Biomechanics of byssal threads outside the Mytilidae: Atrina rigida and Ctenoides mitis

According to Tertullian (ca. 155–230 CE), it was not always possible to find sufficient earthly textiles; luckily, it `also proved possible to fish for clothes'. He goes on to say that `fleeces also come from the sea' (Tertullian, 2005). Though he may have had his tongue held firmly in cheek, Tertullian's `fleeces of the sea' were the byssal threads of Pinna nobilis Linnaeus which,at least by the 18th century, actually were sometimes woven into hats and gloves (Maeder, 2002).

Despite such interesting historical uses, no work in the last 50 years has investigated the mechanical properties of pinnid byssal threads [the most recent study is by Lucas and colleagues(Lucas et al., 1955)]. Instead, most byssal research to date has focused on the byssal threads of mytilids (mussels and their near relatives). This disproportionate emphasis on the Mytilidae is likely due to the presence of mytilids in easily accessible intertidal areas, but it fails to capture the widespread occurrence of byssal attachment among the Bivalvia and the potential diversity in byssal function and properties.

Although the byssus first evolved to aid in post-larval dispersal and settlement (Yonge, 1962; Stanley, 1972; Sigurdsson et al., 1976; De Blok and Tan-Maas, 1977; Lane et al., 1985), a recent catalogue of tropical marine bivalves revealed that about a quarter of the genera surveyed are byssally attached as adults(Todd, 2001). In fact, the only pteriomorph superfamilies without byssate adult representatives are characterized by a different attachment strategy – cementation(Márquez-Aliaga et al.,2005; Bieler and Mikkelsen,2006). Although there has been some research into non-mytilid byssal thread chemical composition, Dreissena polymorpha Pallas is the only bivalve from outside the Mytilidae whose threads have been the subject of biomechanical investigation(Jackson et al., 1953; Pujol, 1967; Pujol et al., 1970; Mascolo and Waite, 1986; Brazee and Carrington,2006).

Biomechanics researchers have thus restricted their study of an attachment structure that appears in every pteriomorph order to only a single family. This narrow focus is a problem that should be remedied, for two reasons:first, because of the phenomenon of phylogenetic non-independence, many interesting evolutionary questions can only be answered through comparative work across a wide range of taxa; and second, as researchers have pointed out the possible engineering applications of simulated `mussel glue', knowledge of a wider range of byssal thread compositions and properties is likely to yield rich insights into potential technological applications(Waite et al., 2005; Waite, 2008).

With more biomechanical data on the threads of both epifaunal and semi-infaunal bivalves from a variety of pteriomorph orders, one would be able to sort out whether life habits are correlated with the mechanical properties of byssal threads. This is an especially interesting question, as both endobyssate (infaunal or semi-infaunal with byssal attachment) and epibyssate(epifaunal with byssal attachment) groups declined during the Paleozoic and Mesozoic, perhaps due to increased predation pressure(Stanley, 1972; Stanley, 1977; Vermeij, 1983; Skelton et al., 1990; Aberhan et al., 2006; Harper, 2006). The surviving byssate groups live in a variety of environments, and their survival probably involved adjustments of their thread mechanics. There are certainly interesting chemical differences between the threads of different bivalve groups, which may translate into differences in mechanical properties. For example, mytilid threads are collagenous, whereas the threads of pinnids,anomiids and dreissenids are not (Jackson et al., 1953; Pujol,1967; Pujol et al.,1970; Mascolo and Waite,1986; Brazee and Carrington,2006).

We investigated the mechanical properties of the byssal threads of two bivalve species from two orders outside the Mytiloida: the pen shell Atrina rigida Lightfoot (Pterioida: Pinnidae) and the flame `scallop'Ctenoides mitis Lamarck (Limoida: Limidae). There is some debate in the literature about the breakdown of pteriomorph orders; we have adopted the classifications of Bieler and Mikkelsen rather than those of Matsumoto but these authors all agree that limids and pinnids belong to different orders(Matsumoto, 2003; Bieler and Mikkelsen, 2006). Thus, although A. rigida and C. mitis are more closely related to one another than to mytilid species, they are still only distantly related. This study does not seek to compare mytilid threads with`non-mytilid' threads in general but instead compares mytilid threads with the threads of two unrelated species from outside the Mytilidae.

