Inhibition of PRL2 Upregulates PTEN and Attenuates Tumor Growth in Tp53-deficient Sarcoma and Lymphoma Mouse Models

Abstract

The phosphatases of regenerating liver (PRL) are oncogenic when overexpressed. We previously found that PRL2 deletion increases PTEN, decreases Akt activity, and suppresses tumor development in a partial Pten-deficient mouse model. The current study aims to further establish the mechanism of PTEN regulation by PRL2 and expand the therapeutic potential for PTEN augmentation mediated by PRL2 inhibition in cancers initiated without PTEN alteration. The TP53 gene is the most mutated tumor suppressor in human cancers, and heterozygous or complete deletion of Tp53 in mice leads to the development of sarcomas and thymic lymphomas, respectively. There remains a lack of adequate therapies for the treatment of cancers driven by Tp53 deficiency or mutations. We show that Prl2 deletion leads to PTEN elevation and attenuation of Akt signaling in sarcomas and lymphomas developed in Tp53 deficiency mouse models. This results in increased survival and reduced tumor incidence because of impaired tumor cell proliferation. In addition, inhibition of PRL2 with a small-molecule inhibitor phenocopies the effect of genetic deletion of Prl2 and reduces Tp53 deficiency–induced tumor growth. Taken together, the results further establish PRL2 as a negative regulator of PTEN and highlight the potential of PRL2 inhibition for PTEN augmentation therapy in cancers with wild-type PTEN expression.

Significance:

Prl2 deletion attenuates Tp53 deficiency–induced tumor growth by increasing PTEN and reducing Akt activity. Targeting Tp53-null lymphoma with PRL inhibitors lead to reduced tumor burden, providing a therapeutic approach via PTEN augmentation.

Introduction

Overexpression of the phosphatase of regenerating liver (PRL) phosphatases—PRL1, PRL2, and/or PRL3, occurs in many human cancers and often correlate with metastasis or poor disease prognosis (1–4). Reports suggest that PRL overexpression could potentiate major signaling pathways that are frequently hyperactivated in cancer, such as the PI3K/Akt signaling cascade (1). This highlights the therapeutic potential for PRL inhibition in cancer. Although the cancer-promoting role of PRLs is almost certain, the mechanisms by which they promote tumorigenesis are under active investigation. To study the mechanistic role of PRL2—the most ubiquitously expressed PRL—in cancer and other physiologic processes, mice without Prl2 gene expression were generated. These Prl2-null mice display impaired spermatogenesis, placental insufficiency, and abnormal hematopoietic stem cell self-renewal (5–8). These deficiencies are the result of impaired Akt signaling in the affected tissues caused by increased PTEN protein level in Prl2-null mice. We found that PRL2 dephosphorylates PTEN on Y336, and this dephosphorylation allows binding of the E3-ubiquitin ligase NEDD4 to PTEN for PTEN polyubiquitination and subsequent proteasomal degradation (9). PTEN is a negative regulator of the PI3K/Akt signaling pathway, in which it dephosphorylates phosphatidylinositol (3–5) trisphosphate (PIP3) to phosphatidyl inositol (4, 5) bisphosphate (PIP2) to hinder Akt activation. Thus, Prl2 deletion promotes PTEN stabilization and accumulation, which would lead to attenuated Akt activation. PTEN is a haploinsufficiency tumor suppressor gene, and even subtle decrease in PTEN protein expression increases cancer susceptibility (10–12). In several cancers, PTEN downregulation occur in later stages or during disease recurrence. The mechanism we advanced where PRL2 downregulates PTEN protein is therefore a plausible mechanism by which PRL2 promotes tumorigenesis (9). Accordingly, we showed that Prl2 deletion in the Pten heterozygous mouse model prevents spontaneous tumorigenesis through PTEN upregulation and attenuated Akt signaling pathway (9). This highlighted the potential of PRL2 inhibition as a therapeutic approach for the treatment of PTEN deficiency–induced cancers. However, based on cBioPortal cancer genomic database, more than 90% of all cancers develop with wild-type (WT) PTEN expression, which emphasizes the need to develop treatments for such cancers. Whether loss of PRL2 can still prevent tumor growth in cancers that develop independently of PTEN downregulation through PTEN augmentation is still unknown.

TP53 is an important tumor suppressor that regulates important cellular processes such as DNA repair, cell cycle, apoptosis, and metabolism (13–15). Indeed, TP53 is the most mutated or downregulated gene in human cancers. Certain cancers, such as osteosarcomas and soft-tissue sarcomas, have a high rate of TP53 mutation or downregulation, and novel approaches to treat cancers with TP53 abnormalities would play a critical role to improve patient outcome (16, 17). Proposed approaches for the treatment of cancers with P53 mutations involve the use of small molecules for the restoration of WT P53 function, P53 stabilization by inhibiting its interaction with negative regulators (such as MDM2 and MDMX), and activation of P53 family members such as P73 (18–20). Although these approaches show promise in laboratory research, clinical trials exploiting these strategies show adverse effect due to off-target effect, that hinder the progress of such therapies (18). Thus, developing novel approaches for the treatment of cancers with P53 downregulation or mutations remains a high priority. Consistent with human patient studies, partial or complete deletion of Tp53 in mice leads to the spontaneous tumorigenesis, with sarcomas and thymic lymphomas being the major cancers observed, respectively (21–23). Thus, providing adequate models to evaluate novel approaches for the treatment of cancers driven by P53 downregulation is highly needed. We know that Prl2 deletion in WT mice can elevate PTEN (5, 7, 24), but it remains unclear whether this can inhibit cancers that develop independently of PTEN downregulation. We hypothesized that PTEN augmentation induced by PRL2 inhibition in cancers with reduced P53 expression could serve as a promising therapeutic approach. Indeed, we found patients with sarcoma with TP53 mutations have higher PRL2 mRNA expression and lower PTEN protein level, supporting our hypothesis that targeting PRL2 to enhance PTEN level could serve as a therapeutic approach for cancer treatment in TP53-driven cancer. By using the well-established P53 deficiency mouse models (heterozygous and homozygous Tp53 deletion), we sought to not only further establish the mechanism of PTEN regulation by PRL2 but also assess the therapeutic potential of PRL2 inhibition for PTEN augmentation therapy in the treatment of cancers that develop without PTEN alteration.

