Dragline Silk: A Fiber Assembled with Low-Molecular-Weight Cysteine

Sep 26, 2014 - Here we report the discovery of novel molecular constituents within the ... that belong to the cystine slipknot superfamily, suggesting...
0 downloads 0 Views 4MB Size
Download PDF
Article pubs.acs.org/Biomac

Dragline Silk: A Fiber Assembled with Low-Molecular-Weight Cysteine-Rich Proteins Thanh Pham,†,‡ Tyler Chuang,†,‡ Albert Lin,†,‡ Hyun Joo,§ Jerry Tsai,§ Taylor Crawford,† Liang Zhao,∥ Caroline Williams,† Yang Hsia,⊥ and Craig Vierra*,† Departments of †Biology and §Chemistry, University of the Pacific, Stockton, California 95211, United States ∥ Center for Alternatives to Animal Testing (CAAT), Department of Environmental Health Sciences, Bloomberg School of Public Health, Johns Hopkins University, Baltimore, Maryland 21205, United States ⊥ Department of Biochemistry, University of Washington, Seattle, Washington 98195, United States ABSTRACT: Dragline silk has been proposed to contain two main protein constituents, MaSp1 and MaSp2. However, the mechanical properties of synthetic spider silks spun from recombinant MaSp1 and MaSp2 proteins have yet to approach natural fibers, implying the natural spinning dope is missing critical factors. Here we report the discovery of novel molecular constituents within the spinning dope that are extruded into dragline silk. Protein studies of the liquid spinning dope from the major ampullate gland, coupled with the analysis of dragline silk fibers using mass spectrometry, demonstrate the presence of a new family of low-molecular-weight cysteinerich proteins (CRPs) that colocalize with the MA fibroins. Expression of the CRP family members is linked to dragline silk production, specifically MaSp1 and MaSp2 mRNA synthesis. Biochemical data support that CRP molecules are secreted into the spinning dope and assembled into macromolecular complexes via disulfide bond linkages. Sequence analysis supports that CRP molecules share similarities to members that belong to the cystine slipknot superfamily, suggesting that these factors may have evolved to increase fiber toughness by serving as molecular hubs that dissipate large amounts of energy under stress. Collectively, our findings provide molecular details about the components of dragline silk, providing new insight that will advance materials development of synthetic spider silk for industrial applications.

■

black widow MaSps.7 In addition, atomic force microscopy has been used to examine the surface of dragline silk threads from black widow spiders.9 Attempts to produce synthetic silk fibers using truncated recombinant fibroins purified from a variety of different heterologous expression systems, such as bacteria, yeast, plants, mammalian cells, and insect cell lines, have had limited success, leading to artificial fibers with lesser properties relative to natural silks.10−13 In part, these inferior properties have been attributed to the use of truncated recombinant proteins instead of full-length fibroins, problems with reproducing the precise chemical and physical properties of the natural extrusion process, and a lack of understanding all of the protein constituents in the spinning dope.14−16 In these studies, we discover and characterize a new family of low-molecular-weight cysteine-rich proteins and demonstrate their expression pattern is restricted to silk glands. These factors are shown to colocalize with the spidroins in both the dope and spun fibers, and have structural implications on the mechanical properties of silk threads. Taken together, these

INTRODUCTION Spider silk has extraordinary mechanical properties, displaying tensile strength and toughness values that exceed some of the most renowned engineering materials, including high tensile steel and Kevlar.1 Many spiders are capable of spinning several fiber types; however, most attention has been focused on dragline silk. Dragline silk, also known as the lifeline of the spider, has attracted the most attention due to its ease of collection and superior mechanical properties. Molecular and chemical studies have suggested that dragline silk consists of MaSp1 and MaSp2, two distinct structural proteins that belong to the spider silk family.2,3 These fibroins are large molecular weight proteins that contain internal block repeats, which are flanked by conserved nonrepetitive N- and C-terminal domains (NTDs).4−6 Secondary structural analysis of the protein structure of natural dragline silk from black widow spiders using X-ray diffraction (XRD) reveals β-sheet nanocrystallites that are stacked and oriented parallel to the fiber axis, along with an amorphous region with isotropic and oriented components.7 XRD and transmission electron microscopy have also been used to characterize the size of the crystallites in black widow dragline silk.8 Solid-state NMR measurements support the crystallite regions are formed from poly(Ala)n blocks found within the internal protein block repeats of the © 2014 American Chemical Society

Received: July 30, 2014 Revised: September 24, 2014 Published: September 26, 2014 4073

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

Table 1. De Novo Sequence of Peptide Ions after Solubilization of Dragline Silk Followed by Tryptic Digestion and MS/MS Analysisa peptide mass

de novo sequence

ID

1560 1611

LEPQPGGPTGPSGNPR GGAGQGGAAAAAAAAAGAGQR

1760

GGAGAAAAAAAAAGGAGQGGYGR

1888

(Q/GA)GGAGGGAAAAAAAAAAGGAGGYGR

1899

QQAYGPGGSGAAAAAAGGAGPGR

2082

···GYGPGGSGAAAAAAAAAGGAG···R

2099

(Q/AG)AGGYGPGGSGAAAAAAAAAGGAG···

2373

(YG)GGQGAGQGGAGAAAAAAAAGGAGQGGAGR

2460

(GAG/GQ)GAGQGGAAATAAAAGGAGQGGQGGYGQR

3212

GGAGQGGAAAAAGAGQGGYGGQGAGQGGAGAAAAAAAAGGAGR

3297

···GGYGQGGYGQGGAGQGGAGAAAAAAAAAGGAGQGGYGR

3460

···DMAFASSVAELAVADGQNVGAATNALSDALR

3917

···GQGGAAAAAGAGQGGYGGQGAGQGGAQAAAAAAAGGAGQGGQGGYGR

unknown MaSp1 (L2) EU177651.1 MaSp1 (L2) EU177651.1 MaSp1 EF595246.1 MaSp2 EU177652.1 MaSp2 DQ379382.1 MaSp2 DQ379382.1 MaSp1 (L1) EU177654.1 MaSp1 AF350273.1 MaSp1 EF595246.1 MaSp1 (L2) EU177653.1 MaSp2 EF595248.1 MaSp1 (L1) EF595246.1

