Structural properties of recombinant nonrepetitive and repetitive parts of major ampullate spidroin 1 from Euprosthenops australis: implications for fiber formation

Spider dragline silk proteins, spidroins, have a tripartite composition; a nonrepetitive N-terminal domain, a central repetitive region built up from many iterated poly-Ala and Gly rich blocks, and a C-terminal nonrepetitive domain. It is generally believed that the repetitive region forms intermolecular contacts in the silk fibers, while precise functions of the terminal domains have not been established. Herein, thermal, pH, and salt effects on the structure and aggregation and/or polymerization of recombinant N- and C-terminal domains, a repetitive segment containing four poly-Ala and Gly rich coblocks, and combinations thereof were studied. The N- and C-terminal domains have mainly alpha-helical structure, and interestingly, both form homodimers. Dimerization of the end domains allows spidroin multimerization independent of the repetitive part. Reduction of an intersubunit disulfide in the C-terminal domain lowers the thermal stability but does not affect dimerization. The repetitive region shows helical secondary structure but appears to lack stable folded structure. A protein composed of this repetitive region linked to the C-terminal domain has a mainly alpha-helical folded structure but shows an abrupt transition to beta-sheet structures upon heating. At room temperature, this protein self-assembles into macroscopic fibers within minutes. The secondary structures of none of the domains are altered by pH or salt. However, high concentrations of phosphate affect the tertiary structure and accelerate the aggregation propensity of the repetitive region. Implications of these results for dragline spidroin behavior in solution and silk fiber formation are discussed.     

Primary structure elements of spider dragline silks and their contribution to protein solubility

Spider silk proteins have mainly been investigated with regard to their contribution to mechanical properties of the silk thread. However, little is known about the molecular mechanisms of silk assembly. As a first step toward characterizing this process, we aimed to identify primary structure elements of the garden spider's (Araneus diadematus) major dragline silk proteins ADF-3 and ADF-4 that determine protein solubility. In addition, we investigated the influence of conditions involved in mediating natural thread assembly on protein aggregation. Genes encoding spider silk-like proteins were generated using a cloning strategy, which is based on a combination of synthetic DNA modules and PCR-amplified authentic gene sequences. Comparing secondary structure, solubility, and aggregation properties of the synthesized proteins revealed that single primary structure elements have diverse influences on protein characteristics. Repetitive regions representing the largest part of dragline silk proteins determined the solubility of the synthetic proteins, which differed greatly between constructs derived from ADF-3 and ADF-4. Factors, such as acidification and increases in phosphate concentration, which promote silk assembly in vivo generally decreased silk protein solubility in vitro. Strikingly, this effect was pronounced in engineered proteins comprising the carboxyl-terminal nonrepetitive regions of ADF-3 or ADF-4, indicating that these regions might play an important role in initiating assembly of spider silk proteins.     

Contributions of the C-terminal domain to poly(A)-specific ribonuclease (PARN) stability and self-association

Poly(A)-specific ribonuclease (PARN) catalyzes the degradation of mRNA poly(A) tail to regulate translation efficiency and mRNA decay in higher eukaryotic cells. The full-length PARN is a multi-domain protein containing the catalytic nuclease domain, the R3H domain, the RRM domain and the C-terminal intrinsically unstructured domain (CTD). The roles of the three well-structured RNA-binding domains have been extensively studied, while little is known about CTD. In this research, the impact of CTD on PARN stability and aggregatory potency was studied by comparing the thermal inactivation and denaturation behaviors of full-length PARN with two N-terminal fragments lacking CTD. Our results showed that K⁺ induced additional regular secondary structures and enhanced PARN stability against heat-induced inactivation, unfolding and aggregation. CTD prevented PARN from thermal inactivation but promoted thermal aggregation to initiate at a temperature much lower than that required for inactivation and unfolding. Blue-shift of Trp fluorescence during thermal transitions suggested that heat treatment induced rearrangements of domain organizations. CTD amplified the stabilizing effect of K⁺, implying the roles of CTD was mainly achieved by electrostatic interactions. These results suggested that CTD might dynamically interact with the main body of the molecule and release of CTD promoted self-association via electrostatic interactions.     

