-1 (10-20 mol%),19 we observed a strong intrinsic blue fluorescence from the cuticle of M. galloprovincialis threads in transverse sections (Figure 1a-c). The signal intensity was not differentiated by the characteristic composite structure of the cuticle (Figure 1d).

Figure 1
(a) Scanning electron micrograph of an M. galloprovincialis thread sectioned perpendicular to the fiber axis to expose the fibrous core and the granular cuticle. The orientation of sections used throughout this study is indicated by the dashed line. (b) ...

Elemental maps of transverse sections of the mussel threads were generated using secondary ion mass spectroscopy (SIMS) (Figure 2). They reveal that both Fe and Ca are concentrated within the cuticle whereas N appears to be more concentrated in the thread interior. Corroborating evidence was obtained from SIMS depth profiles taken on the external cuticle surface (Figure 3a) and using thermogravimetric analysis (TGA) of entire threads followed by electron desorption spectroscopy (EDS) (Figure 4 in Supporting Information). The total inorganic content of the byssal threads was found to be less than 1 wt %, with more than 90% present as Ca.

Figure 2
Maps of Fe, Ca, N, and C distributions in a transverse cross-section of a mussel thread generated using SIMS. The maps illustrate the concentrations of Fe and Ca in the cuticle. N appears to be more concentrated in the thread interior. C is present in ...
Figure 3
(a) Depth profile of Ca and Fe generated using SIMS on threads with and without treatment with 0.2 M EDTA at pH 5.5. The inset shows a burn crater following analysis. The cuticle in the analyzed thread was ~2.5 μm thick. Note the difference ...

Both Ca and Fe were extracted from the thread cuticle by incubation in ethylene diamine tetraacetic acid (EDTA) (Figure 3a). The contribution of the metals to mechanical properties was investigated by comparing nanoindentational hardness values of cuticles in whole threads with and without the metals removed before sectioning. Metal chelation with EDTA from whole threads led to a 50% reduction in cuticle hardness (t-test p < 0.001, Figure 3b), yet the cuticle structure viewed by TEM appeared to be largely unchanged (Figure 3c). When metals were extracted from thin sections of cuticle, however, some structural disintegration was evident (Figure 3c).


The actual dopa levels in cuticle cannot be directly determined because cuticle is not separable from the thread core. However, the fluorescence data indicate that dopa is uniformly distributed throughout the cuticle in that the signal intensity was not differentiated by the composite structure of the cuticle. On the basis of previous analyses, the distal portion of M. galloprovincialis threads contains an average dopa content of ~3 wt%.19 Because the cuticle represents about 10% of the total thread volume and the proteins of the thread core are reported to contain less than 0.5 wt % dopa,20 the dopa content in the cuticle can be estimated to be ~25-30 wt % (assuming that cuticle and the thread core have similar densities). This estimate resembles the dopa content of pure mfp-1, which ranges from 15 to 30 wt %.19

The catecholic side chain of dopa exhibits a moderate affinity (log KS ≈ 7-10) for many metal ions, but in some transition metals, the affinity is much higher (e.g., log KS = 18 for the monocatecholato-FeIII complex).21 In this regard, Taylor et al. demonstrated that when mfp-1 or mfp-1-derived peptides were mixed with FeIII at low iron to dopa ratios, tris-dopa-FeIII complexes (as shown in Figure 1e) with cumulative log stabilities (i.e., log KS) of 37-40 formed.22 A conservative estimate would allow that at the pH of seawater (pH 8.2), about 80% of this affinity would be realized.23 These complexes were later confirmed to be present in byssus by electron spin resonance.24 Both EDS and the histochemical detection of Fe have established its localization to the cuticle.19,25 On the basis of elemental analysis and fluorescence microscopy, our results indicate that dopa has the same distribution as Fe and Ca in the cuticle of M. galloprovincialis mussel threads. The colocalization in situ falls short of proving a chemical interaction between dopa and the metals, but given the accumulated evidence, their interaction is beyond a reasonable doubt.

Recently, the strengths of dopa-metal complexes were measured by single-molecule tensile tests.26 With a force to break of about 0.8 nN in water buffered at pH 8, these bonds are comparable to covalent bonds (~2 nN) yet are completely reversible. Noncovalent bonds have been demonstrated to increase extensibility in several types of biomimetic materials when engineered to replace covalent bonds as cross-links.27,28 During extension, these reversible bonds release stress buildup upon breaking and thereby prevent catastrophic material failure. Having tris-dopa-Fe3+ complexes (as shown in Figure 1e) as reversible cross-links for proteins in the M. galloprovincialis thread cuticle would be consistent with the high cuticle extensibility. Although the extent of tris-dopa-Fe3+ complexation in the M. galloprovincialis cuticle still needs to be determined, it is worth noting that compared to the cuticle of another mussel species (Perna canaliculus) known to contain a high density of cysteinyl-dopa cross-links the M. galloprovincialis thread cuticle exhibits a 300% greater extensibility with a compromise of only 30% in hardness.9,29

