Identification of Redox-Active Cell-Surface Proteins by Mechanism-Based Kinetic Trapping

See allHide authors and affiliations

Science's STKE  18 Dec 2007:
Vol. 2007, Issue 417, pp. pl8
DOI: 10.1126/stke.4172007pl8


A number of thiol-dependent oxidoreductases are released from cells and act on the cell surface. Correspondingly, several cell-surface processes appear to depend on catalyzed thiol-disulfide exchange, including integrin activation and the fusion of viral particles with the host membrane. Tumor cells frequently increase the abundance of secreted and cell-surface forms of particular oxidoreductases, and evidence suggests that oxidoreductases released from tumor cells promote growth and contribute to the remodeling of the cellular microenvironment. Few cell-surface or membrane proteins that are targeted by extracellular redox enzymes have been identified. One major reason for this slow progress is the highly transient nature of thiol-disulfide exchange, making its detection by conventional techniques difficult or impossible. Here we describe the application of an activity-based proteomics approach, also known as "mechanism-based kinetic trapping," to identify individual cell-surface target proteins that engage in disulfide exchange with thiol-dependent oxidoreductases. Although we have applied this approach to thioredoxin-1, it should also be applicable to other members of the thioredoxin superfamily whose activity is based on the CXXC active-site motif.


Disulfide bonds formed in the secretory pathway contribute to the stability of protein domains destined for the extracellular compartment and are usually considered static structural reinforcements. Whereas this is probably true for the majority of ectodomain disulfide bonds, it is now evident that some disulfide bonds are special in that they undergo dynamic redox changes (such as reduction, isomerization, and reoxidation) at the cell surface (1) that serve to regulate protein function. For example, the cell surface–associated protein disulfide isomerase (PDI) mediates conformational changes in the viral protein gp120 that are required for HIV-1 entry (2) and mediates the conversion of integrin αIIBβ3 from a low-affinity to a high-affinity conformation (3). However, the identification of cell-surface signaling receptors regulated by particular thiol oxidoreductases has proven difficult.

Thiol-based oxidoreductases found on the cell surface—including PDI, ERp57, P5, and thioredoxin-1 (Trx1)—belong to the thioredoxin superfamily. Their basic mechanism of action is well understood. In the case of the reduction reaction, the target disulfide bond is attacked by the upstream cysteine in the CXXC motif of the oxidoreductase (Cys32 in Trx1), leading to the formation of a mixed disulfide bond that covalently links the oxidoreductase to its target protein. In a second step, the mixed disulfide is resolved by nucleophilic attack of the downstream cysteine in the CXXC motif (Cys35 in Trx1), thus generating a reduced substrate and an oxidized enzyme (4). In accordance with the reaction mechanism, removal of the resolving (downstream) cysteine substantially prolongs the lifetime of the intermediary enzyme-substrate conjugate (5, 6). Thus, oxidoreductases that have been mutated so that target proteins remain covalently linked to the mutant enzyme can be used to isolate and analyze the target protein (Fig. 1), and we call these mutant enzymes "oxidoreductase trapping mutants."

Fig. 1.

Principle of cell-surface kinetic trapping. (A) In the recombinant Trx1 trapping mutant (C35S), the second (resolving) cysteine of the CXXC motif has been replaced by serine (–OH). The protein is also equipped with an SBP tag. (B) Trx1(C35S) forms a mixed disulfide bond with susceptible proteins on the cell surface. (C) The resulting complexes are solubilized and (D) purified using the SBP tag on the oxidoreductase. The trapped protein is then analyzed by immunoblotting and mass spectrometry.

Oxidoreductase trapping mutants were initially used to study individual enzyme-substrate interactions, such as those between plant thioredoxin f and fructose-1,6-bisphosphatase (7) or between human Trx1 and the nuclear protein redox factor 1 (Ref-1) (8). Later, the principle was exploited to screen for Trx-interacting proteins in yeast (9) and in plant cellular compartments [reviewed in (10)]. Our group applied mechanism-based kinetic trapping to the cell surface of living lymphocytes, which led to the identification of CD30 as the principal Trx1-interacting protein on the surface of these cells (11).

