Nano particle loaded EZH2 inhibitors: Increased efficiency and reduced toxicity for malignant solid tumors

Reagents and antibodies

GSK126 and bovine serum albumin (BSA) were purchased from Sigma Aldrich (St. Louis, MO, USA)., and the purity was determined to be 99.5% based on high-performance liquid chromatography (HPLC). Fetal bovine serum (FBS) was from Gibico-BRL (Grand Island, NY, USA). Antibiotics and trypsin were from Macgene (Beijing, China). De-ionized water was obtained from Milli-Q system (Burlington, MA, USA). CD45, CD3, CD11b, Gr-1, and Ly6G flow antibodies were from BioLegend (San Diego, CA, USA). Coumarin-6 and β-cyclodextrin were from Sigma-Aldrich (Diegem, Belgium)

Mice and cell lines

Female BALB/c or C57BL/6 mice (4 weeks) were purchased from the Department of Laboratory Animal Science of Peking University Health Science Center, and maintained under SPF conditions in a controlled environment of 22°C–25°C approved by the Ethical Guidelines of EIACUC-PKU (A2021230). Mice were kept for 48 h to acclimatize to environment and fasted overnight before treatment.

Murine Melanoma B16F10, 4T1, MC38 cells were obtained from Peking Union Medical College, Cell Bank (Beijing, China). Cells were routinely maintained in high glucose DMEM supplemented with 10% FBS, 100 U/mL penicillin, and 100 μg/mL streptomycin at 37°C in a 5% CO₂ humidified incubator.

Cell proliferation assay

Tumor cells were seeded in 96-well plate at a density of 5 × 103/well and treated with GSK126 of indicated concentration for 36 h. The cells were imaged using phase contrast channel in the Incucyte S3 platform (Sartorius, Göttinggen, Germany). Four sets of phase contrast images from distinct regions within each well were taken at intervals of 3 h using a 10× objective. Incucyte S3 image analysis software was set to detect the edges of the cells and to determine their confluence in percentage.

Tumor xenograft experiments

Tumor cells (5 × 105) were injected into the flanks of 4-week-old BALB/c or C57BL/6 mice to generate xenografts. When tumors reached approximately 100 mm³ in size 5 days later, the mice received an intraperitoneal injection of GSK126 or GSK126 NPs daily at a dose of 50 mg/kg. Tumor volume was determined every day by measuring the two perpendicular diameters of the tumors and using the equation V = length (mm) × width (mm)²/2 (Eqn. 1), and the body weight was recorded every day. After treatment, mice were euthanized with CO₂, then tumors were weighted and fixed with 4% paraformaldehyde for further studies.

RNA sequencing and data processing

RNA sequencing and data processing followed the methods detailed in our previous publication.^[18] Briefly, B16F10 cells were treated with DMSO or GSK126, and total RNA was extracted. Paired-end libraries were prepared and sequenced on the Illumina Novaseq 6000 platform. Reads were aligned to the GRCm38.p4 genome using Hisat2, and RPKM values were calculated. Differential gene expression analysis was performed with edgeR, defining significant differential expressions by FDR-adjusted P values < 0.05 and log2 fold change > 1.0. Gene enrichment analysis for Gene Ontology and KEGG pathways was conducted using gene set enrichment analysis software. The RNA-seq data for this paper is available at NCBI GEO: GSE163078.

Preparation of GSK126-loaded albumin NPs

GSK126-loaded albumin NPs were prepared by a desolvation methods as previously reported.^[26,27] Briefly, albumin (20 mg) was completely dissolved in distilled water (2 mL) and was titrated to pH 8.0 with NaCl (0.1 mol/L). Then GSK126 dissolved in ethanol was dropped into the albumin solution under constant stirring. Afterward, glutaraldehyde (8%, 10 μL) was added and the solution was stirred at room temperature for 8 h. The purified GSK126-loaded albumin NPs were obtained after evaporation and ultrafiltration, and stored at 4℃ for further use.

