Characterization of PTPRG in knockdown and phosphatase-inactive mutant mice and substrate trapping analysis of PTPRG in mammalian cells

Abstract

Receptor tyrosine phosphatase gamma (PTPRG, or RPTPγ) is a mammalian receptor-like tyrosine phosphatase which is highly expressed in the nervous system as well as other tissues. Its function and biochemical characteristics remain largely unknown. We created a knockdown (KD) line of this gene in mouse by retroviral insertion that led to 98-99% reduction of RPTPγ gene expression. The knockdown mice displayed antidepressive-like behaviors in the tail-suspension test, confirming observations by Lamprianou et al. 2006. We investigated this phenotype in detail using multiple behavioral assays. To see if the antidepressive-like phenotype was due to the loss of phosphatase activity, we made a knock-in (KI) mouse in which a mutant, RPTPγ C1060S, replaced the wild type. We showed that human wild type RPTPγ protein, expressed and purified, demonstrated tyrosine phosphatase activity, and that the RPTPγ C1060S mutant was completely inactive. Phenotypic analysis showed that the KI mice also displayed some antidepressive-like phenotype. These results lead to a hypothesis that an RPTPγ inhibitor could be a potential treatment for human depressive disorders. In an effort to identify a natural substrate of RPTPγ for use in an assay for identifying inhibitors, "substrate trapping" mutants (C1060S, or D1028A) were studied in binding assays. Expressed in HEK293 cells, these mutant RPTPγs retained a phosphorylated tyrosine residue, whereas similarly expressed wild type RPTPγ did not. This suggested that wild type RPTPγ might auto-dephosphorylate which was confirmed by an in vitro dephosphorylation experiment. Using truncation and mutagenesis studies, we mapped the auto-dephosphorylation to the Y1307 residue in the D2 domain. This novel discovery provides a potential natural substrate peptide for drug screening assays, and also reveals a potential functional regulatory site for RPTPγ. Additional investigation of RPTPγ activity and regulation may lead to a better understanding of the biochemical underpinnings of human depression.

Figure 1. Generation of knockdown mice.

A. Gene trap disruption of the Ptprg gene locus. Genotyping and RT-PCR/QRT-PCR primers are indicated by arrows, as described in the methods. Primers in Exons 2 and 3 are used for RT-PCR and QRT-PCR to assay the endogenous transcript in wildtype and mutant mice. Primers in the LTR and intron (solid line) are used to genotype the wildtype and mutant mice. B. QRT-PCR analysis of wildtype (+/+) and Ptprg knockdown (−/−) mice. Lung and liver tissues were analyzed from one +/+ and three −/− knockdown mice. Logarithmic scale.

Figure 2. Introduction of a Ptprg point mutation.

A, Targeting strategy used to introduce the point mutation into the Ptprg locus. Homologous recombination (represented by X) between the targeting vector and the Ptprg gene results in the replacement of a G residue with a C resulting in a Cysteine to Serine amino acid change. The LoxP flanked Neo cassette is excised during the chimeras breeding step. B, Southern hybridization indicating proper gene targeting in the embryonic stem cell clones. Lex represents untransfected embryonic stem cell DNA. C. RT-PCR sequence confirmation of the point mutation. Each sample includes a negative control without Reverse Transcriptase, an assay with Ptprg primers alone, and an assay that includes an internal positive control for Actin. D. western blotting. In the top panel, anti-RPTPγ antibody (9E3) detected a band slightly larger than 180 kDa in the wild type and knockin, but not in the knockdown. wt: wild type. KI: knockin. KD: knockdown (gene trap disruption of the Ptprg gene locus). Proteins loaded were 30 micrograms of proteins as membrane preparation from mouse brain without cerebellum. The bottom panel is a loading control showing a band slightly larger than 37 kDa which reacted nonspecifically to anti-RPTPγ (LGI13) and is present in all four samples. The molecular weight (kilodalton) is shown on the left.

Figure 3. Behavioral analysis on the wild type (WT) and knockdown mutant mice (MT).