Through bioinformatics analysis, we show that PRL2 mRNA expression negatively correlates with survival in patients with sarcoma, and positively correlates with cell proliferation markers in these patients. In addition, we found that only a small portion of patients with sarcoma have PTEN mutation, suggesting such cancers are suitable for PTEN augmentation therapy. We show that in both Tp53 heterozygous and Tp53-null mouse models, deletion of Prl2 significantly improves survival and delays tumor growth. Mechanistically, we confirmed increased PTEN protein level and reduced Akt activation in sarcomas and lymphomas obtained from Prl2-null mice. The reduced growth was determined to be a result of reduced cell proliferation, but interestingly, we observed no change in apoptosis in both models. In addition, we show that pharmacologic inhibition of PRL2 with a small-molecule PRL inhibitor—Cmpd-43 (24), significantly hinders tumor growth in Tp53-null mice. Our findings: (i) further substantiate the mechanism of PTEN downregulation by PRL2 in vivo, (ii) demonstrate the potential of PRL2 inhibition for PTEN augmentation therapy in cancers that develop with WT PTEN expression and P53 mutations, and (iii) establish the efficacy of small-molecule inhibition of PRLs in an in vivo mouse tumor model for the first time. Together, the results suggest that pharmacologic inhibition of PRL2 could serve as a novel therapeutic strategy to increase PTEN protein levels, thereby obliterating P53 deficiency–induced malignancies.

Materials and Methods

Reagents and Antibodies

Tamoxifen (CAS 10540-29-1) was obtained from as a powder from Cayman Chemicals (catalog no. 13258). The following antibodies used for FACS analysis were obtained from BioLegend: CD4-488 (BioLegend, catalog no. 100423, RRID: AB_389302), CD8a-PE (BioLegend, catalog no. 100708, RRID: AB_312747), CD45-PeCy7 (BioLegend, catalog no. 103114, RRID: AB_312979), CD25-BV605 (BioLegend, catalog no. 102035, RRID: AB_11126977), CD44-647 (BioLegend, catalog no. 103039, RRID: AB_10895752). Primary antibodies used for Western blotting were: PTEN (Cell Signaling Technology, catalog no. 9188, RRID: AB_2253290), AKT (Cell Signaling Technology, catalog no. 9272, RRID: AB_329827), p-AKT pS473 (Cell Signaling Technology, catalog no. 4060, RRID:AB_2315049), mTOR (Cell Signaling Technology, catalog no. 2972, RRID: AB_330978), p-mTOR pS2448 (Cell Signaling Technology, catalog no. 2971, RRID:AB_330970), β-actin (Santa Cruz Biotechnology, catalog no. sc-47778, RRID: AB_626632), PARP (Cell Signaling Technology, catalog no. 9542, RRID: AB_2160739), Caspase 3 (Cell Signaling Technology, catalog no. 14220, RRID: AB_2798429), p-BAD pS136 (Cell Signaling Technology, catalog no. 4366, RRID: AB_10547878), PCNA (Cell Signaling Technology, catalog no. 2586, RRID:AB_2160343), FoxO1 (Cell Signaling Technology, catalog no. 2880, RRID:AB_2106495), p-FoxO1/3a/4 pT24/32/28 (Cell Signaling Technology, catalog no. 2599, RRID:AB_2106814), GSK3-alpha/beta (Cell Signaling Technology, catalog no. 5676, RRID:AB_10547140), p-GSK3alpha/beta pS21/9 (Cell Signaling Technology, catalog no. 9331, RRID:AB_329830), ERK1/2 (Cell Signaling Technology, catalog no. 4696, RRID:AB_390780), p-ERK1/2 pT202/pY204 (Cell Signaling Technology, catalog no. 4370, RRID:AB_2315112), BAX (Cell Signaling Technology, catalog no. 2772, RRID:AB_10695870), p27 (Santa Cruz Biotechnology, catalog no. sc-528, RRID:AB_632129), p21 (Santa Cruz Biotechnology, catalog no. sc-6246, RRID:AB_628073), cyclin D1 (Santa Cruz Biotechnology, catalog no. sc-20044, RRID: AB_627346), 4E-BP1 (Santa Cruz Biotechnology, catalog no. sc-9977, RRID: AB_626621), p-4E-BP1/2/3 (Santa Cruz Biotechnology, catalog no. sc-271947, RRID:AB_10709575), PUMA (Thermo Fisher Scientific, catalog no. PA5-20007, RRID: AB_11153580), p53 (Leica Biosystems, catalog no. NCL-L-p53-CM5p, RRID: AB_2895247) Secondary antibodies used were: Anti-rabbit IgG, horseradish peroxidase (HRP)-linked (Cell Signaling Technology, catalog no. 7074, RRID: AB_2099233), and Anti-mouse IgG, HRP-linked (Cell Signaling Technology, catalog no. 7076, RRID: AB_330924).

Transgenic Mice

This study utilizes commercially available C57BL/6J Rosa26-CreER^T2 (Strain# 008463, RRID: IMSR_JAX:008463) and C57BL/6J Tp53^flox (Strain# 008462, RRID: IMSR_JAX:008462) mice from Jackson Labs. A Prl2^flox mouse was generated within C57BL/6J strain mice in-house using CRISPR/Cas9 recombination as recently described by Carlock and colleagues (25). Comparable numbers of male and female mice were enrolled in each study. Animals were housed and cared for in accordance with Purdue University Institutional Animal Care and Use Committee guidelines and all experiments conducted followed our approved protocol (protocol number 1511001324).

Genotyping

DNA was isolated from 1 mm tail snips taken from mouse pups during weaning. Tails snips were boiled at 95°C for 20 minutes in a solution of 25 mmol/L NaOH and 200 µmol/L ethylenediaminetetraacetic acid (EDTA) buffered to pH 12. The solution was cooled to room temperature, and then neutralized by the addition of equal volume 40 mmol/L Tris-HCl, pH 5. The following primer sets were used to identify transgenes: Rosa26 WT forward: CTG GCT TCT GAG GAC CG, reverse: CCG AAA ATC TGT GGG AAG TC; Rosa26-CreER^T2 forward: CGT GAT CTG CAA CTC CAG TC, reverse: AGG CAA ATT TTG GTG TAC GG; Tp53^flox forward: GGT TAA ACC CAG CTT GAC CA, reverse: GGA GGC AGA GAC AGT TGG AG; Prl2^flox forward: CAC ACA CTT AAG TAA GTA CCT GGT TGG, reverse: CCA ATC ATC TAC TAT CTG ATT AGG G. PCR reaction mixtures were made using GoTaq Flexi DNA Polymerase (Promega, catalog no. M8295), and cycling protocols were performed according to Jackson Labs recommendation. The in-house Prl2^flox cycling conditions were as follows: 95°C-5 minutes, (95°C-30 seconds, 57°C-45 seconds, 72°C-1 minute) × 34, 72°C-5 minute, 4°C-hold.

Tamoxifen Treatment

Tamoxifen powder was dissolved in corn oil (Sigma-Aldrich, catalog no. C8267) at a stock concentration of 20 mg/mL by rocking overnight in the dark at 37°C. This stock solution was then filtered through a 70 µm mesh filter (Falcon, catalog no. 352350) and stored at 4°C until necessary. Animals were given intraperitoneal injections of tamoxifen at the concentration indicated for each protocol using sterile 25G needles (BD, catalog no. 305125).

