Proteomic and Peptidomic Characterization of the Venom from the

The bird spider Ornithoctonus huwena Wang is a very venomous spider in China. Several ... about 38 000 described spider species.1 Spider venom is a...
1 downloads 0 Views 338KB Size
Download PDF
Proteomic and Peptidomic Characterization of the Venom from the Chinese Bird Spider, Ornithoctonus huwena Wang Chunhua Yuan,† Qihui Jin,† Xing Tang, Weijun Hu, Rui Cao, Shengqing Yang, Jixian Xiong, Chunliang Xie, Jinyun Xie, and Songping Liang* Key Laboratory of Protein Chemistry and Developmental Biology of the Ministry of Education, Life Science College, Hunan Normal University, Changsha 410081, People’s Republic of China Received January 14, 2007

The bird spider Ornithoctonus huwena Wang is a very venomous spider in China. Several compounds with different types of biological activities have been identified previously from the venom of this spider. In this study, we have performed a proteomic and peptidomic analysis of the venom. The venom was preseparated into two parts: the venom proteins with molecular weight (MW) higher than 10 000 and the venom peptides with MW lower than 10 000. Using one-dimensional gel electrophoresis (1-DE), two-dimensional gel electrophoresis (2-DE), and mass spectrometry, 90 proteins were identified, including some important enzymes, binding proteins, and some proteins with significant biological functions. For venom peptides, a combination of cation-exchange and reversed-phase chromatography was employed. More than 100 components were detected by mass spectrometry, and 47 peptides were sequenced by Edman degradation. The peptides display structural and pharmacological diversity and share little sequence similarity with peptides from other animal venoms, which indicates the venom of O. huwena Wang is unique. The venom peptides can be classified into several superfamilies. Also it is revealed that gene duplication and focal hypermutation have taken place during the evolution of the spider toxins. Keywords: spider venom • Ornithoctonus huwena Wang • proteomics • peptide • Edman sequencing • evolution

Introduction During about 400 million years of evolution, there have been about 38 000 described spider species.1 Spider venom is a complex mixture of components which exhibit a diverse array of actions both on prey and on human victims. Most of these components are biologically active proteins that function to kill or immobilize the prey. Despite the chemical diversity in spider venoms, encompassing a range of small organic molecules as well as large proteins, the vast majority of spider toxins are small polypeptides which display structural and pharmacological diversity.2 Previous research has identified a mass of significant components, which show different biological function, including enzymatic, cytotoxic, insecticidal, and hemagglutinic activities.3-5 To date, more than 100 representative peptide toxins which have effects on ion channels (potassium, calcium, sodium, and acid-sensing channels), glutamate receptors, and lipid bilayers have been identified.6-10 Spider venoms contain such rich and diverse resources of novel pharmacologically and agrochemically interesting compounds that all those identified components are just the tip of an iceberg. However, a complete systematical proteomic profile of spider venoms has not been performed yet. * To whom correspondence should be addressed. Phone: 86-731-8861304. Fax: 86-731-8861304. E-mail: [email protected]. † These authors contributed equally to this work.

2792

Journal of Proteome Research 2007, 6, 2792-2801

Published on Web 06/14/2007

The Chinese bird spider, Ornithoctonus huwena Wang, distributed in the hilly areas of Yunnan and Guangxi in southern China, is a very venomous spider. In our previous work, the major components were isolated and purified from its venom by HPLC, and their properties were characterized in detail.11 However, most minor components could not be studied in this way due to the availability. Systematical research on the venom would become very important to understand the full function of the venom. In this study, a proteomic strategy combining multidimensional chromatography (GF, HPLC, SDS-PAGE, 2-DE) with mass spectrometry provided a solution for this purpose. This work presents, to the best of our knowledge, the first systematic investigation of spider venoms and offers much important information for the diversified functions and chemical structure of the toxins of O. huwena Wang.

Materials and Methods A. Materials. Sephadex G-75, immobilized pH gradient (IPG) DryStrips (3-10 linear), IPG buffer (pH 3-10 linear (L)), cover fluid, and Coomassie Blue dye were purchased from Amersham Pharmacia-Biotech (Uppsala, Sweden). DTT, iodoacetamide, trypsin (proteomics sequencing grade), and TFA were obtained from Sigma (St. Louis, MO). Acrylamide, bisacrylamide, urea, glycine, Tris, CHAPS, and SDS were from Amresco (Solon, OH). Acetonitrile was a domestic product (chromatogram grade). 10.1021/pr0700192 CCC: $37.00

