PSMA-specific degradable dextran for multiplexed immunotargeted siRNA therapeutics against prostate cancer

Abstract

Small interfering RNA (siRNA) is ideal for gene silencing through a sequence-specific RNA interference process. The redundancy and complexity of molecular pathways in cancer create a need for multiplexed targeting that can be achieved with multiplexed siRNA delivery. Here, we delivered multiplexed siRNA with a PSMA-targeted biocompatible dextran nanocarrier to downregulate CD46 and PD-L1 in PSMA expressing prostate cancer cells. The selected gene targets, PD-L1 and CD46, play important roles in the escape of cancer cells from immune surveillance. PSMA, abundantly expressed by prostate cancer cells, allowed the prostate cancer-specific delivery of the nanocarrier. The nanocarrier was modified with acid cleavable acetal bonds for a rapid release of siRNA. Cell imaging and flow cytometry studies confirmed the PSMA-specific delivery of CD46 and PD-L1 siRNA to high PSMA expressing PC-3 PIP cells. Immunoblot, qRT-PCR and flow cytometry methods confirmed the downregulation of CD46 and PD-L1 following treatment with multiplexed siRNA.

Graphical Abstract

Prostate cancer continues to have a high incidence,¹ causing approximately 20% of cancer-related deaths in men despite diagnostic and therapeutic advances.² Effective therapy against advanced prostate cancer remains a challenge. Sequence-specific gene silencing by small interfering RNA (siRNA) provides an exciting platform to achieve a precise and personalized treatment of diverse life-threatening diseases including advanced prostate cancer.^3,4 In comparison to conventional small-molecule drugs or proteins, siRNA therapeutics provide precise treatment by downregulating target genes. The flexibility of siRNA design and synthesis makes siRNA therapeutics an appealing option for novel treatment strategies.⁵

To achieve safe and effective systemic delivery that is essential for siRNA therapeutics, carriers such as polymers, lipids, peptides, and exosomes, have been developed, some with stimuli-responsive properties to pH, light or temperature, to enhance siRNA delivery.^6,7 Cell-specific targeting ligands attached to siRNA carriers can also significantly improve the pharmacokinetics, biodistribution, and selectivity of siRNA therapeutics.⁸

PSMA is a transmembrane protein that is highly expressed by castrate-resistant prostate cancer cells.⁹ Because of its prostate cancer cell-specific expression, PSMA has been an active target for various therapeutic approaches, such as antibody-drug conjugates,¹⁰ RNA aptamer conjugates,¹¹ PSMA-based immunotherapy,¹² and PSMA-targeted prodrug therapy.¹³ Urea-based PSMA low-molecular-weight inhibitors have demonstrated high PSMA binding and suitable pharmacokinetic profiles with picomolar to nanomolar affinities, leading to their applications in radiolabeled forms for imaging and radiopharmaceutical therapy.^14,15

Here, we developed a novel PSMA-specific acid biodegradable siRNA nanocarrier with a dextran backbone, using a urea-based small molecule as the PSMA-specific targeting moiety. We chose a 40 kD dextran as the backbone to construct an acid cleavable cationic nanocarrier to deliver siRNA. Due to its biodegradability, easy availability, and ease of modification, dextran has been widely used as a drug carrier in human applications.¹⁶ The biodegradation of the nanocarrier triggered within cells provides a strategy to resolve the conflicting requirements for polymeric siRNA carriers between the condensation and release of siRNA.¹⁷ Amine functional groups with positive charges that electrostatically bind siRNA are conjugated to dextran through acid cleavable acetal bonds.^18,19 Under weakly acidic conditions, such as in endosomes, these acetal bonds are cleaved to rapidly release the amine groups and the siRNA.

