A Cornelia de Lange syndrome NIPBL 5'‑UTR mutation reduces cell proliferation in an in vitro model by downregulating RAD21 and β‑catenin

Introduction

Cornelia de Lange Syndrome (CdLS; OMIM 122470, 300590, 610759, 614701, 300882, 617116, 618736), also referred to as Brachmann-de Lange syndrome, is the most common disorder among the chromatinopathies, with a prevalence estimated between 1 in 10,000 to 1 in 30,000 live births (1). This condition is characterized by a variety of anomalies that affect the craniofacial region, musculoskeletal system, gastrointestinal tract and neurodevelopment (2). CdLS belongs to the group of disorders known as chromatinopathies, disorders caused by variants in proteins involved in chromatin remodeling and transcriptional regulation, resulting in global dysregulation of gene expression (3). CdLS results from mutations in any of the seven genes that constitute the cohesin complex: NIPBL, SMC1A, SMC3, RAD21, HDAC8, ANKRD11, MAU2, AFF2 and BRD4, each playing a role in the construction or regulation of the complex (4,5). The NIPBL gene mutation is most frequently implicated in CdLS cases, particularly those exhibiting the typical phenotype, accounting for ~60-70% of instances (6). Most NIPBL mutations associated with CdLS are located in the coding region, while a few are in the 5'-UTR (untranslated region) (7-9). Previously, a novel mutation c.-467C>T was identified in the NIPBL gene, which is associated with CdLS. Cs (10). Although the 5'-UTR mutation in NIPBL linked to CdLS is recognized for introducing a novel upstream open reading frame (uORF) (7), the underlying pathogenic mechanisms remain unclear.

The 5'-UTR includes structural elements and uORFs, which are essential for regulating translation. The extended length and intricacy of 5'-UTRs underscore the significance of post-transcriptional and translational mechanisms in modulating the expression levels of the proteins they encode, particularly in the context of dose-sensitive genes (11). The NIPBL protein, with >2,500 amino acids, is crucial for the function of the cohesin complex and its role in chromatin organization, gene regulation and developmental processes (12,13). In total, ~10% of CdLS cases involve mutations in cohesin regulators and components, indicating that NIPBL mutations may contribute to CdLS pathogenesis by disrupting cohesin function (14). The expression levels of the NIPBL gene are critical, as even minor decreases in them can lead to CdLS, indicating its dosage sensitivity (14,15). Cells derived from patients with CdLS with a heterozygous pathogenic variant in the NIPBL gene, as well as mice heterozygous for a Nipbl mutation, exhibit decreased overall or regional cohesin association and impaired three-dimensional interactions within the chromatin structure, which is consistent with the role of NIPBL as a component that assists in the attachment of cohesion (16-19).

The present study established a heterozygous cell line with a point mutation to investigate the impact of the 5'-UTR mutation on NIPBL gene expression, cohesin complexes and the transcription of genes involved in cellular development. The present study aimed to further elucidate the preliminary mechanisms by which the 5'-UTR mutation of the NIPBL gene contributes to CdLS development.

Materials and methods

RNA secondary structure prediction

The RNA secondary structure was computed, and the folding energy for RNA sequences was estimated using the RNAfold tool, which is part of the ViennaRNA package (version 2.6.4), as previously described (20).