 2007 American Chemical Society

research articles

Characterization of Chinese Bird Spider Venom

Other chemicals are domestic products (analytical grade). Deionized water was prepared with a tandem Milli-Q system and used for the preparation of all buffers. B. O. huwena Wang Venom Preparation. The venom from adult female O. huwena Wang was collected as described12 and immediately freeze-dried. C. Gel Filtration. Venom samples (100 mg) were loaded onto a Sephadex G-75 column (10 mm × 100 mm) preequilibrated with 50 mM NH4HCO3 (pH 6.8). Elution of venom was carried out using an equilibration buffer with a flow rate of 1.0 mL/ min at room temperature (25 °C). Protein elution was monitored at 280 nm. The eluted fractions were analyzed further by SDS-PAGE to check the separation efficiency. On the basis of the results from SDS-PAGE, the fractions containing proteins with a molecular weight (MW) of less than 10 000 were pooled for HPLC separation, and the remaining fractions were pooled for SDS-PAGE and 2-DE analysis. D. SDS-PAGE and 2-D Electrophoresis. Venom samples collected from GF with an MW of more than 10 000 were denatured and subjected to SDS-PAGE using an 11.5% separation gel and a 4.8% stacking gel. After electrophoresis, the gel was stained with Coomassie Brilliant Blue. 2-DE was performed according to ref 13. Solubilized proteins (1 mg) with an MW of more than 10 000 were mixed respectively with a rehydration solution containing 8 M urea, 4% CHAPS, 0.5% IPG buffer, pH 3-10 L, 18 mM DTT, 20 mM Tris base, and a trace of bromophenol blue to a total volume of 350 µL, and applied to IPG DryStrips (pH 3-10 L, 180 × 30 × 0.5 mm). After rehydration for 12 h, IEF was conduced automatically to a total of 44 kV h at 20 °C. The seconddimensional run was carried out on incontinuity SDS-polyacrylamide vertical slab gels 1 mm thick, with 12.5% separation, and 4.8% stacking gels in a Bio-Rad Protein II electrophoresis apparatus. The condensed gel was run at a 12.5 mA/gel constant current, and a separate gel was run at a 25 mA/gel constant current. An SDS marker for Mr calibration was added. After 2-DE, the gels were stained with Coomassie Blue. E. Protein Digestion. The Coomassie Blue-stained protein lanes or spots were excised and digested in-gel with trypsin as described by Hellmann et al.14 The digested samples were first analyzed using ESI-Q-TOF, and those of which with no results were subjected to MALDI-TOF-TOF analysis. F. Analysis Using ESI-Q-TOF and MALDI-TOF-TOF Mass Spectrometry. Peptide mixtures from in-gel digestion were analyzed by a Q-TOF mass spectrometer (Micromass, Manchester, U.K.) fitted with a nanoelectrospray ionization source (Micromass) as previously described.15 We basically selected the candidate peptides with probability-based Mowse scores (total score) that exceeded their thresholds, indicating a significant (or extensive) homology (p < 0.05) and referred to them as “hits”. The criteria were based on the manufacturer’s definitions (Matrix Science, Ltd.).16 Proteins that were identified with at least two peptides showing a score higher than 40 were validated without any manual validation. Those with at least two peptides showing a score lower than 40 and higher than 20 were systematically checked and/or interpreted manually to confirm or cancel the MASCOT suggestion. For proteins identified by only one peptide, the score must exceed 40, and its peptide sequence was systematically checked manually. Tandem mass spectra acquired on an ESI-Q-TOF mass spectrometer were interpreted de novo using masslynx 4.0 software. Contiguous stretches of seven or more amino acids with a 100% confidence call using the software’s default

parameters were collected and matched to the NCBI nonredundant protein database using the protein BLAST algorithm (http://www.ncbi.nlm.nih.gov/BLAST/). The tryptic digestion mixed peptides from 2-DE gel spots with no results identified by ESI-Q-TOF were loaded onto an AnchorChip target plate according to ref 17. Molecular weight information of the peptides was obtained by using a MALDITOF-TOF mass spectrometer (UltraFlex I, Bruker Daltonics) equipped with a nitrogen laser (337 nm) and operated in reflector/delay extraction mode for the MALDI-TOF peptide mass fingerprint (PMF) or LIFT mode for MALDI-TOF-TOF with a fully automated mode using the flexControl software. An accelerating voltage of 25 kV was used for PMF. The peaks with S/N g 5 and resolution g2500 were selected and used for LIFT-TOF-TOF MS/MS from the same target. LIFT spectra were interpreted with the Mascot software. The MS peptide tolerance was 50 ppm, and the MS/MS tolerance was 1.0 Da. The protein identifications were considered to be confident when the protein score of the hit exceeded the threshold significance score of 70 (p < 0.05). For samples from 2-DE, only the first hit protein was selected. G. Separation of Venom Peptides by HPLC. The fractions with an MW of less than 10 000 were pooled and injected on a Waters AP-1 column (Acell plus CM cation-exchange media, 10 mm × 100 mm) initially equilibrated with buffer A (0.02 M sodium phosphate, pH 8.4). The column was eluted using a linear gradient of 0-80% buffer B (1.0 M sodium chloride, pH 8.4) over 70 min at a constant flow rate of 2.5 mL/min. The peaks were monitored by absorbance at 280 nm. The peaks eluted from CM cation-exchange HPLC were applied to an analytical Vydac C18 reversed-phase HPLC column (300 Å, 4.6 mm × 250 mm) and eluted at a flow rate of 1.0 mL/min. The mobile phases were (A) 0.1% trifluoroacetic acid in water and (B) 0.1% trifluoroacetic acid in acetonitrile. The effluents were monitored at 280 nm. H. MALDI-TOF MS Analysis and Edman Sequencing. The peaks eluted from reversed-phase HPLC were collected and analyzed using MALDI-TOF MS. A 0.3 µL aliquot of each fraction was spotted onto a gold-plated target along with an equal volume of a matrix (15 mg of CCA, 600 µL of acetonitrile, 400 µL of water, and 3 µL of TFA). External mass calibration was performed using a peptide calibration standard (Bruker, Germany), and mass spectrometry was performed using an acceleration voltage of 25 kV. Native peptides with a purity of more than 90% (based on HPLC and MS analysis) were subjected to a Precise 491A sequencer (Applied Biosystems). Sequence homologies were determined using sequences obtained from literature data and a search of nonredundant protein databases, via the BLAST server (http://www.ncbi.nlm.nih.gov/BLAST). The fully sequenced peptides were edited using the Bioedit Sequence Alignment Editor software and then aligned and refined manually using the ClustalW1.8 program. A pairwise distance matrix was calculated on the basis of the proportions of different amino acids. The matrix was then used to construct phylogenetic trees by the neighbor-joining method (MEGA2.1). The reliability of branching patterns was assessed using 1000 bootstrap replications.

