Functionalized MoS2 Nanosheets as Multi-Gene Delivery Vehicles for In Vivo Pancreatic Cancer Therapy.

Transition metal dichalcogenides (TMDCs) are categorized as novel two-dimensional (2D) nanomaterials with unique physical and chemical properties, bearing varied applications in medical and materials sciences. However, only a few works report the application of TMDCs for gene therapy in cancer treatment. Here, we engineer a multi-gene delivery system based on functionalized monolayer MoS2, which can co-deliver HDAC1 and KRAS small interfering RNAs (siRNAs) to Panc-1 cancer cells for combinational cancer therapy. The synergistic effect of gene silencing therapy and NIR phototherapy is demonstrated by inhibition of both genes, in vitro cell growth rate, and in vivo tumor volume growth rate, exemplifying pre-eminent anticancer efficacy. This anti-tumor effect is a result of the photothermal effect of MoS2 induced by NIR excitation and inactivation of HDAC1 and KRAS genes, which consequently bring about apoptosis, inhibit migration, and induce cell cycle arrest in the treated Panc-1 cells. Moreover, good biocompatibility and reduced cytotoxicity of MoS2-based nanocarriers enable their metabolism within in vitro and in vivo mouse models over a prolonged duration without any evident ill-effects. In summary, we demonstrate the promising potential of low-toxicity, functionalized MoS2 nanocarriers as a biocompatible gene delivery system for in vivo pancreatic adenocarcinoma therapy.

1. Introduction

A multitude of nanomaterials have been designed over the past decades as they exhibit promising potential for cancer theranostics applications, both in vitro and in vivo 1-3. Owing to the unique properties 4, 5 such as tunable particle size, multi-functionalized surface, high drug-loading capacity and good biocompatibility, great effort has been made to investigate the potential of nanomaterials for applications in the biomedical field. Numerous kinds of nanomaterials, such as liposomes 6, carbon nanomaterials 7, silica nanoparticles 8, metal-based nanoparticles 3, and quantum dots 9, 10 have been used in biomedical applications. Compared to these nanomaterials, 2D nanomaterials possess exceptional chemical, optical, and electronic properties and are thus, being considered as novel therapeutic agents for biomedicine, especially for cancer treatment 11. There have been some particularly interesting reports that demonstrate the encouraging potential of 2D nanomaterial theranostics in the pre-clinical area and targeted delivery of cancer therapeutics 12-17.

Since its discovery 18, graphene has become a well-known 2D nanomaterial extensively used in various fields, including solar cells 19, sensors 20, electronics 21, and energy storage 22. Subsequently, functionalized graphene nanosheets with biocompatible polyethylene glycol (PEG) have been applied in cancer therapy successfully 23, and were used to generate tumor ablation by utilizing the photothermal effect of graphene 24. Furthermore, the large surface area entitles graphene as the ideal drug delivery system for various therapeutic agents. In the recent years, another class of 2D nanomaterials, TMDCs, have become subjects of great interest for applications in electronics and optics 25. Similar to graphene, TMDCs possess a quasi 2D layered structure, typically composed of MX2, in which M represents a transition metal atom (Mo, W, or V) and X represents a chalcogen atom (S, Se, or Te), forming an X-M-X sandwich structure (Scheme ). As a typical layered transition metal, Molybdenum disulfide (MoS2) is a promising nanomaterial and considered to potentially surpass graphene in sensors and electronics 26, 27. Moreover, due to their structural similarity to graphene, MoS2 nanosheets have also been attracting likewise interest for their potential application in biomedical field. Recently, MoS2-based DNA sensor was used to detect DNA and small molecules 28, and functionalized MoS2 nanosheets were employed as drug delivery agents for cancer treatment 11. Nevertheless, there are only a handful of reports describing the theranostics applications of MoS2 11, 29-32 and thus, further investigation for exploring their biomedical applications, including gene therapy, is necessary.

Gene therapy is a biomedical strategy with the goals of modifying the genetic disease at its source by delivering nucleic acid into cells 33. The most common form of gene therapy is the use of corrective genes to replace a mutated gene, often called as gene silencing, wherein the expression of a certain gene gets suppressed. Since the discovery of recombinant DNA as a revolutionary tool for gene therapy in the 1970s 34, 35, gene therapy has attracted great interest and has been extensively utilized for therapeutic effect in many diseases, for example, cancer 36, immunodeficiency diseases 37, diabetes 38, and cardiovascular diseases 39. Among the several kinds of approaches in gene therapy, RNA interference (RNAi) stands out due to its fast experimental period, feasible operation, dose-dependent manner, and high efficiency of gene knockdown 40, 41. This technique uses a double-stranded siRNA to cause sequence-specific gene silencing at the post-transcriptional level 42. The siRNA hybridizes with the mRNA sequence of the targeted gene and then activates nucleolytic degradation of the targeted mRNA by RNase enzyme in cells 43, 44. The RNAi phenomenon is highly potent in the down-regulation of the targeted gene expression, however, the siRNA in its naked form is highly unstable 45 and thus, is not readily delivered to the target site of interest 46. These drawbacks thus, prevent further applications of RNAi in gene therapy. To overcome these hurdles for achieving successful clinical trials, various nanomaterials, such as liposomes 47, gold nanoparticles 48, and silica nanoparticles 49, are being investigated as gene delivery systems.

