216
Biomacromolecules 2008, 9, 216–221
Structural Disorder in Silk Proteins Reveals the Emergence of Elastomericity Cedric Dicko,*,† David Porter,† Jason Bond,‡ John M. Kenney,§ and Fritz Vollrath† Department of Zoology, Oxford University, OX1 3PS, U.K., Department of Biology, East Carolina University, Greenville, North Carolina 27858, and Department of Physics, East Carolina University, Greenville, North Carolina 27858 Received September 25, 2007
Spider silks combine basic amino acids into strong and versatile fibers where the quality of the elastomer is attributed to the interaction of highly adapted protein motifs with a complex spinning process. The evaluation, however, of the interaction has remained elusive. Here, we present a novel analysis to study silk formation by examining the secondary structures of silk proteins in solution. Using the seven different silks of Nephila edulis as a benchmark system, we define a structural disorder parameter (the folding index, γ). We found that γ is highly correlated with the ratio of glycine present. Testing the correlation between glycine content and the folding index (γ) against a selected range of silks, we find quantitatively that, in order to achieve specialization with changes in mechanical performance, the spider’s silks require higher structural flexibility at the expense of reduced stability and consequently an increased conversion-energy cost. Taken together, our biophysical and evolutionary findings reveal that silk elastomericity evolved in tandem with specializations in the process of silk spinning.
Introduction Elastomeric proteins are an important group of structural macromolecules giving biological systems toughness and flexibility. A recent study showed that structural disorder and associated hydration are critical features of protein elastomericity.1 Molecular dynamic simulations of key protein sequences have demonstrated that the crucial conformational hindrance is provided by proline and often coupled with high chain flexibility provided by glycine residues.1 Yet the origin of elastomericity is only partially explained by the functional interplay between proline and glycine, as we shall show using spider silk as our test case. Spider evolution and cladogenesis underpin the animal’s everincreasing investment in silk production and allow us to correlate spider diversification and silk properties.2,3 Interestingly, the structure–function relationship in spider silks is coupled with drastic changes in the design of the animal’s spinning apparatus (see Table 1). The aciniform glands (simple spherical or pear-shaped structures, see Table 1) are accepted candidates for the most ancestral spinning device making protective silks to shelter eggs and to line burrows.4 More advanced ecological functions (such as capture webs) led to the emergence of more complex glands and silks (see Table 1). The major ampullate gland (dragline and orb-web radial threads) is a case in point, where composition, chemistry, and processing together deliver a highly specialized silk with properties that excite even modern fiber manufacturers. Unlike most other silk-producing animals such as insects, spiders can produce up to nine different types of silks and have the ability to control key aspects of silk production over a wide range of conditions,5 controlling not only (i) silk protein composition and (ii) storage conditions but, crucially, also (iii) * Corresponding author:
[email protected]. † Department of Zoology, Oxford University. ‡ Department of Biology, East Carolina University. § Department of Physics, East Carolina University.
fiber extrusion conditions. The mechanisms by which spiders optimize and adjust silk properties and function are beginning to provide valuable information on structure–function relationships of silk proteins. Studies of the interplay between silks and spinning processes6 demonstrate a strong correlation between strength and relative elasticity with specific molecular architecture.7–9 Genetic analysis10–12 and amino acid composition13 of silk protein composition demonstrate a strong bias toward amino acid content dominated by alanine and glycine residues (Table 1). Furthermore, silk protein sequences show preferential motif arrangements (polyalanine crystalline domains and glycine-rich amorphous domains), which can be correlated to macroscopic properties of silk such as elasticity and strength.11,14 Molecular Basis for Elastomericity: Maintaining Disorder. Advances in molecular modeling of silk fibers are driving a major shift in our understanding of silk.15 The features controlling the mechanical behavior of the material are the ratios of the ordered and disordered fractions as well as the degree of hydration.15 Both parameters are interlinked and controlled by the animal during the extrusion process.16 The “silk’s” ability to implement small chemical changes by modifying hydrogen bonding sites during processing affects the properties of the bulk material significantly and has led to its status as an archetypal model elastomer.17 As such it is interesting to examine our knowledge of silk proteins in solution in the light of the Rauscher1 observations that a fundamental requirement for elastomeric domains to remain disordered in solution is controlled by two major sequence determinants: proline and glycine content. This suggests that elastomeric assembly is, first, assisted by proline residues that conformationally restrict the main chain and, second, helped by glycine residues that modify backbone hydration. Put together, the variation in proline and glycine content in elastomeric proteins and amyloids gives a composition threshold by which disorder is maintained and by which functional elastomericity becomes possible.
