In Vivo Ovarian Cancer Gene Therapy Using CRISPR-Cas9

Abstract

Clustered regularly interspaced short palindromic repeats (CRISPR)-caspase 9 (Cas9) genome editing technology holds great promise for the field of human gene therapy. However, a lack of safe and effective delivery systems restricts its biomedical application. Here, a folate receptor-targeted liposome (F-LP) was used to deliver CRISPR plasmid DNA co-expressing Cas9 and single-guide RNA targeting the ovarian cancer–related DNA methyltransferase 1 (DNMT1) gene (gDNMT1). F-LP efficiently bound the gDNMT1 plasmid and formed a stable complex (F-LP/gDNMT1) that was safe for injection. F-LP/gDNMT1 effectively mutated endogenous DNMT1 in vitro, and then expressed the Cas9 endonuclease and downregulated DNMT1 in vivo. The tumor growth of both paclitaxel-sensitive and -resistant ovarian cancers were inhibited by F-LP/gDNMT1, which shows fewer adverse effects than paclitaxel injection. Therefore, CRISPR-Cas9-targeted DNMT1 manipulation may be a potential therapeutic regimen for ovarian cancer, and lipid-mediated delivery systems represent promising delivery vectors of CRISPR-Cas9 technology for precise genome editing therapeutics.

Introduction

The clustered regularly interspaced short palindromic repeats (CRISPR)-caspase 9 (Cas9) bacterial immune system is a robust and highly efficient novel genome editing tool that can be repurposed to generate deletions, insertions, and replacements easily in the mammalian genome.^1

–4 Therefore, the CRISPR-Cas9 system holds great promise for inactivating or repairing disease-causing genes and has reinvigorated the field of human gene therapy.^5
–7 CRISPR-Cas9 has been used to treat a variety of diseases at the preclinical level, such as hereditary tyrosinemia,⁸ Wiskott–Aldrich syndrome,⁹ human immunodeficiency virus infection,^10,11 cardiovascular diseases and lower levels of low-density lipoprotein cholesterol,¹² neuronal ceroid lipofuscinosis,¹³ surfactant protein B deficiency and experimental asthma,¹⁴ hematologic disorders,^15,16 Duchenne muscular dystrophy,¹⁷ and so on. In the field of cancer research, CRISPR-Cas9 has been widely applied for identifying gene functions and generating animal models of cancer.^18

–21 Furthermore, CRISPR-Cas9 has shown great potential in cancer gene therapy, such as disrupting tumor-promoting mutations and inactivating oncogenic virus genes by introducing insertion and deletion mutations (indels) at specific genome loci of cancer cells.^22
–24

DNA methyltransferase 1 (DNMT1) is a key nuclear enzyme of DNA methylation and an important molecular target for epigenetic therapy of cancer.^25
–27 The abnormal overexpression of DNMT1 inactivates tumor suppressor genes and is indispensable for the maintenance of cancer stem cells.^28,29 Moreover, DNMT1 closely correlates with tumorigenesis, relapse, and resistance of ovarian cancer, and high DNMT1 levels have been frequently observed in patients with poor prognoses.^30,31 Inconsistent with this observation, a series of studies have shown that inhibition of DNMT1 suppressed ovarian cancer growth and resistance.^32
–34 Consequently, DNMT1 is a promising molecular target of ovarian cancer treatment.^35,36 As a targeted genome editing technique, CRISPR-Cas9 has been used to edit endogenous genes in a wide variety of cell types and in organisms.^24,37,38 Therefore, using the CRISPR-Cas9 system to edit DNMT1 of ovarian cancer cells is a potential approach for ovarian cancer therapy.

Delivery systems, including lentiviral, adenoviral, adeno-associated viral, naked plasmid DNA, nanoparticles, and cationic liposomes, have been employed and are known to work well in CRISPR-Cas9-mediated research.^37
–39 However, the lack of safe and effective delivery systems restricts further application of CRISPR-Cas9 in cancer gene therapy.^40,41 To enable the application of Cas9 in vivo broadly, new delivery systems must be researched and developed.⁴² Ideally, Cas9 delivery vehicles should be safe, induce little to no immune response, have a short half-life and few limitations when delivering larger genetic payloads, not integrate into the genome, and efficiently regulate target cell functions.^14,37,42 Cationic liposomes as a nonviral delivery system are easily modified in order to meet the requirements of an ideal Cas9 delivery vehicle.^42,43 Furthermore, cationic liposomes have been utilized to deliver Cas9:single-guide RNA (sgRNA) complexes into cultured human cells in vitro and a mouse inner-ear in vivo and efficiently modified the target genomes (80% in vitro, 20% in vivo).⁴⁴ Therefore, cationic liposomes are an available vector for CRISPR-Cas9 delivery.

In the present study, a folate receptor-targeted cationic liposome (F-LP) was developed to deliver a CRISPR plasmid expressing Cas9 and sgRNA targeting the ovarian cancer-related DNMT1 gene (gDNMT1). The physicochemical properties of the F-LP/gDNMT1 complex were characterized in vitro. Then, the ability of these complexes to edit the genome of ovarian cancer cells precisely was evaluated both in vitro and in vivo. Furthermore, using paclitaxel (PTX) as a positive control, the antitumor effects of F-LP/gDNMT1 were also explored in a mouse model of ovarian cancer.