Estimation of entrapment efficiency and drug loading capacity

The entrapment efficiency (EE) and drug loading capacity (LC) of GSK126 was determined using an indirect method, which involves measuring the amount of free drug available in nano dispersion. The standard curve for GSK126 was established by preparing a series of GSK126 solutions with concentrations ranging from 4 to 20 μg/mL. Subsequently, the absorbance values of these solutions were measured at 280 nm using a UH5300 UV-Visible spectrophotometer, and a standard curve for GSK126 was plotted. Utilizing an indirect method to evaluate the encapsulation efficiency and drug loading, the solution derived from ultrafiltration in the nanoparticle purification process is diluted to an appropriate concentration. Following this, the absorbance at 280 nm is measured to quantify the unbound GSK126 content based on a standard curve. The encapsulation efficiency (EE%) and drug loading capacity (LC%) are then calculated using the provided equations.

Characterization of prepared GSK126 NPs

The mean particle size and size distribution of NPs were measured by dynamic light scattering (DLS) (ZEN1690 Zetasizer Nano-S90, Malvern, UK). The intensity autocorrelation was measured at a scattering angle of 90° at 25°C. The freeze-dried samples were diluted to an appropriate concentration using normal saline and were transferred into a polystyrene cuvette for measurement. The zeta potential of NPs was detected using Zetasizer 2000 (Malvern, UK). The obtained freeze-dried samples were reconstituted and diluted to an appropriate concentration with normal saline before measurement. The morphology of prepared nanoparticles was investigated by transmission scanning electron microscope (TEM). Briefly, the nanoparticles were reconstituted with purified water. Then the nanoparticle suspension was dropped on a 200-mesh copper grid and was left to dry prior to measurement.

Intracellular localization of NPs

For intracellular localization analysis, coumarin-6 was used as a model fluorescent dye. B16F10 cells were seeded in confocal dishes (5 × 10⁴ cells/well) and cultured overnight at 37°C in 5% CO₂. Subsequently, the culture medium was removed and the cells were washed with PBS for three times. Then, the cells were treated with serum-free medium containing BSA or coumarin-6 NPs at a concentration of 200 ng/mL. The cells were then incubated in the dark at 37°C for 6 h. Following this incubation, the culture medium was removed, and the cells were washed three times with cold PBS before being stained with Hoechst 33342 (5 μg/mL) for nuclear staining. Then, the cells underwent three washes with cold PBS, were sealed with phenol red-free medium, and promptly examined under a laser confocal microscope. Specifically, the excitation and detection wavelengths for coumarin-6 were 467 nm and 502 nm, respectively. For Hoechst 33342, the excitation and emission wavelengths were 345 nm and 461 nm, respectively.

Mechanism of the cellular uptake of NPs

As a step forward, the mechanism of cellular-uptake or endocytic pathway of NPs was determined as B16F10 cells (5 ×10⁵ cells/well) were grown for 24 h in 6-well plates supplemented with 2 mL RPMI-1640 at 37°C with 5% CO₂. Then, the cells were supplied with fresh medium containing 10 mM β-cyclodextrin (a known inhibitor of the Gp60/ caveolar transport) for 0.5 h; cells without the inhibitor pretreatment were used as controls. After treatment with the inhibitor, the cells were carefully washed with PBS for three times, and each group was further incubated with coumarin-6 NPs (200 ng/mL) at 37°C with 5% CO₂ for 4 h. Then, the cells were collected and resuspended in PBS for flow cytometry analysis.

Oil Red O staining

After drug treatment, mice were sacrificed, and liver samples were excised and embedded in Tissue-Tek OCT compound for histopathological analysis. The OCT-embedded samples were serially sectioned at 10 μm. Oil Red O staining was performed to evaluate lipid accumulation, and the results were photographed under a microscope.

Flow cytometry

For flow cytometric analysis of TILs, tumors were finely minced with a scalpel blade and enzymatically digested to obtain single-cell suspensions. Splenocytes were isolated through mechanical disruption. Subsequent antibody staining followed the manufacturer’s protocols.

Statistical analysis

All experiments were performed at least three times with triplicate. Statistical data were expressed as means ± standard deviation (SD). Comparisons between different groups were carried out with Student’s t-test and analysis of variance (ANOVA) as appropriate using GraphPad Prism (ver. 8.0). Statistical comparison among groups was determined using Kruskal-Wallis test. Values of P < 0.05 were considered statistically significant (^*P <0.05; ^**P < 0.01; ^***P < 0.001).