Results and Discussion A. Proteomic Strategy To Analyze the Spider Venom of O. huwena Wang. Due to limits in the sensitivity and resolution of routinely used methods of protein separation, the complexity Journal of Proteome Research • Vol. 6, No. 7, 2007 2793

research articles

Yuan et al.

Figure 1. Separation of the venomous proteins by GF (Sephadex G-75 column, 10 mm × 100 mm) (a) and analysis of the eluted fractions by SDS-PAGE with Coomassie Brilliant Blue staining (b).

Figure 2. Separation of the proteins (MW > 10 000) from the spider venom of O. huwena Wang: (a) Coomassie-stained 2-DE gel image of pH 3-10 linear; (b) Coomassie-stained SDS-PAGE gel image.

and diversity of spider venom are usually underestimated. Many components with low abundance are often omitted. To obtain overall identification of peptides and proteins, especially high molecular weight proteins (MW > 10 000), we used a proteomic strategy for this research. The venom was separated into four fractions (named FI, FII, FIII, and FIV) through gel filtration (Figure 1a). On the basis of the results of SDS-PAGE (Figure 1b), fractions FI and FII containing large proteins (MW > 10 000) were pooled and subjected to SDS-PAGE and 2-DE separation. Fractions FIII and FIV composed of peptides or small molecules (MW < 10 000) were pooled and subjected to HPLC separation. B. Proteome Maps of O. huwena Wang Venom. Two methods were adopted to separate venom fractions FI and FII from gel filtration with MW > 10 000: (a) 1-DE-MS/MS, (b) 2-DE-MS. A representative Coomassie Blue-stained 2-D gel is shown in Figure 2a. Approximately 300 spots were counted after background subtraction with a window of pI 4-9 and molecular weight 10000-100000. A majority of the spots with 2794

Journal of Proteome Research • Vol. 6, No. 7, 2007

relatively high abundance were located in the MW region between 40 000 and 100 000. Considering the protein loss in the 2-DE system, we separated the sample with MW > 10 000 by one-dimensional SDS-PAGE. The main protein bands ranged in MW from 30 000 to 100 000 (Figure 2b), and there are some protein bands with MW > 100 000 which were invisible in the 2-DE gel. All the bands were excised, in-gel digested, and identified by ESI-Q-TOF MS. Due to the absence of a spider venom database, the PMF or MS/MS data acquired were matched against the theoretical peptide mass of Metazoa proteins in the NCBI and Swiss-Prot databases. In all, 90 proteins were identified (Supporting Information Table 1), including enzymes, binding proteins, transport proteins, regulatory proteins, and so on, and 35 spots from 2-DE displaying high-quality ESI-Q-TOF MS/MS spectra (Figure 3) had no match. Those spots were subjected to de novo sequencing (Supporting Information Table 2). We categorized the identified proteins according to their function, on the basis of universal GO annotation terms: 27% of the proteins have

Characterization of Chinese Bird Spider Venom

research articles

Figure 3. MS/MS spectra for parent ion 730.59. The amino acid sequence, SKGTLLSLSLTALR, was confirmed by analyzing y- and b-ions derived from the peptide ion of spot 51.

enzyme activity, 18% are involved in cellular binding, 10% behave as receptors or inhibitors, 3% are structural proteins, 8% are transport proteins, and 7% display regulatory activities. In addition, the functions of the other 27% of proteins are unknown. C. Protein Identification. Many significant components have been identified from venomous animals such as snakes, scorpions, spiders, and bees. To date, at least eight toxin families have been analyzed (Kunitz-type protease inhibitors, CRISP toxins, galactose-binding lectins, M12B peptidases, nerve growth factor toxins, cystatin toxins, phospholipase A2, and natriuretic toxins).18 In the venom of O. huwena Wang, we found not only some important components similar to the above toxins, but some novel proteins with significant biological function discussed as follows. 1. Enzymes. Enzymes are always the common components in animal venoms,19 probably related to the derivation of the venom gland from the salivary gland. Here 24 enzymes involved in digestion, hydrolysis, and energetic metabolism have been identified in O. huwena Wang venom. Spot 73 was identified as dual-specificity phosphatase (DSP). It belongs to the protein tyrosine phosphatase (PTP) family, which plays an important role in regulating mitogenic signal transduction and controlling the cell cycle in response to extracellular stimuli. The DSPs are thought to be involved in important signaling events, ranging from the control of MAPKs (mitogen-activated protein kinases) in cell proliferation to the regulation of cyclin-dependent kinases in the cell cycle.20 Another protein, similar to phosphomannomutase 2 (PMM 2), is an isomerase which catalyzes the isomerization of mannose 6-phosphate to mannose 1-phosphate. A serine proteinase was also identified which is present in many snake venoms. Protein guanylyl cyclase GC-B is a receptor of the neuronal Ca2+ sensor (NCS) proteins, a family known to influence a variety of physiological and pathological processes by affecting signaling of different receptors and ion channels.21,22 Cytochrome c oxidase subunit II has been identified from the venomous glands of the scorpion Centruroides exilicauda.23 Nourseothricin acetyl transferase is a member of the acetyltransferase family which has been identified from the

