Regulation of Hippo-YAP signaling by insulin-like growth factor-1 receptor in the tumorigenesis of diffuse large B-cell lymphoma

Lymph node samples from 60 de novo DLBCL patients and 30 reactive lymphoid hyperplasia patients were collected. Histological diagnoses were established according to the World Health Organization (WHO) classification [19]. Normal peripheral blood mononuclear cells (PBMCs) from healthy volunteers were isolated by the Ficoll-Hypaque density gradient centrifugation method (TBD Science, Tianjin, China). Normal B cells were purified from freshly isolated PBMCs using CD19+ magnetic microbeads kit (Miltenyi Biotec, Bergisch Gladbach, Germany) as previously reported [20, 21]. This study was approved by the Medical Ethical Committee of Shandong Provincial Hospital Affiliated to Shandong University. All samples were obtained with informed consent in accordance with the Declaration of Helsinki.

Human DLBCL cell lines (LY1, LY8, LY3, and Val) were routinely cultured in Iscove’s modified Dulbecco’s medium (IMDM) with 10% heat-inactivated fetal bovine serum (Gibco, MD, USA). The medium contained a 1% penicillin/streptomycin mixture and 2 mM glutamine. Verteporfin (VP; a YAP inhibitor, SML0534) was obtained from Sigma (MO, USA). Recombinant human IGF-1 was obtained from PeproTech (100-11, NJ, USA). Doxorubicin (S1208), AG1024 (an IGF-1R inhibitor, S1234), and picropodophyllin (PPP; an IGF-1R inhibitor, S7668) were purchased from Selleckchem (TX, USA).

The sequences for YAP and IGF-1R shRNAs were as follows: shYAP 1#, 5′-GCCACCAAGCTAGATAAAGAA-3′; shYAP 2#, 5′-CCCAGTTAAATGTTCACCAAT-3′. shIGF-1R 1#, 5′-GCCGAAGATTTCACAGTCAAA-3′; and shIGF-1R 2#, 5′-GCCTTTCACATTGTACCGCAT-3′. The control shRNA was synthesized with a scrambled sequence. The shRNAs were cloned into lentiviral vectors (GeneChem, Shanghai, China). Lentivirus infection was carried out according to the manufacturer’s instruction. Stably transfected cells were selected with puromycin (2.0 μg/ml). Seventy-two hours after transfection, the cells were collected and used for subsequent analysis.

YAP-knockout (YAP^−/−) cells were generated using CRISPR/Cas9 genomic editing system. The production and packaging of lentiviral vectors for stably expressing Cas9-gRNA was accomplished by GeneChem. The gRNA target sites for YAP deletion were as follows: sgYAP#1, 5′-TGGGGGCTGTGACGTTCATC-3′; sgYAP#2, 5′-GAGCACTCTGACTGATTCTC-3′; and sgYAP#3, 5′-ACATCGATCAGACAACAACA-3′. Validation of YAP^−/− cells selected by puromycin (2.0 μg/ml) was conducted by PCR analysis of genomic DNA coupled with DNA sequencing. The primers to amplify sgYAP cut sites are listed in Table S1.

Immunohistochemical staining was performed as previously described [20]. The negative control was detected with the primary antibody replaced by PBS. Immunohistochemical staining was scored by two independent observers who were blinded to the patients’ clinical data. The scoring system for grading the level of YAP was reported previously [22]. The expression level was evaluated by immunohistochemistry (IHC) score calculated by multiplying a proportion score and intensity score. The proportion score reflected the fraction of positive-stained cells (0, none; 1, ≤ 10%; 2, 10–25%; 3, 25–50%; and 4, > 50%), and the intensity score revealed the staining intensity (0, no staining; 1, weak; 2, intermediate; and 3, strong). Finally, the total score was calculated. High and low protein expression levels were defined using the mean score of all samples as a cutoff point. With these criteria, tissue staining could be interpreted as “low” or “high.” The primary antibodies used were anti-YAP (Proteintech Group, 13584-1-AP, IL, USA), anti-Ki67 (Proteintech Group, 27309-1-AP), and anti-c-myc (Abcam, ab32072, Cambridge, UK). For hematoxylin-eosin (H&E) staining, fresh subcutaneous tumors isolated from mice were fixed in 4% paraformaldehyde (PFA) and embedded in paraffin for histological examinations. Sections with thickness of 4 μm were cut and stained with H&E.