Cmpd-43 Treatment

A 3 mg/mL solution of Cmpd-43 (24) in PBS with 10% DMSO was obtained. On week 9 after tamoxifen treatment, a volume (of the 3 mg/mL Cmpd-43 solution) equal to 10 times the mouse's mass was injected intraperitoneally daily to achieve the 30 mg/kg daily dosage of Cmpd-43. This treatment was done daily for 3 weeks.

Sample Preparation for Western Blot Analyses

Tissue samples were homogenized on dry ice using a lysis buffer made of 50 mmol/L Tris-HCl, 150 mmol/L NaCl, 10% glycerol, and 0.5% Triton X-100 containing a cocktail of protease inhibitors (Roche, catalog no. 04693132001) and phosphatase inhibitor cocktail (bimake.com, catalog no. B15001). Protein concentration was quantified by Bradford assay using the Coomassie Plus Protein Assay kit (Thermo Fisher Scientific, catalog no. 23236). Samples were electrophoresed on 7.5%–15% SDS-PAGE gels using the Bio-Rad Mini-PROTEAN gel casting system. Samples were transferred to nitrocellulose membranes, blocked with 5% w/v BSA in Tris-buffered saline with Tween 20 (TBST) and then incubated with primary antibodies at a 1:1,000–1:2,000 dilution in TBST overnight at 4°C. Membranes were washed and then incubated with HRP-conjugated secondary antibodies at a 1:3,000–1:5,000 dilution in TBST overnight at 4°C. Membranes were washed and then visualized using SuperSignal West Pico (Thermo Fisher Scientific, catalog no. 34580), or SuperSignal West Femto (Thermo Fisher Scientific, catalog no. 34095) Chemiluminescent Substrate as necessary.

Tissue Sectioning, Staining, and Microscopy

Fixed tumors were stored in 75% ethanol then submitted to the Purdue University histology laboratory where they were embedded in paraffin, sectioned, and stained for proliferating cell nuclear antigen (PCNA; Abcam, catalog no. ab29, RRID: AB_303394) and cleaved caspase 3 (Cell Signaling Technology, catalog no. 9661S) using standard protocol.

Sample Preparation for FACS Analysis

To perform FACS analysis, thymic lymphoma samples were first reduced to single-cell suspensions by compressing the tissue through a 70 µm filter (Falcon, catalog no. 352350) into cold 1x PBS supplemented with 5 mmol/L EDTA using the plunger of a syringe (HSW, catalog no. 4010-200V0). Sample were strained through a second filter and then resuspended in cold blocking buffer made of 5% BSA, 0.5 mmol/L EDTA, and a 1:1,000 dilution of CD16/32. Whole blood collected via cardiac puncture was prepared by lysing red blood cells in a buffer made with 155 mmol/L NH₄Cl, 12 mmol/L NaHCO₃, and 0.1 mmol/L EDTA for 1 minute, washing in 1x PBS, and then blocking buffer. Samples were then incubated with fluorophore conjugated antibodies. Samples were incubated with antibodies at 1:600 dilution in blocking buffer on ice for 2 hours. The following antibody combinations were used for population analysis: CD45, CD4, CD8, CD44, CD25. The gating strategy is shown in Supplementary Fig. S1. After staining samples were washed and resuspended in 5% BSA, 0.5 mmol/L EDTA solution, then added to 5 mL polystyrene tube for analysis (Falcon, catalog no. 352235). Sample data were acquired using a BD LSRFortessa Cell Analyzer, and results were analyzed using FlowJo (RRID: SCR_008520).

The Cancer Genome Atlas Analysis

The Cancer Genome Atlas (TCGA; RRID: SCR_003193) level 3 RNAseqV2 gene expression, reverse phase protein array (RPPA), and clinical data of primary patient clinical samples from Sarcoma (SARC, n = 261) were downloaded from the Broad Institute's Firehose (https://gdac.broadinstitute.org/). Patients were separated into two subgroups (Tp53 WT and TP53 mutated) based on Tp53 mutation status or (low PRL2 and high PRL2) based on their PRL2 mRNA expression levels using a cutoff on the z scores (|z| ≥ 1) for PRL2 mRNA, PTEN, and p53 RPPA analysis and Kaplan–Meier survival analysis. The correlation coefficients between PRL2 mRNA and cell cycle–related genes (Cyclin E1, FoxM1, PCNA, Cyclin B1) were measured by Pearson correlation coefficients analysis. The number of each analysis was also affected by the data availability.

Data Availability

The data generated in this study are available upon request from the corresponding author.

Statistical Analysis

Western blot data were quantified for statistical analysis using ImageJ software (RRID: SCR_003070). Protein abundance was determined as black pixel intensity after subtracting background. Protein concentration was determined relative to loading control concentration. Phosphorylated protein concentration was determined by concentration relative to total protein. Protein concentration datasets were analyzed by the Student t test. Tumor weights were also compared using the Student t test. Survival analysis was performed using the Kaplan–Meier log-rank test. Error bars in figures indicate SEM. P values less than 0.05 were not considered statistically significant. Analysis was performed on all datasets using Microsoft Excel (RRID: SCR_016137) to identify outliers which were then excluded from all figures presented in this article. Outliers were defined as datapoints below the lower limit = 1.5*IQR − Q1 or above the upper limit = 1.5*IQR + Q3. IQR = interquartile range, Q1 = 1st quartile, Q3 = 3rd quartile.

Study Approval

All procedures performed on mice were approved by the Purdue Institutional Animal Care and Use Committee.

Results

PRL2 Expression Negatively Correlates with Survival and PTEN Protein Expression in Human Patients with Sarcoma

TP53 is the most mutated or downregulated tumor suppressor gene in human cancers, thus the development of therapeutic strategies for cancers with TP53 mutations or downregulations remains critical for cancer treatment. In this work, we propose that PRL2 inhibition for PTEN augmentation can impede tumor growth in Tp53 deficiency–driven cancers. Tp53 deficiency in mouse models primarily leads to sarcomas and lymphomas (23, 26–28). We therefore decided to analyze sarcoma and lymphoma patient samples from TCGA database (https://gdac.broadinstitute.org/), but only sarcoma patient samples were available. We first examined the PRL2 mRNA expression in patients with sarcoma with or without TP53 mutations. We found that patients with TP53 mutation have higher PRL2 mRNA expression (Fig. 1A). Meanwhile, these patients also showed reduced PTEN protein level (Fig. 1B). These data support our hypothesis that targeting PRL2 to enhance PTEN level could serve as a therapeutic approach for cancer treatment in TP53 mutant–driven cancer. We found that PRL2 mRNA expression negatively correlates with PTEN protein levels and has no correlation with P53 protein. This is consistent with our previous findings that PRL2 promotes PTEN degradation (9). We also found that high PRL2 expression significantly reduced survival in these patients (Fig. 1C–E). Partial reduction in PTEN expression is sufficient to increase cancer risk or severity (10, 11), thus PTEN degradation resulting from PRL2 overexpression in these cancers could explain the worse patient outcome. Moreover, we found that high PRL2 expression in these tumors positively correlates with the expression of cell cycle–related genes that would promote tumor proliferation (Fig. 1F). To use PRL2 inhibition for PTEN augmentation therapy, the targeted cancers need to express WT PTEN. Increasing the expression of an inactive mutant of PTEN would not provide any antitumor benefits. We show that only a small portion of sarcomas have PTEN mutations (Fig. 1G), meanwhile the incidence of P53 mutations in sarcomas was reported to vary between 10% and 80% (16). Taken together, the results from this study and the bioinformatics analysis supports targeting PRL2 therapeutically for the treatment of P53-deficient sarcomas with WT PTEN expression, for PTEN augmentation cancer therapy.