10.1021/bm701069y CCC: $40.75 2008 American Chemical Society Published on Web 12/14/2007
Emergence of Elastomericity in Silks
Biomacromolecules, Vol. 9, No. 1, 2008 217
Table 1. Spinnerets, Glands, Composition, and CD Spectra of the Studied Spiders
SSC (small side chains) ) Ala + Gly + Ser.b PC (polar chains) )Asp + Thr + Glu + Ser + His + Lys + Tyr + Arg.c Midpoint transition temperature and folding index γ.d Shapes redrawn from refs 26-32. Scale bars 100 µm (Acinous, tubular, Median) and 1 mm (Ma, Mi, Flag, Cyl, Bmx).e Temperature transition determined by differential scanning calorimetry from ref 37. a
Interestingly, depending on the environment which they “inhabit”, glycine residues can promote either highly ordered or rather disordered structures.1,18–20 The measure of this glycine ambivalence is based on an explicit approach of folding that
properly accounts for the chemical nature of the backbone protein.21–25 The important effects of hydration both during and after chain folding coupled with the ambivalent nature of glycine and its interplay with proline (maintaining disorder and promot-
218 Biomacromolecules, Vol. 9, No. 1, 2008
ing elastomeric properties) point toward silk proteins as ideal study models for detailed investigations into the structure–function relationship of natural elastomers.
Dicko et al. spectra; thus, the heat-induced changes were nonreversible. Fractional changes in conformation were plotted against temperature using the equation
fu )
Materials and Methods Species Sampling. Four exemplar spider species were examined. Species selected on the basis of phylogenetic position were taken from basal and derived branches in the Mygalomorphae (Antrodiaetus unicolor (Antrodiaetidae) and Aphonopelma chalcodes (Theraphosidae), respectively) and from basal and derived branches in the araneomorphae (Kukulcania hibernalis (Filistatidae, crevice weaver)) and the benchmark (Nephila edulis (Tetragnathidae, golden orb spider)).26–32 Comparative data for the Lepidoptera were taken for Bombyx mori (Bombycidae). Henceforth the spiders and the insect will be referred respectively as A, T, F, N, and bmx followed by the corresponding gland name. Sample Preparation. Mature female “Golden Silk” spider, Nephila edulis (Tetragnathidae), samples were prepared according to ref 33. A similar procedure for mature Antrodiaetus (A) spiders (male and females), mature Aphonopelma (T) spiders (male and female), and Kukulkania hibernalis (male and females) was performed to obtain: A-Acinous (A), T-Acinous/tubular (T), F-Major Ampullate (F-Ma), and F-Acinous (F-Ac) solutions. Note that to obtain enough samples the glands from two or three conspecific spiders were pooled. Data for Nephila edulis major ampullate, minor ampullate, flagelliform, and cylindriform was taken from the literature.33 Circular Dichroism (CD). Two machines were used to collect the data: a Jasco model J810 (East Carolina University, NC) and synchrotron radiation based circular dichroism (SRCD, Aarhus University, Denmark). Samples from A, T, and F spiders were collected on the Jasco, whereas the Nephila samples were collected on SRCD.33 Bombyx mori data were collected on SRCD and compared with literature.34,35 Folding Index. The folding index is an extension of our previous interpretation of silk protein CD spectra.36 The folding index, previously known ambiguously as the r value, is simply the ratio of two transitions typical of the amide group measured at 20 °C, the π-π* and the n-π* transitions. The n-π* transition is characterized by a shoulder at around 220 nm and the π-π* transition by two bands (exciton coupling) at around 190 and 208 nm. In this context the folding index, γ, measures the ratio of folded structures (from the π-π* transition) to the hydrated sections (from the n-π* nonbonding transition) of the silk protein backbone. We found that γ provides a good measure of structural order/ disorder of silk proteins in solution. Typically a folded structure will have a γ value above 0.5 whereas a partially folded or disordered structure will have γ value below 0.5. It is important to note that the combined effects of conformation, hydration, and side chains will shift the amide band position and intensity. Furthermore the concentration of silk protein native extract is not easily accessible; it is therefore more reliable to use a self-normalizing index such as the folding index. Full size spectra are available in Supporting Information. Temperature-Induced β-Sheet Conversion. Samples were loaded into closed cell