Advertisement

Structural Characterization and Disulfide Assignment of Spider Peptide Phα1β by Mass Spectrometry

  • Kelly L. Wormwood
  • Armand Gatien Ngounou Wetie
  • Marcus Vinicius Gomez
  • Yue Ju
  • Paul Kowalski
  • Marius Mihasan
  • Costel C. DarieEmail author
Focus: 29th Sanibel Conference, Peptidomics: Bridging the Gap Between Proteomics and Metabolomics by MS: Research Article

Abstract

Native Phα1β is a peptide purified from the venom of the armed spider Phoneutria nigriventer that has been shown to have an extensive analgesic effect with fewer side effects than ω-conotoxin MVIIA. Recombinant Phα1β mimics the effects of the native Phα1β. Because of this, it has been suggested that Phα1β may have potential to be used as a therapeutic for controlling persistent pathological pain. The amino acid sequence of Phα1β is known; however, the exact structure and disulfide arrangement has yet to be determined. Determination of the disulfide linkages and exact structure could greatly assist in pharmacological analysis and determination of why this peptide is such an effective analgesic. Here, we used biochemical and mass spectrometry approaches to determine the disulfide linkages present in the recombinant Phα1β peptide. Using a combination of MALDI-MS, direct infusion ESI-MS, and nanoLC-MS/MS analysis of the undigested recombinant Phα1β peptide and digested with AspN, trypsin, or AspN/trypsin, we were able to identify and confirm all six disulfide linkages present in the peptide as Cys1-2, Cys3-4, Cys5-6, Cys7-8, Cys9-10, and Cys11-12. These results were also partially confirmed in the native Phα1β peptide. These experiments provide essential structural information about Phα1β and may assist in providing insight into the peptide’s analgesic effect with very low side effects.

Graphical Abstract

Keywords

Disulfide bridges Mass spectrometry Venom peptides Protein structure 

Abbreviations:

MALDI

Matrix-assisted laser desorption ionization

MS

Mass spectrometry

MALDI-MS

MALDI mass spectrometry

ESI

Electrospray ionization

ESI-MS

Electrospray ionization mass spectrometry

nanoLC

Nanoliquid chromatography

nanoLC-MS/MS

Nanoliquid chromatography tandem mass spectrometry

DTNB

5,5-Dithio-bis-(2-nitrobenzoic acid)

E. coli

Escherichia coli

FA

Formic acid

ACN

Acetonitrile

Q-TOF

Quadrupole-time of flight

UPLC

Ultra performance liquid chromatography

HPLC

High-performance liquid chromatography

MS/MS

Tandem mass spectrometry

HCAA

α-Cyano-4-hydroxycinnamic acid

TFA

Trifluoroacetic acid

TOPA

3,4,6-Trihydroxy-phenylalanine

Cys

Cysteine

Introduction

The spiders of the genus Phoneutria are members of the family Ctenidae, suborder Labdognata, and order Araneidae. They inhabit forests of the neotropical region from Southern Central America (Costa Rica) throughout South America, from the East of the Andes to the North of Argentina. The genus comprises the largest known true spiders, considering their size and weight. In Brazil, they are also known as “armed spiders,” because they display an “armed” position when threatened. The species Phoneutria nigriventer are the most important species of the genus Phoneutria, when considering clinical significance. The venom of this aggressive spider is highly toxic and it is the most studied among the venoms of Phoneutria species. An initial fractionation procedure of the venom, using gel filtration and reversed-phase chromatography, monitored by the assessment of lethal activity and toxic effects, yielded four distinct families of neurotoxic polypeptides, named PhTx1, PhTx2, PhTx3, and PhTx4. The purified fraction PhTx3 contains six neurotoxic peptide isoforms, PnTx3-1 to 6 [1]. The fraction PhTx3-6 has antinociceptive effects [2] and was patented with the name of Phα1β. It reversibly and non-specifically inhibits high-voltage-activated Ca2+ channels, namely L-(Cav1.2), N-(Cav2.2), P/Q-(Cav2.1), and R-(Cav2.3) type, with varying potency (N > R > P/Q > L), in heterologous and native systems [3]. Phα1β toxin can be overexpressed and purified from in Escherichia coli in a recombinant form that shares all the analgesic properties of the native toxin [2].

It has been demonstrated that the Phα1β peptide is more effective and potent as an analgesic than ω-conotoxin MVIIA and produces far less side effects [2]. Because of this, it has been suggested that Phα1β may have potential to be used as a therapeutic for controlling persistent pathological pain, such as cancer pain [4]. The amino acid sequence of Phα1β is known; however, the exact structure and disulfide arrangement has yet to be determined. Determination of the disulfide linkages and exact structure could greatly assist in pharmacological analysis and determination of why this peptide is such an effective analgesic.

Historically, disulfide linkages within proteins and peptides have been determined using a combination of chromatography (i.e., reversed-phase chromatography) under non-reducing and reducing conditions, followed by N-terminal sequencing, and the free, reduced cysteines were identified by Ellman’s reagent (DTNB) [5]. Other methods such as a combination of gel filtration, high-performance liquid chromatography and peptide mapping by paper high-voltage electrophoresis in one direction and paper chromatography in the second dimension [6], or chemical and enzymatic digestion followed by 2D gel electrophoresis under non-reducing and reducing conditions were also useful [7]. Recently, mass spectrometry has become the method of choice for determination of disulfide linkages within proteins and peptides [8, 9]. Specifically, combinations of nanoliquid chromatography-tandem mass spectrometry (nanoLC-MS/MS) with direct infusion electrospray ionization mass spectrometry (ESI-MS) or matrix-assisted laser desorption ionization mass spectrometry (MALDI-MS) of the trypsin, AspN, or chymotrypsin and/or a combination of them became the investigators’ first choice in disulfide assignment in proteins and peptides. Here, we used biochemical and mass spectrometry approaches to determine the disulfide linkages present in the recombinant Phα1β peptide. Using a combination of MALDI-MS, direct infusion ESI-MS, and nanoLC-MS/MS analysis of the undigested Phα1β peptide and peptide digested with AspN, trypsin, or AspN/trypsin, we were able to identify and confirm all six disulfide linkages present in the peptide as Cys 1–2, Cys 3–4, Cys 5–6, Cys 7–8, Cys 9–10, and Cys 11–12. These experiments provide essential structural information about Phα1β and may assist in providing insight into the peptide’s analgesic effect with very low side effects.

Materials and Methods

Reagents

All reagents were purchased from Sigma-Aldrich, Waters Corp., or Fisher Scientific, unless otherwise stated.

Peptide Preparation

The recombinant version of Phα1β was synthesized by Giotto Biotech (www.giottobiotech.com) and was expressed in E. coli. It was purified through a proprietary production process, with a combination of ion exchange and size exclusion chromatography. The yield of the process was 0.5 mg/ml. The peptide’s MW is 6045.12 Da (average mass) and 6040.58 (monoisotopic mass). The native Phα1β peptide was purified by a combination of gel filtration, reversed-phase FPLC/FPLC, and ion exchange HPLC, as previously described [1]. Both the native Phα1β and the overexpressed, recombinant Phα1β are 55 amino acid peptides with the amino acid sequence ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFKCSCAHANKYFCNRKKEKCKKA. The Phα1β recombinant and the natural Phα1β peptide also have in the N-terminal a signal sequence (native peptide) and methionine (recombinant peptide), both removed during expression.

