ML-162 br sensitively detected in transformed ORGs but not
sensitively detected in transformed ORGs, but not wild ORGs. Higher GLuc levels were measured in medium of ORGKrasG12D/P53Del versus that of ORGKrasG12D (Fig. 3E). GLuc levels in transformed ORGs were highly
correlative with BIRC5 expression, as determined by Western blot (R = 0.97) (Fig. 3F and G). Therefore, the assay using the enhanced BIRC5-SPTSTA to drive the secreted reporter, GLuc, was highly sensitive and specific in reflecting BIRC5 expression levels in transformed orga-noids and PDAC ML-162 in vitro.
3.4. Dual reporter genes: serologic and optical imaging Gaussia luciferase and sr39TK microPET/CT imaging reporters can detect and localize minute PDAC
To test the sensitivity of GLuc assay, stably transfected Mia PaCa2CMV−GLuc and Capan2CMV−GLuc cells were studied. 100 PDAC
cells/ml of culture medium produced detectable levels of secreted GLuc and the signal increased proportionally with cell growth (Fig. 4A), suggesting remarkable sensitivity in vitro. To demonstrate that GLuc-2A-sr39TK could serve a dual-diagnostic role for both detection and
Fig. 3. Generation of BIRC5 super-promoter. BIRC5 synthetic promoters were designed based on the regulatory elements of proximate BIRC5 endogenous promoter
(A). The BIRC5-SP was validated using the GLuc assay and demonstrated markedly increased expression of GLuc compared to BIRC5-EP in PDAC cell lines with no expression seen in benign pancreatic cell lines (B). BIRC5-SP activity was further enhanced using two-step transcriptional amplification (BIRC5-SPTSTA), which resulted in 10-fold higher promoter activity than that of BIRC5-SP in PDAC cells, and no significant expression in benign HPPE cells (C). Transient transfection of HPPEKrasG12D, HPPEKrasG12D/P53Del and benign HPPEWT cells with BIRC5-SPTSTA-GLuc, BIRC5-SP-GLuc and CMV-GLuc revealed that BIRC5-SPTSTA-GLuc resulted in significantly greater secreted GLuc levels and specifically in both HPPEKrasG12D and HPPEKrasG12D/P53Del cells compared to BIRC5-SP-GLuc and CMV-GLuc (D). Wild-
type and transformed organoids were infected by lentivirus BIRC5-SPTSTA-GLuc. GLuc culture media levels and BIRC5 expression levels were determined by bio-luminescence assay western blot, respectively. Secreted GLuc levels (E) and BIRC5 expression levels (F) were each progressively higher in ORGKrasG12D, ORGKrasG12D/ P53Del, and MIA PaCa2 (positive controls), which were highly correlated (R = 0.97) (G). No secreted GLuc or expression of BIRC5 was observed in WT ORGs (E, F and G). Therefore, BIRC5-SPTSTA-driven GLuc secretion is a highly sensitive assay of BIRC5 expression in PDAC cell lines.
Fig. 4. Evaluation of activities of GLuc and sr39TK in GLuc-2A-sr39TK transfected cells. GLuc levels in cell culture medium of Mia PaCa2 and Capan-2 cells expressing GLuc demonstrates that 100 cells in one ml medium can be detected, thus GLuc is a highly sensitive serologic reporter
(A). To determine if dual reporter Gluc-2A-sr39TK aﬀects GLuc secretion or sr39TK's ac-tivity, transfection of Mia PaCa2 and Capan-2 cells with CMV-GLuc-2A-sr39TK versus CMV-GLuc or CMV-sr39TK was performed. Gluc-2A-sr39TK neither reduced the secretion of GLuc in Mia PaCa2 and Capan-2 cells (B), nor aﬀected TK's expression (C) and activity (D) in Mia PaCa2 cells, as determined by immuno-fluorescence assay (scale bar = 100 μm) and 18F-FHBG uptake assay, respectively. These in vitro data support the hypothesis that GLuc-2A-sr39TK can be used as dual diagnostic reporters to detect and localize PDAC.
Fig. 5. Xenograft tumor models were gen-erated by subcutaneous injection of Mia PaCa2 or patient-derived PDAC cell line (PDCL-15), both stably expressing GLuc-2A-sr39TK. 50ul of blood were collected on day 2 and 7 after inoculation of PDAC cells. Serum GLuc levels increased with tumor growth (A) and correlation between serum GLuc levels and minute tumor volume was highly significant (B; R = 0.93, p = 0.001). Optical imaging of PDAC tumors demon-strated strong signals in PDCL-15 and Mia PaCa2 tumors expressing GLuc-2A-sr39TK and no signal in control tumors (C) with a highly significant correlation of tumor vo-lume and bioluminescence emitted photons (D; R = 0.92; p = 0.0005). In a second study, PDCL-15 cells stably expressing GLuc-2A-sr39TK were implanted in the right flank and control patient derived PDAC cells (PDCL-15) were implanted in the left flank. In two weeks, [18F]FHBG microPET/CT was performed and demon-strated highly specific imaging of the minute PDCL-15 tumor with no imaging of the control cells and no background noise (E; orange circles), which correlated to bioluminescence imaging with coelenter-azine, (F; orange circles). However, stan-dard [18F]FDG microPET/CT imaging re-vealed non-specific imaging of the both PDCL-15 tumors expressing GLuc-2A-sr39TK and control PDCL-15 tumors with extensive non-specific background noise (G; orange circles). These data demonstrate that GLuc-2A-sr39TK is highly sensitive and specific to localize minute PDAC tumors using microPET/CT imaging, and support