Here, we provide a detailed Protocol for the identification of cell-surface target proteins of thiol-based oxidoreductases that use the CXXC motif. In short, the oxidoreductase trapping mutant is prepared as a recombinant protein, applied to cultured cells of interest, and allowed to interact with the cell surface, thus potentially forming stable disulfide-linked complexes with susceptible proteins. After the removal of any unbound recombinant oxidoreductase, cells are lysed and the oxidoreductase-containing complexes are recovered for analysis.

This procedure allows assessment of the oxidoreductase reactivity of proteins embedded in their native microenvironment, namely, the intact and physiologically active surface of living cells. Most previous studies of Trx1 reactivity have been performed on proteins that have been removed from their natural environment, such as in cellular lysates or membrane preparations. Performing the trapping reaction on intact cells might be closer to the situation in vivo, as it is conceivable that protein disulfide exchange interactions are limited and controlled by native context and location.

Recombinant oxidoreductases require additional modifications to facilitate the efficient purification of covalently captured target proteins. We found it expedient to equip the recombinant proteins with a combination of one high-capacity and one high-affinity tag. The high-capacity hexahistidine tag (His6) facilitates the purification of proteins from bacteria. In contrast, the high-affinity tag is required to efficiently recapture the Trx1–target protein complexes after their formation on the cell surface. As a high-affinity tag, we used the 38 amino acid–containing streptavidin-binding peptide (SBP), which binds to the biotin-binding pocket of streptavidin with nanomolar affinity (12), thus enabling highly efficient and quantitative purification (13). Moreover, the high affinity of the interaction (KD = 2.5 nM) allows for stringent washing conditions, which makes it highly suitable for the purification of small amounts of disulfide-linked complexes for mass spectrometry.

The mixed disulfide complexes that are generated as products of cell-surface kinetic trapping can be analyzed in at least three ways. First, the detection of the oxidoreductase on Western blots will reveal if (and to what extent) oxidoreductase–target protein complexes have been formed. Second, these complexes can be analyzed by antibodies against the known or suspected target proteins. For example, previously established target proteins (e.g., CD4 in the case of Trx1) may serve as positive controls. Third, large-scale preparation of captured complexes can be used to identify unknown target proteins by mass spectrometry. Cell-surface kinetic trapping might also be useful for the comparative analysis of cell-surface redox differences between different conditions or cell types.


Ammonium sulfate

Antibody that recognizes the protein of interest in immunoblotting

BCA Protein Assay Kit (Pierce)


B-PER Bacterial Protein Extraction Reagent (Pierce)

Bromophenol blue

Complete protease inhibitors (Roche)

Coomassie Brilliant Blue G250 (Serva)

Dithiothreitol (DTT)

Escherichia coli DH5-alpha or E. coli XL1-Blue

E. coli M15 (pREP4) (Qiagen)


Heat-inactivated fetal bovine serum

HIS-Select nickel affinity gel (Sigma-Aldrich)

Imidazole (AppliChem)

Iscove’s modified Dulbecco’s medium (IMDM) 1640 (1×) without l-glutamine (Gibco)

Iodoacetamide (IAA)

l-glutamine 200 mM (100×), liquid (Gibco)


Na2-EDTA (0.5 M, pH 8.0)

N-ethylmaleimide (NEM)

Penicillin-streptomycin, liquid (contains penicillin, 10,000 units/ml, and streptomycin, 10,000 μg/ml) (Gibco)

Phosphoric acid

Potassium chloride (KCl)

Potassium phosphate monobasic (KH2PO4)

Protein molecular weight marker [e.g., Precision Plus Protein Dual Color Standard (Bio-Rad)]

QIAprep Spin Miniprep Kit or Maxi Kit (Qiagen)

QuickChange Multi Site-Directed Mutagenesis Kit (Stratagene)

Quick Ligation Kit (New England BioLabs)

Sodium chloride (NaCl)

Sodium dodecyl sulfate (SDS) (10%)

Sodium phosphate dibasic (Na2HPO4; anhydrous)

Streptavidin Sepharose High Performance beads (GE Healthcare)


Tris-HCl (1 M, pH 7.4)

Triton X-100


Bottle-top filters [e.g., 500-ml-capacity filters (Nalgene)]

Economy Mini-Spin columns (Pierce)

Gel electrophoresis system [e.g., SE 400 Sturdier Vertical Slab Gel Electrophoresis power units (Hoefer)]

Gel-drying system [e.g., GelAir Drying System (Bio-Rad)]

Rotating wheel

Semi-dry transfer equipment [e.g., Trans-Blot SD Semi-Dry Transfer Cell system (Bio-Rad)]

Sephadex PD-10 desalting column (GE Healthcare)

Slide-A-Lyzer dialysis cassettes (Pierce)


Transfer membrane [nitrocellulose or polyvinylidene difluoride (PVDF)]

TufRol Roller Bottles, pleated surface/easy on/off (BD Biosciences)


Note: Prepare all buffers with double-distilled water (ddH2O).