Enzymatic Digestion of Phα1β Peptide

Digestion of Phα1β peptide was carried out in solution. Two hundred microliters of 80 mM Tris HCl, pH 7.8, was added to 200 μg of dry peptide. The solution was pipetted up and down, vortexed, and sonicated for 15 min. The peptide did not fully go into solution, so another 200 μl of buffer was added and was again pipetted, vortexed, and sonicated as before. This was repeated once more so the final concentration was 0.3 μg/μl (200 μg/600 μl). Fifty microliters of 6 M urea was added for a final urea concentration of 0.75 M. The solution was pipetted, vortexed, and sonicated as before. Two micrograms of AspN (cleavage at N-terminus of D and E residues) was added to 150 μg of peptide (487.5 μl of solution), and 5 μg of trypsin (cleavage at C-terminus of R and K residues) was added to 200 μg of peptide (650 μl of solution). These were allowed to digest overnight (16–18 h) at 37 °C. Fifty micrograms of peptide (162.5 μl of solution) was left undigested. The undigested peptide was zip tipped using 1000 μg–200 μl capacity C18 TopTips from Glygen. The sample was completely dried using a SpeedVac evaporator and then rehydrated in 0.1% formic acid (FA) and sonicated for 15 min. The sample was then resolublized in 100 μl 0.1% FA and sonicated for 15 min (Glygen TopTip Procedure). The TopTip was gently tapped to displace any packing material that may have been stuck to the cap. The caps were removed from the top and bottom of the tip and 50 μl of 50% ACN + 0.1% FA was added to the top of the TopTip via a pipette tip inserted in the top of the TopTip. This solution was forced through the packed bed by attaching the TopTip to the provided syringe and applying air pressure. The TopTip was held while the plunger was pressed during this process to ensure the TopTip did not release from the syringe due to pressure. The TopTip was removed from the syringe and this was repeated three times. The sample solution was loaded and forced through as above, loading the maximum volume allowed (200 μl) or less at a time. The eluate was then reapplied to ensure thorough binding. The packed bed was then washed two times with 50 μl of 0.1% FA. The bound peptides were then eluted from the packed bed with 50 μl of 50% ACN + 0.1% FA. This eluate was then reapplied to elute the maximum amount of peptide. Another 50 μl of 50% ACN + 0.1% FA was added to the tip and forced through the bed and the eluate was again reapplied for maximum peptide recovery. These two 50 μl aliquots were then combined. This sample was then completely dried in the SpeedVac.

AspN and trypsin digestions were stopped by adding 100% acetic acid until the pH was between 2 and 4. The samples were then completely dried using a SpeedVac evaporator, and the Glygen TopTip Procedure described above was performed. The samples were then resolublized in 100, 150, and 200 μl of 2% ACN + 0.1% FA for undigested, AspN-digested, and trypsin-digested peptides respectively. Samples were then analyzed using nanoliquid chromatography tandem mass spectrometry.

Nanoliquid Chromatography Tandem Mass Spectrometry Analysis

The peptide and/or peptide mixture was analyzed by reversed-phase nanoliquid chromatography tandem mass spectrometry (nanoLC-MS/MS) using a nanoAcuity UPLC (Waters, Milford, MA) coupled to either a Q-TOF Xevo G2 (Waters, Milford, MA) or Q-TOF Ultima API-US mass spectrometer (Waters, Milford, MA) according to published procedures [10, 11, 12, 13] as well as direct infusion mass spectrometry using a Q-TOF Xevo-G2 mass spectrometer (Waters, Milford, MA).
  1. (a)

    For the analysis using the Q-TOF Ultima API, the peptides were loaded onto a 150 μm × 100 mm NanoAquity BEH130 C18 1.7-μm UPLC column (Waters, Milford, MA) and eluted over a 150-min gradient at a flow rate of 300 nl/min as follows: 1–45% organic solvent B (ACN containing 0.1% FA) over 1–20 min, 45–85% B (20–80 min), 85% B constant (80–120 min), 85–1% B (120–135 min), and finally remain at initial conditions of 1% B (135–150 min). The aqueous solvent A was 0.1% FA in HPLC water. The column was coupled to a fused silica tube emitter tip for the nano-electrospray option (Waters, Milford, MA). The MS data acquisition consisted of survey 1 s MS scans in the m/z range 350–1800 and automatic data-dependent analysis (DDA) of the top 5 highest intensity ions with charge of 2+, 3+, 4+, or 5+. The MS/MS, which was obtained over m/z 50–1800, was triggered when MS signal intensity exceeded 10 counts/s. In the survey MS scans, the five most intense peaks were selected for collision-induced dissociation (CID) and were fragmented until the total MS/MS ion counts reached above 10,000 counts/s or for up to 6.3 s each.

     
  2. (b)

    The peptides were then eluted over a 60-min gradient at a flow rate of 300 nl/min as follows: 1–45% organic solvent B (ACN containing 0.1% FA) over 1–37 min, 45–85% B (37–40 min), 85% B constant (40–47 min), 85–1% B (47–50 min), and finally remain at initial conditions of 1% B (50–60 min). The aqueous solvent A was 0.1% FA in HPLC water. The column was coupled to a fused silica tube emitter tip for the nano-electrospray option (Waters, Milford, MA). The MS data acquisition consisted of survey 1 s MS scans in the m/z range 350–1600 and automatic DDA of the top 5 highest intensity ions with charge of 2+, 3+, or 4+. The MS/MS, which was obtained over m/z 50–1900, was triggered when MS signal intensity exceeded 10 counts/s. In the survey MS scans, the five most intense peaks were selected for CID and were fragmented until the total MS/MS ion counts reached above 999,999 counts/s or for up to 6.3 s each.

     
  3. (c)

    The peptides were then eluted over the same 60-min gradient and system setup as previously described in (b). The MS data acquisition consisted of survey 1 s MS scans in the m/z range 300–2200 and automatic DDA of the top 5 highest intensity ions with charge of 2+, 3+, 4+, 5+, or 6+. The MS/MS, which was obtained over m/z 50–1900, was triggered when MS signal intensity exceeded 10 counts/s. In the survey MS scans, the five most intense peaks were selected for CID and were fragmented until the total MS/MS ion counts reached above 999,999 counts/s or for up to 6.3 s each. MS/MS peak selection using an inclusion list was enabled to preferentially select all peaks that were on the inclusion list. The inclusion list consisted of all potential m/z up to the 9+ charge state for the peptide after an AspN digestion. Peaks that were within ± 200 mDa of the entries on the inclusion list were included. The retention time window was ± 10 s.

     
  4. (d)

    The peptides were then eluted over the same 60-min gradient and system setup as previously described in (b). The MS data acquisition was the same as that described in (c). Rather than enabling MS/MS peak selection using an inclusion list to preferentially select peaks that were on the inclusion list, MS/MS peak selection using an inclusion list was enabled to ONLY select peaks that were on the inclusion list. The inclusion list consisted of all potential m/z up to the 9+ charge state for the peptide after a trypsin digestion. All other parameters remained the same.

     
  5. (e)

    The peptides were then eluted over a 120-min gradient at a flow rate of 300 nl/min as follows: 1–55% organic solvent B (ACN containing 0.1% FA) over 1–70 min, 55–85% B (70–85 min), 85% B constant (85–95 min), 85–1% B (95–105 min), and finally remain at initial conditions of 1% B (105–120 min). The same instrument setup as previously described in (a) was used. The MS data acquisition consisted of survey 1 s MS scans in the m/z range 300–2200 and automatic DDA of the top 5 highest intensity ions with charge of 2+, 3+, or 4+. The MS/MS, which was obtained over m/z 50–1900, was triggered when MS signal intensity exceeded 10 counts/s. In the survey MS scans, the five most intense peaks were selected for CID and were fragmented until the total MS/MS ion counts reached above 999,999 counts/s or for up to 6.3 s each. MS/MS peak selection using an inclusion list was enabled to preferentially select all peaks that were on the inclusion list. The inclusion list consisted of all potential m/z up to the 9+ charge state for the peptide after an AspN digestion. Peaks that were within ± 300 mDa of the entries on the inclusion list were included. The retention time window was ± 10 s.

