Anti-EpCAM functionalized graphene oxide vector for tumor targeted siRNA delivery and cancer therapy

Shung Zhng Yifn Wng Yng Yng

a School of Pharmaceutical Sciences, Capital Medical University, Beijing 100069, China

b Engineering Research Center of Endogenous Prophylactic of Ministry of Education of China, Beijing 10069, China

c Beijing Area Major Laboratory of Peptide and Small Molecular Drugs, Beijing 10069, China

d Yanjing Medical College, Capital Medical University, Beijing 101300, China

Keywords: Graphene oxide siRNA delivery Survivin Anti-EpCAM Gene silencing

ABSTRACT Graphene oxide (GO) has emerged as a potential drug delivery vector.For siRNA delivery,GO should be modified to endow it with gene delivery ability and targeting effect.However,the cationic materials used previously usually had greater toxicity.In this study,GO was modified with a non-toxicity cationic material (chitosan) and a tumor specific monoclonal antibody (anti-EpCAM) for the delivery of survivin-siRNA (GCE/siRNA).And the vector (GCE) prepared was proved with excellent biosafety and tumor targeting effect.The GCE exhibited superior performance in loading siRNA,maintained stability in different solutions and showed excellent protection effect for survivin-siRNA in vitro.The gene silencing results in vitro showed that the mRNA level and protein level were down-regulated by 48.24% ±2.50% and 44.12% ±3.03%,respectively,which was equal with positive control (P > 0.05).It was also demonstrated that GCE/siRNA had a strong antitumor effect in vitro,which was attributed to the efficient antiproliferation,and migration and invasion inhibition effect of GCE/siRNA.The results in vivo indicated that GCE could accumulate siRNA in tumor tissues.The tumor inhibition rate of GCE/siRNA 54.74% ± 5.51% was significantly higher than control 4.87% ±8.49%.Moreover,GCE/siRNA showed no toxicity for blood and main organs,suggesting that it is a biosafety carrier for gene delivery.Taken together,this study provides a novel design strategy for gene delivery system and siRNA formulation.

1.Introduction

RNAi has been a unique treatment of genetic diseases,such as malignant tumors [1],cardiovascular diseases [2],hypertension [3],diabetesm [4].

In

vivo

,double-stranded RNA (dsRNA) is firstly degraded to small interfering RNA (siRNA) by the specific enzymes.And then,siRNA associates mRNA according to the principle of base complementation,so as to induce the process of downregulating its expression [5,6].However,it is difficult to introduce exogenous siRNA into the body and produce specific therapeutic effects in treatment position [7].There are many limitations for the successful application of siRNA

in

vivo

,such as half-life,high molecular weight,negative charge,hydrophilicity,instability,degradation and filtration by glomerul,etc.[8,9].It is urgent to develop an effective delivery carrier for introducing exogenous siRNA into the body and producing specific therapeutic effects at the therapeutic site.For efficient gene delivery,it is essential to load gene drugs into a delivery vector,which is divided into viral vector and non-viral vector.The further clinical application of viral vectors is limited due to the biological safety problems such as cytotoxicity,high immunogenicity and potential carcinogenicity [10,11].At present,non-viral gene delivery vectors such as cationic polymers [12],cationic polysaccharides [13],protamine [14],cationic lipids [15]and cationic surfactants [16]have become the focus of gene therapy research due to low toxicity,low immunogenicity,

etc

.Unfortunately,the transfection efficiency of non-viral vectors is not ideal

in

vivo

.Therefore,it has become the most urgent and challenging topic in gene therapy to search a new non-viral vector with high efficiency and low toxicity.

Graphene oxide (GO) is a derivative of graphene,which is a carbon material with two-dimensional (2D) honeycomb structure formed by accumulation of single-layer sphybridized carbon atoms [17–19].There are a lot of oxygen functional groups (carboxyl,hydroxyl,epoxy) on the surface and edges of GO,and large conjugated structure is convenient to covalent and non-covalent functionalization [20–22].In recent years,GO has attracted more and more attention in the area of biomedical research and application due to its high water dispersibility,colloidal stability,large specific surface area and surface modification.Many studies have proved that GO can be widely used in the application of drug and gene carriers [23,24],cell imaging [25],biosensor [26]and antibacterial materials [27]for the treatment and diagnosis of diseases.It is necessary to modify GO properly to ensure the gene loading ability and targeting effect of GO.

As a natural alkaline polysaccharide with positive charge [28],chitosan (CS) has many advantages,such as non-toxicity,good biocompatibility and biodegradability,making it an excellent material for gene delivery.Therefore,CS was selected for the modification of GO in this research.As an inorganic-organic hybrid nanoparticle,the CS modified GO is supposed to be a versatile platform for gene delivery,which has the advantages of both inorganic and organic materials [29–32].Epithelial cell adhesion molecule (EpCAM) is a type I transmembrane glycoprotein which is normally expressed only on the basement membrane of epithelial tissue except squamous epithelium [33].EpCAM is only localized in the basement membrane of epithelial tissue in normal tissues.In pathological cases,EpCAM is usually expressed in inflammatory area and epithelial cells with rapid proliferation.EpCAM is highly expressed in adenocarcinoma,including lung cancer,pancreatic cancer,breast cancer,ovarian cancer

etc

.[34–36].As reported previously [37],EpCAM was abundantly expressed in breast cancer cells,which was 100–1000 times higher than that in normal tissue cells.EpCAM monoclonal antibody (anti-EpCAM) is a specific antibody for highly expressed EpCAM molecule on the surface of cancer cells.In this study,CS was conjugated onto the GO surface through amide bonds (GO–CS),and then anti-EpCAM was connected with GO–CS by