Cell culture and knock-in cell line construction

The 293t cell line was sourced from the American Type Culture Collection repository of the Chinese Academy of Sciences. These cells were maintained in Dulbecco's Modified Eagle Medium supplemented with 10% fetal bovine serum (Shanghai ExCell Biology, Inc.) and cultured at 37˚C in a humidified atmosphere with 5% CO2. A variant of the 293t cell line, designated as NIPBL-mut, was engineered to harbor a 5'-UTR mutation in the NIPBL gene using the CRISPR/Cas9 system. The sgRNA targeting the NIPBL 5'-UTR was designed with CRISPOR: web-based design tool (https://crispor.tefor.net; Kircher Lab, Max Delbrück Center for Molecular Medicine, Berlin, Germany). The NIPBL-specific sgRNA sequence (5'-TTTGTTCTGAGAGGGAGAGA-3'), designed to target the first exon of the NIPBL gene, was cloned into the PX459 vector, which was acquired from Addgene, Inc. The sgRNA-PX459 construct (2 µg in 6 well plate) and the homology repair template (500 ng; 5'-GTCGGCATTTTGTTCTGAGAGGGAGAGATGGAACGAGA-3') were co-transfected into 293t cells using Lipofectamine 2000 transfection reagent (Invitrogen; Thermo Fisher Scientific, Inc.) at 37˚C for 36 h. Following transfection, the cells were subjected to selection with 2 µg/ml puromycin for 48 h, after which the antibiotic-resistant cells were single-cell cloned in the same medium. Genomic DNA was extracted from the edited cells (Tissue DNA Kit; Omega Bio-Tek, Inc.) and a 517-bp fragment spanning the entire sgRNA insertion site (chr5: 36,876,519-36,877,036, hg38) was amplified with PrimeSTAR® HS DNA Polymerase (Takara Biotechnology Co., Ltd.) under standard high-fidelity cycling conditions (98˚C for 30 sec; 35 cycle x [98˚C for 10 sec, 63˚C for 20 sec, 72˚C for 30 sec]; 72˚C for 2 min). The purified 517-bp product was then sent to Tsingke Biological Technology for Sanger sequencing with the forward PCR primer (5'-GCAAACGTCGCTCTGAAGCAT-3') and a reverse primer (5'-CGAGAAGACAAATCGTTGCT-3'). Chromatograms were aligned to the reference gene sequence with SnapGene to confirm 100% identity and absence of indels. The template was verified on three separate occasions to confirm sequence identity.

Luciferase reporter assay

The procedures for plasmid construction and dual-luciferase reporter analysis were conducted as previously described by the authors (10). The data were standardized to account for transfection efficiency by quantifying the activities of firefly and Renilla luciferases, with the results presented as a ratio of these two luciferase activities. Most control samples were set to a value of 100% for normalization purposes. The results are depicted as the mean values ± standard deviations, derived from triplicate experiments.

Immunofluorescence

WT and mut 293t cells were seeded onto glass coverslips in their respective growth media. The cells were cultured for the specified number of days until they reached ~80% confluence. The cell layers were then fixed with a 4% paraformaldehyde solution in phosphate-buffered saline (PBS, pH 7.5) at room temperature (RT, 25˚C) for 20 min. Following fixation, the coverslips were permeabilized with a 0.2% Triton X-100 solution for 15 min and then were blocked with 3% bovine serum albumin (BSA; Sangon Biotech Co., Ltd.) for 1 h at RT. Primary antibodies targeting NIPBL (cat. no. sc-374625; Santa Cruz Biotechnology, Inc.), diluted to a ratio of 1:100 in 1% BSA, were then applied, and the solution was incubated overnight (for >18 h) at 4˚C in a light-protected environment. Subsequently, the Alexa Fluor 488 Goat Anti-Mouse IgG (H+L) secondary antibodies (cat. no. ab150113; Abcam) were incubated for 1 h at RT. The cells were then stained at RT with 0.2 mg/ml DAPI for 7 min to visualize the nuclei. After staining, the slides were air-dried and mounted with anti-fade reagents to preserve the fluorescence. Finally, the samples were analyzed using a fluorescence microscope.

Cell fractionation

This process started with the collection of cells and their subsequent incubation in lysis buffer I, which is composed of 20 mM HEPES at a pH of 8.0, 2 mM MgCl2, 10 mM KCl, 0.5% NP-40 and protease inhibitors, for 10 min, under refrigeration at 4˚C and with the inclusion of protease inhibitors in all solutions. The mixture was centrifuged at 1,500 x g for 5 min after incubation. The supernatant, which was rich in cytosolic components, was then carefully separated and set aside. The remaining pellet was rinsed twice with PBS and then subjected to lysis using nucleus lysis buffer II, which is prepared by incorporating 0.5 M NaCl into lysis buffer I. After lysis, the nuclear lysate was allowed to chill on ice for 5 min before centrifugation at 15,000 x g for 10 min. The supernatant, containing the nuclear components, was subsequently collected and preserved as the nuclear fraction.

