Using Lipid Nanoparticles to Deliver RNAi Against GIPC1 May Overcome Drug Resistance in Pancreatic Cancer

Dulbecco's Modified Eagle Medium (DMEM; Gibco, Waltham, MA, USA), RPMI-1640 medium (Gibco, Waltham, MA, USA), and DMEM/F-12 medium (Gibco) supplemented with fetal bovine serum (FBS; tetracycline-free; Gibco, Waltham, MA, USA), Antibiotic–Antimycotic (Gibco), and Plasmocin (InvivoGen, San Diego, CA, USA) were used for cell culture. Antibodies included anti-β-actin and HRP-conjugated secondary antibodies (Santa Cruz Biotechnology, Dallas, TX, USA) and anti-GIPC antibody (Thermo Fisher Scientific, Waltham, MA, USA). Tet-On–inducible GIPC1 shRNA plasmids were purchased from Dharmacon, Lafayette, CO, USA. Puromycin (2 μg/mL; Sigma-Aldrich, St. Louis, MO, USA) and doxycycline (0.4 μg/mL for in vitro and 0.5 mg/mL for in vivo; Sigma-Aldrich, St. Louis, MO, USA) were used for selection and induction, respectively. Cell viability was assessed using the CellTiter-Glo^® 2.0 Assay (Promega, Madison, WI, USA). For protein extraction and quantification, NP-40 lysis buffer, protease inhibitor cocktail, and Halt phosphatase inhibitor cocktail (Thermo Fisher Scientific, Waltham, MA, USA) were employed, and protein concentrations were determined using the Pierce™ BCA Protein Assay Kit (Thermo Fisher, Waltham, MA, USA, Cat. #23225). For Western blotting, 6× Laemmli SDS sample buffer (Bio-Rad, Hercules, CA, USA), PVDF membranes (Millipore, Burlington, MA, USA), and SuperSignal™ West Pico Chemiluminescent Substrate (Thermo Scientific, Waltham, MA, USA) were used. Liposomal formulations were prepared with KRC18 (custom lipid), DOPC, DOPE, and DSPE-PEG2000-Amine (Avanti Polar Lipids, Alabaster, AL, USA), cholesterol (Sigma-Aldrich, St. Louis, MO, USA), TTLP (custom tumor-targeting ligand peptide), and Rhodamine-PE (Avanti Polar Lipids, Alabaster, AL, USA). GIPC1 siRNA was custom synthesized (Dharmacon, Lafayette, CO, USA). Agarose gels and ethidium bromide (Bio-Rad, Hercules, CA, USA) were used for gel retardation assays. Hydrodynamic diameter and zeta potential were measured using a Zetasizer (Malvern Instruments, Malvern, UK, Zetasizer Software version v7.13), and Milli-Q water (Millipore, Burlington, MA, USA) was used for formulation stability studies. For in vivo imaging, sodium D-Luciferin (Gold Biotechnology, St. Louis, MO, USA, Cat. #LUCK-100) was administered, and animals were imaged using the IVIS Imaging System (Waltham, MA, USA) with Living Image Software (IVIS Living Image Software version v4.7). For RT-qPCR, TRIzol™ Universal Reagent (Thermo Fisher Scientific, Waltham, MA, USA), iScript™ cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA, Gol Cat. #1708890), and SYBR^® Green Premix Ex Taq™ II (Tli RNaseH Plus; Takara, San Jose, CA, USA, Cat. #RR820A) were used, and amplification was performed on an Applied Biosystems QuantStudio™ 5 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA).

Human pancreatic cancer cell lines PANC-1 and AsPC-1 were obtained from the American Type Culture Collection (ATCC), whereas patient-derived xenograft (PDX) cell lines, including 6741, 4636, 4535, and 4911, were established from primary tumor specimens collected from pancreatic cancer patients at the Mayo Clinic. PANC-1 cells were maintained in Dulbecco's Modified Eagle Medium (DMEM, Gibco, Waltham, MA, USA), and AsPC-1 cells were maintained in RPMI-1640 medium. Primary pancreatic cancer cells (6741, 4636, 4535, and 4911) were maintained in DMEM/F-12. All cell culture media were supplemented with 10% fetal bovine serum (FBS); tetracycline-free FBS (10%) was specifically used to prepare the medium for maintaining GIPC-1–inducible cell lines. In addition, media were supplemented with 1% anti-anti (penicillin–streptomycin, Gibco, Waltham, MA, USA) to prevent bacterial contamination and 0.02% plasmocin (InvivoGen, San Diego, CA, USA) to prevent mycoplasma contamination.