The two species under investigation have quite different life habits– A. rigida is semi-infaunal, and usually lives in protected subtidal or low intertidal areas with most of its shell buried in muddy or sandy sediment; C. mitis, in contrast, is epifaunal, and normally lives byssally attached in crevices of reefs and ledges, swimming only if disturbed (Stanley, 1970; Mikkelsen and Bieler, 2003). Because A. rigida is semi-infaunal, data on the properties of its threads can be compared with data from our recent study that includes two semi-infaunal mytilids, Modiolus modiolus Linnaeus and Geukensia demissa Dillwyn [see accompanying paper(Pearce and LaBarbera, 2009)]. The properties of epifaunal C. mitis threads can similarly be compared with those of epifaunal mytilids like Mytilus californianusConrad and Mytilus edulis Linnaeus(Pearce and LaBarbera,2009).

Atrina rigida Lightfoot specimens were ordered from Gulf Specimen Marine Laboratories (Panacea, FL, USA), and kept in a tank at room temperature(approximately 18°C). We buried them as deeply as possible (∼5 cm) in the calcareous gravel in the aquarium, and lightly supported the exposed shell to prevent toppling. Ctenoides mitis Lamarck specimens were ordered from Ward's Natural Science (Rochester, NY, USA) [the `Lima scabra'specimens obtained from Ward's were identified as C. mitis rather than Ctenoides scaber Born, following Mikkelsen and Bieler(Mikkelsen and Bieler, 2003),on the basis of their white tentacles and greater number of radial ribs in the shell]. The C. mitis were kept in individual enclosures (polyethylene freezer containers with sections of the walls replaced with plastic mesh) in the same tank as the A. rigida specimens. Tank salinity was maintained at approximately 31–32 p.p.t. by adding either tap water or Instant Ocean® (Aquarium Systems, Inc., Mentor, OH, USA) sea salt mixture as necessary. Animals were fed daily on an artificial phytoplankton substitute(Kent Marine®, PhytoPlex™, Franklin, WI, USA), producing byssal threads and surviving without obvious ill effects for over 2 months.

We measured the shell length of all animals using digital calipers. To harvest threads from the C. mitis specimens, we opened each enclosure and disturbed the animal inside, causing it to release its threads and swim away. We then lifted the enclosure out of the tank and removed the thread plaques from the plastic walls using a razor blade. To harvest the A. rigida threads, we carefully dug out each animal and transferred it underwater into a smaller tray, which was then lifted out of the tank. We snipped each thread at the proximal end using iris scissors; the plaques usually remained attached to a small piece of gravel. All samples were stored in salt water (31–32 p.p.t.) at 5°C until testing.

Thread mechanical properties were measured using a custom-built tensile tester. The apparatus consisted of a lower grip at the bottom of a Plexiglas tank and an upper grip that could be displaced by turning a crank on a dovetail slider (Velmex, Bloomfield, NY, USA; Model A6027K1M-S6). The upper grip was attached to a 10 lb (∼45 N full scale) force transducer(OmegaDyne®, Sunbury, OH, USA; Model LC703-10). The four strain gauges in the transducer were set up as a full Wheatstone bridge supplied with a constant 5V excitation; the excitation and amplification of the voltage output of the bridge circuit were supplied by a bridge amplifier (Vishay®Micro-Measurements, Shelton, CT, USA; Model 2120A). We calibrated the voltage output of the amplifier to yield a voltage-to-force conversion factor. A linear variable differential transformer (Pickering Controls, Plainview, NY,USA; Model 7308-X2-A0) powered by a constant 5V DC from an external power supply converted the displacement of the upper grip into a voltage, which could then be converted back into a displacement value following calibration. The voltage was digitized using a GW Instruments (Somerville, MA, USA) Model 100B analog-to-digital converter.