Target-gene selection is a critically important aspect of siRNA therapeutics. Immune checkpoint inhibitors are providing novel opportunities for exploiting the immune system to destroy cancer cells.²⁰ The use of siRNA as a gene-specific silencing agent is promising for blocking immune checkpoints, especially if the siRNA can be specifically delivered to cancer cells to avoid the autoimmune complications of antibody–based checkpoint blockades.²¹ In prostate cancer, immunotherapy has failed to achieve the impressive outcomes observed in cancers such as melanoma²² and NSCLC,²³ requiring new strategies to leverage the immune system for this disease, a need that may be met with multiplexed siRNA therapeutics.

While multiplexed siRNA therapy has been previously reported,^24–26 here, for the first time, we applied multiplexed siRNA to concurrently downregulate two immune molecules, CD46 and PD-L1, which facilitate the escape of cancer cells from immune surveillance. CD46 is a membrane protein that protects cells from complement attack.²⁷ A high expression of CD46 has been observed in several cancers including prostate cancer,²⁸ with clinical and experimental data supporting an association between increased CD46 expression and malignant transformation, and metastasizing potential.^29–31 CD46 is also being investigated as a target in early-phase clinical trials for a variety of cancers including advanced prostate cancer (e.g. NCT05011188). Programmed death 1 (PD-1) and its ligands, PD-L1 and PD-L2, play important roles in regulating the balance between T cell activation, tolerance, and immunopathology.³² Cancer cells exploit the PD-1/PD-L1 pathway to escape immune surveillance. The inhibition of PD-L1 can induce a significant immune response against cancer cells.^33–36 Here, we have used multiplexed siRNA to achieve the PSMA-specific combined downregulation of two potent immune molecules, CD46 and PD-L1, in PSMA-expressing prostate cancer cells.

Details regarding the synthesis of the siRNA nanocarrier, presented in Scheme 1, are described in the ESI, Fig. S1.† Briefly, compound 1³⁷ was reacted with neat tris(2-aminoethyl) amine at 100 °C to form compound 2. The dextran (40 kDa) scaffold was reacted with an excess of compound 2 to attach the amine groups to the dextran backbone through acid cleavable acetal bonds to produce compound 3 consisting of non-specific dextran, also termed dextran. ¹H NMR spectra indicated that the functionalized degree of glucose residues was ~0.5. Then, 6-maleimidohexanoic acid N-hydroxysuccinimide ester (EMCS) was reacted with compound 3 to introduce the maleimide group (~12 maleimide groups per dextran) to the dextran backbone (compound 4). The NHS ester of the previously described urea-based PSMA inhibitor¹³ (compound 5) was reacted with heterobifunctional poly(ethylene glycol) (PEG) (2 kDa) to form compound 6. Compound 6 was reacted with compound 4 to produce the penultimate nanocarrier. NMR spectra indicated an average of 8.3 PEG-PSMA on one dextran backbone. Finally, rhodamine, as an imaging probe, was conjugated to the amine groups (1.2 rhodamine molecules per dextran molecule) to form the final PSMA-specific siRNA nanocarrier compound 7, also termed PSMA-dextran.

Scheme 1.

Synthesis procedure of the dextran siRNA carrier, (a) 100 °C, overnight, (b) p-Toluenesulfonic acid, 4 Å molecular sieves, 65 °C, overnight. (c) HEPES buffer, pH 8.4, room temperature, 2 hours, (d) HEPES buffer, pH 8.4, room temperature, 2 hours, (e) 1: PBS pH 7.0 buffer, room temperature, 2 hours; 2: rhodamine-NHS ester, PBS pH 7.4.