GSK126’s limited efficacy with undesirable side effects

To gain a comprehensive understanding of the cellular response to EZH2 inhibition in various solid tumors, we employed the Incucyte S3 long-term dynamic cell image acquisition device to evaluate the proliferative kinetics of B16F10, 4T1, and MC38 treated with GSK126. Our analysis revealed that solid tumor cell lines displayed minimal inhibitory effects at a concentration of 10 μmol/L (Figure 1A). ^[18] Consistent with our previous studies, hematologic tumor cell lines exhibit higher sensitivity to GSK126 compared to solid tumors.^[18] For the assessment of the in vivo antitumor efficacy of EZH2i, we established subcutaneous xenograft models of B16F10 and administrated GSK126 intraperitoneally at a dose of 50 mg/kg/d (Figure 1B). Despite the administration of GSK126, the tumorigenic capacity of B16F10 in vivo remained unaffected, as indicated by the consistent dynamics of tumor volume and weight (Figure 1C and 1D). Consistent with these findings, intraperitoneal administration of GSK126 also did not exhibit a significant inhibitory effect on tumor growth in C57BL/6 mice bearing breast cancer cell line 4T1 or colon cancer cell line MC38 xenografts (Figure 1E-1J). Notably, the group treated with GSK126 experienced a marked decrease in body weight (Figure 1K). Furthermore, Oil Red O staining of liver sections displayed a significant increase in lipid accumulation in the GSK126-treated group (Figure 1L). In addition, obvious inflammatory features such as peritoneal adhesions and ascites accumulation were found when dissecting mice treated with GSK126. Collectively, these findings suggest that EZH2i demonstrate limited effectiveness against solid tumors, as well as the undesirable adverse effects.

Figure 1

Efficacy of EZH2 inhibitor GSK126 in solid tumors. (A) Cell proliferation of B16F10, 4T1 and MC38 cells in GSK126 (10-16 μmol/L) treatment monitored by IncuCyte S3. Mean ± SEM is shown (n = 6). (B) Images, (C) tumor growth curve, and (D) tumor weight of mice bearing B16F10 in the vehicle group (n = 6) and GSK126 (50 mg/kg) treatment group (n = 6). (E) Images, (F) tumor growth curve, and (G) tumor weight of mice carrying 4T1 in the vehicle group (n = 6) and GSK126 (50 mg/kg) treatment group (n = 7). (H) Images, (I) tumor growth curve, and (J) tumor weight of mice carrying MC38 in the vehicle group (n = 5) and GSK126 (50 mg/kg) treatment group (n = 5). Tumor volume was measured with a vernier caliper every 2–3 days. Tumor volume = Length × Width × Width / 2. Tumor weights were measured at day 15. (K) Body weight of B16F10, 4T1, and MC38 xenograft mice. (L) Oil Red O staining of liver sections in B16F10 tumor-bearing mice of Vehicle and GSK126 treatment group. Data in (D), (G) and (J) were analyzed by one-way ANOVA with Tukey multiple comparison posttest. ns, nonsignificant, ^*P < 0.05.

Global toxic effects of GSK126 by RNA-seq profiling

To further dig out the underlying mechanism of the unpleasant anti-tumor effects of the EZH2i, we treated B16F10 cells with GSK126 and proceeded with RNA-seq. GSEA analysis revealed that GSK126 treatment led to the activation of various immune-related pathways, including interferon-gamma response, interferon-alpha response, inflammatory, and IL6/JAK/STAT3 signaling pathways (Figure 2A and 2B). Notably, cytokine-related pathways exhibited significant alterations post-GSK126 treatment (Figure 2C), with volcano plots highlighting substantial changes in multiple inflammatory factors and their associated transcription factors (Figure 2D). The activation or permission of these immune pathways and cytokines may increase the inflammatory level and exacerbate tissue damage, contributing to the adverse effects of GSK126. Furthermore, we also observed that GSK126 treatment upregulated chemokine-related pathways and increase the expression of immune chemokines such as CCL5 (Figure 2E). The upregulation of CCL5 can enhance the chemotactic effect on immune cells,^[28] thereby may contributing to the observed adverse abdominal inflammatory effects associated with GSK126. Additionally, GSEA enrichment analysis also demonstrated that genes associated with the triglyceride metabolism pathway showed significant upregulation in the GSK126-treated group, including Pnpla3, Lipg, and Gpat3 (Figure 2F). This is consistent with our observation of hepatic lipid accumulation in vivo. These observations suggest a potential exacerbation of the local inflammatory response and dysregulated lipid metabolism within the peritoneal cavity triggered by GSK126.