identities 100 100 89 100 100 100 96 96 100 100 100 96

a

Dots represent regions where the amino acid sequence could not be obtained by MS/MS analysis. L1 or L2 represent either locus 1 or 2 for MaSp1 from L. hesperus, respectively. GenBank accession numbers are shown under the ID category. peptide sequence LEPQPGGPTGPSGNPR, a sequence that was obtained by MS/MS analysis of the precursor ion with m/z 1560. A single translated cDNA sequence, which was named CRP1, was discovered and then used to identify four other translated cDNAs with sequence similarity. These were designated CRP2−5. Isolation of Major Ampullate Spinning Dope and SDS-PAGE Analysis. For each analysis, purified spinning dope was obtained by dissecting two adult female black widow spiders. The epithelial layers of four major ampullate glands were stripped off using microscissors and tweezers. The spinning dope was combined and dissolved in 8 M urea supplemented with protease inhibitor cocktail (Sigma-Aldrich). After urea addition, the mixture was split into two fractions, with one of the fractions supplemented with 5 mM DTT. After 5 min of incubation with reducing agent, the samples were size-fractionated by electrophoresis on 12% Mini-PROTEAN TGX precast gels (Biorad). Protein bands were visualized by either silver staining or Western blot analysis. Quantitative Real-Time PCR Analysis. Spider silk glands were removed as previously described from female black widow spiders.18 For silking experiments, dragline threads were collected once every 24 h over the course of 11 days. Spiders were fed once between the collection procedures. Approximately 250 feet of dragline silk was collected during the silking process. Control spiders were not silked but handled in a similar fashion as the silked group. After microdissection and removal of the silk-producing glands, total RNA was isolated using TRIzol reagent (Life Technologies) according to the manufacturer’s protocol. Total RNA was quantified using a NanoDrop 2000c (Thermo Scientific) and expression levels of CRPs were analyzed using the KAPA SYBR FAST qPCR kit (Kapa Biosystems, Inc.). Real-time PCR fluorescence was monitored using an Opticon II instrument (MJ Research Inc.). Amplification products were monitored with SYBR Green detection and routinely checked using melting curve software. Gene-specific forward and reverse primers used to produce specific CRP amplicons for Q-PCR were as follows: 5′- att gta atc tgc aat cct gaa cc-3′ and 5′-act cag aaa tgt gtg aga

results advance our understanding of the molecular constituents and the assembly process of spider silks.

■

MATERIALS AND METHODS

Identification of CRP1 in Dragline Silk Fibers. In solution digests of dragline silk were performed by dissolving 3 μg of silk in 100 μL of 5 M LiBr. Protein solubilization was facilitated by heating at 95 °C for 10 min. Prior to trypsin digestion, the sample was neutralized by the addition of 700 μL of 50 mM NH4HCO3 (pH 7.8). The sample was incubated at 37 °C overnight in the presence of 10 μg of trypsin. Peptides were extracted and desalted with a C18 Zip-Tip and then used for MS or MS/MS analysis. Mass Spectrometry Analysis. Mass spectra were obtained with a MALDI/TOF/TOF mass spectrometer (4700 Proteomics Analyzer, Applied Biosystems, Foster City, CA) operated in reflector mode. MS/ MS spectra were obtained by operating the instrument in MS/MS mode. In-solution tryptic digest samples were separated using HPLC (Magic C18 5u, 200A, 0.15 × 150 mm column, 5−35% acetonitrile in water in 35 min gradients) and fractions were collected onto a MALDI plate. An equal volume of CHCA matrix was mixed with each fraction as it eluted from the column. The plate was subsequently introduced into the 4700 mass spectrometer. De novo peptide sequences were obtained by manual interpretation of the high energy CAD spectra. cDNA Library Construction and Screening. The cDNA library was constructed using all seven different silk-producing glands, forming a composite library as previously described.17 After plating the cDNA library, hundreds of individual colonies were selected and grown overnight to saturation. Plasmid DNA from each colony was isolated using QIAprep Spin Miniprep Kit (Qiagen) according to the manufacturer’s instructions. DNA sequence analysis (University of Florida) was performed to determine the cDNA sequence. cDNA sequences were translated and the longest open reading frames (ORFs) were stored in a custom database. This database was used to search for translated sequences that contained similarities to the 4074

