A CRISPR interference strategy for gene expression silencing in multiple myeloma cell lines

sgRNAs cloning into lentiviral plasmid pLV hU6-sgRNA hUbC-dCas9-KRAB-T2a-Puro

To obtain the sequence of the sgRNA an Addgene library (CRISPRi library was a gift from Jonathan Weissman (Addgene #62217)) was consulted [24]. From the sequences available directed to RND3 promoter, two different sgRNA were selected. In addition, another sgRNA guide to a non-coding region (scrambled) was chosen to be used as negative control. To allow the cloning with the selected sgRNAs, a prior modification must be performed. In the forward and reverse primers, a nucleotide tails, target of Esp3I restriction enzyme must be introduced (Table 1).

Table 1 List of primers used to generate the sgRNAs. The necessary tails in the ends for the cloning were highlighted

The annealing of the designed primers was necessary to prepare a double stranded sgRNAs. This process requires the use of forward and reverse oligonucleotides, and their incubation at 95ºC, allowing a progressive cooling. Finally, T4 PNK enzyme was added to phosphorylate the 5’ end for subsequent cloning ligation. The details of the process are described in Fig. 1A. Then, the annealed sgRNAs were diluted 1:1000 and cloned into lentiviral plasmid construction pLV hU6-sgRNA hUbC-dCas9-KRAB-T2a-Puro (this plasmid was a gift from Charles Gersbach (Addgene plasmid # 71236); [25]). 5 µg of plasmid was linearized with Esp3I enzyme (Thermo Fisher) and then it was dephosphorylated with alkaline phosphatase (Promega) (Fig. 1B). The digested plasmid was analyzed using an agarose gel electrophoresis and the 15 kb expected band was purified using NZYGelpure kit (Nzytech), following the manufacturer’s instructions and quantified by Nanodrop Simplinano (GE Healthcare Life Science). The ligation reaction containing both the linearized plasmid and the annealed sgRNA was prepared and incubated overnight as described in Fig. 1C. Finally, the ligation product was transformed into DH5α competent E. coli and grown on to LB-Agar plates with ampicillin. To check the ligation procedure, plasmid DNA was obtained using GenElute Plasmid Miniprep kit (Sigma-Aldrich) and two different strategies were carried out: an amplification using a PCR with specific primers and DNA sequencing (Fig. 1D). Once checked, bacteria culture was amplified to obtain enough plasmid DNA using PureLink Plasmid Filter Midiprep kit (Invitrogen) following the manufacturer’s instructions. These plasmids containing the sgRNAs cloned into the lentiviral vector were then used to be transfected in HEK 293T cells.

Fig. 1figure 1

Schematic protocol of sgRNA cloning into lentiviral plasmid. First, the annealing of the primers was required to generate the sgRNAs (A). Then, the lentiviral plasmid was digested with Esp3I enzyme to linearize, purify and quantify (B). The ligation of the sgRNA into the linearized lentiviral plasmid was made at 16ºC overnight in presence of T4 ligase enzyme (C). Finally, the ligated plasmids were transformed in DH5α competent E. coli  (D) and colonies were checked by PCR and sequencing, before DNA amplification.

Transfection, production and concentration of lentiviral particles

The production of lentiviral particles was carried out by the calcium phosphate transfection method. HEK 293T lentiviral packaging cells were cotransfected with different vectors that included the sgRNA plasmids and 3rd generation lentiviral envelope and packaging plasmids (Table 2). 24 h prior the transfection, HEK 293T cells were plated in 10 cm plates and reached 60–70% of confluence the day of transfection. Two hours before transfection the medium was removed and replaced with 9 mL of fresh medium. To obtain a high virus titer, five 10 cm plates per each sgRNA construction were needed although the information related to the volumes and quantities hereby is referred to one 10 cm culture plate.

Table 2 List of plasmids used for HEK 293T cells co-transfection

Every plasmid containing the different RND3 sgRNA was co-transfected with 3rd generation lentiviral plasmids (3 µg pENV, 5 µg pRRE and 2,5 µg pREV) in a 15 ml tube, together with 15 µg sgRNA plasmid, in a 150 µL final volume placed in a 15ml tube. Then, 300 µL of TE 0.1X and 50 µL of CaCl2 were added and the DNA mix was incubated 5 min at room temperature. Finally, 500 µL of HBS 2X was drop by drop added to the mixture while vortexed and rapidly transferred to the HEK 293T cells plate. After 14–16 h the transfection medium was removed, 6 ml of fresh medium was added, and cells were kept for 48 h more. Then, to concentrate the lentiviral particles, the medium from the five plates was recollected and centrifuged for 10 min at 2000 g to discard remaining cells and the supernatant was filtered with 0.45 μm filter. To obtain high virus titer, the filtered supernatant was ultracentrifuged at 125,000 g for 150 min at 4ºC using the Optima L-100 XP ultracentrifuge (SW32 rotor, Beckman Coulter) and the pellet was resuspended in 1 ml of PBS for 30–60 min on ice, with occasional mixing. The lentiviral particles were used to freshly transduce the MM cells or they were frozen at -80ºC to further infections.

Transduction of MM cells and obtention of Rnd3 knock-down (KD) stable cell lines

RPMI 8226 and JJN3 cells were transduced with concentrated lentiviral particles in 6-well plates. 100 µL of lentiviral particles and 1 µL of polybrene solution (10 mg/mL, Sigma-Aldrich) were added to 1 × 106 MM cells/well in 1ml final volume and cells were kept at 37ºC and 5% CO2 for 48 h. Then, the transduction medium was removed, and MM cells were maintained in fresh medium. The selection of transduced cells was made by addition of puromycin (2 µg/mL, Sigma-Aldrich) to the culture medium for 24 h. Finally, puromycin resistant cells were amplified and considered as a stable cell line after sequential passages and used to perform all the experiments. To perform the qPCR and western blot experiments, cells were grown to confluence and RNA or protein extracts were collected two weeks post-transduction.