π

-

π

interaction.A novel carrier GO–CS/anti-EpCAM (GCE) for siRNA delivery was successfully prepared.Survivin-siRNA was applied to explore the siRNA delivery by GCE.Silencing of survivin inhibits the development and metastasis of tumors by promoting apoptosis,and arrests the tumor cells at G0/G1 phase.The results showed that GCE could effectively encapsulate survivin-siRNA and inhibit the proliferation of MCF-7 breast cancer cells,producing significant anti-tumor effects

in

vitro

and

in

vivo

(Scheme 1).It is believed that the antibody functionalized graphene oxide material can serve as a promising gene delivery vector,and has a broad clinical application prospect in the future.

Scheme 1–Schematic illustration of mechanisms and preparation of GCE/siRNA.(A) GCE/siRNA were prepared by GO-COOH,functionalization with chitosan (GO-CS),immobilization with anti-EpCAM (GCE) and then mixing with survivin-siRNA (GCE/siRNA) in aqueous solution.(B) After targeted into tumor tissues,GCE could deliver survivin-siRNA into cells successfully,and survivin-siRNA could release from GCE/siRNA to perform the antitumor effect.

2.Materials and methods

2.1.Materials

Graphene oxide,chitosan (Mw=50 kDa),chloroacetic acid,2-(

N

-morpholino) ethanesulfonic acid (MES,99%),

N

-(3-dimethylaminopropyl)-

N

-ethylcarbodiimide hydrochloride (EDC ? HCl,99%) and N-hydroxysuccinimide (NHS,98%) were purchased from Sigma-Aldrich (St Louis,USA).Anti-EpCAM antibodies (anti-EpCAM) produced in mouse was obtained from Abcam (Cambridge,UK).MCF-7 cell line was obtained from the Cell Bank of Chinese Academy of Sciences.Homo-survivin-siRNA,FAM-survivin-siRNA,Cy5-survivin-siRNA,and the negative control siRNA (NC) were purchased from GenePharma Co.,Ltd.(Shanghai,China).Lipo2000 (Lipo),Hoechst 33,342,TRIzol and Alexa Fluor?488 annexin V/Dead Cell Apoptosis Kit was purchased from Invitrogen (Carlsbad,USA).RPMI-1640 medium and trypsin were purchased from Hyclone Laboratories Inc.(Logan,USA).Fetal bovine serum (FBS) was purchased from GIBCO (Grand Island,USA).High Capacity RNA-to-cDNA Kit,High Capacity cDNA Reverse Transcription Kit,TaqMan Gene Expression Master Mix TaqMan Gene Expression Assays (survivin assay,GAPDH assay),and human survivin ELISA kit were bought from Thermo Fisher Scientific (Waltham,USA).DAPI and doxorubicin hydrochloride (DOX ? HCl) was purchase from Sigma-Aldrich (St Louis,USA).Matrigel was purchased from BD Biosciences (New Jersey,USA).

MCF-7 human breast tumor cells were cultured in RPMI-1640 medium containing 10 FBS,1% penicillin and streptomycin,and kept at 37 °C with 5% CO.The cells were subcultured 2,3 times a week till they reached 80% confluence.

Specific pathogen free (SPF) female BALB/c nude mice (4–6 weeks,15–17 g) were purchased from Animal Department of Capital Medical University (Beijing Laboratory Animal Center,Beijing,China).The guideline for all of the animal experiments was the Laboratory Animal-Guideline for Ethical Review of Animal Welfare issued by the National Standard GB/T 35,892–2018 of the People’s Republic of China.

2.2.Synthesis of GO-CS

In order to attach CS with GO more effectively,the hydroxyl,epoxide,and the carbonyl groups of GO were converted to carboxyl group.As shown in Fig.1,carboxylic acid-functionalized GO (GO-COOH) was synthesized by oxidation method [38,39].Firstly,GO (50 mg) was added to 30 ml water and sonicated for 2 h to disperse homogeneously.Then,NaOH (5 g) and ClCH 2 COONa (5 g) were added to the suspension.After ultrasonic reaction for 2 h,the reaction was stirred at room temperature for 3 h.Then,the supernatant was discarded after centrifugation (10 000 ×

g

,4 °C) for 10 min,and the precipitation was washed by water for 3 times.The product was freeze dried for 48 h to obtain GO-COOH.GO-CS was prepared by a typical method.GO-COOH (20 mg) and CS (140 mg) were dispersed in 10 ml of MES buffer (0.1 M,pH 5) and the suspension was sonicated for 1 h.EDC (300 mg) and NHS (300 mg) were slowly added into the mixture and sonicated for 6 h.Then the reaction was stirred at room temperature for 16 h,and the supernatant was discarded after centrifugation (10 000 ×g,4 °C) for 60 min.The sediment was washed by 1% acetic acid,and then washed with water for 3 times [40].GO-CS was obtained after 48 h of freeze-drying.

