Merkel cell carcinoma (MCC) is a rare but aggressive type of skin cancer with increasing incidence, currently at 0.7 cases per 100,000 individuals in the US [1
]. Major factors associated with MCC are UV-light exposure in fair-skinned people, immunosuppression, and presentation in the elderly [2
]. Over the past few years, there has been a growing interest in MCC due to the identification of a new human polyomavirus, Merkel cell polyomavirus (MCPyV), clonally integrated into ~80% of MCCs [3
]. Nevertheless, MCPyV-negative (MCPyV−
) MCCs are a smaller fraction of tumors characterized by a higher mutational burden with UV-light signature [4
Polyomaviruses are non-enveloped viruses with a dsDNA genome of ~5 kb divided into an early and late region separated by a non-coding control region (NCCR). MCPyV-positive (MCPyV+
) MCCs constitutively express two alternatively spliced products of the early gene: the small (sT) and the large (LT) tumor antigens (TAs) [5
]. During the normal viral life cycle, the TAs stimulate S phase entry, so that the virus can hijack the host cell replication machinery to replicate its own genome [6
]. Though MCPyV infection is widespread [7
], MCC is a dead-end for MCPyV replication, as truncating mutations in the LT disrupt the origin binding and helicase/ATPase domains [8
]. Nevertheless, the retinoblastoma (Rb) binding domain, necessary to promote cell growth and tumor progression, is conserved [9
]. Among the diverse functions of MCPyV sT [10
], the LT-stabilization domain (LSD) avoids degradation of the LT and cellular oncoproteins [18
The five-year relative survival of patients diagnosed with metastatic MCC is as low as 18% [19
]. Recently, the FDA has granted approval to avelumab and pembrolizumab as preferable treatments for patients with metastatic MCC, owing to their advantages over the classic chemotherapies. However, not all patients respond to these treatments and some develop resistance. Even when diagnosed at early stages, patients may be ineligible for surgery or radiotherapy due to other comorbidities. Consequently, new therapeutic options are needed [20
Using short hairpin RNAs (shRNAs), Houben et al. showed that MCPyV TAs are required for the maintenance of MCPyV+
MCC cells [21
]. Since MCPyV is integrated into the genome of the tumor cells, the use of gene-editing tools is a promising therapeutic strategy for MCC. In recent years, the CRISPR/Cas9 system has revolutionized the genome-engineering field [23
]. It consists of the Streptococcus pyogenes
Cas9) endonuclease and a single-guide RNA (sgRNA) that contains a spacer sequence matching a target site next to the protospacer adjacent motif (PAM) [24
]. In the absence of a template, the cellular DNA repair machinery resolves the Cas9-induced DNA break by non-homologous end joining (NHEJ), an error-prone mechanism that generates insertions and deletions (indels) in the target sequence [25
]. Contrary to shRNAs that reduce gene expression transitorily, CRISPR/Cas9 editing could generate stable and permanent changes in the genomic sequence of MCPyV TAs. This approach has been successfully used to eliminate JCPyV infection [26
In the present study, we used CRISPR/Cas9 to introduce frameshift mutations in the genomic sequence of MCPyV TAs. Inactivation of the TAs affected cell proliferation, led to cell cycle arrest, increased apoptosis and changed the expression of cellular proteins involved in cell cycle regulation. Importantly, a MCPyV−-non-MCC cell line remained unaffected, as well as those cells expressing a non-targeting sgRNA.
In the present study, we investigated the potential inactivation of MCPyV TAs using CRISPR/Cas9 editing in two MCPyV+
MCC cell lines, MS-1, and WAGA. The efficacy of CRISPR/Cas9 to induce mutations varied according to the sgRNA as well as the cell line used. In agreement with previous reports, the distribution of indels at a given target site was reproducible [30
]. CRISPR/Cas9 editing at a DNA level caused a significant decrease of the LT protein, especially in WAGA cells, in line with the higher transfection and cleavage efficiencies. CRISPR/Cas9 editing could not be reliably evaluated at the RNA level. LT-sgRNA might also affect sT expression to some extent, due to overlapping 3′-coterminal transcripts.
As previously reported [31
], the downregulation of LT protein resulted in the impaired proliferation of MCPyV+
cells. The antiproliferative effects were not observable in HEK293T cells, which are MCPyV−
, suggesting that the TAs-targeting sgRNAs did not exert off-target activity affecting cell proliferation.
