The prime editor 2 system was used to achieve IMP2 knockout in A549, Huh7, and LLC1 cells. pCMV-PE2-P2A-GFP (Addgene plasmid #132776) and pU6- pegRNA-GG-acceptor (Addgene plasmid #132777) were a gift from Anzalone et al. [19]. pU6- pegRNA-GG-acceptor plasmid was employed as a vector to deliver the pegRNA component and the designed pegRNA encoding sequences were inserted into the vector by golden gate cloning. A previously validated spacer targeting exon 6 in IGF2BP2 was used [10], and two different spacers targeting different loci of exon 6 in Igf2bp2 were utilized for pegRNA assembly (Table 1, Ref. [10]). Desired mutations were planned to disrupt the protospacer adjacent motif (PAM) of the spacer sequences. The scaffold sequence used was: 5${}^{\prime }$AGAGCTAGAAATAGCAAGTTAAAATAAGGCTAGTCCGTTATCAACTTGAAAAAGTGGCACCGAGTCG-3${}^{\prime }$ for all pegRNAs.

The knockout procedure: 100,000 cells/well were seeded into 24-well plates and cultivated overnight (16–24 hours). The transfection was initiated when confluence reached approximately 60%. Lipofectamine 3000 (#L3000008, Thermo Fisher Scientific, Darmstadt, Germany) was used for transfection, and an equimolar ratio of the two vectors was employed to obtain a total DNA content of 2 µg, along with lipofectamine reagent in volumes as recommended by the manufacturer. For Huh7 cells, jet-PEI hepatocyte DNA transfection reagent ((#101000053, Polyplus, Illkirch, France) was used according to the manufacturer’s instructions. Based on previous trials, media was changed 4 h post-transfection to prevent toxicity associated with the transfection reagents; 24 h post-transfection, cells were gently washed with 1$\times$ PBS, followed by detachment using trypsin (#T3924, Merck, Taufkirchen, Germany). The cells were resuspended in media to obtain a single-cell suspension. Single GFP-positive cells were manually picked using a glass capillary and transferred into a petri dish, ensuring enough space between the picked cells so that the resulting colonies would not overlap. A sterile-filtered 48 h conditioned medium from wild-type cells, supplemented with 20% FCS, was utilized to support the growth of the clones. Once the clones reached a sufficient growth point, they were then transferred into a 96-well plate and subsequently cultured in a 24-well plate. They were maintained in culture until the knockout of IMP2 was verified by Sanger sequencing using CRISP-ID to analyze the sequencing results [20], followed by western blot analysis.

Design of CRISPR/Cas9 system: The CRISPR/Cas9 system was utilized to achieve IMP2 knockout in A549, Huh7, and LLC1 cells. Here, pSpCas9(BB)-2A-GFP (PX458, Addgene plasmid #48138) plasmid construct was used. Cas9-GFP fusion protein expression cassette was combined with U6-promotor driven sgRNA expression in the single plasmid. Restriction enzyme cloning was employed to insert designed gRNA sequences into the PX458 plasmid construct. Two different gRNAs, namely gRNA1 and gRNA2, were used to target IGF2BP2, while gRNAI and gRNAII were used for editing in Ifg2bp2 (Table 2). The knockout cell lines generation, selection, and isolation procedures were as described in 2.1; however, the total DNA used for transfection was 1 µg.

The transfection of the cells was performed as described in section 2.2. After 24 h of transfection, the cells were washed with 1$\times$ PBS, detached with trypsin (#T3924, Merck, Taufkirchen, Germany), and resuspended to obtain a single-cell suspension; 30–40 GFP-positive cells, as mentioned in section 2.1, were manually picked and added to a PCR tube containing 2 µL of lysis buffer (Phire Tissue Direct PCR Kit, #F170S, ThermoFisher Scientific, Darmstadt, Germany). Heat inactivation of Proteinase K in the lysis buffer was performed at 96 °C for 3 min.

Amplification of the specific target region in IGF2BP2 (Table 2) was carried out by PCR using the following primers: hIMP2 exon 6_for 5${}^{\prime }$-TGTCCTGCTGCATTTCAGAGCC-3${}^{\prime }$ and hIMP2 exon 6_rev 5${}^{\prime }$-AAGGAAGCAAAGGAAGCCCCAC-3${}^{\prime }$, as described previously [10]. For the first PCR, we used the Phire Tissue Direct PCR Master Mix (F170S, Thermo Scientific, Darmstadt, Germany) following the manufacturer’s protocol. The amplified product was purified by agarose gel electrophoresis using the NucleoSpin® Gel and PCR Clean-up kit (#11992242, Macherey-Nagel, Düren, Germany) to obtain the desired size amplicon of 323 bp and to remove PCR residues, such as nucleotides, salt, and primers, that could interfere with the downstream processing.