FIGURE 1.

PRL2 expression inversely correlates with PTEN and survival in human patients with sarcoma. A, Scatter plot of PRL2 mRNA data showing the increased expression of PRL2 mRNA in patients with TCGA SARC with TP53 mutations. Patient samples were divided into TP53 WT group (n = 159) and TP53 mutated group (n = 82) based on TP53 mutation status, and PRL2 mRNA level of the two groups were plotted. B, Scatter plot of RPPA data of PTEN protein showing the decreased expression of PTEN protein in patients with SARC with TP53 mutations. Patient samples were divided into TP53 WT group (n = 122) and TP53 mutated group (n = 67) based on TP53 mutation status, and PTEN protein level of the two groups were plotted. C, Scatter plot of RPPA data showing the reverse correlation between the mRNA levels of PRL2 and protein levels of PTEN in patients with TCGA SARC. Patient samples were divided into low PRL2 (n = 35) and high PRL2 (n = 33) groups based on the PRL2 mRNA level, and then the PTEN protein level of the two groups were plotted. D, Scatter plot of RPPA data showing no correlation between the mRNA levels of PRL2 and protein levels of p53 in patients with TCGA SARC. Patient samples were divided into low PRL2 (n = 35) and high PRL2 (n = 33) groups based on the PRL2 mRNA level, and then the p53 protein level of the two groups were plotted. E, Patients with high PRL2 mRNA level (n = 43) have significantly reduced probability of survival than that with low PRL2 mRNA level (n = 39) in SARC by Kaplan–Meier survival analysis. F, Correlation between PRL2 mRNA and cell cycle–related gene expression in TCGA sarcoma patient samples (n = 223). G, PTEN mutation status in TCGA sarcoma patient samples (n = 253). For statistics, *, P < 0.05. ***, P < 0.001.

PRL2 Deletion Significantly Improves Survival in Tp53 Deficiency Mouse Models

We previously showed that constitutive Prl2 deletion in the context of Pten heterozygosity increased PTEN protein levels in both normal tissues and tumors that resulted from the PTEN deficiency (9). In this study, we wanted to investigate whether PTEN augmentation that results from PRL2 inhibition can hinder the growth of tumors that develop with WT PTEN expression. To achieve this, we decided to use Tp53 deficiency models for spontaneous tumor development. We first generated the constitutive Tp53 heterozygous (Tp53-H) and Tp53 heterozygous; Prl2-knockout (Tp53-H; Prl2-KO) mice to determine the impact of PRL2 loss in the context of partial loss of P53 expression (Fig. 2A). The genotype of these mice was confirmed by PCR (Fig. 2B). As expected, Tp53-H; PRL2-KO mice had significantly improved survival compared with the Tp53-H control (50% survival for Tp53-H approximately 21.5 weeks vs. Tp53-H; PRL2-KO undetermined within timeframe of study, P = 0.0215; Fig. 2C). We also generated constitutive Tp53-knockout (Tp53-KO) and Tp53-knockout; Prl2-knockout (Tp53-KO; Prl2-KO) mice to assess how PRL2 affects tumorigenesis in the Tp53-KO background (Fig. 2A and B). Again, Tp53-KO; Prl2-KO mice also displayed improved survival compared the Tp53-KO control mice (5.2 vs. 8.7 weeks, P = 0.0437; Fig. 2D). Sarcomas were the most common tumors observed in Tp53-H background mice as expected from previous reports (23, 29), and deletion of Prl2 significantly reduced the incidence of sarcomas in these mice (Fig. 2E). These data further support our hypothesis that deletion of Prl2 leads to improved tumor-free survival in constitutive Tp53 deficiency–driven tumors.

FIGURE 2.

Prl2 deletion enhances survival in constitutive Tp53 deficiency mouse models. A, Mating strategy for the generation of constitutive Tp53-deficient mice with or without constitutive Prl2 deletion. B, PCR genotyping to identify different genotypes in the study. Kaplan–Meier survival plot for overall survival in constitutive Tp53-H (C) and Tp53-KO (D) backgrounds with or without constitutive Prl2 deletion. E, Bar graphs representing percent tumor incidence in Tp53-H background with or without Prl2 deletion. P values are indicated in figures and a P value <0.05 is considered statistically significant. WT n = 44, Prl2-KO n = 31, Tp53-H n = 118, Tp53-H; Prl2-KO n = 108.

Constitutive Tp53-KO mice show early tumor development with a median survival of approximately 4 months, with low incidence of birth defects (23, 26, 30). In addition to the constitutive Tp53-KO mice, we also generated tamoxifen-inducible Tp53-knockout (Tp53^−/−) and Tp53-knockout; Prl2-knockout (Tp53^−/−; Prl2^−/−) mice for controlled induction of tumorigenesis. This would circumvent potential developmental defects and pre-adulthood tumorigenesis and allow us to contrast the constitutive and inducible models. Floxed-Tp53 mice (Tp53^flox/flox) mice were obtained commercially, and we generated our own floxed-Prl2 (Prl2^flox/flox) mice as described previously (25). We first bred the Prl2^flox/flox mice with the tamoxifen-inducible ERT2-Cre^+/+ mice to generate Prl2^flox/flox; ERT2-Cre^+/+ mice (Supplementary Fig. S2A). Our Prl2^flox/flox; ERT2-Cre^+/+ mice were then bred with control Tp53^flox/flox; ERT2-Cre^+/+ mice to generate our experimental Prl2^flox/flox; Tp53^flox/flox; ERT2-Cre^+/+ mice (Supplementary Fig. S2A), for which the genotyping was confirmed by PCR (Fig. 3A). To bypass major postnatal development, 8 to 10 weeks old adult mice were given tamoxifen intraperitoneally for 5 consecutive days to induce Tp53 and/or Prl2 whole-body deletion to generate the control Tp53^−/− and experimental Tp53^−/−; Prl2^−/− mice, which were monitored for tumor development and survival (Supplementary Fig. S2B). Following tamoxifen treatment, we found that the experimental Tp53^−/−; Prl2^−/− mice had a significant increase in survival compared with the control Tp53^−/− mice (20.9 vs. 16.6 weeks, P < 0.0001; Fig. 3B). On the other hand, the primary tumor type observed after tamoxifen treatment in Tp53^−/− mice were thymic lymphomas (98% incidence), and deletion of Prl2 in this model drastically reduced the incidence of thymic lymphomas (∼62%; Fig. 3C). Prior to tamoxifen treatment, these mice had normal PRL2 and P53 expression in the thymus, and deletion of Prl2 and/or Tp53 in the thymus was confirmed after tamoxifen treatment (Supplementary Fig. S2C).