     
  6. (f)

    For the analysis using the QTOF Xevo G2, the peptides were loaded onto a 150 μm × 100 mm NanoAquity BEH130 C18 1.7-μm UPLC column (Waters, Milford, MA) and eluted over a 60-min gradient at a flow rate of 600 nl/min as follows: 1–50% organic solvent B (ACN containing 0.1% FA) over 1–31 min, 50–85% B (31–33 min), 85% B constant (33–36 min), 85–1% B (36–37 min), and finally remain at initial conditions of 1% B (37–60 min). The aqueous solvent A was 0.1% FA in HPLC water. The column was coupled to a Picotip Emitter Silicatip nano-electrospray needle (New Objective, Woburn, MA). The MS data acquisition consisted of survey 1 s MS scans in the m/z range 350–2000 and automatic DDA of the top 6 highest intensity ions with charge of 2+, 3+, 4+, 5+, and 6+. The MS/MS, which was obtained over m/z 50–2000, was triggered when MS signal intensity exceeded 500 intensity/s. In the survey MS scans, the top 6 most intense peaks were selected for CID and were fragmented until the total MS/MS ion counts reached above 3000 intensity/s or for up to 7.0 s each.

     
  7. (g)

    The intact peptide was then analyzed by direct infusion ESI-Q-TOF MS using the Xevo G2 previously mentioned. A flow rate of 400 nl/min was used. MS was acquired over m/z 100–2000. A scan time of 1 s was used. MS/MS was collected for the individual peaks with m/z 862, 1005, 1069, 1206, 1508, and 1069. All instruments were calibrated using 1 pmol Glu-fib, (Waters, Milford, MA), prior to sample being run.

     

MALDI Analysis

Saturated HCCA matrix solution was prepared in 50% acetonitrile/50% nanopure water with 0.1% TFA. One microliter of the peptide samples was mixed with 1 μl of the matrix before being spotted onto the Bruker MTP 384 ground steel plate and then the mixture was air-dried at room temperature. MALDI-MS and MS/MS measurements were accomplished using the Bruker UltrafleXtreme MALDI TOF/TOF mass spectrometer (Bruker Daltonics). In the MALDI-MS analysis, ions were ionized by a Smartbeam II laser (modified Nd:YAG laser). After laser ionization, ions were accelerated by a 19.5-kV electric field into the field-free flight tube and were detected in the positive reflectron mode. In the MALDI-MS/MS analysis, laser-induced dissociation (LID) was used to fragment the peptides (LIFT mode with no added gas). MALDI-MS/MS data of the reduced peptide with m/z 3212 was analyzed through Biotools software (Bruker Daltonics) with Mascot server to yield the corresponding sequence.

Structural Biology Analysis

All sequences were from Uniprot database (http://www.uniprot.org/). Sequence alignments were performed using HHalign algorithm implemented in Clustal-Omega [14]. The default transition matrix (Gonnet) and scoring functions (gap opening penalty—6 bits, gap extension—1 bit) were used. The alignments were visualized using Jalview [15].

Disulfide Bond Arrangement Determination

Disulfide bonds were assigned manually using MassLynx Software Version 4.0 along with Protein Prospector (http://prospector.ucsf.edu/prospector/cgi-bin/msform.cgi?form=msproduct).

Data Sharing

The raw data and any additional information regarding the experiments presented will be available upon request, according to Clarkson University’s Material Transfer Agreement policy.

Results and Discussion

All Cysteine Residues Within Phα1β Peptide Are Disulfide-Linked

The theoretical neutral molecular mass of the Phα1β peptide with all 12 cysteine residues reduced is 6040.58 Da, while the theoretical mass of the same peptide with all cysteines oxidized and involved in disulfide bridges is 6028.58 Da. To determine whether the cysteine residues are reduced or linked as disulfide bridges, we analyzed the intact, non-reduced peptide both by electrospray ionization mass spectrometry using direct infusion and by nanoLC-MS/MS. We found peaks with m/z of 862.23 (7+), 1005.76 (6+), 1206.72 (5+), and 1508.12 (4+) that corresponded to the mass of the peptide, minus 12 Da due to the six disulfide bridges (Figure 1). Therefore, Phα1β peptide contains no free cysteine and all cysteine residues are disulfide-linked.
Figure 1

Disulfide linkage determination of Phα1β peptide using ESI-MS. Confirmation that all cysteines present are disulfide-linked. (a) MS spectra showing the peaks with charge states of 4+, 5+, 6+, and 7+ corresponding to the mass of the peptide (6040.58 Da) minus 12 Da (6028.58 Da) indicating all 12 cysteines present are disulfide-linked. (b) MS spectra of individual charge states identified. In the top, the cysteine residues within the Phα1β peptide are numbered

Of the 12 Cysteine Residues Within Phα1β Peptide, the First Eight Cys Are Disulfide-Linked and the Last Four Cys Are Also Disulfide-Linked

The first piece of evidence that the first eight Cys residues are disulfide-linked came from nanoLC-MS/MS analysis of the Phα1β peptide digested with trypsin, where we identified a peak with m/z of 1198.12 (3+), which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK, with all Cys residues intramolecular disulfide-linked. The MS/MS fragmentation of this peak confirmed the amino acid sequence of the peptide listed earlier and that the Cys residues from this peptide are indeed intramolecular disulfide-linked (Figure 2a). Our data was also confirmed by detection of a peak with m/z of 1204.25 (3+), which corresponded to tryptic peptides ACIPR and GEICTDDCECCGCDNQCYCPPGSSLGIFK bound in an intermolecular disulfide bridge between the first and the next seven Cys residues. The MS/MS fragmentation of this peak produced a series of b and y and internal ions from both peptides which corresponded to the amino acid sequence of the two peptides disulfide-linked (Figure 2b).
Figure 2

Disulfide linkage determination of Phα1β peptide using nanoLC-MS/MS. Confirmation of the first eight cysteines and, hence, last four cysteines, being disulfide-linked. (a) MS/MS fragmentation of precursor ion (inset) with m/z 1198.12 (3+). This corresponds to the mass of the peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK minus 8 Da (again, the first eight cysteines are disulfide-linked). (b) MS/MS fragmentation of the precursor ion (inset) with m/z 1204.27 (3+). This corresponds to peptides ACIPR and GEICTDDCECCGCDNQCYCPPGSSLGIFK disulfide-linked (minus 8 Da), through an intermolecular bridge (note the identification of the y1’, y2’, and y3’ that are part of peptide ACIPR and of the b and y ions that are part of peptide GEICTDDCECCGCDNQCYCPPGSSLGIFK. The bridge is between Cys1 and 2 (see Figure 4). (c) MS/MS fragmentation of the precursor ion (inset) with m/z 1068.68 (3+). This corresponds to the mass of the peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG minus 8 Da, suggesting that the first eight cysteines are disulfide-linked. Note that spectra from (a) and (b) were from nanoLC-MS/MS analysis of a tryptic digestion of the peptide, while the spectrum from (c) was from nanoLC-MS/MS analysis of an undigested peptide. Spectrum from (c) was also identified in other experiments

We also tried to determine the disulfide linkages by direct infusion ESI-MSMS of the intact Phα1β peptide, but the disulfide-linked Phα1β peptide did not fragment well (Supplemental Figure 1). However, the peptide fragments in smaller pieces and keeps the cysteines disulfide-linked. Therefore, we reasoned that if we find the same disulfide bridges in these truncated peptides, this information could help us in determining the disulfide bridges of the intact peptide. In a nanoLC-MS/MS analysis of the undigested peptide, we found a precursor ion with m/z of 1068.68 (3+), which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG with all Cys residues disulfide-linked (Figure 2c). MS/MS fragmentation of this peak produced a series of b and y ions and of internal ions that confirmed the amino acid sequence of the identified peptide (Figure 2c). The MS/MS of this 1068.68 (3+) precursor was identified in multiple nanoLC-MS/MS experiments (data not shown). This data further confirmed the nanoLC-MS/MS experiments of the tryptic digests that the first eight cysteine residues are disulfide-linked. Since we clearly demonstrated that (1) the intact peptide has all 12 Cys residues intramolecularly disulfide-linked and (2) the first eight Cys residues are also intramolecularly disulfide-linked, this suggests that the last four Cys are also intramolecularly disulfide-linked.

