The exoskeleton of the lobster Homarus americanus as an example of a smart anisotropic biological material
Patricia Romano, Helge Fabritius and Dierk Raabe
Many biological materials are composed of fibrils arranged according to well defined three-dimensional patterns. These materials often show a strong anisotropy in their properties. Their composite microstructure, comprising soft organic and sometimes also hard inorganic materials, plays a key role in enhancing the toughness, strength, and hardness of materials like arthropod cuticles by restricting crack growth and delocalizing deformation fields. An essential characteristic of biological materials is their hierarchical organization from the nanometer- to the millimeter scale. Lobster cuticle is a good example for this and a suitable model for studying these properties.
In this study the structure of untreated as well as chemically and physically treated cuticle from the exoskeleton of the American lobster (Homarus americanus) is investigated using scanning electron microscopy. Fresh samples have been chemically decalcified and deproteinated and thermally treated in order to evaluate their resistance to degradation. The results show that their structure is more complex that the commonly assumed model for arthropod cuticles. The stacked chitin-protein planes create the characteristic twisted plywood pattern found in arthropod cuticles. However, due to a well developed pore canal system these planes are not simple arrays of parallel chitin-protein fibers, but interconnected fibers forming a planar honeycomb like structure by bending around the continuous lenticellate cavities of the pore canals. The chemically and thermally treated samples show that the organic matrix retains its shape and structure despite the attack of chemical compounds or heat. It is also possible to study the distribution of the biominerals after the removal of the organic matrix. The observed residual structure gives a good impression of how the minerals (mainly calcite) are distributed inside of the polymeric network.
Many skeletal tissues found in nature are composite systems, associating organic fibrils and mineral crystals, where the fibrous matrix, first deposited, orientates the mineral nucleation and growth afterwards. In fact, skeletal tissues are ‘fiber-reinforced composites’, a system where the “whole is more important than the sum of its parts”. Separately, the mineral or the fibrils do not show useful mechanical properties, the mineral being made of brittle crystals and the polymers being supple, but the composite structure resists strong constraints, since crystal fractures stop where they meet fibrils, and those ones do not bend, because the inter-distances are fixed in the mineral []. Good examples of this are the bones of vertebrates, shells of mollusks and the cuticles of crustaceans. The cuticles of crustaceans and arthropods in general act as exoskeletons, functional units which provide mechanical support to the body of the animals, enable movement through the formation of joints and attachment sites for muscles and provide protection against predators. In order to grow, these exoskeletons have to be shed and replaced by a new, larger one regularly. Our model organism, the American lobster Homarus americanus is a decapod crustacean whose relatively large size and good availability makes it an excellent model for studying its exoskeleton.
The cuticle of the Crustacea comprises two main layers, the epicuticle and the procuticle. The epicuticle is a thin waxy layer which acts as a diffusion barrier to the environment. The procuticle is further divided into an exocuticle and an endocuticle which are chiefly designed to resist mechanical loads. The organic matrix of crustacean cuticles is secreted by a single layered epithelium and is composed of chitin associated with proteins where eventually the calcite crystals will grow [, , ]. Chitin is a biopolymer whose ideal structure is a linear polysaccharide of b-(1,4)-2-acetamido-2-deoxy-D-glucopyranose, where all the residues are comprised entirely of N-acetyl-glucosamine, i.e. fully acetylated. However, in nature, the biopolymer exists as a co-polymer together with its deacetylated derivative, chitosan. When the number of acetamido groups is more than 50% the biopolymer is termed chitin [].
In many crustacean groups the hard parts of the exo- and endocuticle are mineralized, essentially by precipitation of calcium carbonate and to a lesser extent, of amorphous calcium carbonate into the twisted lamellar structure of the chitin-protein cholesteric matrix [, , , , , , , ]. The cuticle of the crustaceans is a complex structure where initiation, growth, and orientation of calcite domains are controlled by organic intra- and extracellular elements. This material provides a wide range of mechanical properties through local variations in composition and structure [4, , , , , ]. Characteristic for such composite materials is their strong microstructural hierarchy. Hierarchical structures of the organic matrix are visible already in sections imaged by optical microscopy (in the range of 10 to 50 µm) [].