Materials and Methods

Materials

The pX330 plasmid containing a human-encoded optimized Cas9 and sgRNAs was purchased from Addgene (pX330-U6-Chimeric_BB-CBh-hSpCas9; Addgene: 42230; Addgene, Cambridge, MA).⁴⁵ BbsI, T4 polynucleotide phosphorylase, and T4 ligase were obtained from Life Technologies (Carlsbad, CA). 1, 2-Dioleoyl-3-trimethylammonium-propane chloride salt (DOTAP) was purchased from Avanti Polar Lipids, Inc. (Alabaster, AL), and cholesterol (Chol) was obtained from Shanghai Bio Life Science & Technology Co. Ltd. (Shanghai, China). Methoxy-polyethylene glycol (PEG)-succinyl-Chol and folate-PEG-succinyl-Chol conjugates were synthesized and purified, as previously described.^46,47 F-LP was composed of DOTAP, Chol, methoxy-PEG-succinyl-Chol, and folate-PEG-succinyl-Chol and was prepared by film dispersion, as previously described.^48,49 Coumarin was incorporated into F-LP to prepare F-LP-coumarin for in vivo imaging. Plasmid DNA was extracted according to the EndoFree Plasmid Purification Handbook (Qiagen, Hilden, Germany). Pre-stained protein and DNA ladders were purchased from Fermentas (Thermo Fisher Scientific, Waltham, MA). Glucose (5%) and NaCl (0.9%) injections were obtained from Sichuan Kelun Pharmaceutical Co. Ltd. (Chengdu, China). PXT injection was purchased from Taiji Group Sichuan Taiji Pharmaceutical Co. Ltd. (Chengdu, China).

sgRNA design and Cas9 vector construction

For targeting DNMT1 with the pX330 Cas9 vector, sgRNA sequences (Table 1) were designed using online software (http://zifit.partners.org/ZiFiT/). The enhanced green fluorescent protein (eGFP) guide sequence has been described previously.⁵⁰ All guide sequences are summarized in Table 1. The sense and antisense oligonucleotides of guide sequences were phosphorylated and annealed in 1 × T4 ligase buffer and subjected to thermal cycling as follows: 37°C for 30 min, 95°C for 5 min, and then ramping down to 25°C at 5°C intervals per minute. The pX330 vector was digested by BbsI at 37°C overnight and then purified with a gel extraction kit. The phosphorylated and annealed oligonucleotides were ligated with the pX330 vector with T4 ligase at 16°C overnight.

Table 1.

gRNA sequences

Target names	Vector	Guide sequences
eGFP-gRNA	pX330	GGCGAGGGCGATGCCACCTA
DNMT1-gRNA1	pX330	GGAGACGTGGAGATGCTGTG
DNMT1-gRNA2	pX330	GCTGCCCCAGAAGGGAGACG
DNMT1-gRNA3	pX330	GCGTTCCGGCTGAACAACCC
DNMT1-gRNA4	pX330	GGCCAGCGGCTGCCCCAGAA

eGFP, enhanced green fluorescent protein; DNMT1, DNA methyltransferase 1.

Preparation and characterization of lipoplexes

FR-targeted lipoplexes were prepared by mixing F-LP or F-LP-coumarin with plasmid DNA for 30 min at room temperature. The appearance of F-LP and lipoplexes were photographed with a digital camera (Canon, Tokyo, Japan). The mean particle size, polydispersion index, and zeta potential of lipoplexes were determined by a Zetasizer NanoZS ZEN 3600 (Malvern Instruments Ltd., Malvern, United Kingdom). Morphological characteristics of the F-LP/gDNMT1 complex were examined by a Tecnai G² F20 transmission electron microscope (FEI Co., Hillsboro, OR). After lipoplexes were prepared, agarose gel electrophoresis was conducted in pH 7.4 Tris-base/acetic acid/EDTA buffer containing Gold View nucleic acid stain, as previously described.⁴⁹

A hemolysis study of F-LP/gDNMT1 complexes was performed in normal saline solution, with plasmid weights ranging from 2.5 to 15 μg. Briefly, lipoplexes with various plasmid weights were diluted in 2.5 mL of normal saline and added to 2.5 mL of rabbit erythrocyte suspension (2%) in normal saline at 37°C; normal saline and distilled H₂O were used as negative and positive controls, respectively. The erythrocyte suspensions were centrifuged 3 h later, and the color of supernatants were compared to that of the controls.⁵¹

Cell culture

SKOV-3 human ovarian cancer cells were obtained from American Type Culture Collection (Manassas, VA). The cells were cultured as a monolayer in Dulbecco's modified Eagle's medium (DMEM) supplemented with 10% fetal bovine serum in a humidified atmosphere containing 5% CO₂ at 37°C. Human ovarian cancer PXT-resistant SKOV-3/TR cells were selected from SKOV-3 cells by step-wise increases in PXT concentration (0.1–300 nM).⁵² An eGFP control lentivirus was used to infect SKOV-3 cells to generate eGFP-stable cells (SKOV-3^eGFP+), and stable clones were selected by green fluorescence sorting to purify SKOV-3^eGFP+ cells.