MALDI-MS Analysis of the Phα1β Peptide Confirms the Disulfide Linkages Identified by Direct Infusion ESI-MS and nanoLC-MS/MS Experiments: Cys 3–Cys 4 Are Disulfide Linked; Cys 5–Cys 6 Are Also Disulfide Linked

To confirm the disulfide linkages between the cysteines detected by direct infusion ESI-MS and nanoLC-MS/MS experiments, we analyzed by MALDI-MS the Phα1β peptide either undigested or digested with trypsin or AspN. MALDI-MS analysis of the intact and truncated Phα1β peptide identified a series of peaks that corresponded to intact peptide (m/z 6027.82), in agreement with the theoretical mass of the peptide (6040.58 Da) with all 12 cysteines disulfide-linked (6028.58 Da) and also in agreement with the m/z of this peptide determined by direct infusion ESI-MS for the peaks with charge states of 4+, 5+, 6+, and 7+, reported earlier in Figure 1. In the MALDI-MS spectrum, we also observed a peak with m/z 3204.19 (1+), which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG, with all of its cysteines disulfide-linked. Note that we have already identified this peptide in nanoLC-MS/MS experiments of the undigested peptide (Figure 2c). To further confirm that this peptide contains all eight cysteines disulfide-linked, we analyzed this peak under non-reducing and reducing conditions. As observed, this peak shifted from m/z 3204 to m/z 3212, further demonstrating that the first eight cysteines are disulfide-linked (Figure 3a, insert). MALDI-MSMS analysis of the peak with m/z 3212 (reduced peptide) provided additional evidence that peak with m/z 3204.19 (1+) corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG (Figure 3b), thus confirming our initial findings shown in Figure 2c that the first eight cysteines are disulfide-linked.
Figure 3

Confirmation of amino acid sequence and disulfide linkage determination by MALDI-MS. (a) Analysis of the Phα1β peptide under non-reducing conditions led to identification of a degradation product with m/z of 3204.19 (1+), which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG with the first eight cysteines disulfide-linked (minus 8 Da). Analysis of the same peak under reducing conditions led to the increase of the m/z of 3204 (1+) to m/z 3212 (1+) (inlet). (b) MS/MS fragmentation of peak with m/z 3212 (1+) that corresponded to reduced peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG produced a series of b and y ions that confirmed the amino acid sequence of this peptide. Note that the peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG was also identified by nanoLC-MS/MS (shown in Figure 2c)

Since analyses of peptide fragments were useful in assignment of the disulfide linkages, we also looked in the MALDI-MS spectrum for additional peaks that could aid us in determination of the disulfide linkages within Phα1β peptide. We found a peak with m/z 3592.35 (1+) which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK and a peak with m/z of 3016.05 (1+) which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSS-H2O, both peptides with all cysteines intramolecularly disulfide-linked (Figure 3a). All these peptides had the first eight cysteines disulfide-linked (Figure 3a). Therefore, the first eight Cys residues are all disulfide-linked and the last four Cys residues are also all disulfide-linked.

Cys 1 and Cys 2 Are Disulfide-Linked

Our conclusion that the first eight Cys residues are disulfide-linked with each other (i.e., none of the first eight Cys are connected to the last four Cys) allowed us to restrict the investigation of the Cys linkages to the first eight of them. Therefore, we pursued the investigation of the disulfide linkages within the first eight Cys residues. The first disulfide bridge that we identified was an intramolecular disulfide bridge between Cys 1 and Cys 2. In an AspN digestion, we found a peak with m/z of 588.24 (2+) which corresponded to peptide ACIPRGEICTD with Cys 1 and Cys 2 disulfide-linked (Figure 4a). MS/MS fragmentation of the peak with m/z of 588.24 (2+) (Figure 4a) produced a series of b and y and internal peaks that confirmed the amino acid sequence of this peptide, with the first two Cys residues disulfide-linked. Some of these product ions within the MS/MS spectrum of peak with m/z of 588.24 (2+) such as peaks with m/z of 1042.45 (1+), 941.40 (1+), and 922.39 (1+) corresponded to b10, b9, and b9 + H2O, all with Cys residues disulfide-linked (i.e., 2 Da lower than the theoretical m/z of these ions). This data was nicely complemented by identification of y2 ions, but not of y3, y4, or y5, which should not be observed at their expected m/z, further demonstrating that the Cys1 and Cys2 residues are disulfide-linked. Additional evidence that Cys 1 and Cys 2 are disulfide-linked came from identification of an intermolecular bridge from a nanoLC-MS/MS of an AspN digest. Specifically, we found a peak with m/z of 539.76 (2+) which corresponded to peptides ACIPRG and EICT intermolecularly disulfide-linked. The MS/MS fragmentation of this peak produced some b and y ions from both peptides that confirmed that these two peptides are disulfide-linked (Figure 4b). Taken together, our data suggests that Cys 1 and Cys 2 are disulfide-linked. Since the first eight Cys are disulfide-linked and Cys 1–2 are also disulfide-linked, this suggests that Cys 3 to Cys 8 are also disulfide-linked.
Figure 4

Identification of Cys1–2 disulfide linkage. (a) MSMS fragmentation of the precursor ion (inset) with m/z 588.24 (2+) corresponds to the mass of the peptide ACIPRGEICTD with Cys intramolecularly disulfide-linked (minus 2 Da). (b) MS/MS fragmentation of the precursor ion (inset) with m/z 539.73 (2+) corresponds to the peptides ACIPRG and EICT with the cysteines disulfide-linked through an intermolecular bridge. Note that pairs of a2, -b2, and dehydrated a2-b2 peaks from the EICT peptide, y2’, y3, and y4’ peaks from ACIPRG peptide, and peptides ACIPRG + EIC intermolecularly disulfide-linked provide strong evidence that Cys1–2 are disulfide-linked. The spectra from (a) and (b) resulted from an AspN digestion