We limited each testing run to 10–15 byssal thread samples to minimize drying during preparation. Between one and six byssal threads from each individual were tested, with a total sample of about 20–25 threads per species. To ensure proper gripping, we sandwiched each end of each thread between two small squares of 100% rag paper using a drop of cyanoacrylate adhesive (Loctite® `Gel Control' super glue; Henkel Consumer Adhesives,Inc., Avon, OH, USA) to maximize adhesion. Before testing, we measured the length of each byssal thread sample with digital calipers.

Prior to each test, we secured one end of the thread in the upper grip of the tester and the other end in the lower grip at the base of the tank; the entire thread was immersed in sea water for the duration of the test. The tank was filled with salt water from the 5°C tank (salinity 31–32 p.p.t.)during all tests. Once the thread was secured, we initiated data capture in the application instruNet World Mac (GW Instruments) and displaced the upper grip at approximately 0.5 mm s^–1 until thread failure. At the outset of the test, the samples were slack; the beginning of the tensile test was taken to be the point at which there was a non-negligible force on the sample.

Following testing, we inspected the broken ends of each byssal thread under a dissecting microscope to assess the failure mode (e.g. smooth break,fraying, etc.). We took digital photographs (Nikon D100 camera back) of each broken end through the dissecting microscope at approximately ×100, and measured thread diameter using ImageJ (NIH). Following previous work,cross-sections were assumed to be circular even though byssal threads are often elliptical in cross-section (Brazee and Carrington, 2006). Initially we measured the minimum thread diameter before testing, but discovered that the samples invariably broke at a different (and wider) location, presumably a cryptic weak point in the structure. Thus the diameter at failure was used in all calculations of strain to ensure consistency, although this does result in an underestimate of the inherent strength of byssal thread material.

The stress (force per unit area) and strain (displacement per unit length)for each test were plotted in Microsoft® Excel® to produce a stress–strain curve. Because strains were always in excess of 50%, it was clear that byssal thread cross-sectional area and length changed significantly during the test. Thus instead of `engineering' strain(ϵ<sub>E</sub>=ΔL/L<sub>0</sub>, where Lis length and subscript 0 indicates initial) we used `true' or `logarithmic'strain [ϵ<sub>T</sub>=ln(L/L<sub>0</sub>)], which does not assume constant length or constant volume. Stress is always calculated assuming a certain value for Poisson's ratio (ν), which is defined in this case as the negative of the ratio of transverse to axial strain. The instantaneous diameter of the thread is given by d=d<sub>0</sub>exp(–νϵ<sub>T</sub>). There are two possible approaches. (1) `Engineering' stress (σ_E)assumes constant area: ν=0, thus d=d₀ andσ _E=F/A₀ (where F is force and A is cross-sectional area). (2) `True' stress(σ_T) assumes constant volume: ν=0.5, andσ _T=σ_Eexp(ϵ_T). We conservatively assumed constant volume rather than constant area (see Pearce and LaBarbera, 2009). A number of different mechanical properties can be determined from the stress–strain curve. In almost all cases, there was a sharp drop in stiffness at a characteristic stress level – the yield stress. The slope of the stress–strain curve represents the stiffness of the material;thread stiffness was determined both for the initial loading of the thread and at thread failure. The stress and strain at failure are termed strength and extensibility, respectively. Finally, by fitting a polynomial to the stress–strain curve and integrating over the total strain, the area under the curve was determined; this area is the energy absorbed per unit volume, or the toughness of the material.

A small percentage of the byssal thread stress–strain curves for each species differed dramatically from the characteristic shape of the curve for that species. In almost all cases, the discrepancy appeared to result from splitting and fraying of the thread prior to failure; we did not include the data from these samples in the analysis.

We analyzed the data using StatView 5.0 (SAS Institute, Cary, NC, USA). First, we conducted an ANOVA on the threads of each individual, followed by an ANOVA of all threads of each species, split by individual. Because no significant differences were detected, we then pooled the individuals within each species and ran an overall ANOVA, split by species. We performed post-hoc Scheffe tests to determine the specific differences detected by the ANOVA. We also ran a Kruskal–Wallis test (a non-parametric version of a standard ANOVA), as a normal distribution of the data could not be assumed. Finally, we produced a partial correlation matrix for each species to determine whether any two of the measured variables were significantly correlated when all other variables were held constant.