At the end of this process, we obtained two Rnd3-deficient lines (sgRND3 #3 and #4) for each cell type, together with a scramble line in the case of RPMI cells. In the case of JJN3 cells, it was not possible to obtain a scramble line, so the results shown below were compared with wildtype cells.

CRISPRi technology allows an efficient gene silencing at transcriptional level, leading the dCas9-KRAB to the gene promoter and inhibiting its expression as described above. As shown in Fig. 2A, RPMI cells transduced with RND3 sgRNA lentiviral constructs dramatically decrease RND3 RNA levels compared to either scramble and wildtype cells (1, 1.5 and 0.2-fold change in wildtype, scramble and both RND3 sgRNA, respectively). Same results were obtained with JJN3 RND3 KD cells, showing a 0.06-fold change reduction compared to wildtype cells.

Fig. 2figure 2

Rnd3 expression was reduced in RPMI 8226 and JJN3 transduced cells. RPMI 8226 and JJN3 RND3 mRNA expression levels were analyzed by qPCR and calculated as fold change using the 2−ΔΔCt method (A). Representative western blot images of Rnd3 protein expression in RPMI 8226 and JJN3 cells (B). Bands were quantified using ImageJ software and the densitometric values were normalized to GAPDH and plotted in (C). Mean values from 3 independent experiments are plotted and one-way ANOVA and Tuckey’s post-hoc test show differences between groups: *p < 0.05; **p < 0.01, ***p < 0.001, ****p < 0.0001

According to the decrease of RNA levels, RND3 knockdown cells also show a reduction of Rnd3 protein expression (Fig. 2B). Densitometric quantification of immunoblots reveals that RPMI cells lacking Rnd3 decrease up to 75–78% the levels of protein, compared to wildtype cells. In JJN3 cells, gene silencing seems to be more effective, observing a reduction of 88–89% in protein levels compared to wildtype cells (Fig. 2C). After the verification of gene silencing at transcriptional and translational levels, each cell line was frozen for next experiments.

To confirm long-term maintenance of gene silencing, transduced cell lines were thawed and collected at 1, 4 and 8 weeks after. Then, RND3 expression was quantified using a transcriptomic analysis consisting of 3’ UTR RNA sequencing, as described in Methods. As shown in Fig. 3, both cell lines transduced with RND3 sgRNA lentiviral constructs show a reduction in RND3 expression compared to WT or Scramble cells over time (Fig. 3A). When the data corresponding to each condition were pooled and plotted together, RPMI 8226 cells show two-fold change decreasing expression (p = 0,0009), while JJN3 show five-fold change decreasing between control and transduced cells (p = 0,0002). Taken together, these results confirm that a significant silencing was achieved using this protocol.

Fig. 3figure 3

Long-term maintenance of RND3 silencing is observed in RPMI 8226 and JJN3 transduced cells. RNA extracts were obtained from RPMI 8226 and JJN3 cells at different times (1 week, 4 weeks and 8 weeks post-thawing) and RND3 expression was analyzed by RNA sequencing. (A) Rnd3 time course expression in either RPMI and JJN3 cells. Values plotted represent single RNA levels for every cell line at the indicated times. (B) Mean of the values shown in A) grouped for each time and cell line were also plotted to confirm the consistency of the gene silencing along the time. RPMI 8226 knock-down cells show two-fold change expression decrease (p = 0,0009) while JJN3 knock-down show five-fold change expression decrease (p = 0,0002), compared to WT or scramble cells

Rnd3 KD cells shown changes in transcriptomic profile

The results obtained comparing the transcriptomic profile of control (wildtype and/or scramble) versus Rnd3 KD cells showed 93 genes differentially and consistently expressed in both RPMI 8226 and JJN3 cell lines (Fig. 4). Among these genes, specifically 39% were down-regulated and 61% were up-regulated in Rnd3 KD cells. The Gene Ontology (GO) database was used to carry out an enrichment study according to the categories based on the biological processes associated with these genes. With this criterion, the genes that presented significant differences between both groups were selected. Finally, genes associated with biological processes without relevance in MM were discarded. Based on the information obtained, we established a list of 19 differentially expressed genes that were grouped into 6 functional categories: calcium ion transport and mobilization, proinflammatory cytokine production, cell migration and motility, cell-cell interactions, angiogenesis, and cAMP cell signaling. All these functional categories correspond to cellular processes that have been described as relevant in the pathophysiology of MM. These results confirm that Rnd3 silencing produces downstream gene expression changes in MM cells and validates the methodology here proposed. However, further experiments should be addressed to confirm the potential role of Rnd3 in MM etiology, such as the analysis of Rnd3 expression in MM patients.

Fig. 4figure 4

Rnd3 loss-of-function results in gene expression repression signature. Heatmap showing the RNA-seq data of 93 differentially expressed genes upon Rnd3 KD in RPMI 8226 and JJN3 cell lines. Columns represent individual samples; rows correspond to the genes. Heat map represents the z-scores of the expression value (RPKM) characterized by RNA-seq. The column on the right represents the proportion of the genes that were repressed (green) or upregulated (red) in RND3 knockdown cells as compared with scramble or control samples

留言 (0)

沒有登入
gif