Dynamic light scattering (DLS) was performed to investigate the hydrodynamic radii of the dextran nanocarriers (Fig. 1A). The radius of dextran (40 kDa) was ~4 nm, which increased to ~18 nm after the amine group and the PEG-PSMA targeting moieties were attached. With the attachment of the amine groups, the zeta potential of compound 4 increased to 19.7 ± 3.0 mV, compared to −9.8 ± 1.8 mV for natural dextran. PSMA-specific as well as non-specific dextran formed nanoplexes with siRNA at nitrogen/phosphate (N/P) ratios of 15 within 20 min incubation in water. The zeta potential of compound 7/siRNA nanoplexes at an N/P ratio of 15 was 10.2 ± 2.1 mV. Degradation studies of compound 7 were performed under different pH conditions (acetate buffer, pH 5.5 and PBS buffer, pH 7.4). The amine groups on the dextran backbone were cleaved with incubation in low pH buffer. An Amicon^® Ultra-15 centrifugal filter was applied to remove the salt and free amine in the buffer. The desalted solution was lyophilized and weighed, and a colorimetric assay was applied to measure the absorbance of the rhodamine of the compound 7 solution (1 mg mL⁻¹). As shown in Fig. 1B, when compound 7 was incubated in PBS buffer at pH 7.4, the absorbance of rhodamine at 550 nm was ~1.5 for up to 24 h of incubation indicating no obvious degradation of compound 7. In contrast, the amine groups were cleaved from the dextran scaffold at pH 5.5 starting from 8 h, with the absorbance of rhodamine decreasing significantly as the incubation time progressed. Nearly 80% of compound 7 degraded after 24 h of incubation in pH 5.5 buffer. Aldehyde derivatives predominantly generated acyclic acetals that have a shorter half-life than their cyclic analogues. However, there were small amounts of the more stable cyclic acetals formed as a byproduct¹⁹ that caused a weak absorbance of rhodamine after 24 h incubation in pH 5.5 acetate buffer. Although the proton sponge hypothesis has been proposed as being advantageous for the effective endosomal escape of cationic polymer carriers, the release of molecular agents from the carrier is equally important to achieve good transfection efficiency.³⁸ In both the non-specific and PSMA-specific dextran siRNA nanocarriers, the amine groups that electrostatically bind siRNA were attached to the dextran backbone through acetal bonds that were cleaved under weakly acidic conditions. With the cleavage of acetal bonds, these amine groups rapidly detached from the dextran scaffold to release siRNA. We demonstrated the stability of the siRNA/dextran complex for 24 h in 70% serum as shown in the ESI, Fig. S2.†

Fig. 1.

(A) Hydrodynamic radii of dextran and compound 7. (B) Zeta potential of dextran, compound 7, and the siRNA/compound 7 complex (n =3, values represent mean ± SD). (C) Rhodamine absorbance of 1 mg mL⁻¹ compound 5 solution at pH 5.5 and pH 7.4 buffer at different time points (n = 3, values represent mean ± SD, **P < 0.01).

To evaluate the specificity of the siRNA nanocarrier against PSMA-expressing prostate cancer cells, we performed cell studies with PSMA-positive PC3-PIP cells and PSMA-negative PC3-Flu cells. These isogenic cell lines have well-established differences in PSMA expression. Additionally, compound ZJ-43 (N-[[[(1S)-1-carboxy-3-methylbutyl]amino]carbonyl]-l-glutamic acid), an inhibitor of PSMA,³⁹ was used as a blocking agent in our cell studies to further establish PSMA specificity. Non-specific dextran labeled with rhodamine was used as a control compound in our studies. Confocal microscopy was used to investigate the uptake of the PD-L1 siRNA/compound 7 nanoplex in live cells (Fig. 2), with a magnified image presented in the ESI, Fig. S3.† After 2 h of incubation, the uptake of 1 μg mL⁻¹ compound 7 (PSMA-dextran) with siRNA in PC3-PIP cells was high whereas the uptake of 1 μg mL⁻¹ compound 3 with siRNA (dextran) in PC3-PIP cells was much lower. When excess ZJ-43 was added to PC3-PIP cells to block PSMA, the uptake of PSMA-dextran in PC3-PIP cells decreased to levels comparable to dextran. When PSMA-dextran and dextran were added to PC3-Flu cells that have low PSMA expression, the uptake of these two compounds was comparable and lower than in PC3-PIP cells. These imaging studies confirmed that the siRNA/compound 7 nanoplex demonstrated significant PSMA-specificity in human prostate cancer cells with high PSMA expression. Flow cytometry was used to quantify the cell uptake of the siRNA nanocarrier (Fig. 3A–D) by measuring the fluorescence of labeled rhodamine. In Fig. 3A and B, the uptake of siRNA/PSMA-dextran in PC3-PIP cells was much higher than that in PC3-Flu cells and ZJ-43 treated PC3-PIP cells. The uptake of siRNA/PSMA-dextran in PC3-PIP cells was ~ 4-fold higher than that in ZJ-43 treated PC3-PIP cells. We also observed a higher uptake of the siRNA/PSMA-dextran nanocarrier in ZJ-43 treated cells, compared to that in PC3-Flu cells, which may be due to the partial blocking of PSMA by ZJ-43. In contrast, the non-specific siRNA/dextran nanocarrier did not result in a significant uptake difference between PC3-PIP cells and PC3-PIP cells with PSMA blocked or PC3-Flu cells (Fig. 3C and D), highlighting the importance of a cell-targeted nanocarrier for improved uptake.

