Detection of genes responsible for cetuximab sensitization in colorectal cancer cells using CRISPR-Cas9

Abstract Colorectal cancer (CRC) is a common malignant tumor in digestive tract with highly invasive and metastatic capacity. Drug sensitivity remains a significant obstacle to successful chemotherapy in CRC patients. The present study aimed to explore genes related to cetuximab (CTX) sensitivity in CRC by clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9. Celigo image cytometer was used to detect suitable cells and optimal dosage of CTX. Inhibition rate of CTX on Caco-2 cells was evaluated by cell counting kit-8 (CCK-8) method before and after transfection. 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT) was performed to explore suitable concentration of puromycin and multiplicity of infection (MOI). CRISPR-Cas9, sequencing data quality analysis and cell viability test were used for the selection of genes related to CTX sensitivity in CRC cells. Finally, the selected genes associated with CTX sensitivity in CRC cells were further validated by colony formation and CCK-8 assays. In the present study, Caco-2 cells had a better prolificacy, and CTX 100 μg/ml exhibited a good inhibition trend on the 7th and 14th days of infection. MTT assay indicated that the minimum lethal concentration of puromycin was 2.5 μg/ml. Forty-six candidate genes were preliminarily screened via sequencing data quality analysis. Subsequently, we found that knockout of any of the four genes (MMP15, MRPL48, CALN1 and HADHB) could enhance CTX sensitivity in Caco-2 cells, which was further confirmed by colony formation assay. In summary, MMP15, MRPL48, CALN1 and HADHB genes are related to the mediation of CTX sensitivity in CRC.


Introduction
Colorectal cancer (CRC) is a highly invasive and metastatic malignant tumor of the digestive tract, with high mortality [1]. Most CRC cells are currently assumed to be stem cells or stem cell-like cells. These stem cells are the outcomes of progressive accumulation of genetic and epigenetic changes that activate oncogenes and inactivate tumor-suppressor genes [2]. Due to the low early diagnosis rate, most patients are diagnosed as middle or advanced CRC, resulting in difficult treatment and poor prognosis [3]. At present, chemotherapy and radiotherapy are typical therapies for CRC. Systemic treatment of metastatic CRC usually involves chemotherapy backbone (fluoropyrimidines, oxaliplatin and irinotecan chemotherapies form) combined with biotherapy [4]. However, despite therapeutic advancements, the prognosis for metastatic CRC patients remains poor, with a median overall survival of 18-21 months [5]. Drug resistance is the main reason for the failure of chemotherapy in cancers.
Apart from chemotherapy and radiotherapy, advances have been achieved by the addition of targeted agents. Cetuximab (CTX) (Erbitux) is a monoclonal antibody that targets epidermal growth factor receptor to impede the phosphorylation of receptor-associated kinases and related signaling cascade [6], and plays an anti-tumor role at both cellular and genetic levels [7]. Cetuximab was also demonstrated to enhance overall survival when combined with radiotherapy alone [8]. Segelov et al. have clarified that the median survival of advanced CRC patients can be extended from 6-7 to 24-30 months after treatment with combined targeted drugs such as CTX and chemotherapy [9]. However, after a period of CTX treatment, numerous patients become insensitive to CTX with disease progression. CTX can kill cancer cells that are sensitive to drugs, but the remaining resistant cells can cause tumor recurrence and even metastasis. Currently, the vital challenge for CRC treatment is to improve sensitivity to CTX. Therefore, it is of great significance to find new CTX-sensitized targets in CRC.
Clustered regularly interspaced short palindromic repeats (CRISPR)-Cas9 is a special structure found in the genomes of bacteria and archaea that protect them from phages and plasmids [10]. Scientists have transformed CRISPR-Cas9 into a simple, efficient and accurate genome editing tool, providing an effective method for finding functional genes [11]. For CRC, the use of CRISPR-Cas9 to explore genes related to CTX sensitivity has not been reported. Therefore, in the present study, we screened novel genes involved in CTX sensitivity in CRC cells based on genome-scale CRISPR-Cas9 knockout (GeCKO) screening and coupled with cell viability test.

