Exploration of the regulatory relationship between KRAB-Zfp clusters and their target transposable elements via a gene editing strategy at the cluster specific linker-associated sequences by CRISPR-Cas9

Abstract

Background: Krüppel Associated Box-containing Zinc Finger Proteins (KRAB-ZFPs), representing the largest superfamily of transcription factors in mammals, are predicted to primarily target and repress transposable elements (TEs). It is challenging to dissect the distinct functions of these transcription regulators due to their sequence similarity and diversity, and also the complicated repetitiveness of their targeting TE sequences.

Results: Mouse KRAB-Zfps are mainly organized into clusters genomewide. In this study, we revealed that the intra-cluster members had a close evolutionary relationship, and a similar preference for zinc finger (ZnF) usage. KRAB-Zfps were expressed in a cell type- or tissue type specific manner and they tended to be actively transcribed together with other cluster members. Further sequence analyses pointed out the linker sequences in between ZnFs were conserved, and meanwhile had distinct cluster specificity. Based on these unique characteristics of KRAB-Zfp clusters, sgRNAs were designed to edit cluster-specific linkers to abolish the functions of the targeted cluster(s). Using mouse embryonic stem cells (mESC) as a model, we screened and obtained a series of sgRNAs targeting various highly expressed KRAB-Zfp clusters. The effectiveness of sgRNAs were verified in a reporter assay exclusively developed for multi-target sgRNAs and further confirmed by PCR-based analyses. Using mESC cell lines inducibly expressing Cas9 and these sgRNAs, we found that editing different KRAB-Zfp clusters resulted in the transcriptional changes of distinct categories of TEs.

Conclusions: Collectively, the intrinsic sequence correlations of intra-cluster KRAB-Zfp members discovered in this study suggest that the conserved cluster specific linkers played crucial roles in diversifying the tandem ZnF array and the related target specificity of KRAB-Zfps during clusters' evolution. On this basis, an effective CRISPR-Cas9 based approach against the linker sequences is developed and verified for rapidly editing KRAB-Zfp clusters to identify the regulatory correlation between the cluster members and their potential TE targets.

Keywords: CRISPR-Cas9; KRAB-Zfp cluster; Linker sequence; Transposable elements.

Fig. 1

Analyses of the genomewide distribution, the evolutionary relationship, and the properties of ZnF usage in mouse KRAB-Zfp genes. A A circular genomic map displays the distribution of all KRAB-Zfp genes across the mouse genome. The width of each purple bar positively correlates to the number of KRAB-Zfp genes in a cluster. B A phylogenetic tree demonstrates the genetic distance among all the KRAB-Zfp genes located within KRAB-Zfp clusters 3 and 5. Bootstrap is a computer system function, representing the reliability of the branches of an evolutionary tree, ranging from 0 to 1. The larger the bootstrap value, the more reliable the branches of the evolutionary tree are. C Probability heatmap of the composition of ZnFs for KRAB-ZFPs in clusters 3, 4, and 5, including their identity and the frequency of occurrence. The identity of a ZnF is designated by its unique fingerprint (the amino acid triplets), see Fig. 3A for more details. Each square represents a unique type of ZnF/fingerprint and the intensity of the red color correlates to its frequency of occurrence in the corresponding KRAB-ZFP labeled on the right

Fig. 2

Analyses of the expression profiles of KRAB-Zfp clusters in various mouse cell lines and tissues. The heatmaps of the combined RNA-Seq data for the indicated cell lines (the detailed sample information is indicated in the table) and tissues are shown on the top. A The KRAB-Zfps highly expressed in mESC were selected (transcripts per million base pairs, TPM > = 25) and compared with the other cell types for their expression as indicated. B The KRAB-Zfps highly expressed in the forebrain were selected (TPM > = 4) and compared with the other indicated tissue types. For the main clusters highly expressed in mESC or in the forebrain, the ratios of the highly expressed KRAB-Zfp genes to the total number of KRAB-Zfp genes present in each cluster were calculated and shown in the tables at the bottom