Fig.1–Characterization of GCE/siRNA.(A) The chemical structure of GO,CS,GO-COOH,GO-CS and GCE were determined by FTIR spectra.(B) AFM images of GO (a) and GCE (b).(C) The zeta potential value of GO (a),GO-CS (b),GCE (c) and GCE/siRNA (30:1,w/w) (d),the particle size of GO (e),GO-CS (f),GCE(g) and GCE/siRNA (30:1,w/w) (h).(D) TEM of GO (a) and GCE (b).The data were presented as the mean ±SD,n=3 (For interpretation of the REFERENCES to color in this figure,the reader is referred to the web version of this article.).

2.3.Immobilization of GCE

In order to increase the targeting effect of gene vectors,EpCAM monoclonal antibody (anti-EpCAM) was immobilized onto the surface of GO-CS.Briefly,GO-CS (3 mg) was evenly dispersed in 3 mL water.Anti-EpCAM (6 μg) was added and stirred overnight at room temperature.The supernatant was discarded after centrifugation (10 000 ×g,4 °C) for 1 h,and the sediment was washed with water for 3 times [41].GCE was obtained after 48 h of freeze-drying.

2.4.Preparation of GCE/siRNA

GCE/siRNA complex was prepared as follows:1 OD survivin-siRNA (33 μg) was dissolved in 250 μl of fresh diethylpyrocarbonate (DEPC) water to obtain 10 μmol/l survivin-siRNA solution.GCE solution (1 mg/ml) and survivin-siRNA solution were mixed at a series of mass ratios,and the transfection complex GCE/siRNA could be obtained after the reaction solutionsgently shaken and incubated at room temperature.

2.5.Characterization of GCE

The structure of the GO,GO-CS and GCE were determined by FTIR spectra (Nicolet iS5,USA) and UV absorption spectra (UV-2600,Kyoto,Japan).The morphology of GO and GCE were observed by transmission electron microscope (TEM,Tokyo,Japan).Atomic force microscopy (AFM) images were gotten by atomic force microscopy (Veeco Instruments Inc.,USA).The hydrodynamic size of GCE/siRNA was obtained by Zetasizer Nano ZS (Malvern Instruments Ltd,UK) at room temperature.Zeta potentials were measured by a zeta potential machaine (ZetaPlus,Brookhaven,USA).

2.6.Agarose gel retardation assay

Agarose gel retardation assay was applied to the detection of loading capacity of GCE and ratio of GCE to survivin-siRNA.GCE/siRNA complex was prepared at a certain mass ratio (2–50,w/w).Each complex (10 μl) at different proportions was carefully added to the hole of the agarose gel (1% agarose gel with 10 mg/ml EtBr) in Tris/Borate/EDTA (TBE) buffer,respectively.After electrophoresed under 120 V for 30 min,the siRNA bands were imaged by a UV transilluminator (Geliance 600,PerkinElmer Inc.,Waltham,USA).

2.7.Anti-RNase a degradation assay

Survivin-siRNA (1 μg) and GCE/siRNA loaded with survivin-siRNA (1 μg) (w/w=30:1) were mixed with RNase A solution (0.1 mg/ml) separately at 37 °C for 0.5 h.EDTA (5 μl,5 mM) was added at the determined time point.Then,20 μl,0.8 mg/ml of heparin sodium was added at 37 °C for 0.5 h to replace survivin-siRNA protected by GCE vector.The samples at different time points were operated according to the agarose gel retardation assay mentioned above.

2.8.Cellular uptake

FAM-labeled survivin-siRNA were performed in the experiment.After seeded for 24 h,MCF-7 cells were treated with siRNA (100 nM),GCE/siRNA (contain 100 nM siRNA),and Lipo/siRNA (contain 100 nM siRNA),respectively.After incubation and transfection for 5 h,each medicinal medium was discarded and washed by PBS solution for 3 times.Hoechst 33,342 (1 ml,1 μg/ml) was added into the well.After incubation for 15 min,the staining solution was discarded and the cells were washed 3 times by PBS.The cellular uptake results were observed by confocal microscopy (CLSM,TCSSP5,Leica,Wetzlar,Germany).

2.9.Cell toxicity assay

MTT assay was performed to evaluate the toxicity of GCE against MCF-7 cells.MCF-7 cells (5 ×10cells/well) were seeded and incubated for 24 h.The cells were treated with GCE at various concentrations (10–150 μg/ml).After incubation for 2 d,25 μl,5 mg/ml of MTT was added and the cells were incubated at 37 °C with 5% COfor 4 h.The absorbance of each well was tested by microplate reader (EnSpire,PerkinElmer Inc.,USA),and the detection wavelength was 570 nm.

The antitumor activity of different complexes

in

vitro

was also tested by MTT test.The cells were treated respectively with naked NC,naked siRNA,GCE/NC,GCE/siRNA,Lipo/NC,Lipo/siRNA (containing 20,40,80,120 and 160 nM siRNA).

2.10.Real-time PCR

After MCF-7 cells were transfected for 2 d,TRIzol reagent (2 ml) was used to extract the RNA from each group,and then the RNA concentration was measured by Nanodrop 1000 spectrophotometer.2 μg of total RNA of MCF-7 cells was reverse transcribed to cDNA,and then cDNA was transcribed to survivin-mRNA with housekeeping gene GAPDH as an internal control.The value relative quantity of survivin-mRNA expressions were calculated with average threshold cycle (Ct) by the delta-delta Ct (2) test.

