Fig 2: Upregulation of GRK3 increased GAC cell growth and invasion, while downregulation of GRK3 inhibited malignant behaviors of tumor cells in vitro. A GRK3 expression levels in normal gastric epithelial cell lines (GES-1 and HFE145), GAC cell lines (AGS, MKN45, GT-5, KATO III, SNU-1, and SNU-16), and patient-derived ascites cells (GA0518, GA0804, GA0515, and GA0313) were determined by western blot. Quantification of fold change for each cell line relative to GES-1 normal gastric cells and ß-actin level was performed using Image J. B GRK3 was overexpressed in MKN45 cells by transducing GRK3 cDNA and confirmed by western blot (left) and qPCR (right). C Invasion assay was performed in MKN45-GRK3 overexpression cells compared to that of MKN45-GFP control cells; *** P < 0.0001. D Colony formation assay was performed in MKN45-GRK3 overexpression cells compared to that of MKN45-GFP control cells. E MKN45-GFP control and MKN45-GRK3 OE cells were cultured in media with a series of FBS % for 6 days. Cell survival and proliferation were measured by alamar blue viability assay. x-axis represents the series of FBS % in the culturing media. Values on y-axis are fold changes of growth for each % FBS relative to 0% FBS in either cell line. P values were calculated using the 2-sided Student’s t-test. ***: P < 0.005; **: P < 0.01. F GRK3 was overexpressed in KATO III cells by transfecting wild type (WT; GRK3 OE) or kinase-dead mutant form of GRK3 cDNA and confirmed by western blotting (left) and qPCR (right). G Tumor sphere formation assay was performed in KATO III GRK3 OE or mutant cDNA compared to that of KATO III-GFP control cells and quantification was quantified by image J analyses. *P < 0.01; ** P < 0.001. H GRK3 was KD in GA0518 PC cells using doxycycline inducible system and validated using western blot (left) and qPCR (right). I. GA0518 cells seeded in a 96-well plate were infected with a series of shRNA virus doses of shGRK3 #1 and shGRK3#2 and shScramble, from 20ul to 2.5ul per well. Cell survival and proliferation were measured by alamar blue viability assay 7 days after infection. X-axis represents the series of shRNA virus doses in the infections. Values on Y-axis are relative growth which was normalized to the scramble control shRNA for each virus dose for either cell line. P values were calculated using 2-sided Student’s t-test. ****: P < 0.0001; ***: P < 0.005; *: P < 0.05. J. Colony formation was determined in GA0518 cells with 2 independently GRK3 knockdown clones compared with control. **P < 0.001

Fig 3: The expression of GRK3 is upregulated in primary and metastatic GACs, which is associated with aggressive phenotypes and shorter survival. A Expression of GRK3 is shown in GAC tumor tissues compared with normal tissues from TCGA STAD and GTEX cohort (****P < 0.0001). B Expression of GRK3 is associated with poor survival of 876 GAC patients in KM plotter database (P = 1.2e-08). C Expression of GRK3 mRNA was measured by quantitative polymerase chain reaction (qPCR) and normalized to GAPDH in 23 pairs of primary GAC tissues and adjacent normal tissues. (**P < 0.001). D Immunohistochemistry (IHC) staining was performed using the GRK3 antibody in GAC tissue microarrays (TMAs) of primary tumor tissues and adjacent normal tissues. E Kaplan–Meier analysis of overall survival time according to GRK3 expression level in GAC TMAs. GAC patients with high GRK3 expression showed worse survival than patients with low expression did (P = 0.002). F&G The expression of GRK3 was analyzed with stage and lymph node metastases in our GAC cohort in TMAs. H GRK3 mRNA level was detected on normal tissues, primary tissues, and ascites cells by qPCR. **P < 0.001. I. GRK3 expression was detected using immunofluorescent staining in 22 cases of malignant ascites cells. GRK3/DAPI staining are shown from 4 representative PC samples with high GRK3 (upper panel); Dual staining of GRK3 (Alexa Fluor 555, red) and SOX9 (Alexa Fluor 488, green) are shown in another 4 representative PC samples with high GRK3 (bottom panel). Scale bar: 25µm

Fig 4: GRK3 promoted tumor growth, while inhibition of GRK3 by LD2 significantly suppressed GAC growth in xenografts. A Representative tumor images and tumor volume were shown in MKN45 GRK3 overexpression (OE) vs negative control group (NC) over three-week period after implantation. B-D Tumor burden, tumor weights and tumor volumes were shown at end point. E GRK3 level was detected by qPCR in treated tumor tissues compared to NC group. ***P < 0.001. F Co-immunofluorescent staining of GRK3 and SOX9 was performed in tumors of GRK3 OE group vs NC group. Representative images were shown. Scale bar: 25µm. G Representative GA0518 xenograft tumors in LD2 treatment group vs control were shown after 2-week treatment. H & I tumor weights (H) and tumor volumes (I) were significantly reduced in the LD2-treated group compared with that of the control group. *P < 0.01; **P < 0.001. J Mouse body weight was measured in the indicated group. K Immunofluorescent staining of GRK3, YAP1, SOX9 and proliferation marker Ki67 was performed in LD2 treated tumor tissues compared with control tumor tissues. Images were captured by confocal microscopy. Scale bar: 25µm. L Diagram demonstrates the link between GRK3 and YAP1 and its downstream targets. Novel GRK3 inhibitor LD2 can curtail tumor cell progression and metastases mediated by the GRK3-YAP1 axis

Fig 5: Identification of a novel GRK3 inhibitor LD2 that strongly suppressed the aggressive phenotypes of GAC cells in vitro. A The chemical structure of LD2 was shown. B&C The dose curves of LD2 effects in reducing kinase activities of GRK3 and its closest kinase GRK2. Two types of in vitro kinase assays were carried out to determine IC50 values for LD2 inhibition of GRK3 and GRK2: Invitrogen Z-LYTE and Promega ADP-Glo kinase assay. Plotted on y-axis is the % of remaining kinase activities of GRK3 and GRK2 upon treating with a series of LD2 doses. D LD2 suppressed tumor cell growth in GA0518, GA0804 and AGS cells treated with LD2 at the dosage as indicated. Cell survival was detected by MTS assay at 3 day and 6 day respectively. E& F LD2 suppressed colony formation (E), and cell invasion (F) in a dose-dependent manner. *P < 0.01; **P < 0.001. G Invasion assay was performed in KATO III-overexpression of GRK3 compared to that of KATO III-GFP control cells, and then treated with LD2 in indicated dosage for 48 h. Images of invasion (upper panel) and quantification (bottom panel) were shown. *P < 0.01; ** P < 0.001. H Invasion assay was performed in MKN45-overexpression of GRK3 cells compared to that of MKN45-GFP control cells, and then treated with LD2 in indicated dosage for 48 h. Images of invasion (upper panel) and quantification (bottom panel) were shown. *P < 0.01; ** P < 0.001