Thioredoxin reductase 1 and NADPH directly protect protein tyrosine phosphatase 1B from inactivation during H2O2 exposure

Abstract

Regulation of growth factor signaling involves reversible inactivation of protein tyrosine phosphatases (PTPs) through the oxidation and reduction of their active site cysteine. However, there is limited mechanistic understanding of these redox events and their co-ordination in the presence of cellular antioxidant networks. Here we investigated interactions between PTP1B and the peroxiredoxin 2 (Prx2)/thioredoxin 1 (Trx1)/thioredoxin reductase 1 (TrxR1) network. We found that Prx2 becomes oxidized in PDGF-treated fibroblasts, but only when TrxR1 has first been inhibited. Using purified proteins, we also found that PTP1B is relatively insensitive to inactivation by H₂O₂ but found no evidence for a relay mechanism in which Prx2 or Trx1 facilitates PTP1B oxidation. Instead, these proteins prevented PTP1B inactivation by H₂O₂ Intriguingly, we discovered that TrxR1/NADPH directly protects PTP1B from inactivation when present during the H₂O₂ exposure. This protection was dependent on the concentration of TrxR1 and independent of Trx1 and Prx2. The protection was blocked by auranofin and required an intact selenocysteine residue in TrxR1. This activity likely involves reduction of the sulfenic acid intermediate form of PTP1B by TrxR1 and is therefore distinct from the previously described reactivation of end-point oxidized PTP1B, which requires both Trx1 and TrxR1. The ability of TrxR1 to directly reduce an oxidized phosphatase is a novel activity that can help explain previously observed increases in PTP1B oxidation and PDGF receptor phosphorylation in TrxR1 knockout cells. The activity of TrxR1 is therefore of potential relevance for understanding the mechanisms of redox regulation of growth factor signaling pathways.

Keywords: growth factor signaling; hydrogen peroxide; peroxiredoxin; protein tyrosine phosphatase (tyrosine phosphatase); redox regulation; thiol oxidation; thioredoxin reductase.

Figure 1.

Prx2 dimer formation is induced by PDGF-BB treatment in MEF cells pretreated with auranofin. A, representative images of MEF cells expressing cytosolic HyPer after 0 and 20 min without treatment (Control) or treatment with 100 ng/ml PDGF-BB. B, change in fluorescence over 20 min in untreated or PDGF-BB–stimulated Hyper-expressing MEF cells. Data from 19 transfected cells were analyzed in five different experiments (mean ± S.E.; *, p < 0.05). C, Prx2 immunoblot of non-reducing SDS-PAGE–resolved lysates from MEF cells treated with 50 ng/ml PDGF-BB for the indicated times. D, quantification of Prx2 dimers (percentage of total Prx2) from densitometry analyses of blots represented in C (n = 7). E, immunoblot as in C of lysates from cells pretreated for 1 h with the TrxR1 inhibitor auranofin (1 μm) before PDGF-BB treatment. F, quantitation of Prx2 dimers in auranofin-treated cells determined as in D (mean ± S.E.; *, p < 0.05).

Figure 2.

Reduced Prx2 protects PTP1B from inactivation by H₂O₂. A, analysis of the PTP1B His-tagged cleaved variant and PTP1B catalytic domain on SDS-PAGE stained with Coomassie Blue. The purity was ∼72% and ∼ 99%, respectively, as determined by densitometry using ImageJ. Side-by-side comparison of activity of the reduced PTP1B cleaved variant and PTP1B catalytic domain showed a mean substrate turnover of 269 and 355 min⁻¹, respectively (n = 3). B, H₂O₂-dependent inactivation of recombinant PTP1B. Reduced PTP1B (600 nm) was exposed to 100 μm or 1 mm H₂O₂ for the designated times and subsequently analyzed for PTP activity using a p-nitrophenyl phosphate (pNPP) substrate (n = 3; mean ± S.E.; *, p < 0.05). PTP1B activities are expressed as percentages of untreated controls. C, activity of PTP1B following treatment with H₂O₂ and reduced or oxidized Prx2. Fully reduced PTP1B (600 nm) was incubated for 30 min with 20 μm reduced Prx2 and 20 μm H₂O₂ or reduced or oxidized Prx alone and then assayed for PTP activity (n = 3; mean ± S.E.; *, p < 0.05; H₂O₂-treated compared with controls). D, Prx2 oxidation state after treatment with reduced PTP1B. Equimolar reduced PTP1B (150 nm) and oxidized Prx2 were incubated for 30 min and analyzed by non-reducing SDS-PAGE. A silver-stained gel (representative of three independent experiments) shows the positions of individual proteins as well as reduced Prx2 for comparison. The bottom and top oxidized Prx2 bands correspond to dimers containing one or two disulfides, respectively. E, activity of PTP1B following treatment with H₂O₂ or oxidized Prx1. Fully reduced PTP1B (600 nm) was incubated for 30 min with 10 μm H₂O₂ or 10 μm oxidized Prx alone and then assayed for PTP activity (n = 3; mean ± S.E.; *, p < 0.05; H₂O₂-treated compared with controls).

Figure 3.

