Introduction

Lung cancer carries the highest incidence and mortality of cancer-related diseases, and remains a major public-health issue in China [1]. Rat sarcoma virus (Ras) genes (Kirsten rat sarcoma viral oncogene homolog (KRas), neuroblastoma RAS viral oncogene homolog (NRas), Harvey rat sarcoma viral oncogene homolog (HRas)) are the most common genes driving cancer. KRAS is the most frequently mutated RAS isoform (22% of all tumors), and NRAS and HRAS account for 8% and 3%, respectively [2]. Mutations to 12, 13, and 61 codons convert RAS into an oncoprotein by impairing intrinsic and guanosine triphosphate (GTP)ase-activating protein-mediated GTPase activity [3]. The three isoforms of RAS protein show high similarity, but a distinct bias in mutation-site ‘signatures’ has been discovered. 80% of KRAS mutations occur at codon 12, whereas 60% of NRAS tumors harbor mutations at codon 61 [2]. RAS mutations are present in one-third of patients with lung adenocarcinoma [4]. Mutations of RAS activate the downstream mitogen-activated protein kinase (MAPK) pathway defined by a rapidly accelerated fibrosarcoma (RAF)-mitogen-activated protein kinase kinase (MEK)-extracellular regulated protein kinases (ERK) signaling axis, which is a critical mediator of RAS-driven tumorigenesis and cancer development [5, 6].

Tyrosine phosphatases dephosphorylate proteins and participate in regulation of key signaling [7, 8]. Recent studies have shown dysfunction of tyrosine phosphatases to be involved in sustained activation of MAPK signaling in cancer. Loss of protein tyrosine phosphatase receptor type R (PTPRR) has been found in cancer of the colorectum, cervix, prostate gland, and ovary [8,9,10,11]. Silencing of PTPRR expression caused by aberrant methylation of DNA in the cervix has been reported to facilitate ERK1/2 hyperactivation [9]. Re-expression of PTPRR impairs ERK1/2 phosphorylation, restrains downstream signaling, and delays tumor progression [9, 11].

Despite the tumor-suppressor function of tyrosine phosphatases, tyrosine-protein phosphatase non-receptor type 11 (Ptpn11)-encoded Src homology 2 containing protein tyrosine phosphatase 2 (SHP2) dephosphorylates RAS protein, restores the canonical GTPase cycle, augments RAS binding affinity to effectors and, consequently, promotes MAPK signaling [12]. Recently, SHP2 has been postulated to be a potential drug target for RAS-driven cancers. Pharmacologic inhibition of SHP2 expression has been shown to suppress RAS-MAPK signaling and impair the growth of cancer cells in RAS-mutant cell lines, xenografts, and patient-derived organoids [13, 14]. However, 61 codon mutations of both KRAS and NRAS confer resistance to SHP2 inhibitors, which limited possible future applications of SHP2-targeted therapy on RAS-mutant lung cancers [15, 16].

Targeting the RAS-MAPK pathway has been suggested as a treatment for RAS-mutant lung cancer [17]. Given that the phosphatases PTPRR and SHP2 mediate the dephosphorylation and phosphorylation of ERK1/2, respectively, we wondered if interventions targeting phosphatases had therapeutic efficacy by preventing ERK1/2 overactivation in lung cancer. However, the function and regulation of PTPRR in lung adenocarcinoma, and the anti-tumor effect of targeting the phosphatases PTPRR and SHP2 simultaneously, are not known.

Here, we report that ectopic expression of silenced PTPRR suppressed the proliferation and migration of lung cancer cells. ERK1/2 dephosphorylation by PTPRR and inhibition of SHP2 cooperatively restrained MAPK signaling in RAS-mutant lung adenocarcinoma. Notably, the histone deacetylase (HDAC) inhibitor SAHA recovered PTPRR expression by increasing acetylation of histone H3K9 in the promoter region of PTPRR. Dual inhibition of HDAC and SHP2 had a synergistic effect in RAS-mutant lung cancer. These results suggest that targeting dysregulated phosphatases to suppress RAS-MAPK signaling could be an effective strategy to treat RAS-mutant lung adenocarcinoma.