In our previous work, we utilized gold nanorods (AuNRs) as nanocarriers to co-deliver siRNA and anticancer drug, Doxorubicin, to study both in vitro and in vivo combinatorial therapy of pancreatic cancer 3. We demonstrated superior anticancer efficacy resulting from the interactive chemotherapeutic, gene suppression, and photothermal therapy 3. Compared to AuNRs-based nanocarriers, TMDCs nanomaterials possess large specific surface area and exceptional surface-to-volume ratio, which result in incomparably high loading capacity for drug delivery. In comparison to other nanomaterials, TMDCs possess advantages due to their optical properties, which enable superior photothermal effects to nanomaterials like graphene 11,50. In addition, low toxicity of TMDCs in comparison to graphene 51 signifies a promising alternative NIR photothermal theranostic agent. In this work, we fabricate a multi-gene delivery system comprising of PEGylated MoS2 nanosheets that can successfully co-deliver HDAC1 and KRAS siRNAs (against a G12D mutant KRAS gene for Panc-1 cells) for in vitro pancreatic cancer therapy. The unique function of Histone deacetylase 1 (HDAC1) in sustaining the pluripotency of embryonic and cancer stem cells underlines its important role in epigenetics 52, 53. Furthermore, the appreciable expression of HDAC1 in pancreatic cancer and other forms of cancers 54 promote cell transformation, survival, invasion, and metastasis 55-57. Thus, HDAC1 is considered as a promising target for cancer therapy. Moreover, the mutational KRAS gene has been shown to predominate more than 90% of pancreatic cancers and is thus, a major target for addressing cancer therapy 58. In this work, we utilized functionalized MoS2 nanosheets with folic acid (FA)-polyethyleneglycol (PEG) polymers and polyallylamine hydrochloride (PAH) for complexation with siRNAs. The surface chemistry of the nanosheets allowed their easy internalization in Panc-1 cells, and the synergistic effect of gene therapy and photothermal effect contributed by the MoS2 nanosheets-mediated co-delivery was able to impede the pancreatic cancer cell growth rate by 70%. Interestingly, the synergistic effect of the MoS2 nanosheets in the presence of NIR light reduced the in vivo tumor volume growth rate by 80%. On the other hand, we did not observe any significant weight changes in the mice treated with PEGylated MoS2 nanoformulation. Mice voluntary cage-wheel exercise showed no changes in the learning ability and movements of the treated mice vis-a-vis the untreated mice. Moreover, the major organs of treated mice did not exhibit any deformations or side-effects. Thus, our work establishes the rooted potential of MoS2 nanosheets as a novel and low-toxic multi-gene therapy nanoplatform for targeted gene delivery and photothermal therapy of pancreatic cancer.

2. Results

As a graphene analogue in 2D nanomaterials, the layered structure of MoS2 comprises of a 2D sandwich-type S-Mo-S structure (Scheme ). The most common method to prepare MoS2 nanosheets is the chemical exfoliation method with different lithium intercalating agents according to the reported studies 11, 27. Whilst naked MoS2 exhibits solubility in water, they are unstable in buffer solution and functionalization is necessary to improve their solubility to form stable solution 11. Scheme illustrates the functionalization procedure of MoS2 nanosheets. We used six different MoS2-based nanoformulations in our work: (i) naked MoS2 nanosheets functionalized with lipoic acid (LA)-modified PEG (MoS2/PEG); (ii) naked MoS2 nanosheets functionalized with FA and LA-modified PEG (MoS2/PEG/FA); (iii) MoS2/PEG formulation modified with PAH polymer (MoS2/LA/PEG/PAH, MoS2); (iv) MoS2/PEG/FA formulation modified with PAH polymer (MoS2/PEG/FA/PAH, FA/MoS2); (v) nanoplexes of siRNA and MoS2/PEG/PAH (MoS2/siRNA); (vi) nanoplexes of siRNA and MoS2/PEG/FA/PAH (FA/MoS2/siRNA).

2.1. Synthesis of multifunctional MoS2 nanosheets-based siRNA delivery system

For our experiments, we purchased the monolayer 2D MoS2 nanosheets from 2D-Semiconductors Inc, which exhibited thickness from 1-2.6 nm and lateral size ranging from 58-88 nm, as shown in Figure . For their application in biological media, the MoS2 nanosheets have to be functionalized with PEG to improve its biocompatibility. Due to its chemical structure, functionalization of MoS2 with PEG and folic acid-PEG were performed with LA-modified PEG (LA-PEG) or FA-modified LA-PEG (LA-PEG-FA) in order to allow the disulfide moiety to bind to the defect sites in MoS2 11, 50. Upon functionalization with LA-PEG-FA, the dimensions of the MoS2/PEG/FA were examined. As shown in Figure , the functionalized MoS2 exhibited thickness from 1.8-4.3 nm and lateral size from 42-100 nm. This increased thickness and lateral dimension most likely originated from addition of the PEG/FA groups on the surface of MoS2. FT-IR spectrum of naked MoS2, LA-PEG-FA, and the functionalized MoS2 are displayed in Figure . Unlike naked MoS2, the functionalized MoS2 exhibited notable absorption peaks at 2916 cm-1 and 1101 cm-1. The peaks at 2916 cm-1 and 1101 cm-1 correspond to C-H and C-O bonds in PEG chains in LA-PEG-FA, respectively. Hence, the presence of these peaks further indicates that successful conjugation of LA-PEG-FA onto the MoS2 nanosheets is achieved in MoS2/PEG/FA.

