ΔNP63 Defines an Exocrine-committed Subset of Murine Pancreatic Progenitor Cells

Graphical abstract

The transcription factor ΔNp63, one of the two major isoforms of Trp63, has a key role in embryonic development, regeneration, and tumorigenesis across various epithelial tissues, including the skin.¹ Despite its role as a cellular plasticity regulator in pancreatic cancer² and its expression in rare basal cells in the adult human pancreas,³ ΔNp63 remains unstudied in pancreatic development.

Re-analysis of publicly available single-cell RNA sequencing datasets from the embryonic mouse pancreas^4,5 showed that Trp63 is expressed in the pancreas at E10.5 (Supplementary Figure 1A and B). The proportion of Trp63-expressing cells was highest in E12.5 pancreatic progenitors (PPs), where expression was confined to early epithelial tip (E-Tip) cells and by E14.5, Trp63 expression became restricted to the acinar lineage (Figure 1A and B), and subsequently diminished from E17.5 into adulthood (E17.5-adult, Supplementary Figure 1C–F). Differential gene expression analysis (Supplementary Figure 1G–H) revealed that genes typical of future acinar cells were upregulated in E12.5 Trp63⁺ tip cells (Figure 1A, right panel), accordingly enriching for Gene Ontology (GO) terms “protein digestion and absorption,” “pancreas secretion,” and “fat digestion and absorption.” In contrast, typical trunk cell genes (indicated by asterisks) were enriched in E12.5 Trp63^- tip cells. Besides Trp63, other basal genes such as Krt5 and Krt14 were barely detected at the developmental stages examined (Supplementary Figure 2A–C).

Figure 1.

(A) Dot plot displayingTrp63expression and differentially expressed genes betweenTrp63-positive (Trp63⁺) andTrp63-negative (Trp63^-) tip cells at E12.5 (n = 2 mice). Dot size represents the proportion of expressing cells; color indicates expression level. ∗Indicating trunk markers. Adjusted P value ≤ .05; Wilcoxon rank sum test. (B) Dot plot of Trp63 at E14.5 (n = 2 mice). (C and D) Immunofluorescence for ΔNP63 (purple) and DNA (white) and (C) CPA1 (blue), SOX9 (green) or PDX1 (green) at E12.5 and (D) CPA1 (blue) at E14.5. n = 3 mice (n = 5 for CPA1 at E12.5); 3 sections per mouse were analyzed. (E) qRT-PCR for ΔNp63 and TAp63 at E10.5, E12.5, E14.5, and E17.5 (n = 3 mice/time point). ∗P ≤ .05; ∗∗P ≤ .01; ns, non-significant. Paired two-tailed t-test was performed in between the two conditions. (F) Immunofluorescence staining for mGFP (green), DNA (white) and ΔNP63 (purple), CPA1 (red), PDX1 (blue) or MIST1 (blue) at E14.5 (n = 4; 3 images/mouse). (G) Percentage of mGFP⁺ cells in acinar (red bar) vs CA-TD (blue bar) compartments at E18.5 (n = 6 mice). Results are represented as mean ± SEM. (H) Immunofluorescence for mGFP (green), DNA (white) and AMY (purple), mTomato (red) or KRT7 (blue) at E18.5 (n = 6 mice; 3 sections/mouse). Arrowheads indicate double-positive cells. CA-TD, centroacinar-terminal ductal cells.

Immunostainings confirmed ΔNP63 expression at E10.5 in a subset of PPs (confirmed by staining for PDX1, SOX9 and HNF1B; Supplementary Figure 3A) and in early glucagon⁺ cells (not shown). At E12.5, ΔNP63 staining was confined to a subset (10.2% ± 0.3%) of proliferating tip cells, that express CPA1, SOX9, and PDX1 (Figure 1C), as well as MIST1 and Ki67 (Supplementary Figure 3B). Acinar cell commitment was confirmed at E14.5 co-staining with CPA1 (Figure 1D) and PTF1a (not shown). RNAscope confirmed the ΔN-isoform-specific expression at E12.5 (Supplementary Figure 3C), whereas quantitative polymerase chain reaction (qPCR) demonstrated negligible expression of the TAp63 isoform at all examined timepoints (Figure 1E).

