Skip to main page content
U.S. flag

An official website of the United States government

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2024 May 15;23(1):101.
doi: 10.1186/s12943-024-02021-y.

CRISPR-Cas13d screens identify KILR, a breast cancer risk-associated lncRNA that regulates DNA replication and repair

Lu Wang  1   2 , Mainá Bitar  1   2   3 , Xue Lu  1 , Sebastien Jacquelin  1   4 , Sneha Nair  1 , Haran Sivakumaran  1 , Kristine M Hillman  1 , Susanne Kaufmann  1 , Rebekah Ziegman  1 , Francesco Casciello  1 , Harsha Gowda  1 , Joseph Rosenbluh  5   6 , Stacey L Edwards #  7   8   9 , Juliet D French #  10   11   12
Affiliations

CRISPR-Cas13d screens identify KILR, a breast cancer risk-associated lncRNA that regulates DNA replication and repair

Lu Wang et al. Mol Cancer. .

Abstract

Background: Long noncoding RNAs (lncRNAs) have surpassed the number of protein-coding genes, yet the majority have no known function. We previously discovered 844 lncRNAs that were genetically linked to breast cancer through genome-wide association studies (GWAS). Here, we show that a subset of these lncRNAs alter breast cancer risk by modulating cell proliferation, and provide evidence that a reduced expression on one lncRNA increases breast cancer risk through aberrant DNA replication and repair.

Methods: We performed pooled CRISPR-Cas13d-based knockdown screens in breast cells to identify which of the 844 breast cancer-associated lncRNAs alter cell proliferation. We selected one of the lncRNAs that increased cell proliferation, KILR, for follow-up functional studies. KILR pull-down followed by mass spectrometry was used to identify binding proteins. Knockdown and overexpression studies were performed to assess the mechanism by which KILR regulates proliferation.

Results: We show that KILR functions as a tumor suppressor, safeguarding breast cells against uncontrolled proliferation. The half-life of KILR is significantly reduced by the risk haplotype, revealing an alternative mechanism by which variants alter cancer risk. Mechanistically, KILR sequesters RPA1, a subunit of the RPA complex required for DNA replication and repair. Reduced KILR expression promotes breast cancer cell proliferation by increasing the available pool of RPA1 and speed of DNA replication. Conversely, KILR overexpression promotes apoptosis in breast cancer cells, but not normal breast cells.

Conclusions: Our results confirm lncRNAs as mediators of breast cancer risk, emphasize the need to annotate noncoding transcripts in relevant cell types when investigating GWAS variants and provide a scalable platform for mapping phenotypes associated with lncRNAs.

Keywords: Breast cancer; CRISPR; Cas13 screen; Cell proliferation; DNA replication; GWAS; Genetic variants; Long noncoding RNA; lncRNA.

PubMed Disclaimer

Conflict of interest statement

The authors declare no competing interests.