To enable conjugation and delivery of siRNA molecules, we functionalized PEGylated MoS2 with PAH polymer to obtain positively-charged MoS2/PEG/PAH (MoS2) or MoS2/PEG/FA/PAH (FA/MoS2). The absorbance spectra of MoS2 does not change by PEG or PAH functionalizations (Figure ). We used the dynamic light scattering (DLS) technique to measure the hydrodynamic diameter of the naked MoS2 nanosheets in DI water, which was estimated to be 90.84 ± 3.76 nm (Figure ). The measured zeta potential of the naked MoS2 nanosheets in DI water was -23.86 ± 1.94 mV (Figure ). The values of hydrodynamic diameter and zeta potential varied according to functionalization of different materials on the MoS2 surface. We noted an increase in the hydrodynamic size of MoS2 nanosheets after functionalization with LA-PEG and LA-PEG-FA, in agreement with the atomic force microscope (AFM) characterization results. Further functionalization with PAH was successful in modifying the surface charge of the nanosheets, as observed in the change of zeta potential from negative to positive (Figure 1F). Complexation with siRNA molecules by electrostatic interaction introduce negatively-charged molecules onto the surface of the nanocomplexes, resulting in negatively-charged surface. We noted increments in the zeta potential and hydrodynamic diameter of the complex as we prolonged the incubation time between MoS2/PEG/FA and PAH (Figure S1). Based on these results, we chose 36 hours as the optimized time to mix MoS2/PEG/FA and PAH. Figure 1G and Figure S2 represent the results of gel retardation study used to confirm the mass of FA/MoS2 nanosheets that entirely bind with a given quantity of siRNA. As we increased the amount of FA/MoS2 nanosheets in the MoS2-siRNA nanocomplex, a considerable decline in the quantity of free siRNA molecules was observed. Hence, we optimized the mass ratio between FA/MoS2 and siRNA molecules to be 6 μg: 1.3 μg on the basis of the gel retardation study, and we have used this ratio for all the in vitro experiments.

Figure 1

Characterization of the engineered MoS2-based nanocarriers. AFM characterization of (A) MoS2, (B) MoS2/PEG/FA. (C) FT-IR spectra of naked MoS2, LA-PEG-FA, and MoS2/PEG/FA. (D) Absorption spectra of different MoS2-based nanocarriers. (E) Hydrodynamic size and (F) surface zeta potential of the different complex of MoS2. (G) The quantitative analysis of gene loading ability of MoS2/PEG/FA/PAH nanocarriers. Gel retardation quantitative analysis of siRNA loading by MoS2/PEG/FA/PAH, using free siRNA as reference. The mass ratio between MoS2/PEG/FA/PAH (1 mg/mL) and siRNA (130 μg/mL) are set to be 0.5:1.3, 1:1.3, 2:1.3, 4:1.3 and 6:1.3, respectively. All experiments are performed in duplicates with consistent results. Values are means±SEM, n=3.

2.2. Microscopy fluorescent images of PEGylated or FA-PEGylated MoS2-siRNA nanocomplexes in Panc-1 cells

The overexpression of HDAC1 in most malignant forms of tumors (for e.g., pancreatic cancer, colorectal cancer, lung cancer, etc.), is known to activate specific-kinase signaling pathways in the tumor microenvironment that lead to increased cell proliferation and metastasis. Therefore, in this report, we investigate the feasibility of using a PEGylated MoS2 nanosheet platform to co-deliver HDAC1 and KRAS siRNAs in Panc-1 cancer cells in vitro and evaluate its efficiency as a multi-gene delivery system. First, we treated the Panc-1 cells with various MoS2 nanoformulations and examined them after 4 hours under a fluorescence microscope to observe transfection and distribution of HDAC1 siRNACyc3 and KRAS siRNAFAM inside the cells (Figure ). The Cyc3 (red, Figure ) and FAM (green, Figure ) fluorescence signals demonstrate the successful delivery of the two kinds of siRNA into Panc-1 cells by FA/MoS2 nanosheets. Figure depicts the successful co-delivery of HDAC1 and KRAS siRNAs into Panc-1 cells, mediated by FA/MoS2 nanoformulations. In contrast, we observed weaker fluorescence signals from Panc-1 cells treated with FA-unconjugated MoS2 nanoformulations (Figure ). Expectedly, the cells treated with PBS (Figure ), FA/MoS2 (Figure ), free HDAC1 siRNACyc3 (Figure ), and free KRAS siRNAFAM (Figure ) showed negligible fluorescence. These fluorescent images indicate that the use of MoS2 nanoformulations can successfully transport siRNAs into cells and the cellular uptake of MoS2 nanoformulations can be effectively increased by FA conjugation.

2.3. Flow cytometry analysis of PEGylated or FA-PEGylated MoS2-siRNA nanocomplexes in Panc-1 cells

To assess the HDAC1 siRNACyc3 and KRAS siRNAFAM delivery efficiency of the MoS2 nanoformulations, we performed flow cytometry analysis. Figure depicts the fluorescent snapshots of Panc-1 cells treated with different MoS2 nanoformulations. Negative controls were: Panc 1-PBS, Panc 1-FA/MoS2, Panc 1-HDAC1 siRNACyc3, or Panc 1-KRAS siRNAFAM. While the Panc-1 cells treated with FA/MoS2/siRNACyc3+FAM nanoformulations exhibited strongest FAM and Cyc3 fluorescence signals with 65% and 90% transfection efficiencies, respectively, the cells treated with MoS2/siRNACyc3+FAM nanoformulations yielded 35% and 60% transfection efficiencies, respectively (Figures ). These differences signify the role of FA conjugation in the cellular uptake of MoS2 nanoformulations. Contrary to these results, the cells treated with negative controls showed negligible fluorescent signals. Moreover, these results compliment the cell fluorescence microscopy analysis shown in Figure .