A structural view on spider silk proteins and their role in fiber assembly

Spider silk is the toughest known biomaterial and even outrivals modern synthetic high‐performance materials. The question of understanding fiber formation is how the spider can prevent premature and fatal aggregation processes inside its own body and how the chemical and mechanical stimuli used to induce the fiber formation process translate into structural changes of the silk material, finally leading to controlled and irreversible aggregation. Here, the focus will be on the structure and function of the highly conserved N‐domains and C‐terminal domains of spider dragline silk which, unlike the very long repetitive sequence elements, adopt a folded conformation in solution and are therefore able to control intermolecular interactions and aggregation between other spider silk molecules. The structures of these domains add valuable details for the construction of a molecular picture of the complicated and highly optimized silk assembly process that might be beneficial for large‐scale in vitro fiber formation attempts with recombinant silk material. Copyright © 2012 European Peptide Society and John Wiley & Sons, Ltd.     

Macroscopic fibers self-assembled from recombinant miniature spider silk proteins

Strength, elasticity, and biocompatibility make spider silk an attractive resource for the production of artificial biomaterials. Spider silk proteins, spidroins, contain hundreds of repeated poly alanine/glycine-rich blocks and are difficult to produce recombinantly in soluble form. Most previous attempts to produce artificial spider silk fibers have included solubilization steps in nonphysiological solvents. It is here demonstrated that a miniature spidroin from a protein in dragline silk of Euprosthenops australis can be produced in a soluble form in Escherichia coli when fused to a highly soluble protein partner. Although this miniature spidroin contains only four poly alanine/glycine-rich blocks followed by a C-terminal non-repetitive domain, meter-long fibers are spontaneously formed after proteolytic release of the fusion partner. The structure of the fibers is similar to that of dragline silks, and although self-assembled from recombinant proteins they are as strong as fibers spun from redissolved silk. Moreover, the fibers appear to be biocompatible because human tissue culture cells can grow on and attach to the fibers. These findings enable controlled production of high-performance biofibers at large scale under physiological conditions.     

Characterization of a novel intramolecular chaperone domain conserved in endosialidases and other bacteriophage tail spike and fiber proteins

Folding and assembly of endosialidases, the trimeric tail spike proteins of Escherichia coli K1-specific bacteriophages, crucially depend on their C-terminal domain (CTD). Homologous CTDs were identified in phage proteins belonging to three different protein families: neck appendage proteins of several Bacillus phages, L-shaped tail fibers of coliphage T5, and K5 lyases, the tail spike proteins of phages infecting E. coli K5. By analyzing a representative of each family, we show that in all cases, the CTD is cleaved off after a strictly conserved serine residue and alanine substitution prevented cleavage. Further structural and functional analyses revealed that (i) CTDs are autonomous domains with a high alpha-helical content; (ii) proteolytically released CTDs assemble into hexamers, which are most likely dimers of trimers; (iii) highly conserved amino acids within the CTD are indispensable for CTD-mediated folding and complex formation; (iv) CTDs can be exchanged between proteins of different families; and (v) proteolytic cleavage is essential to stabilize the native protein complex. Data obtained for full-length and proteolytically processed endosialidase variants suggest that release of the CTD increases the unfolding barrier, trapping the mature trimer in a kinetically stable conformation. In summary, we characterize the CTD as a novel C-terminal chaperone domain, which assists folding and assembly of unrelated phage proteins.     

Engineered disulfides improve mechanical properties of recombinant spider silk

Nature's high‐performance polymer, spider silk, is composed of specific proteins, spidroins, which form solid fibers. So far, fibers made from recombinant spidroins have failed in replicating the extraordinary mechanical properties of the native material. A recombinant miniature spidroin consisting of four poly‐Ala/Gly‐rich tandem repeats and a nonrepetitive C‐terminal domain (4RepCT) can be isolated in physiological buffers and undergoes self assembly into macrofibers. Herein, we have made a first attempt to improve the mechanical properties of 4RepCT fibers by selective introduction of AA → CC mutations and by letting the fibers form under physiologically relevant redox conditions. Introduction of AA → CC mutations in the first poly‐Ala block in the miniature spidroin increases the stiffness and tensile strength without changes in ability to form fibers, or in fiber morphology. These improved mechanical properties correlate with degree of disulfide formation. AA → CC mutations in the forth poly‐Ala block, however, lead to premature aggregation of the protein, possibly due to disulfide bonding with a conserved Cys in the C‐terminal domain. Replacement of this Cys with a Ser, lowers thermal stability but does not interfere with dimerization, fiber morphology or tensile strength. These results show that mutagenesis of 4RepCT can reveal spidroin structure‐activity relationships and generate recombinant fibers with improved mechanical properties.     