Six- to eight-week-old SCID mice were obtained from the in-house breeding colony and maintained in the institutional animal facility. All animal experiments were performed with the highest standards of care in compliance with the guidelines of the Association for Assessment and Accreditation of Laboratory Animal Care (AAALAC) and under protocols approved (Protocol A00004002-18) by the Mayo Clinic Institutional Animal Care and Use Committee (IACUC), further ensuring the ethical conduct of our research.

Antibodies for β-actin and horseradish peroxidase-conjugated secondary antibodies were purchased from Santa Cruz Biotechnology. Anti-GIPC antibody was purchased from Thermo Fisher, Waltham, MA, USA.

The plasmids for Tet-on-inducible GIPC1 shRNA were purchased from Dharmacon, Lafayette, CO, USA. Lentivirus stocks for GIPC1 shRNA were prepared and infected into the target cells following standard procedures. For antibiotic selection, 2 μg/mL of puromycin was added to the medium. For the Tet-On-inducible GIPC shRNA, 0.4 μg/mL of doxycycline was used to induce shRNA expression in the stably infected cells.

Approximately 5 × 10³ cells were seeded in 384-well plates. After 18–24 h, cells were treated with increasing concentrations of GEM and incubated for a further 72 h. Cell viabilities were assayed using the CellTiter-Glo^® 2.0 Assay (Promega, Madison, WI, USA) as per the manufacturer's protocol. Briefly, 24 μL of reagent was added to each well. The plates were incubated for 10 min at RT, and luminescence signals were measured using Spectramax i3x (Molecular Devices, LLC (San Jose, CA, USA), version 7.x. Percentage viability is calculated as follows: Viability (%) = 100 × (A_Treated − A_Blank)/(A_Untreated − A_Blank).

For in vitro studies, cells were washed three times with ice-cold PBS and lysed using NP-40 lysis buffer supplemented with 1% protease inhibitor cocktail and 1% Halt phosphatase inhibitor cocktail (Pierce, Rockford, IL, USA ). The lysates were centrifuged at 13,000 rpm for 20 min at 4 °C, and the supernatants were collected. Protein concentrations were quantified using the BCA assay (Pierce BCA Protein Assay Kit, Waltham, MA, USA).

For in vivo samples, proteins were denatured by adding 6× Laemmli SDS sample buffer and heating at 95 °C for 5 min. Equal amounts of protein were loaded into each lane for SDS-PAGE, followed by wet transfer onto PVDF membranes. Membranes were blocked in TBS-T buffer (50 mmol/L Tris-HCl, pH 7.4; 150 mmol/L NaCl; 0.05% Tween-20) containing 5% nonfat milk or BSA. Membranes were then incubated overnight at 4 °C with the primary antibody diluted in TBS-T containing 5% nonfat milk or BSA, followed by incubation for 1 h at room temperature with horseradish peroxidase (HRP)-conjugated secondary antibody (Santa Cruz Biotechnology, Dallas, TX, USA). Protein bands were visualized using the SuperSignal West Pico Chemiluminescent Substrate (Thermo Scientific, Waltham, MA, USA).

Six- to eight-week-old male SCID mice were obtained from in-house breeding and housed in the institutional animal facilities. All animal work was performed under protocols approved by the Mayo Clinic Institutional Animal Care and Use Committee. To establish orthotopic pancreatic tumors, approximately 1 × 10⁶ luciferase-labeled PANC-1 cells resuspended in 100 μL of PBS were injected into the pancreas of each mouse. Tumors were allowed to grow for 4 weeks without treatment. LGIPCsi loaded with IR-780 dye and complexed with FAM-siRNA was administered via the intravenous route. Mice were imaged using the IVIS imager 24 and 48 h after administration.