Western blot analysis

Cells were subjected to lysis using RIPA buffer (cat. no. P0013C; Beyotime Institute of Biotechnology) supplemented with phenylmethylsulfonyl fluoride. The concentration of total proteins in the lysates was quantified using a bicinchoninic acid protein assay kit (Beyotime Institute of Biotechnology). Aliquots of 20 µg of protein samples were resolved using sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 10% acrylamide gel and then transferred onto polyvinylidene fluoride membranes. The membranes were then blocked with a 5% skim milk in TBS-T at RT for 1 h. After blocking, the membranes were incubated overnight at 4˚C with a panel of primary antibodies targeting NIPBL (cat. no. sc-374625; Santa Cruz Biotechnology, Inc.), α-Tubulin (cat. no. ab7291; Abcam), RAD21 (cat. no. DF7520; Abcam), SMC1A (cat. no. AF6439; Affinity Biosciences), CTNNB1 (cat. no. 8480; Cell Signaling Technology, Inc.) and GAPDH (cat. no. ab8245; Abcam). Subsequently, the membranes were treated with HRP-Goat Anti-rabbit IgG (H+L) or HRP-Goat Anti-mouse IgG (H+L) secondary antibodies (BK-R050-50 µl or BK-M050-50 µl΄ Hangzhou Baoke Biotech Co., Ltd.), diluted at 1:5,000, for 1 h. The immunoreactive bands were visualized using an enhanced chemiluminescence detection system (Hangzhou Fude Biological Technology Co., Ltd.) and their intensities were quantified through densitometric analysis using ImageJ 1.54f (National Institutes of Health). The following primary antibodies were used: NIPBL (1:500; Santa Cruz Biotechnology, Inc.); α-Tubulin and GAPDH (1:1,000; both from Abcam); RAD21 and SMC1A (1:1,000; both from Affinity Biosciences) and CTNNB1(1:1,000; Cell Signaling Technology, Inc.).

EDU staining

The DNA replication in 293t cells was detected using an EdU staining kit (Beyotime Institute of Biotechnology), following the manufacturer's instructions. Initially, 293t cells were plated in 24-well plates and cultured for 12 h until they reached a confluency of ~50%. Subsequently, the cells were treated with a 10 µM concentration of EdU solution, which was incorporated into the cell medium and left for 2 h. After this incubation period, the cells were fixed with 4% paraformaldehyde in PBS (pH 7.4) at room temperature for 15 min and then rinsed three times (5 min each) with PBS to remove any unincorporated EdU. A click reaction solution containing Azide 488 was then added to the cells and incubated for 30 min to facilitate the detection of EdU incorporation. To visualize the cell nuclei, 0.2 mg/ml DAPI staining was performed at RT for 7 min, and the stained cells were subsequently examined using a fluorescence microscope.

Statistical analyses

The experiments were repeated at least three times unless otherwise indicated. Statistical analyses were performed using unpaired two-tailed t-tests in GraphPad Prism 10.1.2 (GraphPad Software; Dotmatics). Most control groups were normalized to 100%. The data are presented as the mean ± standard error of the mean. P<0.05 was considered to indicate a statistically significant difference.

Results

Impact of a 5'-UTR mutation on uORFs

To assess the impact of the c.-467C>T variant on post-transcriptional regulation, the 5'-UTR of NIPBL was analyzed for uORFs using NCBI ORFfinder. A total of Four uORFs were identified in the wild-type (WT) sequence, whereas the mutant harbored an additional uORF, yielding a total of five (Fig. 1A). The de novo uORF initiates at a strong Kozak sequence (GAGAUGG) (21) and spans 53 codons, terminating upstream of the NIPBL start codon (Fig. 1B). Secondary structure prediction with ViennaRNA revealed a complex secondary RNA structure for the mutated 5'-UTR, characterized by numerous stem-loops (Fig. 1C), with no significant difference in the lowest free energy between the WT (-2548.10 kcal/mol) and the mutant (-2546.70 kcal/mol) for the initial 10,000 bases.

Figure 1 Impact of 5'-UTR mutation on uORFs. (A) Identification and comparison of uORFs between the WT and mut variants of the 5'-UTR of the NIPBL gene. (B) Kozak sequence within the 5'-UTR of the mut variant of the NIPBL gene. (C) Predicted secondary structures of the RNA for the mutated 5'-UTR of the NIPBL gene, and the WT RNA predicted secondary structure is shown on the left and the mut structure on the right. UTR, untranslated region; uORF, upstream open reading frame; WT, wild-type; mut, mutant.

