The Hippo terminal effector YAP boosts enterovirus replication in type 1 diabetes - PubMed (original) (raw)
doi: 10.1038/s41467-025-64508-6.
Huan Liu # 1, Heena Pahwa # 1, Murali Krishna Madduri # 1, Farah Atawneh 1, Adib Miraki Feriz 2, Sahar Rafizadeh 1, Annabelle Elisabeth Kruf 1, Mona Khazaei 1, Pouria Bahrami 1, David Gotti 1, Mohamed Elawour 1, Ruth M Elgamal 3, Ausilia Maria Grasso 1, David Bund 1, Blaz Lupse 1, Zahra Azizi 1 4, Omar Zabad 1, Karim Bouzakri 5, Marc Horwitz 6, Alberto Pugliese 7 8 9, Kathrin Maedler 10, Amin Ardestani 11 12
Affiliations
- PMID: 41053097
- PMCID: PMC12500894
- DOI: 10.1038/s41467-025-64508-6
The Hippo terminal effector YAP boosts enterovirus replication in type 1 diabetes
Shirin Geravandi et al. Nat Commun. 2025.
Abstract
Type 1 diabetes (T1D) risk has been associated with enteroviral infections, particularly coxsackieviruses B (CVB). Cellular host factors contributing to virus-induced islet autoimmunity remain unclear. We show that the Hippo pathway effector Yes-associated Protein (YAP) is markedly upregulated in the exocrine and endocrine pancreas of T1D and at-risk autoantibody-positive (AAb+) donors, along with its target CTGF. YAP expression correlates with CVB RNA presence, often in or near infected cells. YAP overexpression enhances CVB replication, islet inflammation, and β-cell apoptosis, whereas its inhibition halts viral replication in primary and immortalized pancreatic cells. In exocrine-islet co-cultures, CVB triggers YAP and target gene expression. In mice, chronic β-cell YAP expression impairs glucose tolerance, abolishes insulin secretion, and promotes β-cell dedifferentiation. Mechanistically, YAP, in complex with its transcription factor TEAD, induces its own negative regulator MST1. MST1 inhibition boosts viral replication and reduces β-cell apoptosis, constituting a negative feedback loop in which the reciprocal antagonism between YAP and MST1 balances viral replication and β-cell death during CVB infections. YAP is thus an important host factor for enteroviral amplification, offering a potential antiviral target in T1D.
© 2025. The Author(s).
Conflict of interest statement
Competing interests: The authors declare no competing interests.
Figures
Fig. 1. YAP is highly upregulated in the pancreas of T1D and AAb+ organ donors.
YAP protein and Yap1 mRNA labeling were analyzed in FFPE sections of pancreases from 13 control, 15 AAb+ organ donors without diabetes and 15 donors with T1D from the nPOD pancreas collection. A, B Representative images from different donors (A) and quantification (B) of the percentage of YAP+ area in the exocrine pancreas from FFPE sections of control donors without diabetes (n = 229 independent positions from 13 donors), donors without diabetes but expressing T1D-associated autoantibodies (AAb+) (n = 223 independent positions from 15 donors), and donors with T1D (n = 284 independent positions from 15 donors). C, D Representative images (C) and quantification (D) of the percentage of YAP+ cells within islets of controls (n = 10), AAb+ (n = 10), and donors with T1D (n = 15) of the number of islet cells. E, F Representative images (E) and quantification (F) of YAP (brown), and late endocrine marker chromogranin (green) double-positive cells from controls (n = 3; 16671 islet cells), AAb+ donors (n = 3; 14237 islet cells), and donors with T1D (n = 6; 15116 islet cells). G, H Representative images (G) and quantification (H) of Yap1 mRNA (pink) by RNAscope in situ hybridization of controls (n = 30 independent positions from 3 donors), AAb+ donors (n = 30 independent positions from 3 donors), and donors with T1D (n = 33 independent positions from 3 donors). I Association of YAP protein expression between endocrine islets and exocrine pancreas in AAb+ (n = 10; grey circles) and in donors with T1D (n = 15; black circles). All box plots showing single analytes and median (box and whiskers; min to max show all points). A, C, G Sections were counterstained with Hematoxylin. Data are expressed as means ± SEM. _P-_values were calculated by one-way ANOVA with Holm-Sidak multiple comparisons correction for (B, D, F) and by two-tailed unpaired Student _t-_test (Spearman) for (I). Scale bars depict 50 µm (A, C, G-upper panel) and 10 µm (E, G-lower panel). Source data are provided as a Source Data file.