snake venom of Trimeresurus stejnegeri.24 Prophenoloxidase, a key melanin-synthesizing enzyme, is considered to be an important arthropod immune protein. It has been shown to be involved in refractory mechanisms against malaria parasites. It has been identified from the insect venom of Catalpa speciosa caterpillar.25 The hypothetical protein CBG07634 and Eph2 belong to the Ser/Thr protein kinase family and the protein tyrosine kinase (PTK) family, which are located at the cell membrane and function as transmembrane conductors of signals. 2. Binding Proteins. The heat shock cognate 70 protein (HSC70), a constitutive member of the highly conserved heat shock protein 70 family, was identified in the venom. It is the uncoating ATPase for the disassembly of clathrin-coated vesicles. In a study using the differential display technique to compare the expression profile of breast carcinoma cells and nearby normal epithelial cells, HSC70 was found up-regulated in breast cancer cells.26 These findings make HSC70 an attractive candidate as a tumor-associated gene. Another heat shock protein identified in the higher molecular weight region was GRP78 (glucose-regulated protein, 78 000). It may play a role in facilitating the assembly of multimeric protein complexes known to exist in the venom and has been identified in Australian brown snake snake venom of Pseudonaja textiles.27 Some ATP-binding and GTP-binding proteins have also been identified in the venom. They may be related to ion channel signal transduction. 3. Others. A serine proteinase inhibitor was identified from the venom which has the possibility of development in human therapeutics. The identified proteins of muscle actin, actin, and myosin are ubiquitous components of the cytoskeleton which also have been identified from the venom C. speciosa caterpillar.25 Another protein of immunoglobulin heavy chain variable domain has been identified, which is glycoproteins in the immunoglobulin superfamily that function as antibodies. Previously, Jiraporn Nawarak had identified a similar immunoglobulin heavy chain in snake venom.28 Many nontoxic proteins including cytochrome c oxidase,23 nourseothricin acetyl transferase,24 prophenoloxidase, muscle Journal of Proteome Research • Vol. 6, No. 7, 2007 2795

research articles

Figure 4. CM cation-exchange chromatogram of fractions FIII and FIV (MW < 10 000). Fractions FIII and FIV were pooled and separated into seven major peaks (from CM1 to CM7) using a linear gradient of 0-80% buffer B (1.0 M sodium chloride, pH 8.4) over 70 min at a constant flow rate of 2.5 mL/min. The effluents were monitored by their absorbance at 280 nm.

actin,25 and HSC7026 have also been found in the venoms of other animals, such as scorpions, snakes, and insects. The

Yuan et al.

reason for the presence of the nontoxic components in the venom is complicated and remains a puzzle. As we thought, there are two possible reasons for these nontoxic proteins to be found in animal venoms. First, some nontoxic components may have some synergistic or protectic effects on the toxins in the venom. For example, the crotoxin acidic and nontoxic subunit CA can interact with phospholipase A2 in a synergistic manner,29 and protease inhibitors in snake venom can protect peptide toxins against hydrolyzing. Many other nontoxic proteins may have some kind of role in the venoms, which we do not know so far and need to investigate further. Second, it would also be possible that, in venom gland cells, there are a certain proportion of genes coding house-keeping proteins. These proteins are necessary to keep toxin-producing cells and the venom gland functional, and during the secretion of toxins, some of these house-keeping proteins are somehow secreted to venoms. We think it is still necessary to get more new

Figure 5. Analytical reversed-phase HPLC profiles of seven major peaks. Each major peak was applied to a Vydac C18 column (300 Å, 4.6 mm × 250 mm) and eluted at a flow rate of 1.0 mL/min using a gradient from 0 to 20% buffer B (0.1% trifluoroacetic acid in acetonitrile) over 20 min, followed by a gradient from 20% to 30% buffer B over 40 min. The effluents were monitored by their absorbance at 280 nm. 2796

Journal of Proteome Research • Vol. 6, No. 7, 2007

research articles

Characterization of Chinese Bird Spider Venom

Table 1. Mass Fingerprinting of Peptide Toxins (MW < 10 000) from the Spider Venom of O. huwena Wang retention time (min)

molecular mass [M + H+]

17.7 17.81 18.25 18.91 19.60 20.09 20.33 20.91

3768.4 4440.6 4465.1 4309.6, 4643.5, 4341.5, 4612.6 4483.9, 4513.8 3883.8 4183.9, 4212.8 3483.1, 1331.2

19.0 21.0 21.5 21.81 22.41 23.57 24.31 25.21

3481.0 3424.6 6793.2 4437.8 4450.7, 4418.1 2760.5 3949.5, 2067.4 3508.3

8.8 13.0 13.9 16.8 21.50 22.87 23.06 23.93 24.60 25.41 25.41

2066.3 3996.6 2419.0 4931 4467.5 3928.1 3961.1 3921.8, 3939.9 6432.4 3941.2, 3870.3 3941.2, 3870.3

12.9 14.5 22.7 23.49 25.44 25.6 26.4

5348.4 5675.2 4688.1 3955.1 3696.4 6743.6 6755.3

15.53 16.7 17.4 19 20.98 21.88

3338.9 5754.1 4612.5 4470.0 3646.3 3767.3

16.80 17.20 23.23 24.34 24.76 25.14

2952.3, 2779.1 2931.1 4048.9, 4080.9 4238.0 3971.0, 8795.2 3608.3

18.38 26.55 27.32 28.48 29.60 31.11

2712.1, 4023.5 3810.2 1994.1, 4299.3 3986.7 3980.0 4171.1

retention time (min)

molecular mass [M + H+]

CM1 21.36 22.64 23.92 25.44 26.26 26.80 27.84 28.95

3530.6, 4358.9 4032.8, 3295.0, 3528.1 6791.0, 6759.4 3664.9 4111.6 3562.0 3573.5 3714.1

25.45 27.00 28.05 30.90 37.32 38.60 45.8

5552.5, 5124.9 3374.2, 3515.3 3354.6 3842.4 3541.5 3554.6 6827.5

26.61 27.51 28.4 28.68 30.10 30.37 32.87 35.20 36.13 42.8

3907.1 6881.1 3721.2 4312.3 3671.3 3865.5 3852.8 2661.2, 2632.2, 2571.1 3624.0 3081.5