A single mouse trial (SMT) was used to assess the in vivo tumor inhibition study in combination with GIPC-1 knockdown plus GEM treatment. This strategy is currently being used in 'Avatar' models for the development of personalized therapy for various diseases. These 'Avatar' models are developed by implanting patient tumor samples in mice for subsequent drug efficacy studies. Each tumor-bearing mouse is treated with a different therapeutic regimen to identify the most effective regimen, thereby reducing the cost and toxicity associated with non-targeted therapy. We used the same strategy in our SMT. All experiments were accomplished according to the permitted guidelines. For the pancreatic cancer orthotopic model, 6-8-week-old female SCID mice were obtained from in-house breeding. Approximately 1 × 10⁶ luciferase-labeled Tet-inducible GIPC-1 shRNA vector-transfected AsPC-1 or PANC-1 cells, resuspended in 100 μL of PBS, were injected into the pancreas of each mouse. After three days of cell inoculation, GIPC-1 knockdown was induced by adding doxycycline (0.5 mg/mL) in drinking water for the remainder of this study. After substantial tumor growth was confirmed by bioluminescence imaging, mice were treated with GEM (10 mg/kg) twice a week for four weeks. After completion of the experiment, all tumor-bearing mice were euthanized with CO₂; tumors were harvested, and tumor volumes and weights were measured. We also performed a validation study to confirm the results obtained from the SMT in AsPC-1 tumor-bearing mice with 5 mice per group.

LGIPCsi lipid nanoparticles were prepared using the thin film hydration method. LGIPCsi consisted of KRC18, DOPC, DOPE, cholesterol, DSPE-(PEG-2000)-amine, and TTLP in a weight ratio of 0.5:3.98:0.35:0.484:0.27:0.22. For fluorescent labeling, 50 µg of Rh-PE was added. Ingredients were dissolved in chloroform, and the solvents were evaporated with nitrogen gas and dried under a high vacuum for 4 h. The film was hydrated for 12 h with water or siRNA solution. For LGIPCsi, the film was hydrated with GIPC-siRNA pre-condensed with KRC18. The suspensions were vortexed, sonicated, and subjected to 8 freeze/thaw cycles. Sequential extrusions through 200 nm and 100 nm pore-size membranes yielded 100–150 nm unilamellar liposomes.

The siRNA binding ability of LGIPCsi was assessed by gel retardation assay on a 1.2% agarose gel pre-stained with ethidium bromide. The LGIPCsi complex was incubated with 2% SDS at room temperature for 20–25 min. Then, 2 µL of 6X loading buffer was added, and the solution was loaded into the wells. Samples were electrophoresed at 80 V for 45 min, and siRNA bands were visualized using a Gel Documentation System (Bio-Rad, Hercules, CA, USA).

HDD and zeta potentials (1:100 dilution) were measured in triplicate by using a Zetasizer (Malvern, UK). For stability studies, liposomes were incubated in autoclaved Milli-Q water at 4 °C for up to 30 days, with size measurements taken at various intervals.

To harness the therapeutic efficacy of GIPC-1 knockdown in vivo, we developed a novel liposomal siRNA delivery system with a tumor-targeting ligand (LGIPCsi). Luciferase-labeled PANC-1 cells were used to develop orthotopic PDAC xenografts in 6-8-week-old female SCID mice following a similar procedure described above. After 4 weeks, tumor growth was confirmed by bioluminescence imaging, and tumor-bearing mice were divided into five groups (n = 5). The mice were i.p. Administered with: LNPs, LGIPCsi (0.5 mg/kg) complexed with LGIPCsi, GEM (10 mg/kg), and a combination of both twice a week for three weeks. The untreated group was used as a control. The LGIPCsi and GEM treatments were performed two days apart. After completing the experiment, all tumor-bearing mice were euthanized with CO₂; tumors were harvested for morphologic analysis and immunostaining.

Tumor growth of orthotopic tumors was analyzed by bioluminescent imaging. Sodium-D-Luciferin (Gold Biotechnology, St. Louis, MO, USA) was injected intraperitoneally at 100 mg/kg in PBS 5 min before imaging. Mice were anesthetized using isoflurane and imaged with the IVIS system, Xenogen, Waltham, MA, USA. Signals were displayed as photons/second/cm²/steradian and were measured using the Living Image software (IVIS Living Image Software, v4.7).