Fig. 2.

Live-cell confocal fluorescence microscopy of PSMA-positive PC3-PIP and PSMA-negative PC3-Flu cells with the siRNA/compound 7 nanoplex. Cells were treated with the PD-L1 siRNA/compound 7 nanoplex (concentration of PD-L1 siRNA: 100 nM; N/P ratio: 15). ZJ-43 (100 μg mL⁻¹), as a PSMA inhibitor, was applied to block PSMA.

Fig. 3.

(A) Flow cytometry fluorescence intensity of PD-L1 siRNA/PSMA-dextran (PSMA-dextran: 1 μg mL⁻¹) in PSMA-positive PC3-PIP, PSMA-negative PC3-Flu and PC3-PIP + ZJ43 (100 μg mL⁻¹) cells. (B) Quantification of fluorescence intensity of PD-L1 siRNA/PSMA-dextran in PC3-PIP, PC3-Flu and PC3-PIP + ZJ43 (100 μg mL⁻¹) cells. (C) Flow cytometry fluorescence intensity of PD-L1 siRNA/dextran (dextran: 1 μg mL⁻¹) in PC3-PIP, PC3-Flu and PC3-PIP + ZJ43 (100 μg mL⁻¹) cells. (D) Quantification of fluorescence intensity of PD-L1 siRNA/dextran in PC3-PIP, PC3-Flu and PC3-PIP + ZJ43 (100 μg mL⁻¹) cells (n = 3, values represent mean ± SD, **P < 0.01). Data in (B) and (D) were normalized to the fluorescence intensity obtained from PSMA-dextran in PC3-PIP cells.

We also characterized the effects of concentration and incubation time on the cellular uptake of the siRNA nanocarrier using flow cytometry results of PC3-PIP cells, as shown in Fig. 4A and B. After 2 h of incubation with 1 μg mL⁻¹ of PSMA-dextran or non-specific dextran, the cellular uptake of PSMA-dextran was ~3-fold higher than the uptake of non-specific dextran. As the incubation time progressed, the difference in uptake between PSMA-dextran and dextran became less apparent. By 4 h, the difference was 2.25-fold and continued to decrease with time dropping to 1.3-fold and 1.7-fold at 8 h and 24 h incubation, respectively. At longer time periods of 8 h-24 h, especially at 24 h, the degradation of dextran and the clearance of the amine groups and rhodamine may have also occurred. The rate of uptake was also different between the PSMA-dextran and the non-specific dextran over the 8 h period. Between 2 and 4 h, the uptake rate of PSMA-dextran increased marginally, but that of non-specific dextran increased by a factor of 1.7. Between 4 and 8 h, the uptake rate of PSMA-dextran again increased marginally but that of non-specific dextran increased by almost 2-fold. These data suggest that the initial uptake of PSMA-dextran was an active and rapid internalization process most likely mediated by PSMA, which was saturated with time. In contrast, with non-specific dextran, a non-specific slow but continuous uptake was observed.

Fig. 4.