Review the role of terminal domains during storage and assembly of spider silk proteins

Fibrous proteins in nature fulfill a wide variety of functions in different structures ranging from cellular scaffolds to very resilient structures like tendons and even extra-corporal fibers such as silks in spider webs or silkworm cocoons. Despite their different origins and sequence varieties many of these fibrous proteins share a common building principle: they consist of a large repetitive core domain flanked by relatively small non-repetitive terminal domains. Amongst protein fibers, spider dragline silk shows prominent mechanical properties that exceed those of man-made fibers like Kevlar. Spider silk fibers assemble in a spinning process allowing the transformation from an aqueous solution into a solid fiber within milliseconds. Here, we highlight the role of the non-repetitive terminal domains of spider dragline silk proteins during storage in the gland and initiation of the fiber assembly process.     

C-terminal domain of the Uup ATP-binding cassette ATPase is an essential folding domain that binds to DNA

The Uup protein belongs to a subfamily of soluble ATP-binding cassette (ABC) ATPases that have been implicated in several processes different from transmembrane transport of molecules, such as transposon precise excision. We have demonstrated previously that Escherichia coli Uup is able to bind DNA. DNA binding capacity is lowered in a truncated Uup protein lacking its C-terminal domain (CTD), suggesting a contribution of CTD to DNA binding. In the present study, we characterize the role of CTD in the function of Uup, on its overall stability and in DNA binding. To this end, we expressed and purified isolated CTD and we investigated the structural and functional role of this domain. The results underline that CTD is essential for the function of Uup, is stable and able to fold up autonomously. We compared the DNA binding activities of three versions of the protein (Uup, UupΔCTD and CTD) by an electrophoretic mobility shift assay. CTD is able to bind DNA although less efficiently than intact Uup and UupΔCTD. These observations suggest that CTD is an essential domain that contributes directly to the DNA binding ability of Uup.     

An essential role for the C-terminal domain of a dragline spider silk protein in directing fiber formation

We have employed baculovirus-mediated expression of the recombinant A. diadematus spider dragline silk fibroin rADF-4 to explore the role of the evolutionary conserved C-terminal domain in self-assembly of the protein into fiber. In this unique system, polymerization of monomers occurs in the cytoplasm of living cells, giving rise to superfibers, which resemble some properties of the native dragline fibers that are synthesized by the spider using mechanical spinning. While the C-terminal containing rADF-4 self-assembled to create intricate fibers in the host insect cells, a C-terminal deleted form of the protein (rADF-4-DeltaC) self-assembled to create aggregates, which preserved the chemical stability of dragline fibers, yet lacked their shape. Interestingly, ultrastructural analysis showed that the rADF-4-DeltaC monomers did form rudimentary nanofibers, but these were short and crude as compared to those of rADF-4, thus not supporting formation of the highly compact and oriented "superfiber" typical to the rADF-4 form. In addition, using thermal analysis, we show evidence that the rADF-4 fibers but not the rADF-4-DeltaC aggregates contain crystalline domains, further establishing the former as a veritable model of authentic dragline fibers. Thus, we conclude that the conserved C-terminal domain of dragline silk is important for the correct structure of the basic nanofibers, which assemble in an oriented fashion to form the final intricate natural-like dragline silk fiber.     

Solution structure of a double mutant of the carboxy-terminal dimerization domain of the HIV-1 capsid protein