Recipe 1: Phosphate-Buffered Saline (PBS, 1×, pH 7.4)
NaCl8 g
KCl201 mg
Na2HPO4 (anhydrous)1.42 g
KH2PO4272 mg
ddH2O1000 ml
Prepare 1000 ml and store at room temperature.
Recipe 2: Tris-Buffered Saline (TBS, 1×, pH 7.5)
Tris-HCl (1 M, pH 7.4)10 ml
NaCl8.77 g
ddH2O990 ml
Prepare 1000 ml and store at room temperature.
Recipe 3: DTT Solution (1 M)
DTT1.54 g
ddH2O10 ml
Prepare 10 ml and store at –20°C in 200-μl aliquots. Thaw an aliquot immediately before use, but do not reuse.
Recipe 4: NEM Solution (1 M)
NEM1.25 g
Ethanol (100%)10 ml
Prepare 10 ml and store at –20°C in 200-μl aliquots. Thaw an aliquot before use, but do not reuse.
Recipe 5: Cell Culture Medium
IMDM 1640 (1×) without l-glutamine10,000 ml
Heat-inactivated fetal bovine serum1000 ml
l-glutamine100 ml
Penicillin-streptomycin100 ml
Prepare 10 liters and store at 4°C.
Recipe 6: Lysis Buffer
TBS solution (1×, pH 7.5) (Recipe 2)495 ml
Triton X-1005 ml
Prepare 500 ml and store at 4°C.
Recipe 7: Washing Buffer 1 (1% Triton X-100 in TBS)
TBS solution (1×, pH 7.5) (Recipe 2)495 ml
Triton X-1005 ml
Prepare 500 ml and store at 4°C.
Recipe 8: Washing Buffer 2 (0.1% Triton X-100 in TBS)
TBS solution (1×, pH 7.5) (Recipe 2)100 ml
Triton X-100100 μl
Prepare 100 ml and store at 4°C.
Recipe 9: Elution Buffer (4 mM Biotin in PBS)
PBS solution (1×, pH 7.5) (Recipe 1)10 ml
Biotin 2.44 g
Prepare 10 ml and store at 4°C. Elution buffer is stable for 2 months at 4°C.
Note: Biotin is largely insoluble at higher concentrations than 4 mM.
Recipe 10: SDS Sample Buffer (10×), Nonreducing
SDS (10%)1 ml
Tris-HCl (1 M, pH 7.4)5 ml
Glycerol 2 ml
Bromophenol blue 2 mg
Note: Add 1:10 to samples. For reducing sample buffer, add 200 mM DTT to buffer before use (20 mM DTT final concentration).
Note: SDS powder is very electrostatic and toxic. Wear a mask and gloves when handling SDS.
Recipe 11: Colloidal Coomassie Staining Solution
Ammonium sulfate100 g
Phosphoric acid10 ml
Coomassie Brilliant Blue G2501 g
Prepare 800 ml in water.
Add 200 ml of methanol just before use.


Preparation of Recombinant Tagged Thioredoxin-1 (Trx1) Proteins

Recombinant proteins for trapping may be prepared by various standard procedures. Therefore, detailed cloning, expression, and purification protocols are not provided here. Instead, to provide an example, we briefly outline how to prepare recombinant tagged thioredoxin-1 (Trx1) proteins (wild-type and mutant) for use in trapping experiments.

1. Insert the coding sequence for human Trx1 (or the oxidoreductase of interest) into the pQE-60 bacterial expression vector at the Nco I and Bgl II restriction sites to generate a Trx1 containing a C-terminal hexahistidine tag (His6).

Note: Check which restriction sites are appropriate for other oxidoreductases.

2. Generate an SBP-encoding DNA from four synthetic oligonucleotides (see Table 1), using the Quick Ligation Kit, and insert into the pQE-60-Trx1-His6 construct at the Bgl II site.

Table 1.

Oligonucleotides used for the generation of an SBP-encoding DNA fragment.

