Activation and function of receptor tyrosine kinases in human clear cell renal cell carcinomas

The renal tissue specimens were collected in compliance with local ethics regulations at the Department of Urology, Xin Hua Hospital Affiliated to Shanghai Jiao Tong University School of Medicine, China. The 10 ccRCC patients were five males and five females (Table 1). The mean age at diagnosis was 65 ± 9. The patient information of 3 other kidney cancer samples and 1 benign renal tumor sample are in Table 2. After surgical resection, tissue samples were lysed in lysis buffer (R&D Sytems, AYR001B) for protein lysates on the ice or fixed in neutral buffered formalin (10%) for histology staining, or immediately processed to establish ccRCC patient-derived xenograft models in nude mice.

786–0(CRL-1932), A-498(HTB-44), ACHN(CRL-1611), and Caki-1(HTB-46) cell lines were obtained from ATCC. 786–0 and Caki-1 cell lines were derived from human primary ccRCC. A-498 and ACHN cell lines were derived from human primary papillary RCCs. 786–0 and ACHN cells were cultured in RPMI 1640 Medium (Gibco) with 10% FBS (Gibco). A498 cells were cultured in Dulbecco’s Modification of Eagle’s Medium (Gibco) with 10% FBS. Caki-1 cells were cultured in McCoy’s 5A Medium (Sigma) with 10% FBS.

Fixed tissues were dehydrated using grades of ethanol (70, 80, 90, 95, and 100%). Dehydration was followed by clearing the samples in two changes of xylene. The samples were then impregnated with two changes of molten paraffin wax, embedded, and blocked out. The tissue sections (7 μm) were stained with hematoxylin-eosin by standard procedures. Stained sections were observed and photographs were taken using a Leica microscope.

Human phospho-RTK arrays (R&D Systems, AYR001B) were used according to the manufacturer’s instructions. Briefly, a total of 5 mg protein lysates of in vitro cultured cells, or 10 mg protein lysates of clinical samples and mouse xenografts were diluted in the kit-specific dilution buffer and incubated with blocked membranes overnight. The membranes were washed and incubated with anti-phospho-tyrosine-HRP antibody for 2 h. The membranes were washed and exposed to chemiluminescent reagent. The arrays were photographed using Image Station 4000MM PRO system (Carestream). The pixel densities of various spots were collected and quantified with its software. The average signal (pixel density) of the pair of duplicate spots was determined for each RTK. A signal from the PBS negative control spots was used as a background value. And signals of reference spots in the corners were used for normalization among different arrays. Phospho-RTK relative value was calculated according to the following formula: Phospho-RTKx relative value = (INTx-INTnc)/(INTref-INTnc). INTx is the pixel density of RTKx, INTnc is the pixel density of background,and INTref is the density of reference spots.

Proteins were separated by SDS-PAGE and transferred to a nitrocellulose membrane. The membrane was blocked in TBS containing 0.1% Tween 20 (TBST) and 5% nonfat milk for 1 h at room temperature and then incubated overnight in TBST containing 5% bovine serum albumin and primary antibodies. Membranes were then washed with TBST and incubated with horseradish peroxidase-conjugated secondary antibody for 1 h, and immune complexes were detected by immobilon Western chemiluminescent HRP substrate (WBKLS0500, Millipore). Primary antibodies are anti-phospho-EGFR (#3777), anti-EGFR (#4267), anti-phospho-PDGFRβ (#3161), anti-PDGFRβ (#3169), anti-phospho-InsulinRβ (#3024), anti-InsulinRβ (#3025), anti-phospho-VEGFR2 (#2474), anti-VEGFR2 (#9698), anti-phospho-Met (#3077), anti-Met (#3148), anti-phospho-Akt (#4060), anti-phospho-Erk1/2 (#4370). All antibodies were purchased from Cell Signaling Technology. The membranes were photographed using Azure Biosystems (c300) and were quantified using its software (Analysis Toolbox). The density ratio of interest proteins to GAPDH or β-Actin were calculated.