Fig. 3

The sequence conservation analyses of the zinc finger domains of KRAB-ZFPs. (A) A schematic model of the interaction between a typical C2H2 zinc finger domain of KRAB-ZFP and its target DNA. (B) Sequence conservation analyses of the amino acid sequences (top) and the corresponding nucleic acid sequences (bottom) according to the regions shown in the bean green box or the pale purple box in (A). The related sequences were extracted from all the KRAB-Zfps across the mouse genome (a), within the KRAB-Zfps in cluster 3 (b), within the clusters 4 and 5 (c), or from one randomly selected KRAB-Zfp gene, namely KRAB-Zfp987 (d)

Fig. 4

Screening effective sgRNAs for CRISPR-Cas9 mediated knockout of the KRAB-Zfp gene cluster(s). A A diagram illustrates the methodology of a CRISPR-Cas9 based reporter assay for screening the effective sgRNAs with multi-targets. SgRNAs that target to the cluster-specific linker sequences of KRAB-Zfps were designed and cloned into pX459, respectively. Their targeting sequences (including the corresponding PAM sites) were arranged in an array and fused together with the mCherry encoding sequences (PEST was added to accelerate the turn-over rate of the mCherry protein) and the entire open reading frame was ligated into the mammalian expression vector, pcDNA3.1. B Testing the screening assays in (A) with an empty pX459 vector (a negative control) and two positive sgRNA controls against Zfp809. The representative fluorescence image for each transfection was shown on top. The images were pseudo-colored to facilitate the recognition of some special readers. The bar graph at the bottom demonstrates the measured fluorescence changes of mCherry upon the expression of Cas9 and the indicated positive control sgRNAs. C A diagram demonstrates the target specificity and the numbers of the predicted cleavage sites of the sgRNAs specifically designed for the indicated KRAB-Zfp cluster(s). The color intensity is proportional to the number of the predicted cleavage sites. D The fluorescence changes of the mCherry reporter upon the expression of Cas9 together with the indicated sgRNAs listed in (C). The statistical analyses were carried out to compare each sgRNA’s effect to a negative control with an empty pX459 plasmid being transfected. “**” stands for an extremely significant change (p-value is smaller than 0.01); “*” stands for a significant change (p-value is greater than 0.01 but smaller than 0.05); n.s. stands for no significant change (p-value is greater than 0.05)

Fig. 5

CRISPR-Cas9 mediated gene editing of KRAB-Zfp clusters and the effects on TE transcription in mESC. The mESC cell lines stably expressing the indicated sgRNA and inducibly expressing Cas9 were analyzed. A Verification of the efficiency and specificity of CRISPR-Cas9 mediated cleavage events in the targeted KRAB-Zfp clusters via PCR. Top panel: A Diagram of the primer design strategies. Primer pairs for directly detecting the zinc finger domains of the targeted KRAB-Zfps are indicated by F1 and R1 (the green and brown arrows) and the primers used to detect the intergenic regions of the two targeted KRAB-Zfp genes are indicated by F2 and R2 (the red and dark blue arrows). The bottom panel: The representative PCR or qPCR results for detecting the genomic DNA templates extracted before and after gene editing (−Dox and + Dox). The type of the primers is labeled on top using the arrow pairs with colors consistent with the diagram. For the specificity tests, the sgRNA labeled in red on top of the agarose gel images are the specific sgRNA designed for the KRAB-Zfp cluster of the genes tested, and the sgRNAs in black are sgRNAs designed for some other non-target clusters. The PCR products amplified from the original, non-edited control samples are indicated by pink arrows. An irrelevant DNA fragment was amplified in parallel and used as an internal control (the yellow arrow) for checking the quality and quantity of the templates used in the PCR reactions. The layout of the KRAB-Zfp genes located within the tested clusters are also shown on the top of the results and the ones with the effective PCR data are highlighted in the red boxes. For qPCR results shown at the bottom, the amplicons were located at the downstream intergenic regions of the indicated KRAB-Zfp genes. B RT-qPCR detection of the relative expression levels of the indicated classes of the repetitive DNA elements in mESC before and after gene editing (−Dox and + Dox) triggered by the sgRNAs labeled on the left. The qPCR data of the indicated TEs were normalized to the data of the Gapdh transcript, and the normalized data of −/+Dox samples of the sgRNA expressing mESC were further normalized to the data generated from the control cell line expressing the empty vector without any relevant sgRNA