2.11.ELISA

MCF-7 cells were incubated and transfected for 2 d By the end of culture,the cells were collected and lysed with RIPA buffer.The protein concentration and survivin-protein were calculated by a BCA protein kit and a human survivin ELISA kit,respectively.Plates were read at 450 nm to measure the OD values.

2.12.Apoptotic cell morphology

3 ×10cells of MCF-7 cells were seeded in a 20 mm dish and incubated for 1 d.The cells were treated with siRNA (100 nM survivin-siRNA),GCE/NC (contain 100 nM NC-siRNA),GCE/siRNA (contain 100 nM survivin-siRNA) for 2 d,respectively.The cells were washed 3 times by PBS.10% formalin stationary solution was added for 20 min.Then,the cells were stained with DAPI (1 ml,1 μg/ml) at 37 °C,5% COincubator for 8 min.After staining,the cells were washed by PBS solution,and the apoptotic morphology was observed and photographed by CLSM.

2.13.Analysis of cell cycle and apoptosis rate by FCM

After MCF-7 were incubated and treated with GCE/NC (contain 100 nM NC-siRNA) and GCE/siRNA (contain 100 nM survivin-siRNA) for 2 d,the cells were collected and washed by cold PBS solution,respectively.Then the cells were resuspended by adding binding buffer (250 μl,1 ×) to a concentration of 1 ×10cells/ml.According to instructions of Cell Apoptosis Kit,Alexa Fluor? 488 annexin V (5 μl) and PI (1 μl,100 μg/ml) were added (100 μl),respectively.The cells were incubated for 15 min in darkness at room temperature,and then binding buffer (400 μl,1 ×) was added.The apoptosis rate of MCF-7 cells in each group was calculated by a flow cytometry (BD LSRFortessa,Becton Dickinson,USA) immediately.

2.14.Wound healing assay

Wound healing assay is a commonly used method to evaluate the migration ability of cancer cells [42,43].MCF-7 (4 ×10cells/well) were seeded and incubated until the cell confluence reached 90%–100%.Then,in order to simulate the wound formation,the 10 μL micropipette tip was used to scratch vertically at each well of the 6-well plate to ensure that the four to five straight lines.MCF-7 cells were incubated and treated with GCE/NC (containing 100 nM NC-siRNA) and GCE/siRNA (containing 100 nM survivin-siRNA).The scratch widths of cells in each group were observed at determined time points (0,24 and 48 h) by a fluorescence microscopy (Nikon Eclipse Ti,Tokyo,Japan).

2.15.Transwell assay

Migration assay of the MCF-7 cells was quantified by the cells that directionally migrated through 8 μm pore size polycarbonate filter of the transwell chamber (Corning Inc.,NY,USA) [44].The MCF-7 cells of each group were suspended with serum-free RPMI-1640 containing 0.1% BSA.The cells (100 μl,2 ×10/ml) were seeded in the upper chamber,while RPMI-1640 medium containing 10% FBS (600 μl) was added into the lower chamber.After incubated for 24 h,the culture medium was discarded from the transwell chamber,then the non-migrating cells were carefully wiped from the upper chamber,and the upper chamber was washed twice with PBS.Then the cell was fixed,stained,and washed by PBS.After air-dried naturally,the transwell chamber was turned upside down and placed on the slide under a fluorescent microscopy (DM6000B,Leica,Wetzlar,Germany).Five visual fields were randomly selected for photography (×200).The invasion test was carried out,except that the upper surface of the filters was covered with matrigel in serum-free RPMI-1640 medium at 4 °C.

2.16.Distribution of GCE/siRNA in vivo

In order to explore the targeting and organ distribution of GCE in nude mice,a model of human breast cancer xenograft in nude mice was established.MCF-7 cells (1 ×10cells/mouse) were subcutaneously injected into mouse.After 20 d,the tumor volumes was approximately 100 mm.The mice were randomly divided into four groups,which were injected with normal saline (NS),naked Cy5-siRNA (0.3 mg/kg),GC/Cy5-siRNA (contain 0.3 mg/kg siRNA) and GCE/Cy5-siRNA (contain 0.3 mg/kg siRNA) into tail vein respectively.The nude mice were imaged after injection for 1,2,4 and 8 h.After observation,all of the mice were sacrificed and the main organs (hearts,livers,spleens,lungs and kidneys) were dissected.The fluorescence distribution in organs was photographed by an

in

vivo

fluorescence system (IVIS Spectrum,USA).

2.17.Tumor inhibition assay and toxicity study in vivo

The nude mice with MCF-7 cell transplantation tumor model were divided into 4 groups (

n

=6),and treated respectively with NS,naked siRNA (0.3 mg/kg),GCE/siRNA (containing 0.3 mg/kg siRNA),and DOX (1.2 mg/kg) every other day for 5 times.On the last injection,all mice were sacrificed and the tumors and the main organs of each group were collected to study tumor inhibition and organ toxicity

in

vivo

.The histomorphological structure changes of organs and tumor tissues were observed by H&E staining.