Impact of uORF variants on NIPBL expression

The CRISPR/Cas9 system-mediated editing introduced the c.-467C>T variant into the endogenous NIPBL locus, yielding heterozygous NIPBL-mut cells (Fig. 2A). A dual-luciferase reporter assay was used to quantify the transcriptional impact of the variant. Compared with the WT, the mutant (mut) 5'-UTR produced a modest but significant reduction in reporter activity (1±0.046 vs. 0.693±0.019; P=0.0034; Fig. 2B). Consistently, immunoblot analysis revealed a ~50% decrease in endogenous NIPBL protein levels in mutant cells relative to isogenic controls (0.211±0.027 vs. 0.100±0.016; P=0.0225; Fig. 2C). Immunofluorescence staining further showed that the subcellular distribution of NIPBL was unaltered by the mutation. However, the fluorescence intensity in the mutant cells was significantly reduced by 40% compared with that in the WT (103.000±2.767 vs. 67.680±3.323; P=0.0012; Fig. 2D).

Figure 2 5'-untranslated region mutation of the NIPBL gene affects expression. (A) Schematic illustration depicting the generation of NIPBL-mut cells using the CRISPR/Cas9 System. (B) Luciferase reporter assay measuring the transcriptional activity of the NIPBL gene in WT vs. mutant groups. (C) Western blot comparing NIPBL protein levels in WT and mut cells. (D) Immunofluorescence staining of NIPBL in wild-type and mutant cells; nuclei were counterstained with DAPI (blue). Fluorescence intensity was quantified and normalized to the WT level. Scale bar, 100 µm. Data are presented as the mean ± SEM and the value of NIPBL-WT was normalized as 100%. **P<0.01. WT, wild-type; mut, mutant.

NIPBL haploinsufficiency downregulates RAD21

It was next examined whether the NIPBL mutation affects the steady-state abundance of core cohesin subunits. Whole-cell lysates revealed a significant reduction in RAD21 protein (1.020±0.257 vs. 0.449±0.141; P=0.0297; Fig. 3A). To determine compartment-specific changes, nuclei and cytoplasm were fractionated. Consistent with our previous findings, NIPBL protein levels decreased in both the nucleus (0.476±0.070 vs. 0.199±0.070, P=0.0489 Fig. 3B) and cytoplasm (0.623±0.081 vs. 0.354±0.021, P=0.0326, Fig. 3B). Similarly, RAD21 expression in cellular fractions decreased in the mut compared with the WT (Cytoplasm, 0.283±0.042 vs. 0.103±0.025; P=0.0201; nucleus 0.224±0.041 vs. 0.067±0.013; P=0.0205; Fig. 3B). By contrast, SMC1A levels were comparable between genotypes in both compartments (Fig. 3B).

Figure 3 Cohesin complex integrity. (A) RAD21 protein levels in WT and mut cells. (B) Western blot analysis of cohesin complex proteins (NIPBL, RAD21 and SMC1A) in cytoplasmic (Cyto) and nuclear (Nuc) fractions of wild-type and mutant cells. Data are presented as the mean ± SEM and the value of NIPBL-WT was normalized as 100%. *P<0.05. WT, wild-type; mut, mutant; ns, not significant.

Reduction in NIPBL expression leads to decreased β-catenin and impacts cell proliferation

EdU staining assays were conducted to quantify the proliferative capacity ofNIPBL-mutant cells. Compared with WT controls, the fraction of EdU-positive cells were significantly lower in the mutant population (49.360±1.384 vs. 40.200±2.154; P=0.0025; Fig. 4A). Western blot analysis revealed that total β-catenin protein levels were significantly reduced in mut cells compared with WT controls (1.129±0.039 vs. 0.692±0.028, P=0.0008; Fig. 4B). Since nuclear accumulation of β-catenin is required for canonical Wnt signaling, β-catenin was further resolved into nuclear and cytoplasmic fractions. Nuclear β-catenin was significantly decreased in the mut group (2.301±0.328 vs. 1.213±0.152; P=0.0398; Fig. 4C); cytoplasmic β-catenin levels remained unaffected in both the WT and mutant groups.