Materials and methods

Patient records

Patient records were collected from the Cancer Genome Atlas (TCGA) to analyze PTPRR expression by TNMplot (www.tnmplot.com). Images of histological sections from normal lung tissue and lung adenocarcinoma were obtained from the Human Protein Atlas [18]. GSE30219 from Gene Expression Omnibus repository (GEO) was used to analyze correlation between PTPRR expression and prognosis of people suffering from lung adenocarcinoma by Kmplot (www.kmplot.com) [19, 20].

Cell culture

Human lung adenocarcinoma cell lines (NCI-H1299, NCI-H2087, A549, and NCI-H358) and immortalized human lung epithelial cell line BEAS-2B were obtained from China Infrastructure of Cell Line Resources (Beijing, China). NCI-H1299, NCI-H2087, A549, and NCI-H358 cells were cultured in RPMI-1640 (Corning Life Sciences, Tewksbury, MA, USA) supplemented with 10% inactivated fetal bovine serum (FBS; Gemini, West Sacramento, CA, USA). BEAS-2B cells were cultured by Bronchial Epithelial Cell Growth Medium BulletKit™ (Lonza, Basel, Switzerland). Mycoplasma testing (InvivoGen, San Diego, CA, USA) was undertaken before experimentation.

Chemicals

The compounds SAHA (Vorinostat™), trichostatin A (TSA), and SHP099 were purchased from TargetMol (Boston, MA, USA). For in vitro assays, compounds were dissolved in dimethyl sulfoxide and added to cell culture media to achieve final concentrations.

PTPRR overexpression in lung cancer cells

PTPRR complementary-DNA (NM_002849, 658aa) was cloned to the pCDH lentiviral expression vector and transfected to 293 T cells with pMD2G and pSPAX2 packaging plasmids. Transfections were undertaken using Lipofectamine™ 3000 following manufacturer instructions (Thermo Fisher Scientific, Waltham, USA). After 24 h, 48 h, and 72 h, the supernatant was harvested, filtered, and concentrated by adding lentivirus solution (Yeason, Shanghai, China). Lung cancer cells were seeded to 12-well plates and infected in the presence of polybrene (8 μg/mL; Cyagen, Suzhou, China). After 24 h, cells were placed under fresh medium with puromycin selection (1.5 μg/mL) to generate cells with PTPRR expression.

Knockdown of gene expression via small interfering (si)RNA or short hairpin (sh)RNA

Silencing of SHP2 expression in A549 cells and NCI-H1299 cells was mediated by siRNA with the indicated sequences. Oligonucleotides targeting human HDAC1, HDAC2, HDAC3, and PTPRR were designed to construct the pKO.1 lentiviral vector. The lentivirus was produced in 293 T cells by co-transfecting pMD2G and psPAX2 plasmids. Filtered supernatant was concentrated and used to infect target cells. The sequences of siRNA and shRNA are listed in Supplementary Table 1.

RNA extraction, cDNA synthesis, and real-time reverse transcription-quantitative polymerase chain reaction (RT-qPCR)

Total RNA was isolated from each sample using a RNeasy kit (Qiagen, Hilden, Germany). RNA was reverse-transcribed to cDNA using TransScript® One-Step gDNA Removal and cDNA Synthesis SuperMix (TransGen, Beijing, China) according to manufacturer protocols. Real-time RT-qPCR was undertaken using Hieff® qPCR SYBR Green Master Mix (Yeason Biotechnology, Shanghai, China) in a real-time thermal cycler (qTOWER3; Analytik, Jena, Germany). The primers we employed are listed in Supplementary Table 2.

Chromatin immunoprecipitation (ChIP) assay

The ChIP assay was carried out with a SimpleChIP® Enzymatic Chromatin IP kit (9003; Cell Signaling Technology) according to manufacturer protocols. The antibody used in the process was anti-acetyl-histone H3 (Lys9) (PTM-156; PTM Bio). Input DNA (2%) and eluted DNA were subjected to real-time RT-qPCR. Three pairs of primers that targeted the promoter region of PTPRR were designed, and are shown in Supplementary Table 2.