The smallest subunits in the lobster cuticle are long chitin molecules. 18 to 25 chitin molecules then form nanofibrils with a length of approximately 300 nm and a diameter of about 2-5 nm which are wrapped by proteins. In the lobster the chitin chains are arranged in an antiparallel fashion forming a-chitin which is the most abundant of the three crystalline modifications of chitin occurring in nature. Crystallographic texture analysis of the chitin in lobster cuticle using X-ray wide angle Bragg diffraction has shown that a large fraction of the a-chitin lattice cells is arranged with their longest axis parallel to the normal of the surface of the cuticle. The protein-wrapped chitin nanofibrils cluster to form 50 to 250 nm thick, long chitin protein fibers which are arranged parallel to each other and form horizontal planes. These chitin protein planes are stacked and the longitudinal axis of the fibers in superimposed layers rotates around the normal axis of the cuticle, creating a twisted plywood or Bouligand structure. A stack of fiber layers that completed a 180° rotation is referred to as one Bouligand or twisted plywood layer. Figure 1a shows that the twisted plywood structure of the exocuticle is more densely packed than that of the endocuticle. The lobster cuticle additionally features an extremely well developed pore canal system whose numerous canals penetrate it perpendicular to the surface (Fig. 1b). These canals contain long, soft and probably flexible tubes which play an important role in the transport of ions during the mineralization of the new exoskeleton after the animals molt []. The parallel fibers of the chitin protein planes meander around the lenticellate cavities of the pore canals, a formation which generates a structure that can be described as a twisted honeycomb [].
Figure 1. Microstructure of lobster cuticle. a) SEM micrograph showing a cross section through the three layered cuticle. The different stacking density of the twisted plywood layers (tp) in the exo- and endocuticle can be clearly seen. b) SEM micrograph of obliquely fractured endocuticle displaying several superimposed twisted plywood layers (tp) showing the typical honeycomb-like structure. The arrows indicate the pore canals.
The combination of different types and quantities of materials and their interaction on different scales leads to a pronounced structural, topological and crystallographic directionality in the cuticle material relative to applied mechanical loads. This close connection between matter and construction is referred to as smart anisotropy and is a typical principle in biological materials. However, these internal parameters of the biological materials are not the only ones to define their mechanical properties. They are also influenced by physiological parameters like the environmental conditions the organism lived in (i.e. temperature, pH, salinity, pollution) and its physiological state (i.e. molting stage, nutrition, diseases) together with artificial parameters like storage conditions of the samples, grade of hydration, state of decomposition. All these parameters influence each other and their combination finally defines the measurable mechanical properties of biological materials. The aim of this study is to document the changes in the microstructure of lobster cuticle where selected ingredients have been removed either chemically or by heat treatment using scanning electron microscopy.
Materials and methods
The exoskeleton of an American lobster (Homarus americanus) in intermolt stage acquired from a local seafood supplier was used. Molting stage was determined by the presence of the basal membrane on the inner surface of the cuticle.
Both chemical and thermal treatment were conducted on specimens that were dissected mainly from the chelipeds. Chemical attacks were performed using NaOH (1M, 1week) to remove the protein structure, EDTA (0.15M, 2 weeks) to remove the biominerals and a combination of EDTA (0.15M, 2 weeks) followed by NaOH (1M, 1week) to obtain only the chitin network.
TGA experiments were carried out to determine the decomposition temperature of the different components of the native exoskeleton. The heating of the samples was done up to 1200°C, 5°C/min, in an inert atmosphere using a Setaram SETSYS 16 thermobalance. Thermal treatments were done in a conventional furnace under air.
The obtained samples were air-dried and cleaved either perpendicular to the cuticle surface in order to expose cross section or parallel to the surface to expose the endocuticle. They were then sputter-coated with 5 nm of gold and imagined using a high resolution scanning electron microscope (Zeiss Gemini 1540 XB). Contrast and brightness of the digital images were adjusted where necessary using Adobe Photoshop CS2 (Adobe Inc.).