Nuclear and cytoplasmic proteins were isolated with NE-PER Nuclear and Cytoplasmic Extraction Reagents (Thermo Fisher Scientific, MA, USA) following the manufacturer’s instructions. The levels of GAPDH and Histone H3 were used as loading controls for the nuclear and cytoplasmic fractions, respectively

Cell lysates were extracted with radioimmunoprecipitation assay buffer together with a 1× phosphatase inhibitor cocktail (PhosSTOP, Roche, Basel, Switzerland). Western blotting was performed as previously described [20, 23]. The primary antibodies used were anti-phospho-IGF-1R (Tyr1135/1136) (Cell Signaling Technology, 3024, MA, USA), anti-IGF-1R (Cell Signaling Technology, 9750), anti-Mcl-1 (Cell Signaling Technology, 5453), anti-Caspase-3 (Cell Signaling Technology, 14220), anti-cleaved Caspase-3 (Cell Signaling Technology, 9661), Caspase-8 (Cell Signaling Technology, 9746), anti-cleaved Caspase-8 (Cell Signaling Technology, 8592), anti-Histone H3 (Cell Signaling Technology, 4499), anti-YAP (Proteintech Group, 13584-1-A), anti-TEAD (Abcam, ab197589), anti-MST1 (Abcam, ab124787), anti-Bcl-XL (Abcam, ab32370), and anti-GAPDH (Zhongshan Goldenbridge, TA-09, Beijing, China).

Cell proliferation was measured with Cell Counting Kit-8 (CCK-8) method (Dojindo Laboratories, Kumamoto, Japan) as previously described [20]. DLBCL cells (1 × 10⁴ cells/100 μl/well) given the designated treatment were seeded in 96-well plates. Thereafter, the cells were incubated with 10 μl/well CCK-8 for 4 h according to the manufacturer’s protocol. Cell proliferation was detected by measuring light absorption at 450 nm with the SpectraMax M2 Microplate Reader (Molecular Devices, CA, USA).

Both cell apoptosis assays and cell cycle assays were performed on Navios flow cytometer (Beckman Coulter, CA, USA). Cell apoptosis was detected by Annexin V-PE/7-AAD assay or Annexin V-FITC/ propidium iodide (PI) staining according to the manufacturer’s instructions (BD Biosciences, MA, USA). For cell cycle analysis, cells were washed with PBS, fixed with 70% ethanol overnight at 4 °C, and resuspended in PI/RNase staining solution (BD Biosciences). The percentage of cells in the indicated cell cycle phase was calculated with ModFit LT version 3.2 software.

RNA from YAP^−/− cells was prepared for RNA-sequencing (RNA-seq; three biological replicates for each group). RNA-seq experiments were performed by Novogene (Beijing, China). Briefly, total RNA was isolated with TRIzol reagent (Invitrogen, MA, USA). Sequencing libraries were generated following the manufacturer’s recommendations. After cluster generation, the library preparations were sequenced on an Illumina HiSeq platform. HTSeq v0.6.0 was applied to count the numbers of reads mapped to each gene, and the fragments per kilobase of transcript per million fragments mapped (FPKM) of each gene were calculated. Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) analyses were implemented using the cluster Profiler R package. A hierarchical clustering heat map was generated with the ggplot library.

Total RNA was extracted with RNAiso Plus (TaKaRa, Dalian, China) and reverse transcribed into cDNA with PrimeScript RT reagent kit (TaKaRa). Amplification reactions were conducted with SYBR Green (TaKaRa) on LightCycler 480II real-time PCR system (Roche). The primers for quantitative real-time PCR (qRT-PCR) are listed in Table S2. Real-time PCR for each sample was performed in triplicate. Relative quantification was determined by means of the 2^−ΔΔCT method with LightCycler 480 Gene Scanning version 1.5 software.

Cells given the designed treatment were seeded on glass slides in a liquid thin layer cell smear, fixed in 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, and then blocked with 5% normal goat serum for 1 h at room temperature. The slides were incubated with primary antibodies at 4 °C overnight, followed by incubation for 1 h at room temperature with Dylight 488-conjugated goat anti-rat IgG antibody (Abbkine, Beijing, China). The slides were washed with PBS and mounted with DAPI. Images were acquired with Pannoramic DESK Scanner (3DHistech Ltd., Budapest, Hungary) and viewed on CaseViewer version 2.3.