Cys 7 and Cys 8 Are Disulfide-Linked; Cys 3 to Cys 6 Are Also Disulfide-Linked

Since we determined that the Cys 1 and Cys 2 are disulfide-linked and that the first eight Cys are also disulfide-linked, we sought to determine the disulfide linkage in Cys 3 to Cys 8. In nanoLC-MS/MS analysis of the Phα1β peptide digested with AspN, we found a peak with m/z of 863.36 (2+) which corresponded to unmodified peptide DNQCYCPPGSSLGIFK (calculated m/z of 864.88 (2+)) that had the Cys 7 and Cys 8 disulfide-linked (Figure 5a). We also found a peak with m/z of 879.34 (2+) that corresponded to peptide DNQCYCPPGSSLGIFK modified by 32 Da (also with the Cys 7 and Cys 8 disulfide-linked; Figure 5b). This 879.34 (2+) peak had a mass increase of the precursor ion with 32 Da, as compared with the theoretical mass of the peptide, with the cysteines disulfide-linked. Such modification is most likely a dihydroxylation or oxidation of one of the two proline residues to glutamate, a trisulfide bond (additional sulfur in disulfide bond), or a modification of tyrosine residue to 3,4,6-trihydroxy-phenylalanine (TOPA). MS/MS fragmentation of the precursor ions with m/z of 863.36 (2+) and 879.38 (2+) produced a series of b and y ions (b2, b3, b11 + H2O, b14 − 2H2O, y2, y4, y9, and y10), as well as a few internal ions that led to identification of this peptide, with the Cys residues disulfide-linked. In addition, identification of the b11 + H2O and b14 − 2H2O with cysteine residues disulfide-linked strengthened our conclusion that these two cysteines are disulfide-linked. Since we could easily observe in the MS/MS spectrum y9 (PGSSLGIFK) and y10 (PPGSSLGIFK) ions which contained unmodified proline residues, this suggests that the modification in this peptide was most likely due to formation of TOPA on the tyrosine residue. These two peaks were also identified in a different experiment (Supplemental Figure 2A & B). Another peak with m/z 895.82 (2+) was identified in a AspN digest, which corresponded to peptide DNQCYCPPGSSLGIFK that had the Cys 7 and Cys 8 disulfide-linked, but modified by 64 Da (calculated m/z of 864.88 (2+) (Supplemental Figure 2C). The MS/MS fragmentation of these three precursor ions produced an almost identical fragmentation pattern and similar to the ones shown in Figure 5a, b, further confirming that peptide DNQCYCPPGSSLGIFK has the Cys 7 and Cys 8 disulfide-linked. Because (1) the first eight Cys residues are disulfide-linked, (2) Cys 1 and Cys 2 are disulfide-linked, and (3) Cys 7 and Cys 8 are also disulfide-linked, this suggests that Cys 3 to Cys 6 are also disulfide-linked.
Figure 5

Identification of Cys7–8, Cys9–10, and Cys11–12 disulfide linkages. (a) MS/MS fragmentation of the precursor ion (inset) with m/z 863.36 (2+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFK with cysteines disulfide-linked (minus 2 Da). (b) MS/MS fragmentation of the precursor ion (inset) with m/z 879.38 (2+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFK with cysteines disulfide-linked (minus 2 Da) and modified by 32 Da. (c) MS/MS fragmentation of the precursor ion (inset) with m/z 676.27 (4+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFKCSCAHANKY with cysteines disulfide-linked (minus 4 Da). The spectra from (a), (b), and (c) resulted from an AspN digestion

Cys 9 and Cys 10 Are Disulfide-Linked; Cys 11 and Cys 12 Are Also Disulfide-Linked

We have already determined that the last four Cys (Cys 9 to Cys 12) are disulfide-linked. However, we did not have any information about how are these Cys residues linked. The answer to our question came from nanoLC-MS/MS analysis of the Phα1β peptide digested with AspN. Specifically, we found a peak with m/z of 676.26 (4+) that corresponded to peptide DNQCYCPPGSSLGIFKCSCAHANKY (calculated m/z 677.28 (4+)), with all four Cys residues (Cys 7 to Cys 10) disulfide-linked (calculated m/z 676.28 (4+)) (Figure 5). MS/MS fragmentation of the 676.26 (4+) peak produced a series of peaks including b2, y2, y3, y4, and the internal ions PPGSSLG and PPGSSLGI that confirmed the amino acid sequence of peptide DNQCYCPPGSSLGIFKCSCAHANKY, with all four Cys residues disulfide-linked (Figure 5c). This MS/MS was found multiple times in multiple experiments. Additional evidence that Cys 7 to Cys 10 are disulfide-linked came from identification of a peak with m/z 1044.84 (2+), which corresponded to peptide DNQCYCPPGSSLGIFKCSCA, with Cys 7 to Cys 10 disulfide-linked (Supplemental Figure 3). While the MS/MS spectrum did not have a good series of ions, we could still find b2 and b3 ions and the internal fragment PPGSSLGI, which could be a good indication that this identification is a true positive.

Since we have already determined that Cys 7 and Cys 8 are disulfide-linked, now we provide additional evidence that these two Cys residues are indeed disulfide-linked. In addition, we also can infer that Cys 9 and Cys 10 are disulfide-linked. Furthermore, identification of an internal fragment GIFKCSCAHA that contained the Cys 9–Cys 10 disulfide-linked further confirmed that these Cys 7 to Cys 10 are disulfide-linked, and that (1) Cys 7–Cys 8 are disulfide-linked and (2) Cys 9–Cys 10 are also disulfide-linked. Lastly, since we have demonstrated that (1) Cys 9–Cys 10 are disulfide-linked (Figure 4b), (2) Cys 1 to Cys 8 are disulfide-linked, and (3) Cys 9 to Cys 12 are also disulfide-linked, this suggests that Cys 11 and Cys 12 are also disulfide-linked.

MALDI-MS Analysis of the Phα1β Peptide Under Non-reducing Conditions Confirms the Disulfide Linkages Identified by Direct Infusion ESI-MS and nanoLC-MS/MS Experiments

In our initial MALDI-MS experiments (Figure 3a), we identified a series of peaks that corresponded to (N-terminal) truncated peptide with cysteines disulfide-linked. Examples of such peaks include peak with m/z 3592.35 (1+), which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK, m/z 3204.19 (1+) which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG, and m/z 3016.05(1+) which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSS-H2O (all singly charged and with cysteines disulfide-linked and shown in Figure 3a). To further confirm the disulfide linkage within some of these peaks observed, we subjected peaks with m/z 3592.35 (1+) and 3204.19 (1+) to MALDI-MS/MS analysis (Figure 6a, b). MS/MS of peak with m/z 3592.35 (1+) which corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK with first eight Cys disulfide-linked produced a series of peaks that corresponded to peptides that contained all eight Cys (m/z 3463.9, 2589.7), or Cys 1–2 (m/z 1271.7 and 1155.7) disulfide-linked (Figure 6a). Furthermore, MS/MS of the precursor ion with m/z 3204.19 produced a series of ions that corresponded to peptides containing the first eight Cys (m/z 3106.04, 3016.19, and 2592.95), first four Cys (m/z 1605.21), and Cys 1–2 (m/z 940.95) disulfide-linked (Figure 6b). Since we found (1) a peak (m/z 1605.21) that corresponded to peptide ACIPRGEICTDDCEC with first four Cys (Cys 1 to 4) disulfide-linked and (2) a peak (m/z 940.95) that corresponds to peptide ACIPRGEIC with Cys1–2 disulfide-linked and (3) we know that Cys 1–2 are disulfide-linked (Figure 4) and Cys 3 to Cys 6 are also disulfide-linked (see earlier text), this suggests that Cys 3–4 are disulfide-linked and Cys 5–6 are also disulfide-linked.
Figure 6

Confirmation of previously determined disulfide linkages, as well as determination of Cys3–4 and Cys 5–6 disulfide linkages by MALDI-MS. (a) LIFT of peak with m/z 3591 (1+) showing fragment ions corresponding to the peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK. The precursor ion and two of the b ions (b33-H2O and b24) indicate that the first eight Cys are disulfide-linked. Two other b ions (b11 and b12) indicate that Cys 1–2 are disulfide-linked. (b) LIFT of peak with m/z 3204 (1+) showing fragment ions corresponding to the peptide ACIPRGEICTDDCECCGCDNQCYCPPGSSLG. The precursor ion and the ions y31, b29, and b24 indicate that the first eight Cys are disulfide-linked. Ions b15 and b9 indicate that the first four Cys are disulfide-linked and Cys 1–2 are disulfide-linked, suggesting that Cys 3–4 are disulfide-linked

Taken together, our MALDI-MS analysis further confirmed the direct infusion ESI-MS and nanoLC-MS/MS experiments and also provided additional information that Cys 3–4 are disulfide-linked and Cys 5–6 are also disulfide-linked. In addition, all MALDI-MS, ESI-MS, and nanoLC-MS/MS data are in agreement that Cys 1–2 are disulfide-linked, Cys 3–4 are disulfide-linked, Cys 5–6 are disulfide-linked, Cys 7–8 are disulfide-linked, Cys 9–10 are disulfide-linked, and Cys 11–12 are also disulfide-linked.