2.4. Gene expression analysis of Panc-1 cells post treatment with MoS2 nanoformulations

Since the overexpression of HDAC1 and mutant KRAS mRNA results in uncontrolled proliferation of pancreatic cancer cells, our goal in this study is to realize MoS2 nanoformulations-mediated delivery of specific siRNAs into Panc-1 cells which can target HDAC1 or mutant KRAS gene. Consistent with the flow cytometry analysis, the results in Figure demonstrate both HDAC1 and KRAS gene silencing in MoS2 nanoformulations-treated Panc-1 cells. While the cells treated with negative controls exhibited negligible changes in HDAC1 or KRAS mRNA and protein levels, we observed a notable suppression of HDAC1 or KRAS gene expression in the cells treated with FA/MoS2/HDAC1 siRNA, FA/MoS2/KRAS siRNA, or FA/MoS2/HDAC1+KRAS siRNAs with gene silencing effect around 50% for each gene. Thus, the results of RT-PCR and Western Blotting analysis reveal the potential of MoS2-based nanoformulations as efficient nanovehicles of multiple siRNAs to address targeted gene knockdown in pancreatic cells.

2.5. Viability of Panc-1 cells post treatment with MoS2 nanoformulations

Due to its overexpression in most malignant tumors, HDAC1 is a driving factor in sustaining the pluripotency of cancer stem cells and regulating vital movements of cancer cells (survival, invasion, metastasis, etc.) 52. Meanwhile, the mutational oncogene KRAS exists in almost all pancreatic cancers, which regulates cell proliferation, transformation, and apoptosis 59. The ability of MoS2 nanoformulations-mediated delivery of HDAC1 or KRAS siRNA to suppress the expression of HDAC1 or KRAS gene in Panc-1 cells is sufficiently established by the gene knockdown results (Figure ).

Subsequently, we further examined the effect of MoS2 nanoformulations on cell inhibition (Figure and Figure ). We treated Panc-1 cells with varying concentrations of FA/MoS2/PEG/PAH (5 μg mL-1 to 200 μg mL-1) and recorded around 80% cell viability even at 200 μg mL-1 concentration 72 hours post treatment (Figure ). Hence, PEGylated MoS2-based nanocarriers strongly demonstrate good biocompatibility and low-toxicity. Furthermore, we observed minor cell growth inhibition upon treatment with FA/MoS2 or FA/MoS2/scramble siRNA nanoformulations. In contrast, we recorded obvious inhibition in cases of FA/MoS2/HDAC1 siRNA (62.15±1.23%), FA/MoS2/KRAS siRNA (58.34±3.54%), and FA/MoS2/HDAC1+KRAS siRNA (53.34±4.71%). We can thus infer that the co-delivery of HDAC1 and KRAS siRNAs caused a stronger inhibition rate of Panc-1 cell growth. In addition, the strong NIR absorbance and typical optical properties of MoS2 nanosheets 50 endow them as potent representatives for photothermal therapy, and hence, we subsequently investigate the anti-proliferative effects of the combination of MoS2 nanoformulation and NIR light. Upon irradiation with NIR light, MoS2 solution exhibit increase in temperature, which was observed in a concentration-dependent manner (Figure . For this study, we treated Panc-1 cells with FA/MoS2 or FA/MoS2/HDAC1+KRAS siRNA nanoformulations for 4 hours, and then exposed them under 808 nm NIR light for 5 minutes. In the additional 72 hours, we observed 40% inhibition rate for FA/MoS2 under NIR light, and the strongest inhibition rate (over 70%) for the nanoformulation, FA/MoS2/HDAC1+KRAS siRNA under NIR light. In addition, the possibility of enhanced MoS2 release from endosome and release of genetic material from nanocarriers upon irradiation of NIR light might also contribute into the enhanced therapeutic effect of the MoS2 nanosheets carriers 60. These results suggest that both HDAC1 and KRAS genes are prospective candidates for gene therapy in pancreatic cancer. Furthermore, the combination of gene therapy and photothermal effect demonstrates synergistic effects for pancreatic cancer treatment.

2.6. Cell Migration Test post treatment with different MoS2 nanoformulations

Collective cell migration is the process of a large scale coordinated movement of cells, which form the basis of invasion and metastasis in malignant tumors 61. To study the effect of HDAC1 and KRAS genes in collective cell migration, we performed the wound healing migration assay 62 in Panc-1 cells treated with different MoS2 and siRNA nanoformulations (Figure ). Briefly, we seeded Panc-1 cells in 6-well plates and treated with PBS, FA/MoS2, FA/MoS2 with NIR light, FA/MoS2/scramble siRNA, FA/MoS2/HDAC1 siRNA, FA/MoS2/KRAS siRNA, FA/MoS2/(HDAC1+KRAS) siRNA, or FA/MoS2/(HDAC1+KRAS) siRNA with NIR light, respectively. Thereafter, the cells were wounded by manual scraping four hours post treatment to produce a gap and incubated for additional 68 hours. Compared with the initial wound window width, the 72 hours wound-closure of cells treated with PBS, FA/MoS2, or FA/MoS2/scramble siRNA reached 86.37±6.43%, 81.59±2.71%, or 80.25±2.05%, respectively, while the cells treated with FA/MoS2 with NIR light, FA/MoS2/HDAC1 siRNA, FA/MoS2/KRAS siRNA, FA/MoS2/(HDAC1+KRAS) siRNA or FA/MoS2/(HDAC1+KRAS) siRNA with NIR light showed obvious reduction in the wound healing process. The wound healing of cells transfected with both HDAC1 and KRAS siRNAs was much slower than that of cells transfected with either HDAC1 or KRAS siRNA. Interestingly, the wound healing process was most predominantly suppressed in the cells treated with FA/MoS2/si(HDAC1+KRAS) and NIR light. The wound closed to around 16.31±2.49% of the initial wound window width post 72 hours. These results pressingly suggest that the synergistic effect of photothermal therapy and MoS2 nanoformulations-mediated co-delivery of HDAC1 and KRAS siRNA significantly decreases the migration ability of pancreatic cancer cells.