of 1 mm path length (Suprasil quartz-Hellma 124-QS) then scanned from 20 to 90 °C, with a 5 °C step size. At each step, the sample was allowed to equilibrate for 10 min. On the Jasco the CD scans at each temperature were recorded at 200 nm/min and 10 accumulations. On the SRCD each scan was recorded three times with a 6 s reading time. The experiments were repeated at least three times with different spider samples. Note, that even though two different CD spectrometers were used in this study, the temperature-induced changes were made using cells of the same material and path length, the temperature was regulated by devices using the same mechanism (i.e., pelletier), and the volumes of the material converted were similar. At the end of each temperature melt, the system was slowly cooled to 20 °C. In all cases we did not observe a recovery of the starting
θ - θmin θmax - θmin
(1)
where θ is the ellipticity at a temperature T and θmax and θmin represent the ellipticities of the final and initial states, respectively. Because the temperature effects are not reversible, no quantitative thermodynamics of the change could be obtained. The transition midpoint Tm, independent of the model, was determined by fitting the fractions fu with a simple Boltzmann sigmoid. Note that the conversion temperature of Bombyx mori was taken from Tanaka et al.37 Discriminant Analysis (DA). Linear DA (LDA) is a multivariate method that predicts classification variables based on a known continuous response.38 LDA is also closely related to principal component analysis (PCA) and Factor Analysis in that both look for linear combinations of variables which best explain the data. LDA explicitly attempts to model the difference between the classes of data. In our particular case the responses are the amino acid composition (continuous) and the predictors the amino acids (category). Without further assumption the data are classified and the results plotted against two canonical vectors (or principal components) that explain most of the data variation. What one evaluates then is how the data group (type of silks) and which predictor (amino acids) explains most of separation between groups. The central plot in the DA graph identifies the amino acids and their strength along each of canonical axis to explain separation between groups. Data Processing. The analysis of the folding index γ correlation with the amino acids of Nephila edulis silk proteins was performed as follows: The amino acid compositions were measured on a truncated scale (0-100%) and, therefore, by definition not normally distributed.39 An [Arcsine (square-root ())] transformation was applied, after which the amino acid composition did not deviate from normality (Shapiro Wilks’ W; p < 0.05). The following step, a multiple regression of the folding index against all the amino acids, was used to eliminate collinear amino acids. The p value (at the 95% confidence level is p ) 0.05 to reject the null hypothesis) was changed using Bonferroni adjustment to correct for any regression being significant by chance. The corrected p value was 0.0011. This resulted in threonine, alanine, proline, tyrosine, valine, methionine, cysteine, isoleucine, leucine, phenylalanine, and lysine being collinear and asparagine/aspartic acid (Asp), glutamine/ glutamic acid (Glu), serine (Ser), glycine (Gly), and arginine (Arg) being independent. To estimate the best predictor or combination of predictors to explain the variation in folding index γ, we used a mixed stepwise (forward and backward) regression of the folding index γ against Asp, Glu, Ser, Gly, and Arg and the most significant collinear predictor from the multiple regression, namely, threonine (Thr). We found as a result that glycine composition alone explained significantly 67% of the variation in folding index (seven predictors, n ) 38, S ) 0.124, adjusted R2 ) 66%, p < 0.000). Note that the next combination to explain significantly the variation of the folding index was Gly + Glu with an R2 of 77%. The addition of Glu to the model improved the fit by only 11% (seven predictors, n ) 38, S ) 0.104, adjusted R2 ) 77%, p < 0.000), we therefore considered that glycine alone was enough to interpret the variation in folding index γ with amino acid composition. Finally, after selecting glycine as the main predictor, the regression was fully tested using a general linear model (GLM) analysis, n ) 36, F ) 98.11, p < 0.000, adjusted R2 ) 67%. The data processing and analysis were performed using Minitab v14 statistical package.