Tumor xenograft model

Male Balb/c nude mice (8–10 weeks) were purchased from Charles River Laboratories (Beijing, China). NCI-H1299 pCDH- and PTPRR-overexpressed cells (5 × 107/mL) were prepared in PBS, mixed with Matrigel™ (BD Biosciences, Franklin Lakes, NJ, USA) and injected (s.c.) into the left flank and right flank of mice. To study the efficacy of combination therapy using SAHA and SHP099, NCI-H1299 or A549 tumor tissues were cut to pieces (1 mm × 1 mm) and inoculated to the right flank of male Balb/c nude mice. Once the tumor volume reached 100 mm3, mice were grouped randomly and received vehicle, SAHA (75 mg/kg, once-daily), SHP099 (75 mg/kg, every other day), or a combination of SAHA (75 mg/kg, once-daily) plus SHP099 (75 mg/kg, every other day). Tumor volume and bodyweight were monitored twice a week. When the length of tumor surpassed 15 mm, mice were sacrificed and tumor xenografts were harvested. The protocol for animal experiments was approved by the Ethics Committee for Animal Experiments of the Institute of Materia Medica, Chinese Academy of Medical Sciences & Peking Union Medical College.

Statistical analysis

IC50 values were analyzed by nonlinear regression and calculated by Graphpad Prism 8. Student’s unpaired t test or one-way ANOVA were used to evaluate differences between groups. For all experiments, *p < 0.05, **p < 0.01 and ***p < 0.001 were considered statistically significant.

Results

Re-expression of silenced PTPRR suppresses malignant phenotypes of lung cancer

Lower PTPRR expression in lung adenocarcinoma tissues than that in adjacent normal lung tissues were observed in patient samples from the Cancer Genome Atlas and Human Protein Atlas (Fig. 1a). Representative results from immunochemical staining of tissues from the Human Protein Atlas also showed diminished expression of PTPRR in sections of lung tissue with adenocarcinoma (Fig. 1b). Similarly, PTPRR expressions in lung cancer cell line including NCI-H1299, NCI-H2087, and A549 were significantly downregulated relative to the lung normal cell line BEAS-2B (Fig. 1c). Clinical data from GEO dataset was used to investigate the relationship between PTPRR expression and tumor progression. As it shown in Fig. 1d, PTPRR expression was positively co-related with overall survival for patients suffering from lung adenocarcinoma.

Fig. 1
figure 1

PTPRR was underexpressed in lung cancer. a Diminished expression of PTPRR was found in lung adenocarcinoma tissues (n = 57) versus adjacent normal lung tissues (n = 57) by TMNplot. b Tissue sections documented on the Human Protein Atlas indicated low expression of PTPRR in malignant lung carcinoma. Scale bar, 200 μm c PTPRR expression in the immortalized human normal lung cell line BEAS-2B and three lung cancer-derived cell lines were examined by immunoblotting. d Kaplan–Meier survival analysis of GSE30219 data set from GEO showed positive correlation of PTPRR (Affymetrix ID: 206084_at) expression with overall survival. e Growth curve of NCI-H1299 cells and NCI-H2087 cells infected with pCDH or pCDH-PTPRR lentivirus. f and g Motility and migration capabilities of PTPRR-expressing cells were assessed by wound-healing (f) and Transwell™ assays (g). Scale bar, 100 μm. Data are shown as mean ± SD. n = 3, two-tailed unpaired Student’s t-test was used for statistical analysis. *p < 0.05 and ***p < 0.001

Full size image

Exogenous expression of PTPRR suppressed the growth of human lung cancer (NCI-H1299, NCI-H2087) cells (Fig. 1e). Moreover, PTPRR overexpression inhibited the migration ability of lung cancer cells significantly as determined by the wound-healing assay (Fig. 1f) and Transwell assay (Fig. 1g). These findings suggested that expression of PTPRR phosphatase was silenced in lung adenocarcinoma, and had an important role in the proliferation and migration of cancer cells.