CRISPR-Cas9 knockout screening
To identify genes associated with enhanced sensitivity of CRC to CTX, GeCKO screening with a pooled GeCKO library was performed. We applied a GeCKO library of 123411 single-guide RNAs (sgRNAs) targeting 20914 human genes, including 19050 annotated protein-coding genes and 1864 microRNA expression genes, with an average coverage of six sgRNAs per gene. A single lentiviral vector system (lentiCRISPRv2) of GeCKO library was used to transfect HT-29 and Caco-2 cells at a low multiplicity of infection (MOI) to ensure that most cells gained only one viral construct. After 72 h of transduction, HT-29 and Caco-2 cells were screened with puromycin, and the cells that survived after 7 days were successfully transfected cells. A total of 3 × 10 7 cells were directly harvested as day 0 group after puromycin selection. Then the remaining cells were treated with dimethyl sulfoxide (DMSO) vehicle and 100 μg/ml CTX for 7 and 14 days, respectively.

Cell growth curve assay
HT-29 and Caco-2 cells were plated in triplicate in 96-well plates at a density of 2000 cells per well. After culturing at 37 • C in a 5% CO 2 incubator for another day, Celigo image cytometer (Cyntellect Inc., CA, U.S.A.) was used to capture cell images once a day for 4 days. Cell numbers in each well were quantified by Celigo image cytometer, and cell growth curves were generated for each cell.

MTT assay
To explore the most suitable concentration of puromycin, 3-(4,5-dimethylthiazol-2-yl)2,5-diphenyl tetrazolium bromide (MTT) (Sigma Chemical Company, MO, U.S.A.) was used to assess the proliferation of Caco-2 cells. Briefly, Caco-2 cells were seeded in 96-well plates in triplicate at the density of 2000 cells per well, followed by addition of puromycin at various concentrations (0, 1.25, 2.5, 5, 10 μg/ml). Then the culture medium was discarded, and 20 μl of MTT (5 mg/ml) were added into each well. After 7 days of incubation, the MTT crystals were dissolved in DMSO and the absorbance at 490 nm was measured.

Cell proliferation analysis via Celigo image cytometer
To find out the best dosage of CTX, Celigo image cytometer was used to quantify the cell number. The Caco-2 cells were placed in 96-well plates in triplicate and then adding CTX at a range of concentrations (0, 50, 100, 200, 400, 600, 800, 1200 μg/ml). After culturing at 37 • C in a 5% CO 2 incubator for another day, Celigo image cytometer was used to capture cell images for 14 days. Cell numbers in each well were quantified by Celigo image cytometer, and cell growth curves were generated for each cell.

Determination of optimal MOI
Caco-2 cells with and without puromycin were divided into group A and group B, respectively. Cells were infected with different MOI values (0.25, 0.5, 1, 2, 10) and seeded in 96-well plates at the density of 6000 cells per well. Puromycin (2.5 μg/ml) was then added to each well in group A only. Forty-eight hours later, the culture medium was discarded, and 20 μl of MTT (5 mg/ml) were added to each well. Following a 4-h incubation, the crystals were dissolved by DMSO and the absorbance at 490 nm was read. The functional MOI was defined as the MTT value of group A/the MTT value of group B [12].

Sequencing data quality analysis
NEBNext ® Ultra™ DNA Library Prep Kit for Illumina ® kit (QIAGEN Company, Hilden, Germany) was used for the construction of library. Briefly, more than 50 ng PCR products purified by gel cut or magnetic beads were adopted to construct the library directly. End Prep Enzyme Mix was used for end repair, including 5 -end phosphorylation and 3 -end of ' A' plus, with addition of sequencing adapter at both ends, followed by purification via AxyPrep Mag PCR Clean-up. Finally, P5 and P7 primers were used for amplification, in which the index containing 6-bp base at the end of P7 was used for subsequent sample discrimination. After purification, the quality and concentration of library were measured by Agilent 2100 bioanalyzer (

Cell viability test
The infected cells in logarithmic phase were seeded in 96-well plates and were treated with 100 μl of CTX 100 μg/ml or DMSO, respectively. Meanwhile, normal control (NC) group was set up and treated with CTX 100 μg/ml or DMSO, respectively. Five days later, 100 μl of CellTiter-Glo (thawed 24 h at 4 • C in advance) was added to each well and shook on an oscillated instrument for 10 min to induce cell lysis. Subsequently, an enzyme mark instrument was adopted to measure the absorbance of each well in Luminescent mode. The PC interference background, multiples of inhibited proliferation and sensitization multiples of each group of cells were calculated by absorbance of each well.

