Protein tyrosine phosphatase, PTP1B, expression and activity in rat corneal endothelial cells

Abstract

Purpose: The current studies were conducted to determine whether the protein tyrosine phosphatase, PTP1B, plays a role in regulating epidermal growth factor receptor (EGFR) Tyr992 phosphorylation and cell cycle entry in rat corneal endothelial cells.

Methods: Corneas were obtained from male Sprague-Dawley rats. PTP1B mRNA and protein expression were compared in confluent and subconfluent cells by RT-PCR and western blots. Immunocytochemistry was used to determine the subcellular localization of both PTP1B and EGFR following epidermal growth factor (EGF) stimulation. Western blots were used to analyze the time-dependent effect of EGF on phosphorylation of EGFR Tyr992 plus or minus CinnGEL 2Me, an inhibitor of PTP1B activity. The effect of PTP1B inhibition on cell cycle entry was determined by calculating the percent of Ki67-positive cells following EGF treatment.

Results: PTP1B mRNA expression was similar in confluent and subconfluent cells, but PTP1B protein was expressed at 3 fold higher levels in subconfluent cells. Positive staining for PTP1B was localized in vesicular structures below the plasma membrane. EGFR staining was located at cell-cell borders in untreated endothelium, but was mainly cytoplasmic by 15 min after EGF treatment. In control cultures, phosphorylation of EGFR Tyr992 peaked by 5 min following EGF stimulation and rapidly decreased to basal levels by 30 min. In cultures pretreated with CinnGEL 2Me, Tyr992 phosphorylation peaked 2 min following EGF addition and was consistently sustained at a higher level than controls until 60 min after treatment. By 18 h following EGF treatment, cultures pretreated with CinnGEL 2Me exhibited a 1.7 fold increase in the number of Ki67-positive cells compared with control cultures.

Conclusions: Comparison of PTP1B mRNA and protein levels indicates that PTP1B expression is regulated mainly at the protein level and is higher in subconfluent cells. PTP1B was located in vesicles below the plasma membrane. The fact that EGFR is internalized in response to EGF stimulation suggests that it could interact with and be regulated by PTP1B. The ability of PTP1B inhibitor to sustain EGFR Tyr992 phosphorylation and increase the number of Ki67-positive cells indicates that PTP1B plays a role in the negative regulation of EGF-induced signaling and helps suppress cell cycle entry.

Figure 1

Representative results comparing PTP1B mRNA and protein expression in confluent and subconfluent rat corneal endothelial cells. A: RT-PCR for PTP1B. At left are molecular weight markers in 100 bp increments with the brightest band at 600 bp. Lane 1: Control sample containing all reagents, but without cDNA; Lane 2: cDNA extracted from confluent cells; Lane 3: cDNA extracted from subconfluent cells. Position of 443 bp band for PTP1B is indicated. B: RT-PCR for G3PDH control. At left are molecular weight markers; Lane 4: Control sample containing total RNA extracted from confluent cells; Lane 5: cDNA extracted from confluent cells; Lane 6: Control sample containing total RNA extracted from subconfluent cells; Lane 7: cDNA extracted from subconfluent cells. Position of 452 bp G3PDH band is indicated. C: Western blot for PTP1B (48 kDa) showing position of PTP1B in control SW480 cells and relative expression of PTP1B in subconfluent (SC) and confluent (C) rat corneal endothelial cells. D: Same membrane as in (C) reprobed for β-actin to control for protein load.

Figure 2

Representative fluorescent confocal images showing immunostaining for PTP1B and β-catenin in corneal endothelial cells of ex vivo rat corneas. PTP1B (red) localization is compared with that of β-catenin (green), a protein closely associated with cell-cell adhesion junctions. The staining pattern of PTP1B strongly suggests that PTP1B is associated with vesicular structures subjacent to the plasma membrane. The higher magnification image in (B) is included to more clearly demonstrate this localization pattern. The final magnification in A was 2100X and in B was 3120X.

Figure 3

Fluorescent confocal images of EGFR localization in corneal endothelial cells of ex vivo rat corneas following EGF stimulation. In the absence of EGF, EGFR is located mainly at cell borders (arrow), although some cytoplasmic punctate staining is visible (A). After EGF stimulation for 15 min (B), EGFR staining is greatly reduced at cell borders (arrow). After 60 min (C), little-to-no EGFR is visible at cell borders; however, ZO-1 staining of the same tissue (D) indicates that cell borders remain intact. No staining is observed in the endothelium of ex vivo corneas incubated in secondary antibody alone (E). Note that small dots of intense stain can be observed scattered in the cytoplasm of cells in (A-C). The specific nature of this staining is unclear, since all antibodies were centrifuged at high speed prior to dilution to prevent nonspecific antibody deposition. Final magnification: 1000X.

Figure 4

Time-course of EGFR Tyr992 phosphorylation. Subconfluent rat corneal endothelial cells were serum-starved and then treated with 25 ng/ml EGF in Medium 199 plus gentamicin. Samples were taken at various times for western blot analysis. A: Representative western blot showing the time-course of EGFR Tyr992 phosphorylation. Phosphorylated EGFR Tyr992 yields a 175 kDa band. β-Actin (42 kDa) was used for a loading control. B: Graph showing the average level of phosphorylated EGFR Tyr992 in duplicate gels from three separate experiments. β-Actin was used for normalization. Bars represent SEM.

Figure 5

Effect of the PTP1B inhibitor, CinnGEL 2Me, on the time-course of EGFR Tyr992 phosphorylation. Serum-starved, subconfluent rat corneal endothelial cells were pre-incubated for 1 h in Medium 199, gentamicin, and DMSO(-) or supplemented with 1 mM CinnGel 2Me(+). EGF (25 ng/ml) was then added and samples were taken for western blot analysis at various times following EGF addition. Representative western blot in (A) shows phosphorylated Tyr992 at 175 kDa. Densitometric analysis was conducted using β-actin for normalization. The graph in (B) shows average levels of phosphorylated Tyr992. Results are the average of duplicate gels from four separate experiments. Bars represent SEM.

Figure 6

Effect of CinnGEL 2Me inhibition of PTP1B on cell cycle entry. Serum-starved subconfluent rat corneal endothelial cells were preincubated for 1 h in medium containing either DMSO(-) or 1 mM CinnGel 2Me in DMSO(+) followed by addition of 25 ng/ml EGF for 6 or 18 h. A: Cultures were immunostained for Ki67 (green), a marker of proliferating cells, and counterstained with propidium iodide (PI) to reveal all nuclei (red). Magnification: 400X. B: Bar graph shows the average number of Ki67-positive cells at each time-point in duplicate samples from three separate experiments. Bars represent SEM and the asterisk indicates a 1.7 fold increase over control.