2.7. In vivo Assessment of Antitumor activities of MoS2-based nanoformulations

To evaluate the tumor growth suppression effect of MoS2-based nanoformulations in vivo, tumor-bearing mice were treated with different MoS2 nanoformulations (Figure ). The mice treated with FA/MoS2 and NIR light (Figure ) or FA/MoS2/siRNA(HDAC1+KRAS) (Figure ) exhibited 47% and 56% tumor volume growth rate, respectively. Furthermore, the combination of FA/MoS2/siRNA(HDAC1+KRAS) and NIR light irradiation (Figure ) yielded the highest 78% suppression of in vivo tumor growth effect. Tumors-bearing mice treated with PBS (Figure ), FA/MoS2 (Figure ), or FA/MoS2/scramble siRNA (Figure ) exhibited negligible inhibition of tumor growth. The tumor weights in Figure also demonstrated that the combination of FA/MoS2/siRNA(HDAC1+KRAS) and NIR light irradiation resulted in the strongest suppression of tumor growth. It is worth mentioning that there were minimal changes in mice treated with different MoS2 nanoformulations during the examination period (Figure ). In addition, we examined gene knockdown efficiency of MoS2-based nanoformulations in vivo. As shown in Figure , tumor tissues were obtained at day 28 and are prepared for RT-PCR. In agreement with the in vitro results, Figure shows suppression of HDAC1 and KRAS expression by FA/MoS2/siRNA(HDAC1+KRAS) with or without NIR light. Therefore, the in vivo tumor inhibition assay clearly demonstrates the pivotal role of both HDAC1 and KRAS genes in driving pancreatic tumor growth, and that they both might be potential target genes for pancreatic cancer treatment.

2.8. In vivo toxicity test of MoS2 nanoformulations

The results of the in vitro and in vivo studies substantiate the promising feasibility of PEGylated MoS2-based nanosheets as an efficient siRNA delivery system for pancreatic cancer therapy. However, the successful use of such nanomaterials for clinical therapeutic applications in vivo cannot be realized without their toxicity assessments and time-bound harmless elimination from the body. We therefore, conducted a mice voluntary cage-wheel exercise assay, an ICP-MS (Inductively coupled plasma mass spectrometry) assay, and performed histological analyses to examine the in vivo toxicity and biodistribution of MoS2 nanoformulations. The effect of MoS2 nanoformulations on the motor learning ability of mice was first investigated through a mice voluntary cage-wheel exercise assay (Figure ). For this, we randomly divided C57Bl/6 mice into two groups and subcutaneously injected them with 50 mg kg-1 of PBS or FA/MoS2. Starting from day 0 through day 20 post injection, a steady rise in the voluntary running cycles of both the groups was observed. Moreover, the negligible changes of mice treated with different FA/MoS2 nanoformulations were consistent with the results in Figure .

Furthermore, we measured the concentrations of molybdenum in blood and tissues by ICP-MS analysis to quantify the accumulation and biodistribution of FA/MoS2 (Figure and Figure ). The ICP-MS analysis of blood samples at 0.5 hour yielded the concentration of molybdenum in blood to be 0.678±0.062 μg mL-1. The concentration decreased with injection time, where at 6 hours, the concentration was 0.095±0.009 μg mL-1, which suggested the rapid clearance in blood of MoS2-based nanoformulations. As shown in ICP-MS analysis of tissue samples, molybdenum element has predominantly accumulated in liver, spleen and kidney. At 24 hours after subcutaneous injection, the concentrations of molybdenum in liver, spleen, and kidney were 7.87 ± 1.51 μg mL-1, 3.60 ± 1.07 μg mL-1, and 2.78 ± 0.78 μg mL-1, respectively. As time passed, the concentrations of molybdenum decreased gradually. On day 21, the concentrations of molybdenum in liver, spleen and kidney are 2.49 ± 0.33 μg mL-1, 2.35 ± 1.35 μg mL-1 and 2.23 ± 0.13 μg mL-1, respectively. Because molybdenum is one of the trace elements in organism 63, it can be detected in the tissues of untreated mice. The concentrations of molybdenum in treated mice were higher than that of untreated ones, which indicate the injected molybdenum remained in major organs for more than 21 days.

Lastly, we performed histological analysis to assess the in vivo toxicity of MoS2 nanoformulations (Figure ). For this, hematoxylin and eosin (H&E) staining of tissue sections was done to investigate any histological changes in major organs (heart, liver, spleen, lung, kidney, and brain) 21 days post FA/MoS2 injection. We did not record any observable evidence of organ lesions in the FA/MoS2-treated mice with respect to the PBS-treated mice. Thus, the in vivo toxicity results pressingly support the use of MoS2-based nanoformulations as a low-toxic and biocompatible nanocarrier for cancer therapy.