Figures

Fig. 1
Fig. 1
CRISPR-Cas13d screens identify lncRNAs that modulate breast cell proliferation. a Schematic for CRISPR-Cas13d screens. b Scatterplots from CRISPR-Cas13d screen data showing differentially represented crRNAs (red/blue dots; log2[fold-change] > 0.1 and p value < 0.05) targeting candidate genes and lncRNAs. Labels are unannotated breast cancer-associated lncRNAs with FDR ≤ 0.3. c qPCR for KILR expression and (d) cell confluence measured over time using Incucyte in MCF7 cells after Cas13d-KILR knockdown with two independent crRNAs (crKILR1-2). The crCON contains a non-targeting control. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (**p < 0.01, ****p < 0.0001). e WashU genome browser (hg19) showing GENCODE annotated genes (blue) and KILR (green). The breast cancer risk variants are shown as red vertical lines (Signals 1–3). The H3K27ac, H3K4me3 and ER (estrogen receptor) binding tracks from MCF7 cells are shown as black histograms. f A linear schema of KILR. SNP (single nucleotide polymorphism); 5’ TOP (terminal oligopyrimidine tract); NLS (nuclear localization signal). g qPCR for KCTD1, KCTD1-5 and KILR expression in T47D cells after CRISPRa activation of the KCTD1-5 promoter to overexpress KILR with two independent gRNAs (CRa-gKILR1-2). The CRa-gCON contains a non-targeting control. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). h KILR RNA stability assay in MDAMB361 cells after treatment with actinomycin D (ActD), then qPCR for KILR RNA relative to CDKN2A mRNA levels. KILR mRNA half-life (t1/2) was calculated by linear regression analysis. Error bars, SEM (n = 3). i Boxplot of KILR read counts in normal breast tissue from scRNA-seq data clustered based on NB-lncRNA expression [6]. j Boxplot of KILR TPM (transcript per million) in breast tumor samples from TCGA RNA-seq data stratified by tumor subtype. k qPCR for KCTD1-5 and KILR expression in estrogen-stimulated T47D cells. Error bars, SEM (n = 3). p values were determined by two-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). l qPCR after nuclear/cytoplasmic/chromatin fractionation of T47D cells detecting the distribution of the indicated transcripts. Error bars, SD (n = 2). m Representative confocal microscopy images of KILR in MCF7 cells after CRISPRa (CRa-gKILR1-2) stained with Stellaris KILR RNA FISH probes (red). The CRa-gCON contains a non-targeting control. Nuclei were stained with DAPI (blue). Scale bar, 5 μm
Fig. 2
Fig. 2
KILR overexpression inhibits breast cell proliferation and induces apoptosis. a Cell confluence measured over time using Incucyte in breast cells after CRISPRa activation of the KCTD1-5 promoter to overexpress KILR with two independent gRNAs (CRa-gKILR1-2). The CRa-gCON contains a non-targeting control. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). b Representative apoptosis analysis of breast cells after CRISPRa (CRa-gKILR1-2) by double staining with annexin V and PI. The CRa-gCON contains a non-targeting control. The quadrants (Q) were defined as Q1 = live (Annexin V- and PI-negative), Q2 = early stage of apoptosis (Annexin V-positive/PI-negative), Q3 = late stage of apoptosis (Annexin V- and PI-positive) and Q4 = necrosis (Annexin V-negative/PI-positive). c The percentage of cells in early and late-stage apoptosis in each group (Q2 + Q3). Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (*p < 0.05). d Representative apoptosis analysis of breast cells after doxycycline induction of ectopic KCTD1-5 or KILR expression by double staining with annexin V and PI. The Tet-CON represents an empty vector control. The quadrants were defined in (b). e The percentage of cells in early and late-stage apoptosis in each group (Q2 + Q3). Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (*p < 0.05)
Fig. 3
Fig. 3
KILR inhibits DNA replication by sequestering the RPA1 protein. a Representative confocal microscopy images of KILR and RPA1 in MCF7 cells after doxycycline induction of ectopic KCTD1-5 or KILR expression stained with Stellaris KILR RNA FISH probes (red) and immunostained with anti-RPA1 (green) (n = 3). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). White arrows highlight KILR/RPA1 co-localization. Scale bar, 5 μm. b Left panels: Representative images of DNA fibers in MCF7 cells after Cas13d-KILR knockdown with two independent crRNAs (crKILR1-2) then labelling with CldU and IdU. Right panel: Replication fork speed was calculated by length of track/time of CIdU pulse. Data are presented from two independent fiber assays. Error bars, SEM (n = 154). