Fig 2: PPM1A is critical for proliferation of intestinal organoids.(A, B) Double depletion of PPM1A and PPM1B resulted in an enhanced level of phospho-YAP (S127), when compared with depletion of PPM1A alone (A). Depletion of PPM1B by a CRISPR-based strategy in PPM1A KO HEK293 cells led to a complete loss of colony formation (B). (C, D) Intestinal crypts were isolated from WT and PPM1A KO mice at 6-week age and cultured in vitro (C). A dramatic lower number of de novo crypts (D, top panel) and the decreased size of organoids (D, bottom panel) were seen in organoids from PPM1A KO mice, with roughly 50 intestine organoids each group examined. (E) Edu integration, which mostly localized in the ends of crypts and represented the proliferating cells, was significantly lower in intestinal organoids without PPM1A. (F) Anti-YAP/TAZ immunofluorescence imaging of organoid sections detected a substantial subset of cells containing nucleo-YAP/TAZ in the WT crypts and with active proliferation (Edu staining), in a sharp distinction of Ppm1a-/- crypts, where YAP/TAZ were located in the cytoplasm even in some proliferating cells. (G) The mRNA levels of YAP/TAZ target genes were significantly lower in Ppm1a-/- intestinal organoids, examined at Day 3. (H) Lentiviral delivery into Ppm1a-/- intestinal organoids of the hYAP 2SA mutant (S109A/S127A), which was resistant to the PPM1A-mediated dephosphorylation, restored the mRNA levels of CTGF and CYR61. (I, J) The decreased number of de novo crypts (I) and size of organoids (J) in PPM1A KO intestinal organoids were failed to be recovered, when TGF-ß signaling was inhibited by small molecule inhibitor SB431542 or when IFN-I signaling was blocked by the neutralizing antibody targeting IFNAR1, examined at Day 6, with roughly 20 intestine organoids each group. (K) The mRNA levels of YAP/TAZ target genes in Ppm1a-/- intestinal organoids were unable to be restored, under inhibition of TGF-ß signaling or IFN-1 signaling, examined at Day 3. Unprocessed images of blots are shown in S1 Raw Images. Statistics source data are provided in S1 Data. IFN-I, type I interferon; KO, knockout; phospho-YAP, phosphorylating forms of YAP; PPM1A, protein phosphatase magnesium-dependent 1A; TAZ, transcriptional coactivator with PDZ-binding motif; TGF-ß, transforming growth factor beta; WT, wild-type; YAP, Yes-associated protein.

Fig 3: PPM1A facilitates both nuclear distribution and transcription potency of YAP/TAZ.(A) Transcription potency of YAP (5 ng), which drove the TEAD-responsive promoter, was substantially attenuated by coexpression of MST1 (10 ng). Using this setting, a cDNA library comprised of 40 members of serine/threonine phosphatase was screened for their individual contribution in regulating the Hippo-YAP pathway. The phosphatome screen identified PPM1A/PP2Ca as a potent suppressor of Hippo-YAP signaling, which completely restored the MST1-induced suppression of the YAP-TEAD transactivation. (B–D) Transcription potency of YAP (5 ng), which was suppressed by coexpression of MST1 (50 ng) (B), LATS1 (500 ng) (C), or MAP4K1 (50 ng) (D), was markedly recovered by cotransfection of wild-type PPM1A but not the enzyme-dead PPM1A (D239N). (E) Phosphorylation of YAP at the S127 residue, which was revealed by immunoblotting and represented an activation status of Hippo signaling, was profoundly increased under 2-DG-triggered energy stress. Dox-induced PPM1A expression significantly attenuated this energy stress-triggered Hippo signaling. (F) The LATS1-trigged phospho-YAP (S127) was eliminated in a dose-dependent manner by coexpression of WT PPM1A, but not its phosphatase-dead form (D239N). (G) Coexpression of WT PPM1A stabilized TAZ in a dose-dependent manner, which otherwise was rapidly degraded upon activation of Hippo signaling. (H) The mRNA levels of CTGF and CYR61 were significantly up-regulated in the presence of PPM1A induction. (I) Serum starvation resulted in a substantial increase of YAP in the cytoplasm, which was significantly prevented by the induction of PPM1A (left panels). Percentage of cells with the relative ratio of nucleo-YAP vs cytoplasmic-YAP was counted (right panels). (J) Inducible PPM1A expression decreased the cytoplasmic-YAP but promoted nuclear accumulation of YAP, as analyzed by the nucleocytoplasmic fraction assay. Figs 1–7, unless otherwise indicated, n = 3 independent experiments (mean ± SEM). *, P < 0.05, **, P < 0.01, and ***, P < 0.001, compared with control condition (ANOVA test and Bonferroni correction). Unprocessed images of blots are shown in S1 Raw Images. Statistics source data are provided in S1 Data. Dox, doxycycline; LATS1, large tumor suppressor kinase 1; MST1, mammalian sterile 20-like kinase 1; PPM1A, protein phosphatase magnesium-dependent 1A; TAZ, transcriptional coactivator with PDZ-binding motif; TEAD, transcriptional enhanced associate domain; WT, wild-type; YAP, Yes-associated protein.