FIGURE 3.

Prl2 deletion enhances survival and hinders tumor growth rate in tamoxifen-inducible Tp53^−/− mouse but does not affect tumor initiation. A, PCR genotyping to identify different genotypes in the study. B, Kaplan–Meier survival plot for thymic lymphoma-free survival in tamoxifen-inducible Tp53^−/− n = 43 and inducible Tp53^−/−; Prl2^−/− n = 33. C, Bar graph representing percent tumor incidence in inducible Tp53^−/− background with or without inducible Prl2 deletion. Dot plots representing: The mass of thymus/thymic lymphoma over time in inducible Tp53^−/− n = 9–20 per timepoint and inducible Tp53^−/−; Prl2^−/− n = 8–14 per timepoint (D). E, Thymocyte population ratios 9 weeks after inducing Tp53 deletion with or without Prl2 deletion. F, Thymocyte population ratios 12 weeks after inducing Tp53 deletion with or without Prl2 deletion. DP: Double positive (CD4+ CD8+). DN: Double negative (CD4− CD8−). log-rank test performed for Kaplan–Meier survival curves and Student t test performed for all other analysis. Data are shown as mean ± SEM. *, P < 0.05; **, P < 0.01; ***, P < 0.001.

Loss of P53 in mice leads to dysregulation in T-cell development and promotes the selection and expansion of lymphoblasts which results in significant increase in thymic mass (21, 29). Because we have temporal control on gene deletion, we assessed the growth rate of thymic lymphomas in our model. Our results show that simultaneous inducible deletion of Prl2 and Tp53 significantly hinders thymic lymphoma growth rate (Tp53^−/− vs. Tp53^−/−; Prl2^−/− at 12 weeks 0.419 g vs. 0.098 g, P = 0.0017; at 16 weeks 1.083 g vs. 0.594 g, P = 0.0085; Fig. 3D). Interestingly, there was a modest albeit insignificant reduction in tumor size in the Tp53^−/−; Prl2^−/− mice at death. The tumor growth is limited to the size of the thoracic cavity, and thus both groups are expected to have comparable tumor size when death is because of the thymic lymphoma. Analysis of thymic lymphomas using flow cytometry showed thymic lymphoma onset in some mice as early as 9 weeks after tamoxifen treatment, albeit normal thymus size and morphology at this time for both groups. At 12-weeks after tamoxifen treatment, we observed depletion of the DN2 and DN3 thymocyte populations along with thymic enlargement (Fig. 3E and F), in agreement with previous models of thymic lymphomagenesis (31, 32). DN2 and DN3 thymocyte depletion was observed to a lesser extent at 12 weeks in the Tp53^−/−; Prl2^−/− mice, which further highlights hindered thymic lymphoma development upon Prl2 deletion. Thymocyte population distribution in WT mice is shown in Supplementary Fig. S2D. Taken together, these results support that loss of PRL2 significantly hinders tumorigenesis in both constitutive and inducible P53 deficiency models. In addition, we determined the tumor growth rate in the inducible Tp53^−/− tumor model, which will allow us to assess the therapeutic efficiency of pharmacologic PRL2 inhibition.

Deletion of PRL2 in Tp53 Deficiency Models Inhibits Tumor Cell Proliferation but not Apoptosis

Our results suggest that tumor growth rate is significantly reduced because of Prl2 deletion in our Tp53 deficiency models. To determine whether this is a result of changes in tumor cell proliferation and/or apoptosis, we assessed the expression of proliferation and apoptotic markers in the thymic lymphomas of both control and experimental mice 12 weeks after tamoxifen treatment. We used PCNA as a proliferation marker, and through both immunoblotting and IHC, we observed a significant 43% reduction in thymic lymphoma proliferation in Tp53^−/−; Prl2^−/− mice compared with Tp53^−/− controls (Fig. 4A–C). This observation is consistent with our previous report that Prl2 deletion reduces proliferation in the PTEN heterozygous tumor model (9). The tumor suppressor P53 is a major regulator of apoptosis, and we previously showed that deletion of Prl2 in the PTEN heterozygous model increases tumor cell apoptosis in a P53-dependent manner (9). Because of complete loss of Tp53 in the tamoxifen-inducible Tp53^−/− model, we anticipated that Prl2 deletion would have no effect on apoptosis. Immunoblotting for cleaved-PARP and cleaved-caspase 3 as markers for apoptosis confirmed that there is no difference in apoptosis between the Tp53^−/−; Prl2^−/− and Tp53^−/− mice (Fig. 4A and B). Interestingly, similar results were observed when comparing sarcomas obtained from Tp53-H; Prl2-KO with those derived from Tp53-H mice, where we observed a significant 34% reduction in tumor proliferation and no change in apoptosis (Supplementary Fig. S3A–S3C). Our results suggest that loss of PRL2 in Tp53 deficiency–derived tumors inhibits tumor growth through downregulation of tumor cell proliferation, with no effect on apoptosis.

FIGURE 4.

Prl2 deletion inhibits tumor cell proliferation but not apoptosis in tamoxifen-inducible Tp53^−/− mice. A, Representative Western blot analysis for proliferation marker PCNA and apoptosis markers c-PARP and c-caspase 3 from thymic lymphoma samples obtained 12 weeks after tamoxifen treatment. B, Dot plots for quantification of proliferation and apoptosis markers relative to actin and total protein control respectively for Western blots in A. Data are shown as mean ± SEM. Tp53^−/− n = 12–20, Tp53^−/−; Prl2^−/− n = 11–14. C, Representative IHC stain using PCNA marker to compare proliferation in inducible Tp53^−/− and Tp53^−/−; Prl2^−/− thymic lymphomas. *, P < 0.05.