All animal experimental procedures were performed in accordance with protocols approved by the Institutional Animal Care and Research Advisory Committee of Shandong Provincial Hospital Affiliated to Shandong University. Specific pathogen-free (SPF)-grade 5-week-old female severe combined immunodeficiency (SCID) beige mice (n = 6 per group) were housed in individually ventilated cages. The in vivo tumor growth study was performed as previously described [20]. A total of 1 × 10⁷ LY1 cells resuspended in 100 μl PBS mixed with 100 μl Matrigel (Corning, MA, USA) were subcutaneously injected into the flanks of mice. Tumor size was measured with a digital caliper. For in vivo therapeutic studies with AG1024, SCID beige mice were injected subcutaneously with 1 × 10⁷ LY1 cells (resuspended in 100 μl PBS mixed with 100 μl Matrigel) in the left inferior limb. One week later, the mice were blindly randomized and treated with daily intraperitoneal injections of AG1024 (30 μg/day), or vehicle control for 10 days (n = 6 per group). Tumor dimensions were measured every 2 days, and tumor volumes were calculated using the equation V = (l × w²) × 0.5, where l is the largest dimension and w is the perpendicular diameter.

Data are represented as the mean ± standard deviation (SD) from at least three separate experiments. Differences between groups were analyzed by one-way analysis of variance (ANOVA) or t tests. Overall survival time was measured from the date of diagnosis to the date of death or last follow-up. Survival analyses were performed using the Kaplan-Meier method, and the log-rank test was used to identify significant differences. Univariate and multivariate analyses were performed using the Cox proportional-hazards regression model. All statistical analyses were performed with SPSS Statistics version 20.0 and GraphPad Prism version 6.0 statistical software. P < 0.05 was considered statistically significant.

To elucidate the potential role of YAP in human cancers, we first examined the expression of YAP in data from the Oncomine database [24]. YAP expression levels were upregulated (tumor versus normal) in 6 out of 29 lymphoma datasets using the threshold of > 2-fold change and p value < 0.0001 (Figure S1). We next analyzed the microarray datasets [25] obtained from the Oncomine database to illuminate the YAP mRNA transcriptional alterations between normal B cells and DLBCL samples. As shown in Fig. 1a, the mRNA level of YAP was significantly elevated in the DLBCL tissue samples (p < 0.01). To assess the protein expression level of YAP in DLBCL patients, YAP expression was detected by IHC in a cohort of DLBCL primary samples (n = 60) diagnosed at Shandong Provincial Hospital Affiliated to Shandong University. Compared to reactive lymphoid hyperplasia, DLBCL patients showed significantly higher levels of YAP (Fig. 1b). High YAP expression (YAP^high) was detected in 60% (36/60) of the DLBCL primary samples but only 23.3% (7/30) of the reactive lymphoid hyperplasia tissue samples (p = 0.001). Upregulation of YAP expression was validated in DLBCL cell lines. Consistently, the YAP expression level was significantly higher in human DLBCL cell lines than in normal B lymphocytes (Fig. 1c).

To address the clinical significance of YAP upregulation in DLBCL patients, the correlations between YAP expression and clinicopathological characteristics were analyzed. High levels of YAP expression were associated with B symptoms (p = 0.015), extranodal involvement (p = 0.023), and a high International Prognostic Index (IPI) score (p = 0.023) (Table 1), suggesting that upregulation of YAP expression was associated with DLBCL disease progression. Moreover, survival analysis of the enrolled patients revealed that higher expression of YAP was associated with a more aggressive disease process (p = 0.014) (Fig. 1d).

The above findings prompted us to further investigate the potential function of YAP in DLBCL. Functional enrichment analysis of YAP expression in DLBCL microarray profiles was performed. GO analysis indicated that YAP was closely related to cellular processes, biological regulation, and multicellular organismal processes (Fig. 1e). KEGG analysis revealed that YAP was enriched in pathways including cancer pathways and PI3K-Akt signaling (Fig. 1f). To validate the bioinformatics results, two lentivirus-mediated RNA interference vectors targeting YAP were used and exhibited remarkable silencing of YAP at the mRNA and protein levels in LY1, LY8, and LY3 cells (Fig. 1g, h), with shYAP#2 demonstrating higher efficacy. To examine the effect of YAP on the proliferation of DLBCL cells, a CCK-8 assay was performed. The results indicated that stable transfection of shYAP could significantly suppress the proliferation of DLBCL cells with high (LY1 and LY8) or low YAP expression (LY3) (p < 0.01) (Fig. 1i).

Flow cytometry was performed to evaluate the cell cycle distribution. The results demonstrated that knocking down of YAP expression induced a blockade in cell cycle progression at the G0/G1 phase (Fig. 1j, k). The apoptotic rates of DLBCL cells were also evaluated; however, no significant difference was detected after YAP knockdown. Therefore, silencing YAP restrained the growth of DLBCL cells predominantly through inhibition of cell proliferation and induction of cell cycle arrest.