Consistent with previous findings [21
], the slow-growing MCPyV+
cells exhibited a prominent peak in the G1 phase. When the expression of TAs was impaired, WAGA cells presented a significant reduction in cell cycle progression. Though the same was suspected to occur in MS-1, no significant changes were observed among the different conditions. Importantly, neither cells expressing the non-targeting ctr-sgRNA nor HEK293T cells expressing the TAs-targeting sgRNAs showed this reduction in cell cycle progression, suggesting a cell cycle arrest upon CRISPR/Cas9 targeting of TAs.
As evidenced by the cell death profiles in WAGA cells, electroporation, and presence of exogenous DNA caused certain damage in MCPyV+ cells, including cells expressing the ctr-sgRNA. Nevertheless, they progressively recovered to the levels of the cell control, contrary to those cells expressing the TAs-targeting sgRNAs. This observation could also explain the delay in cell proliferation. HEK293T cells were transfected with a lipid-based method, which requires a lower amount of DNA and is less cytotoxic.
During normal G1/S progression, cyclin-dependent protein kinases (Cdks)-mediated phosphorylation of Rb inhibits its binding to the transcription factor E2F [32
]. Nevertheless, proteins containing the conserved LXCXE motif can bind Rb and reduce Rb–E2F complex formation, promoting the expression of E2F target genes [33
] (Figure 6
B). LT from all polyomaviruses contain this motif [6
], yet Rb-binding is not unique to polyomaviruses, i.e., HPV E7 [34
]. Thus, the effects of CRISPR/Cas9 editing of MCPyV TAs could be further explained by alterations in the Rb-E2F pathway (Figure 7
). We found a marked decrease in products of E2F target genes that promote S phase entry: cyclin A2, Cdk2, and survivin. P-Chk1 (phosphorylated checkpoint kinase 1), another E2F-responsive element that accumulates in S phase to ensure successful DNA replication, was also significantly decreased [35
]. Rb phosphorylation is opposed by several Cdk-inhibitors (CKIs) such as p27 [36
]. In normal cells, Rb inhibits the ubiquitin-mediated degradation of p27, which promotes cell cycle arrest [37
]. In accordance with this model, TAs knockdown resulted in increased levels of p27. On the contrary, Cyclin D1 and Cdk6 were decreased. Cyclin D1 induces Rb-phosphorylation and sequestration of p27 [36
]. Generally, these findings provide strong evidence that alterations in cell cycle regulators upon TAs knockdown contribute to cell cycle arrest.
Our findings were restricted by the efficacy to transiently transfect MCPyV+ MCC cells and edit the viral TAs. Hence, MS-1 cells experienced lower transfection efficiency and higher electroporation-induced cytotoxicity, resulting in the slight effects of TAs editing. The introduction of specific modifications with CRISPR/Cas9 might allow targeting each MCPyV TA individually to elucidate their respective functions in MCC. However, targeting only the sT could also affect the levels of LT protein, due to lack of the LSD domain.
The trend of increasing incidence of MCC is expected to persist, owing to the aging of the population with prolonged UV-light exposure and immunosuppression [1
]. Moreover, a mortality rate between 33–46% makes MCC one of the most aggressive types of skin cancer [19
]. In view of the majority of MCCs being caused by MCPyV, viral TAs are an attractive target for a therapeutic CRISPR/Cas9 strategy. Currently, efforts are being made for the development of a CRISPR/Cas9-based therapeutic tool against viruses causing diseases, such as HPV [38
] and HIV [39
], and to treat hereditary genetic disorders [40
In summary, we report the use of CRISPR/Cas9 to target MCPyV TAs in MCC. Our data confirmed that MCPyV TAs have a crucial role in the maintenance of MCPyV+
MCC cells. In addition, we obtained insights into the effects of the TAs on cell cycle regulators, which could aid to identify targets for novel therapies for MCC, e.g., agents that avoid p27 degradation [42
]. Future experiments should focus on the development of a safe and efficient delivery system, especially in cases where the lesions are not directly reachable [43
4. Materials and Methods
4.1. Cell Lines
MS-1 was obtained from the European Collection of Authenticated Cell Cultures (ECACC Cat#09111802). The WAGA cell line was kindly provided by Roland Houben (University Hospital Würzburg, Germany). These cells were tested for mycoplasma contamination (MycoAlert™ detection kit, Lonza, Verviers, Belgium) and grown in RPMI 1640 + GlutaMAXTM-l medium supplemented with 20% FBS. MCPyV+ MCC cell lines grow as cell suspensions: WAGA grow as single-cell suspensions but MS-1 necessarily needs to form spheroid cell clusters to maintain cell viability. For this reason, prior to analysis MS-1 cell clusters were disrupted by incubation with trypsin-EDTA (0.25%) and gelatin. HEK293T cells were maintained in DMEM with 10% FBS. Media were supplemented with 1× MEM NEAA, 1 mM sodium pyruvate 100 mM, 1× Penicillin/Streptomycin/Glutamine 100× and 10 mM HEPES 1 M. All media and supplements were purchased in Thermo Fisher Scientific (Merelbeke, Belgium).
