The Effects of Glypican-3 Deficiency on Radiosensitivity in Liver Cancer Cells

PURPOSE/OBJECTIVE(S)

Glypican-3 (GPC-3), a heparan sulfate proteoglycan involved in cellular proliferation, modulates signaling of FGF/FGFR, IGF/IGFR, HGF/Met, Wnt/Frizzled, among others and correlates with survival. GPC-3 is overexpressed in the majority of hepatocellular carcinoma and hepatoblastoma, but not in normal hepatocytes. Accordingly, it is being investigated as a liver cancer-selective target for radiopharmaceutical imaging and therapy. However, the potential linkage between GPC-3 expression and radiosensitivity has not yet been defined. In this study, we investigated the effects of GPC-3 deficiency on radiosensitivity in liver cancer cell lines.

MATERIALS/METHODS

CRISPR/Cas9 system was used to engineer GPC-3 knockout variants of liver cancer cell lines, HepG2 & Hep3B, both of which natively express GPC-3. Confirmation of knockout of GPC-3 was evaluated by RT-PCR, western blotting, flow cytometry, immunocytochemistry, and gDNA sequencing. Cell growth and migration were evaluated by BrdU insertion and wound-healing assays, respectively. In vitro radiosensitivity was examined by radiation-induced apoptosis/necrosis (Annexin V-APC and PI staining), cell cycle modification, γH2AX foci formation, and clonogenic assays (6 Gy). Wildtype and knockout lines were engrafted into athymic mice to assess tumor growth kinetics.

RESULTS

RT-PCR, western blotting, flow cytometry, and immunocytochemistry all confirmed GPC-3 knockout in both HepG2 and Hep3B cell lines. Nucleotide deletion at exon 3 of the GPC-3 gene was confirmed by gDNA sequencing in HepG2ΔGPC3 and Hep3BΔGPC3. GPC-3 deficiency reduced liver cancer cell proliferation (HepG2ΔGPC3, p = 0.027, and Hep3BΔGPC3, p = 0.031) and migration (HepG2ΔGPC3: 1.5-fold, p<0.001, and Hep3BΔGPC3: 2.3-fold, p<0.001) significantly when compared with wild type. GPC-3 deficiency reduced cell survival and clonogenicity (HepG2ΔGPC3: DEF = 1.23, Hep3BΔGPC3: DEF = 1.23) in liver cancer cells exposed to irradiation (6 Gy). The delayed repair of double-stranded DNA damage was observed in irradiated GPC-3 deficient liver cancer cells. Tumor growth was dramatically delayed by GPC-3 deficiency. Tumor weight measured at 50 (Hep3B) and 60 (HepG2) days after liver cancer cell inoculation corroborated these effects.

CONCLUSION

Knockout lines of HepG2 and Hep3B exhibited decreased cell proliferation, migration, and in vivo tumor growth compared to wildtype. GPC-3 deficiency was associated with increased sensitivity to radiation therapy. Studies identifying the pathways through which this radiosensitivity is mediated are ongoing.

Nanobody-Based ImmunoPET for Hepatocellular Carcinoma

PURPOSE/OBJECTIVE(S)

HCC accounts for 75-90% of all primary liver cancers, the majority of which are treated with liver-directed therapy. Treatment response and recurrence are difficult to discern using conventional imaging with MR/CT. Tumor-selective PET imaging could help with clinical management in this setting. Here, we engineer HN3, a single-domain antibody (nanobody) specific to GPC3, a histopathologically-defining HCC marker, as an immunoPET agent. We compared both conventional and sortase-based site-specific modification methods for synthesizing HN3 immunoPET tracers.

