The H2A.Z histone variant integrates Wnt signaling in intestinal epithelial homeostasis

Abstract

The Tip60/p400 chromatin-modifying complex, which is involved in the incorporation and post-translational modification of the H2A.Z histone variant, regulates cell proliferation and important signaling pathways, such as Wnt. Here, we study the involvement of H2A.Z in intestinal epithelial homeostasis, which is dependent on the finely-tuned equilibrium between stem cells renewal and differentiation, under the control of such pathway. We use cell models and inducible knock-out mice to study the impact of H2A.Z depletion on intestinal homeostasis. We show that H2A.Z is essential for the proliferation of human cancer and normal intestinal crypt cells and negatively controls the expression of a subset of differentiation markers, in cultured cells and mice. H2A.Z impairs the recruitment of the intestine-specific transcription factor CDX2 to chromatin, is itself a target of the Wnt pathway and thus, acts as an integrator for Wnt signaling in the control of intestinal epithelial cell fate and homeostasis.

Fig. 1

H2A.Z-dependent regulation of intestinal epithelial cell proliferation. a HIEC cells were transfected using siRNAs targeting H2A.Z mRNA or control. Two different series of siRNAs were used. Cell number was measured and represented relative to 1 for day 1 after transfection and plating. One representative experiment out of two is shown for each serie. b Caco-2/15 cells were transfected using siRNAs control or directed against H2A.Z and the cell number was measured and represented relative to 1 for day 1 after transfection and plating. The mean and standard deviation are shown (n = 3 independent experiments)

Fig. 2

Negative control of differentiation marker expression by H2A.Z. a Caco-2/15 cells were transfected using siRNA targeting H2A.Z or control. 72 h later, total RNA was extracted and subjected to RT-qPCR analysis for the expression of the indicated genes (calculated relative to β2m mRNA level). The mean and standard deviation are shown (n = 5 independent experiments). Statistical analysis was done using the Student’s t-Test (*p < 0.05; **p < 0.02 vs control siRNA). Also shown is the expression of the corresponding proteins analyzed by western blot from total extracted proteins of the same samples. b HIEC2F cells were transfected using siRNA control or targeting H2A.Z, and treated or not with 10 µg/ml doxycycline for 3 days. Total RNA was extracted and RT-qPCR analysis for the expression of the indicated genes was performed as in a

Fig. 3

Regulation of markers accumulation in tissues upon H2a.z depletion. a Immunofluorescence experiments on sections of small intestine from mice expressing (+CRE) or not (−) the CRE recombinase and treated for 10 days with a tamoxifen-containing feed before dissection. The indicated specific antibodies were used, as well as DAPI to stain nuclei (in blue in merge panels), for staining tissues from representative individuals. b Alcian blue staining, counterstained with nuclear fast red, was performed on sections, obtained as in a, to reveal mucin-producing cells. c Same as in a using the indicated antibodies and DAPI staining. All panels are representative of the same animals than in b

Fig. 4

H2a.z controls the expression of differentiation marker mRNAs in vivo. 10–15 mice per indicated genotypes were fed a tamoxifen-containing diet for 10 days, sacrificed and the small intestinal mucosae were harvested and total RNA extracted, as indicated in the Materials and Methods. qPCR was then performed for the detection of enterocyte (a), goblet (b), entero-endocrine (c), and Paneth (d) cell markers. β2m-normalized values for each mouse were plotted relative to 1 for the mean of control (-CRE) mice. Means and standard errors are represented and the Student’s t-Test p-value is indicated for each gene. e Same as in a–d for intestine-specific transcription factors

Fig. 5

Epistatic relationship between H2A.Z and CDX2. a Caco-2/15 cells were transfected using siRNA targeting CDX2 mRNA or control, upon or not siRNA-mediated H2A.Z depletion. 72 h later, total RNA was extracted and subjected to RT-qPCR analysis for the expression of the indicated genes (calculated relative to β2m mRNA level). The mean and standard error are shown (n = 3 independent experiments). Statistical analysis was done using Student’s t-Test (*p < 0.05; **p < 0.02 vs control siRNA). b Caco-2/15 cells were transfected using siRNA targeting H2A.Z or control, as indicated. 72 h later, ChIPs were performed using an H2A.Z antibody or no antibody as a control (No Ab). The amount of LPH or MUCDHL promoters, or an intergenic sequence were analyzed by qPCR, and the percentage of input was calculated. The mean and standard error are shown (n = 3 independent experiments). Statistical analysis was done using unilateral Student’s t-Test. c Same in b for CDX2 ChIP samples