In the second PCR, the primers bound to NGS-sequencing adapters (Table 3) were used, and the KAPA2G Fast Hot Start Genotyping Mix (KK5620, Sigma Aldrich, Taufkirchen, Germany) was employed following the manufacturer’s instructions. The PCR product was subjected to agarose gel purification. Subsequent sequencing was performed on an Illumina NextSeq platform using single-end sequencing. The sequencing results were analyzed using the CRISPResso2 software package [21].

Western blots were performed as described earlier [6], employing antibodies specific for IMP2/p62 [22] and $\alpha$-tubulin (#T9026, Merck, Taufkirchen, Germany). The secondary antibodies used were IRDye680-conjugated anti-rabbit IgG (#926-68071, LI-COR Bioscience, Bad Homburg, Germany) and IRDye800- conjugated anti-mouse IgG (#926-32210, LI-COR Bioscience, Bad Homburg, Germany). The Odyssey near-infrared imaging system from LI-COR Bioscience (Bad Homburg, Germany) was used to measure the signal intensities for IMP2 and its splice variant p62. The quantification of the western blot signal intensities was performed using StudioLite software version 5.0 (LI-COR Bioscience, Bad Homburg, Germany).

The IncuCyte® S3 system (Sartorius, Goettingen, Germany) was employed to monitor 2D cell proliferation; 3000 cells per 100 µL per well were seeded into a 96-well plate. For compound treatment, solvent control (0.1% DMSO dissolved in respective media) was also tested, and compounds were added to the cell the day after seeding. Cell confluency was measured every 8 h starting from the time point of treatment or seeding (for experiments without compound treatment) until 72 h and analyzed using the basic analyzer software. Cell confluency was normalized to the 0 h time point. Metabolic activity was measured 72 h after seeding using the MTT assay. For each compound, the inhibition of cell viability was calculated for each concentration and normalized to its respective DMSO control or untreated control (if DMSO control showed viability above 90%).

60,000 cells per 100 µL per well were seeded into a 96-well plate. For compound treatment, solvent control (0.1% DMSO dissolved in respective media) was also tested, and compounds were added to the cells the day after seeding. Metabolic activity was measured 72 h after treatment using the MTT assay. For each compound, the inhibition of cell viability was calculated for each concentration and normalized to its respective DMSO control or untreated control (if DMSO control showed viability above 90%).

To assess the spheroid-forming ability of cells in 3D, the IncuCyt® S3 system was used. 3000 cells per 100 µL per well were seeded into low-attachment U-bottom 96-well plates. Spheroid formation was monitored over 6 days after seeding.

Using serial dilution, 300 cells per 2 mL per well were seeded into a 6-well plate. For compound treatment, solvent control (0.1% DMSO dissolved in respective media) was also tested, and compounds were added to the well at the time of seeding. Cells were allowed to grow for 1–2 weeks to assess the colony formation ability. The media was removed, and cells were washed once carefully using 1$\times$ PBS. The live colonies were imaged using the IncuCyte® S3 system. The confluence area was measured for colonies consisting of at least 50 cells per colony, which corresponded to an object area of at least 1 $\times$ 10${}^4$–2 $\times$ 10${}^4$ µm${}^2$, depending on the cell type, and the cut-off mask was applied accordingly. The object counts per well module provided the total count of the colonies per well. The object average area module provided the average colony area per well. The inhibition of colony formation activity was calculated by normalizing to its untreated control (if solvent control did not show any difference in the number and size of colony in comparison to the untreated).

A total of 70,000 cells were seeded per 100 µL per well into an Image Lock 96-well plate. On the following day, the wound maker tool was used to perform scratches into the cell monolayers. The media containing the detached cells were removed and replaced with media without FCS. Migration was observed every 8 h up to 48 h after scratching. Cell confluency in the wound area was analyzed using IncuCyte® S3 migration software.