In vitro transfection experiments

In vitro transfection experiments were performed, as previously described.⁵³ F-LP/gDNMT1 and F-LP carrying CRISPR plasmid DNA co-expressing Cas9 and sgRNA targeting eGFP (geGFP) (F-LP/geGFP) were used to transfect SKOV-3 and/or SKOV-3^eGFP+ cells. SKOV-3 cells transfected with F-LP/gDNMT1 were stained with an anti-DNMT1 antibody (1:100; Proteintech Group, Inc., Chicago, IL) and 4′,6-diamidino-2-phenylindole and photographed using a Leica TCS SP5 confocal laser-scanning microscope (Leica Microsystems CMS GmbH, Mannheim, Germany). After cellular nuclei were stained with Hoechst 33258, the transfection efficiency of F-LP/geGFP into SKOV-3^eGFP+ cells was determined using an Olympus IX 70 inverted fluorescence microscope (Olympus Corp., Shinjuku, Tokyo, Japan) and FACSCalibur flow cytometry (BD Biosciences, San Jose, CA).

Target site (DNMT1) identification

Genomic DNA was prepared from SKOV-3 cells transfected with an empty CRISPR vector or CRISPR vectors harboring DNMT1 target sequences. DNA sequences containing each sgRNA target site were amplified by polymerase chain reaction (PCR). These PCR products were analyzed by T7E1 or restriction enzyme digestion, and the DNA was sequenced to screen for possible effective sgRNAs and quantify the indel efficiency.^54,55 The forward and reverse PCR primers for DNMT1 were 5′-TCTTCTGGAGAGTTGAAAGC-3′ and 5′-GGGCTTCTACCCGTTTACC-3′, respectively.

DNA dot blot

DNA dot blot was performed, as described by Blaschke et al. ⁵⁶ Briefly, isolated genomic DNA was denatured in 0.1 M of NaOH for 10 min at 95°C, neutralized with 1 M of NH₄CH₃CO₂ on ice, and then serially diluted twofold in TE buffer (10mM Tris-HCl 1mM EDTA PH 8.0). DNA samples were spotted on a nitrocellulose membrane and dried at 80°C for 5 min. The membrane was subjected to ultraviolet crosslinking at 120,000 μJ/cm², blocked in blocking buffer (Tris-buffered saline/Tween 20 with 5% milk) at room temperature for 2 h, and then incubated with a mouse anti-5-methylcytosine monoclonal antibody (1:500; Active Motif, Carlsbad, CA) at 4°C overnight. After washing the membrane three times for 10 min each in Tris-buffered saline/Tween 20, it was incubated with sheep anti-mouse horseradish peroxidase-conjugated immunoglobulin G secondary antibody (1:2,000; ZSGB-BIO, Beijing, China) for 3 h at room temperature. The membrane was then washed three times for 10 min each in Tris-buffered saline/Tween 20 and visualized with a ChemiScope 5300 Integrated Chemiluminescence Imaging System (Clinx Science Instruments Co. Ltd., Shanghai, China).

Animal experiments

Athymic female nude mice (BALB/c, 4–6 weeks old) were purchased from Vital River Laboratories (Beijing, China). The mice were housed and maintained under specific pathogen-free conditions. All mice were handled according to the Guide for the Care and Use of Laboratory Animals (National Research Council, US), and all experiments performed on animals followed the guidelines and regulations of Sichuan University (Chengdu, China). All procedures were approved and supervised by the State Key Laboratory of Biotherapy Animal Care and Use Committee (Sichuan University, Chengdu, China).

Experiments with these mice began when they were 6–8 weeks old. For in vivo imaging, F-LP/gDNMT1 tagged by coumarin was injected into the abdominal cavity. After 6 or 12 h, the vital organs (heart, liver, spleen, lung, kidney, brain, stomach, and intestinal tract) were harvested and observed under a Bio-Real QuickView 3000 in vivo imaging system (Labatech GmbH, Salzburg, Austria). Tumor models were established by intraperitoneal (i.p.) injection of SKOV-3 or SKOV-3/TR cells (5 × 10⁶ cells in 0.2 mL serum-free DMEM), and mice were randomly allocated into four groups. To assess tumor growth, treatment began 7 days after inoculation. Mice were administered i.p. every 3 days with liposomal plasmid DNA (2.5 or 5 μg) in 200 μL of glucose solution. Tumor development (n = 8 per group) were monitored every 3 days. Two days after the 10th treatment, mice were anesthetized and sacrificed. Tumor tissues were harvested and processed, as previously described.^48,57

Reverse transcription PCR

For reverse transcription (RT)-PCR, total RNA from tumor tissues, which were treated as indicated in the section “In vivo gene editing using F-LP/gDNMT1,” was extracted using TRIzol reagent (Invitrogen, Carlsbad, CA) and quantified by spectrophotometry. cDNA was synthesized using a RevertAid First Strand cDNA Synthesis kit (Thermo Fisher Scientific, Rockford, IL) following the manufacturer's guidelines. The primers were designed as follows: Cas9, 5′-ACCATCGACCGGAAGAGGTA-3′ (forward) and 5′-CGATCCGTGTCTCGTACAGG-3′ (reverse); β-actin, 5′-AGCGAGCATCCCCCAAAGTT-3′ (forward) and 5′-GGGTGGCTTTTAGGATGGCA-3′ (reverse).