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

Figure 1. MALDI tandem TOF analysis of a tryptic fragment generated from an in-solution digestion of solubilized dragline silk from L. hesperus identifies the presence of a non-MaSp peptide sequence, LEPQPGGPTGPSGNPR. The high energy CAD (MS/MS) spectrum of precursor ion with m/z 1560.7 is shown. Y ions and their corresponding masses are shown above the peaks. aac-3′ (CRP1), 5′-aag gaa gct caa gct gac tgc-3′ and 5′-ccc aaa cat gtg tga aaa at-3′ (CRP2), 5′-gaa gaa gat aca tcg aga g-3′ and 5′-aaa caa aga tga taa atg t-3′ (CRP3), 5′-caa tgt gga gat tgc agt ggc-3′ and 5′-atg cga gaa agc tcc ttc taa a-3′ (CRP4), and 5′-gtt tgc ttg cct gaa aga t-3′ and 5′-ctg aaa ctt tac aat gtt ct-3′ (CRP5). Normalization was performed using the beta-actin sense primer 5′-ccc tga gag gaa gta ctc cgt-3′ and antisense primer 5′-atc cac atc tgc tgg aag gtg-3′. DNA Constructs. The cDNA coding for CRP1 was performed using the forward and reverse primers 5′-cat atg gaa ttc att gta atc tgc aat cct gaa cc-3′ and 5′-tta gga tcc ctcgag ctt gtc atc gtc atc ctt gta gtc gat gtc atg atc ttt ata aat cac cgt cat ggt ctt tgt agt cgt ttc tca aca ttt ctg agt-3′, respectively. The forward primer was engineered to incorporate a SalI restriction site on the 5′-terminus, while the reverse primer integrated a BamHI restriction site on the 5′ terminus. The forward primer was designed to exclude the predicted secretion signal from the CRP1 protein. Following PCR, the amplified product was ligated into the cloning vector pBAD-Topo according to the manufacturer’s instructions (Life Technologies). Prior to expression studies, the nucleic acid sequence was confirmed by DNA sequencing. The pET19b-Sumo-CRP3 and pET19b-Sumo-CRP4 constructs were generated by amplifying the coding regions of CRP3 and CRP4 from the composite cDNA library by PCR. The forward and reverse primers used to amplify the CRP3 cDNA were 5′-gaa ttc gaa gaa gat aca tcg aga g-3′ and 5′-gg tcc aaa caa aga tga taa atg t-3, respectively, whereas the forward and reverse primers used to amplify the CRP4 cDNA were 5′-ctc gag caa tgt gga gat tgc agt ggc-3′ and 5′-gga tcc atg cga gaa agc tcc ttc taa a-3′, respectively. The primers were designed to remove the predicted secretion signals. The pET19b-Sumo vector contains an N-terminal Sumo tag, along with a 10× histidine tag. Computational Analysis. The CRP1 sequence was modeled by using the Modeler 9.13 program.19 Because no structural homologues could be identified through sequence alignment, known slipknot proteins were analyzed based on the number of cysteines and remote sequence similarity. Based on these criteria, the human chorionic gonadotropin (PDB ID: 1HRP) structure was chosen as a template for the initial build.20 Of the 12 cysteines, three pairs were arranged to form a cysteine knot structure and an additional two other disulfide bonds were formed. These five disulfide bonds left two cysteines free to potentially form intermolecular disulfide bonds with other proteins.

Two models are designed that differ only in the six residues that make up the slipknot. Model 1’s slipknot consists of cysteines at residue positions 4, 30, 33, 73, and 79. Model 2’s slipknot consists of cysteines at residue positions 4, 15, 19, 32, 36, and 79. The UCSF Chimera program was utilized to visualize and produce images of the models.21

■

RESULTS

Dragline Silk Contains Cysteine-Rich Proteins. In order to identify new constituents with structural roles in spider dragline silk, we dissolved dragline fibers in chaotropic reagents and digested the solubilized material with trypsin. After sequencing the peptide ion mixture using MS/MS analysis, the peptides were searched against the nonredundant NCBI protein database using BLASTP. The majority of the peptides were derived from the translated sequence of L. hesperus MaSp1 or MaSp2 (Table 1). Although 3 distinct MaSp1 loci have been identified in black widow spiders, the MaSp1 locus 2 mRNA has been shown to be expressed in the highest levels in the major ampullate (MA) gland.22 It has also been proposed that these loci might be differentially expressed for different biological functions.22 In our studies, we detected peptides from both MaSp1 locus 1 and 2, as well as MaSp2 (Table 1). This is consistent with previous reports that dragline silk is a two protein fiber, a conclusion largely drawn from the similarities between amino acid composition profiles of natural fibers and the translated MaSp1 and MaSp2 cDNA sequences.3 We also identified one peptide that displayed a non-MaSp-like amino acid sequence, lacking the characteristic fibroin motifs reported in the translated genetic blueprints of MaSp1 or MaSp2 (Table 1). In particular, MS/MS analysis of peptide ion with m/z 1560 displayed the amino acid sequence LEPQPGGPTGPSGNPR (Figure 1). This sequence was used to search a database of translated cDNAs that were obtained by randomly sequencing clones plated from a library prepared from silk-producing glands of a black widow spider. This led to 4075

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

Figure 2. Cob-weaving spiders contain a family of low molecular weight proteins that are cysteine-rich. (A) Predicted amino acid sequence of CRP1. (B) Alignment of CRP1 to other CRP family members. Conserved cysteine residues are shown in red bolded text. Underlined regions represent putative secretion signals identified by SignaIP 4.1.36 The bolded blue text represents the peptide identified by MS/MS analysis after solubilized dragline silk fibers were subject to in-solution tryptic digestion. Asterisks represent conserved residues. Theoretical molecular masses after the removal of the predicted secretion signals are 9.5, 8.9, 11.3, 9.3 and 7.6 kDa for CRP1 to 5, respectively. The isoelectric points (pIs) for CRP1−5 are 8.31, 4.32, 5.06, 6.25, and 4.44, respectively.