The raw data was analyzed using Microsoft Excel, and statistical analysis was performed using Origin Pro® version 2023b software (Northampton, MA, USA). Non-linear regression analysis was used to calculate the inhibitory concentration 50 (IC${}_{\mathrm{50}}$). Data are presented as means $\pm$ SEM unless otherwise indicated. To analyze the data distribution, the Shapiro-Wilk test was performed, and subsequent analysis was conducted based on the distribution and group size. Statistical differences between groups were calculated on the group number using Student’s t-test for two groups and one-way ANOVA with Tukey’s or Bonferroni’s post hoc analysis for more than two groups at a single time point. Statistical differences between groups at several time points were calculated using two-way ANOVA with Bonferroni’s post hoc analysis. *p 0.05, **p 0.01, ***p 0.001.

Prime editing, a CRISPR/Cas9 variant, had previously achieved biallelic IMP2 HCT116 knockout cells [10]. However, despite multiple trials, IMP2 knockout generation using this approach (Table 1) in HepG2, Huh7, SW480, A549, and LLC1 cells was not successful.

Multiple attempts to generate IMP2 knockouts in A549 and Huh7 cells using the CRISPR/Cas9 approach also failed. To rule out any effect of the manual single-cell picking procedure and further culture, we applied the entire single-cell colony expansion procedure on non-transfected A549 and Huh7 cells. While both non-transfected and GFP-negative cells, which had undergone the same transfection technique, formed colonies, no edited clones in either cell line were obtained.

To test the hypothesis that the inability of the successfully edited single cells to proliferate into a colony was since IMP2 is necessary for cell survival and proliferation, we performed next-generation sequencing of the GFP-positive cells picked up 24 h post-transfection that were successfully transfected with gRNA1 or gRNA2 individually in A549 cells or combination in Huh7 cells. Furthermore, as a control for the next-generation sequencing method, we simultaneously analyzed a separate batch of A549 and Huh7 cells that were not transfected at all. These cells are referred to as the non-transfected cells.

The sequencing analysis of GFP-positive A549 cells transfected with gRNA1 revealed a modification of 21.97% (Fig. 1A), while transfection with gRNA2 exhibited a higher editing efficiency of 83.66% (Fig. 1B). Similarly, GFP-positive Huh7 cells transfected with gRNA1 showed a comparable editing efficiency (Supplementary Fig. 1D), providing further evidence that gRNA editing efficiency is independent of the cell line.

Fig. 1.

Editing efficiency of gRNA1 and gRNA2 in A549 cells. Panel (A) represents gRNA1-transfected A549 cells analyzed using gRNA1 adapter primers, while panel (B) represents gRNA2-transfected A459 cells using gRNA2 adapter primers. The pie charts display the percentage and number of reads, indicating the presence of unmodified alleles (blue) and modified alleles (orange).

In the case of non-transfected A549 cells, the analysis using gRNA1 and gRNA2 adapter primers showed that 99% (Supplementary Fig. 1A) and 98.9% (Supplementary Fig. 1B) of the alleles matched the reference sequence near the gRNA target site, respectively. Similar results were observed for non-transfected Huh7 cells analyzed using gRNA1 adapter primers (Supplementary Fig. 1C). It is important to note that the modifications depicted in Fig. 1 and Supplementary Fig. 1 include base insertions, deletions, and substitutions.

The next-generation sequencing analysis of Huh7 cells transfected with both gRNA1 and gRNA2 revealed a deletion of 96 base pairs between the expected cut sites of the two sgRNAs, as determined using the gRNA1 adapter primer system (Fig. 2). This finding provides additional evidence supporting the high editing efficiency of gRNA1 and gRNA2.

Fig. 2.

Editing efficiency of gRNA1 and gRNA2 in combination in Huh7 cells. The sequence alignment with the reference genome is presented in the sequence view. Nucleotides indicated by the pink bar represent the sequencing result. Additionally, the red bar represents a 96 bp deletion between the gRNA1 and gRNA2, represented by blue and green bars, respectively. This deletion occurs in the target region of exon 6 of IGF2BP2. The image was generated using SnapGene Viewer version 6.2 (San Diego, CA, USA).