2.18.Hemocompatibility test

The fresh rat blood (5 ml) was centrifuged (2 000 ×g,10 min),and washed by PBS for 3 times to obtain the red-blood-cell (RBC).Then,2% (v/v) RBC suspension was prepared with PBS solution.Naked siRNA,GCE,GCE/siRNA and Lipo/siRNA (100 μl) with a series of concentrations were mixed with 2% RBC suspension (900 μl).After incubation for 1 h at 37 °C,the samples were centrifuged (3 000 ×g,10 min).The absorbance was measured at 540 nm.

2.19.Blood biochemistry test

Hematological examination was performed to explore the potential toxicity of drugs

in

vivo

.BALB/c nude mice were divided into four groups (

n

=5),and injected by NS,naked siRNA (0.3 mg/kg),GCE/siRNA (containing 0.3 mg/kg siRNA),and DOX (1.2 mg/kg) every other day for 5 times.After the drug administration,blood samples were collected from all nude mice,and blood routine test was performed by an automatic hematology analyzer (MEK6400,Nihon Kohden,Japan).

The whole blood was taken from the blood collection vessel without anticoagulant for blood biochemical examination.The whole blood was kept at room temperature for 2 h,and centrifuged (3 000 ×g,4 °C) for 10 min to obtain the upper serum.The important blood biochemical indicators of liver function (ALT:alanine aminotransferase,AST:aspartate aminotransferase,ALP:alkaline phosphatase,TB:total protein,and ALB:albumin) and renal function (CREA:creatinine,UA:uric acid,and UREA) of nude mice were detected by an automatic biochemical analyzer (BS-350E,Mindray,China).

2.20.Statistics

Data were represented as the Mean ± SD.The results were treated statistically by paired two-sample Student’s

t

-test.A value of

P

<

0.05 was considered statistically significant,and a value of

P

<

0.01 was considered very significant.

3.Results and discussion

3.1.Synthesis and characterization of GCE

The novel siRNA vector GCE was prepared as described in Scheme 1 A.The chemical structure of GCE nanoparticles was confirmed by FTIR (Fig.1 A) and UV spectroscopy (Fig.S1).Compared with the FTIR spectrum of GO-COOH (blue),there was a characteristic peak of C-N group at 1374 cmin that of GO-CS (yellow),and the

C

=

O

stretching vibration peak shifted from 1574 cm1 to 1558 cm1 .These results indicated that CS was successfully grafted onto GO-COOH surface.When compared the spectra of GO-CS and GCE (pink),the

C

=

O

stretching vibration peak shifted from 1558 cmto 1583 cm,and the absorption peak became stronger in the spectrum of GCE,which indicated that anti-EpCAM was successfully modified on GO-CS and GCE composite.The thermal stability of GO was also different when compared with GO-CS,GCE and siRNA loaded GCE,which further confirmed the successful preparation of GCE/siRNA (Figs.S2 and S3).

The structures of GO and GCE were observed by AFM (Fig.1 B) and TEM (Fig.1 D).The diameter of unmodified GO was in the range of 70–280 nm,while the diameter of GCE increased to 80–350 nm.The thickness was also increased after modification.These results indicated that CS and anti-EpCAM had been successfully modified on the surface of GO.Meanwhile,the edges of GO particles are sharper but thinner than that of GCE particles,mainly due to the packaging and folding of polymer CS.The average hydrodynamic radius and zeta potential of GO,GO-CS,GCE and GCE/siRNA nanoparticles were determined by DLS (Fig.1 C).In our previous research,GO was decorated with cationic materials such as octaarginine (R8) and poly-L -lysine (PLL) for siRNA delivery [45–47].The zeta potential of CS functionalized GO was (41.94 ±0.34) mV (Fig.1 C),which is even more positive than R8 or PLL functionalized GO.This character makes GO-CS a promising vector for siRNA.

3.2.Drug loading and stability analysis

GCE/siRNA nanoparticles at the different weight ratios (2:1,5:1,10:1,20:1,30:1,40:1,and 50:1) were prepared.In the group of GO/siRNA,the bright siRNA bands were observed at all ratios.While in GCE/siRNA group,the bands of GCE/siRNA disappeared completely when the ratio reached 30:1 (Fig.2 B).These results suggested that 1.0 mg of siRNA could be loaded to 30 mg GCE carrier completely.Based on the agarose gel retardation assay,GCE/siRNA nanoparticles by a weight ratio of 30 were used in the following experiments.

Fig.2–Drug loading ability and stability of GCE.(A) The stability and dispersibility of GO (a),GCE (b),Tyndall phenomenon of GO and GCE (c).(B) Agarose gel retardation test of siRNA treated with GO (a) and GCE (b).(C) Degradation of naked siRNA (a) and GCE/siRNA (b) with Heparin and RNase A.

Fig.3–Intracellular delivery and gene silencing of GCE/siRNA in vitro .(A) Confocal images treated by PBS,naked siRNA,GCE/siRNA,and Lipo/siRNA;siRNA was labled with FAM fluorescent molecule (green) to show the amount of siRNA into cells,and Hoechst 33,342 (blue) for cell nucleus.(B) Survivin-mRNA levels of the MCF-7 cells treated with different survivin-siRNA formulations were tested by RT-PCR.(C) Survivin protein levels treated with different survivin-siRNA formulations were tested by ELISA assay.The cells with no treatment was considered as a control.The data were presented as the mean ±SD,n=3.(For interpretation of the REFERENCES to colour in this figure legend,the reader is referred to the web version of this article.)