3. Discussion and Conclusion

Pancreatic cancer stands among the most fatal cancers globally because of extensive local metastasis and advanced systemic dissemination 64. The limitations of conventional diagnostic modalities to detect the disease in early stages has restricted the post-diagnosis average lifetime of patients to usually less than one year 65. The traditional methods including surgery, radiation therapy, and chemotherapy can assist to prolong the lifetime of patients to some extent, but they are not suitable for long-term treatment of the disease 3. In the last decade, a myriad of nanomaterials have been extensively applied for diagnostic and therapeutic outcomes 2, 6, 48, 49. As representative of TMDCs, MoS2 has been widely utilized in electronics and optics 26. Although MoS2 nanosheets possess similar properties as graphene such as large surface area, facile surface-functionalization, and good biocompatibility, there are only a handful of studies that describe the relevance MoS2 nanosheets as drug delivery systems for cancer therapy 11, 60. Furthermore, the limited toxicity studies impede further biomedical applications of MoS2.

In this study, we explored a novel class of 2D nanomaterials-MoS2 nanosheets, and functionalized them with FA-PEG for combinatorial therapy of pancreatic cancer. We demonstrated the use of multifunctional PEG-modified MoS2 nanosheets as a multi-gene delivery system to co-deliver HDAC1 and KRAS siRNAs with photothermal effect for pancreatic cancer therapy. The cytotoxicity assessment show that the functionalized MoS2 nanosheets exhibited low cytotoxicity and were biocompatible for in vitro application. Following the successful transfection of the MoS2-based nanocarriers, fluorescence microscopy images, flow cytometry analysis, RT-PCR, and Western Blotting results validate the co-delivery of HDAC1 and KRAS siRNA molecules into Panc-1 cells with high transfection efficiency of 60% and 90%, respectively. The successful release of siRNAs inside the cells resulted in around 50% down-regulation of gene expression of both genes. The suppression of HDAC1 and KRAS gene expression essentially led to significant inhibition of cell proliferation and migration. These results underline the pressing roles of both, HDAC1 and KRAS genes, in regulating pancreatic cell proliferation, invasion, and migration, in agreement with previous reports 56, 57, 66. Furthermore, the use of NIR light in the application of the nanocarriers was also shown to be advantageous towards the therapeutic efficacy, due to the unique optical properties of MoS2 nanosheets. In summary, our studies discover the new and promising applications of TMDCs in cancer therapy, with combinational gene and photothermal therapy, which provide new insights for the development of 2D nanomaterials for cancer theranostics.

4. Materials and methods

4.1. Chemicals and reagents

We purchased monolayer MoS2 nanosheets (62 μg mL-1, dissolved in ethanol) from 2D-Semiconductors. mPEG-NH2 (5K) and NH2-PEG-Boc (5K) from Laysan Bio, Inc (USA). We purchased Lipoic acid (LA), Folic acid (FA), trifluoroacetic acid (TFA), dichloromethane, N,N'-dicyclohexylcarbodiimide (DCC), triethylamine (TEA), sodium bicarbonate, N-Hydroxysuccinimide (NHS), dimethyl sulfoxide (DMSO), diethyl ether, Poly(allylamine hydrochloride) (PAH, molecular weight 15,000), and 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromides (MTT) from Sigma Aldrich. Antibodies against HDAC1 (ab53091), KRAS (ab55391), and Actin (ab8227) were obtained from Abcam. We used ultrapure DI water for conducting all experiments (Milli-Q Integral 5).

4.2. Synthesis of LA-PEG and LA-PEG-FA polymers

: The synthesis of mPEG-LA polymer followed a previous protocol with subtle changes 67. Briefly, we mixed 500 mg of 5K mPEG-NH2 with 50 mg LA, 10 mg DCC and 6 μL TEA in 3 mL dichloromethane and stirred for 24 hours at room temperature. After the reaction solution got evaporated, we added 10 mL water to the solid precipitate in the bottle. The insoluble solid was removed by filtration. Sodium bicarbonate (0.1M) was added to this solution to adjust its pH to 8. Thereafter, we extracted the filtrate solution by dichloromethane for three times. When the organic solvent evaporated, the product was dissolved in water and lyophilized to obtain LA-PEG.

: To synthesize LA-PEG-FA, a Boc protected bi-functional NH2-PEG-Boc polymer was used to replace mPEG-NH2. Briefly, we mixed NH2-mPEG-Boc (500 mg) with LA (50 mg), DDC (10 mg), and TEA (6 μL) in dichloromethane (3 mL) under nitrogen and kept for magnetic stirring for 24 hours at room temperature. Subsequently, the dichloromethane solvent was evaporated and then, TFA (2 mL) was added and left for magnetic stirring for 3 hours at room temperature to de-protect the Boc group. After the TFA solvent evaporated, the leftover solid was dissolved in 10 mL water. The same purification procedure was carried out to obtain the LA-PEG-NH2 mid-product in solid. FA-conjugated LA-PEG (LA-PEG-FA) was prepared by mixing the amine-functionalized LA-PEG-NH2 with activated FA. Briefly, we mixed FA (100 mg) with DCC (50 mg) and NHS (35 mg) in anhydrous DMSO (3 mL) for 15 minutes at room temperature. We then added LA-PEG-NH2 (500 mg) in DMSO (5 mL). After 24 hours stirring at room temperature, cold diethyl ether (50 mL) was added to precipitate PEG from the mixture solution. We rinsed the product with diethyl ether (5 mL) 3 times and dried in vacuum. Then, we dissolved the yellowish product in water and filtered to remove any insoluble solid. The final product, LA-PEG-FA, was lyophilized and stored at -20 °C until further use.