Loss of PRL2 in TP53-deficient Tumors Attenuates Tumor Growth through PTEN Augmentation and Inhibition of Akt Signaling

Our Tp53^−/−; Prl2^−/− and Tp53^−/− mice have WT PTEN gene, and we showed that inducible Prl2 deletion in the thymus and other tissues also leads to higher PTEN protein level in these tissues (25). PTEN is a major negative regulator of the PI3K/Akt signaling pathway and we thus hypothesized that increased PTEN protein level in developing thymic lymphomas inhibits Akt signaling to hinder the growth of these tumors as observed. As expected, by performing immunoblotting, we observed significantly higher PTEN protein levels in thymic lymphomas derived from Tp53^−/−; Prl2^−/− mice compared with tumors derived from Tp53^−/− mice (119% increase, P = 0.0014; Fig. 5A and B). Prl2 deletion also resulted in significantly higher PTEN protein in sarcomas from the Tp53-H model (68.4% increase, P = 0.0434; Supplementary Fig. S4A and S4B). Accordingly, the increased PTEN protein level observed in tumors upon Prl2 deletion was coupled with suppressed Akt activation in both the inducible Tp53^−/− background (57.8% reduction, P = 0.0015; Fig. 5A and B) and the Tp53-H background by both Western blot and IHC staining (51.8% reduction, P = 0.0086; Supplementary Fig. S4A–S4D).

FIGURE 5.

Prl2 deletion upregulates PTEN to inhibit Akt signaling and hinder tumor cell proliferation in inducible Tp53^−/− mice. A, Representative Western blot analysis from inducible Tp53^−/− and inducible Tp53^−/−; Prl2^−/−–derived thymic lymphomas to determine expression of PTEN, PUMA, BAX, p21, cyclin D1, p27, as well as the expression and activation of Akt and Akt targets. The actin loading control shown is the same as that in Fig. 4A. B, Dot plot for quantification of PTEN and Akt activation. C, Dot plot for quantification of Western blot analysis for the activation of Akt targets mTOR and GSK3β. D, Dot plot for quantification of the expression of Tp53 targets PUMA, BAX, p21, and cyclin D1. E, Dot plot for quantification of FoxOs activation and p27 expression. Data shown as mean ± SEM for samples obtained 12 weeks after tamoxifen treatment, Tp53^−/− n = 10–20, Tp53^−/−; Prl2^−/− n = 10–14; *, P < 0.05; ***, P < 0.001; ****, P < 0.0001.

Akt activation regulates several downstream signaling processes through various Akt substrates that can affect tumor cell proliferation. We then performed further analysis to corroborate the reduced Akt activation we observed upon Prl2 deletion with reduced Akt substrate phosphorylation that would support reduced proliferation. Cell growth and protein synthesis are cellular processes regulated by mTOR, which can be activated by Akt-mediated phosphorylation (33). As expected, we observed a significant reduction in mTORC1 phosphorylation in Tp53^−/−; Prl2^−/− mice, which is consistent with lower Akt activation (39% reduction, P = 0.0151; Fig. 5A and C). In addition, Akt inhibits glycogen synthase kinase 3β (GSK3β) through S9 phosphorylation which prevents accessibility to GSK3β substrates to promote cell proliferation and metabolism (33). The PRL2-deficient thymic lymphomas had reduced pS9 in GSK3β, which supports reduced proliferation observed in tumors derived from these mice (61% reduction, P = 0.0271; Fig. 5A and C). The ERK signaling pathway is known to regulate tumor growth and PRL2 has been shown to modulate this pathway in vitro (2, 8, 34), but not in vivo (9). We therefore assessed ERK activation, and no significant difference was observed between experimental and control groups (Supplementary Fig. S5A and S5B). Sarcomas derived from Tp53-H; Prl2-KO mice also showed reduced mTORC1 and GSK3β phosphorylation compared with Tp53-het–derived tumors, in accordance with reduced tumor proliferation observed (29.2% reduction, P = 0.0473, and 44.7% reduction, P = 0.0465, respectively Supplementary Fig. S4A–S4C).

The tumor suppressor P53 is a major regulator of cell apoptosis and we previously showed that Prl2 deletion in Pten heterozygous mice promotes P53-dependent apoptosis. In our models of Tp53 deficiency–induced tumorigenesis, we did not observe any difference in apoptosis upon Prl2 deletion. Nonetheless, we assessed the protein expression level of some P53-regulated genes in thymic lymphomas derived from Tp53^−/−; Prl2^−/− and Tp53^−/− mice, such as puma, BAX, p21. Our results confirm that Prl2 deletion in this model does not affect the expression of these proteins, hence apoptosis is not affected upon Prl2 deletion in this model (Fig. 5A and D). The forkhead box O (FoxO) proteins are transcription factors that regulate several cellular processes such as apoptosis, cell cycle, and stress response. FoxOs are direct Akt substrates, and their phosphorylation by Akt enables 14-3-3 binding to FOXO (33). This interaction prevents nuclear localization of FoxO, and thus inhibits its activity as a transcription factor. Thymic lymphomas from Tp53^−/−; Prl2^−/− mice had significantly lower FoxO-1 and -3 phosphorylation compared with the control mice (73.6% reduction, P = 00014, and 86.1% reduction, P = 0034; Fig. 5A and E). We showed that there are no differences in apoptosis and apoptotic signaling that could also be regulated by FoxOs upon Prl2 deletion in this model. In addition to apoptosis, FoxOs also regulate genes involved in the cell cycle, and this regulatory function could impact cell proliferation. P27 and cyclin D1 are target genes for FoxO that are known to regulate cell-cycle progression. No changes in cyclin D1 were observed, but we saw a significant increase in p27 protein levels in Tp53^−/−; Prl2^−/− mice compared with Tp53^−/− mice (113.1% increase, P = 0.0136; Fig. 5A and E). P27 promotes cell-cycle arrest, and this would be consistent with reduced proliferation observed in thymic lymphomas upon Prl2 deletion. Taken together, our results show that loss of PRL2 in thymic lymphomas derived from Tp53^−/− mice leads to PTEN augmentation and attenuation of Akt and downstream signaling. This inhibition of Akt signaling results in reduced tumor cell proliferation and reduced tumor burden in the mice, while the apoptotic pathway remains unaffected.

Pharmacological Inhibition of PRL2 in Tamoxifen-inducible Tp53 Deletion Model Significantly Inhibits Thymic Lymphoma Growth