Total RNA was isolated from tumor samples subjected to different treatments using the TRIzol™ Universal Reagent (Thermo Fisher Scientific, Waltham, MA, USA) according to the manufacturer's protocol. Complementary DNA (cDNA) was synthesized from purified RNA using the iScript™ cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA RR037). Quantitative real-time PCR was performed with SYBR^® Green Premix Ex Taq™ II (Tli RNaseH Plus, RR820A; Takara, San Jose, CA, USA) following the supplier's instructions. Gene-specific primers used for amplification are provided below. PCR reactions were conducted on an Applied Biosystems QuantStudio™ 5 Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). Relative mRNA expression was normalized to β-Actin as the internal control and calculated using the 2^−ΔΔCT method. All assays were performed in triplicate to ensure reproducibility. GeneSpeciesForwardReverseGIPC-1HumanCCTGATGGTGGACCAGAGGTGGTACACCCCTCCTAATGCCIGFR betaHumanTTTTGACTCCGCGTTTCTGCCAAAGAAAGGGGGCAAAGCC

Data were analyzed using Microsoft Excel (v2312, Microsoft Office 365) and GraphPad Prism (v9.4.0). One-way ANOVA or unpaired two-tailed t-tests were used to assess differences between treatment groups, as appropriate. For tumor growth curves, endpoint or same-day tumor volumes were compared using unpaired two-tailed t-tests. Statistical significance was defined as p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p < 0.0001 (****). Error bars represent standard deviation (SD).

Before experiments to evaluate the effect of GIPC1 depletion in sensitizing pancreatic cancer cells towards chemotherapeutic drugs, we analyzed the GIPC1 expression levels in AsPC-1, PANC-1, and some PDX cell lines. As shown in Figure 1A, GIPC1 expression could be seen in all the cell lines, corroborating the previous notion that GIPC1 is an essential survival gene in pancreatic cancer. Next, we intended to see whether GIPC1 depletion in pancreatic cancer cells can increase their sensitivity towards chemotherapeutic drugs such as GEM. Towards this end, we utilized lentiviral-based GIPC1 shRNA plasmids to perform GIPC1 knockdown in four cell lines. Unfortunately, we failed to generate stable clones of GIPC-depleted PDAC cell lines since the growth of the cells was not adequate to generate stable cell lines. Therefore, we had to use the Tet-on inducible smart-GIPC1 shRNA system. We analyzed three different clones of Tet-on inducible GIPC1 shRNA for their GIPC1 knockdown efficacy in AsPC-1 GFP-Luciferase cells and selected the best clone demonstrating the highest knockdown after induction with 400 ng/ml doxycycline for 72 h for further in vitro drug sensitivity experiments (Figure 1B). For these experiments, we added increasing concentrations of GEM to doxycycline-pretreated or untreated AsPC-1 and PANC-1 cells across a concentration range from 0.5 µM to 50 µM. The MTS assay was performed after 72 h. GIPC-depleted cells exhibited a steeper growth inhibition curve compared to the control cells, suggesting that GIPC-depletion sensitizes pancreatic cancer cells towards GEM in vitro (Figure 1C,D).

To further validate our hypothesis in vivo, we performed a single mouse trial using two pancreatic cancer cell lines, namely AsPC-1 and PANC-1. We observed that GIPC-depleted tumors were significantly smaller in both PANC-1 (Figure S1A,B) and AsPC-1 (Figure S1C,D) orthotopic xenografts. Also, GIPC-depleted tumors exhibited a significantly higher response to GEM treatment for both cell lines. To validate the results obtained from the single mouse trail, we repeated the same experiment in AsPC-1 xenografts with five mice per treatment group as usual. We observed more or less similar results in that experiment (Figure 2A,B). Histological evaluation of H&E- and Ki67-stained tumor sections revealed that GIPC1-depleted tumors displayed pronounced morphological alterations accompanied by a marked downregulation of proliferative markers (Figure 2C). Taken together, these results suggest that GIPC knockdown sensitizes pancreatic cancer xenografts towards GEM in vivo. Additionally, the single mouse trial's usefulness in identifying the best treatment strategy for combating cancer was also substantiated.