The stability of GCE was tested in different solutions (PBS,NS and RPMI-1640 medium).As shown in Fig.2 A,GO was precipitated in all the solutions,while GCE could disperse homogeneously,indicating that the modification could increase the stability of GO.The brilliant stability of GCE may be due to the increase of the layer resistance effect and the electrostatic shielding effect.Moreover,the result of cumulative release proved that GCE/siRNA had a sustained release effect for siRNA,which could prolong the effect of siRNA

in

vivo

(Fig.4 S).

Fig.4–Safety of GCE and antiproliferation effect of GCE/siRNA in vitro .(A) The cytotoxicity of GCE on MCF-7 cells using MTT method.(B) Anti-proliferation effect of GCE/siRNA in MCF-7 cells.The results were presented as the mean ±SD,n=3.(C) Effect of GCE/siRNA on apoptosis morphology of MCF-7 cells.(D) Cell cycle distribution of MCF-7 cells treated by survivin-siRNA for 48 h.(a) Distribution of MCF-7 cell cycle detected by flow cytometry in different treatment groups.(b) Statistical chart of cell cycle distribution of MCF-7 cells in different treatment groups.(E) The apoptosis of MCF-7 cells induced by siRNA for 48 h.(a) Flow cytometric plot of rate of apoptosis by annexin V and PI staining of MCF-7 cells.(b) Apoptosis rate of MCF-7 cells in different treatment groups.The cells with no treatment was considered as a control.The data were presented as the mean ±SD,n=3.

The protective effect of GCE on siRNA was also evaluated by gel retardation assay.The band of siRNA was degraded by RNase A within 5 min.For GCE/siRNA,a bright band of survivin-siRNA could still be observed from GCE/siRNA even after incubating with RNase A for 6 h (Fig.2 C).These results indicated that GCE vector could effectively protect survivin-siRNA from degradation

in

vivo

.As it’s reported,one of the main elements for a successful siRNA delivery system is protection of siRNA from degradation [48].The RNase A degradation study suggested that GCE is a promising system for siRNA delivery.

3.3.Intracellular delivery and in vitro gene silencing effect of GCE/siRNA

Efficient cellular uptake is a critical step for the gene silencing effect of survivin-siRNA.FAM-siRNA was applied to trace the siRNA uptake in MCF-7 cells.As shown in Fig.3 A,the green fluorescence of FAM in GCE/siRNA were much stronger than that in control and naked siRNA group.The fluorescence of FAM distributed around cell nucleus.The result also displayed that GCE/siRNA had the same delivery behavior as Lipo/siRNA group (positive control group),suggesting that GCE vector could deliver survivin-siRNA into MCF-7 cells effectively.

Fig.5–Migration and invasion inhibition effects of GCE/siRNA on MCF-7 cells.(A) Effect of survivin-siRNA on MCF-7 cells migration by wound healing assay.(a) Photographs of MCF-7 cells in different treatment groups at 0,24 and 48 h by wound healing assay.(b) Statistical chart of cell migration rate of MCF-7 cells in different treatment groups.(B) Migration by transwell assay.(a) Control,(b) GCE/NC,(c) GCE/siRNA,(d) Statistical chart of number of migrated MCF-7 cells in different treatment groups.(C) Effect of invasion by transwell assay.(a) Control,(b) GCE/NC,(c) GCE/siRNA,(d) Statistical chart of number of invaded MCF-7 cells in different treatment groups.The cells with no treatment were considered as a control.The data were presented as the mean ±SD,n=3.

The

in

vitro

gene silencing effect of GCE/siRNA was confirmed by RT-PCR analysis.Compared with control and negative controls (naked siRNA,GCE/NC,Lipo/NC),the relative amount of mRNA in GCE/siRNA group was downregulated by 48.24% ± 2.50%,and that was lower than Lipo/siRNA group (45.76% ± 1.98%,

P

<

0.01) (Fig.3 B).The results showed that GCE/siRNA downregulate the expression of RNA effectively.The downregulated expression of protein was determined by ELISA analysis (Fig.3 C).Compared with the control,the expression level of survivin-protein in GCE/siRNA group was decreased (

P

<

0.01).The downregulation rates of the protein were 44.12% ± 3.03% and 41.77% ± 1.97% for GCE/siRNA and Lipo/siRNA,respectively.There was no significant difference on protein downregulation rates for GCE/siRNA group and Lipo/siRNA group (

P

>

0.05).The above results showed that GCE/siRNA could regulate the expression of protein.

3.4.In vitro antiproliferation assay

A series of antiproliferation assays were carried out to explore the inhibition effect of survivin-siRNA in MCF-7 cells.Firstly,the toxicity of GCE carriers was tested by MTT assay,suggesting that GCE had no effect on the MCF-7 cells (Fig.4 A).Then,the antiproliferation effect of GCE/siRNA was tested (Fig.4 B).As the increase of siRNA concentration,the survival rate of MCF-7 cells in GCE/siRNA group decreased significantly,which had the same effect as Lipo/siRNA.The results indicated that GCE/siRNA had excellent antiproliferation effect

in

vitro

.Besides,the apoptotic morphology observed by CLSM also suggested that GCE/siRNA treated could make cell nucleus shrinking and irregular (Fig.4 C),which was a typical character of apoptotic nuclear morphology [49,50].All of these results above indicated that the GCE/siRNA could inhibit MCF-7 cells growth by antiproliferation and cell apoptosis effects.