: For the PEG functionalization, modified protocol was adopted from literature. In brief, 1 mg of the MoS2 solution (62 μg mL-1, ethanol) was evaporated to remove the ethanol solvent, yielding a solid, to which 10 mg of LA-PEG or LA-PEG-FA, redispersed in 2 mL of ultrapure DI water, was added. Then, we sonicated the resulting PEGylated MoS2 nanosheets mixture for 30 minutes and stirred overnight vigorously. This led to the removal of the excess unbound PEG or FA-PEG polymers upon two cycles of centrifugation at 10000 rpm for 25-30 minutes. We resuspended the precipitate and sonicated again in 1 mL of ultrapure water. This way, we obtained 1 mg mL-1 of PEGylated MoS2 nanosheets. Subsequently, we predissolved PAH in ultrapure DI water for siRNA loading, and made a 1:1 mixture of PEGylated MoS2 nanosheets (1 mg mL-1) and PAH (10 mg mL-1) in 2 mL DI water. This mixture was kept under 36 hours stirring under room temperature, followed by centrifugation to remove free PAH.

4.3. Characterizations of MoS2 nanocomplexes

The UV-visible absorption spectra of the prepared samples were recorded on a Shimadzu UV-2450 spectrophotometer with water as the reference. We measured the hydrodynamic size distribution profile and the zeta potential of the PEGylated MoS2 nanocomplex by a particle size analyzer (90 Plus, Brookhaven Instruments). We obtained the AFM images with an Asylum Research ARC2™ SPM instrument and the Fourier Transform Infrared (FTIR) spectrum by a Shimadzu FTIR spectrometer.

4.4. Gel retardation assay

We employed agarose gel electrophoresis to study the binding capacity of the synthesized PEGylated MoS2 nanocarriers with siRNA. We mixed MoS2/PEG/PAH (1 mg mL-1) with siRNA (130 μg mL-1) at five different mass ratios (0.5 μg: 1.3 μg, 1 μg: 1.3 μg, 2 μg: 1.3 μg, 4 μg: 1.3 μg and 6 μg: 1.3 μg). For electrophoresis, 1% agarose gel was mixed in TAE buffer solution at 100 V for 15 min. The intensity of ethidium bromide staining upon UV illumination was proportional to the amount of siRNA bound to the nanoplexes.

4.5. Tumor cell line and culture

The human pancreatic cancer cells, Panc-1 (CRL-1469) (American Type Culture Collection), were grown in Dulbecco's Modified Eagle's Medium (DMEM, Hyclone), supplemented with 10% fetal bovine serum (FBS, Hyclone), penicillin (Gibco, 100 μg mL-1), and streptomycin (Gibco, 100 μg mL-1). Cells were cultured and maintained in the UV incubator at 37 °C in a humidified atmosphere with 5 % CO2.

4.6. siRNA transfection

We seeded Panc-1 cells onto 6-well plates in DMEM medium with 10% FBS for 24 hours. Once the cells reached 30% - 50% confluence, they were ready for transfection. We then gently mixed the MoS2/PEG/PAH or MoS2/PEG/FA/PAH nanosheets dispersion (1 mg mL-1, 6 μL) with a G12D mutant KRAS siRNAFAM (130 μg mL-1, 10 μL) or HDAC1 siRNACyc3 and let them to rest for 30 min. Thereafter, a new culture medium with OPTI-MEM (950 μL, Invitrogen) was added to this PEGylated MoS2/siRNA mixture followed by continuous cell culturing. After four hours, we added 500 μL of DMEM + 30% FBS to the medium. In another parallel experiment, we used free siRNAFAM or siRNACyc3 at the same dose. We monitored the gene expression 72 hours post-transfection. The transfection efficiency was evaluated 4 hours post-treatment using fluorescence imaging and flow cytometry assays.

4.7. Flow Cytometry

The Panc-1 cells were trypsinized before being subjected to flow cytometry. The quantitative assessment of the siRNA transfection efficiency meditated by the synthesized MOS2 nanocarriers was done by a FACScalibur flow cytometer (Becton Dickinson, Mississauga, CA). Filter sets for KRAS siRNA (FAM) and HDAC1 siRNA (Cyc3) were used.

4.8. RNA isolation and quantitative RT-PCR

TRIzol reagent (Invitrogen) extracted the total RNA from Panc-1 cells 48 hours post transfection. The total RNA (2 μg), measured by a spectrophotometer (Nano-Drop ND-2000), was fed to a reverse transcriptase kit (Promega) to form cDNA. Evaluation of the mRNA levels of the target genes was done using an ABI Prism 7500 real-time PCR system (Applied Biosystems) by SYBR green (Promega).