Through a genetic approach, we successfully showed that deletion of Prl2 inhibits thymic lymphoma growth in Tp53^−/− background mice. In Figure 3D, we show that the thymus mass is similar between both experimental and control groups 9 weeks after tamoxifen treatment, but the tumors are significantly smaller in Tp53^−/−; Prl2^−/− mice at 12 weeks after tamoxifen. Tumor growth is significant between week 9 and week 12 for the Tp53^−/− mice, and this provides an adequate window to assess pharmacologic inhibition of PRL2. We previously reported Cmpd-43 as a PRL inhibitor that disrupts PRL trimerization, which is essential for PRL function (24). Cmpd-43 efficiently inhibits tumor growth in a xenograft model (3, 24) as well as proliferation of leukemia cell lines in vitro and in vivo (35–37). To evaluate the therapeutic potential of PRL2 inhibition in Tp53-deficient models, we therefore assessed the ability of Cmpd-43 to suppress tumorigenesis in our inducible Tp53^−/− mouse model. This will be the first time this compound is used in a genetically engineered cancer model. Thymic lymphoma development was induced as previously described in Tp53^−/− mice, and 9 weeks after tamoxifen treatment, the mice were treated with either 30 mg/kg Cmpd-43 or vehicle control daily for 3 weeks, after which the mice were euthanized, and the tumors were analyzed (Fig. 6A). Consistent with previous reports using compound 43 in vivo with similar or longer dosage regiment than in our study presented here, we did not observe toxicity because of compound 43 treatment (24, 37). Mice treated with Cmpd-43 showed a significantly lower tumor burden, highlighting the efficacy of our PRL2 inhibitor to prevent tumorigenesis in this model (0.4189 g vs. 0.0929 g, P = 0.0020; Fig. 6B). Immunoblotting of the tumors derived from compound and vehicle-treated mice with the proliferation marker (PCNA) and apoptosis markers (PARP and caspase-3) showed a 50% reduction in tumor cell proliferation upon compound treatment, with no change in apoptosis (Fig. 6C and D). These results are consistent with our observations for the genetic inhibition of PRL2 in this model. We showed that genetic deletion of Prl2 leads to PTEN augmentation and suppression of Akt signaling in this Tp53^−/− model. We therefore immunoblotted for PTEN and Akt phosphorylation to determine whether pharmacologic inhibition of PRL2 with Cmpd-43 also elevates PTEN protein and inhibits Akt signaling to prevent tumorigenesis as observed. We observed approximately 85% increase in PTEN, and approximately 70% reduction in Akt phosphorylation following Cmpd-43 treatment (Fig. 6C and E). Reduced Akt activation was accompanied by reduced phosphorylation of downstream targets such as FoxOs (∼45% reduction), GSK3α (∼55% reduction), and GSK3β (∼35% reduction), which corroborates our observations in the genetic model (Fig. 6F and G). Interestingly, we did not observe any changes in mTOR phosphorylation. Reduced phosphorylation of FoxOs and GSK3s affects the expression of P27 (∼110% increase) and cyclin D1 (∼46% reduction) after pharmacologic inhibition of PRL2 (Supplementary Fig. S6A and S6B). These changes in Akt target phosphorylation as well as p27 and cyclin D1 expression support the smaller tumor size observed after Cmpd-43 treatment. These results suggest that trimer disruption with Cmpd-43 leads to PTEN augmentation and Akt inhibition in our inducible Tp53^−/− tumor model. In addition, Cmpd-43 was previously shown to inhibit ERK signaling in a melanoma xenograft model and in cell culture. Just like in the genetic model, pharmacologic inhibition of PRL2 with Cmpd-43 did not affect ERK activation (Fig. 6C and E). In addition, no changes in the expression of the P53 targets p21, BAX and puma were observed upon treatment with Cmpd-43 (Supplementary Fig. S6A and S6B). Taken together, pharmacologic inhibition of PRL2 with Cmpd-43 in our inducible Tp53^−/− model phenocopied the results observed with genetic deletion of Prl2 in this model.

FIGURE 6.

Pharmacologic inhibition of PRL2 in tamoxifen-inducible Tp53^−/− mouse inhibits tumor proliferation through PTEN augmentation to hinder tumor growth. A, Experimental design for the treatment of inducible Tp53^−/− with compound 43. B, Dot plot for mass of thymic lymphoma derived from inducible Tp53^−/− treated with or without compound 43, data are shown as mean ± SEM, Tp53^−/− + vehicle n = 20, Tp53^−/− + compound 43 = 13. C, Representative Western blot analysis from thymic lymphomas derived from inducible Tp53^−/− mice treated with vehicle or compound 43 to assess the levels PCNA, c-PARP, c-caspase 3, PTEN, and the activation of Akt and ERK1/2. D, Dot plot for quantification of C. E, Representative western blot from thymic lymphomas derived from inducible Tp53^−/− mice treated with vehicle or compound 43 to assess the activation of Akt targets. F, Dot plot for quantification of E. Data are shown as mean ± SEM for samples obtained 12 weeks after tamoxifen treatment, Tp53^−/− + vehicle n = 7–15, Tp53^−/− + compound 43 n = 7–11; *, P < 0.05; **, P < 0.01.

Discussion

The PRL phosphatases have garnered interest as therapeutic targets for cancer treatment due to their overexpression and oncogenicity to drive malignancies that correlate with poor survival prognosis (38–41). We recently showed that deletion of Prl2 in a Pten heterozygous background results in PTEN augmentation and impedes Pten deficiency–induced spontaneous tumorigenesis by inhibiting tumor cell proliferation and promoting their apoptosis (9). In that work, we showed that mechanistically, PRL2 dephosphorylates PTEN on Y336 to promote its ubiquitination by NEDD4 and subsequent degradation. The work we are reporting here shows that PTEN augmentation resulting from genetic or pharmacologic inhibition of PRL2 can be exploited therapeutically to prevent cancer development in cancers that develop independently from PTEN inactivation and PI3K/Akt signaling hyperactivation. On the basis of cBioPortal cancer genomic database, more than 90% of all cancers develop without PTEN alterations, which emphasizes the importance of proving the efficacy of PTEN augmentation therapy in such cancers. We used the well-established Tp53 deficiency tumor models to test this and confirmed PTEN augmentation as a result of both Prl2 genetic deletion and PRL inhibition in Tp53-deficient mice. The tumor profile we observed from both models was consistent with literature reports (23, 27, 42, 43).

We developed a model for conditional Prl2 deletion to control the where and when PRL2 could be deleted in mice, which would allow us to better study the function(s) of PRL2 in specific cancer types and in normal physiology. In this work, we induced whole body deletion of both Prl2 and Tp53 in the mice at the same time. Tp53-null mice predominantly developed thymic lymphomas and Prl2 deletion in this model significantly delayed the development of these lymphomas. This model allowed us to primarily study the effect of Prl2 deletion on Tp53-null thymic lymphomas, which accumulate hallmark mutations observed in T-cell acute lymphoblastic leukemia in humans (21). Recent work showed that deletion of Akt1 in Tp53^−/− mice prevents the development of thymic lymphomas (30), and thus we would expect that inhibition of PRL2 to increase PTEN protein level and attenuate Akt signaling in such a model would also hinder tumor growth. The deletion of Akt1 in Tp53^−/− mice resulted in approximately 85% increase in survival time (30), compared with approximately 30% increase in the work we present here. As expected, we showed that Prl2 deletion prevents tumorigenesis in this model through PTEN augmentation and Akt inhibition. Unlike the model that used Akt-null mice, and thus completely prevented all Akt signaling, we only have partial inhibition of Akt signaling upon Prl2 deletion. Thus, such differences in the effect of Prl2 deletion compared with Akt deletion can be expected. Although the inhibition of PI3K/Akt pathway component appears promising for cancer treatment based on laboratory research, the development of approved medication that directly target members of this signaling pathway has been challenging because of on-target and off-target toxicities associated with direct Akt inhibition (44, 45). As an alternative approach, an increase in PTEN level is expected to oppose PI3K/Akt signaling activation and may have therapeutic benefits against cancer. Indeed, PTEN overexpression in mice has little adverse effect and has been shown to confer a tumor resistant phenotype (46, 47). Also, constitutive Prl2^−/− mice have a normal life span even though their PTEN levels were 25% higher than the WT controls, and these mice show no other physiologic defect besides reduced body weight compared with control, which is consistent with observations in PTEN-overexpressing mice (7, 9). Consequently, PTEN restoration or augmentation therapy is a viable approach for cancer treatment that could circumvent the challenges faced by the inhibition of other members of the PI3K/Akt signaling pathway (48–50). In addition, PTEN has been shown to have anticancer function that is independent of its canonical role in the PI3K/Akt signaling pathway (12, 48, 50). Consequently, inhibition of PRL2 for PTEN augmentation would provide a viable anticancer approach, and the work presented by our group strongly supports the anticancer benefits of this approach.