Fig. 2. Re-analysis of scRNA-seq data from pancreas of healthy and type 1 diabetic donors.
A UMAP visualization of cells colored by diabetic status (ND = Non-Diabetic, T1D = Type 1 Diabetes). B UMAP showing the sample sources from the Gaulton study (nPOD: Network for Pancreatic Organ Donors with Diabetes, UPenn: University of Pennsylvania). C UMAP representation of original study-defined cell types. D Dot plot of the top ten significantly enriched GO Biological Process (GO BP) terms in β-cells (T1D vs. ND). E Dot plot of T1D-relevant pathways (GO BP) significantly enriched in β-cells (T1D vs. ND). F GSEA of YAP target genes and Hippo signaling pathway in β-, alpha-, ductal-, and activated stellate cells. GSEA was performed using the GSEAPY tool with the GO Biological Process 2023 gene sets, applying a Kolmogorov–Smirnov-like enrichment score. Statistical significance was assessed via permutation testing followed by false discovery rate (FDR) correction.
Fig. 3. CTGF is upregulated in the pancreas of T1D and AAb+ organ donors.
RNAscope in situ hybridization for CTGF (turquois) was performed on controls (n = 76 independent positions from 5 donors), AAb+ donors (n = 62 independent positions from 4 donors), and donors with T1D (n = 78 independent positions from 5 donors); double RNAscope was performed for YAP (pink). Quantification of CTGF + puncta/cell is presented as (A) all independent positions, (B) the mean value for each analysed pancreatic section from each donor, and (C) the percentage of cells with <5 (grey), 5-15 (blue) and >15 CTGF+ (pink) puncta. D Representative images display double RNAscope for CTGF (turquois) and Yap1 mRNA (pink), counterstained with Hematoxylin, shown larger (upper) and smaller (lower) magnification. Both box plots showing single analytes and median (box and whiskers; min to max show all points). Data are expressed as means ± SEM. _P-_values were calculated by one-way ANOVA with Holm-Sidak multiple comparisons correction. Scale bars depict 10 µm. Source data are provided as a Source Data file.
Fig. 4. YAP colocalizes and correlates with enteroviral RNA expression in the pancreas.
A –D Detection and quantification of Yap1 mRNA (pink) and viral RNA-CVB3/4 (turquois) by RNAscope in situ hybridization from FFPE nPOD pancreas sections of AAb+ (n = 9) and T1D donors (n = 10). A Representative images of Yap1/CVB-RNA double labelling from AAb+ and T1D pancreatic sections and (B, C) total distribution and quantification throughout the whole pancreas section differentiated in YAP-viral RNA double positive cells (YAP+/CVB+; purple), CVB-positive cells in close proximity of YAP-positive neighbor cells (n-YAP+/CVB+; blue) or YAP-negative but CVB-RNA-positive cells (YAP-/CVB+; gray). D Quantification of all viral RNA-positive cells throughout the whole pancreas section in AAb+ and T1D donors presented as the mean number of single (white; 5-10 single puncta/cell) or cluster (black; >10 single puncta/cell) infected cells. E Association between YAP protein expression and number of enterovirus-positive cells by smFISH for enteroviral RNA detection in AAb+ (n = 9) and T1D (n = 14) donors. F Representative microscopical images of enteroviral RNA (red; Stellaris probes) and YAP protein (brown; IHC) expression in the pancreas showing YAP+/Enterovirus+ cells (YAP+/V+) and enteroviral positive cells in close proximity of YAP-positive neighbor cells (n-YAP+/V+). G Quantification of YAP-protein+/viral smFISH+ cells (YAP+/CVB+; purple), viral smFISH+ cells in close proximity of YAP-positive neighbor cells (n-YAP+/CVB+; blue) or YAP-negative but viral smFISH+-positive cells (YAP-/CVB+; gray) throughout the whole pancreas; control represents only one single enteroviral RNA+ cell, which was YAP-negative (n = 3 for both AAb+ and T1D donors). Data are expressed as means ± SEM. _P-_values were calculated by one-way ANOVA with Holm-Sidak multiple comparisons correction for C, and two-tailed unpaired Student _t-_test for D, E (Spearman) and G. *P = 0,0219 YAP+/V+ vs. n-YAP+/V+; **p = 0,00053 YAP+/V+ vs. YAP-/CVB+ for Aab+ group; § p = 8.5 × 10⁻7 YAP+/V+ vs. n-YAP+/V+; §§ p = 0,00011 YAP+/V+ vs. YAP-/CVB+ for T1D group. Scale bars depict 10 µm. Source data are provided as a Source Data file.