the identification of a translated cDNA that displayed nearly a 100% match across the 16 amino acids (Figure 2A). A total of 14 of the 16 residues were identical. The two differences between the de novo sequence of the peptide and translated cDNA sequence involved isobaric residues, including Leu (L) and Ile (I), as well as Gln (Q) and Lys (K), respectively (see underlined region in peptide). Multiple cDNA molecules amplified from the library using PCR, followed by DNA sequencing, confirmed these residues were I and K, supporting the in-solution tryptic product sequenced by MS/MS resulted from one missed cleavage event (Figure 2A). Inspection of the translated full-length cDNA revealed a protein with an expected molecular mass of 11.4 kDa and pI = 8.27. Analysis of the translated sequence using SignaIP 4.0 predicted the presence of an N-terminal hydrophobic signal sequence, a property expected for proteins synthesized in epithelial cells of the major ampullate tail region that are secreted in the ampulla prior to fiber extrusion (Figure 2A). Based on the theoretical amino acid composition profile from the translated cDNA, Cys, Val, Asn, and Pro were found to be 11.3, 9.4, 8.5, and 8.5%, respectively. Because of the high Cys content, this protein was named cysteine-rich protein 1 (CRP1). Collectively, our results support CRP1 is extruded, along with MaSp1 and MaSp2, into dragline silk fibers. Black Widow Spiders Contain a Family of CRPs. In order to identify proteins related to CRP1, we searched a custom database containing spider cDNA sequences generated from screening our composite cDNA library constructed from the seven silk-producing glands. Four additional cDNAs were discovered that shared similar translated protein sequences, suggesting the presence of a low-molecular-weight cysteine-rich family (Figure 2B). These other clones were designated CRP2−5 [GenBank Accession ADV40319 (CRP2),

ADV40315 (CRP3), ADV40217 (CRP4), ADV40350 (CRP5)]. Predicted molecular masses for these proteins spanned 7.6 to 11.3 kDa. Alignment of the CRP1 amino acid sequence to these products showed sequence identities ranging from 34 to 60%. Strikingly, alignment of the CRP amino acid sequences revealed almost perfect conservation with respect to the spacing and number of Cys residues in their polypeptide chains, but showed differences in flanking residues (Figure 2B). Protein−protein BLAST searches using CRP1 against the nrNCBI database revealed 27% sequence identities to a protein classified as a cystine slipknot polypeptide from the spider Dolomedes mizhoanus (GeneBank Accession KC480132.1).23 The majority of the conservation was localized to the Cys residues. Analysis of the secondary structure of the CRPs supports the presence of a common structural motif that is similar to a cystine slipknot, a signature reported in toxins and growth factors.24,25 Dynamical modeling has shown cystine knots within growth factors and toxins produce exceptional resistance to stretching, requiring some of the highest forces required to unravel its tertiary structure relative to other proteins.26 Expression of CRPs. Because dragline silk is synthesized by the MA gland, we investigated whether CRP1 was coexpressed with MaSp1 and MaSp2. To test this assertion, we examined CRP transcript levels in 6 of the 7 different silk-producing glands (major and minor ampullate-, aciniform-, tubuliform-, aggregate-, and flagelliform glands) using QRT-PCR analysis. The pyriform gland was excluded from this analysis due to the intrinsic difficulty in isolating intact RNA from this structure. CRP1, CRP2, CRP3, and CRP4 mRNAs were shown to be predominantly expressed in the major ampullate gland (Figure 3A). As a negative control, fat tissue showed no appreciated levels of CRP expression (Figure 3A). CRP2 transcript levels 4076

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

mRNA levels were approximately 43 times higher in the aggregate gland relative to the tubuliform gland (Figure 3A). In order to investigate the influence of removing dragline silk from the major ampullate gland and its effect on CRP mRNA synthesis, we collected dragline silking from spiders over several days using a free falling methodology (gravity silking), and then extracted the MA glands to analyze changes in gene expression profiles. When spiders were subject to dragline silking, CRP2, CRP3, CRP4, MaSp1, and MaSp2 transcript levels were elevated 6.4-, 5.5-, 2.5-, 1.3-, and 2-fold, respectively (Figure 3B−C). Conversely, CRP1 levels were decreased 330-fold, suggesting a differential function relative to CRP2 and CRP4 (Figure 3B−C). CRP5 levels remained essentially the same in control versus silked spiders. Taken together, our findings support the assertion that four of the five CRP family members (CRP1 to 4) have expression profiles linked to pulling dragline silk while CRP5, which was not influenced by dragline silking, appears to be under a different mode of transcriptional regulation, which agrees with the mRNA analysis as its transcript was not detected in the MA gland in significant amounts (Figure 3A−C). MA Spinning Dope. To demonstrate CRP family members were present within the spinning dope with the MaSps, we developed a technique to remove the epithelial tissues of the MA gland, leaving only pure spinning dope constituents (Figure 4A,B). Solubilized spinning dope proteins were then

Figure 3. CRP family members are predominantly expressed in the MA gland and display expression profiles linked to dragline silk production. (A) Real-time RT-PCR analysis of CRP1−5 transcript levels in six different silk-producing glands and fat from resting spiders. (B) Real-time RT-PCR analysis of CRP1−5 mRNA levels in the major ampullate gland from resting (control) or dragline silked spiders (silked). (C) Real-time PCR analysis of CRP2-, CRP4-, MaSp1-, and MaSp2 transcript levels in the major ampullate gland from resting (control) or dragline silked spiders (free falling methodology). *Note in (A), expression of CRP3 in the MA gland was detectable, but due to the current scale selection, it is not readily apparent.

