Oxidative stress-induced expression and modulation of Phosphatase of Regenerating Liver-1 (PRL-1) in mammalian retina

Abstract

The phosphatase of regenerating liver-1, PRL-1, gene was detected in a screen for foveal cone photoreceptor-associated genes. It encodes a small protein tyrosine phosphatase that was previously immunolocalized to the photoreceptors in primate retina. Here we report that in cones and cone-derived cultured cells both PRL-1 activity and PRL-1 gene expression are modulated under oxidative stress. Oxidation reversibly inhibited the phosphatase activity of PRL-1 due to the formation of an intramolecular disulfide bridge between Cys104 within the active site and another conserved Cys, Cys49. This modulation was observed in vitro, in cell culture and in isolated retinas exposed to hydrogen peroxide. The same treatment caused a rapid increase in PRL-1 expression levels in cultured cells which could be blocked by the protein translation inhibitor, cycloheximide. Increased PRL-1 expression was also observed in living rats subjected to constant light exposure inducing photooxidative stress. We further demonstrated that both oxidation and overexpression of PRL-1 upon oxidative stress are greatly enhanced by inhibition of the glutathione system responsible for cellular redox regulation. These findings suggest that PRL-1 is a molecular component of the photoreceptor's response to oxidative stress acting upstream of the glutathione system.

Figure 1

Protein tyrosine phosphatase specificity of recombinant PRL-1. (A) Phosphatase activities were measured using the phosphatase substrate, DiFMUP (50 µM) whose reaction product, DiFMU, has an excitation/emission maxima of ~358 /450 nm. The starting concentration of recombinant PRL-1s was 800 µg/ml. The enzymes were serially diluted in reaction buffer (25 mM MOPS, 50 mM NaCl, 1 mM DTT and 0.05% Tween-20). The activity of C104S mutant PRL-1 was completely abolished. (B) Concentrations of the inhibitors used as indicated on the x-axis and activities relative to that in the absence of any inhibitors (100%) are shown. (C) Recombinant WT PRL-1, C104S PRL-1 and PTEN were incubated with 200 µM water soluble diC8-PIP₃ substrate at 37°C for 30 min followed by incubation with 200 µl malachite green solution for 15 min at room temperature. Release of phosphate was measured at an absorbance of 620 nm. The amount of free phosphate was calculated from the standard curve derived using phosphate solution standards. Activities are expressed as nmol of phosphate released per min per mg. Amount of enzyme used in the assay are indicated on the x-axis. Results are means (±SE) from representative experiments. Na₃VO₄, sodium orthovanadate; NaF, sodium fluoride; OA, okadaic acid.

Figure 2

In vitro reversible oxidative inhibition of PRL-1. (A) PRL-1 dialyzed in buffer containing 50 mM Tris, 50 mM Bis-Tris, 100 mM Na-Acetate (pH 7.0), adjusted to a final concentration of 2 mg/ml was incubated in the absence or presence of H₂O₂ in a total volume of 100 µl for 10 min at room temperature. The reaction was stopped by adding 12.5 U catalase, and the PRL-1 phosphatase activity was measured immediately after as described in experimental procedures using DiFMUP. (B) PRL-1 reactions, partially inactivated by 500 µM H₂O₂, were further incubated with 4 mM DTT for 30 min at room temperature. Remaining PTP activity was assayed with DiFMUP as described for Fig. 1 and expressed as % activity relative to the untreated control PRL-1. Treated PRL-1 was also shown on silver-stained SDS gel (top panel). Results are means from three independent experiments.

Figure 3

Intramolecular disulfide bond formation between Cys104 and Cys49. Recombinant WT and Cys mutant PRL-1s with or without the presence of varying concentrations (as indicated) of H₂O₂ or DTT, were subjected in SDS-PAGE under non-reducing conditions followed by silver staining. The reduced and oxidized forms of PRL-1 have different electrophoretic mobilities. Redu-, reduced; Oxi-, oxidized.

Figure 4

Oxidative status of PRL-1 in 661W cells (A, B) and porcine retina (C) exposed to H₂O₂. (A) Mouse photoreceptor-derived 661W cells were grown to 90% confluence. Cells were either incubated with varying concentrations of H₂O₂ (as indicated) for 30 min or (B) incubated with 1 mM H₂O₂ for varying times (as indicated). (C) Porcine retinas in eyecups were incubated with indicated concentrations of H₂O₂ (added to DMEM) for 30 min. Cell or retinal lysates were alkylated with 40 mM NEM to prevent disulfide exchange. Proteins were resolved by SDS-PAGE followed by immunoblot analysis with anti-PRL-1 antibody. Equal sample loading was monitored by reprobing the Western blot with anti-γ-tubulin antibodies (data not shown).

Figure 5

Time-dependent increase in PRL-1 protein levels in 661W cells exposed to low concentrations of H₂O₂. 661W Cells without pretreatment (-BSO) or pretreated with 1 mM BSO (+BSO) for 17 h, were incubated with 0.2 mM H₂O₂ for the indicated times. Equal amounts of protein from each sample were separated by SDS-PAGE followed by immunoblot analysis with anti-PRL-1. Equal sample loading was monitored by reprobing the Western blot with anti-γ-tubulin antibodies.

Figure 6

Influence of the translation inhibitor cycloheximide (CHX) on the time-dependent increase in PRL-1 protein levels in 661 W cells exposed to H₂O₂. 661W cells were exposed to 1 mM H₂O₂ for increasing times. One set of cells was pretreated with 100 µg/ml CHX for one hour prior to exposure to H₂O₂ (+CHX) and the other set were treated only with the H₂O₂ (-CHX). Equal amounts of protein from each well were separated by SDS-PAGE followed by immunoblot analysis with anti-PRL-1 and equal sample loading was monitored by reprobing with anti-tubulin antibodies.

Figure 7

Light microscopy of rat retina following 7 days constant light exposure (A) Rat retina exposed to 7 days constant light (B) Rat retina exposed for 7 days to constant light, at nine weeks after return to normal 12 hr cyclic light. RPE, retinal pigment epithelium; OS, outer segments; ONL, outer nuclear layer of photoreceptors; INL, inner nuclear layer, D, days. Scale bars equal 20 microns.

Figure 8

Increasing expression of PRL-1 mRNA (A) and protein (B) in rat retinas due to constant moderate light exposure. Albino rats were exposed to the constant cool white light for 2, 5 and 7 days. Rats kept in normal cyclic light/dark served as control animals. (A) Real-time qRT-PCR summaries of PRL-1 gene expression in constant light-exposed retinas (Day 2,5,7) relative to expression in retina at Day 0 (no constant light) (B) Equal amounts protein (50 µg) of each retinal lysate was subjected to SDS-PAGE followed by immunoblot analysis with anti-PRL-1 antibody. Data are representative of three independent experiments. Trx, thioredoxin; Loading ctrl, non-specific bands on the same blot used as loading control.