Cell counting kit-8 assay
After transfection, Caco-2 cells were seeded in 96-well plates in triplicate at the density of 2 × 10 3 cells per well, followed by addition of CTX 100 μg/ml. After culturing for 24, 48, 72, 96 and 120 h, cell counting kit-8 (CCK-8) solution 100 μl (Sigma) was added into each well, and plates were incubated at 37 • C for 1 h. Microplate reader was used to test optical density at the wavelength of 450 nm.

Colony formation assay
Colony formation assay was adopted to evaluate the effects of Caco-2 cells proliferation in vitro. Cells were plated in a six-well plate containing 1 × 10 3 cells per well in triplicate. After 2 weeks, the cells were fixed with 4% paraformaldehyde for 15 min and stained with 0.1% Crystal Violet. The visible colonies were counted.

Statistical analysis
Statistical analysis was performed by SPSS (version 20.0) and the experimental measurement data were expressed as mean + − SD. Statistical differences between groups were determined by one-way or two-way ANOVA. Differences were considered significant when P<0.05.

Growth curve results
From Figure 1A,B, the proliferation multiples of Caco-2 and HT-29 cells significantly increased in the first 4 days. However, on the 3rd day, the proliferation multiples of Caco-2 cells were higher than 2, while that of HT-29 cells was lower than 2, indicating that infected Caco-2 cells had a better prolificacy. Therefore, Caco-2 cells were selected for the following experiments.

Optimal dosage of CTX
As shown in Figure 3, CTX suppressed Caco-2 cell proliferation in a dose-dependent manner, and the IC 50 of CTX was 245.7 μg/ml. On the 14th day, when the CTX concentration was 50, 100, 200, 400, 600, 800 and 1200 μg/ml, the inhibitory rate was 11.67, 42.33, 43.67, 56.33, 57.67, 76.67 and 97.46%, respectively, indicating that CTX had a significant inhibitory function on Caco-2 cells in a dose-dependent manner. In the case that three CTX concentrations of 100, 200, 400 μg/ml had similar inhibitory effect and considering the high cost of CTX, CTX 100 μg/ml was selected as the optimal concentration for subsequent experiments.

Optimal MOI results
After 48 h of infection with Caco-2 cells, we found that the fluorescence rate augmented with the increase in infection intensity in group A, but not in group B. When MOI was 0.5, the fluorescence rate was 43% in group A, and 86% in group B, with 50% Functional MOI. MOI = 0.5 was selected for the subsequent experiments (Table 1).

Sequencing data quality analysis results
Sequencing data quality analysis was used to screen genes meeting the screening criteria (|FC Caco2| < 2, FCER-BITUX ≤ −2; BF Caco2 < 0, BFERBITUX > 2) between Day 7 vs Day 0 and Day 14 vs Day 0. A total of 16 genes were obtained from the intersection in accordance with 209 genes acquired from the preliminary selection of Day 7 vs Day 0 and 307 genes obtained from the preliminary selection of Day 14 vs Day 0. In addition, another 30 genes were    selected according to the screening criteria (day 14 ERBITUX logFC was sorted from small to large, Cacao2 logFC was less than 0.5) ( Table 2). However, previous report has reported that BCR and GEM genes are closely related to CTX sensitivity, so we only focused on the remaining 44 genes [13].

Cell viability by CellTiter-Glo luminescence assay
To measure the PC background interference, multiples of inhibited proliferation and sensitization multiples of the 44 genes, CellTiter-Glo luminescence assay was used to determine the proliferation of cells. Genes with PC background interference multiples between 0.7 and 1 (closer to 1 indicated less background interference), multiples of inhibited proliferation greater than 1.5, and sensitization multiples greater than 1.2 were selected. There were four genes meeting the selection criteria, namely MMP15, MRPL48, CALN1 and HADHB (Table 3).