Figure 4. Microdissection of a female black widow and visualization of the MA gland. (A) An intact MA gland; (B) Skinning of the MA gland and isolation of the pure spinning dope constituents. Images were photographed at 10× magnification using a Leica MZ16 stereo dissecting microscope equipped with a DFC320 digital camera.

size fractionated using SDS-PAGE analysis and the proteins visualized by silver staining. Following visualization, large molecular weight proteins were detected at >250-kDa (Figure 5). These species correlated well to the predicted molecular masses of MaSp1 and MaSp2 (Figure 5). Treatment of the spinning dope with reducing agent resulted in the appearance of prominent bands that migrated at approximately 10 kDa as well as stronger banding profiles for the MaSps (Figure 5, compare lanes 2 and 3). The detection of these bands required breaking disulfide bonds, an observation that supports the

were approximately 56 and 6 times higher relative to CRP1 and CRP4, respectively. CRP3 transcripts were detected in the major ampullate gland, but were substantially lower relative to CRP1, CRP2, and CRP4 (Figure 3A). Elevated levels of CRP4 transcripts in the MA gland of black widows is consistent with a recent transcriptome analysis, despite being unidentified and characterized.22 CRP5 mRNA was preferentially expressed in the aggregate- and tubuliform glands (Figure 3A). CRP5 4077

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

Figure 5. Spinning dope from the major ampullate gland contains MaSps and CRPs. Proteins from the spinning dope were sizefractionated using SDS-PAGE analysis and then visualized using silver staining.

assertion that these proteins are assembled in the spinning dope via disulfide linkages into macromolecular complexes. Excision of the ∼10-kDa bands, followed by in-gel tryptic digestion and MS analysis, revealed the presence of peptide fingerprints consistent with the presence of the CRPs. For this analysis, we observed sequence coverage of 60.38, 45.71, 78.33, and 25.23% for CRP1−4, respectively, after theoretical digestion of their translated cDNA with trypsin and comparison to experimental masses (data not shown). Collectively, these data support CRPs are present within the spinning dope and are assembled into large molecular weight complexes via disulfide bond linkages. CRPs Have Self-Polymerization Activity. To characterize the biochemical properties of the CRPs, we expressed these proteins in bacteria and analyzed the products by Western blot analysis using antibodies specific for epitope tags that were engineered on either the N- or C-termini of the recombinant proteins. Expression of a C-terminal FLAG-tagged CRP1 construct, followed by analysis under reducing conditions, resulted in the detection of monomeric species, while analysis using nonreducing conditions led to the observation of high molecular weight immunoreactive species (Figure 6A). These species represented self-polymerized multimeric CRP1 complexes, requiring the oxidation of cystine residues (Figure 6A). Similar, but less pronounced higher molecular weight immunoreactive complexes were observed for CRP2 and CRP4 fusion proteins that contained engineered N-terminal His-tag epitopes under nonreducing conditions (Figure 6B). The majority of the higher molecular weight complexes could be broken with reducing agent, but in some instances, the slower migrating, immunoreactive bands remained resistant to reducing agent. This implies that either the cystine bonds are resistant to reducing agents or another type of covalent bond is also present within the immunoreactive complexes. Overall, these observations support that the CRPs can assemble into larger molecular weight complexes via disulfide linkages. Molecular Modeling of CRP1. To investigate structural properties of the CRP family, we performed a computational modeling analysis of CRP1. Computational structural modeling of the CRP1 sequence supported the presence of a putative

Figure 6. CRPs form large molecular weight complexes under oxidizing condition. Western blot analysis of crude lysates containing recombinant proteins expressed in bacteria. Lysates were treated with Laemmli buffer supplemented with β-mercaptoethanol (+BME) or lacking reducing agent (-BME). (A) Crude bacterial lysate after CRP1 expression; (B) Crude bacterial lysate after the induction of CRP3 and CRP4. CRP1 proteins were detected using an anti-FLAG monoclonal antiserum, while CRP3 and CRP4 were detected using an anti-his monoclonal antibody antiserum.

slipknot structure as well as the formation of higher order oligomerized structures through intermolecular disulfide bonds. The CRP1 amino acid sequence contains 12 cysteines at residue positions 4, 10, 15, 19, 30, 32, 33, 36, 46, 73, 79, and 86. These residues have the potential to participate in a maximum number of six intramolecular disulfide bonds. The slipknot requires a total of three disulfide bonds: two pairs form a core ring and another pair connects through the ring. The arrangement of cysteines is important, and four cysteine residues within a stretch of seven residues (C30, C32, C33, and C36) can form the core ring of the slipknot structure. Two models were computationally built, as shown in Figure 7. In the first model (Figure 7A,C; Model 1), two disulfide bonds, C30− C79 and C33−C73, form the core ring, and C4 and C46 form the disulfide linkage penetrating the ring. Potentially, the C15− C19, and C36−C86 disulfide bonds reinforce the structure, while C10 and C32 are free to disulfide bond with cysteines from another CRP1 molecule to form larger structures. Aggregation of recombinant CRP1 has been observed experimentally, which supported leaving two or more cysteines free for the oligomerization (Figure 6A). The second model 4078

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

Figure 7. Schematic representation of the cysteine slipknot of CRP1 (a) model 1 and (b) model 2. The cysteines involved with cystine knots are shown as red circles, and the cysteines involved in additional disulfide bonds are shown as yellow circles. The green circles represent the free cysteines that can participate in intermolecular disulfide bonds to form dimer or oligomers. The disulfide bonds are represented in blue lines connecting the circles. The residue at the amino terminus is labeled as N and the carboxyl terminus is labeled as C. The ribbon diagrams of modeled structures are shown for (c) model 1 and (d) model 2. The side chain atoms of all the cysteine residues are shown and so are the disulfide bonds. The details of the cystine knots are shown in the blowup images that follow the same coloring scheme as in (a) and (b). The residues involved in the slipknots are shown in red and the residue numbers of cysteines are listed. The three pairs of cystines side chain atoms that form the disulfide slipknot are 4−46, 30−73, and 33−79 in model 1 are shown in (c), and 4−79, 15−36, and 19−32 in model 2 are shown in (d). The cysteines shown in green (32 in model 1 and 33 in model 2) are free unoxidized cysteines that can participate in polymerization.