(A) Uptake of PD-L1 siRNA/PSMA-dextran and PD-L1 siRNA/dextran in PSMA-positive PC3-PIP cells at different time points (concentration of PSMA-dextran and dextran: 1 μg mL⁻¹ (not including the weight of 5 nM siRNA), n = 3, values represent mean ± SD, *P < 0.05, and **P < 0.01). (B) Uptake of PD-L1 siRNA/PSMA-dextran and PD-L1 siRNA/dextran in PSMA-positive PC3-PIP cells at different concentrations (treatment time: 2 h, n = 3, values represent mean ± SD, *P < 0.05, and **P < 0.01).

Because dextran can be degraded at longer time periods, we investigated the effects of concentration with 2 h incubation, after which PSMA-dextran showed significantly more accumulation than non-specific dextran in PSMA-positive PC3-PIP cells at all investigated concentrations (Fig. 4B). With the time-course study, PSMA targeting increased the accumulation of the nanocarrier in cells with high PSMA expression. However, as the concentration increased, the difference between the uptake of PSMA-specific and non-specific dextran became less pronounced. At a concentration of 0.2 μg mL⁻¹, the difference in uptake was 11-fold. When the concentration was increased from 0.2 μg mL⁻¹ to 1 μg mL⁻¹, the difference in uptake dropped to 3-fold, and at 5 μg mL⁻¹, the difference in uptake dropped further to 1.7-fold. As observed with PSMA-specific cell uptake with time, a saturation of cell uptake with increasing concentration was observed, again suggesting the active PSMA-specific internalization of the nanocarrier, which became saturated as the concentration increased.⁴⁰ The non-specific dextran uptake continued to increase with increasing concentration but remained significantly lower than the PSMA-specific uptake. These data demonstrate the importance of PSMA-specific targeting to improve cellular uptake.

The amine functional groups of the nanocarrier carried positive charges to electrostatically bind negatively charged siRNA. These amine groups were conjugated to the dextran backbone through cleavable acetal bonds designed to break under weakly acidic conditions, which detaches the amine groups that are distributed throughout the cytoplasm with a reduced positive charge, rapidly releasing the siRNA. The dextran scaffold was depolymerized to small molecules using α-1-glucosidases present in cells and various organs,¹⁶ making our nanocarrier a highly biocompatible and effective siRNA carrier. The dextran nanocarrier did not alter cell viability in 72 h of treatment at concentrations of less than 100 μgmL⁻¹ as shown in the ESI, Fig. S4.† The rapid accumulation of PSMA-specific siRNA/dextran nanoplexes in cells with a high PSMA expression resulted in an enhancement of PD-L1 siRNA transfection efficiency, as is evident from the downregulation of PD-L1 in Fig. 5. A quantitative reverse transcription-polymerase chain reaction (qRT-PCR) was performed to measure the changes in the PD-L1 messenger RNA (mRNA) in PC3-PIP cells. Cells were treated with the PD-L1 siRNA/compound 7 nanoplex for 24 h using a previously validated PD-L1 siRNA sequence 5′-CCUACUGGCAUUUGCUGAACGCAUUU̲-3′,⁴¹ collected and processed for qRT-PCR to determine PD-L1 messenger RNA (mRNA). As shown in Fig. 5, the PD-L1 siRNA/compound 7 nanoplex resulted in significant inhibition of PD-L1 mRNA, which significantly decreased PD-L1 mRNA levels to ~60%, comparable to transfection with lipofectamine. When PC3-PIP cells were pretreated with ZJ-43 (100 μg mL⁻¹) for 1 h to block PSMA, no downregulation of PD-L1 was observed, further confirming the PSMA-specific downregulation of PD-L1.

Fig. 5.

Fold-change in PD-L1 mRNA levels in PSMA-positive PC3-PIP cells with different siRNA treatments (siRNA concentration: 100 nM; ZJ43 concentration: 100 μg mL⁻¹; PSMA-dextran concentration: 15 μg ml⁻¹). Cells were treated with siRNA-dextran for 24 h and values represent the geometric mean at a 95% confidence interval (n = 3). mRNA levels were normalized to untreated cells. *P < 0.05 and **P < 0.005. PC3-PIP cells treated with dextran were used as an additional control.