Western blot and hematoxylin and eosin staining

Total protein concentrations of tumor tissue lysates were determined with a Pierce BCA assay reagent kit (Thermo Fisher Scientific). DNMT1, Cas9, heterochromatin protein (HP)1α and p15^INK4b proteins were separated by 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis under reducing conditions and then transferred to polyvinylidene difluoride membranes (Millipore Corp., Billerica, MA). Membranes were blocked with 5% skim milk and incubated with anti-DNMT1 (Proteintech Group, Inc.), anti-Cas9, anti-HP1α, and anti-p15^INK4b primary antibodies (Abcam, Boston, MA) at 4°C overnight. Antibodies were detected with a horseradish peroxidase-conjugated secondary antibody and developed with an enhanced chemiluminescence detection kit (Luminata Crescendo Western horseradish peroxidase substrate; Millipore Corp.); β-tubulin was used as a control. Tumor tissue sections were also stained with hematoxylin and eosin for histomorphometric analysis. All sections were observed or counted by two investigators or pathologists in a blinded fashion.

Statistical analysis

Statistical analysis was performed with one-way analysis of variance using IBM SPSS Statistics for Windows v19.0 (IBM Corp., Armonk, NY). Differences were considered statistically significance at p < 0.05.

Results and Discussion

In vitro reporter gene editing based on CRISPR-Cas9 system

Ovarian cancer SKOV-3 cells infected by lentivirus-encoded eGFP and eGFP-stable cells SKOV-3^eGFP+ were purified after green fluorescence sorting. Fluorescence microscopy (Fig. 1a) and flow cytometric analysis (Fig. 1b) showed that eGFP expression was disrupted in about 70% of SKOV-3^eGFP+ cells after F-LP/geGFP transfection. This in vitro genome modification efficiency was similar to that mediated by other cationic lipid/Cas9:sgRNA nuclease complexes.⁴⁴ These results indicate that F-LP is a potential nonviral vector for delivering the CRISPR-Cas9 system. Therefore, the CRISPR plasmid was further designed to target DNMT1, and F-LP was utilized to deliver the gDNMT1 plasmid and disrupt DNMT1 expression in ovarian cancer cells.

Figure 1.

Targeted genome editing for reporter gene based on folate receptor-targeted liposome (F-LP) delivery. (a) Fluorescence images of F-LP mediated enhanced green fluorescent protein (eGFP) reporter gene editing based on the clustered regularly interspaced short palindromic repeats (CRISPR)-caspase 9 (Cas9) system (arrows point to eGFP-negative cells). (b) Flow cytometric analysis of F-LP-mediated eGFP reporter gene editing based on the CRISPR-Cas9 system.

Physicochemical properties of F-LP/gDNMT1 complexes

The gDNMT1 plasmid is large, with about 8,500 base pairs, and complete F-LP binding did not occur until the weight ratio between F-LP and gDNMT1 was 12:1 or 16:1 (Fig. 2a). The results showed that the lipoplexes (F-LP/vehicle and F-LP/gDNMT1) were formulated successfully and no free plasmid DNA was observed on the agarose gel (Fig. 2b). The appearances of the lipoplexes were clear under the light, similar to F-LP alone (Fig. 2c). The particle size and polydispersion index of F-LP/gDNMT1 were about 150 nm and 0.15, respectively (Fig. 2d and e). The morphology of F-LP/gDNMT1 was spheroidal or rod-like, and F-LP/gDNMT1 exhibited a membrane structure via transmission electron microscopy (Fig. 2f, classical morphology indicated by red arrows). The zeta potential of F-LP/gDNMT1 was about 35 mV, with a uniform distribution (Fig. 2g and h). The results of hemolytic tests demonstrated that F-LP/gDNMT1 did not cause hemolysis under test conditions (Fig. 2i). Overall, the F-LP/gDNMT1 complex displayed good physicochemical properties for injection. Therefore, F-LP/gDNMT1 was further evaluated for targeted genome editing in ovarian cancer cells in vitro.

Figure 2.