PTPRR suppresses ERK1/2 phosphorylation and sensitizes RAS-mutant lung cancer cells to SHP2 inhibition

It has been reported that PTPRR removes phosphate groups from ERK1/2 and inhibits the growth of cervical cancer cells [9]. Similarly, exogenous expression of PTPRR in HEK293T cells downregulated ERK1/2 phosphorylation significantly (Fig. 2a). Gain-of-function mutation of the RAS proto-oncogene activates RAS-MAPK pathway abnormally, and is involved in the proliferation and signal transduction of lung cancer cells. Following ectopic PTPRR expression, NRAS Q61 mutant NCI-H1299 cells and NCI-H2087 cells showed a marked reduction of phospho-ERK1/2 while negligible changes were found in KRAS G12-mutant A549 cells and NCI-H358 cells (Fig. 2a), The activator protein-1 (AP-1) transcription factors are known to be targets of ERK signaling [21]. As one pivotal component of AP-1 complex, FOS-related antigen 1 (FRA-1) was reported to linearly responded to ERK1/2 activation and was inhibited by PTPRR expression in cervical cancer [9, 22] Likewise, downstream MAPK-dependent FRA-1 transcription showed the same trend with ERK1/2 phosphorylation in PTPRR-overexpressed lung cancer cells (Fig. 2b).

Fig. 2
figure 2

ERK1/2 activation in lung cancer cells was regulated by PTPRR and SHP2. a PTPRR inhibited ERK1/2 phosphorylation in NCI-H1299 cells and NCI-H2087 cells, but not in A549 cells or NCI-H358 cells. b Fra1 (downstream target of the MAPK pathway) expression was reduced by PTPRR expression in NCI-H1299 cells and NCI-H2087 cells. c ERK1/2 phosphorylation in lung cancer cells after SHP099 (50 μM) treatment for 24 h. d PTPRR downregulated ERK1/2 activation in SHP2-deficient cells significantly. e Western immunoblots in PTPRR-expressing NCI-H1299 cells and A549 cells treated with SHP099 (50 μM). f PTPRR-expressing NCI-H1299 cells and A549 cells were subjected to viability assessment with increasing concentrations of SHP099 for 72 h. Data are shown as mean ± SD. n = 3, two-tailed unpaired Student’s t-test was used for statistical analysis. *p < 0.05, **p < 0.01, and ***p < 0.001. n.s., no significance

Full size image

The Ptpn11-encoded tyrosine phosphatase SHP2 is an activator of MAPK signaling in cancer due to RAS mutations. The allosteric SHP2 inhibitor SHP099 reduced the cycling of RAS GTPase and impaired its binding to effector proteins. In contrast to the effects caused by PTPRR expression, addition of SHP099 suppressed ERK1/2 phosphorylation in KRAS G12-mutant cells but not in NRAS Q61-mutant cells (Fig. 2c). Therefore, we wondered whether lung cancer cells harboring a different RAS status or mutant “hotspot” codons could be treated with PTPRR re-expression in combination with SHP2 inhibition. For NRAS Q61-mutant NCI-H1299 cells, although silencing of SHP2 expression by siRNA transfection did not induce a noticeable change of phospho-ERK1/2, a stronger inactivation of ERK1/2 was detected in cells overexpressing PTPRR (Fig. 2d). In KRAS G12-mutant A549 cells, knockdown of SHP2 expression showed a mild effect, whereas PTPRR expression led to a significant reduction of ERK1/2 phosphorylation (Fig. 2d). Similar effects were observed by treating PTPRR-expressing NCI-H1299 cells and A549 cells with SHP099 for 24 h (Fig. 2e). Then, a wide range of SHP099 concentrations was used to determine the change in the half-maximal inhibitory concentration (IC50) following PTPRR expression. As expected, PTPRR-sensitized NCI-H1299 cells and A549 cells underwent SHP2 inhibition (NCI-H1299 cells: IC50 85.9 μM vs. 33.3 μM; A549 cells: IC50 17.4 μM vs. 3.54 μM) (Fig. 2f). Therefore, PTPRR in combination with a SHP2 inhibitor was an effective strategy to target lung cancer because it cooperatively increased the removal and reduced the addition of phosphate groups to ERK1/2.

Inhibition of HDAC restores PTPRR expression in lung cancer cells

It has been reported that DNA methyltransferase 3 beta (DNMT3B)-mediated high methylation of DNA is responsible for PTPRR silencing in colorectal cancer and cervical cancer [9]. A shRNA targeting DNMT3B was packaged to a lentivector and NCI-H1299 cells were infected. Surprisingly, PTPRR expression was not affected by knockdown of DNMT3B expression (Fig. 3a). Meanwhile, an inhibitor of DNA methyltransferase, decitabine, also did not induce an increase in mRNA expression of PTPRR (Fig. 3b). This result prompted us to speculate that silencing of PTPRR expression in lung cancer might be the result of another pattern of epigenetic modification.