As in other retroviruses, the HIV-1 capsid (CA) protein is composed of two domains, the N-terminal domain (NTD) and the C-terminal domain (CTD), joined by a flexible linker. The dimerization of the CTD is thought to be a critical step in the assembly of the immature and mature viral capsids. The precise nature of the functional form of CTD dimerization interface has been a subject of considerable interest. Previously, the CTD dimer was thought to involve a face-to-face dimerization observed in the early crystallographic studies. Recently, the crystallographic structure for a domain-swapped CTD dimer has been determined. This dimer, with an entirely different interface that includes the major homology region (MHR) has been suggested as the functional form during the Gag assembly. The structure determination of the monomeric wt CTD of HIV-1 has not been possible because of the monomer-dimer equilibrium in solution. We report the NMR structure of the [W184A/M185A]-CTD mutant in its monomeric form. These mutations interfere with dimerization without abrogating the assembly activity of Gag and CA. The NMR structure shows some important differences compared to the CTD structure in the face-to-face dimer. Notably, the helix-2 is much shorter, and the kink seen in the crystal structure of the wt CTD in the face-to-face dimer is absent. These NMR studies suggest that dimerization-induced conformational changes may be present in the two crystal structures of the CTD dimers and also suggest a mechanism that can simultaneously accommodate both of the distinctly different dimer models playing functional roles during the Gag assembly of the immature capsids.     

A proposed model for dragline spider silk self‐assembly: Insights from the effect of the repetitive domain size on fiber properties

Dragline spider silk has been intensively studied for its superior qualities as a biomaterial. In previous studies, we made use of the baculovirus mediated expression system for the production of a recombinant Araneus diadematus spider silk dragline ADF4 protein and its self‐assembly into intricate fibers in host insect cells. In this study, our aim was to explore the function of the major repetitive domain of the dragline spider silk. Thus, we generated an array of synthetic proteins, each containing a different number of identical repeats up to the largest recombinantly expressed spider silk to date. Study of the self‐assembly properties of these proteins showed that depending on the increasing number of repeats they give rise to different assembly phenotypes, from a fully soluble protein to bona fide fibers with superior qualities. The different assembly forms, the corresponding chemical resistance properties obtained as well as ultrastructural studies, revealed novel insights concerning the structure and intermolecular interactions of the repetitive and nonrepetitive domains. Based on these observations and current knowledge in the field, we hereby present a comprehensive hypothetical model for the mechanism of dragline silk self‐assembly and fiber formation. © 2009 Wiley Periodicals, Inc. Biopolymers 93: 458–468, 2010. This article was originally published online as an accepted preprint. The “Published Online” date corresponds to the preprint version. You can request a copy of the preprint by emailing the Biopolymers editorial office at biopolymers@wiley.com     

Structural characterization of Saccharomyces cerevisiae prion-like protein Ure2

Sacchromyces cerevisiae prion-like protein Ure2 was expressed in Escherichia coli and was purified to homogeneity. We show here that Ure2p is a soluble protein that can assemble into fibers that are similar to the fibers observed in the case of PrP in its scrapie prion filaments form or that form on Sup35 self-assembly. Ure2p self-assembly is a cooperative process where one can distinguish a lag phase followed by an elongation phase preceding a plateau. A combination of size exclusion chromatography, sedimentation velocity, and electron microscopy demonstrates that the soluble form of Ure2p consists at least of three forms of the protein as follows: a monomeric, dimeric, and tetrameric form whose abundance is concentration-dependent. By the use of limited proteolysis, intrinsic fluorescence, and circular dichroism measurements, we bring strong evidence for the existence of at least two structural domains in Ure2p molecules. Indeed, Ure2p NH2-terminal region is found poorly structured, whereas its COOH-terminal domain appears to be compactly folded. Finally, we show that only slight conformational changes accompany Ure2p assembly into insoluble high molecular weight oligomers. These changes essentially affect the COOH-terminal part of the molecule. The properties of Ure2p are compared in the discussion to that of other prion-like proteins such as Sup35 and mammalian prion protein PrP.     

Biophysical Analysis of the MHR Motif in Folding and Domain Swapping of the HIV Capsid Protein C-Terminal Domain