Figure 4 NIPBL downregulation reduces β-catenin levels, thereby affecting cell proliferation. (A) EdU assays assessing proliferation in WT and NIPBL-mut cells; EdU-positive nuclei were quantified and normalized to total DAPI counts. Scale bar, 100 µm. (B) Total β-catenin protein levels in WT and mut cells. (C) Subcellular β-catenin levels in WT and mut cells (cytoplasmic vs. nuclear fractions). Data are presented as the mean ± SEM and the value of NIPBL-WT was normalized as 100%. *P<0.05, **P<0.01 and ***P<0.001. WT, wild-type; mut, mutant; ns, not significant.

Discussion

In our previous publication, a novel NIPBL variant (c.-467 C>T) was identified in a Chinese boy affected by CdLS (10). In the present study, it was demonstrated that this mutation creates a strong uORF that represses NIPBL translation without affecting mRNA stability. Mechanistically, the uORF stalls ribosome scanning, preventing initiation at the downstream NIPBL start codon (22). Critically, reduced NIPBL diminishes cohesin loading onto chromatin, disrupting long-range chromatin interactions, a core defect in CdLS pathogenesis (23,24). These data establish uORF-mediated translational inhibition as a recurrent mechanism underlying NIPBL haploinsufficiency in CdLS.

Reduced NIPBL dosage, in turn, perturbs the cohesin complex. Western blotting and subcellular-fractionation analyses revealed a decrease in RAD21 protein. This finding contrasts with some studies in which NIPBL mutations altered cohesin chromosomal occupancy without affecting total cohesin subunit abundance (25,26). However, it aligns with recent evidence from small-cell lung cancer showing that NIPBL protein promoted RAD21 gene transcription by enhancing H3K27 demethylation via recruiting lysine demethylase 6B to the RAD21 gene promoter (27), reconciling our findings with the emerging view that NIPBL dosage can exert both occupancy and abundance-dependent effects on cohesin. Because cohesin dysfunction underlies the group of disorders termed chromatinopathies, CdLS being the archetype, the observed RAD21 downregulation provides a direct mechanistic link between the NIPBL 5'-UTR mutation and CdLS pathogenesis (28,29). These findings suggest that downregulation of RAD21 may be caused by pathogenic mutations in the NIPBL 5'-UTR. This interpretation is further supported by clinical genetics: Heterozygous RAD21 variants can themselves cause CdLS, but the resulting phenotype is typically mild (30-32), mirroring the attenuated presentation associated with NIPBL 5'-UTR mutations. Thus, converging molecular and clinical evidence indicates that reduced NIPBL dosage elicits RAD21-mediated disruption of cohesin that contributes to the milder end of the CdLS phenotypic spectrum.

To determine how NIPBL-RAD21 deficiency translates into cellular pathology, the Wnt/β-catenin pathway was interrogated. Functional assays revealed modest yet consistent defects in cell proliferation that were accompanied by a marked reduction in nuclear β-catenin. This result is consistent with the downregulation of the classical Wnt signaling pathway observed in the CdLS zebrafish model (33,34), and extends it to human cells. Importantly, this finding also supports the notion that NIPBL or RAD21 knockdown leads to a decrease in cell proliferation ability, suggesting a causal rather than correlative relationship (15,33,35). Supporting this notion, transcriptomic profiling of CdLS-associated cardiac dysplasia demonstrates that RAD21 depletion alone is sufficient to derail Wnt signaling (8). Consequently, it was hypothesized that NIPBL mutations trigger the downregulation of RAD21, leading to the dysregulation of β-catenin and ultimately to defects in cell proliferation.

In conclusion, it was demonstrated that a single 5'-UTR variant (c.-467C>T) creates a de novo uORF, quantitatively reduces NIPBL and cohesin levels, and thereby disrupts Wnt signalling and proliferation. These findings establish a direct molecular link between a non-coding mutation and mild Cornelia de Lange phenotypes.

Acknowledgements

Not applicable.

Funding

Funding: The present study was supported by the National Natural Science Foundation of China (grant no. 82270837).

Availability of data and materials

The data generated in the present study are included in the figures and/or tables of this article.

Authors' contributions

CW and CZ conceived and designed the research. QC and YC performed the experiments and analyzed the data. QC, YC and CW wrote the manuscript. All authors read and approved the final version of the manuscript. QC and YC confirm the authenticity of all the raw data.

Ethics approval and consent to participate

Not applicable.

Patient consent for publication

Not applicable.

Competing interests

The authors declare that they have no competing interests.

References