Fig. 3
figure 3

Upregulated H3K9ac level in the promoter region initiates PTPRR transcription. a DNMT3B expression in NCI-H1299 cells was knocked down using the lentiviral vector pKO.1 expressing shDNMT3B. The change in PTPRR mRNA expression was measured by RT-qPCR. b NCI-H1299 cells were treated with decitabine for 24 h and the transcription of PTPRR was measured. c NCI-H1299 cells were infected by lentivirus containing the empty pLKO.1 vector or pLKO.1-shHDAC1, HDAC2, and HDAC3. d Lung cancer cells were treated with SAHA (0, 2.5, 5, 10, or 20 μM) or TSA (2 μM) for 12 h and the acetylation level of histone H3K9 was determined. e After treating NCI-H1299 cells and NCI-H2087 cells with TSA (2 μM) for 12 h, PTPRR expression was detected by real-time RT-qPCR. f Expression of PTPRR mRNA was detected in SAHA (0, 10, or 20 μM)-treated lung cancer cells. g ChIP-qPCR of H3K9ac enrichment around the promoter region of PTPRR in SAHA-treated NCI-H1299 cells. h Higher acetylation level of histone H3K9 was found in BEAS-2B cells than NCI-H1299, NCI-H2087, and A549 cells. Data are shown as mean ± SD. n = 3, two-tailed unpaired Student’s t-test was used for statistical analysis. *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

Histone acetylation has a pivotal role in chromatin remodeling and regulation of gene expression. HDAC1, HDAC2, and HDAC3 are class-I HDACs that regulated cancer development by deacetylating histones and regulating cellular homeostasis [23]. Higher level of HDAC1, HDAC2, and HDAC3 were found in lung adenocarcinoma than adjacent normal lung tissues (Fig. S1). Suppression of HDAC1, HDAC2, and HDAC3 via shRNA released PTPRR expression in NCI-H1299 lung cancer cells, (Fig. 3c), suggesting that PTPRR expression was silenced in lung adenocarcinoma due to changes in histone coding. Inhibition of HDAC activity by the clinically available drug Vorinostat (SAHA) and a robust anti-HDAC agent (TSA) increased acetylation of histone H3 at Lys 9 residues significantly (Fig. 3d). Consequently, treatment with TSA and SAHA upregulated PTPRR expression markedly (Fig. 3e, f). Accordingly, chromatin immunoprecipitation by antibody against H3K9 acetylation and quantitative real-time PCR using primers targeting PTPRR promoter was performed. The results showed that H3K9 acetylation at the promoter region of PTPRR were dramatically elevated in the presence of SAHA compared to DMSO control (Fig. 3g). Moreover, histone H3K9 acetylation level of PTPRR promoter region in human lung BEAS-2B cells were significantly higher than that in lung cancer cells, which also explains the diminished PTPRR expression in lung adenocarcinoma (Fig. 3h). Taken together, these data suggested the role of histone acetylation in maintaining PTPRR gene expression by transcriptional regulation.

HDACi treatment restrains ERK1/2 signaling and sensitizes RAS-mutant lung cancer cells to SHP2 inhibition

As PTPRR dephosphorylates ERK1/2, we next studied the effect of HDAC inhibition on ERK1/2 signaling. Knockdown of HDAC1, HDAC2, and HDAC3 expression suppressed ERK1/2 activation in NCI-H1299 cells significantly (Fig. 4a). In accordance with the effect of PTPRR overexpression on RAS-mutant lung cancer cells, the HDAC inhibitor SAHA reduced ERK1/2 signaling in NRAS Q61-mutant cells dose-dependently, but showed little effect on KRAS G12-mutant cancer cells (Fig. 4b). In NCI-H1299 cells, inhibition of ERK1/2 signaling accompanied with PTPRR expression were induced by SAHA treatment and lasted from 6 to 36 h (Fig. 4c). The transcription of Fra-1, which is downstream of MAPK signaling, was impaired severely by SAHA treatment in NRAS-mutant lung cancer cells (Fig. 4d).