We previously showed that Cmpd-43 (24), a small-molecule PRL detrimerizer, can prevent tumorigenesis or leukemogenesis in cancer cell-derived xenograft models and patient-derived xenograft models (24, 36, 37). For the first time here, we assessed the efficacy of these compounds in suppressing tumorigenesis in vivo in a genetically modified mouse cancer model. Because we had determined the growth kinetics of thymic lymphoma in Tp53^−/− mice, we used this model to assess the efficacy of our compound. We observed significant tumor growth inhibition after 3 weeks of treatment with Cmpd-43, and just like we saw with genetic deletion of Prl2, we found that pharmacologic inhibition of PRL2 suppresses Akt signaling through PTEN augmentation to inhibit tumor cell proliferation with no effect on apoptosis in the thymic lymphomas. Also, although both Akt and ERK attenuation has been previously reported upon PRL downregulation in vitro, the effect on ERK has not been observed in in vivo models. Interestingly, treatment of these mice with Cmpd-43 over 3 weeks for PRL2 inhibition had comparable effect on tumor size as the genetic deletion of PRL2 which occurred at the same time as Tp53 deletion, with no obvious adverse effect as also noted previously (24, 37). These results are much better than one would expect, and this is potentially due to Cmpd-43 being able to inhibit PRL1 and PRL3 in addition to PRL2 to provide synergistic effect on inhibition of tumor growth (24). Although we did not observe any improvement in survival when Prl1 or Prl3 were independently deleted in Tp53-null mice background (Supplementary Fig. S7), it is possible that combined inhibition of PRL1 and PRL2 could synergistically affect tumorigenesis. We previously observed such synergistic effect upon deletion of both PRL1 and PRL2 to impair spermatogenesis, while PRL1 deletion alone had no effect on spermatogenesis (6). Thus, inhibition of multiple PRLs by pan-PRL inhibitors such as Cmpd-43 in cancer cells could be more effective compared with single PRL inhibition. An additional advantage of Cmpd-43 is that it exploits an oligomerization mechanism unique to PRL therefore ensuring high selectivity over other PTPs. However, future studies are needed to develop PRL isoform-specific detrimerizers or inhibitors to further assess the therapeutic effectiveness of targeting a single PRL protein.

Interestingly, previous in vitro studies showed that PRL1 and PRL3 can downregulate p53 expression, and the expression of these PRLs is also transcriptionally regulated by p53 conversely (51–53). However, no such observation was made for PRL2. We previously found that Prl2 deletion stabilized p53 though PI3K-Akt-MDM2-p53 axis, which promoted tumor cell apoptosis in lymphomas induced by PTEN deficiency (9). In the current study with Tp53 deficiency tumor models, we did not observe any impact on the expression of p53 targets or apoptosis upon Prl2 deletion in sarcoma or thymic lymphoma from both Tp53-H and Tp53-null tumors. These data suggest that the impact of Prl2 on p53 is tissue/tumor type specific. More studies are required to further assess the underlying mechanism of whether or how PRL2 controls p53 under different conditions.

The tumor suppressor P53 is the most mutated and inactivated tumor suppressor in human cancers, and its dysregulation promotes a more severe cancer phenotype. Current therapeutic approaches under development for the treatment of such cancers involve inhibition of p53 degradation for cancers expressing WT p53 with or without lower protein expression and restoration of WT p53 function in cancers with mutated or truncated p53 expression (18, 54). Unfortunately, none of these approaches have been approved for clinical use due to p53-dependent and independent toxicity (54). This highlights the need for novel therapies that target cancers with p53 mutation or downregulation, and through this work we provide the proof-of-concept for PTEN augmentation approach via PRL2 inhibition. Previous work on PRL2, and the work we report here, show that loss of PRL2 expression or PRL2 inhibition has no adverse effect on survival, but instead promotes an antitumor phenotype through PTEN augmentation. Up to 80% of patients with sarcoma have P53 mutation or downregulation, but almost no PTEN mutation, which makes these tumors good targets for PTEN augmentation therapy. In mouse models of Tp53 deficiency, genetic loss of Prl2 or pharmacologic inhibition of PRL2 led to PTEN augmentation and attenuation of Akt signaling, which ultimately inhibited tumor cell proliferation to prevent tumor growth and promote survival in these tumor models. In conclusion, through this work, we not only further established the mechanistic regulation of PTEN by PRL2 but also demonstrated that genetic and pharmacologic inhibition of PRL2 in vivo could be exploited for PTEN augmentation therapy for the treatment of Tp53-deficient tumors.

Supplementary Material

Figure S1

Representative flowchart for flow cytometry analysis

Figure S2

Experiemental design for tamoxifen inducible Tp53 deletion w/o Prl2 deletion

Figure S3

Prl2 deletion inhibits tumor cell proliferations

Figure S4

Prl2 deletion upregulates PTEN to inhibit Akt signaling and hinder tumor cell proliferation in Tp53 heterozygous mice

Figure S5

Prl2 deletion does not affect Erk activation in inducible Tp53-/- mice derived thymic lymphomas

Figure S6

Effect of pharmacological inhibition of PRL2 in inducible Tp53-/- model on P53 and FoxO targets

Figure S7

Loss of PRL1 or PRL3 does not improve tumor-free survival of Tp53-null mice

Authors’ Disclosures

J. Miao reports grants from NIH during the conduct of the study; grants from NIH and Purdue University outside the submitted work. No disclosures were reported by the other authors.

Authors’ Contributions

F. Nguele Meke: Conceptualization, data curation, formal analysis, investigation, writing-original draft. Y. Bai: Conceptualization, data curation, formal analysis, investigation, writing-review and editing. D. Ruiz-Avila: Data curation. C. Carlock: Data curation. J. Ayub: Data curation. J. Miao: Resources. Y. Hu: Data curation. Q. Li: Resources. Z.-Y. Zhang: Conceptualization, supervision, funding acquisition, project administration, writing-review and editing.