Base Editing of Human Pluripotent Stem Cells for Modeling Long QT Syndrome

Human pluripotent stem cells (hPSCs) have great potential for disease modeling, drug discovery, and regenerative medicine as they can differentiate into many different functional cell types via directed differentiation. However, the application of disease modeling is limited due to a time-consuming and labor-intensive process of introducing known pathogenic mutations into hPSCs. Base editing is a newly developed technology that enables the facile introduction of point mutations into specific loci within the genome of living cells without unwanted genome injured. We describe an optimized stepwise protocol to introduce disease-specific mutations of long QT syndrome (LQTs) into hPSCs. We highlight technical issues, especially those associated with introducing a point mutation to obtain isogenic hPSCs without inserting any resistance cassette and reproducible cardiomyocyte differentiation. Based on the protocol, we succeeded in getting hPSCs carrying LQTs pathogenic mutation with excellent efficiency (31.7% of heterozygous clones, 9.1% of homozygous clones) in less than 20 days. In addition, we also provide protocols to analyze electrophysiological of hPSC-derived cardiomyocytes using multi-electrode arrays. This protocol is also applicable to introduce other disease-specific mutations into hPSCs. Graphical abstract


Introduction
Human pluripotent stem cells (hPSCs), including human embryonic stem cells (hESCs) and the closely related human induced pluripotent stem cells (iPSCs), are characterized by self-renewal and can differentiate into a huge number of different functional cell types via directed differentiation [1]. The ability to proliferate indefinitely allows large number of differentiated derived cells to be obtained in a short period. It plays a vital role in regenerative medicine. The ability to directionally differentiate into somatic cells allows stem cells to play an essential role in disease models [2], drug screening [3], cell development [4], and cell fate choice [5]. Patient tissue-derived iPSCs [6], which are then differentiated into cardiac [7], neural [8], endothelial [9], and other cells, are widely used in disease modeling. Using patient-derived iPSCs, we can study many genetic diseases, such as long QT syndrome [10], Brugada syndrome [11], hypertrophic cardiomyopathy [12], etc. However, some diseases are not genetically inherited [13], and such diseases-derived iPSCs are likely to be non-phenotype observed. For example, some diseases in which methylation is involved in regulation may be lost during reprogramming [14]. In addition, there is a lack of ideal control when compared with patient-derived iPSCs. Researchers usually select healthy people of the same family [15] or unrelated healthy people [16] as control. This may considerably reduce the reliability of the studies, as the genetic backgrounds of these individual are different. Moreover, the reprogramming process from tissue cells to iPSCs is time-consuming [17]. Using gene-editing technology to introduce disease mutations in hPSCs and using unedited hPSCs as control can be a perfect solution to overcome these limitations.
Base editing technique is an evolution of the Clustered Regularly Interspaced Short Palindromic Repeat (CRISPR) system, introducing point mutations without requiring DNA double-strand breaks or donor templates [18]. Under the guidance of sgRNA, the catalytically impaired Cas9 protein fused with a single-stranded DNA deaminase enzyme is guided to the target sequence and then substituted the bases [19]. Because it consists of inactivated Cas9, which undergoes point substitution without producing double-stranded DNA breaks, the non-specific activity is greatly reduced and is considered to be the safest. Three main classes of base editors have been developed to date: cytosine base editors (CBEs), which catalyze the conversion of C•G base pairs to T•A base pairs [20]; and adenine base editors (ABEs), which catalyze A•T-to-G•C conversions [21]; Glycosylase base editors (GBEs), which catalyze the conversion of C•G base pairs to A•G (in bacteria) [22] and catalyze the conversion of C•G base pairs to G•A (in mammalian cells) [23]. These techniques could theoretically be used to correct or introduce human pathogenic SNPs [24]. Conventional base editing techniques require two vectors. One expresses catalytically impaired Cas9 protein fused with a single-stranded DNA deaminase enzyme, the other expressing sgRNA. Gene editing is only possible happened if two plasmids enter a cell at the same time. However, although the efficiency of gene editing is high, the low efficiency of hPSCs transfection is a huge obstacle [25]. Therefore, the introduction of disease mutations in hPSCs is a very time-consuming and challenging work, but it is easier and shorter than reprogramming.
In our previous work [26], inactivated Cas9, a single guide RNA (sgRNA) with an adenine base editor (ABE) or a cytosine base editor (CBE), were all co-expressed in one episomal vector (refer to epi-BE). The episomal vector can replicate during cellular division in eukaryotes permitting the continuous expression of Cas9, base editor, and sgRNAs in hPSCs. epi-BE also contains a drug resistance gene. Despite the low efficiency of plasmid delivery, we were able to greatly enrich the target cells through longterm drug screening and the proliferation of drug-resistant cells. We introduced mutations in three pathogenic genes of LQT, KCNQ1, KCNH2, and SCN5A, and screened a total of 328 clones, of which 104 were heterozygous (31.7%) and 30 were homozygous (9.1%) (Note 12). In this paper, we will show how to introduce LQT disease mutation loci into hPSCs step by step. To model LQT syndrome, the diseased hPSCs are differentiated into cardiomyocytes for phenotypic and functional characterization Figs. 1, 2, 3 and 4.  Place the above samples in a 95 °C water bath, switch off the power, and cool naturally to room temperature. Alternatives: You can anneal the oligonucleotides using a thermocycler instead of a hot water bath. Incubate the reaction solution at 95 °C for 5 min and then slowly cool it down to room temperature (20-30 °C) using a thermocycler-the temperature decreases by 1 °C per 10s.