With total inorganic content at less than 1 wt % of dry byssus (Figure 4 in Supporting Information) and all of the Fe and Ca sequestered within the cuticle (as judged by the elemental maps in Figure 2), we calculate a total metal content in the cuticle at 10 wt % or less using estimates of cuticle volume and density as above. Previous work, particularly by Taylor et al.,22 Wilker,24,30,31 and Messersmith,26,32 supports a significant albeit complex role for FeIII-dopa chemistry within the cohesiveness of byssal cuticle. With Fe measured at less than 1/10 of the total inorganic content, we estimate an average molar ratio of between 3:1 and 4:1 of dopa/Fe for the cuticle after converting our weight percent measurements into molar equivalents. This estimate is stoichiometrically within the range of the tris-dopa-Fe3+ complex proposed by Taylor et al.22 However, given how variable FeIII content is in individual byssal threads19 this conjecture, although plausible, may be premature and awaits further mechanical testing of byssal cuticle with and without Fe. Indeed, given the redox tendencies of both dopa and FeIII, the outcome of mixing the two aerobically is still rather uncertain. For example, tris-dopa-Fe3+ complexes were reported to generate oxygen-free radicals that could lead to the formation of covalent didopa adducts.24 Such cross-links were detected in byssal plaques by McDowell et al.,33 but their existence in thread cuticle has not been shown.

Calcium, the other metal ion in cuticle, is consistently present and more abundant, but little thought has been given to its function. The catecholate portion of dopa binds Ca2+ with moderate stability (e.g., log Ks = 8);21 however, in contrast to FeIII, binding is usually through mono- to biscatecholato-Ca2+ complexation at pH 8. Moreover, Ca2+ significantly lowers the pKa of the catecholic OH groups, and the dopa semiquinone is stabilized by Ca2+ ions.34 Consistent with a cross-linking role, Ca2+ was observed to insolubilize films of tea tannins by catecholate bridging in the short term and by enhancing covalent cross-linking of tannins in the long term.35 Alternatively, if all of the available dopa in the cuticle, as suggested above, is tied up with tris-dopa-Fe3+ complexes, then cuticular Ca may be largely unconnected with dopa chemistry. Fatty acids have been reported to be present in byssus,36 and we have previously proposed that Ca2+ may interact with the polar head groups of such species.13 The observation that the nitrogen (N) content of the cuticle appears to be somewhat lower than the interior of the thread (Figure 2) supports the conjecture that macromolecules besides proteins are present in the cuticle.

The extraction data in conjunction with the homogeneous distribution of Fe and Ca within the cuticle supports the hypothesis that both metals are present as cross-linkers. Fe and Ca can be completely removed from the cuticle with a significant impact on its material properties as evinced by the 2-fold decrease in hardness (Figure 3a,b). Furthermore, metal removal from whole threads before sectioning has no apparent structural effect on the cuticular microstructure as viewed by TEM (Figure 3c). In contrast, treating thin thread sections with EDTA leads to detectable cuticle disintegration (Figure 3c) in agreement with the expected effect of breaking up intermolecular cross-links in a section that is only 80 nm thick. A better future assessment of the role of metal ions in the structural and mechanical integrity of thread cuticle would entail the testing of thread cuticles before and after metal removal and again following the restoration of Fe and Ca in the cuticles. Such studies were quite conclusive in the Zn- and histidine-rich proteins of Nereis jaws.37

We have demonstrated that small amounts of metals lead to significant (2-fold) elevations in the hardness of cuticle from the mussel M. galloprovincialis. The results are consistent with the hypothesis that the metals form complexes with dopa that serve as effective noncovalent cross-linking agents. Lessons learned from such studies on natural coatings may guide the design and synthesis of future biomimetic coatings with desirable combinations of hardness and extensibility.

Supplementary Material

Related data


This work was supported by the National Institutes of Health under awards nos. R01 DE015415 and R01-GM65354. This work made use of the MRL central facilities at UCSB supported by the MRSEC Program of the National Science Foundation under award no. DMR00-80034.