Figure 2

(A) Enrichment plots generated by Gene Set Enrichment Analysis (GSEA) for gene ontology chemokine production genes (left) and hit genes expression in heat map (right), n = 3. (B) Enrichment plots generated by GSEA for gene ontology cytokine production. (C) Enrichment plots generated by GSEA for signaling pathways probably related to cytokine production. (D) Volcano plot of differential expressed genes between GSK126-treated cells and vehicle (UP, red color indicating log2 Fold Change > 1.0 and adjust P value < 0.05; DOWN, blue color indicating Log2 Fold Change < -1.0 and adjust P value < 0.05; other genes are colored in grey. (E) Enrichment plots generated by GSEA for signaling pathways probably related to chemokine production. (F) Enrichment plots generated by GSEA for glycerolipid metabolism.

Preparation and characterization of GSK126-loaded albumin nanoparticles

In light of the limited efficacy and severe adverse effects associated with EZH2i in solid tumors, we are exploring the potential of utilizing an albumin-nanoparticle delivery system to facilitate the targeted transportation of EZH2i. Albumin-GSK126 NPs were prepared using the modified desolvation method, which generated ultrafine NPs at a rapid rate according to the optimum conditions. The successful encapsulation of GSK126 into albumin NPs was confirmed through UV analysis, revealing characteristic peaks at 285 and 310 nm (Figure 3A). Notably, the prepared GSK126 NPs exhibited a small spherical core with an average diameter of 30.09 ± 1.55 nm and a narrow size distribution, and demonstrate high drug loading capacity (16.59% ± 2.86%) and good entrapment efficiency (99.53% ± 0.208%) (Figure 3B). Furthermore, the GSK126 NPs displayed a zeta potential of –28 mV, indicating their stability in a physiological environment (Figure 3C). The surface morphologies of GSK126 NPs were examined through SEM. As depicted in Figure 3D, the SEM images showed the synthesized GSK126 NPs have a spherical appearance. To assess the cellular transport capability of GSK126 NPs, a lipophilic fluorescent dye, coumarin-6 (Cou), was incorporated as an indicator of cellular uptake efficiency. The cellular uptake of coumarin-6 NPs by B16F10 cells significantly increased with 6 h incubations (Figure 3E). Flow cytometry confirmed that coumarin-6 NPs efficiently penetrated the cell membrane, while the cellular uptake was inhibited by methyl β-cyclodextrin (β-CD), a known inhibitor of the Gp60/caveolar transport (Figure 3F). These findings verified that the optimized nanoplatforms demonstrated excellent behaviors in vitro, promoting us to further explore their therapeutic potential in vivo.

Figure 3

Morphological and physical characterizations of GSK126 NPs. (A) UV spectrum of BSA, GSK126, and GSK126 NPs. (B-C) The size and surface charge of GSK126 NPs were determined by dynamic light scattering using Zetasizer Nano ZS. (D) Surface morphology of GSK126 NPs was determined by TEM. (E) The intracellular localization of Cou NPs was examined by confocal microscope, and Cou was used as a model fluorescent dye. (F) Flow cytometry analysis of the cellular uptake of Cou NPs.

Nano-assembled delivery system presents effective tumor suppression

The in vivo anti-tumor efficacy of GSK126 NPs, vehicle, and free GSK126 were validated in B16F10 bearing C57BL/6 mice. Both free GSK126 and GSK126 NPs were administered via intraperitoneal injection at a dosage of 50 mg/kg/d. As depicted in (Figure 4A-4B), the tumors in the vehicle treated group grew very rapidly to approximately 700 mm³ at day 12 post-treatment. While free GSK126 did not exhibit a significant inhibitory effect on tumor growth, its nanoformulations demonstrated notable tumor growth inhibition compared to the vehicle group (Figure 4C). This was evidenced by reduced tumor volume and decreased tumor weight, suggesting that the NPs of GSK126 may have superior effects on improving systemic circulation compared to free GSK126. Surprisingly, although nanocarriers effectively enhance the anti-tumor activity of EZH2i in solid tumors, we also noticed a significant decrease in body weight in tumor-bearing mice (Figure 4D). Oil Red O staining results demonstrated that GSK126 NPs significantly ameliorated the hepatic lipid accumulation caused by free GSK126, effectively reducing the hepatotoxicity of GSK126. (Figure 4E). This highlights the need for further exploration of the intravenous administration route.