Results and Discussion Amino Acid Classification Bias in Silk Proteins. In order to quantify the amino acid (AA) composition of the different
Emergence of Elastomericity in Silks
Biomacromolecules, Vol. 9, No. 1, 2008 219
Figure 1. Discriminant analysis. Canonical plot showing the different silk glands according to their amino acid composition. Four factors, amino acids, sufficed to significantly group the observation. Among these four amino acids serine and glycine discriminate the most (Manova test statistics, N ) 36, p < 000). The canonical components 1 and 2 are equivalent to principal components with the exception that they are optimized for class separability.
Figure 2. CD spectra at 20 °C of a typical silk protein. The folding index γ is defined as the ratio of the two minima. The folding index γ is indicative of the degree of unfolded structure present in solution at 20 °C. Low γ correspond to predominantly unfolded structures and high γ to folded structures (e.g., a γ of 1 is typical of the fully helical structure of myoglobin).
spider proteins (spidroins), we analyzed the amino acid composition of the silk feedstock proteins of all glands taken from N Nephila edulis spiders and performed a multivariate classification (discriminant analysis) on their fractional AA content. The grouping of the data sets was determined principally by two amino acid predictors, namely, glycine (Gly) and serine (Ser) (Figure 1). The dependence on the Gly and Ser predictors led us to the conclusion that the casual relationship between AA composition and structural behavior in Nephila silks could depend on two structural factors: (i) molecular chain flexibility (Gly) and available hydrogen bonding (Ser). Definition of the Folding Index γ. To be able to compare spiders’ precursor silks further, we used circular dichroism (CD) spectroscopy to determine the characteristics of the silk protein folds and thermal stability.36 We measured the temperatureinduced structural transition of freshly extracted silk protein feedstock and determined their folding index, γ (see Figure 2, Table 1). γ is defined as
γ)
θ1 θ2
(2)
where θ1 and θ2 are the measured ellipticities (mdeg) of the two minima found respectively around 220 and 200 nm in each of the silks’ solution CD spectra (see Figure 2 and Table 1). The folding index γ represents a normalized measure of disorder at 20 °C, typically a folded helical structure gives a
Figure 3. Inverse correlation between the folding index γ and glycine content. The regression was calculated solely using the Nephila edulis seven glands and analyzed by a general linear model (GLM), n ) 36, F ) 98.11, p < 0.000, adjusted R2 ) 67%. The gray area is the 95% confidence interval. The correlation was tested with T, A, F-Ma, F-Ac, and bmx silks. We find that (i) the model quantitatively explained the structure–function relationship found in the selected silks and (ii) to achieve specialization and performance, silks require higher structural flexibility at the expense of reduced stability and increased conversion energy. Legend: Nephila edulis (Tetragnathidae) major ampullate (N-Ma), minor ampullate (N-Mi), flagelliform (N-Flag), cylindriform (N-Cyl), aciniform (N-Ac), pyriform (N-Pyr), median (NMed), Kukulkania hibernalis (Filistatidae) major ampullate (F-Ma), and acinous (F-Ac). Antrodiaetus unicolor (Antrodiaetidae) single type glands (A), and Aphonopelma chalcoldes (Theraphosidae) acinous (T). Bombyx mori (Bombicidae) (bmx).