Note: The resulting construct encodes a Trx1 with two consecutive C-terminal tags (Trx1-SBP-His6).

3. Generate the Trx1 trapping mutation by replacing the second catalytic cysteine (Cys35) with serine, using the QuickChange Multi Site-Directed Mutagenesis Kit according to the manufacturer’s protocol.

4. Prepare DNA with the use of the QIAprep Spin Miniprep Kit or the HiSpeed Plasmid Maxi Kit.

5. Transform E. coli M15 (pREP4) cells with the expression plasmid, using a standard protocol.

6. Induce protein expression by adding 1 mM IPTG to bacterial cultures for 3 hours.

7. Lyse the bacteria harvested from 1 liter of culture in 40 ml of B-PER Bacterial Protein Extraction Reagent supplemented with 10 mM imidazole.

8. Purify recombinant protein by Ni chelate affinity purification either manually (His-Select Nickel Affinity Gel) or with the use of 1-ml HisTrap FF columns on the ÄKTA purifier system equipped with a FRAC-950 fraction collector.

9. Dialyze protein-containing fractions against PBS overnight, using Slide-A-Lyzer dialysis cassettes.

10. Determine protein concentrations after dialysis, using the BCA Protein Assay Kit according to the manufacturer’s instructions.

11. Add DTT to a final concentration of 5 mM and EDTA to a final concentration of 5 mM, and store 1-ml aliquots at 4°C until needed.

Reduction and Desalting of Recombinant Proteins Before Use in Trapping Experiments

Typically, cell-surface kinetic trapping is initially performed on a small scale, so as to analyze the resulting oxidoreductase-substrate complexes on a Western blot and obtain basic information about the prominence and abundance of the interactions. The small-scale approach also offers the opportunity to test specific candidate proteins for cell-surface disulfide exchange activity, as revealed by differences in the mobility of proteins under nonreducing and reducing conditions. Subsequently performed on a larger scale, cell-surface kinetic trapping can be used to purify novel oxidoreductase target proteins for analysis by mass spectrometry. We provide protocols for both applications: For small-scale analysis of complexes on Western blots, follow the instructions and quantities marked "SS"; for the large-scale approach, follow the instructions marked "LS." Sections lacking specific attributions apply to both large- and small-scale analyses. For small-scale analysis (SS), 100 μg of recombinant Trx1 per trapping reaction or sample is used. For large-scale analysis (LS), 1000 μg of recombinant Trx1 per trapping reaction or sample is used. Even when stored in the presence of DTT, recombinant oxidoreductases may become oxidized during prolonged storage. In the case of Trx1, oxidation leads to the formation of dimers, intramolecular disulfide bonds, or both. Therefore, before each experiment, fresh DTT is added to reduce the oxidoreductase. The DTT is later removed by desalting.

1. Incubate recombinant Trx1 (SS: 100 μg/sample; LS: 1 mg) in a total volume of 1 ml of PBS (Recipe 1) containing 20 mM freshly dissolved DTT for 30 min on ice.

Note: If the volume of recombinant protein solution needed exceeds 1 ml, larger incubation volumes may be used. However, the increased volume must be considered when desalting the protein on the column.

2. Equilibrate a Sephadex PD-10 desalting column with 25 ml of PBS and discard the flow-through.

3. Apply the freshly reduced Trx1 (1 ml) to the column and allow it to enter the resin.

4. Apply 2.5 ml of PBS to the column and discard the flow-through.

Note: If the volume of recombinant protein solution exceeds 1 ml, subtract the excess volume from the 2.5 ml of PBS applied to the column.

5. Apply 500 μl of PBS to the column and collect the flow-through in a 1.5-ml microfuge tube. Repeat this process four times, collecting each fraction in a separate tube.

6. Determine the protein content of each fraction by measuring its absorbance at 280 nm (OD280) in a spectrophotometer.

7. Pool the appropriate fractions and place them on ice.

Note: Covalent dimers formed by the Trx1 trapping mutant can be quite resistant to reduction by DTT, and reduction of dimers may be incomplete. However, in our experiments, the presence of small amounts of dimers in the trapping reaction did not influence the results.

Note: Ellman’s reagent (DTNB) may be used to verify successful removal of DTT. For a detailed protocol, refer to the manufacturer’s manual.

Note: For small-scale desalting of Trx1 in a volume of 30 to 130 μl, Zebra Desalt Spin Columns (Pierce) may be used.