Fig 4: PPM1A deficiency impedes the liver regeneration.(A) H&E staining of liver sections from the 10-day postnatal mice showed a reduced level of hepatocyte proliferation (Ki67 positive) but an elevated level of immune cell infiltration and vacuoles in PPM1A KO mice. (B) Wild-type and PPM1A KO mice at 8–10 weeks age were carried out with hepatectomy surgery. The ratio of liver/body weight was measured at day 4 post hepatectomy, which exhibited a marked decrease of compensatory liver regeneration in Ppm1a-/- mice. n = 6 mice in each group. (C) A compromised level of hepatocyte proliferation in compensatory liver regeneration of Ppm1a-/- mice was detected by Ki67 staining examined at day 3 and day 4 post hepatectomy. (D) Immunofluorescence imaging of liver sections revealed a dramatic decrease of the nucleo-YAP in livers from Ppm1a-/- mice examined at day 3 post hepatectomy, exhibiting by the fewer YAP proteins in the nucleus (darker imaging in the nucleus, left panel) and the substantially lesser cells with the nucleo-YAP localization (circled, and right panel). (E) Administration of SB431542 or anti-IFNAR1 failed to restore the suppressed hepatocyte proliferation in compensatory liver regeneration in Ppm1a-/- mice, examined at Day 2. (F, G) Administration of MST inhibitor XMU-MP-1 decreased the level of phospho-YAP (S112) (F) and increased the proliferating hepatocyte (G) in Ppm1a-/- livers. (H) A schematic model for the PPM1A-regulated Hippo-YAP signaling. PPM1A associates with YAP/TAZ both in the cytoplasm and in the nucleus and directly eliminates the key phosphorylation modifications on YAP, which determines the nucleocytoplasmic localization and transcription potency of YAP/TAZ. Accordingly, PPM1A functions as a critically physiological regulator of the Hippo-YAP pathway in mammals, including the intestinal and liver regeneration upon injuries. Unprocessed images of blots are shown in S1 Raw Images. Statistics source data are provided in S1 Data. H&E, hematoxylin and eosin; KO, knockout; MST, mammalian sterile 20-like kinase; phospho-YAP, phosphorylating forms of YAP; PPM1A, protein phosphatase magnesium-dependent 1A; TAZ, transcriptional coactivator with PDZ-binding motif; YAP, Yes-associated protein.

Fig 5: A graphic conclusion.NDRG2-PPM1A interaction in astrocytes after SAH promoted MMP-9 expression by enhancing Smad2/3 phosphorylation in the nucleus, thereby aggravating BBB disruption (left), and the targeted peptide TAT-QFNP12 inhibited MMP-9 production and alleviated BBB injury after SAH by blocking NDRG2-PPM1A binding (right).