To explore the function of YAP in DLBCL, we sought to examine the effect of YAP inhibition on DLBCL cells. VP, a Food and Drug Administration (FDA)-approved drug used as a photosensitizer for age-related macular degeneration [26], was shown to disturb the interaction between YAP and TEAD independent of light activation [27]. Accumulating evidence suggests that VP can inhibit YAP expression in several human malignancies [28, 29]. LY1, LY8, and LY3 cells were exposed to VP at the indicated concentrations for 24~96 h. We evaluated the effect of VP on cell viability in vitro and determined that the 50% inhibitory concentration (IC₅₀) at 24 h for LY1 (7.172 ± 0.628), LY8 (5.687 ± 0.491), and LY3 (5.601 ± 1.027) cells aligned with the VP activity previously reported in other hematological malignancies [30]. Incubation with VP decreased the proliferation of DLBCL cells in a dose- and time-dependent manner (Fig. 2a). LY1 and LY3 cells treated with VP for 24 h were also subjected to apoptosis analysis by flow cytometry. As shown in Fig. 2b, c, compared to untreated cells, VP-treated DLBCL cells displayed robust apoptosis.

We next examined whether VP-induced apoptosis in DLBCL cells occurs through the regulation of YAP. As expected, decreased expression levels of YAP and TEAD were induced by VP in dose-dependent manner in LY1 and LY3 cells (Fig. 2d), indicating that the proapoptotic effect of VP on DLBCL cells occurs by abrogating the expression of YAP and TEAD. In addition, we found that treatment of VP (8 μM) strongly restrained the mRNA expression of YAP target genes CTGF and CYR61 (Fig. 2e). Taken together, these results demonstrate that VP restrains cell proliferation and induces cell apoptosis by inhibiting YAP expression.

We further validated the involvement of YAP in DLBCL pathogenesis by deleting YAP with a CRISPR/Cas9 genome editing system in LY1 and LY3 cells. As indicated in Fig. 3a, three sgRNAs against YAP were designed. YAP deletion induced an increase in the proportion of DLBCL cells in the G0/G1 phase (p < 0.01) (Fig. 3b, c). Moreover, YAP^−/− cells showed a significant decrease in cell proliferation ability (Fig. 3d). We then tested the impact of YAP deletion on the response of DLBCL to a chemotherapeutic drug. Wild-type or YAP^−/− LY1 and LY3 cells were treated with doxorubicin, a chemotherapeutic reagent that is used to treat DLBCL. As shown in Fig. 3d–f, loss of YAP significantly enhanced the induction of growth inhibition and apoptosis by doxorubicin in these DLBCL cell lines.

A DLBCL xenograft mouse model was established with YAP^−/− LY1 cells to explore the effects of YAP inhibition in vivo. Compared to control tumors, tumors with YAP knocked out displayed a significantly reduced growth rate (Fig. 3g). Decreased expression levels of cell proliferation-related proteins, including Ki67 and c-myc, were also observed in tumors with YAP knocked out (Fig. 3h).

To further explore the mechanism underlying the transcriptional regulation of YAP in DLBCL, mRNA profiles were acquired by RNA-seq. A total of 528 mRNAs were shown to be significantly differentially expressed (DE) in YAP^−/− cells, with 364 downregulated and 164 upregulated. The most DE genes included RPL6 and RNF220 (Fig. 3i, j). Subsequent GO and KEGG pathway analyses revealed that the DE molecules were mainly related to cellular metabolic processes and carbon metabolism (Fig. 3k, Figure S2). We also detected changes in the expression of proteins involved in the proliferation-related mTOR and NF-κB pathways, such as LAMTOR1 and NFKBID (Figure S3), which was consistent with a previous study describing a multiprotein supercomplex involved in the control of oncogenic signaling in lymphoma [31].

IGF-1R, a receptor tyrosine kinase (RTK), is implicated in the development and progression of malignant tumors [32]. Our previous study indicates that Klotho plays an antitumor role by regulating IGF-1R signaling [20]. Therefore, we hypothesized that targeting IGF-1R may have potential therapeutic value in DLBCL. DLBCL cell lines were treated with two structurally unrelated IGF-1R inhibitors, AG1024, a member of the tyrphostins molecule class that specifically inhibit the autophosphorylation of the tyrosin residue on IGF-1R [33], and PPP, a cyclolignan alkaloid that specifically inhibits the activity and downregulates the cellular expression of IGF1R without interfering with the activities of other growth factor receptors [34]. Either of these inhibitors could suppress the proliferation of DLBCL cells in a concentration-dependent manner (Fig. 4a). Flow cytometry was performed to determine the effects of IGF-1R inhibitors on the apoptosis of DLBCL cells. The apoptosis rates exhibited concentration-dependent increases following exposure to the IGF-1R inhibitors (Fig. 4b, c). Decreased expression levels of Mcl-1 and Bcl-XL, as well as cleavage of the apoptotic markers Caspase-3 and Caspase-8, were observed in LY1 and LY8 cells treated with AG1024 (Fig. 4d). These results show the antitumor effect of IGF-1R inhibitors on DLBCL.