Preparation and characterization of F-LP/ DNA methyltransferase 1 (DNMT1) gene (gDNMT1) complexes. (a) The optimal weight ratio between F-LP and gDNMT1 Cas9 plasmid. When the ratio between F-LP and gDNMT1 reaches 12:1, F-LP completely binds with gDNMT1 Cas9 plasmid, and the lipoplex (F-LP/gDNMT1) is formulated. (b) Agarose gel electrophoresis of naked Cas9 plasmids and lipoplexes. (c) The appearance of liposome and lipoplexes. (d) The particle size of liposome and lipoplexes. (e) The size distribution of F-LP/gDNMT1. (f) The morphology of F-LP/gDNMT1 by transmission electron microscopy observation (arrows: classical lipoplexes). (g) The zeta potential of liposome and lipoplexes. (h) The zeta potential distribution of F-LP/gDNMT1. (i) Hemolytic test of F-LP/gDNMT1 (containing gDNMT1 2.5 μg, 5 μg, 10 μg, or 15 μg, respectively. Normal saline (NS) was the negative control, and water H₂O) was the positive control.

In vitro targeted genome editing of endogenous DNMT1

Four gDNMT1 plasmids containing different sgRNAs were constructed (Table 1), and then four F-LP/gDNMT1 complexes were transfected into SKOV-3 cells to identify functional DNMT1 sgRNA. After transfection, PCR and T7E1 assay were performed to determine if there were indels in sgRNA target sites within DNMT1. One out of four sgRNAs (DNMT1-gRNA1; Table 1) were found to be effective, and the indel percentage was 28.6% (Fig. 3a). Moreover, the resultant mutations were further verified by Sanger sequencing (Fig. 3c). DNA dot blots indicated that 5-methylcytosine was downregulated after SKOV-3 cells were transfected with F-LP/gDNMT1 (Fig. 3b). Endogenous DNMT1 was disrupted, and DNMT1 expression was inhibited according to immunofluorescence (Fig. 3d). To conclude, F-LP effectively delivered CRISPR plasmid DNA to ovarian cancer cells, and the CRISPR-Cas9 system expressed in ovarian cancer cells efficiently mediated the targeted inactivation of DNMT1 in vitro.

Figure 3.

In vitro targeted genome editing in the endogenous DNMT1 gene. (a) T7E1 assay of polymerase chain reaction (PCR) products. (b) Reduced 5-methylcytosine (5mC) level in the F-LP/gDNMT1-treated cells by DNA blot. (c) Read sequences obtained for DNMT1 after treatment with F-LP/gDNMT1. Wild-type DNMT1 sequence is shown in the top line. (d) The reduced expression of DNMT1 in the F-LP/gDNMT1-treated cells by immunofluorescence staining.

In vivo distribution and antitumor effects of F-LP/gDNMT1

F-LP/gDNMT1 tagged by coumarin was injected into the peritoneal cavity of mice to explore the distribution of F-LP/gDNMT1 in vivo. As shown in Fig. 4a, coumarin-tagged F-LP/gDNMT1 sustained its high levels at mesenteric tissues along the intestinal tract 12 and 24 h after administration i.p. Due to higher peritoneal levels of F-LP/gDNMT1 and its retention, the antitumor effects of F-LP/gDNMT1 was similar to PTX (1.25 mg/kg) and better than that of F-LP/vehicle (p < 0.01; Fig. 4b and c). Compared to the negative control, F-LP/gDNMT1 had a tumor inhibition rate of 84% and significantly inhibited the growth of PTX-sensitive tumors (p < 0.001; Fig. 4b and c). Because 1.25 mg/kg of PTX had little antitumor effect on PTX-resistant tumors, the dose was increased to 10 mg/kg. Moreover, F-LP/gDNMT1 also inhibited the growth of PTX-resistant tumors, whereas F-LP/vehicle did not (p < 0.05; Fig. 4d). Furthermore, F-LP/gDNMT1 showed an approximate antitumor effect like high-dose PTX (10 mg/kg) and reduced the visible toxicity of PTX (10 mg/kg) (Fig. 4e). Cationic lipid and bacterial plasmid DNA complexes could elicit adaptive cellular immunity in murine i.p. tumor models,⁵⁸ thus cationic liposomes (containing cationic lipid DOTAP 50%, mole ratio) and plasmid DNA complexes induced cellular immunity and activated macrophages to inhibit tumor growth in BALB/c nude mice in this study. The tumor growth inhibition rate elicited by F-LP/vehicle was 53.6% (0.41 vs. 0.22 g; Fig. 4b) in PTX-sensitive tumors and 45.9% (0.61 vs. 0.28 g; Fig. 4d) in PTX-resistant tumors compared to F-LP/gDNMT1 (100% in both tumor types). Thus, F-LP/gDNMT1 is a novel and promising antitumor therapeutic agent based on gene editing technology.

Figure 4.

In vivo distribution and antitumor effects of F-LP/gDNMT1 complexes. (a) In vivo distribution of F-LP/gDNMT1 tagged by Coumarin. (b and c) Tumor inhibition effect of F-LP/gDNMT1 on the paclitaxel (PTX)-sensitive tumor model (***p < 0.001, PTX [1.25 mg/kg], F-LP/Vehicle and F-LP/gDNMT1 v.s NS; ^†† p < 0.01, F-LP/gDNMT1 vs F-LP/vehicle, n = 8, M ± SD). (d and e) Tumor inhibition effect of F-LP/gDNMT1 on the PTX-resistant tumor model (*p < 0.05, ***p < 0.001, F-LP/vehicle, PTX [10 mg/kg] and F-LP/gDNMT1 vs. NS; ^† p < 0.05, F-LP/gDNMT1 vs. F-LP/vehicle, n = 6 for NS, n = 8 for PTX, F-LP/vehicle, and F-LP/gDNMT1, M ± SD).