26.92 29.27 30.66 37.70 40.45 42.45

4721.8 4671.3 4691.0, 3900.6 3580.2 3137.2 6972.4

22.33 23.07 23.87 24.03 26.64 38.36

3591.2 3750.2 3925.4 3636.9 3784.2, 4132.3, 3799.3, 4144.7 3589.7, 8485.3

25.82 26.19 26.65 27.34 27.7 33.8

4253.1 4091.6, 3708.1 4108.8 4268.0, 3733.9 4087.4, 4332.8 4006.1

31.90 32.77 33.73 34.29 37.05

4072.4, 5412.4 6166.3 3490.6 4286.0 4088.7

CM2

CM3

CM4

CM5

CM6

CM7

experimental data to answer the question of why these nontoxic proteins are present in the venoms. Although many proteins have been identified in this paper, some common toxic components such as L-amino acid oxidase and hyaluronidases30 were not identified in this study. In our previous work, a component with a molecular weight of 40 700 was purified from O. huwena Wang spider venom and has been identified with the activity of hyaluronidase.31 Furthermore, 35 spots displaying high-quality ESI-Q-TOF MS/MS spectra matched no proteins. This may be due to two factors: (i)

absence of a database specific for spider venom; (ii) the low abundance of those components. D. Peptidomic Profile of Venom Peptides of the Chinese Bird Spider, O. huwena Wang. The pool of venom peptides (MW e 10 000) from gel filtration was applied to a cationexchange HPLC apparatus which gave seven different fractions (Figure 4). The fractions were further separated by reversedphase HPLC using the same gradients, which showed dissimilar elution patterns (Figure 5). The series of effluents were submitted to MALDI-TOF MS analysis. As shown in Table 1, about Journal of Proteome Research • Vol. 6, No. 7, 2007 2797

research articles

Yuan et al.

Table 2. Full or Partial Sequences, Molecular Weights, and Biological Activity of Peptides from the Spider Venom of O. huwena Wanga peptide

mass ([M + H]+)

CM1-17.7 CM1-17.81 CM1-20.09 (mHWTX-V) CM1-26.26 (HWTX-V) CM2-19 CM2-21 CM2-21.5 CM2-25.21 CM2-28.05 CM2-30.90 CM2-37.32 (SHL-I) CM2-38.60 CM2-45.8

3768.4 4440.6 3883.8 4111.6 3481 3424.6 6793.2 3508.3 3354.6 3842.4 3541.5 3554.6 6827.5

CM3-8.8 CM3-13 CM3-13.9 CM3-16.8

2066.3 3996.6 2419 4903.1

CM3-24.60

6432.4

CM3-27.51 CM3-28.4 CM3-30.10 (HWTX-IIIa) CM3-32.87 (HWTX-) CM3-42.8 CM4-12.9 CM4-14.5 CM4-22.7 CM4-25.44 CM4-25.6

6881.1 3721.2 3671.3 3852.8 3081.5 5348.4 5675.2 4688.1 3696.4 6743.2

CM4-26.4 CM4-40.45 CM4-42.45 CM5-16.7 CM5-17.4 CM5-19 CM5-23.07 (HWTX-) CM5-23.87 CM5-24.03 (HWTX-Ia) CM6-17.20 (HWTX-X) CM6-25.14 CM6-26.65 (HWTX-IV ) CM6-27.7 CM6-33.8 CM7-26.55 CM7-28.48 (HWTX-VII) CM7-31.11 (HWTX-VIII) CM7-32.77 (HWTX-XI)

6755.3 3137.2 6972.4 5754.1 4612.5 4470 3750.2 3925.4 3636.9 2931.1 3608.3 4108.8 4087.4 4006.1 3810.2 3986.7 4171.1 6166.3

CM7-34.29 (HWTX-II)

4286

a

partial or full sequence

SCAKPRETCNRMNILCCRGECVCPIIGDCFCYGD NCIGEQVPCDENDPRCCSGLVVLKKTLHGIWIKSSYCYKCK ECRWYLGGCSQDGDCCKHLQCHSNYEWCVWDGT ECRWYLGGCSQDGDCCKHLQCHSNYEWCVWDGTFS ECWSQADCSDGHCCAGSSFSKNCRPYGGDGAQCE ECWSQADCSDGHCCAGSSFSKNCRPYGGDGAQC ECWSQADCSDGHCCAGSSFSKNCRPYGGDGAQCEPRN... GCFGYKCDYYKGCCSGYVCSPTWKWCVRPGP GCLGDKCDYNNGCCSGYVCSRTWKWCVLAGP GCLGDKCDYNNGCCSGYVCSKTWKWCVLAGPWM... GCLGDKCDYNNGCCSGYVCSRTWKWCVLAGPW GCLGDKCDYNNGCCSGYVCSRTWKWCVLAGPW ECWSQADCSDGHCCAGSSFSKNCRPYGGDGAQCEPRNKYEVYSTGCPCDENLICSVINRCQSV FVK DCAKEGEVCSWGKICCDLDNFYCPMEFIPHCKGYP A(/V/D/G)LD(/S)G(/K)Q(/I)I(/G)GDRNESNPDGTTVS DRCKSNCDCCGTTVTCGTIYVGGKEVNQCMDKSSDNAVLNG IGKGWN ACSKNPGESCTNNCECCGATVVCASVYVAGVEKKSCKSKTSDNGFLNIIGQAANAVQNAAS LCV ECASQADCSDGHNILGSRFEFNNRP... GCLGDKCDYNNGCCSGYVCSRTWKWCVLAGPWRR DCAGYMRECKEKLCCSGYVCSSRWKWCVLPAP DCAGYMRECKEKLCCSGYVCSSRWKWCVLPAPW CGGWMAKCADSDDCCETFHCTRFNVCGK ACSKQIGDRCKSNCDCCGTTVS... ACSKQIGDRCKSNCDCCGTTVSCGTIYVGGCEVNQCM... CWGENVPCENKNSPRCCGLS... DCAKEGEVCSWGKKCCDLDNFYKPKEFIP... ACSKQIGDRCKSNCDCCGTTVSCGTIYVGGKEVNQCMDKSSDNAVLNGIGKGWNFVKNGFSFCV ACSKQIGDRCKSNCDGYGTT... IICAPEGGPCVAGIGCCAGLRCSGAKLGLAGSCQ ECASPFETNPMHHAAGSNFCVNPR... SVIKEKSTKENDCECCGMST... SCKKIKKEI‚‚‚ ACKGVFDACTPGKNECCP... ACKGVFDACTPGKNECCPNRVCSDKHKWCKWKL ACKGVFDACTPGKNECCPNRVCSDKHKWCKWKL ACKGVFDACTPGKNECCPNRVCSDKHKWCKWQ KCLPPGKPCYGATQKIPCCGVCSHNKCT ECKGFGKSCVPGKNECCSG... ECLEIFKACNPSNDQCCKSSKLVCSR KTRWCKYQI ECLEIFKACNPSNDQCCKSSKLVCSRKTRWCKYQI FECSISKEIECKGES... ACWKQIGDRCKSNNDCCGTCVTYGTIYV... FECSISCEIEKKGESCKPKKCKGGWKCKFNMCVKV FECSFSCEIEKEGDKPCKKKKCKGGWKCKFNMCVKV IDTCRLPSDRGRCKASFERWYFNGRTCAKFIYGGCGGNGNKFPTQEACMKRCAKA LFECSFSCEIEKEGDKPCKKKKCKGGWKCKFNMCVKV