IMP2 biallelic knockout LLC1 cells were generated using gRNAI and gRNAII (Table 2). The editing was confirmed through Sanger sequencing analysis, and six different mutation types were observed (Supplementary Fig. 2). A representative clone of each mutation type was selected for further investigation. The reduction in IMP2 expression was confirmed through western blot analysis (Supplementary Fig. 3). However, the clone bKO6 exhibited up to 65% expression of IMP2 compared to LLC1 wild-types (WT1) and was therefore excluded from further analysis. The remaining five clones, designated as bKO1, bKO2, bKO3, bKO4, and bKO5, were utilized for subsequent characterization. To assess the potential off-target effects of the CRISPR/Cas9-mediated knockout procedure, an editing control LLC1 clone named WT2 was included in the experimental analysis. This control clone was obtained from an experiment in which cells had undergone transfection with both gRNAI and gRNAII but were selected from non-GFP positive cells. Sanger sequencing confirmed that WT2 was a wild-type clone.

To conduct the in vitro IMP2 target analysis, biallelic knockout clones were utilized where available. However, for cell lines in which achieving a biallelic knockout (bKO) was not possible, monoallelic knockout (mKO) clones were employed. The IMP2 protein levels in the monoallelic knockouts have been previously published. The mean values were close to 80% for Huh7, 44% for HepG2, and 57% for SW480 [10], while the expression of IMP2 in the biallelic LLC1 knockouts is shown in Supplementary Fig. 3.

3.4.1 IMP2 Knockout Reduces Proliferation in 2D Cell Culture

Compared to the WT2 control, which had undergone the same treatment as the knockout clones, clones bKO1-3 showed reduced proliferation (bKO1: p = 0.037; bKO2: p = 0.005; bKO3: p 0.0001). However, the growth rate of the two wild-type controls varied quite strongly (Fig. 3A). No apparent morphological changes were observed in the wild-type and biallelic IMP2 knockout LLC1 cells (Supplementary Fig. 4A).

Fig. 3.

Effect of IMP2 knockout on 2D cell proliferation. (A) LLC1, (B) Huh7, (C) HepG2, and (D) SW480 were assessed for their effect on 2D cell proliferation using the IncuCyte® S3 live cell imaging system. LLC1 IMP2 knockouts were biallelic (bKO), while the Huh7, HepG2, and SW480 cells were monoallelic IMP2 knockouts (mKO). Data were normalized to time point 0 h and are represented as mean $\pm$ SEM, n = 3, quintuplicates. Significance values were determined by comparing the knockout cells to those obtained from the wild-type (WT) cells. The proliferation rate was determined based on cell confluence.

No effect was seen in the proliferation of the monoallelic knockout clones of Huh7 (Fig. 3B). Although not statistically significant, a trend towards reduced proliferation was observed for the SW480 monoallelic knockout clone at later time points (Fig. 3D). Importantly, a significant reduction in proliferation was observed in the HepG2 IMP2 knockout clone (Fig. 3C).

3.4.2 IMP2 Knockout Cells Lose Their Ability to Form Compact 3D Spheroids

3D cell culture models, also known as multicellular tumors (MCTs) or spheroids, closely resemble the structural organization, oxygen and nutrient gradients, and pH conditions of in vivo solid tumors [23]. Based on the compactness, the spheroid/multicellular tumors (MCTs) are categorized as compact spheroids and loose aggregates of cells [23]. Compact spheroids are tightly bound to each other, making it difficult to distinguish single cells, whereas loose spheroids cannot form complete spheres and can be easily disintegrated [23]. In some tumor cell types, they fail to aggregate into one single entity and exist as several small colonies of cells, which are referred to as satellite colonies. We observed a loss in the ability to form compact spheroids upon the knockout of IMP2. All wild-type clones except for Huh7 successfully formed compact spheroids three days after seeding. In contrast, the biallelic IMP2 knockout LLC1 cells and the monoallelic IMP2 knockout SW480 cells showed a change in cell adhesion and formed loose aggregates, while the monoallelic HepG2 IMP2 knockouts formed satellite colonies. This characteristic persisted even after 6 days of observation (Fig. 4). Parental Huh7 cells also showed the formation of satellite colonies and did not form spheroids throughout the 6-day observation period (data not shown).

Fig. 4.

Comparison of spheroid formation between wild-type and IMP2 knockout clones. (A) LLC1, (B) HepG2, and (C) SW480 clones were assessed for their spheroid forming ability using the live cell imaging system IncuCyte® S3 at the 6-day time point after cell seeding. Representative images were chosen, and the scale was adjusted to 1 mm. The Huh7, HepG2, and SW480 IMP2 knockout cells were monoallelic knockouts.