Preparation of Cells

1. Culture cells at a maximal density of 0.5 × 106 to 1 × 106 cells/ml in the appropriate medium (e.g., Recipe 5).

2. SS: Harvest 5 × 107 cells per sample and pellet by centrifugation at 500g for 10 min at 4°C. LS: Culture 2.5 × 109 cells per sample in 1.5-liter roller bottles at 0.5 × 106 to 1 × 106 cells/ml; 12 to 16 hours before harvesting the cells, place the roller bottles in an upright position so that the cells can settle. Gently aspirate most of the supernatant without disturbing the sedimented cells.

Note: If roller bottles are transferred from 37°C to room temperature for settling, thermal convection interferes with the sedimentation of cells. To circumvent this problem, allow the cells to settle at 37°C.

Note: Suspension cells can be harvested by centrifugation; adherent cells should be released from culture flasks with cell dissociation buffer (Sigma) or 0.5 mM EDTA. Do not use trypsin, as it digests extracellular proteins.

3. Wash the cells once with 50 ml of ice-cold PBS.

4. Pellet the cells by centrifugation at 500g for 3 min at 4°C.

5. Optional: Resuspend the cell pellet in 50 ml of ice-cold PBS supplemented with 5 mM IAA and incubate on ice for 5 min.

Note: A brief treatment of cells with IAA alkylates accessible thiol groups on the cell surface, which may be helpful for certain control purposes (see Notes and Remarks).

6. Wash the cells twice with 50 ml of ice-cold PBS.

7. Pellet the cells by centrifugation at 500g for 3 min at 4°C.

8. SS: Resuspend the cells in ice-cold PBS so that the final incubation volume (including prereduced Trx1 solution) is 1.5 to 2 ml. Transfer the cell suspension to a 1.5-ml or 2.0-ml Eppendorf tube. LS: Resuspend the cells in 50 ml of ice-cold PBS.

Incubation of Live Cells with the Oxidoreductase Trapping Mutant

1. Add prereduced Trx1 to the cells: Add the volume of flow-through equivalent to 1 mg (LS) or 100 μg (SS) of recombinant Trx1 to 2.5 × 109 cells per sample (LS) or 5 × 107 cells per sample (SS).

Note: The final concentration of Trx1 will be 1 to 5 μM, depending on the incubation volume.

2. Incubate on a rotating wheel for 1 hour at 4°C.

Note: After the incubation, check the viability of the cells by Trypan blue exclusion. It is essential that cells are kept intact during the Trx1 incubation, as released intracellular proteins may interact with Trx1.

3. SS: Stop the reaction by transferring the cell suspension to a tube containing 10 ml of ice-cold PBS supplemented with 5 mM IAA. LS: Stop the reaction by pelleting the cells by centrifugation at 500g for 3 min at 4°C and then resuspending the pellet in 50 ml of ice-cold PBS supplemented with 5 mM IAA.

4. Pellet the cells by centrifugation at 500g for 3 min at 4°C.

5. Wash the cells once with ice-cold PBS (SS: 10 ml; LS: 50 ml) supplemented with 5 mM IAA.

6. Pellet the cells by centrifugation at 500g for 3 min at 4°C.

7. Wash the cells once with ice-cold PBS (SS: 10 ml; LS: 50 ml) supplemented with 10 mM NEM.

8. Pellet the cells by centrifugation at 500g for 3 min at 4°C.

Note: Washing cells with IAA blocks cell-surface thiol groups without compromising cell integrity. In the second wash, the cell-permeable alkylating agent NEM blocks intracellular thiols, thus preventing reduction of trapped complexes upon subsequent lysis.

Purification of Oxidoreductase–Target Protein Complexes

1. Immediately before use, supplement lysis buffer (Recipe 6) (SS: 10 ml; LS: 50 ml) with 20 mM NEM and complete protease inhibitor (according to manufacturer’s instructions). Lyse cells by incubation in supplemented lysis buffer for 1 hour at 4°C on a rotating wheel.

2. Obtain the postnuclear supernatant by centrifugation at 4600g for 10 min at 4°C.

Note: Keep an aliquot of the postnuclear supernatant for SDS-PAGE analysis.

3. Equilibrate streptavidin sepharose beads (bead volume, SS: 100 μl; LS: 500 μl) by washing them twice with 10 ml of washing buffer 1 (Recipe 7) and collecting the beads by centrifugation at 300g for 3 min at 4°C.