value of γ around 1, whereas values below 0.5 correspond to partially and fully unfolded structures. Elastomericity Depends on Spinning Conditions. Combining the structural (γ) and composition data, we found that only the fraction of glycine residues (Gly) showed a strong inverse relationship with the folding index (see Figure 3). No other amino acid found in the many silks we examined provided a comparable, significant correlation with γ (see Materials and Methods). Overlaying the known function and evolution of the silks (Table 1) onto the γ/Gly plot led us to the hypothesis that one can predict silk protein specialization, relatedness, and the prerequisites for fiber formation by correlating the protein folding index (γ) with the protein’s structural flexibility in solution (i.e., its glycine content). We tested this hypothesis on other silks, i.e., spidroins and fibroins taken from (i) another (but only distantly related) derived spider, (ii) two ancestral spiders, and (iii) an unrelated insect. The data from the analysis show that the two ancestral silk spidroins of the Antrodiaetus and Aphonopelma spiders lay on the γ/Gly plot close to the cylindriform, pyriform, median, and acinous silks of our advanced Nephila spider. This correlated very well with similarities in overall gland morphology as well as the perceived evolutionary basic functions of these silks (see Table 1). The position of the insect silk fibroins on the plot indicates that, regardless of its evolutionary divergence with spiders, the highly specialized (and complex) silk of the Bombyx silkworm requires high chain flexibility prior to spinning. Interestingly the silks from the derived cribellate Kukulcania spider appear at an intermediate position-perhaps in keeping with its evolutionary position. When plotting the amino acid composition of our silks against the proline-glycine (PG) peptide sequences studied by Rauscher,1 we found that most of the “ancestral” silks were “amyloidic” whereas the more derived silks were elastomeric. Indeed, an interpretation of the results from Figure 3 in the light of elastomeric self-assembly classification1 suggests that elastomericity in silks correlates with gland and silk specialization. This qualitative deduction further suggests the possibility of
220 Biomacromolecules, Vol. 9, No. 1, 2008
elastomeric emergence in silks from an amyloid-like “ancestor” protein construct, which would be a hypothesis supported by other observations.40,41
Conclusion Overall, we found a highly significant causal relationship between the concentration of glycine residues and the structural behavior of all the silks produced by a “modern” orb weaver such as Nephila edulis. This discovery suggest that a simple parameter set of a silk’s molecular structure might provide novel, intertaxon insights into silk evolution as well as, ultimately, resolve the interaction of protein sequence, spinning processes, and mechanical properties. Most significantly, in support of the elastomeric self-assembly model by Rauscher1 we conclude that the emergence of elastomericity in silks correlates with increased functional specialization and higher glycine content. The absence of proline effects on the observed classification suggests a possible dichotomization of silk processing, with glycine controlling the soluble precursor assembly and proline governing the solid fibers behavior.42 Undoubtedly, spiders’ silks allow the study of protein folding and evolved structure–function relationships on a number of levels. However important this may be academically, there is also an interesting commercial side to such insights. Clearly, only with a fuller understanding of these complex molecular inter-relationships will we be able to design synthetic silks and spin them with the spider’s benign production methods. And success in such a truly biomimetic endeavor will be the ultimate test of our understanding of protein folding in what we consider to be an archetypal elastomer. Acknowledgment. C.D. and F.V. acknowledge the support of the European Commission (ARI-RII3-2004-506008, Grants G5RD-CT-2002-00738 and EC-MTKD-CT-2004-014533), the Royal Society (Grant 2004/R2-CH) and the AFOSR of the United States of America (Grant F49620-03-1-0111) for funding. C.D. is supported by St Edmund Hall Oxford and EPSRC (Life science interface fellow EP/E039715/1). J.B. acknowledges the NSF (DEB 0108575). We thank Professor A. Rodger, Drs. M. Wise, D. Knight, S. V. Hoffmann, and A. Terry for their critical comments on the manuscript. Drs. J. Ferwerda and A. Weir as well as the Statistics Department of Oxford University kindly provided advise for the statistical analysis. Supporting Information Available. CD spectra. This material is available free of charge via the Internet at http:// pubs.acs.org.