Infection by human immunodeficiency virus (HIV) depends on the function, in virion morphogenesis and other stages of the viral cycle, of a highly conserved structural element, the major homology region (MHR), within the carboxyterminal domain (CTD) of the capsid protein. In a modified CTD dimer, MHR is swapped between monomers. While no evidence for MHR swapping has been provided by structural models of retroviral capsids, it is unknown whether it may occur transiently along the virus assembly pathway. Whatever the case, the MHR-swapped dimer does provide a novel target for the development of anti-HIV drugs based on the concept of trapping a nonnative capsid protein conformation. We have carried out a thermodynamic and kinetic characterization of the domain-swapped CTD dimer in solution. The analysis includes a dissection of the role of conserved MHR residues and other amino acids at the dimerization interface in CTD folding, stability, and dimerization by domain swapping. The results revealed some energetic hotspots at the domain-swapped interface. In addition, many MHR residues that are not in the protein hydrophobic core were nevertheless found to be critical for folding and stability of the CTD monomer, which may dramatically slow down the swapping reaction. Conservation of MHR residues in retroviruses did not correlate with their contribution to domain swapping, but it did correlate with their importance for stable CTD folding. Because folding is required for capsid protein function, this remarkable MHR-mediated conformational stabilization of CTD may help to explain the functional roles of MHR not only during immature capsid assembly but in other processes associated with retrovirus infection. This energetic dissection of the dimerization interface in MHR-swapped CTD may also facilitate the design of anti-HIV compounds that inhibit capsid assembly by conformational trapping of swapped CTD dimers.     

Carbonic Anhydrase Generates CO2 and H+ That Drive Spider Silk Formation Via Opposite Effects on the Terminal Domains

Spider silk fibers are produced from soluble proteins (spidroins) under ambient conditions in a complex but poorly understood process. Spidroins are highly repetitive in sequence but capped by nonrepetitive N- and C-terminal domains (NT and CT) that are suggested to regulate fiber conversion in similar manners. By using ion selective microelectrodes we found that the pH gradient in the silk gland is much broader than previously known. Surprisingly, the terminal domains respond in opposite ways when pH is decreased from 7 to 5: Urea denaturation and temperature stability assays show that NT dimers get significantly stabilized and then lock the spidroins into multimers, whereas CT on the other hand is destabilized and unfolds into ThT-positive β-sheet amyloid fibrils, which can trigger fiber formation. There is a high carbon dioxide pressure (pCO₂) in distal parts of the gland, and a CO₂ analogue interacts with buried regions in CT as determined by nuclear magnetic resonance (NMR) spectroscopy. Activity staining of histological sections and inhibition experiments reveal that the pH gradient is created by carbonic anhydrase. Carbonic anhydrase activity emerges in the same region of the gland as the opposite effects on NT and CT stability occur. These synchronous events suggest a novel CO₂ and proton-dependent lock and trigger mechanism of spider silk formation.     

Formation of Amyloid Fibers by Monomeric Light Chain Variable Domains

Systemic light chain amyloidosis is a lethal disease characterized by excess immunoglobulin light chains and light chain fragments composed of variable domains, which aggregate into amyloid fibers. These fibers accumulate and damage organs. Some light chains induce formation of amyloid fibers, whereas others do not, making it unclear what distinguishes amyloid formers from non-formers. One mechanism by which sequence variation may reduce propensity to form amyloid fibers is by shifting the equilibrium toward an amyloid-resistant quaternary structure. Here we identify the monomeric form of the Mcg immunoglobulin light chain variable domain as the quaternary unit required for amyloid fiber assembly. Dimers of Mcg variable domains remain stable and soluble, yet become prone to assemble into amyloid fibers upon disassociation into monomers.     

Retroviral Capsid Assembly: A Role for the CA Dimer in Initiation

In maturing retroviral virions, CA protein assembles to form a capsid shell that is essential for infectivity. The structure of the two folded domains [N-terminal domain (NTD) and C-terminal domain (CTD)] of CA is highly conserved among various retroviruses, and the capsid assembly pathway, although poorly understood, is thought to be conserved as well. In vitro assembly reactions with purified CA proteins of the Rous sarcoma virus (RSV) were used to define factors that influence the kinetics of capsid assembly and provide insights into underlying mechanisms. CA multimerization was triggered by multivalent anions providing evidence that in vitro assembly is an electrostatically controlled process. In the case of RSV, in vitro assembly was a well-behaved nucleation-driven process that led to the formation of structures with morphologies similar to those found in virions. Isolated RSV dimers, when mixed with monomeric protein, acted as efficient seeds for assembly, eliminating the lag phase characteristic of a monomer-only reaction. This demonstrates for the first time the purification of an intermediate on the assembly pathway. Differences in the intrinsic tryptophan fluorescence of monomeric protein and the assembly-competent dimer fraction suggest the involvement of the NTD in the formation of the functional dimer. Furthermore, in vitro analysis of well-characterized CTD mutants provides evidence for assembly dependence on the second domain and suggests that the establishment of an NTD-CTD interface is a critical step in capsid assembly initiation. Overall, the data provide clear support for a model whereby capsid assembly within the maturing virion is dependent on the formation of a specific nucleating complex that involves a CA dimer and is directed by additional virion constituents.     