4. Resuspend beads in washing buffer 1 to obtain a 50% bead slurry (SS: 100 μl; LS: 500 μl).

5. Incubate the postnuclear supernatant with the equilibrated beads (SS: 200 μl of 50% slurry; LS: 1 ml of 50% slurry) on a rotating wheel for 1 hour at 4°C.

6. Collect the beads by centrifugation at 300g for 3 min at 4°C.

7. Wash the beads twice with 10 ml of washing buffer 1 supplemented with 500 mM NaCl and 10 mM NEM.

8. Pellet the beads by centrifugation at 300g for 3 min at 4°C.

9. Wash beads once with 10 ml of washing buffer 1 supplemented with 1 mM NEM.

10. Pellet the beads by centrifugation at 300g for 3 min at 4°C.

11. Wash the beads twice with 10 ml of washing buffer 1 supplemented with 1 M urea and 1 mM NEM.

12. Pellet the beads by centrifugation at 300g for 3 min at 4°C.

13. Wash the beads once with 10 ml of washing buffer 2 (Recipe 8).

14. Pellet the beads by centrifugation at 300g for 3 min at 4°C.

15. Transfer the beads to 1.5-ml microfuge tubes and collect the beads by centrifugation at 300g for 3 min at 4°C.

16. Elute protein complexes from the beads by resuspending the beads in elution buffer (Recipe 9) (SS: 100 μl; LS: 500 μl) and incubating on a rotating wheel for 30 min at 4°C.

17. Discard beads (e.g., by using Economy Mini-Spin columns) and store the eluted proteins at 4°C.

Note: If you do not proceed with the SDS-PAGE analysis on the same day, store samples at –20°C after boiling in SDS sample buffer.

Sample Preparation and SDS-PAGE

1. Prepare two aliquots of equal volume from each sample. Add 5 μl (SS) or 25 μl (LS) of 10× nonreducing SDS sample buffer (Recipe 10) to the first aliquot and 5 μl (SS) or 25 μl (LS) of 10× reducing SDS sample buffer to the second aliquot, and boil the samples for 5 min.

2. Cool the samples to room temperature on the bench.

3. To protect samples against subsequent in-gel thiol reduction or oxidation, add NEM to each sample to give a final concentration of 40 mM.

4. Load whole samples onto the appropriate SDS-PAGE gel.

Note for SS: As an alternative to large gels, small gels may also be used [e.g., Mini-PROTEAN 3 Cell (Bio-Rad)]. However, only half of the sample (~25 μl) may be loaded onto small gels.

Analysis of Oxidoreductase–Target Protein Complexes by Western Blotting

The sample is analyzed by Western blotting with Trx1-specific antibodies to confirm the presence of Trx1 in the analyzed samples. This procedure provides verification that Trx1–target protein complexes were formed and isolated.

1. Transfer proteins from the SDS gel to a membrane (nitrocellulose or PVDF), using a standard transfer protocol.

2. Block the membrane with a blocking solution for 30 min at room temperature.

3. Incubate the membrane with the appropriate antibody according to the manufacturer’s instructions. Recombinant Trx1 and its complexes may also be visualized by using a streptavidin–horseradish peroxidase (HRP) conjugate to detect the SBP tag.

Note: Streptavidin-HRP may produce background in combination with standard membrane blocking procedures. Use of a special blocking solution [e.g., Starting Block (Pierce)] is recommended.

Note: As a positive control, a sample from the postnuclear supernatant (see step 2 under Purification of Oxidoreductase–Target Protein Complexes) should be included.

Colloidal Coomassie Staining of Oxidoreductase–Target Protein Complexes

1. Wash the SDS-PAGE gel twice in ddH2O for 5 min.

2. Add colloidal Coomassie staining solution (Recipe 11) and incubate overnight at room temperature.

3. Exchange the staining solution and incubate for 2 hours at room temperature.

4. Destain the gel in ddH2O.

Note: After destaining, the bands of interest can be excised from the gel. Alternatively, the gel may be dried with an appropriate gel drying system [e.g., the GelAir Drying System (Bio-Rad)] and subsequently used for further analysis.