4.9. Western Blotting

We seeded the cells in 6-well plates and left undisturbed for 24 hours after treating with different nanoformulations. We isolated the protein from cells by washing them with PBS and harvesting in lysis buffer. A Nano-Drop ND-2000 spectrophotometer was used to determine the total cellular protein concentration. 20 μg of denatured cellular extracts were resolved by 10% SDS-PAGE gels. The protein bands were transferred to Nitrocellulose Blotting membranes and incubated overnight at 4 °C with the appropriate primary antibody in the ratio: 1:1000 for KRAS, 1:1000 for HDAC1, and 1:1000 for actin, followed by washing in PBS buffer (0.05% Tween 20). Furthermore, they were incubated with goat anti-rabbit or anti-mouse secondary antibodies for 1 hour at room temperature. Finally, we observed proteins with chemiluminescent substrates.

4.10. Cell viability test

Following the MTT (Sigma) assay protocol, we seeded cells in a 96-well plate with 5 × 103 cells per well and incubated with different PEGylated MoS2 nanoformulations for specified time points. Four hours post incubation, some of the groups were irradiated by an 808 nm, 1W cm-2 NIR light for 5 minutes. The cells were incubated with MTT (5 mg mL-1, 20 μL) for 4 hours at 37 °C and 5% CO2. The precipitate was solubilized on adding 150 μL DMSO (Dimethylsulfoxide, Sigma) with 5 min of gentle shaking. Then, a microplate reader (Bio-Rad) was used to record the absorbance at 490 nm. The normalized absorbance of the sample well (with respect to that of the control well) was expressed as a percentage, with the viability of non-treated cells being 100%.

4.11. Mice voluntary cage-wheel exercise

C57Bl/6 mice (female; 5 weeks old) from the Medical Laboratory Animal Center (MLAC), Guangdong Province, China were allowed an acclimatization period of 1 week at 22 ± 2 °C with a 12-hour light:dark cycle (8 am, ON: 8 pm, OFF). Then, we randomly divided C57Bl/6 into 2 groups (3-4 mice per group) and injected with PBS or FA/MoS2 (50 mg kg-1) subcutaneously. Voluntary running was performed by these two groups following the start of exercise 68. The voluntary running system comprised of six separated chambers (Chengdu TME Technology Co., Ltd, China). During the training session, we placed the mice on a 30 mm diameter motorized rod in the chamber and gradually increased the rotation speed from 0 to 100 rpm over the course of 100s. The speed at which at the animal fell off was noted. The average rotation speed achieved by the animals in 7 trials was indicative of the performance. The two different groups were trained at the same continuous time points (day 0, day 5, day 10, day 15, and day 20). No significant changes were found between the two groups.

4.12. Preparation of paraffin sections for histological analysis

On the final day, we fixed the organ tissues in 4% buffered formalin-saline at room temperature for 24 hours. Then, we embedded the tissues in paraffin blocks and mounted 4 mm-thick paraffin sections on a glass slide for H&E staining. An Olympus BX51 microscope was used for examining the stained slices.

4.13. Inductively coupled plasma mass spectrometry (ICP-MS) analysis

FA/MoS2 (4 mg kg-1) was injected subcutaneously into 27 healthy C57BL/6 mice as the experimental groups, other three mice were injected with PBS as the control groups. Thereafter, we sacrificed three mice in each group, 0.5 hour, 1 hour, 2 hours, 4 hours, 6 hours, 24 hours, 7 days, 14 days and 21 days after injection, and collected the blood and tissue samples. Before ICP-MS analysis, a Milestone ETHOS ONE microwave acid digestion apparatus was used to digest the samples with 6 mL 65% HNO3 and 2 mL 30% H2O2 at 180 °C for 30 minutes. The sample was then diluted in 10 times DI water and the standard elemental concentrations of molybdenum were measured using an ICP-MS system (Agilent 7500C1).

4.13. PEGylated-MoS2 nanocarriers as dual, in vivo gene and photothermal therapeutic agents

All in vivo experiments were performed according to the protocol laid by the National Standard of Animal Care and Use Procedures at MLAC of Shenzhen University, Guangdong Province, China (Permit number: SZU-HC-2014-02). Following a 1-week acclimatization, we maintained the mice in an isolated biosafety facility meant for specific pathogen free (SPF) animals with bedding, food, and water. For tumor suppression assay, we inoculated athymic 6-weeks old nude mice with in vitro propagated Panc-1 cells (1 x 107) subcutaneously in the lower flank of mice. As the tumor grew in 10-15 days, we randomly divided the tumor-bearing mice (tumor size > 100-150 mm3) into 6 groups of 3-4 mice per treatment group and subcutaneously injected each group with different PEGylated MoS2 nanoformulations, summarized in Table 1. The injections were performed on every fourth day, starting on day 0. We used calipers (accuracy = 0.02 mm) to measure tumor volumes every other day, according to the expression V = L×W2/2 (W, the shortest dimension; L, the longest dimension). Change in size of each tumor was estimated relative to the corresponding size at day 0. We used one-way analysis of variance to test the statistical significances between groups.

Table 1

Different nanoformulations for in vivo gene and photothermal therapy

Group Number	Nanoformulations
1	no treatment (PBS)
2	FA/MoS2 (4 mg kg-1, injection)
3	FA/MoS2 (4 mg kg-1, injection) with NIR light
4	FA/MoS2/scramble siRNA (FA/MoS2 4 mg kg-1, siRNA 16 μg, injection)
5	FA/MoS2/siRNA(HDAC1+KRAS) (FA/MoS2 4 mg kg-1, siRNA 16 μg, injection)
6	FA/MoS2/siRNA(HDAC1+KRAS) (FA/MoS2 4 mg kg-1, siRNA 16 μg, injection) with NIR light