The female BALB/c nude (nu/nu) mice were purchased from Beijing Vital River Laboratory Animal Technology Co., Ltd. and used for implantation at the age of 6–8 weeks. They were maintained under specific pathogen-free conditions, and food and water were supplied ad libitum. Housing and all procedures were performed according to protocols approved by the Ethics Committee of Shanghai institute of materia medica. Subcutaneous xenografts were established by injection of 5× 10⁶ cells or one piece (2mm³) tumor per mouse to right flank. Tumor formation was monitored each week. Each subcutaneous tumor was measured using a caliper, and tumor volumes were calculated as follows: 0.5× length× width². Nude mice with ccRCC patient-derived xenografts of approximately 100 mm³ were allocated randomly into 4 experimental groups and orally treated with 3 mg/kg/d Crizotinib (n = 6), 30 mg/kg/d Lapatinib (n = 6), combination of Crizotinib and Lapatinib(n = 6), or vehicle (n = 6) for 21 days. Mice were euthanized in a CO₂ chamber for 2 h after the last treatment. Crizotinib and Lapatinib were purchased from Selleck Chemicals.

Cryosections were blocked in PBS containing 5% normal donkey serum for 2 h at room temperature. Sections were incubated over night at 4 °C with the primary antibodies against PDGFRβ (ab32570, rabbit Anti-PDGF Receptor beta antibody, 1:50, Abcam), Pan-Keratin (#4545, mouse anti-pan-keratin antibody,1:50, CST), Vimentin (sc-7557, goat anti-vimentin antibody, 1:50, Santa Cruz). After washed with PBS three times, the sections were incubated for 1 h at room temperature with Alexa Fluor 594-labeled donkey anti-rabbit IgG (A21207,1:400, Invitrogen), Alexa Fluor 488-labeled donkey anti-mouse IgG (A21202,1:400, Invitrogen) and Alexa Fluor 555-labeled rabbit anti-goat IgG (A21431,1:400, Invitrogen). Sections were washed three times in PBS, followed by mounting tissue with Dako fluorescence mounting medium. Photographs were taken using a Leica DMi8.

Data were represented as mean ± SEM. T test was used in human phospho-RTK studies. Two-way ANOVA with Tukey post hoc test was used in mouse xenograft treatment studies. Statistical significance was established for P < 0.05, P < 0.01, and P < 0.001.

To examine the histopathology of the kidney tumors, HE staining was performed. Gross examination of the resected tumor samples revealed that the ccRCCs were all bright yellow in color, due to their intracellular lipid accumulation (Fig. 1a). In contrast, the adjacent normal tissues of the ccRCCs showed normal flesh color (Fig. 1b). In HE staining sections, the ccRCC cells showed transparent and empty (water clear) cytoplasm with well-defined cell borders (Fig. 1c). The nuclei of ccRCCs were round. Architecturally, the ccRCCs displayed compact-alveolar or acinar growth patterns. The small nests were surrounded by a well-developed network of thin-walled vessels. An abundance of extravasated red blood cells were observed in the tumors. The glomerulus, proximal convoluted tubules, and distal convoluted tubules in the cortex of the kidney could be seen in adjacent tissues (Fig. 1d).

To understand the expression and phosphorylation of the RTKs in the ccRCCs, we analyzed 10 pairs of primary ccRCCs and their adjacent non-tumor kidney tissues using human phospho-RTK arrays which evaluate the relative phosphorylation levels of 49 receptor tyrosine kinases (Additional file 1: Fig. S1). 9 RTKs (EGFR1–3, Insulin R, PDGFRβ, VEGFR1, VEGFR2, HGFR, and M-CSFR) were found to be phosphorylated in the ccRCC samples (Fig. 2 and Fig. 3). Comparing to their adjacent normal tissues, Insulin R, HGFR, PDGFRβ, M-CSFR, VEGFR1, and VEGFR2 were specific to the ccRCCs. Among them, the phosphorylation levels of Insulin R, PDGFRβ, VEGFR1, and VEGFR2 were significantly increased in all the ccRCC samples. The phosphorylation levels of HGFR (spot #5) and M-CSFR (spot #7) varied among the samples. HGFR was highly phosphorylated in RE0370 and RE0410 samples while M-CSFR was highly phosphorylated in RE0370, RE0440, and RE0450 samples. This RTKs activation patterns of ccRCCs were different from that of their paired adjacent tissues in which only the EGFR family members, particularly EGFR and ErbB4, were significantly phosphorylated. These findings were further verified by Western blotting analyses. The phosphorylation levels of Insulin Rβ (Tyr1150/1151), PDGFRβ (Tyr751), VEGFR2 (Tyr996), and HGFR (Met Tyr1234/1235) were found to be increased in the tumor tissues in comparison to the paired adjacent tissues (Fig. 4). In addition, the protein levels of some of the RTKs (Insulin Rβ, PDGFRβ, VEGFR2, or Met) were also increased in certain tumors. The protein expression patterns of PDGFRβ and VEGFR2 in tumors were also different from their adjacent tissues (Fig. 4a, d).