See details in the text. A and B. Repeated open field (A) and tail suspension tests (B) on the same animals. * - P<0.05, *** - P<0.001, compared to WT on the same day. C. Immobility time in the forced swim test. * - P<0.05, *** - P<0.001, compared to WT for the same measure.

Figure 4. Behavioral analysis on the wild type (WT) and knockin mutant mice (MT).

See details in the text. A. Immobility time in the tail suspension test. Male: WT, n = 8, MT, N = 7; Female: WT, n = 15, MT, n = 10. * - P<0.05, compared to WT of the same sex. B. Immobility time in the forced swim test. * - P<0.05, compared to WT for the same measure.

Figure 5. RPTPγ substrate-trapping mutants in HEK293 cells.

Phosphotyrosine-containing protein of about 190 kDa size “trapped” by RPTPγ C1060S or RPTPγ D1028A is RPTPγ. Recombinant wild type RPTPγ and substrate-trapped mutants were immunoprecipitated from transiently transfected HEK293F lysate with anti-myc mAb. Lanes 1 and 1′were IP product from pCDNA3 transfected cells, 2 and 2′ were from wild type RPTPγ transfected cells, 3 and 3′ were from RPTPγ C1060S transfected cells and 4 and 4′ were from RPTPγ D1060A transfected cells. Lane 1–4 were western reacted with anti-phosphotyrosine 4G10 mAb and lanes 1′-4′ were with anti-myc mAb.

Figure 6. RPTPγ dephosphorylated itself in vitro.

Wild type RPTPγ or RPTPγ C1060S plasmids were transiently transfected into HEK293F and isolated from lysates with anti c-myc/protein G sepharoses. RPTPγ wild type or C1060S on the protein G sepharose was denatured in 8 M urea to linearized protein as substrates for reaction following. The Purified RPTPγ wild type, C1060s sepharose beads was then incubated with recombinant, active purified RPTPγ cyto enzymes in assay buffer with urea at a final concentration of 0.375 M. Lanes 1 and 2 are RPTPγ WT and RPTPγ C1060S on protein G sepharose in a mock reaction without phosphatase in the same buffer, and lanes 1′and 2′ are RPTPγ wild type and RPTPγ C1060S reacted with RPTPγ cytoplasmic region as phosphatase. Samples were subjected to western blot analysis using anti-phosphotyrosine 4G10 mAb. The densitometry of each band was determined to estimate the extent of removal of phosphotyrosine on the protein. The reacted IgG heavy chain bands were used as internal control to ensure equal loading of samples. It is estimated by densitometry that 75% of phosphotyrosine from RPTPC1060S was removed by addition of RPTPγ enzyme.

Figure 7. Identification of tyrosine residue at amino acid 1307 (Y1307) as auto-dephosphorylation site.

There are a total of eleven tyrosine residues that can be found in D2 domain (figure S1). Each of these tyrosine residue was individually mutated to phenylalanine by PCR on the parental plasmid pCDNA3.1 RPTPγ C1060S. We then performed transfection of pcDNA3.1 RPTPγ C1060S (lanes “P” in panels A and B) and the 11 different pcDNA3.1 RPTPγ C1060S/Y->F mutants (lanes1 to 11 in panels A and B) to HEK293 cells. Lane “N” is vector control only. The immunoprecipitates by anti-myc were loaded to SDA-PAGE and subjected to western blot by anti-myc (A) and anti-4G10 (B). The results revealed that all mutants but RPTPγ C1060S/Y1307F (lane 8 in panel B) contained tyrosine-phosphorylation reacted to antibody 4G10 (B). We concluded that Y1307 is the tyrosine residue that was phosphorylated, by activity of an unknown kinase, and then dephosphorylated by active RPTPγ.

Figure 8. Determination of Km of phosphatase RPTPγ on peptide substrate ATQDD(pY)VLEVR (derived from peptide adjacent to Y1307).

Figure here shows Pi released (microM per minute) was plotted against phospho-peptide in a reaction in which Vmax was 2.1 µM/mim/14 nM RPTPγ and Km is 49.6 µM, using one site nonlinear binding algorithm in Graphpad Prism.