In vivo gene editing using F-LP/gDNMT1

The presence of CRISPR-Cas9 transcripts in tumor tissues treated with F-LP/vehicle or F-LP/gDNMT1 was determined by RT-PCR (Fig. 5a). Furthermore, F-LP/gDNMT1 downregulated DNMT1 and HP1α expression in tumor tissues (Fig. 5a). Because DNMT1 interacts with HP1 resulting in increased DNA methylation,⁵⁹ low DNMT1 and HP1α expression implies that the CRISPR-Cas9 system based on F-LP/gDNMT1 worked to decrease DNA methylation in ovarian cancer cells in vivo. Furthermore, some necrotic cells were observed in tumor tissues treated with F-LP/gDNMT1 (Fig. 5b and c), suggesting that the antitumor effect of F-LP/gDNMT1 works to decrease methylation and induce necrosis of tumor cells.

Figure 5.

Reverse transcription (RT)-PCR, Western blot, and hematoxylin and eosin (H&E) staining of F-LP/gDNMT1-treated tumor tissues. (a) RT-PCR and Western blot of tumor tissues. (b) H&E staining of PTX-sensitive tumor tissues. (c) H&E staining of PTX-resistant tumor tissues (arrows: classic necrotic cells).

In conclusion, F-LPs loaded with a CRISPR-Cas9 plasmid were successfully prepared and transfected into cancer cells in vitro and in vivo. These cationic liposomes were able to deliver the plasmid efficiently and produce targeted genome editing based on CRISPR-Cas9 technology. Regarding application of CRISPR-Cas9 technology for cancer gene therapy, the present study created ovarian cancer–targeted F-LP/gDNMT1 complexes that were stable for injection. F-LP/gDNMT1 introduced indels into the genome of ovarian cancer cells, inactivated endogenous DNMT1, and reduced genome methylation within ovarian cancer cells in vitro. Furthermore, F-LP/gDNMT1 downregulated expression of DNMT1 and HP1α and then decreased methylation of genomic DNA in ovarian cancer tissues in vivo. In addition, F-LP/gDNMT1 also induced necrosis of cancer cells in vivo. Consequently, F-LP/gDNMT1 effectively inhibited growth of PTX-sensitive and -resistant ovarian tumors in vivo and showed few adverse effects compared to high-dose PTX injection. However, more studies on the nonspecific antitumor effects of the control F-LP/vehicle complex need to be performed. Overall, cationic liposomal vectors were shown to be effective delivery systems for CRISPR-Cas9 technology, and F-LP/gDNMT1 was found to be a promising cancer therapeutic using CRISPR-Cas9 technology to produce precise genomic editing of cancer cells and tissues.

Footnotes

Acknowledgments

This work was financially supported by the National Natural Science Foundation of China (No. 81602699), the National High-Tech R&D Program of China (Nos. 2015AA020309 and 2014AA020708), the China Postdoctoral Science Foundation funded project (No. 2015M570791), and Salubris Academician Workstation for Innovative Biopharmaceuticals (No. 2017B090904017).

Author Disclosure

No competing financial interests exist.

References

Slaymaker

, Gao

, Zetsche

, et al. Rationally engineered Cas9 nucleases with improved specificity. Science, 2016; 351:84–88.

Suzuki

, Tsunekawa

, Hernandez-Benitez

, et al. In vivo genome editing via CRISPR/Cas9 mediated homology-independent targeted integration. Nature, 2016; 540:144–149.

Barrangou

, Doudna

. Applications of CRISPR technologies in research and beyond. Nat Biotechnol, 2016; 34:933–941.

Modell

, Jiang

, Marraffini

. CRISPR-Cas systems exploit viral DNA injection to establish and maintain adaptive immunity. Nature, 2017; 544:101–104.

Nelson

, Hakim

, Ousterout

, et al. In vivo genome editing improves muscle function in a mouse model of Duchenne muscular dystrophy. Science, 2016; 351:403–407.

Dever

, Bak

, Reinisch

, et al. CRISPR/Cas9 beta-globin gene targeting in human haematopoietic stem cells. Nature, 2016; 539:384–389.

Zhang

. CRISPR-Cas9: prospects and challenges. Hum Gene Ther, 2015; 26:409–410.

Yin

, Xue

, Chen

, et al. Genome editing with Cas9 in adult mice corrects a disease mutation and phenotype. Nat Biotechnol, 2014; 32:551–553.

Gutierrez-Guerrero

, Sanchez-Hernandez

, Galvani

, et al. Comparison of ZFNs versus CRISPR specific nucleases for genome edition of the Wiskott–Aldrich syndrome locus. Hum Gene Ther, 2017 Oct 25 [Epub ahead of print]; DOI: 10.1089/hum.2017.047.

10.