To determine whether the RTK phosphorylation patterns in the ccRCCs are specific, we evaluated the RTK phosphorylation patterns in 2 ccRCC cell lines, 2 papillary RCC cell lines and 4 other types of kidney tumor samples. The RTK phosphorylation patterns of the four human RCC cell lines were similar with each other (Fig. 5). The EGFR family and HGFR were highly phosphorylated in all the four cell lines. In contrast, the RTK phosphorylation patterns of the four other types of tumor samples, namely a papillary RCC (RE0020), an oncocytoma (RE0150), a renal pelvic carcinoma (RE0210), and a cystic nephroma (RE0500), were different from each other and were also different from that of the ccRCCs, except EGFR, which was highly phosphorylated in all samples (Fig.6). ErbB4, Insulin R, and IGF-1R were phosphorylated in the papillary RCC (RE0020), Mer (Axl family) was phosphorylated in the oncocytoma (RE0150), and HGFR, PDGFRα, and PDGFRβ were phosphorylated in the renal pelvic carcinoma (RE0210, Fig.6). In the benign renal tumor, namely the cystic nephroma sample (RE0500), only EGFR was phosphorylated (Fig.6). These data demonstrated that the RTK phosphorylation patterns of the ccRCCs were specific.

In order to treat the tumors with tyrosine kinase inhibitors based on their RTK phosphorylation patterns, we tried to establish tumor xenograft models using the patient-derived tumor samples as well as the cancer cell lines. Thirty-five tissue pieces from the 10 samples of the ccRCCs were subcutaneously implanted into 35 nude mice. Only one xenograft (RE0410) grew successfully. We then analyzed the RTK phosphorylation pattern of this ccRCC explant. The RTK phosphorylation pattern of the xenograft was different from its original primary sample (RE0410). Only the phosphorylation of EGFR family (EGFR, ErbB2 and ErbB3) and HGFR were maintained at high levels while that of the other RTKs decreased (Fig.7a). In contrast to the poor tumorigenicity of the ccRCC samples from patients, the established cell lines of ccRCC and papillary RCC were highly tumorigenic. Both EGFR and HGFR remained phosphorylated in all four of the cell line-derived xenograft samples, although their phosphorylation levels decreased in vivo (Fig.7b, c). These data demonstrated that the RTK phosphorylation patterns in the xenografts changed and the success rate of subcutaneous grafting of ccRCC samples was low in nude mice.

Phospho-RTK array of the ccRCC explants from the xenograft mice showed that three of the EGFR family members and the HGFR were highly phosphorylated in the xenograft tumors. We therefore used the RTK inhibitors targeting EGFR family and HGFR to treat the ccRCC xenograft nude mice. As shown in Fig. 8a, the change of body weight in treatment groups was similar with that in vehicle group. The EGFR inhibitor lapatinib or the HGFR inhibitor crizotinib alone slightly inhibited the tumor growth (Fig.8b). In comparison, the combination of the two inhibitors was much more efficient than the single treatment to inhibit the tumor growth (Fig. 8b). The average inhibition rate of crizotinib, lapatinib, or a combination of them on the ccRCC were 38.24 ± 22.40%, 35.43 ± 37.15%, and 62.79 ± 21.95% respectively (Fig. 8c, d).

To understand the effects of the combination treatment at the molecular level, we examined the effects of crizotinib, lapatinib, or the combination of them on the phosphorylation/activation of their target proteins and their downstream signaling molecules Erk1/2 and Akt. As shown in Fig. 8e and f, the combination treatment synergistically inhibited the phosphorylation of Met, EGFR, and Erk1/2. These data suggested that a combination treatment of the RTK inhibitors based on the RTK phosphorylation patterns synergistically inhibited the RTK-mediated signaling and the tumor growth.

PDGFRs are usually expressed in stroma cells. To understand the function of the PDGFRβ in the ccRCCs, we analyzed the expression of PDGFRβ in the patient-derived ccRCCs and their adjacent tissues. The PDGFRβ was mainly expressed in glomerulus in the tumor adjacent tissues (Fig. 9a). In the ccRCC tumor tissues, PDGFRβ was present in the vimentin-positive stroma cells surrounding the tumor islands and blood vessels (Fig. 9b, c). But the keratin-positive epithelial cells were mainly localized in the tumor islands which were PDGFRβ-negative (Fig. 9b, c). These results suggest that the PDGFRβ expressing cells were periepithelial stroma cells in the ccRCCs.