biological activity

a neurotoxin toxic to mammals a natural mutant of HWTX-V an insecticidal neurotoxin 77% similarity with SHL-I hemagglutination activity hemagglutination activity modifier of SHL-I 94% similarity with SHL-I a natural mutant of HWTX-III a neurotoxin a neurotoxin toxic to mammals N-type calcium channel inhibitor modifier of HWTX-I a natural mutant of HWTX-I N-type calcium channel inhibitor sodium channel inhibitor an inhibitor of TTX-S Na+ channels modifier of HWTX-IV like that of HWTX-II like that of HWTX-II trypsin inhibitor and Kv1.1 channel inhibitor neurotoxin

Peptides named by HWTX- were identified previously. Key: -, unknown biology activity; ..., partial sequence.

133 components were detected, 77.4% of which fall in the 3000-5000 Da mass range. Components with high purity were subjected to sequencing. Peptides with full sequences and partial N-terminal sequences are listed in Table 2. The nomenclature of the peptides was coded by CMx (x represents the fraction number by cation-exchange HPLC) followed by the retention time on the RP-HPLC chromatogram. Most of the fully sequenced peptides contain six Cys residues which may form three disulfide bonds, except for peptides CM1-17.7, CM2-45.8, CM3-24.6, and CM4-25.6 containing eight Cys residues and CM3-13.9 having no Cys residues. The number of disulfide bonds was determined by comparing the theoretical molecular mass and the measured monoisotopic 2798

Journal of Proteome Research • Vol. 6, No. 7, 2007

mass. In those peptides with six Cys residues, according to our previous work, three patterns of disulfide bond linkage and consequentially three structural motifs have been determined. Most of the peptides have two adjacent Cys residues linked in the manner of 1-4, 2-5, 3-6 and folded as an ICK motif like HWTX-I and HWTX-IV.32,33 HWTX-II adopts a second structural scaffold of the DDH motif with six separated Cys residues, forming a 1-3, 2-5, 4-6 disulfide connectivity.34 HWTX-VII and HWTX-VIII might share the same structural scaffold with HWTX-II. The third important structure scaffold was determined for HWTX-XI, which adopts a Kunitz-type motif with a unique disulfide bond linkage of 1-6, 2-4, 3-5 (unpublished data). Other structural scaffolds for the peptides with eight Cys

Characterization of Chinese Bird Spider Venom

research articles

Figure 6. Phylogenetic relationships of venom peptides from O. huwena Wang (A) and comparison of amino acid sequences of peptides in cluster I (B) and cluster II (C). The numbers on the branches are the bootstrap percentages supporting a given partition. A dot indicates the residue identical with that of SHL-I or HWTX-II. The dash means blank.

residues and peptides devoid of Cys residues remain unknown. Further study should shed light on the three-demensional structure determination. The pharmacological effects of the sequenced peptides were characterized on voltage-gated channel level or animal models. HWTX-I was the first peptide identified from O. huwena Wang venom which selectively blocked N-type Ca2+ channels.35 In this study, two natural mutants of HWTX-I named CM5-24.03 and CM5-23.87 have been isolated. In CM5-24.03, replacement of the two amino acids Lys and Leu from the C-teminus of HWTX-I by Gln caused loss of the calcium channel of inhibitory activity of HWTX-I. A possible reason is that the C-terminal amino acid replacement causes spacial structure transformation. The peptide CM5-23.87 shares the same amino acid sequence with HWTX-I, but the molecular weight was 175.2 more than that of HWTX-I. Preliminary sequence analysis showed that this is possible due to some uncertain modification of this peptide. Another two calcium channel blockers, CM117.81 and HWTX-X, isolated from the fractions of CM1 and CM6, respectively, can moderately inhibit the calcium channel current in rat dorsal root ganglion cells.36 HWTX-IV is an important toxin with the activity of affecting neuronal TTXsensitive voltage-gated sodium channels.33 Here we isolated a new component named CM6-25.14, which can inhibit TTX-R sodium currents in adult rat dorsal root ganglion neurons, sharing 57% sequence similarity with the sodium channel toxin of HNTX-I.37 Eight components were isolated from the venom with activity affecting mammals or insects. The peptides of HWTX-II,