cystine participating in two different parts of the backbone via the ring.28 These structures are known to be found within some of the most stable proteins, providing tremendous thermal and mechanical stability as well as resistance to enzymatic degradation,26 which are also characteristics of spider silks. The discovery of the CRP family members in dragline silk has escaped many investigations, largely due to the chemical properties of these molecules. Although the amino acid composition of natural dragline silk from black widow spiders has been determined, most procedures utilize acid hydrolysis.29 This method destroys Cys residues, reducing the potential to detect the CRPs. Additionally, the removal of the spinning dope is often accompanied by contaminating epithelial tissue, which represents a large cellular component of the MA gland that complicates protein analysis. Other reasons for the lack of CRP detection include the limited number of MS/MS analyses performed on solubilized dragline silk coupled with the enzymatically resistant nature of cystine slipknot family members, which introduce their own set of unique challenges using mass spectrometry. We show that treatment of the MA spinning dope solution with reducing agent increases the accumulation of monomeric CRPs, supporting the assertion that these molecules are stored in macromolecular complexes in the glands (Figure 5). Based on SDS-PAGE analysis and silver staining, the CRPs, along with MaSp1 and MaSp2, are extremely abundant in the MA dope (Figure 5). Our data is supportive that the CRPs are

(Figure 7B,D; Model 2) was built by forming the core ring with C15−C36 and C19−C32 disulfide bonds, with C4 forming the disulfide bond with C79 through the ring. Additionally, two pairs of disulfide bonds, C10−C86 and C30−C73, function to stabilize the protein. The cysteines C33 and C46 remain unoxidized, and can allow for higher order oligomerization. More discrete secondary structures, such as α-helices and βsheets, could potentially form in the structure. For example, in model 1, resides 71−82 could represent an α-helix while still forming the same cystine slipknot structure. Also, another disulfide bond variation is possible in model 2, where C86 can form a disulfide bond with C46, leaving C10 free. Taken together, molecular modeling of the primary sequence of CRP1 reveals the potential for formation of a slipknot structure.

■

DISCUSSION Our observations demonstrate that a family of low-molecular weight cysteine-rich proteins are expressed in the MA gland of black widow spiders, assembled into high molecular weight complexes, and are extruded into dragline silk. Further evidence that these molecules represent a family of proteins that display silk-gland restricted expression is supported by a recent transcriptome analysis.27 Molecular modeling implies these factors share structural similarities to proteins that belong to the cystine slipknot family. Cystine knot motifs involve three pairs of cysteines connected by disulfide bonds, with two of the cystines involved in the formation of a closed ring and the third 4079

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

Article

most synthetic fibers spun from MaSp1 recombinant proteins have displayed lower breaking strains relative to natural fibers, a property that many investigators have attributed to the absence of MaSp2 molecules and their GPGXX modules.4 Although this may be one of the reasons, synthetic fibers spun from composites containing truncated recombinant MaSp1 and MaSp2 have also failed to produce breaking strain and toughness’ values that are comparable to natural spider silk. Interestingly, the CRPs share some structural similarities to toxins and spider venom peptides, a group of molecules known to have thermal and structural stability.34,35 This may imply that these molecules have evolved to improve the mechanical properties of dragline silk. Many interesting research questions about the CRP family remain to be elucidated, including what chemical processes control their assembly into large macromolecular complexes, their impact on the mechanical properties of dragline silk, and whether they are present in other species, such as the golden orb-weaver, Nephila clavipes. PCR analysis using a cDNA library prepared from the silk-producing glands of N. clavipes as a template appears to support the presence of an amplified product that matches the approximate size of the black widow CRP cDNA family members, suggesting the CRPs may be found in other spider species. For several years, progress toward production of synthetic spider fibers that mimic natural fibers has remained elusive. The discovery of this family of proteins undoubtedly opens an exciting doorway for the scientific community, having the potential to provide missing molecules that will advance materials development of synthetic spider silks.

preassembled into large molecular weight complexes that are stored in the MA gland. Recently, a new paradigm that is based on structural mechanics instead of materials science has demonstrated that the insertion of simple knots or sliders into fibers can produce unprecedented toughness, a property resulting from the structures functioning as frictional elements that dissipate energy.30 Mathematical modeling demonstrates that the insertion of slipknots into fibers can mimic the plastic constitutive law of spider silk.30 One intriguing question regarding the discovery of the CRPs within the spinning dope and spun fibers is their biological function. Given the large number of cysteine residues within the amino acid sequence of the CRP family and the potential to form disulfide bridges, one hypothesis is that these factors serve a structural role in dragline silk. Could spiders have evolved molecules with slipknots that attach to the fibroin protein chains at the nanoscale level? If the CRPs are covalently linked to fibroins in natural silk threads, the conserved Cys residue in the C-terminal domain of MaSp1 or MaSp2 would seem to represent a plausible molecular target. Analysis of the N-terminal protein sequences of MaSp1 and MaSp2 also reveal the presence of Cys residues, however, these residues are likely removed from the mature protein.5 Although we have demonstrated the CRPs can form higher molecular weight complexes through self-polymerization, further investigations will be required to determine whether CRPs interact with the fibroins or hetero-oligomerize with other CRP family members. For black widow spiders, biochemical data support a 3:1 molecular ratio of MaSp1 to MaSp2 in spun fibers.29 Based upon our preliminary amino acid composition analysis of black widow dragline silk under oxidizing conditions, a method designed to allow quantitation of cysteine residues, our data reveal the presence of a higher cysteine content relative to theoretical values calculated from the full-length primary amino acid sequences of MaSp1 and MaSp2, which both share a single cysteine residues in the conserved C-terminal domains. If we assume the cysteine residue content detected by our amino acid analysis correlates with the presence of CRP and MaSp molecules, we can calculate an approximate ratio of 1:3:1 of CRP to MaSp1 to MaSp2 molecules in the fibers, respectively. To date, the nanostructure of silk fibrils has been described as two major constituents. These include a beta-sheet nanocrystal region embedded in a semiamorphous phase.31 The semiamorphous domains are rich in glycine and poly(Ala) segments that are less oriented relative to the poly(Ala)-rich nanocrystal regions, which consist of highly ordered antiparallel beta sheets.32 The toughness of dragline silk has been attributed to the highly extensible semiamorphous domains, which represent regions that unfold as hydrogen bonds are broken during fiber pulling. Intriguingly, the integration of slipknots fits the behavior of the plastic region from natural dragline silk fibers, and potentially could represent a secondary mechanism that contributes to the toughness of spider silk. The function of slipknots into fibroin chains would not have to be mutually exclusive of the current model that describes the role of the semiamorphous region and its contributions to toughness. In future studies, it will be interesting to examine the structural impact of the CRP family members on synthetic spider silk fibers spun from composites consisting of recombinant expressed MaSp and CRP molecules. One study, which used recombinant MaSp1 protein molecules approaching the molecular mass of native MaSp1, was able to approach the mechanical properties of the natural fibers.33 To date, however,