Based on our findings and corroborating evidence from previous studies, GIPC1 has emerged as a promising therapeutic target in PDAC. Nonetheless, earlier attempts to strongly modulate this pathway did not yield sufficient therapeutic efficacy [10,11,17]. We therefore hypothesized that selective inhibition of GIPC1-mediated signaling could represent a more effective strategy for PDAC treatment. To overcome the lack of efficient small-molecule inhibitors against GIPC1, our group developed a tumor-targeting lipopeptide (TTL) capable of directing therapeutics to tumor tissue [7,18]. Using this lipopeptide, we engineered a novel tumor-targeted liposomal formulation encapsulating GIPC1-specific siRNA (LGIPCsi).

Using a combination of ionic lipids, phospholipids, helper lipids, PEGylated lipids, TTL, and GIPC1-specific siRNA, we developed the LGIPCsi formulation. The physicochemical properties of the liposomes were characterized using dynamic light scattering (DLS), which revealed their hydrodynamic diameter and surface potential (Figure 3A,B; Supplementary Table S1). We evaluated the stability of the LGIPCi formulation in four different media: Milli-Q water, DMEM buffer, DMEM supplemented with 10% FBS, and DMEM supplemented with 55% FBS, for up to seven days. In Milli-Q water (103 nm) and DMEM buffer (106 nm), the hydrodynamic diameters remained stable without significant changes over the seven-day period. In contrast, in DMEM containing 10% FBS (176 nm), the particle size remained stable only up to five days, after which a marked increase in size was observed, rendering the formulation unstable and unmeasurable. Similarly, in DMEM containing 55% FBS (234 nm), stability was maintained for only two days before the particle size increased drastically, preventing further measurements (Figure S2). Compared with water and DMEM buffer, the presence of serum proteins in FBS appeared to promote particle aggregation, likely due to the interaction of negatively charged serum proteins with the positively charged lipid nanoparticles, resulting in increased particle size and reduced stability (Supplementary Table S1). To verify siRNA encapsulation, we performed agarose gel electrophoresis. Naked GIPC1 siRNA (lane I) exhibited a distinct band, whereas no free siRNA was detected when encapsulated within the liposomes (lane II), confirming near-complete encapsulation efficiency. Upon liposomal disruption with 0.1% SDS, the encapsulated siRNA was released and clearly detected (lane III), further validating the efficient entrapment of GIPCSi within the liposomal system (Figure 3C). Fluorescence microscopy demonstrated efficient cellular uptake of FITC-siRNA–loaded, Rhodamine-PE–labeled liposomes in both PANC1 and AsPC-1 pancreatic cancer cell lines. After 4 h of incubation, strong green (FITC-siRNA) and red (Rhodamine-PE) signals were observed in the cytoplasm, with substantial overlap around the perinuclear region, confirming successful delivery of siRNA via the liposomal carrier. Hoechst counterstaining of nuclei further confirmed intracellular localization of the nanocomplexes (Figure 3D).

To evaluate the functional effect of delivery, GIPC1 gene knockdown was assessed following liposomal siRNA transfection. Both PANC1 and AsPC-1 cells treated with GIPC1-siRNA liposomes showed a marked reduction in GIPC1 expression compared with untreated controls and cells transfected with control siRNA (Figure 3E). These findings confirm that the formulated liposomes not only facilitate efficient siRNA uptake but also induce robust gene silencing in PDAC cells.

The results from the above experiments prompted us to assess the therapeutic potential of GIPC1 depletion in vivo. Towards this end, we developed a novel tumor-targeted liposomal formulation (LGIPCsi) for siRNA delivery in vivo. To evaluate the tumor-targeting efficacy of the LGIPCsi formulation, we performed an in vivo tumor uptake study in PANC-1 orthotopic tumor-bearing mice. FAM-siRNA complexed with tumor-targeted (LGIPCsi) liposomal formulations was injected intraperitoneally in orthotopic PANC-1 tumor-bearing mice. In vivo tumor uptake was monitored by IVIS imaging after 24 h and 48 h. There is no autofluorescence interfering with the signal intensity from mice fur in this region as well. As shown in, our novel tumor-targeting liposomal formulation LGIPCsi clearly showed significant tumor uptake compared to the control. However, we did not observe a clear signal of FAM-siRNA due to strong autofluorescence in the green region of the spectrum. Figure S3

The therapeutic efficacy of GIPC1 silencing in combination with GEM was evaluated in an orthotopic AsPC-1 pancreatic xenograft mouse model. Mice were randomized into five groups (n = 5 per group): untreated control, LNPs, GEM (10 mg/kg), GIPC1 siRNA (0.5 mg/kg), and the combination of GEM with GIPC1 siRNA. Treatments were administered twice weekly for three weeks.