HWTX-VII, and HWTX-VIII can reversibly paralyze cockroaches and block neuromuscular transmission in the isolated mouse phrenic nerve-diaphragm preparation. Two other components, HWTX-III and HWTX-V, can reversibly paralyze cockroaches and locusts.11 Their natural mutants, mHWTX-III and mHWTXV, which are only truncated one or two amino acid residues from the C-terminus, lost the activity affecting mammals or insects. A new component named CM4-40.45 was isolated, which shows low homology with other toxins in the database. It can paralyze mice for 12 h by intracisternal injection of 4 µg. Peptides with other pharmacological effects were isolated from the venom. SHL-I, a relatively abundant component, can agglutinate human and mouse erythrocytes at minimal concentrations of 125 and 31 mg/mL, respectively.38 The agglutination activity of SHL-I can be inhibited by D-mannosamine.39 Four components (CM2-28.05, CM3-38.60, CM5-30.90, and CM5-25.21), sharing high sequence similarity with SHL-I, were isolated from the venom. The effects of the sequence changes to the activity remain uncertain. HWTX-XI, a component different from all the above toxins, was separated from the fraction of CM7. It functions as a Kunitz-type protease inhibitor and potassium channel blocker. Besides the above characterized toxins, the biological activities of many peptides in the venom remain unknown. For better understanding the pharmacological diversity of the venom and to predict the function of the peptides in Table 2, a phylogenetic tree was constructed using the NJ algorithm based on the sequence similarity (Figure 6A). Peptides in the venom of Journal of Proteome Research • Vol. 6, No. 7, 2007 2799

research articles O. huwena Wang display evident sequence diversity, and the 31 peptides form at least 6 clusters. Two of them (labeled I and II) are comparatively big, each containing six peptides with high sequence similarity. Cluster I contains SHL-I and HWTXIII and their homologues. Although these toxins exhibit different biological activities, they are evolutionarily close to one another. Cluster II includes HWTX-II and its homologues. These six peptides containing six or eight Cys residues display a close relationship in evolution, suggesting that they might share similar biological activities and structural scaffolds. The abundant toxic components of HWTX-I and HWTX-IV sharing high sequence similarity with HNTX-III and HNTX-IV, respectively, are located at two close clusters, indicating that the four toxins were evolutionarily close. The other clusters contain peptides sharing little sequence similarity with toxins in the database. What are the evolutionary mechanisms leading to such intense molecular diversification of animal venoms? It has been proposed that gene duplication and focal hypermutation based on a limited set of genes may have resulted in the enormous pharmacological diversity of conotoxins.40 Our results on spider venom support this theory. Figure 6 displays the alignment of the amino acid sequence of SHL-I and HWTX-II and their homologues. SHL-I and HWTX-IIIa have high similarity in the prepro regions, and their coding sequence together with that of the four components (HWTX-III, CM2-25.21, CM2-28.05, and CM3-28.4) share high similarity (Figure 6B), indicating that they perhaps evolved from the same ancestor, and the minor change in the coding sequence may be due to gene duplication and focal hypermutation. This is also true in the case of HWTX-II and its homologues (Figure 6C). However, SHL-I and HWTXII share low similarity in the prepro region, suggesting that they may belong to two subfamilies. This evolving process is double: most residues are mutated in just about any possible isoform, but at the same time, molecular mechanisms of transcription preserve the cysteine residues, resulting in a high conservation of the molecular scaffold.41 In the venom of O. huwena Wang, most peptides share a common ICK scaffold, but display evident sequence diversity. The venom components should originate from a limited set of gene superfamilies. It has been reported that snake venoms possess a variety of protein isoforms resulting from post-translational modifications, sequence homologues, and protein degradations.42 It is interesting that many components with slightly different molecular weights and close elution times were detected in the venom of O. huwena Wang (Table 1). Some of them were confirmed to share the same sequence, such as HWTX-I and CM5-23.87, which indicates that the difference in molecular mass was possibly caused by post-translational modifications. The study on peptide isoforms would provide much information on the variety of the venom and the relationship of the bioactivity and post-translational modification of the peptides.

Yuan et al.

and focal hypermutation have taken place during the evolution of the spider toxin.

Acknowledgment. This work is supported by a grant from the National Natural Science Foundation of China (30430170). We are grateful to Dr. Zhonghua Liu, Dr. Ying Wang, Dr. Lijun Zhang, and Dr. Zhi Liao in our laboratory and Dr. Jialie Luo of the Hong Kong University of Science & Technology for good suggestions and critical reading of this manuscript. Supporting Information Available: Supporting Tables 1 and 2. This material is available free of charge via the Internet at http://pubs.acs.org. References (1) (2) (3) (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) (25) (26) (27) (28)

Conclusion In this study, we utilized various chromatography technologies, SDS-PAGE, 2-DE, and MS, for the visualization of venom proteomes. This study demonstrates the great variation and diversity of spider venom components and provides a better understanding of venom complexity and toxic properties. Combining the methods of HPLC and physiology provided identification of many significant peptides which will be interesting to exploit in the areas of medicine and pesticides. Our phylogenetic study presents evidence that gene duplication 2800

Journal of Proteome Research • Vol. 6, No. 7, 2007

(29) (30) (31) (32) (33) (34) (35) (36)