Liao

, Gu

, Diaz

, et al. Use of the CRISPR/Cas9 system as an intracellular defense against HIV-1 infection in human cells. Nat Commun, 2015; 6:6413.

11.

, Yao

, Xiao

, et al. Simultaneous knockout of CXCR4 and CCR5 genes in CD4+ T cells via CRISPR/Cas9 confers resistance to both X4- and R5-tropic human immunodeficiency virus type 1 infection. Hum Gene Ther, 2018; 29:51–67.

12.

Ran

, Cong

, Yan

, et al. In vivo genome editing using Staphylococcus aureus Cas9. Nature, 2015; 520:186–191.

13.

Yao

, Liu

, Zhang

, et al. Gene therapy of adult neuronal ceroid lipofuscinoses with CRISPR/Cas9 in zebrafish. Hum Gene Ther, 2017; 28:588–597.

14.

Mahiny

, Dewerth

, Mays

, et al. In vivo genome editing using nuclease-encoding mRNA corrects SP-B deficiency. Nat Biotechnol, 2015; 33:584–586.

15.

, Natanson

, Dunbar

. Gene editing of human hematopoietic stem and progenitor cells: promise and potential hurdles. Hum Gene Ther, 2016; 27:729–740.

16.

Mussolino

, Alzubi

, Pennucci

, et al. Genome and epigenome editing to treat disorders of the hematopoietic system. Hum Gene Ther, 2017; 28:1105–1115.

17.

Long

, Amoasii

, Mireault

, et al. Postnatal genome editing partially restores dystrophin expression in a mouse model of muscular dystrophy. Science, 2016; 351:400–403.

18.

Xue

, Chen

, Yin

, et al. CRISPR-mediated direct mutation of cancer genes in the mouse liver. Nature, 2014; 514:380–384.

19.

Platt

, Chen

, Zhou

, et al. CRISPR-Cas9 knockin mice for genome editing and cancer modeling. Cell, 2014; 159:440–455.

20.

Sanchez-Rivera

, Jacks

. Applications of the CRISPR-Cas9 system in cancer biology. Nat Rev Cancer, 2015; 15:387–395.

21.

Van Der Weyden

, Arends

, Campbell

, et al. Genome-wide in vivo screen identifies novel host regulators of metastatic colonization. Nature, 2017; 541:233–236.

22.

Yao

, He

, Chen

. CRISPR/Cas9-mediated genome editing of epigenetic factors for cancer therapy. Hum Gene Ther, 2015; 26:463–471.

23.

Kennedy

, Kornepati

, Goldstein

, et al. Inactivation of the human papillomavirus E6 or E7 gene in cervical carcinoma cells by using a bacterial CRISPR/Cas RNA-guided endonuclease. J Virol, 2014; 88:11965–11972.

24.

Fellmann

, Gowen

, Lin

, et al. Cornerstones of CRISPR-Cas in drug discovery and therapy. Nat Rev Drug Discov, 2017; 16:89–100.

25.

Cai

, Tsai

, Yen

, et al. Critical threshold levels of DNA methyltransferase 1 are required to maintain DNA methylation across the genome in human cancer cells. Genome Res, 2017; 27:533–544.

26.

Ahuja

, Easwaran

, Baylin

. Harnessing the potential of epigenetic therapy to target solid tumors. J Clin Invest, 2014; 124:56–63.

27.

Baubec

, Colombo

, Wirbelauer

, et al. Genomic profiling of DNA methyltransferases reveals a role for DNMT3B in genic methylation. Nature, 2015; 520:243–247.

28.

Zagorac

, Alcala

, Fernandez Bayon

, et al. DNMT1 inhibition reprograms pancreatic cancer stem cells via upregulation of the miR-17-92 cluster. Cancer Res, 2016; 76:4546–4558.

29.

Pathania

, Ramachandran

, Elangovan

, et al. DNMT1 is essential for mammary and cancer stem cell maintenance and tumorigenesis. Nat Commun, 2015; 6:6910.

30.

Xiang

, Ma

, Wang

, et al. MiR-152 and miR-185 co-contribute to ovarian cancer cells cisplatin sensitivity by targeting DNMT1 directly: a novel epigenetic therapy independent of decitabine. Oncogene, 2014; 33:378–386.

31.

Hao

, Luo

, Krawczyk

, et al. DNA methylation markers for diagnosis and prognosis of common cancers. Proc Natl Acad Sci U S A, 2017; 114:7414–7419.

32.

Flanagan

, Wilson

, Koo

, et al. Platinum-based chemotherapy induces methylation changes in blood DNA associated with overall survival in patients with ovarian cancer. Clin Cancer Res, 2017; 23:2213–2222.

33.

Stewart

, Tamayo

, Wilson

, et al. KRAS genomic status predicts the sensitivity of ovarian cancer cells to decitabine. Cancer Res, 2015; 75:2897–2906.

34.

Fang

, Bi

, Zhou

, et al. Nicotinamide adenine dinucleotide (NAD) may affect DNA methyltransferase 1 through regulation of BRCA1 in ovarian cancer. Am J Cancer Res, 2015; 5:1199–1206.