In order to explore the apoptosis mechanisms of GCE/siRNA,cell apoptosis and cell cycle analysis were performed by FCM.The results of cell cycle were shown in Fig.4 D.As for GCE/siRNA treated group,the G0/G1 proportion of cells increased greatly.The cell cycle analysis suggesting that GCE/siRNA could inhibit cell proliferation by blocking at G0/G1 phase.The results were shown in Fig.4 E.The apoptosis rates in control group and GCE/NC group were 3%.9 ± 1.3% and 5.8% ±2.1%,while the apoptosis rate in GCE/siRNA group was increased remarkably to 23.9% ± 2.6%.The results indicated that GCE/siRNA could induce cell apoptosis in MCF-7 cells effectively.

Fig.6–In vivo fluorescence image of survivin-siRNA injected into tail of BALB/c nude mice.(A) Images were taken at different time point after tail vein injection.(B) Image of collected organs from each group was taken at 8 h after tail vein injection.

3.5.Cell migration and invasion assays

It was reported that survivin-siRNA could suppressed tumor growth by inhibiting cell migration and invasion [51,52].So wound healing assay and transwell assay were performed to evaluate the effects of GCE/siRNA on cell migration and invasion.As shown in Fig.5 A,the migration rate of GCE/siRNA group was significantly inhibited at 1 d and 2 d (

P

<

0.01).It demonstrated that downregulation of survivin expression could inhibit MCF-7 cell migration.The migration inhibition effect of survivin-siRNA was also confirmed by transwell assay (Fig.5 B).The migration rate of MCF-7 cells treated with GCE/siRNA was 0.49 ±0.04,which was much lower than that in control and GCE/NC (

P

<

0.01).In order to confirm the invasion inhibition effect of GCE/siRNA,transwell chamber covered with matrigel was applied to simulate the invasion environment.The cell invasion rate of MCF-7 cells in GCE/siRNA was also lower significantly (

P

<

0.01) (Fig.5 C).All of the results showed that GCE/siRNA inhibit the cell migration and cell invasion by silencing survivin expression.

3.6.In vivo drug distribution

The successful delivery of survivin-siRNA into tumor tissues is the main step for the anti-tumor effect of survivin-siRNA,so targeting effect and distribution

in

vivo

were investigated in MCF-7 xenograft models.After intravenous injection of NS (Control),naked Cy5-siRNA (Naked siRNA),GC/Cy5-siRNA (GC/siRNA) and GCE/Cy5-siRNA (GCE/siRNA),fluorescence of the tumor-bearing mice was performed at different time points (1,2,4 and 8 h).After intravenous injection for 2 h,there was fluorescence in tumor for GC/siRNA and GCE/siRNA groups,and the fluorescence intensity increased gradually (Fig.6 A).It suggested that survivin-siRNA loaded on carriers could accumulate in tumor through the blood circulation system of mice.Interestingly,at 2,4 and 8 h,the fluorescence intensity of tumor in GCE/siRNA group was higher than that in GC/siRNA group.Compared with GC/siRNA,GCE/siRNA was modified with anti-EpCAM,which is a specific antibody for highly expresses EpCAM molecule on the surface of MCF-7 cells [53].So it was confirmed that modification with anti-EpCAM ensured the targeting effect of GCE/siRNA

in

vivo

.After 8 h,the mice were executed,and the organs were gotten to observe the siRNA distribution.As shown in Fig.6 B,all of the tested groups had a strong fluorescence in kidney,indicating that the kidney was the main metabolic organ for siRNA.While for GC/siRNA group and GCE/siRNA group,there was also distinct fluorescence in liver.It was speculated that graphene oxide nanoparticles were mainly excreted through bile after liver metabolism.In addition,the fluorescence intensity of the main organs in GCE/siRNA group was weaker than that in GC/siRNA group,which further confirmed that more survivin-siRNA was accumulated in tumor tissue.

3.7.Tumor inhibition assays

According to the results above,it suggested that GCE/siRNA would have a brilliant anti-tumor effect

in

vivo

,therefore,the tumor inhibition effect of GCE/siRNA was investigated.It could be seen that the tumor inhibition effect of GCE/siRNA was higher significantly (

P

<

0.01),and GCE/siRNA had an equal anti-tumor effect as DOX (Fig.7 A).And as shown in Fig.7 B,the tumor volumes of GCE/siRNA and DOX were smaller than those of control and siRNA groups.In addition,the tumor inhibition rate of GCE/siRNA group 54.74% ±5.51% were significantly higher than those of siRNA group 4.87 ±8.49%.These results indicated that GCE/siRNA could have a positive therapeutic effect on tumor inhibitory

in

vivo

.What’s more,according to the ratios of organ to body weight,GCE/siRNA would not induce serious damage to the main organs

in

vivo

(Fig.7 C).

Fig.7–In vivo tumor therapeutic efficiency of GCE/siRNA.(A) Tumor volume changes of mice were measured every two days.(B) Image of tumors harvested from the mice on the day after the last injection.(C) The ratios of organ weight to body weight.(D) H&E staining of MCF-7 tumor tissues in nude mice injected with (a) NS (Control),(b) naked siRNA,(c) GCE/siRNA,(d) DOX.(100 ×).The data were presented as the mean ±SD,n=6.

