Silencing of the lncRNA Zeb2-NAT facilitates reprogramming of aged fibroblasts and safeguards stem cell pluripotency

Abstract

Aging imposes a barrier to somatic cell reprogramming through poorly understood mechanisms. Here, we report that fibroblasts from old mice express higher levels of Zeb2, a transcription factor that activates epithelial-to-mesenchymal transition. Synthesis of Zeb2 protein is controlled by a natural antisense transcript named Zeb2-NAT. We show that transfection of adult fibroblasts with specific LNA Gapmers induces a robust downregulation of Zeb2-NAT transcripts and Zeb2 protein and enhances the reprogramming of old fibroblasts into pluripotent cells. We further demonstrate that Zeb2-NAT expression is precociously activated by differentiation stimuli in embryonic stem (ES) cells. By knocking down Zeb2-NAT, we were able to maintain ES cells challenged with commitment signals in the ground state of pluripotency. In conclusion, our study identifies a long noncoding RNA that is overlapping and antisense to the Zeb2 locus as a target for rejuvenation strategies.

Fig. 1

Inefficient reprograming of old fibroblasts. a Schematics of the reprogrammable mouse line i4F in which the OSKM reprogramming factors (Oct4, Sox2, Klf4, and c-Myc) are expressed from a single genomic locus using a doxycycline-inducible transgene. Mouse embryonic fibroblasts (MEFs) and fibroblasts isolated from adult (10–30-week old) and old (70–100-week old) mice were induced with doxycycline for 2 weeks. b Quantification of AP⁺ clones. c Reprogramming efficiency relative to MEFs. d, e qRT-PCR analysis of Zeb2 (d) and Snail (e) mRNA in MEFs and fibroblasts from adult and old mice that were either not exposed to doxycycline or induced for 2 weeks. For all graphics depicted, Student’s t test (two-tailed) statistics, *p < 0.05, **p < 0.01, and ***p < 0.001; error bars represent standard deviation; n=experiments with independent cell cultures. f Immunofluorescence for Zeb2 in MEFs and old fibroblasts. The corresponding images stained with DAPI are shown (scale bars correspond to 10 μm). g Immunoblot for Zeb2 and β-actin in total cell lysates from two independent early-passage cultures of MEFs and old fibroblasts

Fig. 2

. Zeb2-NAT expression is upregulated in old fibroblasts and correlates with retention of Zeb2 intron 1. a Schematics of the murine Zeb2 and Zeb2-NAT locus. An internal ribosome entry site (IRES) is located in the first intron of the Zeb2 transcript. Double arrows indicate the position of primers used to detect Zeb2-NAT RNA. b qRT-PCR analysis of Zeb2-NAT RNA in MEFs and fibroblasts from adult and old mice. Transcript levels were normalized to GAPDH mRNA and depicted as fold change relative to MEFs. c qRT-PCR analysis of Zeb2-NAT RNA in MEFs and in a iPS cell line derived from i4F fibroblasts. d qRT-PCR analysis of spliced and unspliced Zeb2 RNA in MEFs and old fibroblasts. Double arrows indicate the position of primers. For all graphics depicted, Student’s t test (two-tailed) statistics, *p < 0.05, **p < 0.01; error bars represent standard deviation; n=experiments with independent cell cultures

Fig. 3

Knocking down Zeb2-NAT affects Zeb2 expression. Fibroblasts from adult mice were transfected with either control LNA Gapmers or oligonucleotides targeting Zeb2-NAT and Zeb2. a qRT-PCR analysis of Zeb2-NAT RNA. b qRT-PCR analysis of total Zeb2 transcripts (primer Zeb2 #1, see Supplementary Table 2). c Immunoblot for Zeb2 and β-actin in total cell lysates after transfection of anti-Zeb2-NAT Gapmers #1 and #2 (see Supplementary Fig. 2b). d qRT-PCR analysis of spliced and unspliced Zeb2 RNA. Double arrows indicate the position of primers. For all graphics depicted, Student’s t test (two-tailed) statistics, *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent standard deviation; n=experiments with independent cell cultures