Platnick, N. I. Biodiversity Sci. Policy 1999, 33-52. Escoubas, P. Mol. Diversity 2006, 10 (4), 545-54. Lee, S.; Lynch, K. R. Biochem. J. 2005, 391 (Pt 2), 317-23. Cohen, E.; Quistad, G. B. Toxicon 1998, 36 (2), 353-8. Tu, A. T.; Satoh, R. Tanpakushitsu Kakusan Koso 2001, 46 (4), 363-72. Lee, S. Y.; MacKinnon, R. Nature 2004, 430 (6996), 232-5. Xiao, Y.; Tang, J.; Hu, W.; Xie, J.; Maertens, C.; Tytgat, J.; Liang, S. J. Biol. Chem. 2005, 280 (13), 12069-76. Liu, Z.; Dai, J.; Dai, L.; Deng, M.; Hu, Z.; Hu, W.; Liang, S. J. Biol. Chem. 2006, 281 (13), 8628-35. Chen, X.; Kalbacher, H.; Grunder, S. J. Gen. Physiol. 2005, 126 (1), 71-9. Corzo, G.; Escoubas, P. Cell. Mol. Life Sci. 2003, 60, 2409-2426. Liang, S. P. Toxicon 2004, 43, 575-585. Liang, S. P.; Qin, Y. B.; Zhang, D. Y.; Pan, X.; Chen, X. D.; Xie, J. Y. Zool. Res. 1993, 14, 65-70. Zhang, L.; Liu, X.; Zhang, J.; Liang, S. P. Proteomics 2006, 6, 487497. Hellmann, U.; Wemstedt, C.; Gonez, J.; Heldin, C. H. Anal. Biochem. 1995, 224, 451-455. Cao, R.; Li, X.; Liu, Z.; Liang, S. P. J. Proteome Res. 2006, 5, 634642. Perkins, D. N.; Pappin, D. J.; Creasy, D. M.; Cottrell, J. S. Electrophoresis 1999, 20, 3551-3567. Gobom, J.; Schuerenberg, M.; Mueller, M.; Theiss, D.; Lehrach, H.; Nordhoff, E. Anal. Chem. 2001, 73, 434-438. Fry, B. G.; Wuster, W. Mol. Biol. Evol. 2004, 21, 870-883. Serrano, S. M.; Shannon, J. D.; Wang, D.; Fox, J. W. Proteomics 2005, 5 (2), 501-10. Takagaki, K.; Satoh, T.; Tanuma, N.; Masuda, K.; Takekawa, M.; Shima, H.; Kikuchi, K. Biochem. J. 2004, 383, 447-455. Brackmann, M.; Schuchmann, S.; Anand, R.; Braunewell, K. H. J. Cell Sci. 2005, 118, 2495-2505. Li, P. L.; Jin, M. W.; Campbell, W. B. Hypertension 1998, 31, 3038. Valdez-Cruz, N. A.; Davila, S.; Licea, A.; Corona, M.; Zamudio, F. Z.; Garcia-Valdes, J.; Boyer, L.; Possani, L. D. Biochimie 2004, 86, 387-96. Nawarak, J.; Sinchaikul, S.; Wu, C. Y.; Liau, M. Y.; Phutrakul, S.; Chen, S. T. Electrophoresis 2003, 24, 2838-2854. Shevchenko, A.; de Sousa, M. M.; Waridel, P.; Bittencourt, S. T.; de Sousa, M. V.; Shevchenko, A. J. Proteome Res. 2005, 4, 862-9. Kao, R. H. Tzu. Chin. Med. J. 2003, 15, 21-25. Birrell, G. W.; Earl, S.; Masci, P. P.; de Jersey, J.; Wallis, T. P.; Gorman, J. J.; Lavin, M. F. Mol. Cell. Proteomics 2006, 5, 379-89. Nawarak, J.; Phutrakul, S.; Chen, S. T. J. Proteome Res. 2004, 3, 383-392. Choumet, V.; Lafaye, P.; Demangel, C.; Bon, C.; Mazie, J. C. Biol. Chem. 1999, 380 (5), 561-8. Kreil, G. Protein Sci. 1995, 4 (9), 1666-9. Li, S. Y.; Liang, S. P. Chin. Biochem. J. 1995, 11 (2), 155-160. Qu, Y.; Liang, S.; Ding, J.; Liu, X.; Zhang, R.; Gu, X. J. Protein Chem. 1997, 16 (6), 565-74. Peng, K.; Shu, Q.; Liu, Z.; Liang S. J. Biol. Chem. 2002, 277 (49), 47564-71. Shu, Q.; Lu, S. Y.; Gu, X. C.; Liang, S. P. Protein Sci. 2002, 11 (2), 245-52. Peng, K.; Chen, X. D.; Liang, S. P. Toxicon 2001, 39, 491-498. Liu, Z. H.; Dai, J.; Dai, L. J.; Deng, M. C.; Hu, Z.; Hu, W. J.; Liang, S. P. J. Biol. Chem. 2006, 281, 8628-35.

research articles

Characterization of Chinese Bird Spider Venom (37) Li, D. L.; Xiao, Y. C.; Liang, S. P. FEBS Lett. 2003, 555, 616-622. (38) Liang, S. P.; Pan, X. Toxicon 1995, 33, 875-882. (39) Liang, S. P.; Lin, L. Chin. J. Biochem. Mol. Biol. 2000, 16, 9295. (40) Espiritu, D. J.; Watkins, M.; Dia-Monje, V.; Cartier, G. E.; Cruz, L. J.; Olivera, B. M. Toxicon 2001, 39 (12), 1899-916.

(41) Duda, T. F., Jr.; Palumbi, S. R. Proc. Natl. Acad. Sci. U.S.A. 1999, 96 (12), 6820-3. (42) Westbrook, J. A.; Yan, J. X.; Wait, R.; Welson, S. Y.; Dunn, M. J. Electrophoresis 2001, 22, 2865-2871.

PR0700192

Journal of Proteome Research • Vol. 6, No. 7, 2007 2801