■

CONCLUSIONS We have identified a new family of low-molecular weight cysteine-rich proteins that display silk-gland restricted patterns of expression. These proteins are secreted into the spinning dope of the major ampullate gland and are assembled into large molecular weight complexes, which is dependent upon disulfide bridge formation. In addition, we demonstrate their expression patterns are linked to dragline silk production. Recombinant expression of the CRPs also reveals these proteins have the ability to polymerize into large molecular weight complexes. Collectively, our data support a role of the CRPs in the silk assembly process and a potential structural role in the extruded dragline silk fibers.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: cvierra@pacific.edu. Phone: 209-946-3024. Author Contributions ‡

These authors contributed equally to this work (T.P., T.C., and A.L.). Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was funded by the National Science Foundation RUI Grant DMR-1105310 entitled “The Molecular Mechanisms and Mechanical Behavior of Spider Glue Silks”. We are grateful for the MS and MS/MS contributions from Dr. Arnie Falick, senior scientist at HHMI in the San Francisco Bay Area. 4080

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081

Biomacromolecules

■

Article

(21) Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. UCSF Chimeraa visualization system for exploratory research and analysis. J. Comput. Chem. 2004, 25 (13), 1605−12. (22) Lane, A. K.; Hayashi, C. Y.; Whitworth, G. B.; Ayoub, N. A. Complex gene expression in the dragline silk producing glands of the Western black widow (Latrodectus hesperus). BMC Genomics 2013, 14, 846. (23) Jiang, L.; Liu, C.; Duan, Z.; Deng, M.; Tang, X.; Liang, S. Transcriptome analysis of venom glands from a single fishing spider Dolomedes mizhoanus. Toxicon 2013, 73, 23−32. (24) Vitt, U. A.; Hsu, S. Y.; Hsueh, A. J. Evolution and classification of cystine knot-containing hormones and related extracellular signaling molecules. Mol. Endocrinol. 2001, 15 (5), 681−94. (25) McDonald, N. Q.; Lapatto, R.; Murray-Rust, J.; Gunning, J.; Wlodawer, A.; Blundell, T. L. New protein fold revealed by a 2.3-A resolution crystal structure of nerve growth factor. Nature 1991, 354 (6352), 411−4. (26) Sikora, M.; Cieplak, M. Formation of cystine slipknots in dimeric proteins. PLoS One 2013, 8 (3), e57443. (27) Clarke, T. H.; Garb, J. E.; Hayashi, C. Y.; Haney, R. A.; Lancaster, A. K.; Corbett, S.; Ayoub, N. A. Multi-tissue transcriptomics of the black widow spider reveals expansions, co-options, and functional processes of the silk gland gene toolkit. BMC Genomics 2014, 15, 365. (28) Sun, P. D.; Davies, D. R. The cystine-knot growth-factor superfamily. Annu. Rev. Biophys. Biomol. Struct. 1995, 24, 269−91. (29) Casem, M. L.; Turner, D.; Houchin, K. Protein and amino acid composition of silks from the cob weaver, Latrodectus hesperus (black widow). Int. J. Biol. Macromol. 1999, 24 (2−3), 103−8. (30) Pugno, N. M. The “Egg of Columbus” for making the world’s toughest fibres. PLoS One 2014, 9 (4), e93079. (31) Simmons, A. H.; Michal, C. A.; Jelinski, L. W. Molecular orientation and two-component nature of the crystalline fraction of spider dragline silk. Science 1996, 271 (5245), 84−7. (32) Du, N.; Liu, X. Y.; Narayanan, J.; Li, L.; Lim, M. L.; Li, D. Design of superior spider silk: from nanostructure to mechanical properties. Biophys. J. 2006, 91 (12), 4528−35. (33) Xia, X. X.; Qian, Z. G.; Ki, C. S.; Park, Y. H.; Kaplan, D. L.; Lee, S. Y. Native-sized recombinant spider silk protein produced in metabolically engineered Escherichia coli results in a strong fiber. Proc. Natl. Acad. Sci. U.S.A. 2010, 107 (32), 14059−63. (34) Craik, D. J.; Daly, N. L.; Waine, C. The cystine knot motif in toxins and implications for drug design. Toxicon 2001, 39 (1), 43−60. (35) Schalle, J.; Kampfer, U.; Schurch, S.; Kuhn-Nentwig, L.; Haeberli, S.; Nentwig, W. CSTX-9, a toxic peptide from the spider Cupiennius salei: amino acid sequence, disulphide bridge pattern and comparison with other spider toxins containing the cystine knot structure. Cell. Mol. Life Sci. 2001, 58 (10), 1538−45. (36) Bendtsen, J. D.; Nielsen, H.; von Heijne, G.; Brunak, S. Improved prediction of signal peptides: SignalP 3.0. J. Mol. Biol. 2004, 340 (4), 783−95.