Fig. 4
figure 4

HDAC inhibition restrains ERK1/2 signaling. a Lentivirus packaging empty pLKO.1 vector or shRNAs targeting HDAC1, HDAC2, and HDAC3 were used to infect NCI-H1299 cells. b Lung cancer cells were treated with SAHA (0, 5, 10, or 20 μM) for 24 h. c NCI-1299 cells were treated with SAHA (10 μM) for different times to assess expression of PTPRR and ERK1/2 phosphorylation. d NCI-H1299 and NCI-H2087 cells were treated with SAHA for 12 h and subjected to real-time RT-qPCR to determine fra-1 mRNA expression. ***p < 0.001. n.s., no significance

Full size image

To demonstrate the central role of PTPRR in SAHA-induced inhibition of MAPK signaling, we interfered with expression of PTPRR by shRNA in NCI-H1299 cells with or without SAHA treatment. Silencing of PTPRR expression rescued SAHA-induced ERK1/2 inhibition, thereby indicating that PTPRR was the main regulator of MAPK signaling under HDAC inhibition (Fig. 5a). To ascertain if the HDAC inhibitor SAHA could sensitize RAS-mutant lung cancer cells to SHP2 inhibition, lung cancer cell lines (NRAS Q61-mutant NCI-H1299, KRAS G12-mutant A549) were selected to determine the effect of SHP2-targeting siRNAs in combination with SAHA. As expected, silencing of SHP2 expression by siRNAs had a negligible effect on ERK1/2 phosphorylation in NCI-H1299 cells (Fig. 5b). However, in combination with SAHA, ERK1/2 activation in NCI-H1299 cells was impaired severely. Similarly, in A549 cells, although SAHA did not alter ERK1/2 signaling, addition of siRNAs targeting SHP2 showed a significant synergistic effect (Fig. 5b). Meanwhile, knockdown of SHP2 expression plus SAHA represented reduced proliferation of cancer cells (Fig. 5c).

Fig. 5
figure 5

SAHA treatment sensitizes RAS-mutant lung cancer cells to SHP2 inhibition. a Lentivirus containing shRNAs targeting PTPRR were used to infect NCI-H1299 cells with or without SAHA treatment (10 μM). b siRNAs were transfected to NCI-H1299 cells and A549 cells for 48 h, SAHA (10 μM) was added and incubated for 12 h. c Silencing of SHP2 expression in NCI-H1299 and A549 cells in combination with SAHA treatment (10 μM) restrain cell proliferation. Relative cell growth was determined by crystal violet staining and solubilized in 10% acetic acid. Data are shown as mean ± SD. n = 3, two-tailed unpaired Student’s t-test was used for statistical analysis. **p < 0.01 and ***p < 0.001

Full size image

SAHA synergizes with SHP099 in RAS-mutant lung cancer cells

It has been found that SAHA treatment and SHP2 inhibition were complementary in regulating MAPK signaling. Hence, we assessed the sensitivity of RAS-mutant lung cancer cell lines to the SHP2 inhibitor SHP099 upon SAHA treatment. Interestingly, addition of SAHA at low doses sensitized lung cancer cells to SHP099 significantly, with IC50 values lower in combinations compared with those using SHP099 alone (Fig. 6a). To determine the synergistic effect of SAHA and SHP099, we utilized the Chou–Talalay method to provide the combination index (CI) equation that allows quantitative evaluations of drug combinations, where CI < 1 indicates a synergistic inhibition [24]. As the fraction affected-CI plot shown in Fig. 6b, SAHA and SHP099 exerted synergistic effects when fraction of cells affected by treatment was 0.5. Mechanistically, treatment of SAHA plus SHP099 led to markedly reduced ERK1/2 phosphorylation in lung cancer cells (NRAS Q61-mutant NCI-H1299, KRAS-G12 mutant A549) (Fig. 6c). Incorporation of 5-Ethynyl-2’-deoxyuridine (EdU) and staining enabled accurate and image-based analyses of cell proliferation. The number of cells labeled with EdU was reduced dramatically upon treatment with SAHA plus SHP099 (Fig. 6d). Similarly, although a reduction in clonogenic ability by SAHA was not consistent in NCI-H1299 cells harboring a NRAS mutation and A549 cells with a KRAS mutation, combinations with SHP099 resulted in elimination of cell-colony formation (Fig. 6e). These results provided evidence that combining SAHA with SHP099 had a synergistic effect, and led to stronger inactivation of MAPK signaling in RAS-mutant lung cancer cells.