2.1 Chemically treated material (decalcified, deproteinated and both)
High resolution scanning electron micrographs of untreated lobster cuticle (Fig. 2a) show its characteristic microstructure formed by parallel oriented fibers with diameters between 25 and 50 nm and blocky appearance. The walls of the honeycomb-like structure formed by the stacked fiber layers appear homogeneous and solid. All over the investigated area broken fibers with square ends can be observed. This leads to a rough surface of the sample with hardly any continuity in the fibers visible. In cuticle decalcified with EDTA (Fig. 2b) individual fibers are hard to discern. The honeycomb-like structure around the pore canals is well defined but the walls formed by the stacked fiber bundles appear very smooth. The walls of adjacent pore canals do not merge homogeneously but show separated fiber bundles contacting each other in certain areas and then separating again to form another pore canal. The only occasionally observed interrupted fibrous structures have rounded ends. Samples which were deproteinated using NaOH (Fig. 2c) display a well defined honeycomb-like structure with visible contact areas between the walls of individual pore canals. Single fibers can not be discerned. The whole structure is covered with small spherical structures with diameters of 20 to 50 nm. The structure of decalcified and deproteinated cuticle (Fig. 2d) closely resembles that of decalcified cuticle, but the walls of the pore canals are slightly thinner and appear more fragile. The honeycomb-like structure appears also very smooth and slightly less well defined.
Figure 2. Microstructure of lobster cuticle. (a) Untreated cuticle, (b) decalcified cuticle (EDTA, 0.15M), (c) deproteined cuticle (NaOH, 1M) and (d) decalcified and deproteined cuticle (EDTA, 0.15M + NaOH, 1M).
2.2 Thermally treated material
Figure 3. Thermogravimetrical analysis of the untreated cuticle.
The thermogravimetrical analysis (TGA) diagram (Fig. 3) shows that at temperatures between 50 and 270°C water and proteins are removed from the untreated cuticle. The decomposition of the chitin starts at about 270°C. At temperatures between 500 and 600°C the chitin is completely removed and above 650°C the decarboxylation of the CaCO3 starts.
In untreated cuticle heated up to 370°C the honeycomb-like structure is well preserved (Fig. 4a). The individual fibers have a blocky appearance and are broken displaying sharp, square fractures. They form solid shaped honeycomb walls which appear slightly porous due to numerous small cavities between them. At 750°C the honeycomb structure is not nearly as well preserved, although the shape and the dimensions of the lenticellate pore canals can still be distinguished (Fig. 4b). The wall structures are much thinner and consist of spherical particles with diameters between 120 and 150 nm.
Figure 4. Microstructure of heat-treated cuticle. (a) 370°C heated and (b) 750°C heated.
Samples with minerals and proteins chemically removed before heat treatment and subsequently heated to 220°C show a pronounced fibrous structure. The holes of the pore canal system appear narrower and the planar parallel fibers are very long and continuous showing no signs of breakage (Fig. 5a). At higher magnifications the parallel arrangement of individual fibers in superimposed layers becomes much more obvious than in untreated or chemically treated material (Fig. 5b). The fibers have diameters of about 30 nm and adjacent fibers seem to be connected to each other through very small fibrilar structures.
Figure 5. Microstructure of purified chitin heat-treated at 220°C. (a) overview of the structure parallel to the cuticle surface and (b) detail image of a pore canal.
Our results show that both chemical and thermal treatments have severe impact on the microstructure of lobster cuticle. During decalcification with EDTA the CaCO3 is gently removed and the remaining structure is formed by the chitin-protein fibers, which themselves are composed of protein-wrapped nanofibrils. The smoothness of the structure without CaCO3 in contrast to the blocky appearance of the untreated cuticle indicates that the biominerals were indeed located inside of the chitin proteins fibers. The absence of fractured fibers shows that the demineralized material looses the brittleness present in the natural state. The small spherical particles appearing in the deproteinated sample are most likely biomineral crystals which become visible after the removal of the proteins. This indicates that the proteins stabilize not only the structure but the crystalline state by binding the mineral together with the chitin polymer. At the present state it is not possible to determine whether the particles represent original crystallites formerly enclosed in the fibers or some recrystallization process has taken place due to the chemical attack. After removal of both proteins and minerals the residual chitin fibers retain their shape. Therefore it can be assumed that even though the proteins are responsible for the fiber shape of the chitin polymer, the removal of them does not make the polymer loose the honeycomb-like structure. This is most likely due to the formation of new bondings between the now unwrapped chitin nanofibrils.
