Article pubs.acs.org/jpr
Comparative Proteomics Reveal Diverse Functions and Dynamic Changes of Bombyx mori Silk Proteins Spun from Different Development Stages Zhaoming Dong,† Ping Zhao,† Chen Wang, Yan Zhang, Jianping Chen, Xin Wang, Ying Lin, and Qingyou Xia* State Key Laboratory of Silkworm Genome Biology, Southwest University, 216, Tiansheng Road, Beibei, Chongqing 400716, China S Supporting Information *
ABSTRACT: Silkworms (Bombyx mori) produce massive amounts of silk proteins to make cocoons during the final stages of larval development. Although the major components, fibroin and sericin, have been the focus for a long time, few researchers have realized the complexity of the silk proteome. We collected seven kinds of silk fibers spun by silkworm larvae at different developmental stages: the silks spun by new hatched larvae, second instar day 0 larvae, third instar day 0 larvae, fourth instar day 0 larvae, and fourth instar molting larvae, the scaffold silk used to attach the cocoon to the substrate and the cocoon silk. Analysis by liquid chromatography−tandem mass spectrometry identified 500 proteins from the seven silks. In addition to the expected fibroins, sericins, and some known protease inhibitors, we also identified further protease inhibitors, enzymes, proteins of unknown function, and other proteins. Unsurprisingly, our quantitative results showed fibroins and sericins were the most abundant proteins in all seven silks. Except for fibroins and sericins, protease inhibitors, enzymes, and proteins of unknown function were more abundant than other proteins. We found significant change in silk protein compositions through development, being consistent with their different biological functions and complicated formation. KEYWORDS: silk, fibroin, sericin, protease inhibitor, enzyme, proteomics, Bombyx mori
■
8 and 13 kDa, respectively.13 It has been suggested that the seroins and protease inhibitors are involved in protection of the cocoon against predators and microbes. Besides fibroins, sericins, seroins, and protease inhibitors, few other proteins are known as components of the silk of B. mori. Fibroins are secreted from the posterior silk gland cells, and sericins are secreted from the middle silk gland cells. We speculate that other secreted proteins from silk gland cells may be associated with the silk fiber. Abundant proteins from the silk gland were identified by mass spectrometry after separation by 2D gel electrophoresis, including 74 proteins from the middle silk gland and 88 proteins from posterior silk gland of the fifth instar silkworm larvae.14,15 Besides fibroins and sericins, middle silk gland proteins included abundant cytoskeleton proteins, oxidoreductases and storage proteins; and posterior silk gland proteins included abundant heat shock proteins, ribosomal proteins, cytoskeleton proteins, oxidoreductases, and proteases.14,15 Oxidoreductases, proteases, and storage proteins
INTRODUCTION Silk fibers from the cocoon of the domestic silkworm (Bombyx mori) have been used in textiles for nearly 5000 years. B. mori silk is mainly composed of fibroins and sericins.1 Fibroins are oriented fiber protein complexes that consist of 350 kDa heavy chains, 25 kDa light chains, and a 30 or 27 kDa fibrohexamerin/ P25 depending on the degree of glycosylation.2−4 The heavy and light chains are linked by a single disulfide bond, and P25 associates with disulfide-linked heavy and light chains by noncovalent interactions.4−6 Quantitative ELISA assay revealed that molar ratios of the heavy chain, light chain, and P25 were 6:6:1 in the cocoon silk.4 Sericins are soluble glue proteins that coat and cement silk fiber. 400 kDa sericin 1 (sericin M) and 250 kDa sericin 3 (sericin A) are major coating proteins of cocoon silk, whereas 220−230 and 120−130 kDa sericin 2 appear to be major coating proteins of pad silk and scaffold silk.7−9 The pad silk provides a firm holding for the larva, and the scaffold silk fixes the cocoon to a suitable substrate.7 Previous studies of the B. mori cocoon extracts revealed the presence of four small proteins. Two are cysteine-rich serine protease inhibitors, a 6 kDa member of the Kunitz family and a 5 kDa member of the Kazal family.10−12 The other two proteins, seroin 1 and seroin 2, are glycoproteins with molecular weight of © XXXX American Chemical Society
Special Issue: Agricultural and Environmental Proteomics Received: June 18, 2013
A
dx.doi.org/10.1021/pr4005772 | J. Proteome Res. XXXX, XXX, XXX−XXX
Journal of Proteome Research
Article
scan were automatically selected for fragmentation by higher energy collisional dissociation with normalized collision energies of 27%. The maximum ion injection times for the survey scan and the MS/MS scans were 20 and 60 ms, respectively, and the ion target value for both scan modes was set to 1 × 106. Repeat sequencing of peptides was allowed after 18 s.