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). c, d Left panels: Representative images of DNA fibers in MCF7 (c) and Hs578T (d) cells after doxycycline induction of KCTD1-5 or KILR, labelling with CldU and IdU then treatment with 4 mM HU for 4 h. Right panels: Ratio of IdU/CldU. Data are presented from two independent fiber assays. Error bars, SEM (n = 150). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). e Representative confocal microscopy images of breast cells after doxycycline induction of ectopic KCTD1-5 or KILR expression immunostained with anti-ɣH2AX (red). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). Scale bar, 10 μm. f Quantification of ɣH2AX foci in three breast cell lines. A cell with > 10 distinct ɣH2AX foci in the nucleus was considered as positive. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (****p < 0.0001). g Representative confocal microscopy images of MCF7 cells after doxycycline induction of ectopic KILR expression with or without RPA1 overexpression immunostained with anti-ɣH2AX (red). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). Scale bar, 10 μm. h Quantification of ɣH2AX foci in MCF7 cells. A cell with > 10 distinct ɣH2AX foci in the nucleus was considered as positive. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (***p < 0.001)
Fig. 4
Fig. 4
KILR overexpression inhibits HR-based repair. a Representative confocal microscopy images of RPA1 and RAD51 in MCF7 cells after doxycycline induction of ectopic KCTD1-5 or KILR expression and exposure to 6-Gy IR (n = 3). 6 h post-IR, cells were immunostained with anti-RPA1 (red) and anti-RAD51 (green). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). Scale bar, 10 μm. b Quantification of RPA1 or RAD51 foci in MCF7 cells. A cell with > 5 distinct RPA1 or RAD51 foci in the nucleus was considered as positive. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (**p < 0.01). c Representative confocal microscopy images of KILR and RPA1 in MCF7 cells after doxycycline induction of ectopic KCTD1-5 or KILR expression and exposure to 6-Gy IR (n = 3). 6 h post-IR, cells were stained with Stellaris KILR RNA FISH probes (red) and immunostained with anti-RPA1 (green). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). White arrows highlight KILR/RPA1 co-localization. Scale bar, 10 μm. d Representative confocal microscopy images of MCF7 cells after doxycycline induction of ectopic KILR expression with or without RPA1 overexpression and exposure to 6-Gy IR (n = 3). 6 h post-IR, cells were immunostained with anti-RAD51 (green). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). Scale bar, 5 μm. e Quantification of RAD51 foci in MCF7 cells. A cell with > 5 distinct RAD51 foci in the nucleus was considered as positive. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (**p < 0.01). f Representative confocal microscopy images of KILR and RPA1 in MCF10A cells after doxycycline induction of ectopic KCTD1-5 or KILR expression and exposure to 6-Gy IR (n = 3). 6 h post-IR, cells were stained with Stellaris KILR RNA FISH probes (red) and immunostained with anti-RPA1 (green). The Tet-CON represents an empty vector control. Nuclei were stained with DAPI (blue). White arrows highlight KILR/RPA1 co-localization. Scale bar, 5 μm. g Representative confocal microscopy images of ɣH2AX in MCF7 cells after Cas13d-KILR knockdown with two independent crRNAs (crKILR1-2) and exposure to 6-Gy IR (n = 3). 6 h post-IR, cells were immunostained with anti-ɣH2AX (red). The crCON contains a non-targeting control. Nuclei were stained with DAPI (blue). Scale bar, 10 μm. Quantification of ɣH2AX foci in MCF7 cells. A cell with > 10 distinct ɣH2AX foci in the nucleus was considered as positive. Error bars, SEM (n = 3). p values were determined by one-way ANOVA and Dunnett’s multiple comparisons test (*p < 0.05, **p < 0.01)
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

References

    1. Edwards SL, Beesley J, French JD, Dunning AM. Beyond GWASs: illuminating the dark road from association to function. Am J Hum Genet. 2013;93:779–797. doi: 10.1016/j.ajhg.2013.10.012. - DOI - PMC - PubMed
    1. Umans BD, Battle A, Gilad Y. Where are the disease-associated eQTLs? Trends Genet. 2021;37:109–124. doi: 10.1016/j.tig.2020.08.009. - DOI - PMC - PubMed
    1. Mattick JS, et al. Long non-coding RNAs: definitions, functions, challenges and recommendations. Nat Rev Mol Cell Biol. 2023;24:430–447. doi: 10.1038/s41580-022-00566-8. - DOI - PMC - PubMed
    1. Statello L, Guo CJ, Chen LL, Huarte M. Gene regulation by long non-coding RNAs and its biological functions. Nat Rev Mol Cell Biol. 2021;22:96–118. doi: 10.1038/s41580-020-00315-9. - DOI - PMC - PubMed
    1. Bridges MC, Daulagala AC, Kourtidis A. LNCcation: lncRNA localization and function. J Cell Biol. 2021;220:e202009045. doi: 10.1083/jcb.202009045. - DOI - PMC - PubMed