Secondary structure and NMR resonance assignments of the C-terminal DNA-binding domain of Uup protein

ATP-binding cassette (ABC) systems belong to a large superfamily of proteins that couple the energy released from ATP hydrolysis to a wide variety of cellular processes, including not only transport of various molecules, but also gene regulation, and DNA repair. Mutations in the bacterial uup gene, which encodes a cytosolic ABC ATPase, lead to an increase in the frequency of precise excision of transposons Tn10 and Tn5, suggesting a role of the Uup protein in DNA metabolism. Uup is a 72?kDa polypeptide which comprises two ABC domains, separated by a 75-residue linker, and a C-terminal domain (CTD) of unknown function. The Uup protein from Escherichia coli has been shown to bind DNA in vitro, and the CTD domain contributes to the DNA-binding affinity. We have produced and purified uniformly labeled ¹⁵N- and ¹⁵N/¹³C Uup CTD domain (region 528?C635), and assigned backbone and side-chains resonances using heteronuclear NMR spectroscopy. Secondary structure evaluation based on backbone chemical shifts is consistent with the presence of three ??-helices, including two long ones (residues 564?C590 and 601?C632), suggesting that Uup CTD may fold as an intramolecular coiled coil motif. This work provides the starting point towards determining the first atomic structure of a non-ATPase domain within the vast REG subfamily of ABC soluble ATPases.     

KCTD5 is endowed with large,functionally relevant,interdomain motions

The KCTD family is an emerging class of proteins that are involved in important biological processes whose biochemical and structural properties are rather poorly characterized or even completely undefined. We here used KCTD5, the only member of the family with a known three-dimensional structure, to gain insights into the intrinsic structural stability of the C-terminal domain (CTD) and into the mutual dynamic interplay between the two domains of the protein. Molecular dynamics (MD) simulations indicate that in the simulation timescale (120 ns), the pentameric assembly of the CTD is endowed with a significant intrinsic stability. Moreover, MD analyses also led to the identification of exposed β-strand residues. Being these regions intrinsically sticky, they could be involved in the substrate recognition. More importantly, simulations conducted on the full-length protein provide interesting information of the relative motions between the BTB domain and the CTD of the protein. Indeed, the dissection of the overall motion of the protein is indicative of a large interdomain twisting associated with limited bending movements. Notably, MD data indicate that the entire interdomain motion is pivoted by a single residue (Ser150) of the hinge region that connects the domains. The functional relevance of these motions was evaluated in the context of the functional macromolecular machinery in which KCTD5 is involved. This analysis indicates that the interdomain twisting motion here characterized may be important for the correct positioning of the substrate to be ubiquitinated with respect to the other factors of the ubiquitination machinery.     

N-Terminal domain of Bombyx mori fibroin mediates the assembly of silk in response to pH decrease

Fibroins serve as the major building blocks of silk fiber. As the major component of fibroin, the fibroin heavy chain is a considerably large protein comprising N-terminal and C-terminal hydrophilic domains and 12 highly repetitive Gly-Ala-rich regions flanked by internal hydrophilic blocks. Here, we show the crystal structure of the fibroin N-terminal domain (FibNT) at pH?4.7, revealing a remarkable double-layered anti-parallel β-sheet with each layer comprising two FibNT molecules entangled together. We also show that FibNT undergoes a pH-responsive conformational transition from random coil to β-sheets at around pH?6.0. Dynamic light scattering demonstrates that FibNT tends to oligomerize as pH decreases to 6.0, and electron microscopy reveals micelle-like oligomers. Our results are consistent with the micelle assembly model of silk fibroin and, more importantly, show that the N-terminal domain in itself has the capacity to form micelle-like structures in response to pH decrease. Structural and mutagenesis analyses further reveal the important role of conserved acidic residues clustered in FibNT, such as Glu56 and Asp100, in preventing premature β-sheet formation at neutral pH. Collectively, we suggest that FibNT functions as a pH-responsive self-assembly module that could prevent premature β-sheet formation at neutral pH yet could initiate fibroin assembly as pH decreases along the lumen of the posterior silk gland to the anterior silk gland.