4.2. Design of sgRNAs and Plasmid Preparation
CRISPR/Cas9 editing was performed using the GeneArt™ CRISPR Nuclease Vector with OFP Reporter (Thermo Fisher Scientific, Merelbeke, Belgium). Two target-specific sgRNAs with a reduced number of off-target sites were designed with CRISPRdirect (https://crispr.dbcls.jp/
], using the genome sequence of MS-1 (Accession no. JX045709) as input. To explore off-target activity more exhaustively, a BLAST search of the human genome was performed. The candidates showed a difference of at least more than two nucleotides with any other human genomic sequence. The sT/LT-sgRNA targets a region (nts 259–277) in the exon shared by MCPyV TAs. The LT-sgRNA targets the exon 2 of LT (nts 960–978) and upstreams the Rb-binding domain.
Target-specific oligonucleotides (all oligonucleotides used in this study are described in Table S1
) were synthesized with 3’-overhangs with compatible ends for cloning into the GeneArt™ Vector. Then, the circularized vector was transformed into One Shot TOP10 chemically competent E. coli
cells (Thermo Fisher Scientific, Merelbeke, Belgium). Ampicillin-resistant colonies were cultured overnight for DNA extraction using the Wizard®
Plus SV Minipreps DNA Purification System (Promega, Leiden, Netherlands). The presence of the correct insert was confirmed by Sanger sequencing, using primers flanking the cloning site of the vector (named primers OFP vector, Table S1
). Following successful cloning, 50 mL midipreps were performed using the PureLink™ HiPure Plasmid Midiprep Kit (Thermo Fisher Scientific, Merelbeke, Belgium). Plasmid concentration and purity were assessed with Nanodrop ND-1000 (Isogen Life Science, Sint-Pieters-Leeuw, Belgium).
4.3. Transient Cell Transfection
ScreenFect A-Plus (Incella GmbH, Eggenstein-Leopoldshafen, Germany) was used to transfect HEK293T cells following the manufacturer’s instructions. For MCPyV+ cells, the 24-well optimization protocol of the Neon™ Transfection System 10 µL Kit (Thermo Fisher Scientific, Merelbeke, Belgium) was applied. Briefly, 2 × 105 cells/condition were electroporated with 1 µg of DNA and plated in 24-well plates. Transfection efficiency was quantified by flow cytometry (BD Accuri C6, BD Biosciences, San Jose, CA, USA). Once the optimal conditions were selected, 10 µg of DNA was used to electroporate 3–5 × 106 cells for further analysis using the 100 µL kit.
4.4. Sorting, DNA Extraction, PCR and Sequencing
To determine the ability of the TAs-targeting sgRNAs to induce indels, OFP+
cells were sorted using the BD FACSAria III Cell Sorter (BD Biosciences, San Jose, CA). The rest of the experiments described in the present study were performed with pools of transfected cells, unsorted. DNA was extracted from transfected cells using the QIAamp®
DNA Blood Mini Kit (Qiagen, Benelux BV, Antwerpen, Belgium). FastStartTM
High Fidelity PCR (Roche, Mannheim, Germany) with primers flanking the sgRNAs target sites (Table S1
) was used to amplify each sgRNA target region as a single amplicon. PCR cycling conditions were 95 °C for 5 min, followed by 38 cycles of 94 °C for 1 min, 59 °C for 20 s, and 72 °C for 1 min, with a final extension step of 72 °C for 5 min, performed on an Eppendorf Mastercycler ProS (Eppendorf, Hamburg, Germany).
Amplicons were sequenced using the BigDyeTM
Terminator v3.1 Cycle Sequencing Kit (Thermo Fisher Scientific, Merelbeke, Belgium) on a VeritiTM
Thermal Cycler (Applied Biosystems, Foster City, CA, USA), using the same primers and the following conditions: 96 °C for 1 min followed by 25 cycles of 96 °C for 10 s, 50 °C for 5 s, and 60 °C for 4 min. Sequencing products were separated by size on an ABI 3730 Genetic Analyzer (Applied Biosystems, Foster City, CA, USA). Amplicon sequences were compared to the amplicons obtained from DNA of cells expressing the ctr-sgRNA, using the software SeqScape v2.7 (Applied Biosystems, Foster City, CA, USA). Sequencing results of at least three independent experiments were subjected to TIDE (https://tide.nki.nl/
) analysis (limited to indels of size 0–5 that passed a significant cutoff, p
< 0.001) to determine the frequencies and distribution of Cas9-induced mutations.