are generally secreted proteins, but there are no studies showing they can be found in the solid silk fiber. To better understand the complexity of protein constituents of the silkworm silks, we used shotgun liquid chromatography− tandem mass spectrometry (LC−MS/MS) to analyze seven silk proteomes, including the silks produced by each larval instar, the scaffold silk used to attach the cocoon to the substrate, and the cocoon silk. Many proteins of unknown function were identified, some of which showed very high abundances and may be novel fibroins or sericins. This study suggested that silk protein components of silkworms are more complex than previously believed. Our results provide the first comprehensive view of the changing composition of the silks of B. mori as their functions change through development.
■
Data Analysis
Raw MS files were analyzed by MaxQuant version 1.3.0.1.16 MS/ MS spectra were searched against an integrated silkworm proteome database, containing 22 117 protein sequences from NCBI and silkDB (downloaded on June 18, 2012)17,18 using the Andromeda search engine19 with the initial precursor and fragment mass tolerances set to 6 and 20 ppm, respectively. The search included variable modifications of methionine oxidation and N-terminal acetylation and fixed modification of carbamidomethyl cysteine. Minimal peptide length was set to six amino acids, and a maximum of two miscleavages was allowed. Both peptide and protein identifications were filtered at a 1% false discovery rate. In the case of identified peptides that are shared between two proteins, these are combined and reported as one protein group.20 A minimum of one unique peptide was required for protein identification (Supplementary Figure S1 and Tables S2 in the Supporting Information). Protein tables were filtered to eliminate the identifications from the reverse database and common contaminants. Peptide and protein tables are given as Supplementary Tables S1 and S2 in the Supporting Information.
MATERIALS AND METHODS
Material
B. mori, strain p50 (DaZao), were provided by the State Key laboratory of Silkworm Genome Biology, Southwest University. The silkworms were reared on mulberry leaves at a stable temperature of 25 °C. A total of seven silks were analyzed, including silks produced by animals from five different developmental stages (newly hatched larvae (baby silk), second instar day 0 (II-0 silk), third instar day 0 (III-0 silk), fourth instar day 0 (IV-0 silk), molting fourth instar (IV-M silk)), scaffold silk (the silk used to attach the cocoon to the substrate), and cocoon silk. All were collected and stored in 1.5 mL microfuge tubes at 4 °C until required.
Label-Free Quantification
To compare the abundances of different proteins within a single sample, we used the iBAQ (intensity-based absolute quantification) algorithm,21 which essentially normalizes the summed peptide intensities by the number of theoretically observable peptides of the protein. The iBAQ algorithm is now implemented into MaxQuant and can also be used to estimate protein abundances. For comparison between samples, we used label-free quantification (LFQ) with a minimum of two ratio counts to determine the protein intensity.20,22 Both unique peptides and razor peptides were chosen for use in label-free quantification. We assumed that the relative intensity of fibroin heavy chain to other proteins was essentially the same in all of the silks (its intensity was normalized as 100 000); then, the relative intensity of each protein was normalized to the fibroin heavy chain. The estimates of protein intensity are presented in Supplementary Table S2 in the Supporting Information.