For all measured variables, ANOVA revealed no significant differences between threads of a given individual or between individuals of a given species; thus the threads for each species were pooled in the overall analysis.

As shown in Table 1, the overall ANOVA for diameter, which included mytilid species from a previous study (see Pearce and LaBarbera,2009), revealed a clear division between semi-infaunal and epifaunal species: the threads of all epifaunal species were significantly thicker than those of all semi-infaunal species (Scheffe test: P<0.012). Threads of epifaunal species were 2–4 times the diameter of threads of infaunal species. However, the threads of the epifaunal C. mitis were significantly thinner than those of one of the three other epifaunal species, M. californianus (Scheffe test: P=0.007). While the shells of the mytilid species fell into a similar size range (60–70 mm on average), those of C. mitis were somewhat smaller and those of A. rigida were much larger.

`True' stress and strain were used to construct the stress–strain curves for all of the byssal thread samples. The curve of a representative byssal thread from each species is given in Fig. 1. Mytilid stress–strain curves from a previous study(Pearce and LaBarbera, 2009)have been included for comparison. The curve for C. mitis (green)exhibits a dramatically different shape from those of other threads examined to date – it has an early yield point and then a very long region of relatively uniform, low stiffness, finally stiffening slightly and breaking at an extremely high strain.

Strikingly, the A. rigida curve(Fig. 1, yellow) exhibits two distinct yield points. Thus A. rigida threads have a stress–strain curve similar to those of M. modiolus(Fig. 1, light blue), which display the same double-yield behavior(Fig. 2). Atrina rigida and M. modiolus, both semi-infaunal species, have threads that yield twice before failure, while the threads of all tested epifaunal species exhibit only a single distinct yield point.

In terms of byssal thread mechanical properties, C. mitis and A. rigida differed significantly from each other as well as from species within the Mytilidae. As Fig. 1 suggests, the threads of C. mitis were consistently weaker and less stiff than those of other species, often significantly so(Table 2). In addition, C. mitis threads yielded at a significantly lower stress than all other threads tested. However, despite their low strength and stiffness, C. mitis threads proved highly extensible, with an average final strain of 81% (Table 2). This extensibility was significantly greater than that of all other byssal threads tested, which ranged between 44% and 67%. Nevertheless, even with this higher strain to failure, C. mitis threads were significantly less tough than those of most mytilid species due to their low strength(Table 2).

The A. rigida threads were significantly weaker, in terms of both strength and toughness, than the majority of mytilid threads(Table 2). The threads of most species tested were significantly stiffer at the end of the test (paired t-test: P<0.0001; paired sign test: P<0.0001), but A. rigida and G. demissa threads did not exhibit a pronounced stiffening at higher strains. However, A. rigida threads were quite stiff at the outset, and had the highest average value for initial stiffness: 609 MPa(Table 2). Although they were stiff, the threads of A. rigida had a low extensibility. In stark contrast to the extremely stretchy threads of C. mitis, those of A. rigida had an average strain at failure of only 44%, significantly lower than that of all other threads tested(Table 2). Overall, the mechanical properties of the threads of C. mitis and A. rigida differed substantially from those of mytilid threads.

For all bivalve species examined to date, stronger threads tended to be tougher, and threads with a higher yield stress had a higher initial stiffness(Table 3). Whereas the significant correlations between properties of C. mitis threads followed patterns similar to those of mytilid threads, the A. rigidacorrelations seemed specifically to parallel those of M. modiolus. Like the byssal threads of M. modiolus but unlike those of all other species, the strength of A. rigida threads was not highly correlated with their failure stiffness (Table 3). Again like M. modiolus, tougher A. rigidabyssal threads tended to exhibit a higher final stiffness, the opposite of the relationship found for the other species. Finally, initial and final stiffness for A. rigida and M. modiolus threads were not highly correlated, which contrasts with the results for G. demissa and M. edulis threads (Table 3). (For a complete list of property values for all individual threads tested in this study, see supplementary material Table S1.)