REFERENCES

(1) Gosline, J. M.; Guerrete, P. A.; Ortlepp, C. S.; Savage, K. N. The mechanical design of spider silks: from fibroin sequence to mechanical function. J. Exp. Biol. 1999, 202, 3295−3303. (2) Xu, M.; Lewis, R. V. Structure of a protein superfiber: spider dragline silk. Proc. Natl. Acad. Sci. U.S.A. 1990, 87 (18), 7120−4. (3) Hinman, M. B.; Lewis, R. V. Isolation of a clone encoding a second dragline silk fibroin Nephila clavipes dragline silk is a twoprotein fiber. J. Biol. Chem. 1992, 267 (27), 19320−4. (4) Hayashi, C. Y.; Lewis, R. V. Molecular architecture and evolution of a modular spider silk protein gene. Science 2000, 287 (5457), 1477− 9. (5) Ayoub, N. A.; Garb, J. E.; Tinghitella, R. M.; Collin, M. A.; Hayashi, C. Y. Blueprint for a high-performance biomaterial: fulllength spider dragline silk genes. PLoS One 2007, 2, e514. (6) Guerette, P. A.; Ginzinger, D. G.; Weber, B. H.; Gosline, J. M. Silk properties determined by gland-specific expression of a spider fibroin gene family. Science 1996, 272 (5258), 112−5. (7) Jenkins, J. E.; Sampath, S.; Butler, E.; Kim, J.; Henning, R. W.; Holland, G. P.; Yarger, J. L. Characterizing the secondary protein structure of black widow dragline silk using solid-state NMR and X-ray diffraction. Biomacromolecules 2013, 14 (10), 3472−83. (8) Trancik, J. E.; Czernuszka, J. T.; Bell, F. I.; Viney, C. Nanostructural features of a spider dragline silk as revealed by electron and X-ray diffraction studies. Polymer 2006, 47 (15), 5633− 5642. (9) Gould, S. A.; Tran, K. T.; Spagna, J. C.; Moore, A. M.; Shulman, J. B. Short and long range order of the morphology of silk from Latrodectus hesperus (Black Widow) as characterized by atomic force microscopy. Int. J. Biol. Macromol. 1999, 24 (2−3), 151−7. (10) Stark, M.; Grip, S.; Rising, A.; Hedhammar, M.; Engstrom, W.; Hjalm, G.; Johansson, J. Macroscopic fibers self-assembled from recombinant miniature spider silk proteins. Biomacromolecules 2007, 8 (5), 1695−701. (11) Ittah, S.; Cohen, S.; Garty, S.; Cohn, D.; Gat, U. An essential role for the C-terminal domain of a dragline spider silk protein in directing fiber formation. Biomacromolecules 2006, 7 (6), 1790−5. (12) Lazaris, A.; Arcidiacono, S.; Huang, Y.; Zhou, J. F.; Duguay, F.; Chretien, N.; Welsh, E. A.; Soares, J. W.; Karatzas, C. N. Spider silk fibers spun from soluble recombinant silk produced in mammalian cells. Science 2002, 295 (5554), 472−6. (13) Scheller, J.; Guhrs, K. H.; Grosse, F.; Conrad, U. Production of spider silk proteins in tobacco and potato. Nat. Biotechnol. 2001, 19 (6), 573−7. (14) Teule, F.; Cooper, A. R.; Furin, W. A.; Bittencourt, D.; Rech, E. L.; Brooks, A.; Lewis, R. V. A protocol for the production of recombinant spider silk-like proteins for artificial fiber spinning. Nat. Protoc. 2009, 4 (3), 341−55. (15) An, B.; Hinman, M. B.; Holland, G. P.; Yarger, J. L.; Lewis, R. V. Inducing beta-sheets formation in synthetic spider silk fibers by aqueous post-spin stretching. Biomacromolecules 2011, 12 (6), 2375− 81. (16) Dicko, C.; Vollrath, F.; Kenney, J. M. Spider silk protein refolding is controlled by changing pH. Biomacromolecules 2004, 5 (3), 704−10. (17) Hu, X.; Kohler, K.; Falick, A. M.; Moore, A. M.; Jones, P. R.; Sparkman, O. D.; Vierra, C. Egg case protein-1. A new class of silk proteins with fibroin-like properties from the spider Latrodectus hesperus. J. Biol. Chem. 2005, 280 (22), 21220−30. (18) Jeffery, F.; La Mattina, C.; Tuton-Blasingame, T.; Hsia, Y.; Gnesa, E.; Zhao, L.; Franz, A.; Vierra, C., Microdissection of black widow spider silk-producing glands. J. Vis. Exp. 2011, (47). (19) Sali, A. Comparative protein modeling by satisfaction of spatial restraints. Mol. Med. Today 1995, 1 (6), 270−7. (20) Lapthorn, A. J.; Harris, D. C.; Littlejohn, A.; Lustbader, J. W.; Canfield, R. E.; Machin, K. J.; Morgan, F. J.; Isaacs, N. W. Crystal structure of human chorionic gonadotropin. Nature 1994, 369 (6480), 455−61. 4081

dx.doi.org/10.1021/bm5011239 | Biomacromolecules 2014, 15, 4073−4081