3.4.3 IMP2 Knockout Clones and Treatment of Wild-Type Cells with IMP2 Inhibitors Exhibited Impaired Colony Formation Ability

The colony formation assay (CFA) provides insight into the ability of single cells to survive and reproduce [24] and has been extensively used to test the effect of cytotoxic agents on a cancer cell’s ability to form a viable colony [25]. The CFA was employed to understand the role of IMP2 in tumor growth. The LLC1 IMP2 biallelic knockout cells exhibited a significant decrease in both the number of colonies formed (Fig. 5A) and the average area of the colonies (Fig. 5B). While the wild-type cells formed colonies in the form of spheroids, the knockout clones lost this ability (Fig. 5C). Furthermore, monoallelic IMP2 knockouts of Huh7, HepG2, and SW480 cells also exhibited a significant reduction in colony formation ability, as observed in terms of both colony number and area (Fig. 6).

Fig. 5.

Effect of IMP2 knockout in LLC1 cells on the colony formation ability. The results are presented as follows: (A) number of colonies, (B) average area of colonies, and (C) representative images of colonies with the scale adjusted to 1 mm. Colony counting and measurement of average area were performed using the IncuCyte® S3 live cell imaging system. It should be noted that WT1 represents the LLC1 wild-type clone, while WT2 serves as the LLC1 editing control. The data was normalized to the WT1 cells and are represented as mean $\pm$ SEM, n = 4, triplicates. **p 0.01 and ***p 0.001 when compared to values of WT1 cells. n.s. indicates no significant difference.

Fig. 6.

Effect of IMP2 knockout on colony formation ability of Huh7, HepG2, and SW480. Figures illustrate the number of colonies for wild-type (WT) or monoallelic IMP2 knockout (mKO) (A) Huh7, (B) HepG2, and (C) SW480 cells, and the average area of colonies for (D) Huh7, (E) HepG2, and (F) SW480. (G) Representative images of colonies seen in a 6-well plate. The colonies were counted, and the average colony area was measured using the IncuCyte® S3 live cell imaging system. Huh7, HepG2, and SW480 were monoallelic IMP2 knockouts. Data were normalized to wild-type cells/control (Co) and represented as mean $\pm$ SEM, n = 3, duplicates. *p 0.05 and ***p 0.001 when compared to values of control.

Furthermore, we investigated the impact of IMP2 inhibitor compounds # 4, 6, and 9, which had been identified by a screening approach [10], on the 2D proliferation and metabolic activity of wild-type LLC1 cells. The compounds demonstrated a significant reduction in the proliferation rate at concentrations of 15 µM and 30 µM (Fig. 7A–C), and a dose-dependent decrease in the cells’ metabolic activity, as assessed by MTT assay, was observed at concentrations ranging from 5 µM to 100 µM (Fig. 8A–C). The cells did not exhibit distinct morphological changes (Supplementary Fig. 4B).

Fig. 7.

Effect of IMP2 inhibitors on the 2D cell proliferation of wild-type and biallelic IMP2 knockout LLC1 cells. The effects on 2D proliferation of LLC1 cells upon treatment with published IMP2 inhibitors, i.e., (A,D) compound 4, (B,E) compound 6, and (C,F) compound 9 at the indicated concentrations over 3 days were assessed using the IncuCyte® S3 live cell imaging system. The time point shown in panels (D–F) is 72 h post-treatment. Data is normalized to time point 0 h of the respective control and represented as mean $\pm$ SEM, n = 3, triplicates. *p 0.05, **p 0.01, and ***p 0.001 when compared to values of control (Co) for graphs (A–C) and to the values of WT1 LLC1 cells for graphs (D–F). The proliferation rate was determined based on the cell confluence.

Fig. 8.

Effect of IMP2 inhibitors on the metabolic activity (MTT assay) of wild-type and biallelic IMP2 knockout LLC1 cells. The effects on the metabolic activity of LLC1 cells upon treatment with published IMP2 inhibitors, i.e., (A,D) compound 4, (B,E) compound 6, and (C,F) compound 9 at the indicated concentrations after 3 days were assessed by MTT assay. Data are normalized to solvent controls (Co) and represented as mean $\pm$ SEM, n = 3, triplicates. *p 0.05, **p 0.01, and ***p 0.001 when compared to values of control (Co) for graphs (A–C) and to the values of WT1 LLC1 cells for graphs (D–F).