MALDI-MS and MALDI-MS/MS Analysis of the Recombinant and Native Phα1β Peptide Shows Similar Fragmentation

The m/z as determined by MALDI-MS for the intact recombinant (m/z 6029.68 (1+)) and native (m/z 6029.90 (1+)) Phα1β peptides (Figure 7a, b) indicates a good agreement that both peptides have all their 12 cysteines disulfide-linked. In addition, peaks with m/z of 3591.21 (1+), 3204.01 (1+), and 3015.49 (1+) (Figure 7a) correspond to truncated recombinant peptides ACIPRGEICTDDCECCGCDNQCYCPPGSSLGIFK, ACIPRGEICTDDCECCGCDNQCYCPPGSSLG, and ACIPRGEICTDDCECCGCDNQCYCPPGSS (all N-terminal peptides), with all first eight cysteines disulfide-linked. Alternatively, peak 3015.90 could be the doubly charged intact peptide. In addition, inspection of the MALDI-MS spectrum of the native peptide (Figure 7b) also indicates that not only the recombinant, but also the native Phα1β peptide is fragmented in a similar way: it loses the C-terminus (it is not clear whether it has a functional significance). At least one peak from the MALDI-MS spectrum of the native peptide (m/z 3015.94 (1+)) corresponds to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSS with all eight Cys disulfide-linked (Figure 7b). This behavior of both recombinant and native Phα1β peptide may suggest a possible functional role of maintaining the N-terminal intact, or losing the C-terminal, or both. However, more important for the current study, it suggests that both the recombinant and native Phα1β peptides have a similar disulfide pattern. Clearly, at least the first eight cysteines from the native peptide are disulfide-linked and the last four cysteines are also disulfide-linked.
Figure 7

MALDI-MS analysis of the recombinant (a) and native (b) Phα1β peptides. Both peptides have a m/z of 6029.68 (1+) (recombinant) and 6029.90 (1+) (native), consistent with all cysteines disulfide-linked. Some peaks in (a) correspond to some N-terminal fragments of the recombinant Phα1β peptide, with all cysteines disulfide-linked. The amino acid sequence of these fragments is indicated. Note that one such peak from (a) with m/z 3015.49 (1+) was also identified as a major peak in (b) with m/z 3015.94 (1+); in both cases, the peak corresponded to peptide ACIPRGEICTDDCECCGCDNQCYCPPGSS, with all eight cysteines disulfide-linked

We then sought to obtain additional information on the recombinant and native Phα1β peptides by analyzing them by MALDI-MS/MS (Figure 8a, b). The MS/MS of both peptides did not look great due to poor fragmentation. However, we observed that fragmentation of both peptides produced at least two identical fragments that we could identify. One of them had m/z 1233.26 (1+) (recombinant peptide) and m/z 1233.56 (1+) (native peptide) and corresponded possibly to peptide YCPPGSSLGIFK-NH3 or FKCSCAHANKY-H2O (with cysteines disulfide-linked). The second fragment had m/z of 2133.82 (1+) (recombinant peptide) and m/z 2134.02 (1+) (native peptide) and corresponded possibly to peptide CIPRGEICTDDCECCGCDNQ-2H2O and with cysteines disulfide-linked. While only two matched fragments are not ideal, it still helps in suggesting that at least the recombinant and native peptides have a similar behavior, both when they are truncated (Figure 7a, b). Additional MALDI-MS and MALDI-MS/MS spectra are provided in Supplemental Figures 47, where the disulfides identified for the recombinant and native peptides further confirmed the results of our experiments presented in Figures 3 and 7.
Figure 8

Alignment of the Phα1β synthetic polypeptide with (a) highly similar naturally occurring ctenitoxins from Phoneutria nigriventer (Pn), Phoneutria reidyi (Pr), and Phoneutria keyserlingi (Pk), (b) closest sequence homologs with known structures—spiderines from Oxyopes species—and (c) representatives of spider toxins targeting calcium (Ca2+) channels with known structures. Solid blue lines indicate experimentally determined disulfide bonds, and dotted blue lines indicate homology-predicted disulfide bonds. Those marked with an asterisk are the proteins with known structures deposited in Z—spiderine—PDB|2n86, hexatoxin—PDB|1g9p, theraphotoxin—PDB|1koz, and agatoxins PDB|1iva and PDB|1agg

Structural Analysis

The predicted structure of the Phα1β peptide indicates that all cysteines are disulfide-linked except two, Cys 3 and Cys 6 (Figure 8). The specific disulfide linkages expected based on the predicted structure are Cys 1–5, Cys 2–7, Cys 4–10, Cys 8–9, and Cys 11–12 (Figure 8b). This differs greatly from our results, as we found all cysteines to be disulfide-linked in the manner of Cys 1–2, Cys 3–4, Cys 5–6, Cys 7–8, Cys 9–10, and Cys 11–12 (Figure 8a).

The closest sequence homolog with an experimentally determined structure of the Phα1β synthetic polypeptide is spiderine-1a, a classic knottin fold containing polypeptide from Oxyopes takobius. The two sequences share 46% sequence identity. This low level of identity makes homology modeling unreliable for the prediction of high-quality 3D structures required for S–S bridge assignment [16, 17]. Moreover, in the NMR-based 3D structure of spiderine-1a, two of the five disulfide bonds were determined from preliminary structure calculations, while the other three disulfide bonds were assigned based on homology [18] using ω-oxotoxin-Ol1b from Oxyopes lineatus [19] as template. Spiderine-1a and ω-oxotoxin-Ol1b share only 14.81% sequence identity, while the level of sequence identity between Phα1β synthetic polypeptide and ω-oxotoxin-Ol1b is 31%. This could be one potential reason why the disulfide pattern we determined experimentally is so different from the predicted models.

The MS data collected for the Phα1β synthetic polypeptide not only shows a completely different disulfide bridge pattern compared to the predicted model (Figure 8a), but also shows a totally different S–S bridge pattern compared with the knottin fold containing proteins such as spiderine or other spider toxins targeting calcium (Ca2+) channels (Figure 8b). Whether this different pattern is compatible with the knottin motif is a matter of debate. Variations in disulfide-bonding patterns do not always lead to significant changes in the overall fold [20]. One clear example in this direction is the SMB domain of human vitronectin. The four disulfide bonds in human vitronectin exist in three possible arrangements, including two overlap patterns similar to the knottin fold and one linear pattern, as shown here, for the Phα1β polypeptide. All these arrangements are compatible with the same overall fold and with the PAI-1 binding function [21, 22].

This could suggest the possibility that the Phα1β peptide may exist in different disulfide arrangements while still maintaining the same activity and function. A study performed by our collaborator compared the antinociceptive effects of native Phα1β and recombinant Phα1β on rodent pain models with formalin, a chronic constriction injury, capsaicin, and cancer melanoma. In these pain models, the antihypersensitivity effects that were produced by the native form of Phα1β were fully mimicked by the recombinant version of Phα1β without causing any side effects [23]. The recombinant peptide used in our experiments came from the same batch as was used in this study.

While the disulfide bridges determined experimentally were initially focused on the recombinant peptide, one could argue that the native peptide has a different conformation and also a different disulfide pattern. However, in our experiments, we also analyzed the native peptide and found clear evidence that the native peptide, just like the recombinant one, had all cysteines disulfide-linked and that the first eight cysteines were disulfide-bridged; the last four cysteines were also disulfide-bridged. Therefore, again, our data do not support the predicted disulfide linkage for the Phα1β peptide.