35.

Saunthararajah

, Sekeres

, Advani

, et al. Evaluation of noncytotoxic DNMT1-depleting therapy in patients with myelodysplastic syndromes. J Clin Invest, 2015; 125:1043–1055.

36.

Sander

, Joung

. CRISPR-Cas systems for editing, regulating and targeting genomes. Nat Biotechnol, 2014; 32:347–355.

37.

Cox

, Platt

, Zhang

. Therapeutic genome editing: prospects and challenges. Nat Med, 2015; 21:121–131.

38.

, Men

, Qin

, et al. Non-viral and viral delivery systems for CRISPR-Cas9 technology in the biomedical field. Sci China Life Sci, 2017; 60:458–467.

39.

Wang

, Mou

, Li

, et al. Adenovirus-mediated somatic genome editing of Pten by CRISPR/Cas9 in mouse liver in spite of Cas9-specific immune responses. Hum Gene Ther, 2015; 26:432–442.

40.

, Song

, Liu

, et al. Artificial virus delivers CRISPR-Cas9 system for genome editing of cells in mice. ACS Nano, 2017; 11:95–111.

41.

, He

, Wei

, et al. Challenges in CRISPR/Cas9 delivery: potential roles of nonviral Vectors. Hum Gene Ther, 2015; 26:452–462.

42.

Yin

, Kauffman

, Anderson

. Delivery technologies for genome editing. Nat Rev Drug Discov, 2017; 16:387–399.

43.

Yin

, Kanasty

, Eltoukhy

, et al. Non-viral vectors for gene-based therapy. Nat Rev Genet, 2014; 15:541–555.

44.

Zuris

, Thompson

, Shu

, et al. Cationic lipid-mediated delivery of proteins enables efficient protein-based genome editing in vitro and in vivo . Nat Biotechnol, 2015; 33:73–80.

45.

Cong

, Ran

, Cox

, et al. Multiplex genome engineering using CRISPR/Cas systems. Science, 2013; 339:819–823.

46.

, Yu

, Zhang

, et al. Gene delivery with active targeting to ovarian cancer cells mediated by folate receptor alpha. J Biomed Nanotechnol, 2013; 9:833–844.

47.

, He

, Yu

, et al. Micelles based on methoxy poly(ethylene glycol)-cholesterol conjugate for controlled and targeted drug delivery of a poorly water soluble drug. J Biomed Nanotechnol, 2012; 8:809–817.

48.

, Wei

, Luo

, et al. Folate-linked lipoplexes for short hairpin RNA targeting claudin-3 delivery in ovarian cancer xenografts. J Control Release, 2013; 172:679–689.

49.

, Deng

, Wei

, et al. Ovarian cancer treatment with a tumor-targeting and gene expression-controllable lipoplex. Sci Rep, 2016; 6:23764.

50.

, Sander

, Reyon

, et al. Improving CRISPR-Cas nuclease specificity using truncated guide RNAs. Nat Biotechnol, 2014; 32:279–284.

51.

Fan

, Liao

, Guo

, et al. Dexamethasone-loaded poly(D, L-lactic acid) microspheres/poly(ethylene glycol)-poly(epsilon-caprolactone)-poly(ethylene glycol) micelles composite for skin augmentation. J Biomed Nanotechnol, 2014; 10:592–602.

52.

Duan

, Feller

, Penson

, et al. Discovery of differentially expressed genes associated with paclitaxel resistance using cDNA array technology: analysis of interleukin (IL) 6, IL-8, and monocyte chemotactic protein 1 in the paclitaxel-resistant phenotype. Clin Cancer Res, 1999; 5:3445–3453.

53.

Liang

, Luo

, Wei

, et al. A folate receptor-targeted lipoplex delivering interleukin-15 gene for colon cancer immunotherapy. Oncotarget, 2016; 7:52207–52217.

54.

, Zhang

, Yao

, et al. A PCR based protocol for detecting indel mutations induced by TALENs and CRISPR/Cas9 in zebrafish. PLoS One, 2014; 9:e98282.

55.

Hsu

, Scott

, Weinstein

, et al. DNA targeting specificity of RNA-guided Cas9 nucleases. Nat Biotechnol, 2013; 31:827–832.

56.

Blaschke

, Ebata

, Karimi

, et al. Vitamin C induces Tet-dependent DNA demethylation and a blastocyst-like state in ES cells. Nature, 2013; 500:222–226.

57.

Landen

, Chavez-Reyes

, Bucana

, et al. Therapeutic EphA2 gene targeting in vivo using neutral liposomal small interfering RNA delivery. Cancer Res, 2005; 65:6910–6918.

58.

Lanuti

, Rudginsky

, Lambright

, et al. Cationic lipid:bacterial DNA complexes elicit adaptive cellular immunity in murine intraperitoneal tumor models. Cancer Res, 2000; 60:2955–2963.

59.

Smallwood

, Esteve

, Pradhan

, et al. Functional cooperation between HP1 and DNMT1 mediates gene silencing. Genes Dev, 2007; 21:1169–1178.