References and Notes (1) Rauscher, S.; Baud, S.; Miao, M.; Keeley, F. W.; Pomes, R. Proline and Glycine Control Protein Self-Organization into Elastomeric or Amyloid Fibrils. Structure 2006, 14 (11), 1667–1676. (2) Bond, J. E.; Opell, B. D. Testing adaptive radiation and key innovation hypotheses in spiders. EVolution 1998, 52 (2), 403–414. (3) Craig, C. L. Spiderwebs and silks: tracing eVolution from molecules to genes to phenotypes; Oxford University Press: New York, 2003. (4) Schultz, J. W. The origin of the spinning apparatus in spiders. Biol. ReV. 1987, 62, 89–113. (5) Vollrath, F.; Knight, D. P. Liquid crystalline spinning of spider silk. Nature 2001, 410 (6828), 541–548. (6) Vollrath, F. Coevolution of behaviour and material in the spider’s web. In Biomechanics in Animal BehaViour; Domenici, P., Ed.; Bios: Oxford, 2000. (7) Gosline, J.; Lillie, M.; Carrington, E.; Guerette, P.; Ortlepp, C.; Savage, K. Elastic proteins: biological roles and mechanical properties. Philos. Trans. R. Soc. London, Ser. B 2002, 357 (1418), 121–132.
Dicko et al. (8) Riekel, C.; Madsen, B.; Knight, D.; Vollrath, F. X-ray diffraction on spider silk during controlled extrusion under a synchrotron radiation X-ray beam. Biomacromolecules 2000, 1 (4), 622–626. (9) Beek, J. D. v.; Hess, S.; Vollrath, F.; Meier, B. H. The molecular structure of spider dragline silk: Folding and orientation of the protein backbone. Proc. Natl. Acad. Sci. U.S.A. 2002, 99 (16), 10266–10271. (10) Xu, M.; Lewis, R. V. Structure of a protein superfiber spider dragline silk. Proc. Natl. Acad. Sci. U.S.A. 1990, 87 (18), 7120–7124. (11) Hayashi, C. Y.; Shipley, N. H.; Lewis, R. V. Hypotheses that correlate the sequence, structure, and mechanical properties of spider silk proteins. Int. J. Biol. Macromol. 1999, 24 (2–3), 271–275. (12) Guerette, P. A.; Ginzinger, D. G.; Weber, B. H. F.; Gosline, J. M. Silk properties determined by gland-specific expression of a spider fibroin gene family. Science 1996, 272 (5258), 112–115. (13) Andersen, S. O. Amino acid composition of spider silks. Comp. Biochem. Physiol. 1970, 35, 705–711. (14) Bini, E.; Knight, D. P.; Kaplan, D. L. Mapping domain structures in silks from insects and spiders related to protein assembly. J. Mol. Biol. 2004, 335 (1), 27–40. (15) Porter, D.; Vollrath, F.; Shao, J. Z. Predicting the mechanical properties of spider silk as a model nanostructured polymer. Eur. Phys. J. E 2005, 16 (2), 199–206. (16) Holland, C. A.; Terry, A. E.; Porter, D.; Vollrath, F. Comparing the rheology of native spider and silkworm spinning dope. Nat. Mater. 2006, 5, 870–874. (17) Vollrath, F.; Porter, D. Spider silk as an archetypal protein elastomer. Soft Matter 2006, 2, 377–385. (18) Chakrabartty, A.; Schellman, J. A.; Baldwin, R. L. Large differences in the helix propensities of alanine and glycine. Nature 1991, 351 (6327), 586–8. (19) Finkelstein, A. V.; Ptitsyn, O. B. Protein Physics; Academic Press: San Diego, CA, 2002; p 354. (20) Serrano, L.; Neira, J. L.; Sancho, J.; Fersht, A. R. Effect of alanine versus glycine in alpha-helices on protein stability. Nature 1992, 356 (6368), 453–5. (21) Uversky, V. N. Protein folding revisited. A polypeptide chain at the folding- misfolding-nonfolding cross-roads: which way to go. Cell. Mol. Life