Prx2 redox cycling with the Trx system protects PTP1B against inactivation during exposure to H₂O₂. A, effects of Prx2 and Trx system components during treatment of PTP1B with H₂O₂. Reduced PTP1B catalytic domain (600 nm) was preincubated for 30 min with 2 μm Trx1, 50 nm TrxR1 (specific activity, 17 units/mg), 200 μm NADPH, and 0–20 μm Prx2. H₂O₂ (100 μm) was then added, and samples were taken at 5, 15, and 30 min for measurement of PTP activity (n = 3; mean ± S.E.; *, p < 0.05). B, parallel analysis of Prx2 redox status from the experiment shown in A. Aliquots were taken from the assay shown in A after 2, 20, and 30 min of H₂O₂ treatment and analyzed by non-reducing SDS-PAGE. A Coomassie-stained gel (representative of three independent experiments) shows the positions of reduced and oxidized Prx2, the latter (arrows) showing two Prx2 bands corresponding to one or two disulfides.

Figure 4.

The Trx system reverses H₂O₂-inactivated PTP1B. A, reduced PTP1B (600 nm) was preincubated for 30 min either alone or with 2 μm Trx1, 0.5 μm TrxR1 (specific activity, 9.75 units/mg), and 200 μm NADPH, with or without Prx2 (10 μm). H₂O₂ (100 μm) was then added, and samples were taken at 5, 15, and 30 min for measurement of PTP activity (n = 3; mean ± S.E.; *, p < 0.05). B, H₂O₂ consumption by PTP1B and components of the Trx/Prx2 system. Combinations of Trx1 (2 μm), TrxR1 (0.5 μm, 18 units/mg), Prx2 (10 μm), and NADPH (200 μm) with and without reduced PTP1B (600 nm) were treated with 100 μm H₂O₂ at 22 °C, and concentrations of H₂O₂ at the indicated time points were determined by ferrous oxidation of xylenol orange assay (n = 3; mean ± S.E.; *, p < 0.05). C, reactivation of H₂O₂-inactivated PTP1B (cleaved form). Reduced PTP1B was treated with 1 mm H₂O₂ for 5 min and then with catalase to remove residual H₂O₂ and reactivated with 10 mm DTT or TrxR1 (2.5 μm) (specific activity, 18 units/mg) and NADPH (1 mm) with or without Trx1 (10 μm). After 45 min at 22 °C, samples were analyzed for PTP activity (n = 3; mean ± S.E.; *, p < 0.05). D, reactivation of H₂O₂-inactivated PTP1B catalytic domain. Reduced PTP1B was treated with 1 mm H₂O₂ for 5 min and then with catalase and reactivated with 10 mm DTT or TrxR1 (2.5 μm) (specific activity, 22 units/mg), NADPH (300 μm), and Trx1 (10 μm). After 5, 10, and 40 min at 22 °C, samples were analyzed for PTP activity (n = 3; mean ± S.E.; *, p < 0.05).

Figure 5.

TrxR1 activity protects PTP1B inactivation during exposure to H₂O₂. A, protection of PTP1B from H₂O₂ requires only TrxR1 and NADPH. PTP1B was treated with H₂O₂ in the presence of various components of the Trx system and assayed for PTP activity as shown in Fig. 3B. Rates of inactivation with Trx1/TrxR1 or TrxR1 without NADPH were indistinguishable from that with PTP1B alone (Fig. 3B) (n = 3; mean ± S.E.; *, p < 0.05). B, reduced PTP1B cleaved and PTP1B catalytic domain variant (600 nm) were treated with the indicated concentrations of TrxR1 (0.5 μm) and 200 μm NADPH and exposed to 100 μm H₂O₂ for 5, 15, and 30 min (n = 3; mean ± S.E.; *, p < 0.05). C, concentration-dependent protection of PTP1B activity by TrxR1. Reduced PTP1B (600 nm) was treated with the indicated concentrations of TrxR1 and 200 μm NADPH and exposed to 100 μm H₂O₂ for 5 min (n = 3; mean ± S.E.; *, p < 0.05). D, reduced DEP-1 (80 nm) was treated with the indicated concentrations of TrxR1 (0.5 μm) and 200 μm NADPH, exposed to 150 μm H₂O₂ for 5 min, and then assayed for PTP activity as in panel A (n = 3; mean ± S.E.; *, p < 0.05).

Figure 6.

The active-site selenocysteine residue of TrxR1 is indispensable for protection of PTP1B from H₂O₂-mediated inactivation. A, inhibition of TrxR1 protection by auranofin. Reduced PTP1B (600 nm) was treated with or without TrxR1 (0.5 μm, 9.75 units/mg), NADPH (200 μm), and auranofin (AU, 1 μm). PTP activity was measured at the indicated times after addition of 100 μm H₂O₂ (n = 3; mean ± S.E.; *, p < 0.05). B, lack of protection by active-site mutants of TrxR1. Reduced PTP1B was exposed to 100 μm H₂O₂ in the presence of wild-type TrxR1 (18 units/mg) or variants where the active-site SelCys was mutated to Cys (TrxR1Cys) or Ser (TrxR1Ser) and then assayed for PTP activity as in A (n = 3; mean ± S.E.; *, p < 0.05). C, schematic of how the redox activity of the Trx/TrxR/Prx system could regulate PTP1B activity. The bold arrows show the reaction of TrxR1 (green) with the sulfenic acid of PTP1B that, we propose, explains protection of PTP1B activity (blue) by TrxR1 as observed in this study (reaction 1). Additional activities of the Trx system in relation to PTP regulation, as described previously in the literature, are schematically shown in gray. The full scheme is discussed in the text.