To investigate the specificity of the compounds on IMP2 in terms of their effect on 2D proliferation (Fig. 7D–F) and metabolic activity (Fig. 8D–F), we compared their action on bulk WT1 and the editing control WT2 and the five biallelic IMP2 knockout LLC1 clones. Notably, the knockout clones bKO1-4 displayed a lower sensitivity to the treatments compared to wild-type LLC1 cells (WT1 and WT2). The bKO5 displayed a similar response to the wild-type LLC1 cells.

Based on the IC${}_{\mathrm{50}}$ of the compounds on LLC1 metabolic acitivity, a treatment concentration of 25 µM was chosen for our study of their effect on colony formation (Table 4). For HepG2, Huh7, and SW480, the previously published IC${}_{\mathrm{50}}$ values of the three compounds were selected for the treatment [10]. In all tested cell lines, all three compounds exhibited a significant reduction in both the number and average size of the formed colonies. These findings provide further evidence of the essential role of IMP2 in colony formation ability (Figs. 9,10).

Table 4.IC${}_{\mathrm{50}}$ of IMP2 inhibitors in wild-type LLC1 cells.

Compound (#)	IC${}_{\mathrm{50}}$
4	25.23 µM
6	23.90 µM
9	28.45 µM

IC${}_{\mathrm{50}}$, Inhibitory concentration 50.

Fig. 9.

Effect of IMP2 inhibitors on wild-type LLC1 cells in colony formation assay. Figures illustrate the (A) number of colonies, (B) average area of colonies, and (C) representative images of colonies seen in a 6-well plate. Colonies were counted, and the average area of colonies was determined using the IncuCyte® S3 live cell imaging system. A treatment concentration of 25 µM was applied for each compound. Data was normalized to untreated wild-type LLC1 cells/control (Co) and represented as mean $\pm$ SEM, n = 3, duplicates. *p 0.05 and **p 0.01 when compared to values of control cells.

Fig. 10.

Effect of IMP2 inhibitors on Huh7, HepG2, and SW480 cells in colony formation assays. Results present the number of colonies for (A) Huh7, (B) HepG2, and (C) SW480, as well as the average area of colonies for (D) Huh7, (E) HepG2, and (F) SW480 cells. Additionally, (G) representative images of colonies are shown as seen in a 6-well plate. The colonies were counted, and the average colony area was determined using a live cell imaging system, IncuCyte® S3. The Huh7, HepG2, and SW480 were monoallelic IMP2 knockouts. Data was normalized to untreated wild-type cells and represented as mean $\pm$ SEM, n = 2, triplicates. The treatment doses were as follows: For Huh7, comp 4: 35 µM, comp 6: 45 µM, and comp 9: 25 µM. For HepG2, comp 4: 30 µM, comp 6: 40 µM, and comp 9: 35 µM. For SW480, comp 4: 20 µM, comp 6: 50 µM, and comp 9: 35 µM. *p 0.05, **p 0.01, and ***p 0.001 when compared to values of respective wild-type (WT)/control (Co) cells.

3.4.4 IMP2 Knockout Cells Show Reduced Cell Migration

The effect of IMP2 knockout on cell migration was investigated (Fig. 11). The two parental LLC1 cell lines, namely WT1 and WT2, differed significantly (p 0.0001), with the WT2 migrating slower than the bulk WT1 cells. The IMP2 biallelic LLC1 knockout cells were heterogenous regarding their migration ability (Fig. 11A). A significant reduction in cell migration was observed in the monoallelic IMP2 knockout clones Huh7, HepG2, and SW480 cells.

Fig. 11.

Effect of IMP2 knockout on cell migration in a wound healing assay. (A) LLC1 (B) Huh7, (C) HepG2, and (D) SW480 cells were analyzed for cell migration using the IncuCyte® S3 system over a 48 h period. Representative pictures show the wound area in color at the starting point, 0 h, and 48 h after wounding, and the scale was set to 1 mm. The LLC1 wild-type clone is referred to as WT1, and the LLC1 editing control is referred to as WT2. The Huh7, HepG2, and SW480 were monoallelic IMP2 knockouts. Data are normalized to time point 0 h. Data is represented as mean $\pm$ SEM, n = 3 (quintuplicates). Panels (B–D) display p-values for WT vs. KO cells.