We used Trp63-CreERT2; R26-mTmG mice to map the fate of early Trp63⁺ pancreatic cells. Following tamoxifen (4-OHT) administration at E12.5 (earlier labeling was unsuccessful), the recombined progeny at E14.5 co-expressed membrane-GFP (mGFP) (Supplementary Figure 3D), PP markers PDX1 and CPA1, and the acinar-lineage factor MIST1 (Figure 1F). By E18.5, most mGFP⁺ descendants had adopted an AMY⁺ acinar cell identity, with some contribution to centroacinar (CA) cells, and KRT7⁺ terminal ductal (TD) cells⁶ (Figure 1G and H).

We then studied a ΔNp63GFP/+ knock-in mouse model, with loss of ΔNp63 and GFP replacement confirmed by stainings (Supplementary Figure 4A and B).⁷ At E14.5 to E16.5, some pancreatic lobes in the homozygous knockout (KO) mice appeared less dense (Figure 2A and B), and there was a reduction in pancreatic weight relative to heart weight, which is unaffected by the loss of ΔNp63 (Figure 2C; Supplementary Figure 4C). Additionally, the acini exhibited a smaller diameter, indicative of immaturity⁸ (Figure 2D). Consistent with these findings, we observed decreased expression of acinar markers, but not endocrine or ductal markers (Figure 2E). Notably, as early as E12.5, bulk RNA sequencing revealed reduced RNA expression of pro-acinar transcription factors Mecom (fold change [FC] = −2.12) and Gata4 (FC = −1.59), as well as lower levels of Ctrb1 (FC = −7.1), Cela1 (FC = −1.59) (Figure 2F), together with downregulation of genes associated with the ‘Exocrine Pancreas Development’ GO term (NES score = −0.56). We also confirmed a reduced number of CPA1⁺-staining tip cells (Figure 2G) but no significant differences in proliferation or apoptosis (Supplementary Figure 4D−F), suggesting mainly maturation defects.

Figure 2.

(A) Three-dimensional rendering³of gastrointestinal tract of ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E14.5.Rectangles highlight affected regions (splenic and ventral lobes). (B) Immunofluorescence for DNA (white) in ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E16.5 (n = 3 mice; 3 sections/mouse). (C) Quantification of pancreatic weight (mg) in ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E16.5 (n = 4 mice). ∗∗∗P ≤ .001; unpaired two-tailed t-test was performed. (D) Quantification of acinar diameter (μm) in ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E16.5 (n = 3 mice). ∗∗∗∗P ≤ .0001; two-way ANOVA for repeated measures was performed. (E) qRT-PCR of exocrine-specific genes in ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E16.5 (n = 3 mice). ∗P ≤ .05 and ∗∗P ≤ .01; Unpaired two-tailed t-test was performed. (F) Volcano plot highlighting downregulated genes of interest from differential gene expression analysis between ΔNp63 KO (ΔNp63^-/-) and WT (ΔNp63^+/+) mice at E12.5 (n = 3 mice). Adjusted P value ≤ .05; Wald test with Benjamini-Hochberg correction was performed. (G) Immunofluorescence for CPA1 (yellow), PDX1 (blue), and DNA (white) in ΔNp63 WT (ΔNp63^+/+) and KO (ΔNp63^-/-) mice at E12.5 (n = 3 mice; 3 sections/mouse).

In conclusion, ΔNp63 is a previously unrecognized marker of exocrine-committed PP cells and contributes to proper acinar cell maturation, warranting further studies in pancreatic development and disease.

Supplementary Material

Supplementary Figure 1.

Supplementary Figure 2.

Supplementary Figure 3.

Supplementary Figure 4.