Design of the Plasmids for Base Editing and Functional Analysis
3. Dilute the annealed oligonucleotides 20 folds with ddH2O. Clone the annealed oligonucleotide into the sgRNA expression plasmid as indicated below.

Single Cell-Derived Clone Screen
11. The antibiotic-iPSCs were passaged with EDTA. Then, 1 × 10 5 cells were seeded on a Matrigel pre-coated 10 cm dish using mTeSR-1 cell culture medium with 5 μM of Y-27632. (See note 7) 12. Twenty-four hours later, the mTeSR-1 media was replaced by new media without Y-27632. This media was changed every two days. 13. Ten days after seeding, the single cell-derived clones were picked up using a 1 ml sterile syringe and divided into two halves.  15. For each target site, five potential off-targets were selected based on Cas-Offinder and PCR-amplified for Sanger sequencing.

Cardiomyocyte Differentiation
16. Cells (∼90% confluency) were seeded on a Matrigel pre-coated 6-well plate at a ratio of 1:6 in mTeSR-1 media. (See note 10) 17. The media was changed to CDM3 supplemented with 6 μM of CHIR99021 when the cells reached ∼75% confluency. 18. After 48 h, the media was changed to CDM3 [7] supplemented with 2 μM of Wnt-C59. After 2 days, the media was changed to CDM3 and refreshed every 2 days. After differentiation for 7-8 days, spontaneous contracting cells could be observed. 19. On day 12, Cardiomyocytes (CMs) were purified using a metabolic-selection method. The medium consisted of RPMI 1640 without glucose, 213 μg/ml of L-ascorbic acid 2-phosphate, 500 μg/ml of Oryza sativa-derived recombinant human albumin, and 5 mM of sodium DL-lactate. 20. After purification, CMs were cultured with RPMI 1640 and B27 (with insulin). For cellular maintenance, the medium was changed every 3 days.
LQT Phenotype Identification Using MEA 21. CMs were digested with Accutase. (See note 11) 22. CytoView MEA24 plates (Axion Biosystems, Inc., Atlanta, United States) were pre-coated overnight using a 0.5% Matrigel phosphate-buffered saline (PBS) solution. 23. 15,000 CMs were plated on each multi-electrode array (MEA) well with RPMI/B27 medium and cultured for three days. 24. When the cellular electrophysiological activity became stable, the experimental data were recorded using Maestro EDGE (Axion Biosystems, Inc., Atlanta, United States) according to the MEA manual. The data were analyzed using the AxIS Navigator, Cardiac Analysis Tool, and IGOR software.

Notes
1. The potential therapeutic targets were further screened according to the ABE/CBE sgRNA design rules, which must be in the form of 20 nt + PAM. The mutation site is in the sgRNA edit window 2-8. Therapeutic targets eligible for gene editing were obtained. To avoid bystander editing during gene editing, we recommend only one of the mutation loci in the edit window 2-8 of sgRNA. Other databases like ClinVar (https:// www. ncbi. nlm. nih. gov/ clinv ar/)are also recommended. 2. ttt and aac are the sequences that matche the sticky end produced by the enzymatic cleavage of the plasmid used in this paper. You should add the appropriate sequence to the sticky end sequence produced by the plasmid you are using. 3. The sgRNA expression plasmids in this protocol contain epi-ABEmax, epi-AncBE4max, which is an allin-one episomal vector expressing a single guide RNA (sgRNA) with an adenine base editor (ABE) or a cytosine base editor (CBE). If you are using two plasmids for sgRNA and base editor expressing, you only need to ligate oligonucleotides to the sgRNA expression vector. 4. You should choose the appropriate antibiotic according to the resistance expressed by your plasmid. 5. You should choose the appropriate sequencing primer according to your plasmid. 6. We recommend the use of the LONZA 4D Nuclear Transfection System to transfer plasmids into cells, programmed as CA137. 7. EDTA is less toxic to cells. To obtain single cells, we try to increase the contact time between EDTA and the cells until the cells are observed as single cells under the microscope. Typically contact time is10-20 min. 8. The cell clone must not be too small, or the clone will fail to pick up. Generally, under a 10x microscope, the cell clone should fulfill the entire field. The microscope should be transferred to a biosafety cabinet in advance and UV irradiated for at least half an hour. 9. PCR conditions should be referred to the instructions for the polymerase used. We should determine primer Tm values in advance. The amount of DNA extracted by our method is minimal, and therefore the number of cycles of PCR should be increased, with a recommended setting of 38 cycles. 10. Matrigel was diluted using pre-cooled PBS solution at a ratio of 1:500. The process of CM differentiation is susceptible to mycoplasma, and we strongly recommend testing the mycoplasma before CM differentiation. The cell culture medium supernatant is aspirated and used as a template for PCR amplification using mycoplasma-specific primers and agarose gels for validation. Mycoplasma-specific primer [ Code Availability Not applicable. Data Availability Not applicable.

Conflict of Interest
The authors indicate no potential conflicts of interest.
Ethics Approval Not applicable.

Consent for Publication Not applicable.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http:// creat iveco mmons. org/ licen ses/ by/4. 0/.