Fig. 6
figure 6

Combination therapy of SAHA and SHP099 strongly impairs the proliferation of lung cancer cells. a Increasing concentrations of SHP099 plus SAHA (1.25 μM) were added to a panel of RAS-mutant lung cancer cells. Cell viabilities were assessed by Cell-titer Glo at 72 h post-treatment. b Fraction affected (Fa)-combination index (CI) plots were obtained by determining the combination of a series concentrations of SAHA and SHP099. c Single and combination treatment of SAHA (10 μM) and SHP099 (25 μM) were added to four RAS-mutant lung cancer cell lines and western immunoblots obtained. d EdU staining of NCI-H1299 cells and A549 cells treated with SAHA (10 μM), SHP099 (50 μM), or their combination for 24 h. Scale bar, 100 μm. e NCI-H1299 cells and A549 cells were treated with SAHA (1.25 μM), SHP099 (50 μM for NCI-H1299, 10 μM for A549) or their combinations for 7 days and subjected to crystal-violet staining. Data are the mean ± SD. n = 3, two-tailed unpaired Student’s t-test was used for statistical analysis. *p < 0.05, **p < 0.01, and ***p < 0.001

Full size image

Combination of SAHA and SHP099 reduces the tumor burden in vivo

NCI-H1299 cells stably expressing the pCDH empty vector or ectopic PTPRR were injected into the left flank and right flank of male Balb/c nude mice to study the effect of PTPRR expression on tumor progression. Similar to the results of the in vitro study, PTPRR expression in NCI-H1299 cells led to markedly reduced tumor volume and tumor weight, thereby indicating that re-expression of PTPRR was sufficient to restrain the growth of tumor xenografts (Fig. 7a–d). To determine the efficacy of SAHA and SHP099, male Balb/c nude mice bearing NCI-H1299 xenografts were administered (p.o.) the vehicle control, SAHA, SHP099, or SAHA plus SHP099. Treatment with SAHA or SHP099 showed modest efficacy upon NCI-H1299 xenografts, whereas combination treatment resulted in an improved response and significantly delayed tumor growth without causing severe body weight loss (Fig. 7e-g). In the A549 xenograft model, combination treatment of SAHA and SHP099 also significantly suppressed tumor progression (Fig. 7i-l). Consistent with those detected in vitro, immunoblots of representative tumor lysates indicated significant suppression of ERK1/2 phosphorylation upon combination treatment, which was reduced dramatically compared with that in the vehicle control or if a single drug was used (Fig. 7m, n). These results supported the notion of clinical application of a HDAC inhibitor in combination with SHP2 inhibition.

Fig. 7
figure 7

Dual inhibition of HDAC and SHP2 reduced the tumor burden in RAS-mutant NCI-H1299 and A549 lung cancer xenografts. a NCI-H1299 cells expressing empty pCDH or ectopic PTPRR were mixed with Matrigel™ and transplanted to the left flank and right flank of male Balb/c nude mice, respectively. b Tumor volumes of NCI-H1299 pCDH and PTPRR were tracked for 30 days (n = 4). c Tumor weight of xenografts were obtained at the endpoint (n = 4). d Representative image of NCI-H1299 pCDH and PTPRR tumor xenografts (n = 4). e Tumor volume of NCI-H1299 xenografts treated with SAHA (75 mg/kg, p.o.), SHP099 (75 mg/kg, p.o.), or their combination (n = 5). f Tumor weight of NCI-H1299 xenografts (n = 5). g Representative image of NCI-H1299 tumor xenografts (n = 5). h Body weight change of mice bearing NCI-H1299 xenografts (n = 5). i Tumor volume of A549 xenografts treated with SAHA (75 mg/kg, p.o.), SHP099 (75 mg/kg, p.o.), or their combination (n = 5). j Tumor weight of A549 xenografts (n = 5). k Representative image of A549 tumor xenografts (n = 5). l Body weight change of mice bearing A549 xenografts (n = 5). m and n Expression of phosphorylated ERK1/2 and total ERK1/2 in NCI-H1299 (m) and A549 (n) tumor lysates was measured by western blotting. Data are presented as the mean ± SEM. One-way ANOVA was used for statistical analysis. *p < 0.05, **p < 0.01, and ***p < 0.001. n.s., no significance

Full size image

Discussion

By analyzing datasets from public database, lower PTPRR expression was observed in patient samples with lung adenocarcinoma compared with normal adjacent lung tissues, and positively correlated with the prognosis. Together with studies reporting PTPRR expression to be silenced and identified as a prognostic marker in cancer of the cervix, ovary, and oral squamous cell carcinoma [9, 10, 25], our research reinforced the tumor-suppressor role of PTPRR in cancer development.