To obtain insights into the function of IGF-1R in DLBCL, we stably transfected DLBCL cell lines with either IGF-1R-specific shRNAs (shIGF-1R) or an empty vector control (shCon). Transfection of shIGF-1R resulted in a significant reduction in cellular proliferation (p < 0.01) (Figure S4). Silencing IGF-1R significantly promoted cell apoptosis and downregulated Mcl-1 expression in DLBCL cells (Figure S4). In addition, the YAP mRNA level was positively correlated with IGF-1R in DLBCL samples deposited in The Cancer Genome Atlas (TCGA) database [35] (r = 0.57, p < 0.001) (Fig. 4e). Interestingly, the IGF-1R mRNA level was lower in DLBCL tissue samples than in normal tissue samples in the TCGA database (Figure S4). To clarify this, we examined the level of the active form p-IGF-1R in normal B cells and DLBCL cells. As shown in Figure S4, the p-IGF-1R level was upregulated in the DLBCL cells. This indicates that the active IGF-1R level in DLBCL cells is higher than that in normal B cells.

We sought to explore the modulatory effect of IGF-1R on YAP expression in DLBCL. LY1 cells were incubated with various concentrations of IGF-1R inhibitors for 24 h. The results revealed that AG1024 and PPP significantly suppressed the phosphorylation of IGF-1R, which was accompanied by decreased YAP expression (Fig. 5a, b). To further validate the regulatory effect of IGF-1R on YAP expression, LY1 and LY8 cells were stably infected with either an shIGF-1R lentivirus or an shCon lentivirus. In accordance with the inhibitory effect of IGF-1R inhibitors, YAP expression was also significantly downregulated in LY1 and LY8 cells with IGF-1R knockdown (Fig. 5c).

The YAP level in the nucleus is modulated by the activity of Hippo signaling, and inactivated YAP is localized in the cytoplasm, where it is subsequently degraded. To further verify whether IGF-1R alters YAP activity by regulating the Hippo-YAP signaling pathway, we examined the effects of IGF-1R inhibitors on the subcellular expression of YAP in DLBCL cells by separating the nuclear and cytoplasmic proteins in DLBCL cells and detecting the subcellular expression of YAP protein. As shown in Fig. 5d, both the nuclear and cytoplasmic YAP levels were decreased, indicating increased inactivation and degradation of YAP following PPP treatment. This effect was confirmed by immunofluorescence assay (Fig. 5e, f). In addition, increased expression of MST1, a key protein in Hippo-YAP signaling, was observed in DLBCL cells treated with PPP or AG1024 (Figure S5). Enforced expression of MST1 was also detected in LY1 cells transfected with shIGF-1R (Figure S5). Collectively, these results indicate that IGF-1R depletion contributes to the modulation of Hippo-YAP signaling in DLBCL.

The above observations prompted us to confirm whether IGF-1 can rescue the decreased YAP expression caused by IGF-1R inhibition. Treatment with IGF-1 resulted in enhanced phosphorylation of IGF-1R in a dose-dependent manner (Fig. 6a). LY1 cells were serum starved for 24 h and treated with 50 ng/ml IGF-1 for 30 min. Western blot analysis was performed to evaluate the expression of YAP. As shown in Fig. 6b, IGF-1 induced the expression of YAP in LY1 cells. Additionally, IGF-1 rescued the decreased expression levels of YAP induced by AG1024 or PPP (Fig. 6c–f).

Based on these results, we sought to study the antitumor effect of AG1024 in DLBCL in vivo. A mouse DLBCL xenograft model was established and treated with AG1024. Compared to no treatment, AG1024 significantly inhibited tumor growth (Fig. 6g, h) and induced a decrease in Ki67 expression (Fig. 6i), indicating the great therapeutic potential of IGF-1R inhibitors in the treatment of DLBCL.

Overall, our results revealed the modulatory effect of IGF-1R on Hippo-YAP signaling in DLBCL (Fig. 6j).