4.5. RNA Extraction and RT-qPCR
At day 3 post-transfection, 5 × 105
cells were pelleted, washed with DPBS and disrupted/homogenized using QIAzol Lysis Reagent (Qiagen, Benelux BV, Antwerpen, Belgium). The rest of the protocol was performed with RNeasy Mini Kit (Qiagen, Benelux BV, Antwerpen, Belgium), including a DNase digestion step with RNase-Free DNase Set (Qiagen, Benelux BV, Antwerpen, Belgium). The RNA concentration was determined by A260
reading with Nanodrop ND-1000. One-step RT-qPCR was performed using SuperScript™ III Platinum™ One-Step qRT-PCR Kit (Thermo Fisher Scientific, Merelbeke, Belgium). Two primer sets amplifying the MCPyV TAs were used, with internal TaqMan probes (Table S1
). The RPLP0 (Ribosomal Protein Lateral stalk subunit P0) reference gene was used as the endogenous control. Reactions were run on an ABI 7500 Fast RT-PCR System (Applied Biosystems, Foster City, CA, USA). Cycling conditions were as follows: 50 °C for 30 min, 95 °C for 2 min, followed by 45 cycles of 95 °C for 3 s and 60 °C for 30 s. Each sample, from four independent experiments, was run in triplicate. Data were analyzed using 7500 Fast System SDS software v1.4 (Applied Biosystems, Foster City, CA, USA) and to quantify mRNA expression relative to the endogenous control the 2−ΔΔCt
method was applied.
4.6. Cell Proliferation Assay
One day after transfection, cells were counted with a Coulter counter and seeded in 96-well plates (5000 cells/well for HEK293T, 10,000 cells/well for WAGA and 15,000 cells/well for MS-1). Cell growth was monitored using the tetrazolium salt WST-1 (Roche, Mannheim, Germany). After adding the reagent, plates were incubated at 37 °C for 4 h. Then, the formazan product absorbance at 450 and 690 nm was measured with a SpectraMax Plus 384 microplate reader. For each time point, three different wells were measured to determine the mean absorbance in each of at least three individual experiments.
4.7. Protein Extraction and Immunoblotting
Protein extracts were obtained with RIPA Buffer (Thermo Fisher Scientific, Merelbeke, Belgium) containing cOmplete™, Mini, EDTA-free Protease Inhibitor Cocktail (Roche, Mannheim, Germany). Proteins were separated by SDS-PAGE on 10% Criterion XT Bis-Tris gels (Bio-Rad Laboratories, Hercules, CA, USA), transferred to PVDF membranes and incubated overnight at 4 °C with the correspondent primary antibody. The following day, membranes were incubated with HRP-conjugated secondary antibodies for 1 h at room temperature. Peroxidase activity was detected with SuperSignalTM
West Femto Maximum Sensitivity Substrate (Bio-Rad, Hercules, CA, USA). Images were captured with a ChemiDocTM
MP Imaging System and analyzed with Image LabTM
v6 software (Bio-Rad, Hercules, CA, USA). All antibodies used in the present study are listed in Table S2
. Densitometry analysis was performed with ImageJ software. Blots showing molecular weight markers and densitometry readings are included in Figures S3 and S4
4.8. Cell Cycle Analysis
BD Cycletest™ Plus DNA Kit (BD Biosciences, San Jose, CA, USA) was used to stain cellular nuclei of 2 × 105 cells with propidium iodide (PI), following the manufacturer’s instructions. Cell cycle phase distribution was evaluated using BD Accuri C6 flow cytometer, acquiring 2 × 104 events for each sample. Data were analyzed with FlowJo v10 (Tree Star, Williamson Way, Ashland, OR, USA).
4.9. Apoptosis Assay
Apoptosis assays were performed at days 1, 2, 3, 5, and 8 post-transfection. Briefly, 3–5 × 106 cells were pelleted and labeled with eBioscience™ Fixable Viability Dye eFluor™ 520 (Thermo Fisher Scientific, Merelbeke, Belgium). Cells heated up at 65 °C for 5 min served as positive controls. Apoptotic cells were detected with eBioscience™ Annexin V-APC (allophycocyanine) Apoptosis Detection Kit (Thermo Fisher Scientific, Merelbeke, Belgium) using BD Accuri C6. For each sample, 2 × 104 events were recorded and data were analyzed with FlowJo v10.
Statistical analyses (paired t-test) were performed with GraphPad Prism 7 (GraphPad Software Inc., La Jolla, CA, USA). Significance was defined with the following p-values: * p < 0.05, ** p < 0.01, *** p < 0.001, and **** p < 0.0001.