Target Protein Identification by Mass Spectrometry

To provide an example, we briefly summarize the procedure we followed in our experiments. Gel pieces excised from the colloidal Coomassie–stained gel were washed, reduced with DTT, alkylated with IAA, and incubated with trypsin (Promega). After desalting on Poros R2 material (Applied Biosystems), peptides were eluted into a precoated borosilicate nanoelectrospray needle (MDS Protana). Electrospray ionization mass spectrometry (ESI-MS) analysis was done on a quadrupole time-of-flight hybrid mass spectrometer (QStar, Applied Biosystems). Doubly and triply charged precursor ions were selected for fragmentation (unit resolution), using three different fragmentation energies (22, 27, and 31 V). A sequence tag was calculated from the fragment spectra and searched against the NCBI database by means of Mascot query.


Oxidation of Recombinant Oxidoreductases

A common problem experienced with purified thiol-dependent oxidoreductases is their susceptibility to oxidation in air, which may also lead to inactivation. Oxidation affects catalytic and/or noncatalytic cysteines and may lead to intra- and intermolecular disulfide bond formation. Although the oxidoreductase is always treated with fresh DTT before its application in the experiment (see Reduction and Desalting of Recombinant Proteins Before Use in Trapping Experiments), some oxidation products, such as Trx1(C35S) dimers, are quite resistant to reduction.

Removal of noncatalytic cysteines by site-directed mutagenesis offers one way to minimize the overall formation of oxidation products during storage. For example, Trx1 contains three additional cysteine residues distal to the active site (cysteines 62, 69, and 73). These residues are dispensable for catalytic activity but can cause oxidative inactivation by either intra- or intermolecular disulfide bond formation (14, 15). In our experiments with Trx1, the replacement of extra cysteines by alanines did not alter the experimental outcome.

In any case, we recommend using degassed buffers for storage and experimentation. Long-term storage of oxidoreductase preparations may also be facilitated by the freezing of aliquots. However, we found that the freezing and thawing of Trx1 preparations led to substantial protein precipitation. We therefore preferred to use freshly prepared protein, which was stored with DTT at 4°C until used. There is also the possibility of oxidative inactivation taking place during incubation of the trapping oxidoreductase with cells. However, we found that under the conditions of the experiment, the efficiency of the trapping reaction was not limited by oxidative inactivation.

Partial Degradation of Recombinant Oxidoreductases

After one-step metal chelate purification of recombinant protein, trace amounts of bacterial proteases may still be present in the sample. In the case of the Trx1-SBP-His6 protein, we sometimes observed a progressive loss of the C-terminal tag during prolonged storage. Degradation of the C-terminal tail was easily visualized on reducing SDS-PAGE gels. We strongly recommend testing the integrity of the protein preparation on a regular basis, especially after prolonged storage. It is important to always keep recombinant proteins at 4°C during preparation and storage. In addition, we found that EDTA in the storage buffer helped to suppress proteolysis.

Nonspecific or Nonreproducible Interactions

Occasionally, we observed that dying and lysing cells gave rise to additional and variable oxidoreductase interactions. These are likely to be nonspecific, as cell-surface proteins may lose their native environment or conformation. In addition, a loss of membrane integrity may allow the trapping oxidoreductase to form conjugates with intracellular proteins. It is therefore critical to maintain cellular integrity during the washing and trapping phases. During these steps, the cell suspension should remain homogeneous. Cell lysis is indicated by cell clumping and the partial clearing of the suspension. In any case, the integrity of cells should be determined by Trypan blue exclusion after the trapping step.

Notes and Remarks

Specificity Controls

To verify that the overall trapping procedure works, one can perform Western blotting to verify the capture of a previously established interaction partner. For Trx1, either CD4 (16) or CD30 (11) may serve as a positive control if the appropriate cell types are used. A nontrapping version of the oxidoreductase with an intact CXXC motif should always be used as a negative control to verify that the trapping is mechanism-based. It may also be useful to include in the experiments a trapping version of the oxidoreductase that lacks noncatalytic cysteines, so as to formally exclude the possibility that conjugation occurs outside the active site. To exclude the theoretical possibility that the trapping mutant of the oxidoreductase captures proteins by de novo disulfide bond formation between its active-site thiol and a free thiol on the cell surface (rather than by the disulfide exchange mechanism), a brief pretreatment of the cell surface with the membrane-impermeable alkylating agent IAA may be used (see Preparation of Cells).

