Cytosine Base Editor (hA3A-BE3-NG)-Mediated Multiple Gene Editing for Pyramid Breeding in Pigs

Abstract

Pig is an important agricultural economic animal, providing large amount of meat products. With the development of functional genomics and bioinformatics, lots of genes and functional single nucleotide polymorphisms (SNPs) related to disease resistance and (or) economic traits in pigs have been identified, which provides the targets for genetic improvement by genome editing. Base editors (BEs), combining Cas9 nickase and cytidine or adenine deaminase, achieve all four possible transition mutations (C-to-T, A-to-G, T-to-C, and G-to-A) efficiently and accurately without double strand breaks (DSBs) under the protospacer adjacent motif (PAM) sequence of NGG. However, the NGG PAM in canonical CRISPR-Cas9 can only cover approximately 8.27% in the whole genome which limits its broad application. In the current study, hA3A-BE3-NG system was constructed with the fusion of SpCas9-NG variant and hA3A-BE3 to create C-to-T conversion at NGN PAM sites efficiently. The editing efficiency and scope of hA3A-BE3-NG were confirmed in HEK293T cells and porcine fetal fibroblast (PFF) cells. Results showed that the efficiency of hA3A-BE3-NG was much higher than that of hA3A-BE3 on NGH (H = A, C, or T) PAM sites (21.27 vs. 2.81% at average). Further, nonsense and missense mutations were introduced efficiently and precisely via hA3A-BE3-NG in multiple pig economic trait-related genes (CD163, APN, MSTN, and MC4R) in PFF cells by one transfection. The current work indicates the potential applications of hA3A-BE3-NG for pyramid breeding studies in livestock.

Keywords: NGN PAM; base editing; hA3A-BE3-NG; multiple gene editing; pyramid breeding.

Figure 1

Construction of hA3A-BE3-NG vector for expanded targeting scope. (A) Pie chart shows the proportion of porcine genomic sites that can be targeted by SpCas9 or SpCas9-NG with distinct protospacer adjacent motif (PAM) specificities (NGG or NG). Pig reference genome (Sscrofa11.1) was used for analysis. (B) Schematic of the pCMV-hA3A-BE3-NG vector. Compared to SpCas9 (D10A) in hA3A-BE3, SpCas9-NG (D10A) in Target-AID-NG contained seven amino acids variants: R1335V, L1111R, D1135V, G1218R, E1219F, A1322R, and T1337R. hA3A-BE3-NG was constructed by in-fusion cloning of a restriction fragment digested via BsrGI and PmeI from hA3A-BE3, PCR fragment 1 amplified from hA3A-BE3 via Fw1/Rv1 primers, and PCR fragment 2 amplified from Target-AID-NG via Fw2/Rv2 primers. Overlapping sequences exist in the junction of the three different fragments. (C) The gel image indicates that hA3A-BE3 was digested into two fragments by BsrGI and PmeI. The PCR fragment 1 (387 bp) was amplified from hA3A-BE3 via Fw1/Rv1 primers, and the PCR fragment 2 (2,556 bp) was amplified from Target-AID-NG via Fw2/Rv2 primers. The large fragment (5,570 bp) from hA3A-BE3, PCR fragment 1 and PCR fragment 2 were fused into a recombinant vector, hA3A-BE3-NG, which was confirmed by a PCR product (540 bp) amplified via Fw3/Rv3 primers. (D) The chromatograms of Sanger sequencing show the junctional sequence was accurate among the above three fragments in recombinant hA3A-BE3-NG.

Figure 2

Efficient C-to-T conversion in human HEK293T cells by hA3A-BE3-NG. (A) C-to-T editing by hA3A-BE3, hA3A-BE3-NG, and Target-AID-NG at four endogenous EMX1 gene sites in human HEK293T cells. The target base in the editing window is shown, counting the end distal to the PAM as position 1. Data are represented as the mean ± SEM (n = 3). (B) Sanger sequencing results of HEK293T cells transfected with hA3A-BE3, hA3A-BE3-NG, or Target-AID-NG. Red boxes indicate PAMs and blue lines indicate sgRNA sequences. Red arrows indicate substituted nucleotides. (C) Statistical analysis of the C-to-T editing frequency induced by hA3A-BE3, hA3A-BE3-NG, or Target-AID-NG at NGH PAM sites in (A). The median and interquartile range (IQR) are shown; ^** p < 0.01.

Figure 3

Precision missense mutation using hA3A-BE3-NG to expand the editing scope in porcine fetal fibroblast (PFF) cells. (A) Representation of the C-to-T conversion induced by base editors to generate stop codons. The base editors convert CAA, CAG, and CGA codons to stop codons (red) in the sense strand. The TGG codon is converted to stop codons (blue) through G-to-A conversion. (B) Schematic of the target sites at the porcine CD163, APN, and MSTN loci. The target sites indicated by the black arrows can generate stop codons using base editors (BEs). The forward direction of arrow indicates sgRNA-matched anti-sense strand, and vice versa. Total of 32 sgRNAs were designed (A1–19, C1–7, and M1–6). (C) Base editing at 32 NGN PAM sites by hA3A-BE3 and hA3A-BE3-NG. The target sites covered all 16 possible NGN PAM combinations, counting the end distal to the PAM as position 1. (D) Statistical analysis of the C-to-T editing frequency induced by hA3A-BE3 or hA3A-BE3-NG at a total of 32 endogenous target sites. The median and IQR are shown. (E) Schematic of the target site at MC4R locus. MC4R c.893G>A could be produced by hA3A-BE3-NG. The PAM sequence and substituted base are shown in blue and red, respectively. (F) Base editing at the MC4R locus by hA3A-BE3 and hA3A-BE3-NG. In (C,F) values were shown as mean ± SEM (n = 3); ^** p < 0.01 and ^* p < 0.05.

Figure 4

hA3A-BE3-NG-mediated base editing at multiple genes in PFF cells. (A) Summary of multiple sites base editing by hA3A-BE3-NG in PFF cells. (B) Sanger sequencing results of selected single-cell colonies. 11# and 49# colonies have mutations on three genes, and 50# colony has mutations on all four genes. The red box indicates the PAMs and the blue line indicates the sgRNA sequence. The red arrow indicates the substituted nucleotide. The amino acid in the red line indicates expected substitutions at target sites.