Striking in this study was the consistent identification of N-terminal fragments of both the recombinant and native Phα1β peptides. This finding somehow suggests that the N-terminal Phα1β peptide may have a consistent, functional role (i.e., antinociceptive), while the C-terminal end may not be required for this antinociceptive activity, but rather possibly an interaction domain for other peptides, either the same or different peptides. This is not the first time when two polypeptides with a similar disulfide pattern in the N-terminal region, but distinct in the C-terminal region, have a different role for the two different N- and C-terminal regions. Clear examples include zona pellucida (ZP) proteins, uromodulin, and vitelline envelope (VE) proteins, which contain a ZP domain with eight cysteines disulfide-linked, in which the N-terminal domain has the first four Cys disulfide-linked (Cys 1–4 and Cys2–3) and is involved in protein polymerization, while the C-terminal domain has different disulfide linkages in different ZP and VE proteins and regulates the interaction of the N-terminal domains [7, 8, 24, 25, 26, 27, 28]. Whether peptide Phα1β has different disulfide bridges in different situations, this remains to be determined.

Conclusions

By using 1) nanoLC-MS/MS-, 2) direct infusion ESI-MS- and 3) direct infusion ESIMS/MS-based analysis of the a) undigested, b) tryptically digested, c) AspN digested, and d) AspN/trypsin digested peptide, we have determined that in the synthetic spider peptide Phα1β, all cysteines are disulfide linked. We have determined the specific disulfide linkages to be Cys 1–2, Cys 3–4, Cys 5–6, Cys 7–8, Cys 9–10, and Cys 11–12. The synthetic peptide has identical sequence and displays the same functionality and activity as that of the native peptide and thus can be used as a model for the native peptide. We also determined that in the native peptide, at least the first eight cysteines are disulfide-linked, the last four cysteines are disulfide-linked, and that all 12 cysteines are disulfide-linked. Although the linkages we determined contradict all predicted homology models, it is possible that the peptide exists in different forms containing varying disulfide bond arrangements, while maintaining the same function and activity, as in human vitronectin.

Notes

Acknowledgements

The authors thank the members of the lab for fruitful discussions.

Funding Information

KLW was supported by the ASPIRE Graduate Student Fellowship through Clarkson University’s CUPO Office. MM was supported by the Fulbright Senior Postdoctoral Fellowship awarded by the Romania-USA Fulbright Commission to MM (guest) and CCD (host). MVG’s work was supported through REDE Fapemig (CBB-RED-00006-14), CNPq (471070/2012-2), Capes Toxinology (AUX-PE 1444/2011), and Fapemig (PPM-00482-15).

Supplementary material

13361_2018_1904_MOESM1_ESM.pptx (351 kb)
ESM 1 Supplemental Figure 1 A) MS spectra of the undigested Phα1β peptide demonstrating of the presence of the fully disulfide linked peptide by m/z 862.9 (7+), 1006.6 (6+), 1207.7 (5+) and 1509.1 (4+), which correspond to the mass of the Phα1β peptide minus 12 Da and B) The MS/MS fragmentation for each of the previously mentioned precursor ions demonstrating the inability for the peaks to fragment sufficiently enough to assign disulfide arrangements. Supplemental Figure 2 Identification of Cys7-8 disulfide linked. A) MSMS fragmentation of the precursor ion (inset) with m/z 863.37(2+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFK with cysteines disulfide linked (minus 2 Da). B) MSMS fragmentation of the precursor ion (inset) with m/z 879.33(2+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFK with cysteines disulfide linked (minus 2 Da), and modified by 32 Da. C) MSMS fragmentation of the precursor ion (inset) with m/z 895.82(2+) corresponds to the mass of the peptide DNQCYCPPGSSLGIFK with cysteines disulfide linked (minus 2 Da), and modified by 64 Da. The spectra from A, B & C resulted from an AspN digestion. Supplemental Figure 3 Confirmation of the Cys7-8 and Cys9-10 disulfide linkage by MSMS fragmentation of the precursor ion (inset) with m/z 1044.88 (2+). This corresponds to the peptide DNQCYCPPGSSLGIFKCSCA with Cys 7-to Cys 10 disulfide linked (minus 4 Da). Supplemental Figure 4 Confirmation of all disulfide linkages present in the peptide by MALDI-MS analysis of the Phα1β peptide. A) MS spectrum of the intact peptide and the peptide fragments; B) LIFT MS/MS spectrum of the intact peptide. These experiments suggest that all Cys 1 to Cys 12, Cys 1 to Cys 8 and Cys 9 to Cys 12 are disulfide linked (A). These experiments also suggest that Cys 3 to Cys 12, Cys 9 to Cys 12 and Cys 11-12 are also disulfide linked. Supplemental Figure 5 Zoom into the 6 kDa region of MALDI-MS spectra for the peaks that correspond to the intact recombinant (A) and native (B) Phα1β peptide, which demonstrates that all cysteine residues are disulfide linked. Supplemental Figure 6 MALDI-MS spectra for the peaks that correspond to the intact and truncated recombinant (A) and native (B) Phα1β peptide, as analyzed in linear (A) and reflective (B) positive mode. Supplemental Figure 7 MALDI-MS/MS fragmentation of the recombinant (A) and native (B) Phα1β peptides. In both spectra, we identified two fragments with m/z 1233.26(1+) and 2133.82(1+) (A) or 1233.56(1+) and 2134.02(1+) (B), which corresponded to peptides FKCSCAHANKY and CIPRGEICTDDCECCGCDNQ, each with their cysteines intramolecularly disulfide linked. (PPTX 351 kb)