The corresponding CXXS mutants of related oxidoreductases, such as glutaredoxin (Grx), can be used to investigate the specificity of oxidoreductase-substrate interactions. Oxidoreductase–target protein complexes should always be analyzed under nonreducing and reducing conditions in parallel to confirm the formation of direct disulfide bond formation. Disulfide-linked conjugates are susceptible to reduction, leading to a corresponding mobility shift on SDS-PAGE gels.

After identification of a new target protein, the interaction should be verified with wild-type oxidoreductase. For example, membrane-impermeant thiol-specific biotinylation agents, such as N-maleimidylpropionyl biocytin, can be used to demonstrate that the wild-type oxidoreductase generates new thiol groups in the target protein by reducing target disulfide bond(s).

Identification of TNFRSF8/CD30 as a Cell-Surface Target Protein of Trx1

We present a published experiment as an example of the trapping of target proteins on the cell surface of lymphocytic cell lines. Details not mentioned here can be found in the relevant publication (11). The purpose of this experiment was to identify receptors on the cell surface of lymphocytic cell lines that are targeted by extracellular Trx1. Cells were first incubated with the Trx1 trapping mutant or with wild-type Trx1 (as a control) in a small-scale experiment, and disulfide-linked Trx1 complexes were analyzed on Western blots (Fig. 2). One prominent trapping product of about 160 kD was detectable under nonreducing conditions; this product was susceptible to reduction, indicating a mixed disulfide conjugate. To identify the Trx1-interacting protein, we performed a large-scale trapping experiment and analyzed the results by colloidal Coomassie staining (Fig. 3A, left panel). Corresponding bands were analyzed by liquid chromatography (LC)–MS/MS and were identified as TNFRSF8/CD30, a member of the tumor necrosis factor receptor (TNFR) superfamily. To confirm the presence of both Trx1 and CD30 in the disulfide-mixed complex, we analyzed an aliquot of trapped complexes from the same experiment by Western blotting with Trx1- and CD30-specific antibodies, respectively (Fig. 3A, middle and right panels). Additional Western blotting experiments demonstrated that nanomolar concentrations of Trx1 are sufficient for the interaction with CD30 (Fig. 3B) and confirmed that the formation of stable mixed disulfide intermediates depended on the presence of the N-terminal and the absence of the C-terminal catalytic cysteine (Fig. 3C). Moreover, an interaction with CD30 was not observed for a Grx1 trapping mutant, indicating the specificity of the Trx1-CD30 interaction.

Fig. 2.

Trx1 targets one prominent interaction partner on the surface of a lymphocytic cell line. CCRF-CEM cells were incubated with 5 μM of the Trx1 trapping mutant (CSAAA), and disulfide-linked complexes were analyzed by immunoblotting with antibody to human Trx1 (αhTrx1) under nonreducing (–DTT) and reducing (+DTT) conditions. Conjugation of the unknown cell-surface protein is indicated (Trx1-S-S-X). [Figure reproduced from (11)]

Fig. 3.

Identification of TNFRSF8/CD30 as Trx1-interacting protein on the cell surface of lymphocytic cell lines. (A) LCL-721.220 cells (2.5 × 109) were incubated with 5 μM Trx1 trapping mutant (CSAAA) or wild-type Trx1 (CCAAA) as negative control. Disulfide-linked Trx1 complexes were analyzed by colloidal Coomassie staining under nonreducing (–DTT) and reducing (+DTT) conditions (left panel). Indicated bands were subjected to tryptic digestion and LC-MS/MS analysis. In parallel, part of the same sample was analyzed by immunoblotting with antibody to Trx1 (middle panel). The blot was stripped and reprobed with antibody to human CD30 (right panel). The Trx1-CD30 conjugate and monomeric CD30 are indicated. (B) CCRF-CEM T cells were incubated with different amounts of the Trx1 trapping mutant (CSAAA), and disulfide-linked Trx1 complexes were analyzed by immunoblotting with antibody to CD30 under nonreducing (–DTT) and reducing (+DTT) conditions. (C) CCRF-CEM T cells were incubated with recombinant Trx1 constructs or a Grx1 trapping mutant as indicated, and Trx1 complexes were analyzed as described in (B). The difference in signal intensity between the nonreduced and reduced form of CD30 is due to less efficient recognition of the reduced form by the CD30-specific antibody. [Figure reproduced from (11)]


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
View Abstract

Stay Connected to Science Signaling

Navigate This Article