MATERIALS/METHODS

Stochastic lysine conjugation with deferoxamine (DFO-NCS) was done to synthesize nHN3-DFO. ssHN3-DFO was engineered utilizing sortase-mediated conjugation of HN3 containing an LPETG C-terminal tag and a triglycine-DFO chelator. Biolayer interferometry (BLI) and radioligand saturation assays were done to determine binding affinity pre- and post-Zirconium-89 labeling. Following, PET/CT with a terminal 3-hour biodistribution was done in mice inoculated with isogenic A431 and A431-GPC3+ xenografts to determine conjugate specificity for GPC3. Finally, conjugates were evaluated in a HepG2 liver cancer model via ex vivo biodistribution studies and a comparative PET/CT study in mice bearing HepG2 tumors that were imaged with both [18F]FDG and 89Zr-ssHN3.

RESULTS

Both conjugates exhibited nanomolar binding affinity for GPC3 in vitro (11-30 nM for nHN3 and 10-15 nM for ssHN3). A431 and A431-GPC3+ PET/CT and biodistribution studies showed specificity to GPC3 by both probes, with more favorable tumor uptake by 89Zr-ssHN3 at 3 hours post-injection (14% IA/g vs. 7% IA/g for nHN3). Both tracers also displayed uptake in HepG2 (GPC3+) liver tumors, again with the site specifically conjugated probe having higher tumor accumulation and lower liver signal than the conventionally modified HN3 (7% IA/g vs. 5 % IA/g for tumor and 2% IA/g vs. 4% IA/g for liver at 1-hour post-injection). PET/CT studies in mice imaged with [18F]FDG and 89Zr-ssHN3 demonstrated more consistent tumor accumulation for the nanobody conjugate (4/4 mice had uptake by the tumor vs. 1/4 for FDG).

CONCLUSION

We successfully designed, synthesized, and characterized novel GPC3-selective nanobody PET probes that can image liver tumors in vivo. The site-specifically conjugated tracer showed more favorable biodistribution and pharmacokinetic properties, resulting in a much higher tumor: liver signal compared to 89Zr-nHN3. We also show the superiority of the 89Zr-ssHN3 imaging over conventional [18F]FDG, highlighting a clear advantage in using targeted tumor imaging for this cancer type. Successful translation of the site-specifically conjugated nanobody may ultimately aid in characterizing lesions following liver-directed therapy and allow for more comprehensive screening, early diagnosis, and post-treatment surveillance of HCC.

Identification of Hepatocellular Carcinoma-Specific Targets for Imaging and Therapy

PURPOSE/OBJECTIVE(S)

Despite advances in the diagnosis and treatment of hepatocellular carcinoma (HCC), there are no tumor-selective agents that are clinically approved in the US for this disease. Here, we aimed to identify and validate molecules that are overexpressed on HCC plasma membrane compared to normal tissues, which could be facilitate the design, engineering, and testing of tumor-selective imaging and therapeutic agents.

MATERIALS/METHODS

We analyzed next-generation sequencing (NGS) public datasets (TCGA and TIGER-LC) and NCI single cell RNA-sequencing datasets to identify overexpressed plasma membrane molecules and aimed to validate these targets using immunohistochemical staining (IHC) of patient tissue microarrays (TMAs), and flow cytometry using liver cancer cell lines (Huh7, HepG2, and Hep3B).

RESULTS

NGS data identified GPC3, EGFR, MET, MUC13, and ROBO1 molecules overexpressed in HCC relative to non-tumor tissues. In HepG2 cell line, EGFR (p<0.05) and MET (p<0.01) demonstrated statistically significant increased median fluorescence intensity (MFI) relative to controls in flow cytometry. In the Hep3B cell line, MET, GPC3, and EGFR demonstrated an increased MFI relative to the control (p<0.01). No statistically significant difference was observed in Huh7 cell lines. IHC staining of TMAs for GPC3, MET, MUC14, and ROBO1 showed statistically significantly higher staining relative to the normal tumor tissue (p<0.001).

CONCLUSION

We identified and validated plasma membrane molecules overexpressed in HCC compared to non-tumor tissue. Because GPC3, a well-known HCC-specific marker that is expressed in 75% of HCC, was identified using our approach, we are confident that that additional molecules may also represent promising HCC-selective targets. This work could facilitate the design, engineering, and testing of novel precision oncology imaging and therapeutic agents for HCC.