(1) Sun Y, Rogers JA. AdV. Mater. 2007;19:1897–1916.
(2) Ghosh SK. Functional Coatings. Wiley; Weinheim, Germany: 2006.
(3) Fabbri P, Singh B, Leterrier Y, Maanson J-AE, Messori M, Pilati F. Surf. Coat. Technol. 2006;200:6706–6712.
(4) Messori M, Toselli M, Pilati F, Fabbri E, Fabbri P, Pasquali L, Nannarone S. Polymer. 2004;45:805–813.
(5) Tsui TY, McKerrow AJ, Vlassak JJ. J. Mater. Res. 2005;20:2266–2273.
(6) Leyland A, Matthews A. Wear. 2000;246:1–11.
(7) Das D, Datta M, Chavan RB, Datta SK. J. Appl. Polym. Sci. 2005;98:484–489.
(8) Holten-Andersen N, Slack N, Zok F, Waite JH. Nano-Mechanical Investigation of the Byssal Cuticle, A Protective Coating of a Bio-Elastomer. In: Viney C, Katti K, Ulm F-J, Hellmich C, editors. Mechanical Properties of Bioinspired and Biological Materials; Materials Research Society Symposium, Fall 2004; Materials Research Society: Boston, MA. 2005.pp. 155–160.
(9) Holten-Andersen N, Fantner GE, Hohlbauch S, Waite JH, Zok FW. Nat. Mater. 2007:669–672. [PubMed]
(10) Moeser GM, Carrington E. J. Exp. Biol. 2006;209:1996–2003. [PubMed]
(11) Waite JH, Holten-Andersen N, Jewhurst S, Sun C. J. Adhes. 2005;81:297–317.
(12) Coyne KJ, Qin X-X, Waite JH. Science. 1997;277:1830–1832. [PubMed]
(13) Holten-Andersen N, Waite JH. J. Dent. Res. 2008;87:701–709. [PMC free article] [PubMed]
(14) Smith GJ, Haskell TG. J. Photochem. Photobiol. B. 2000;55:103–108. [PubMed]
(15) Andersen SO, Roepstorff P. Insect Biochem. Mol. Biol. 2005;35:1181–1188. [PubMed]
(16) Belli SI, Wallach MG, Luxford C, Davies MJ, Smith NC. Eukaryotic Cell. 2003;2:456–464. [PMC free article] [PubMed]
(17) Ferguson DJP, Belli SI, Smith NC, Wallach MG. Int. J. Parasitol. 2003;33:1329–1340. [PubMed]
(18) Belli SI, Smith NC, Ferguson DJP. Trends Parasitol. 2006;22:416–423. [PubMed]
(19) Sun C, Waite JH. J. Biol. Chem. 2005;280:39332–39336. [PubMed]
(20) Qin X-X, Coyne KJ, Waite JH. J. Biol. Chem. 1997;272:32623–32627. [PubMed]
(21) Martell AE. Stability Constants of Metal-Ion Complexes. supp 1. Vol. 25. Chemical Society; London: 1982. part II.
(22) Taylor SW, Luther GW, Waite JH. Inorg. Chem. 1994;33:5819–5824.
(23) Avdeef A, Sofen SR, Bregante TL, Raymond KN. J. Am. Chem. Soc. 1978;100:5362–5370.
(24) Sever MJ, Weisser JT, Monahan J, Srinivasan S, Wilker JJ. Angew. Chem., Int. Ed. 2004;43:448–450. [PubMed]
(25) Coulon J, Truchet M, Martoja R. Ann. Instit. Oceanogr. 1987;63:89–99.
(26) Lee H, Scherer NF, Messersmith PB. Proc. Natl. Acad. Sci. U.S.A. 2006;103:12999–13003. [PMC free article] [PubMed]
(27) Kong HJ, Wong E, Mooney DJ. Macromolecules. 2003;36:4582–4588.
(28) Kushner AM, Gabuchian V, Johnson EG, Guan Z. J. Am. Chem. Soc. 2007;129:14110–14111. [PMC free article] [PubMed]
(29) Zhao H, Waite JH. Biochemistry. 2005;44:15915–15923. [PMC free article] [PubMed]
(30) Monahan J, Wilker JJ. Langmuir. 2004;20:3724–3729. [PubMed]
(31) Westwood G, Horton TN, Wilker JJ. Macromolecules. 2007;40:3960–3964.
(32) Lee H, Lee BP, Messersmith PB. Nature. 2007;448:338–341. [PubMed]
(33) McDowell LM, Burzio LA, Waite JH, Schaefer J. J. Biol. Chem. 1999;274:20293–20295. [PubMed]
(34) Lebedev AV, Ivanova MV, Timoshin AA, Ruuge EK. ChemPhysChem. 2007;8:1863–1869. [PubMed]
(35) Yamada K, Abe T, Tanizawa Y. Food Chem. 2007;103:8–14.
(36) Cook M, Manly RS. Adhes. Biol. Syst. 1970:xxx–xxx.
(37) Broomell CC, Mattoni MA, Zok FW, Waite JH. J. Exp. Biol. 2006;209:3219–3225. [PMC free article] [PubMed]
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


  • Compound
    PubChem Compound links
  • PubMed
    PubMed citations for these articles
  • Substance
    PubChem Substance links

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...