In the 370° C heated sample the proteins and the major part of the chitin polymer are eliminated. The removal was done at slow heating rate in order to avoid structural damage. The residual structure consists of the biominerals and closely resembles the natural state of the cuticle. The slightly higher porosity originates from the evaporated organic components. Recrystallization cannot take place due to the lack of solvents. At 750°C the decarboxylation of the CaCO3 has already started, leading to a loss of material. The reaction that takes place produces CaO and carbon dioxide. Therefore the residual structure contains both calcium carbonate and calcium oxide. It is possible that due to time and temperature, a diffusion process has occurred which could explain the presence of the relatively large spherical particles, a phenomenon that resembles sintering processes.
The chemically purified chitin was heat treated at 220°C, just before its thermal decomposition starts. The structure preservation is remarkably good and the stacked rotating planes of parallel fibers which represent the basic construction principle of the cuticle become clear. The heat has probably helped the pore canals to constrict almost to the point of collapsing. Individual chitin fibers become easier to discern, probably due to the temperature eliminating all possible organic residues left by the chemical attack.
Chemical treatment is already used for purifying chitin obtained from nature. Attempts have been done to produce chitin networks with defined porosity []. The lobster cuticle has a well developed pore canals system that provides continuous pores with a defined size of about 2µm. Our investigation shows that it is possible to treat the cuticle to obtain a pure chitin matrix which retains the porosity of the original material. Materials with directional porosity and defined pore size, which can be produced through a simple process, could be made suitable for example for medical applications. The complete absence of proteins that were removed by heat treatment is an additional advantage because they are known to be source of allergic reactions.
We have compared the microstructure of cuticle from the exoskeleton of the American lobster Homarus americanus in its natural state with cuticle that was chemically decalcified and deproteinated as well as cuticle that was heat treated. The results show that both types of treatment have little influence on the overall structure. However, on the scale of the individual fibers the structure is altered significantly through the selective removal of the different cuticle components. Through chemical etching and subsequent heat treatment of the native material, a defined porous matrix of highly purified chitin can be obtained. This product could be regarded as an example for the development of innovative medical applications.
The authors would like to thank Dr. Frank Stein (Max Planck Institute for Iron Research) for the help with the thermal experiments and the Gottfried-Wilhelm-Leibniz program of the Deutsche Forschungsgemeinschaft (German Research Foundation) for the financial support.
- Y. Bouligand. C. R. Paleovol., 2004, 3:617-628.
- Y. Bouligand. Tissue & Cell, 1972, 4(2):189-217.
- M. M. Guiraud-Guille. Current Opinion in Solid State & Material Science, 1998, 3:221-227.
- J.F.V.Vincent. Structural Biomaterials. Princeton University Press. 1990.
- E. Khor. Chitin: Fulfilling a biomaterials promise. Elsevier Science. 2001.
- H.A. Lowenstam. Science, 1981, 211:1126-1131.
- S. Mann, J. Webb and R.J.P. Williams. On biomineralization. VCH Press. 1989.
- H.A. Lowenstam, S. Weiner. On Biomineralization. Oxford University Press. 1989.
- S. Weiner and L. Addadi. J. Mater. Chem., 1997, 7:689-702.
- S. Mann. J. Mater. Chem., 1995, 5: 935-946.
- F. Manoli, S. Koutsopoulos and E. Dalas. J. Crystal Growth, 1997, 182:116-124.
- G. Falini, S. Albech, S. Weiner and L. Addadi. Science, 1996, 271:67-69.
- G. Luquet, F. Marin. C. R. Palevol., 2004, 3:515-534.
- D. F. Travis. Ann. N.Y. Acad. Sci., 1963, 109:177-245.
- N. F. Hadley. Sci. Am., 1986, 255:98-l06.
- J. F. V. Vincent. Composites: Part A; 2002, 33:1311–1315.
- A.C. Neville. Biology of Fibrous Composites. Cambridge University Press. 1993.
- J. F. V. Vincent, U. G. K. Wegst. Arthropod structure and development, 2004, 33:187–199.
- M.M. Guiraud-Guille, E. Belamie, G. Mosser. C. R. Paleovol., 2004, 3:503-513.
- J. N. Cameron. J. exp. Biol., 1989, 143:285-304.
- D. Raabe, P. Romano, C. Sachs, A. Al-Sawalmih, H.-G. Brokmeier, S.-B. Yi, G. Servos, H. G. Hartwig, J. Crystal Growth, 2005, 283:1-7.
- K. S. Chow, E. Khor. Biomacromolecules, 2000, 1:61-67.