Protein Digestion
Proteins were extracted from silks using three different methods. Baby silk, II-0 silk, III-0 silk, and IV-0 silk were sonicated at 120 W for 5 min in 1 mL of 9 M LiSCN and then vortexed for 2 h. IVM silk was dissolved in 1 mL of 6 M urea after grinding in liquid nitrogen. The remaining insolubilized IV-M silk protein was recovered by centrifugation (12 000g, 10 min, 4 °C) and then dissolved in 1 mL of 9 M LiSCN with vortexing for 2 h. The ureasolubilized and LiSCN-solubilized IV-M silk proteins were then merged into one solution. The scaffold silk and cocoon silk were cut into small fragments and then dissolved in 1 mL of 9 M LiSCN by vortexing for 2 h. The solubilized proteins were recovered by centrifugation (12 000g, 10 min, 4 °C). Soluble proteins were reduced by incubation with dithiotreitol (DTT) (10 mM) for 150 min at 37 °C and then alkylated by incubation with iodoacetamide (IAA) (50 mM) for 40 min in the dark. The samples were then recovered in an ultrafiltration tube (MWCO 10 000, Millipore, USA) using centrifugation at 12 000g, 4 °C for 40 min and washed twice with urea (8 M) and then twice with NH4HCO3 (50 mM). Silk proteins were digested with trypsin (1 μg trypsin per 25 μg protein) overnight at 37 °C in 150 μL of 50 mM NH4HCO3. Tryptic peptides were recovered by centrifugation in the ultrafiltration tube, resuspended in 1% formic acid, and then lyophilized.
Gene Ontology
Gene ontology (GO) 23 assignments were made using Blast2GO.24 The BLASTp searches were done against the nonredundant database with an expectation value maximum of 1 × 10−3. Annotation was made using the following criteria: Evalue-hit-filter, 1 × 10−3; annotation cutoff, 30; GO weight, 5. Building and Visualization of the Extracellular Matrix Proteins Interaction Network
Liquid Chromatography−Tandem Mass Spectrometry
To determine which of the proteins identified from the silks might be derived from the extracellular matrix of the silk gland, human extracellular matrix protein sequences were downloaded from MatrixDB25,26 and used as queries to search among the identified silkworm silk proteins by BLASTp (1 × 10−5). The UniProtKB Accession Numbers of the putative extracellular matrix proteins were used to build an interaction network model using the tools provided on the website of MatrixDB.25,26 Cytoscape, a software environment for integrated models of
Tryptic peptides were separated on a Thermo Fisher Scientific EASY-nLC 1000 system using a Thermo Fisher Scientific EASYSpray column (C18, 2 μm, 100 Å, 75 μm × 50 cm) with a 2− 100% acetonitrile gradient in 0.1% formic acid over 120 min at a flow rate of 250 nL/min. The separated peptides were analyzed using a Thermo Scientific Q Exactive mass spectrometer operating in data-dependent mode. Up to 10 of the most abundant isotope patterns with charge ≥2 from an initial survey B
dx.doi.org/10.1021/pr4005772 | J. Proteome Res. XXXX, XXX, XXX−XXX
Journal of Proteome Research
Article
Figure 1. Number of the identified proteins in seven B. mori silks. Two repeat experiments were performed for each silk. Unique proteins from one experiment are shown in blue, and unique proteins from the other experiment are shown in yellow, while proteins identified in both are shown in green.
translation-related proteins, and folding-related proteins. The observed extracellular matrix proteins mainly included cuticular proteins, heparin sulfate proteoglycan core proteins, and fibrillins. The identified protease inhibitors mainly belonged to serpin family and cysteine-rich protease inhibitor families. Cytoskeleton-related proteins found in the silks were actin, tubulin, and their binding proteins.
biomolecular interaction networks, was used to visualize the interaction with further manual adjustment.27 Prediction of Signal Peptides
SignalP 4.0 Server was used to predict the presence of the signal peptides (http://www.cbs.dtu.dk/services/SignalP/).
■
RESULTS
Comparison of the Relative Abundance of Proteins in the Seven B. mori Silk Proteomes
Proteome-Wide Identification and Classification of Seven Silks from B. mori
SDS-PAGE shows that the pattern of proteins changes between silks from different life stages (Supplementary Figure S2 in the Supporting Information). We estimated the relative molar abundance of the proteins within each silk using LC-MS/MS and the iBAQ algorithm, as described in the experimental procedures (Figure 2A and Supplementary Table S2 in the Supporting Information). Unsurprisingly, the classic silk proteins (fibroins and sericins) were the most abundant proteins, ranging from 46.2 to 74.7% by molar abundance. Proteins of unknown functions were relatively abundant in all seven silks (12.1 to 22.4%). Enzymes appeared to be more abundant in III-0 silk (13.9%), IV-0 silk (16.7%), and IV-M silk (12.9%) over baby silk (1.8%) and cocoon silk (0.7%). Protease inhibitors were very abundant in the scaffold silk (28.8%). The total abundance of proteins from the other four categories was