Fig. 4

Knocking down Zeb2-NAT is sufficient to enhance reprogramming of old fibroblasts. Fibroblasts were transfected with the indicated LNA Gapmers and induced to reprogram for 3 weeks. a, b Reprogramming efficiency in adult and old fibroblasts based on quantification of AP⁺ colonies. Student’s t test (two-tailed) statistics, *p < 0.05; error bars represent standard deviation; n=experiments with independent cell cultures. c Proportion of colonies that survived re-plating in medium containing LIF and 2i, without doxycycline; colonies were counted at day 8 after replating. At least ten independent colonies where picked from each condition. Fischer’s exact test was used for statistical analysis. d Immunofluorescence for Nanog and SSEA1 in reprogrammed old fibroblasts transfected with anti-Zeb2-NAT oligonucleotides; the corresponding images stained with DAPI are shown (scale bars, 20 μm). e Histological section of a subcutaneous teratoma containing mesoderm (mes), endoderm (end), and ectoderm (ect) 15 weeks after injection of cells into immune-deficient mice. Magnifications as indicated

Fig. 5

Expression of 2′OMe RNA oligonucleotides mimicking Zeb2-NAT. a Position of 2′OMe RNA oligonucleotides relative to Zeb2 RNA. b, c qRT-PCR analysis of Zeb2 transcripts (primers Zeb2 #1 and Zeb2 intron, see Supplementary Table 2) in fibroblasts from old mice transfected as indicated. d Immunoblot for Zeb2 and β-actin in total cell lysates of old fibroblasts transfected as indicated. e Representative reprogramming experiment stained for Alkaline Phosphatase after 3 weeks in culture with doxycycline. f Reprogramming efficiency of old fibroblasts transfected as indicated. For all graphics depicted, Student’s t test (two-tailed) statistics, *p < 0.05. At least three independent experiments were carried per condition

Fig. 6

Zeb2-NAT expression in ES cells increases in response to differentiation stimuli. a qRT-PCR analysis of Zeb2-NAT and Zeb2 transcripts in MEFs and in the embryonic stem cell line E14. b Immunoblot for Zeb2 and tubulin in total cell lysates from MSFs (mouse skin fibroblasts), E14 cells and iPSCs. c qRT-PCR analysis of the indicated transcripts in E14 cells cultured in medium with LIF/2i or incubated for 20 and 72 h after removal of LIF and 2i. d qRT-PCR analysis of the indicated transcripts in old fibroblasts and TNG-A cells grown in the presence or absence of 2i. For all graphics depicted, Student’s t test (two-tailed) statistics was used, *p < 0.05, **p < 0.01, ***p < 0.001; error bars represent standard deviation; n=experiments with independent cell cultures

Fig. 7

Blocking expression of Zeb2-NAT in ES cells enhances pluripotency and self-renewal. a qRT-PCR analysis of Nanog RNA in TNG-A cells transfected with the indicated oligonucleotides and grown in the presence (+) or absence (−) of 2i for 48 h. Transcript levels were normalized to GAPDH mRNA and depicted as fold change. Student’s t test (two-tailed) statistics, *p < 0.05, ***p < 0.001; error bars represent standard deviation. At least three independent experiments were carried per condition. b Immunoblot for Nanog and β-actin in total cell lysates from TNG-A cells transfected as indicated and grown in the presence (+) or absence (-) of 2i for 48 h. Relative quantification was carried out using Scion Image Software. c qRT-PCR analysis of E-cadherin RNA in TNG-A cells transfected as indicated and grown in the presence (+) or absence (−) of 2i for 48 h. Transcript levels were normalized to GAPDH mRNA and depicted as fold change. Student’s t test (two-tailed) statistics, *p < 0.05; error bars represent standard deviation. At least three independent experiments were carried per condition. d Histological section of a subcutaneous teratoma containing mesoderm (mes), endoderm (end), and ectoderm (ect), at 7 weeks after injection of cells treated with anti-Zeb2-NAT oligonucleotides. e Quantification of tumor volume (mm³). Student’s t test (two-tailed) statistics, *p < 0.05; error bars represent standard deviation; n=teratomas from independent cell cultures. f Representative images of tumors observed at 7 weeks after cell transplantation. Magnifications as indicated