The correlation found here between life habit (epifaunal vssemi-infaunal) and byssal thread diameter mirrors a similar relationship among mytilids [see accompanying paper (Pearce and LaBarbera, 2009)]. The association between semi-infaunal life habits and small thread diameters found within the Mytilidae might have been due to the fact that tested mytilid species with similar life habits were closely related to one another – although Geukensia is not a sister taxon of Modiolus, the two epifaunal mytilids tested are both in the genus Mytilus (Distel,2000). However, the data on byssal thread diameter presented here for three different pteriomorph orders strengthens considerably the connection between life habit and byssal thread size. And although shell length (i.e. size) is likely involved in determining thread diameter within species(Brazee and Carrington, 2006),it is striking that threads produced by the A. rigida specimens,whose shells measured over 100 mm and were the largest in our study, were significantly thinner than those produced by the much smaller mytilids.

As previously observed among mytilids(Meadows and Shand, 1989),semi-infaunal species seem to produce a very large number of thin threads,whereas epifaunal species produce a smaller number of thicker threads. One reason for this difference might be that having a larger number of thin threads is more effective in anchoring semi-infaunal animals within a particulate substrate, as the threads can create an extensive network of individual attachments to small particles. The M. modiolus examined in a previous study tended to leave the glass plates to which we tried to confine them and bury themselves in the gravel, from which they were difficult to extricate without digging (Pearce and LaBarbera, 2009). For Mytilus species, on the other hand,which attach to rocks and other hard substrates, a smaller number of thick threads may provide a more reliable tether against wave action or predation(Bell and Gosline, 1996; Bell and Gosline, 1997; Carrington, 2002; Carrington and Gosline, 2004). The stalk-like byssus of some arcoids and pterioids may be an extreme form of this tendency to consolidate material into a smaller number of thicker threads(Oliver and Holmes, 2006; Tëmkin, 2006). Ctenoides mitis appears to represent a somewhat different case, as it only uses byssal threads for temporary attachment, and not for predator resistance; thus a set of weak but stretchy threads of intermediate thickness allows it to hang inside crevices, ready to drop the threads and swim away on disturbance.

The stress–strain curve of A. rigida threads exhibits two very clear yield points (Fig. 2), as seen previously for M. modiolus byssal threads(Brazee and Carrington, 2006). It would be interesting to re-examine existing molecular analyses of M. modiolus and A. rigida threads(Mascolo and Waite, 1986; Rzepecki et al., 1991) in light of this unusual yield pattern, which seems to imply an underlying two-phase molecular structure. Atrina rigida and M. modiolusbyssal threads have a great deal in common: they display an obvious double-yield behavior; they have a comparatively high initial stiffness and a comparatively low extensibility; and they share correlations not seen in the byssal threads of other species between stiffness at failure and other mechanical variables (Fig. 2, Table 2, Table 3). As M. modiolus and A. rigida are members of different orders within the Pteriomorphia (the Mytiloida and Pterioida) that are not sister taxa(Giribet and Wheeler, 2002; Matsumoto, 2003; Bieler and Mikkelsen, 2006),these commonalities suggest that their thread characteristics may be the product of convergent evolution. However, the threads of the only other semi-infaunal species tested, G. demissa, do not exhibit any of these characteristics. Geukensia demissa individuals often occur clumped together in marshy areas populated by sea grasses – the roots of the sea grasses, combined with the network of thin threads produced by the animals,may provide a strong attachment, mitigating any selection for stiffer threads or complex yield behavior (Stanley,1970). In contrast, A. rigida and M. modiolusare often found living singly in the absence of sea grasses. To further examine the correlation between a semi-infaunal life habit and byssal thread characteristics, one could test the threads of other semi-infaunal mytilids and pinnids, e.g. Modiolus americanus Leach, which often lives in grass flats, or Pinna nobilis Linnaeus, which does not, to see whether they exhibit similar properties(Peterson and Heck, 2001).