Fig. 5. Coxsackievirus infection induces YAP expression and activity in islet-exocrine co-cultures.
A Islet-exocrine co-cultures from the same donor were infected with CVB4 (MOI = 10) for 6–24 h. B qPCR analysis of Yap1, AMOTL2, ANKRD1 and CTGF mRNA expression, normalized to actin (n = 4 organ donors). C Representative images of double RNAscope for CTGF (turquoise) and Yap1 mRNA (pink) in human islets transduced with LacZ or YAP or infected with CVB4. D Representative Western blot image and (E) pooled quantitative densitometry analysis of human islet-exocrine co-cultures, showing total, phosphorylated (S127), and active YAP (n = 3 organ donors). Data are expressed as means ± SEM. _P-_values were calculated by two-tailed paired Student _t-_test. Scale bar depicts 20 µm. A Image adapted from Servier Medical Art (
), licensed under CC BY 4.0 (
https://creativecommons.org/licenses/by/4.0/
). Source data are provided as a Source Data file.
Fig. 6. YAP enhances coxsackievirus replication and potentiates coxsackievirus-induced islet inflammation and β-cell apoptosis.
A, C, D, L, M INS-1E cells and (B, E–H, N–X) human islets transduced with Ad-YAP or Ad-LacZ control and then infected with CVB4 (MOI = 5) for 24 h (INS-1E) or CVB3 and -4 (MOI = 10) for 48 h (human islets). A, B Intracellular CVB3 or -4 RNA genome of (A) INS-1E cells (n = 3 independent experiments) and (B) human pancreatic islets (n = 4 organ donors). C –F Representative Western blots and pooled quantitative densitometry analysis of VP1 in (C, D) INS-1E cells (n = 7 independent experiments) and (E, F) human islets (n = 6 organ donors). G, H Representative images (G) and quantitative percentage of VP1-positive β-cells (H) are shown (n = 4 organ donors). I –K Human islets (50%) co-cultured with exocrine cells infected with CVB3 and −4 (MOI = 10) for 48 h and treated with or without 2.5 uM verteporfin (VP) for the last 24 h. I Quantitative percentage of %VP1/insulin+ cells (n = 4 organ donors). J Representative images and (K) quantitative percentage of %VP1/CK19+ cells (n = 6 independent positions from two organ donors). L –O Representative Western blots and pooled quantitative densitometry analysis of cleaved caspase 3 in (L, M) INS-1E cells (n = 7 independent experiments) and (N, O) human islets (n = 6 organ donors; endogenous YAP expression under control conditions stems from exocrine cells, which typically remain even in highly purified human islets cultures). P, Q Representative images (P) and quantitative percentage of TUNEL-positive β-cells (Q) are shown (n = 3 organ donors). R –W qPCR analysis for (R) IFNB1, (S) CXCL10, (T) OAS1, (U) IFIH1, (V) DDX58, and (W) TLR3 mRNA expression in isolated human islets normalized to actin (R, U, W: n = 5, S, V: n = 4, T: n = 3 organ donors). X Secreted CXCL10 analyzed by ELISA in the culture media (n = 8 independent samples from five organ donors). Data are expressed as means ± SEM. _P-_values were calculated by two-tailed paired (A, B, D, F, H, I, K, M, O, Q) or unpaired (X) or ratio paired (R–W) Student _t-_test. Scale bars depict 50 µm (G, P) and 10 µm (J). Source data are provided as a Source Data file.
Fig. 7. YAP’s pro-inflammatory effect depends on viral amplification.
