A TALEN genome-editing system for generating human stem cell-based disease models

Abstract

Transcription activator-like effector nucleases (TALENs) are a new class of engineered nucleases that are easier to design to cleave at desired sites in a genome than previous types of nucleases. We report here the use of TALENs to rapidly and efficiently generate mutant alleles of 15 genes in cultured somatic cells or human pluripotent stem cells, the latter for which we differentiated both the targeted lines and isogenic control lines into various metabolic cell types. We demonstrate cell-autonomous phenotypes directly linked to disease-dyslipidemia, insulin resistance, hypoglycemia, lipodystrophy, motor-neuron death, and hepatitis C infection. We found little evidence of TALEN off-target effects, but each clonal line nevertheless harbors a significant number of unique mutations. Given the speed and ease with which we were able to derive and characterize these cell lines, we anticipate TALEN-mediated genome editing of human cells becoming a mainstay for the investigation of human biology and disease.

Figure 1. Schematic of System for Efficient and Rapid Genome Editing with TALENs

GFP = green fluorescent protein; RFP = red fluorescent protein. See also Figure S1.

Figure 2. APOB is Important for HCV Replication

(A) Generation of APOB knockout HuH-7 clones with TALENs targeting exon 13. Boxes indicate the TALEN binding sites. Deletions, insertions, and duplications in the two alleles of each clone are indicated. The 26-bp insertion and 8-bp duplication (asterisk) in clone A are 5’-GAGTCGCTTCTCCGGGAGATAAGTCA-3’ and 5’-GACTGGCT-3’, respectively. (B) Left panel, Western blot using whole cell lysates from two wild-type and four knockout HuH-7 cell lines (clones A–D). Right panel, Western blot from a wild-type clone and a knockout clone (clone A) infected with or without JFH-1 virus and incubated with or without LDL particles. The same wild-type clone and knockout clone (clone A) were used for all subsequent experiments. (C) Left panel, apoB ELISAs performed on conditioned media from cells; values are normalized to level from wild-type clone. Right panel, APOB mRNA expression by qRT-PCR from whole cell lysates; expression is indicated as fold change of 2^−ΔΔCt with reference to 18S rRNA, normalized to level in wild-type clone. (D) Immunocytochemistry for apoB. (E) HCV RNA levels by qRT-PCR from clones infected with JFH-1 virus; expression is indicated as fold change of 2^−ΔΔCt with reference to GAPDH, normalized to level in wild-type clone. Error bars show s.e.m. from experiments with biological replicates, N = 3. P values calculated with unpaired t test. See also Figure S2.

Figure 3. SORT1 Reduces Hepatocyte Secreted ApoB Mass, is Important for Insulin-Responsive Glucose Transport in Adipocytes, and Mediates proBDNF-Induced Motor Neuron Death

(A) Generation of SORT1 knockout hPSC clones with TALENs targeting exon 2 or exon 3. Boxes indicate the TALEN binding sites. Deletions and insertions in the two alleles of each clone are indicated. The 17-bp insertion (asterisk) in clone B is 5’-TGCTATCTCCAACCAGG-3’. (B) Western blot for sortilin and qRT-PCR for SORT1 mRNA in wild-type and knockout HUES 1 clones (clones A–C); mRNA expression is indicated as fold change of 2^−ΔΔCt with reference to 18S rRNA, normalized to mean level in wild-type clones. (C) Albumin and apoB mass measured by ELISA in media collected from wild-type and knockout HLCs (two clones each; A and B for HUES 1, D and E for HUES 9), normalized to mean levels in wild-type HLCs. N = 3 for HUES 1, N = 6 for HUES 9. (D) Western blots of lysates and ELISAs in media from wild-type and knockout HLCs (one clone each, A for HUES 1; two clones each, D and E for HUES 9) infected with SORT1- or GFP-expressing lentivirus, normalized to mean levels in wild-type HLCs. N = 2 for HUES 1, N = 6 for HUES 9. (E) Ratios of glucose uptake to total protein content in wild-type and knockout HUES 1 adipocytes (one clone each; clone A) infected with SORT1-expressing or control lentivirus and treated with or without insulin, all normalized to mean ratio in wild-type adipocytes without insulin. N = 6. (F) Immunocytochemistry for TUJ1 and ISL-1 in wild-type and knockout HUES 9 motor neurons. Arrows indicate representative double-positive cells. (G) Counts of wild-type and knockout HUES 9 (2 clones each, D and E) motor neurons (TUJ1/ISL-1 double-positive cells) treated with BDNF vs. proBDNF. N = 12. Error bars show s.e.m. from experiments with biological replicates. P values calculated with unpaired t test. See also Figure S3, Figure S4, and Figure S5.