H&E staining was applied to revealing the histological change by GCE/siRNA.It was obvious that the tumor cells in control group and naked siRNA group were closely arranged.In GCE/siRNA group,the tumor cells became sparse,and the intercellular space became large,which were the typical characters of necrosis.And this necrosis phenomenon was also observed in DOX group (Fig.7 D).These results showed that GCE/siRNA effectively inhibit the proliferation of MCF-7 tumor by promoting cell apoptosis.

3.8.Safety study in vivo

Since most of siRNA delivery systems are constructed by cationic materials,they may cause serious blood and tissue toxicity after administration

in

vivo

.So it is necessary to evaluate the biosafety of siRNA vectors.Firstly,the organs were stained by H&E to explore the organ toxicity of GCE/siRNA.As shown in Fig.8 A,there was no obvious pathological feature in heart,liver,spleen,lung and kidney of each group,suggesting that GCE/siRNA had no obvious toxicity to the main organs

in

vivo

.The hemolysis test also suggested that GCE/siRNA had good biocompatibility.It was safety to RBC even when siRNA increased to a high concentration (160 nM) (Fig.8 B).Since it has been reported that GO could induce hemolysis [54],the hemolysis test proved that proper surface modification could effectively improve the biocompatibility of GO.Besides,blood biochemistry test was carried out to research the hepatic or renal toxicity of GCE/siRNA.As shown in the Fig.8 C,neither the liver function indexes (ALT,AST,ALP,TB and ALB) nor the renal function indexes (CREA,UA and UREA) between treatment groups and control group had any significant difference in function.Moreover,the blood routine analysis was also performed to further prove the blood safety of GCE/siRNA.The results revealed that there was no significant difference in blood indexes between each treatment group and control group (Table 1).It was widely reported that cationic gene drug carriers would induce blood toxicity by hemolysis as well as organic damage after administration

in

vivo

[55].However,the positive charged GCE was demonstrated to be a brilliant biosafety vector both

in

vitro

and

in

vivo

,which may be due to the non-toxicity and good biocompatibility of CS decorated on GO.According to previous research,CS was a non-toxicity positive material with brilliant biocompatibility,and decorated with CS would significantly reduce toxicity of drug delivery matertials [56].So the CS modified GCE exhibited adorable biosafety as a gene drug delivery system.

Table 1– Blood routine test.The data were presented as the mean ±SD,n=5.

Fig.8–Biosafety of GCE/siRNA in vivo .(A) H&E staining of the major organs from mice treated with NS (Control),naked siRNA,GCE/siRNA and DOX.(100 ×).(B) Hemolysis activity of GCE/siRNA,hemolysis rate was evaluated in rat RBC after treatment with naked siRNA,GCE,GCE/siRNA or Lipo/siRNA.The data were presented as the mean ±SD,n=3.(C) Blood biochemistry data including liver function markers:ALT,AST,ALP,TB and ALB,and kidney function markers:CREA,UA and UREA.The data were presented as the mean ±SD,n=5.

4.Conclusion

In summary,a graphene oxide-based siRNA delivery vector was successfully synthesized.The novel siRNA delivery system could not only deliver siRNA into tumor cellular efficiently

in

vitro

,but also target siRNA into tumor tissues successfully

in

vivo

.GCE exhibited excellent stability under different conditions,and could protect siRNA from degradation by RNase A.The cellular uptake research suggested that GCE could promote the uptake of survivin-siRNA by MCF-7 cells,which had almost the same effect as Lipo2000.The efficient delivery effect of GCE/siRNA ensured the gene silencing efficiency of survivin-siRNA.The downregulation rates of survivin-mRNA and survivin-protein had no significant differences between GCE/siRNA and Lipo/siRNA.The

in

vitro

experiments proved that GCE/siRNA could suppress MCF-7 cell proliferation,which is attributed to the apoptosis promoting and metastatic inhibition effects of siRNA.Moreover,GCE exhibit a brilliant tumor targeting effect,since the modification of anti-EpCAM.Besides,the toxicity and biosafety research

in

vitro

and

in

vivo

were carried out.The results indicated that GCE was a safety material,and had no damage to the main organs as a gene delivery system.All of the results in this research depict a novel vector with remarkable gene delivery ability,and also highlight the importance of rational delivery-system design for RNA interference technology in cancer treatment.

Conflicts of interest

The authors report no conflicts of interest.The authors alone are responsible for the content and writing of this article.

Acknowledgments

This work was supported by the National Natural Science Foundation (81502688),Beijing Natural Science Foundation Program and Scientific Research Key Program of Beijing Municipal Commission of Education (KM201810025019),and a basic-clinical key research Grant (16JL72,17JL67) from Capital Medical University,the Importation and Development of High-Caliber Talents Project of Beijing Municipal Institutions (2013–2015),Natural Science Foundation of Capital Medical University (1210020232).The authors gratefully acknowledge the support from Beijing Area Major Laboratory of Peptide and Small Molecular Drugs,Engineering Research Center of Endogenous Prophylactic of Ministry of Education of China,and Beijing Laboratory of Biomedical Materials.

Supplementary materials

Supplementary material associated with this article can be found,in the online version,at doi:10.1016/j.ajps.2021.04.002 .