INS-1E cells (A, B) or human islets (C, D) were transduced with Ad-YAP or Ad-LacZ control and then treated (INS-1E) or transfected (human islets) with Poly(I:C). A Representative Western blot and (B) pooled quantitative densitometry analysis of cleaved caspase 3 in INS-1E cells (n = 3 independent experiments). C qPCR analysis for CXCL10, IFNB1, OAS1, IFIH1, DDX58 and IL6 expression in isolated human islets, normalized to actin (n = 3 organ donors). D secreted CXCL10 analyzed by ELISA in the culture media (n = 3 organ donors). E, F Human islets were transduced with Ad-YAP or Ad-LacZ control for 24 h and then infected with CVB4 (MOI = 10) for 48 h with or without treatment with 10 mM pleconaril (n = 3 organ donors). E Intracellular CVB4 RNA genome. F qPCR analysis for CXCL10, IFNB1, and OAS1 expression in isolated human islets, normalized to actin. Data are expressed as means ± SEM. _P-_values were calculated by two-tailed paired Student _t-_test. Source data are provided as a Source Data file.
Fig. 8. YAP re-expression induces diabetes by impairing insulin secretion and inducing β-cell dedifferentiation.
A Scheme how β-YAP mice were generated by crossing RIP-rtTA with TetO-YAPSer127A mice. B IHC and C Western Blot confirmation of YAP induction in pancreatic islets after 2 days i.p injection of Dox. D YAP was transiently induced by doxycycline (DOX) administration in drinking water for 2 weeks (β-YAP) and results compared to -DOX/-YAP (control; C). E –H intraperitoneal glucose tolerance test (ipGTT) and respective AUC analyses in β-YAP and control male (E, F; n = 16 C, n = 18 β-YAP) and female (G, H, n = 12/group) mice. I, J Insulin levels during an ipGTT measured before (0 min) and 15/30 min after glucose injection in β-YAP and control male (I; n = 11/group) and female (J; n = 9 C, n = 11 β-YAP) mice. K, L Islets were isolated from β-YAP and control mice, cultured overnight and subjected to an in vitro GSIS. K Insulin secretion during 1 h-incubation with 2.8 mM (basal) and 16.7 mM glucose (stimulated), normalized to insulin content and (L) stimulatory index denotes the ratio of stimulated to basal insulin secretion (n = 11). M RT-PCR for MafA, Nkx6.1, Slc2a2, NeuroD1, GCK, Ins1, Ins2, Pdx1, Glis3 (n = 6 C, n = 11 β-YAP mice), Nkx2.2 (n = 6 C, n = 10 β-YAP mice), Abcc8, and Kcnj11 (n = 3 C, n = 4 β-YAP mice). Microscopical analyses of β-proliferation by Ki67 (N, O) and pHH3 (P, Q) in both (N, P) male and (O, Q) female mice expressed as percentage of Ki67- (n = 6 mice/group) or pHH3- (n = 6 male and n = 5 female mice/group) positive β-cells. Insulin-positive area (R, T) and β-cell mass (S, V) in both (R, S) male and (T, U) female mice (n = 6 mice/group). Data are expressed as means ± SEM. _P-_values were calculated by two-tailed unpaired Student _t-_test for all except by a mixed-effects model with Holm-Sidak multiple comparisons correction for (E, G). ***p < 0.001 compared to control; Scale bars depict 20 µm. Source data are provided as a Source Data file.
Fig. 9. A YAP-TEAD axis regulates the level of MST1.