Clonal formation results
To verify the sensitization effect of the above four gene knockout and evaluate its long-term sensitization effect, the clonal formation ability of Caco-2 cells was further tested through cell cloning experiment. As shown in Figure 4C, MMP15, MRPL48, CALN1 and HADHB genes knockout alone reduced the cloning potential of Caco-2 cells. On this basis, the addition of CTX 100 μg/ml inhibited the clonal formation of Caco-2 cells more significantly than the gene knockout alone group (P<0.0001).

Discussion
In the present study, GeCKO screening and cell viability test results revealed that MMP15, MRPL48, CALN1 and HADHB genes might be involved in drug sensitivity to CTX in CRC cell (Caco-2). This finding suggested that MMP15, MRPL48, CALN1 and HADHB genes were potential targets for the treatment of CRC, which fills in the gap of genes responsible for CTX sensitivity in CRC. CRISPR-Cas9 forms a complex by using nuclease Cas9 protein and sgRNA [14]. To our knowledge, sgRNA determines the specificity of target sequence through base complementary pairing; Cas9 protein acts as nuclease to cut genomic DNA complementary to the spacers on the sgRNA, causing double-stranded DNA damage, thereby introducing genetic mutations through mechanisms [15]. As the third generation of gene editing technology, CRISPR-Cas9 has attracted wide attention due to its advantages of simple operation, flexible design, low price and multi-point editing. With the continuous development and improvement of CRISPR-Cas9 system, a series of drug-resistance genes or virus-resistance genes have been screened [16]. Chen and Zhang have found that CRISPR-Cas9 can revert resistance gene mutations [17]. Zhang et al. also believed that CRISPR-Cas9 could be used to construct drug-resistant cancer cell lines for drug screening [18]. In this study, we preliminarily obtained 44 candidate genes in Caco-2 CRC cells related to CTX resistance by CRISPR-Cas9. Subsequently, based on the selection criteria of PC background interference multiples between 0.7 and 1, multiples of inhibited proliferation greater than 1.5, and sensitization multiples greater  Metalloproteinases (MMPs) is a family of proteolytic enzymes that can maintain the dynamic equilibrium of extracellular matrix by degrading and remodeling extracellular matrix [19]. Several studies have shown that MMPs are related to inflammation, atherosclerosis, liver cirrhosis, connective tissue disease and cancer [20,21]. As the active center of mitochondrial ribosomes, mitochondrial ribosomal proteins (MRPs) are the basis for the normal expression of mitochondrial DNA and responsible for translation of encoding proteins of mitochondrial DNA [22]. Wang et al. have noticed that down-regulation of MRPL18 gene expression can cause metabolic disorders in mouse insulinoma cells (MIN6) [23]. It is well known that changes in the concentration and activity of intracellular Ca 2+ have a certain effect on the generation of drug resistance in tumors [24]. Human CALN1 gene is only expressed in the brain, especially in the hippocampus and cerebral cortex. Sequence alignment done by Wu et al. has showed that CALN1 gene is highly similar to the ubiquitous calcium mediator protein, calmodulin [25]. Calmodulin is the principle mediator of the Ca 2+ signal found in many eukaryotic cells [26]. The proliferation, migration and metastasis of tumor cells require a large amount of energy, which means they need a more vigorous metabolism than normal cells. In addition to sugars that provide energy for the active proliferation of tumor cells, fatty acids can also produce ATP through β-oxidation [27]. HADHB mainly participates in the last three steps of fatty acid β-oxidation, including hydration of 2-dienoyl-CoA, dehydrogenation of fatty acid-2-hydroxyacyl-CoA, and thiolysis of β-ketoacyl CoA [28]. With a 3-year follow-up of 91 patients diagnosed with hilar cholangiocarcinom, Zhang et al. have found that HADHB not only shows high expression in hilar cholangiocarcinom, but also have a close relationship with tumor infiltration and lymphatic metastasis [29]. Nevertheless, no studies have shown the relationship between MMP15, MRPL48, CALN1 and HADHB genes and CTX sensitivity.
On the basis of our prophase research, we speculated that MMP15, MRPL48, CALN1 and HADHB genes may be related to CTX sensitivity in CRC. To verify the idea, CCK-8 assay was performed on cells that had MMP15,