Figure 4

Effects of GSK126 NPs on reducing side effects and increasing efficacy. (A) Images, (B) tumor growth curve, (C) tumor weight, (D) body weight and (E) Oil Red O staining of liver sections of mice carrying B16F10 in the control group (n = 7), GSK126 (50 mg/kg) treatment group (n = 7) and GSK126 NPs (50 mg/kg) treatment group (n = 7). Tumor volume was measured with a vernier caliper every 2 days. Tumor volume = Length × Width× Width / 2. Tumor weights were measured at day 12 after engraftment. Data were analyzed by one-way ANOVA with Tukey multiple comparison posttest. ns, nonsignificant, ^*P < 0.05, ^**P < 0.01.

Previous investigations have indicated that GSK126 may promote the generation of myeloid-derived suppressor cells (MDSCs) or recruitment within the tumor microenvironment, which could partially account for the compromised effectiveness in combating tumor progression.^[29] We then examined immune cells in the peripheral blood, spleen, and tumor microenvironment of B16F10-bearing mice treated with vehicle control, free GSK126, and GSK126 NPs using flow cytometry, with a focus on myeloid subsets (Figure 5A). In peripheral blood, both forms of GSK126 can reduce the proportion of CD3+ T cells while increasing the proportion of CD11b+ myeloid cells, with these effects being more pronounced in the GSK126 NPs treatment group (Figure 5B). Besides, free GSK126 significantly reduced the proportion of CD3+ T cells in the spleen, whereas GSK126 NPs had no significant impact on the CD3+ T cells in the spleen (Figure 5C). Neither form of GSK126 altered the proportion of CD11b+ myeloid cells within the CD45+ white blood cells in the spleen (Figure 5C). Upon analyzing tumor-infiltrating immune cells, we found that both forms of GSK126 had no significant impact on tumor-infiltrating CD3+ T cells and CD11b+ myeloid cells (Figure 5D). Based on these myeloid immune cells’ phenotyping, we further detect MDSC subsets in peripheral blood, spleen, and tumor microenvironment to evaluate whether the GSK126 NPs could reverse this effect to enhance the anti-tumor efficacy. We identified CD11b⁺Ly6G^-Ly6C^hi as M-MDSCs, and CD11b⁺Ly6G⁺Ly6C^low as PMN-MDSCs, and the sum of these two populations as total MDSCs. Both forms of GSK126 could increase the proportion of PMN- and total MDSC cells in peripheral blood, but not M-MDSCs (Figure 5E and 5F). Meanwhile, both free GSK126 and GSK126 NPs increase the proportion of M-and total MDSC cells in spleen, but have limited effect on PMN-MDSCs (Figure 5G and 5H). However, in the tumor microenvironment, GSK126 NPs significantly reduced the proportion of M-MDSCs which is the main subtype functioning the immune suppressive role, whereas free GSK126 did show the increasing tendency in M-MDSC proportion (Figure 5I and 5J).

Figure 5

GSK126 NPs affect the generation and infiltration of MDSCs. (A) Gating strategy for CD3+ and MDSCs in flow cytometry. (B-D) Proportions of CD3+ T lymphocytes and CD11b+ myeloid cells among CD45+ leukocytes in (B) peripheral blood, (C) spleen, and (D) tumor. (E-F) Effects of vehicle control, free GSK126, and GSK126 NPs on M-MDSCs, PMN-MDSCs, and total MDSCs in peripheral blood. (G-H) Effects of vehicle control, free GSK126, and GSK126 NPs on M-MDSCs, PMN-MDSCs, and total MDSCs in spleen. (I-J) Effects of vehicle control, free GSK126, and GSK126 NPs on M-MDSCs, PMN-MDSCs, and total MDSCs in tumor-infiltrating cells. ns: nonsignificant, ^*P < 0.05, ^**P < 0.01, ^***P < 0.001.

These findings suggest that GSK126 NPs and free GSK126 exhibit similar immune regulatory effect in peripheral blood and spleen by tuning either M-MDSC or PMNMDSC percentage. However, upon nanoformulation, it does reduce M-MDSC infiltration in the tumor microenvironment, thereby improving the immune-suppressive microenvironment and enhancing the antitumor capacity.