A, B, E INS-1E cells and (C, D) human islets transduced with Ad-YAP or Ad-LacZ control for 48 h. A –D Representative Western blot and pooled quantitative densitometry analysis of MST1 in (A, B) INS-1E cells (n = 6 independent experiments) and (C, D) human islets (n = 7 organ donors). E qPCR for STK4 mRNA expression in INS-1E cells normalized to actin (n = 3 independent experiments). F, G, H INS-1E cells and (I, J) human islets transduced with Ad-YAP or Ad-LacZ control for 48 h treated with or without 1–5 μM verteporfin (VP) for last 6 h (INS-1E) or 24 h (human islets). F qPCR for STK4 mRNA expression in INS-1E cells normalized to actin (n = 3 independent experiments). G–J western blots and pooled quantitative densitometry analysis of MST1 in (G, H) INS-1E cells (n = 3 independent experiments) and (I, J) human islets (n = 4 donors). K–N Hela or INS-1E cells transduced with Ad-YAP or Ad-LacZ control for 48 h treated with or without 1 μM verteporfin (VP) for the last 24 h. K Hela cells culture media was analyzed for activities of both GLuc and SEAP and data presented as the relative change in normalized GLuc to SEAP (n = 8 independent experiments). L–N ChIP from INS-1E cells was performed with control IgG, or YAP antibody as indicated (n = 3 independent experiments). The presence of (L, M STK4 and (N) ANKRD1 promoters was detected by PCR. Data presented as fold enrichment in which ChIP signals are divided by the IgG-antibody signals, representing the fold increase in signal relative to the background signal. Data are expressed as means ± SEM. _P-_values were calculated by one-way (K–N) and two-way (F, H, J) ANOVA with Holm-Sidak multiple comparisons correction, and two-tailed paired Student _t-_test for B,D,E. Source data are provided as a Source Data file.
Fig. 10. A YAP-TEAD-MST1 feedback loop controls CVB replication and cell death.
A –E INS-1E cells transfected with siMST1 or control siScr and then transduced with Ad-YAP or Ad-LacZ control for 48 h. All cells were infected with CVB4 (MOI = 5) for last 24 h. A, B Representative Western blot and pooled quantitative densitometry analysis of MST1, VP1 and cleaved caspase 3 in INS-1E cells (n = 3 independent experiments). C, D Representative images (C) and quantitative percentage of VP1-positive cells (D) are shown (n = 28 independent positions for C-LacZ; n = 30 for YAP; n = 29 for siMST1-YAP). E Intracellular CVB4 RNA genome of INS-1E cells (n = 4 independent experiments). F –J INS-1E cells transfected with MST1-K59 or control GFP constructs and then transduced with Ad-YAP or Ad-LacZ control for 48 h. All cells were infected with CVB4 (MOI = 5) for last 24 h. F, G Representative Western blot (F) and pooled quantitative densitometry analysis (G) of MST1, VP1 and cleaved caspase 3 in INS-1E cells (n = 3 independent experiments). H, I Quantitative percentage of VP1-positive cells (H) and representative images (I; n = 50 independent positions for C-LacZ; n = 42 for YAP; n = 40 for MST1-K59-YAP). J Intracellular CVB4 RNA genome of INS-1E cells (n = 3 independent experiments). Data are expressed as means ± SEM. _P-_values were calculated by one-way ANOVA with Holm-Sidak multiple comparisons correction for (B, D, G, H) and by two-tailed paired Student _t-_test for (E, J). Scale bars depict 10 µm. Source data are provided as a Source Data file. K, L Our model how a vicious cycle of YAP expression and CVB replication in the human pancreas may lead to T1D. K At the molecular level, YAP induces the expression of its own negative regulator MST1, through a feedback mechanism, thereby limiting YAP-driven viral replication and promoting apoptosis of infected cells. YAP is highly elevated in the pancreas of patients with T1D where it boosts enteroviral replication, induces a strong IFN response, and promotes islet inflammation, ultimately leading to β-cell apoptosis and destruction. L Persistently infected exocrine cells, where YAP promotes viral replication, may drive T1D by serving as viral reservoirs that facilitate islet infection and by triggering local inflammation that attracts immune cells and damages β-cells. We extend our gratitude to Richard E. Lloyd for kindly providing the enterovirus image. K, L Image adapted from Servier Medical Art (
), licensed under CC BY 4.0 (
https://creativecommons.org/licenses/by/4.0/
). Source data are provided as a Source Data file.
References
MeSH terms
Substances
Grants and funding
- UC4 DK098085/DK/NIDDK NIH HHS/United States
- UC4 DK112232/DK/NIDDK NIH HHS/United States
- U01 DK123594/DK/NIDDK NIH HHS/United States
- UC4 DK112217/DK/NIDDK NIH HHS/United States
- U01 DK123716/DK/NIDDK NIH HHS/United States
LinkOut - more resources
Full Text Sources
Medical
Molecular Biology Databases
Research Materials
Miscellaneous