As shown in Figure 4A,B, combination therapy with GEM and GIPC1 siRNA produced a significant reduction in both tumor volume and tumor weight compared with monotherapies or control groups. Neither empty liposomes nor single-agent treatments achieved comparable tumor suppression, underscoring the synergistic benefit of the combination regimen. Histological analyses further supported these findings. H&E staining revealed reduced tumor cellularity in the combination group, while Ki67 immunostaining demonstrated a marked decrease in proliferative index compared with GEM alone, GIPC1 siRNA alone, or controls (Figure 4C). These results confirm that concurrent downregulation of GIPC1 enhances GEM efficacy, leading to robust inhibition of tumor growth in PDAC xenografts.

The impact of combined GIPC1 silencing and GEM treatment was further evaluated in an orthotopic PANC-1 pancreatic xenograft model. Mice were assigned to five groups (n = 5 each): untreated control, empty liposome (LNPs), GEM (10 mg/kg), GIPC1 siRNA (0.5 mg/kg), or the combination of GEM with GIPC1 siRNA, administered twice weekly for three weeks. As in Figure 5A,B, the combination treatment produced a marked reduction in both tumor volume and tumor weight compared with GEM alone, GIPC1 siRNA alone, or control groups. Neither empty liposomes nor monotherapies demonstrated comparable tumor inhibition. Histopathological analysis provided further confirmation. H&E staining revealed reduced tumor cellularity in the combination group, while Ki67 immunostaining demonstrated a substantial decrease in proliferative activity relative to single-agent or control groups (Figure 5C). Collectively, these findings show that concurrent GIPC1 knockdown significantly enhances the antitumor efficacy of GEM in PANC-1 xenografts.

Western blot analysis demonstrated efficient downregulation of GIPC-1 expression in tumors treated with GIPC-1 siRNA, confirming effective gene silencing in vivo (Figure 6A). RT-PCR analysis further supported these findings, showing marked reductions in GIPC-1 mRNA levels in siRNA-treated tumors compared with controls. Notably, the combination of GIPC-1 siRNA and GEM resulted in the most pronounced suppression of GIPC-1 expression relative to either monotherapy (Figure 6B). In addition, IGFR1 mRNA expression, a downstream signaling mediator of GIPC-1, was significantly decreased in tumors from the combination treatment group, suggesting effective disruption of GIPC-1–IGFR1–mediated oncogenic signaling (Figure 6C). These molecular changes correlated with the enhanced antitumor efficacy observed in the combination group compared with single-agent or control treatments.

In the single-mouse orthotopic pancreatic xenograft experiments using 6741, 4636, 4535, and 4911 tumor models, combination treatment with GIPC-1 siRNA and GEM consistently demonstrated superior antitumor efficacy compared to individual treatments or controls (). In the 6741 tumors, both tumor volume and tumor weight were markedly reduced in the combination group relative to the control, empty liposome, GEM alone, or GIPC-1 siRNA alone. Similar trends were observed in the 4636 tumors, where the combination treatment induced substantial inhibition of tumor growth, confirming that dual targeting of GIPC-1 and chemotherapy enhances therapeutic outcomes. In the 4535 and 4911 tumor models, combination-treated mice exhibited the most significant suppression of tumor progression, as reflected by reductions in both tumor burden and final tumor weights, whereas monotherapies produced only modest effects. Collectively, these findings across four independent pancreatic tumor models validate the robust and reproducible antitumor activity of combined GIPC-1 knockdown and GEM therapy, highlighting the synergistic potential of this strategy to overcome the limited efficacy of monotherapies. Figure S4