Lucas and colleagues (Lucas et al.,1955) performed a tensile test on a single P. nobilisbyssal thread submerged in (presumably distilled) water at 20°C and generated a stress–strain curve. They report `stress' as force per linear density (grams per denier), a variable commonly used in the textile literature which unfortunately confounds volumetric density and cross-sectional area. Thus, without knowing the volumetric density of the P. nobilis thread tested, it is impossible to calculate its breaking stress as defined in the engineering and biomechanics literature. (A similar problem applies to their reported value for strain rate.) Nonetheless, a comparison of strain values is possible: the P. nobilis thread broke at an `engineering' strain of about 56%, a value close to the average`engineering' extensibility of A. rigida threads, 57%; moreover, the yield strain of the P. nobilis thread was in the same range as that of A. rigida threads, although the former yielded at only a single point whereas the latter exhibited a second yield point at a higher strain. It would be unwise to place too much weight on this comparison, however, given that it is based on a single P. nobilis thread that was likely dried– and shipped from Milan to Manchester – before being re-hydrated and strained at an unknown rate.

As shown above, C. mitis threads have mechanical properties that differ dramatically from those of other species. Its threads are not strong,stiff or tough, but they are highly extensible. The `true' strain at failure of these threads, 81%, corresponds to an `engineering' strain of 126%, by far the highest ever recorded for distal or whole byssal threads(Bell and Gosline, 1996; Lucas et al., 2002; Brazee and Carrington, 2006). These properties of C. mitis point to a possible trade-off in thread design. As with many engineered materials, it may be difficult to build a thread that is both very stiff and very extensible. This interpretation is supported by the observation that the stiffest threads are the least extensible (A. rigida, M. modiolus), whereas the highly extensible C. mitis threads have the lowest stiffness(Table 2). Assuming this trade-off, there are (at least) two possible hypotheses for the properties of C. mitis threads: (1) there has been no selection for costly strength/stiffness, and the high extensibility is a by-product; or (2) there has been selection for high extensibility, and the low strength/stiffness is a by-product. One explanation for (1) might be that high strength and stiffness are primarily important for resisting predator manipulation and wave action,which are perhaps not important factors for nestling, mobile C. mitis. On the other hand, a possible explanation for (2) is that, as with viscid spider silk, the highly extensible threads act as single-use shock absorbers, allowing C. mitis to absorb heavy currents or sudden shocks without abandoning a preferred crevice(Denny, 1976). Investigation of the thread properties of unrelated nestling bivalve species could provide evidence for or against (2). The ancestral condition for byssal thread properties is unfortunately unknown, as no one has studied the chemistry or mechanics of juvenile bivalve byssal threads.

Interestingly, the threads of A. rigida seem to share certain properties with those of each of the other two semi-infaunal species tested. Like those of G. demissa, its threads have a low yield stress, low strength and low toughness, but like those of M. modiolus, they have a high initial stiffness and a low extensibility(Table 2). Commonalities such as these are easy to explain away as being related to functional requirements:with a large number of threads and external support from the substrate,strength and toughness may be less important; and if the threads have very little give (high stiffness, low extensibility), that could stop predators from easily manipulating the animal. However, post-hocexplanations such as these do not solve the problem of why the three semi-infaunal species diverge in certain of their properties. This problem is impossible to fully address without research into the threads of other semi-infaunal mytilids and pinnids, as well as unrelated species with similar life habits.

As the data presented here demonstrate, pinnid and limid byssal threads have mechanical properties that often differ significantly from those of mytilid threads. Despite these differences, however, our data suggest a connection between the semi-infaunal life habit and certain thread properties,e.g. small diameter and double-yield behavior. A wider survey of bivalve byssal thread properties, both within and beyond the orders examined to date,would provide a wealth of information about connections between thread properties and evolutionary patterns within the Bivalvia. Moreover, with further work on byssal thread composition outside the Mytilidae, connections between microscopic molecular structures and macroscopic material properties might suggest new avenues for ongoing biomimetic research(Yu and Deming, 1998; Yamada et al., 2000; Tonegawa et al., 2004; Waite et al., 2005; Lee et al., 2006; Lee et al., 2007; Waite, 2008). Thus the comparative biomechanics of bivalve byssal threads has much to offer to both evolutionary biology and materials engineering.