Sci. 2003, 60 (9), 1852–1871. (22) Yang, A. S.; Hitz, B.; Honig, B. Free energy determinants of secondary structure formation: III. beta-turns and their role in protein folding. J. Mol. Biol. 1996, 259 (4), 873–82. (23) Yang, A. S.; Honig, B. Free energy determinants of secondary structure formation: I. alpha-Helices. J. Mol. Biol. 1995, 252 (3), 351–65. (24) Yang, A. S.; Honig, B. Free energy determinants of secondary structure formation: II. Antiparallel beta-sheets. J. Mol. Biol. 1995, 252 (3), 366–76. (25) Kim, W.; Conticello, V. P. Protein engineering methods for investigation of structure-function relationships in protein-based elastomeric materials. Polym. ReV. 2007, 47 (1), 93–119. (26) Kovoor, J. Comparative Structure and Histochemistry of silk-producing organs in Arachnids. In Ecophysiology of Spiders; Nentwig, W., Ed.; Springer: Berlin-Heidelberg-New York, 1987; pp 160–186. (27) Palmer, J.; Coyle, F.; Harrison, F. Structure and cytochemistry of silk glands of the mygalomorph spider Antrodiaetus unicolor (Araneae, Antrodiaetidae). J. Morphol. 1982, 174, 269–274. (28) Foelix, R. F., Biology of Spiders, 2nd ed.; Oxford University Press: New York, 1996. (29) Hajer, J. Spinning apparatus of the spider Filistata insidiatrix (Aranea: Filistatidae). Acta Entomol. BohemosloV. 1990, 86, 401–413. ¨ ber den Spinnapparat von Nephila madagascariensis (30) Peters, H. M. U (Radnetzspinnen, Fam. Argiopidae). Z. Naturforsch. 1955, 10b, 395– 404. (31) Coddington, J. A.; Levi, H. W. Systematics and Evolution of Spiders (Araneae). Annu. ReV. Ecol. Syst. 1991, 22, 565–592. (32) Goloboff, P. A. A reanalysis of mygalomorph spider families (Araneae). Am. Mus. NoVit. 1993, 3056, 1–32. (33) Dicko, C.; Knight, D.; Kenney, J.; Vollrath, F. Major and Minor ampullate, Flagelliform and Cylindrical glands secondary structures. Concentration and temperature effects. Biomacromolecules 2004, 5 (6), 2105–2115. (34) Canetti, M.; Seves, A.; Secundo, F.; Vecchio, G. CD and Small-Angle X-Ray-Scattering of Silk Fibroin in Solution. Biopolymers 1989, 28 (9), 1613–1624. (35) Fournier, A. Quantitative data on the Bombyx mori L. silkworm: a review. Biochimie 1979, 61, 283–320. (36) Dicko, C.; Kenney, J. M.; Knight, D.; Vollrath, F. Transition to a beta-sheet-rich structure in spidroin in vitro: the effects of pH and cations. Biochemistry 2004, 43 (44), 14080–7.
Emergence of Elastomericity in Silks (37) Tanaka, T.; Magoshi, J.; Magoshi, Y.; ichi Inoue, S.; Kobayashi, M.; Tsuda, H.; Becker, M.; Nakamura, S. Thermal properties of bombyx mori and several wild silkworm silks Phase transition of liquid silk. J. Therm. Anal. Calorim. 2002, 70 (3), 825–832. (38) McLachlan, G. J. Discriminant analysis and statistical pattern recognition; Wiley: New York, 2004. (39) Zar, J. H. Biostatistical Analysis, 4th ed.; Prentice Hall: London, 1999.
Biomacromolecules, Vol. 9, No. 1, 2008 221 (40) Kenney, J. M.; Knight, D.; Wise, M. J.; Vollrath, F. Amyloidogenic nature of spider silk. Eur. J. Biochem. 2002, 269 (16), 4159–4163. (41) Dicko, C.; Kenney, J.; Vollrath, F. β-silks: Enhancing and controlling aggregation. In. AdV. Protein Chem. 2006, 73, 17–53. (42) Liu, Y.; Sponner, A.; Porter, D.; Vollrath, F. Proline and processing of spider silks. Biomacromolecules in Press.
BM701069Y