DNA methylation can silence PTPRR expression in cervical cancer [9]. However, knockdown of DNMT3B expression or treatment with an inhibitor of DNA methyltransferase, decitabine, did not alter PTPRR expression in lung adenocarcinoma. Interference of HDACs by treatment with shRNA and a HDAC inhibitor enhanced acetylation of histone H3 Lys 9, which were sufficient to boost PTPRR expression epigenetically. Furthermore, the ChIP assay validated that acetylation of histone H3K9 was enriched in the promoter region of PTPRR gene, which suggested a different regulatory mechanism of silencing of PTPRR expression in lung cancer.

Several studies have mentioned that HDAC inhibitors suppressed ERK phosphorylation, but the underlying molecular mechanism is not known [26, 27]. In our study, silenced expression of PTPRR was restored by histone acetylation in lung cancer, and interference with PTPRR expression reversed (at least in part) SAHA-induced ERK1/2 inhibition. Hence, our investigations suggested a key role of PTPRR in suppression of MAPK signaling by HDAC inhibition.

We speculated that MAPK signaling inhibition is only one of the mechanisms of HDAC inhibition, as SAHA treatment showed little effect on ERK1/2 phosphorylation while severely inhibited the proliferation of KRAS-mutant cells. It is worth noticing that SAHA was identified as a pan HDAC inhibitor and suppressed both class I and class II HDAC [28]. Aside from histone acetylation, non-histone acetylation and protein non-acetyl acylations also have been found regulated by SAHA or pan HDAC inhibition, which facilitated tumor suppression though diverse cellular processes including gene transcription, cell cycle, DNA damage repair, signaling pathways, and cytoskeleton organization [29, 30].

Allosteric inhibitors targeting KRAS G12C directly have been applied to clinical therapeutics [31], but treatment of patients with lung adenocarcinoma harboring different RAS mutations is a challenge. Studies have shown that SHP2 is required for RAS-driven cancer progression, and that targeting SHP2 could be therapy for RAS-mutant cancer [13, 32]. It has been reported that, compared with the G12 mutation, the Q61 mutation of RAS shows impaired GTP hydrolysis and a minimal impact by phosphorylation [15]. Meanwhile, the RAS Q61 mutation binds RAF preferentially and activates MAPK signaling rather than other downstream cascades [33]. Therefore, the Q61 mutation not only protects RAS from SHP2 regulation, it also drives persistent MAPK hyperactivation. In line with recent reports, our results supported the notion that the Q61 mutation of NRAS in lung adenocarcinoma was a predictor of resistance to SHP2 inhibitors. Interestingly, we revealed that PTPRR-mediated dephosphorylation of ERK1/2 could aid suppression of Q61 mutant RAS-driven persistent MAPK activation, highlighting a novel strategy to treat RAS-mutant lung cancers.

As one of the main regulators during tumor progression, suppression of the MAPK pathway by dual inhibition of SHP2 and MEK/ERK has been demonstrated to be effective in multiple types of cancer cell [16, 34,35,36]. Our findings revealed that dysregulated phosphatases PTPRR and SHP2 maintained aberrant activation of ERK1/2 cooperatively and investigated the synergistic efficacy of HDAC and SHP2 co-inhibition. Hence, targeting of disordered phosphatases to suppress RAS-MAPK signaling may be a promising strategy against RAS-mutant lung cancer (Fig. 8).

Fig. 8
figure 8

Graphical summary of suppressing RAS-MAPK pathway by regulating dysfunctional phosphatase PTPRR and SHP2. Histone acetylation facilitates PTPRR expression in lung adenocarcinoma. A combination of the HDAC inhibitor (SAHA) and SHP2 inhibitor (SHP099) suppressed RAS-mutant lung cancer progression markedly through regulation of ERK1/2 signaling. The figure was created with BioRender.com

Full size image