Fig. 6

β-catenin positively regulates H2A.Z and impairs differentiation marker expression. a Caco-2/15 cells were transfected using siRNA targeting β-catenin mRNA or control. 72 h later, total RNA was extracted and subjected to RT-qPCR analysis for expression of the indicated genes (calculated relative to β2 M and RPLP0 (p0) mRNA levels). The mean and standard errors are shown (n = 4 independent experiments). Statistical analysis was done using Student’s t-Test (**p < 0.02 vs control siRNA). b Same as in a for Wnt target genes (**p < 0.02 vs control siRNA). c Same as in a for enterocytes differentiation marker genes (*p < 0.05 vs control siRNA)

Fig. 7

H2A.Z is a target of the Wnt pathway. a Analysis of ChIP-seq data from Frietze et al. for the binding of TCF7L2 on H2AFZ and LGR5 gene promoters in three cells lines (HEPG2, HEK293, and HCT116). Representation of peaks of enrichment were obtained using Integrated Genome Browser 9.0.2 (BioViz, https://bioviz.org). Note that the drawing scale is not identical due to the relative sizes of both genes. b HCT116 cells were transfected using β-catenin –targeting or control siRNA. 72 h later, total RNA was extracted and subjected to RT-qPCR analysis for expression of the indicated genes (calculated relative to β2 M and RPLP0 (p0) mRNA level). The mean and standard errors are shown (n = 3 independent experiments). Statistical analysis was done using Student’s t-Test (*p < 0.05; **p < 0.02 vs control siRNA). c Chromatin from sub-confluent HCT116 and Caco-2/15 cells were harvested and ChIPs were performed using a TCF7L2(TCF4) antibody. The amount of H2AFZ or LGR5 promoters, or an intergenic sequence were analyzed by qPCR, and the percentage of input was calculated. The mean and standard errors are shown (n = 3 technical replicates)

Fig. 8

H2A.Z is an effector of the Wnt pathway on differentiation markers. a Chromatin from sub-confluent CRISPR-Cas9-modified HCT116 cells to homozygously delete the major TCF7L2 binding region of the H2AFZ gene promoter, using indicated H2AFZ promoter—specific or intergenic sequence—targeting primers. Results and standard errors (n = 3 technical replicates) are shown for two independent experiments. b Control or CRISPR-modified HCT116 cells were transfected using β-catenin—targeting or control siRNA. 72 h later, total RNA was extracted and subjected to RT-qPCR analysis for expression of the indicated genes (calculated relative to GAPDH and β-actin mRNA levels). The mean and standard errors are shown (n = 4 independent experiments). The Student’s t-Test p-value was calculated (for β-catenin -depleted samples from control vs CRISPR-modified cells) and is indicated for each gene

Fig. 9

H2A.Z is an effector of the Wnt pathway on differentiation markers. a Caco-2/15 were treated using siRNA targeting β-catenin mRNA or control and transfected with vectors encoding or not H2A.Z to analyse the impact of H2A.Z forced expression on the effects of β-catenin knockdown. The expression of the indicated genes was measured for each condition (calculated relative to β2 M and RPLP0 (p0) mRNA levels). The mean and standard error are shown (n = 6 independent experiments). Statistical analysis was done using Student’s t-Test (*p < 0.05; **p < 0.02 vs β-catenin siRNA + empty vector condition). b Same as in a for Wnt target genes. c Working model: in our study, we showed that the Wnt/β-catenin/TCF7L2(TCF4) signaling pathway positively regulates the expression of the H2A.Z histone variant. The incorporation of the variant, mediated by chromatin modifying complexes, is then increased and responsible for the low recruitment of the intestine-specific transcription factor CDX2 on its target genes and thus, limits the expression of such genes. Since it is known that H2A.Z occupancy of a specific promoter is a signature for intestinal stem cells and that this enrichment is reduced upon differentiation, our data show that H2A.Z participates in the regulation of epithelial differentiation through Wnt signaling. Thus, our study demonstrates that a structural chromatin mark can control the cell fate of normal progenitors and, thereafter, intestinal epithelial homeostasis