Our ultimate goal was to downregulate two important immune molecules, PD-L1 and CD46, to demonstrate the capability of PSMA-specific multiplexed siRNA delivery for immunotherapy. Exploiting the power of the immune system to eliminate cancer cells has led to some of the most exciting advances in the treatment of some cancers, such as melanoma.⁴² Developing strategies that can lead to other cancers, such as advanced prostate cancer, being as effectively controlled or eliminated is an important need. Multiple checks and balances built into the immune system are exploited by cancer cells to escape immunosurveillance.⁴³ The ability to target two or more of these immune molecules concurrently and specifically in prostate cancer cells may provide a potent strategy for treatment.^44,45 As shown in Fig. 6, we co-delivered 50 nM PD-L1 and 50 nM CD46 siRNA using our nanocarrier.

Fig. 6.

Fold-change in (A) PD-L1 and (B) CD46 mRNA levels in PSMA-positive PC3-PIP cells with different siRNA treatments. Cells were treated with siRNA-dextran for 24 h and the values represent the geometric mean at a 95% confidence interval (n = 3). mRNA levels were normalized to untreated cells. *P < 0.05, **P < 0.005, ***P < 0.0005.

We tested three different sequences of CD46 siRNA: (1) 5′-CUAUGGAGCUCAUUGGUAAUU-3′, (2) 5′-AGAUAUCAGGAUUUGGAAAUU-3′, and (3) 5′-GGAAGGAAUACUUGACAGUUU-3′, where nucleotides with UU underlined correspond to overhangs. Sequence 2 showed optimal single and multiplexed target-gene-specific downregulation and was selected for further characterization. As shown in Fig. 6A, PD-L1 mRNA levels significantly decreased with the single or multiplexed PD-L1 siRNA treatment of PSMA-positive PC-3 PIP cells but did not alter with single CD46 siRNA treatment. CD46 mRNA levels significantly decreased with single or multiplexed CD46 siRNA treatment. A small but significant increase in PD-L1 mRNA was observed with CD46 siRNA treatment. Immunoblots presented in Fig. 7 confirmed the mRNA reduction, with the downregulation of CD46 and PD-L1 proteins observed with 50 nM or 100 nM of CD46 or PD-L1 siRNA. Because of the lower basal levels of PD-L1 in PC-3 PIP cells,⁴⁶ the downregulation of the PD-L1 protein was not as dramatic as the downregulation of CD46, which was highly expressed. These differences were also identified in the flow cytometry data presented in the ESI, Fig. S5.† The high expression of CD46 supports its use as an important target for imaging-based detection and therapy of prostate cancer.

Fig. 7.

Representative immunoblot assays of PD-L1 and CD46 protein expression in PSMA-positive PC3-PIP cells with different siRNA treatments. Lane 1: no transfection control; Lane 2: dextran control; Lane 3: CD46 siRNA 100 nM; Lane 4: CD46 siRNA 50nM; Lane 5: PD-L1 siRNA 100 nM; Lane 6: PD-L1 siRNA 50 nM; Lane 7: CD46 100 nM + PD-L1 siRNA 100 nM; Lane 8: CD46 50 nM + PD-L1 siRNA 50 nM. Cells were treated with siRNA for 24 h.

Conclusions

Here, for the first time, we demonstrated the ability of a novel PSMA-specific degradable dextran nanocarrier to deliver multiplexed siRNA to simultaneously downregulate two important immune molecules, PD-L1 and CD46, which are exploited by cancer cells to escape immune surveillance in PSMA-expressing human prostate cancer cells. The PSMA-specific nanocarrier degraded under weakly acidic conditions while remaining stable under conditions of neutral pH. While we selected PD-L1 and CD46 as target genes, the flexibility of siRNA design can allow two or more different genes to be targeted specifically and concurrently. In the next stage, in vivo preclinical studies should further validate the ability of this PSMA-specific nanocarrier for harnessing the power of the immune system using siRNA therapeutics for the effective treatment of advanced prostate cancer in humans.