References

  1. 1.
    Cordeiro Mdo, N., de Figueiredo, S.G., Valentim Ado, C., Diniz, C.R., von Eickstedt, V.R., Gilroy, J., Richardson, M.: Purification and amino acid sequences of six Tx3 type neurotoxins from the venom of the Brazilian ‘armed’ spider Phoneutria nigriventer (Keys). Toxicon. 31, 35–42 (1993)CrossRefGoogle Scholar
  2. 2.
    Souza, A.H., Ferreira, J., Cordeiro Mdo, N., Vieira, L.B., De Castro, C.J., Trevisan, G., Reis, H., Souza, I.A., Richardson, M., Prado, M.A., Prado, V.F., Gomez, M.V.: Analgesic effect in rodents of native and recombinant Ph alpha 1beta toxin, a high-voltage-activated calcium channel blocker isolated from armed spider venom. Pain. 140, 115–126 (2008)CrossRefGoogle Scholar
  3. 3.
    Vieira, L.B., Kushmerick, C., Hildebrand, M.E., Garcia, E., Stea, A., Cordeiro, M.N., Richardson, M., Gomez, M.V., Snutch, T.P.: Inhibition of high voltage-activated calcium channels by spider toxin PnTx3-6. J. Pharmacol. Exp. Ther. 314, 1370–1377 (2005)CrossRefGoogle Scholar
  4. 4.
    Rigo, F.K., Trevisan, G., Rosa, F., Dalmolin, G.D., Otuki, M.F., Cueto, A.P., de Castro Junior, C.J., Romano-Silva, M.A., Cordeiro, M.d.N., Richardson, M., Ferreira, J., Gomez, M.V.: Spider peptide Phα1β induces analgesic effect in a model of cancer pain. Cancer Sci. 104, 1226–1230 (2013)CrossRefGoogle Scholar
  5. 5.
    Yang, C.Y., Kim, T.W., Weng, S.A., Lee, B.R., Yang, M.L., Gotto Jr., A.M.: Isolation and characterization of sulfhydryl and disulfide peptides of human apolipoprotein B-100. Proc. Natl. Acad. Sci. U. S. A. 87, 5523–5527 (1990)CrossRefGoogle Scholar
  6. 6.
    van der Wel, H., Iyengar, R.B., van Brouwershaven, J., van Wassenaar, P.D., Bel, W.J., van der Ouderaa, F.J.: Assignment of the disulphide bonds in the sweet-tasting protein thaumatin I. Eur. J. Biochem. 144, 41–45 (1984)CrossRefGoogle Scholar
  7. 7.
    Darie, C.C., Biniossek, M.L., Jovine, L., Litscher, E.S., Wassarman, P.M.: Structural characterization of fish egg vitelline envelope proteins by mass spectrometry. Biochemistry. 43, 7459–7478 (2004)CrossRefGoogle Scholar
  8. 8.
    Darie, C.C., Biniossek, M.L., Gawinowicz, M.A., Milgrom, Y., Thumfart, J.O., Jovine, L., Litscher, E.S., Wassarman, P.M.: Mass spectrometric evidence that proteolytic processing of rainbow trout egg vitelline envelope proteins takes place on the egg. J. Biol. Chem. 280, 37585–37598 (2005)CrossRefGoogle Scholar
  9. 9.
    Sokolowska, I., Ngounou Wetie, A.G., Woods, A.G., Darie, C.C.: Automatic determination of disulfide bridges in proteins. J Lab Autom. 17, 408–416 (2012)CrossRefGoogle Scholar
  10. 10.
    Darie, C.C., Deinhardt, K., Zhang, G., Cardasis, H.S., Chao, M.V., Neubert, T.A.: Identifying transient protein-protein interactions in EphB2 signaling by blue native PAGE and mass spectrometry. Proteomics. 11, 4514–4528 (2011)CrossRefGoogle Scholar
  11. 11.
    Sokolowska, I., Dorobantu, C., Woods, A.G., Macovei, A., Branza-Nichita, N., Darie, C.C.: Proteomic analysis of plasma membranes isolated from undifferentiated and differentiated HepaRG cells. Proteome Sci. 10, 47–47 (2012)CrossRefGoogle Scholar
  12. 12.
    Sokolowska, I., Gawinowicz, M.A., Ngounou Wetie, A.G., Darie, C.C.: Disulfide proteomics for identification of extracellular or secreted proteins. Electrophoresis. 33, 2527–2536 (2012)CrossRefGoogle Scholar
  13. 13.
    Spellman, D.S., Deinhardt, K., Darie, C.C., Chao, M.V., Neubert, T.A.: Stable isotopic labeling by amino acids in cultured primary neurons: application to brain-derived neurotrophic factor-dependent phosphotyrosine-associated signaling. Mol Cell Proteomics : MCP. 7, 1067–1076 (2008)CrossRefGoogle Scholar
  14. 14.
    Soding, J.: Protein homology detection by HMM-HMM comparison. Bioinformatics. 21, 951–960 (2005)CrossRefGoogle Scholar
  15. 15.
    Waterhouse, A.M., Procter, J.B., Martin, D.M., Clamp, M., Barton, G.J.: Jalview Version 2—a multiple sequence alignment editor and analysis workbench. Bioinformatics. 25, 1189–1191 (2009)CrossRefGoogle Scholar
  16. 16.
    Vyas, V.K., Ukawala, R.D., Ghate, M., Chintha, C.: Homology modeling a fast tool for drug discovery: current perspectives. Indian J. Pharm. Sci. 74, 1–17 (2012)CrossRefGoogle Scholar
  17. 17.
    Xiang, Z.: Advances in homology protein structure modeling. Curr. Protein Pept. Sci. 7, 217–227 (2006)CrossRefGoogle Scholar
  18. 18.
    Nadezhdin, K.D., Romanovskaia, D.D., Sachkova, M.Y., Oparin, P.B., Kovalchuk, S.I., Grishin, E.V., Arseniev, A.S., Vassilevski, A.A.: Modular toxin from the lynx spider Oxyopes takobius: structure of spiderine domains in solution and membrane-mimicking environment. Protein Sci. 26, 611–616 (2017)CrossRefGoogle Scholar
  19. 19.
    Villegas, E., Adachi-Akahane, S., Bosmans, F., Tytgat, J., Nakajima, T., Corzo, G.: Biochemical characterization of cysteine-rich peptides from Oxyopes sp. venom that block calcium ion channels. Toxicon. 52, 228–236 (2008)CrossRefGoogle Scholar
  20. 20.
    Cheek, S., Krishna, S.S., Grishin, N.V.: Structural classification of small, disulfide-rich protein domains. J. Mol. Biol. 359, 215–237 (2006)CrossRefGoogle Scholar
  21. 21.
    Kamikubo, Y., De Guzman, R., Kroon, G., Curriden, S., Neels, J.G., Churchill, M.J., Dawson, P., Ołdziej, S., Jagielska, A., Scheraga, H.A., Loskutoff, D.J., Dyson, H.J.: Disulfide bonding arrangements in active forms of the somatomedin B domain of human vitronectin. Biochemistry. 43, 6519–6534 (2004)CrossRefGoogle Scholar
  22. 22.
    Zhou, A.: Functional structure of the somatomedin B domain of vitronectin. Protein Sci. 16, 1502–1508 (2007)CrossRefGoogle Scholar
  23. 23.
    Rigo, F.K., Trevisan, G., De Prá, S.D.-T., Cordeiro, M.N., Borges, M.H., Silva, J.F., Santa Cecilia, F.V., de Souza, A.H., de Oliveira Adamante, G., Milioli, A.M., de Castro Junior, C.J., Ferreira, J., Gomez, M.V.: The spider toxin Phα1β recombinant possesses strong analgesic activity. Toxicon. 133, 145–152 (2017)CrossRefGoogle Scholar
  24. 24.
    Boja, E.S., Hoodbhoy, T., Fales, H.M., Dean, J.: Structural characterization of native mouse zona pellucida proteins using mass spectrometry. J. Biol. Chem. 278, 34189–34202 (2003)CrossRefGoogle Scholar
  25. 25.
    Han, L., Monne, M., Okumura, H., Schwend, T., Cherry, A.L., Flot, D., Matsuda, T., Jovine, L.: Insights into egg coat assembly and egg-sperm interaction from the X-ray structure of full-length ZP3. Cell. 143, 404–415 (2010)CrossRefGoogle Scholar
  26. 26.
    Jovine, L., Darie, C.C., Litscher, E.S., Wassarman, P.M.: Zona pellucida domain proteins. Annu. Rev. Biochem. 74, 83–114 (2005)CrossRefGoogle Scholar
  27. 27.
    Monne, M., Han, L., Schwend, T., Burendahl, S., Jovine, L.: Crystal structure of the ZP-N domain of ZP3 reveals the core fold of animal egg coats. Nature. 456, 653–657 (2008)CrossRefGoogle Scholar
  28. 28.
    Zhao, M., Boja, E.S., Hoodbhoy, T., Nawrocki, J., Kaufman, J.B., Kresge, N., Ghirlando, R., Shiloach, J., Pannell, L., Levine, R.L., Fales, H.M., Dean, J.: Mass spectrometry analysis of recombinant human ZP3 expressed in glycosylation-deficient CHO cells. Biochemistry. 43, 12090–12104 (2004)CrossRefGoogle Scholar

Copyright information

© American Society for Mass Spectrometry 2018

Authors and Affiliations

  • Kelly L. Wormwood
    • 1
  • Armand Gatien Ngounou Wetie
    • 1
  • Marcus Vinicius Gomez
    • 2
  • Yue Ju
    • 3
  • Paul Kowalski
    • 3
  • Marius Mihasan
    • 1
    • 4
  • Costel C. Darie
    • 1
    Email author
  1. 1.Biochemistry & Proteomics Group, Department of Chemistry & Biomolecular ScienceClarkson UniversityPotsdamUSA
  2. 2.Institute of Education and Research Santa Casa Belo Horizonte-Laboratory of ToxinsBelo HorizonteBrazil
  3. 3.Bruker DaltonicsBillericaUSA
  4. 4.Department of BiologyAlexandru Ioan Cuza University of IasiIasiRomania

Personalised recommendations