Figure 4. AKT2 E17K is a Dominant, Activating Mutation

(A) Generation of AKT2 knockout and E17K knock-in hPSC clones with TALENs targeting exon 2. The underline indicates the codon encoding E17K change. Boxes in wild-type sequence indicate the TALEN binding sites. The box in the ssODN sequence indicates the RsaI restriction site created by the synonymous mutation in blue; the nucleotide in red indicates the missense mutation for E17K. Deletions and insertions in the two alleles of each clone are indicated. The 15-bp insertion (asterisk) in clone A is 5’-GACCTCCAGGTCCTG-3’. (B) Western blot for AKT2 and AKT1 in wild-type, knockout, and E17K HUES 9 HLCs (A and B for knockout; C and D for E17K). (C) Immunocytochemistry for FoxO1 in HLCs (clones A and C) at baseline and after 15 min insulin stimulation. (D) Ratios of glucose production to secreted albumin mass (ELISA) in wild-type, knockout, and E17K HUES 9 HLCs (two clones each; A and B for knockout; C and D for E17K) treated with or without dexamethasone and forskolin with or without insulin, all normalized to mean ratios in wild-type HLCs without additives. Left panel, N = 4; right panel, N = 2. (E) Ratios of triglyceride content to total protein content in wild-type, knockout, and E17K HUES 9 adipocytes (two clones each; A and B for knockout; C and D for E17K), normalized to mean ratio in wild-type adipocytes. N = 5. (F) Ratios of glucose uptake to total protein content in wild-type, knockout, and E17K HUES 9 adipocytes (two clones each; A and B for knockout; C and D for E17K) treated with or without insulin, all normalized to mean ratio in wild-type adipocytes without insulin. N = 6. (G) IL-8 and adiponectin mass measured by ELISA in media collected from wild-type, knockout, and E17K HUES 9 adipocytes (two clones each; A and B for knockout; C and D for E17K), normalized to mean levels from wild-type adipocytes. N = 4. Error bars show s.e.m. from experiments with biological replicates. P values calculated with unpaired t test. See also Figure S3 and Figure S6.

Figure 5. PLIN1 Frameshift Mutations Produce Dominant-Acting and Truncated Proteins

(A) Generation of PLIN1 frameshift mutant hPSC clones with TALENs targeting exon 8. Boxes indicate the TALEN binding sites. Deletions in the one allele of each clone are indicated. (B) Immunocytochemistry for perilipin and BODIPY in wild-type, PLIN1⁵⁵⁸ (clone A), and PLIN1⁴¹⁵ (clone B) adipocytes. (C) Western blot for N-terminus of perilipin in wild-type, PLIN1⁵⁵⁸, and PLIN1⁴¹⁵ adipocytes; arrow indicates truncated protein. (D) Ratios of triglyceride content to total protein content in wild-type, PLIN1⁵⁵⁸, and PLIN1⁴¹⁵ adipocytes. N = 3. (E) Ratios of lipolysis activity (as measured by glycerol release) to total triglyceride content in wild-type, PLIN1⁵⁵⁸, and PLIN1⁴¹⁵ adipocytes treated with or without 10 µM forskolin. N = 4. Error bars show s.e.m. from experiments with biological replicates. P values calculated with unpaired t test. See also Figure S7.