Differential phenotypes in mice with smooth muscle-specific heterozygous versus homozygous deletion of Stk11

Abstract

Liver kinase B1 (Lkb1), encoded by the serine/threonine kinase gene STK11, functions as a critical tumor suppressor associated with Peutz-Jeghers syndrome (PJS). Previous studies have shown that mesenchymal-specific deletion of Stk11 promotes the development of PJS-like polyps in murine models, highlighting its essential role in mesenchymal homeostasis. In this study, we generated tamoxifen-inducible Stk11^flox/flox;Myh11-Cre/ERT2 mice to achieve targeted deletion of Stk11 in mature smooth muscle cells (SMCs), allowing a comparative analysis of heterozygous versus homozygous loss-of-function phenotypes. We found that homozygous ablation of Stk11 in mature Myh11⁺ SMCs resulted in systemic arteriomegaly and extensive dilatation of gastrointestinal and genitourinary organs—including intestinal segments, gallbladder, ureters, and bladder. These structural abnormalities were associated with significant thinning of the muscularis propria layer and compromised tissue integrity. Interestingly, heterozygous deletion of Stk11 in mature Myh11⁺ SMCs fully recapitulated polyp formation characteristic of PJS, whereas complete knockout of Stk11 in these cells unexpectedly failed to induce gastrointestinal polyposis. Our results establish Stk11 as a fundamental regulator of both vascular and visceral SMC homeostasis. The genotype-dependent phenotypic divergence in SMC-specific Stk11 deficiency models underscores distinct mechanistic contributions to tissue-specific tumor suppression and PJS pathogenesis.

Introduction

Liver kinase B1 (Lkb1), a tumor-suppressive serine/threonine kinase encoded by the STK11 gene, is a key driver in the pathogenesis of Peutz-Jeghers syndrome (PJS)—an autosomal dominant cancer predisposition disorder characterized by gastrointestinal hamartomatous polyposis and increased susceptibility to malignant transformation. Clinical evidence have firmly established that germline mutations in STK11 lead to the development of PJS^1,2. Correspondingly murine models recapitulate this pathology: constitutive Stk11 knockout results in embryonic lethality with characteristic vascular malformations and neural tube defects mediated by aberrant mesenchymal apoptosis³, whereas heterozygous deletion induces spontaneous intestinal polyposis mirroring human PJS phenotypes⁴. Importantly, lineage-specific studies demonstrate that mesenchymal Stk11 ablation alone is sufficient to drive polyposis in mice^5–7, underscoring its essential role in maintaining mesenchymal homeostasis.

Building on our previous studies using tamoxifen-inducible Stk11^flox/flox;Myh11-Cre/ERT2 mice—a model enabling spatiotemporal Stk11 deletion in mature smooth muscle cells (SMCs)—we reported that homozygous Stk11 loss triggers aortic aneurysms and premature mortality^8,9. In the present study, we extend these findings by systematically comparing the effects of heterozygous versus homozygous Stk11 deletion in mature SMCs, aiming to elucidate their distinct contributions to vascular and visceral pathophysiology.

Methods

Ethics statement

This study was performed in accordance with relevant guidelines and regulations in Shanghai Chest Hospital affiliated to Shanghai Jiao Tong University School of Medicine. All methods were reported in accordance with ARRIVE guidelines.

Animal models

The Myh11-Cre/ERT2 (Stock No. 019079), Pdgfrα-Cre/ERT2 (Stock No. 032770), and Stk11^flox/flox (Stock No. 014143) mouse strains were obtained from The Jackson Laboratory to generate lineage-specific Stk11 knockout models via tamoxifen-inducible Cre-loxP recombination. To achieve smooth muscle-specific deletion, Myh11-Cre/ERT2 transgenic mice were initially crossed with Stk11^flox/flox mice to produce Stk11^flox/+;Myh11-Cre/ERT2 offspring. These heterozygous mice were then intercrossed to yield three experimental groups: Stk11^+/+;Myh11-Cre/ERT2 (wild-type controls), Stk11^flox/+;Myh11-Cre/ERT2 (heterozygous knockout), and Stk11^flox/flox;Myh11-Cre/ERT2 (homozygous knockout). Similarly, for Pdgfrα-lineage targeting, Pdgfrα-Cre/ERT2 mice were crossed with Stk11^flox/flox mice to generate Stk11^flox/+;Pdgfrα-Cre/ERT2 founders, which were intercrossed to produce littermates of the following genotypes: Stk11^+/+;Pdgfrα-Cre/ERT2, Stk11^flox/+;Pdgfrα-Cre/ERT2, and Stk11^flox/flox;Pdgfrα-Cre/ERT2.

To initiate cell-type-restricted recombination, post-weaning mice (4–8 weeks old) underwent daily intraperitoneal administration of tamoxifen (1 mg/mouse in 100 µL sunflower oil vehicle) for five consecutive days, achieving Cre-mediated Stk11 excision specifically in target mesenchymal lineages. All animals were maintained under standardized conditions (22 ± 1 °C, 12/12-h light/dark cycle) with free access to food and water throughout the experimental period.

As previously reported, all homozygous Stk11^flox/flox;Myh11-Cre/ERT2 mice exhibited premature mortality within 8 months following tamoxifen induction⁸. Given that aortic aneurysm development in this model becomes visibly apparent after 4 months post-induction, comprehensive phenotyping—including assessments of vascular, gastrointestinal, and other organs—was performed at 4–5 months post-tamoxifen induction. In heterozygous Stk11^flox/+;Myh11-Cre/ERT2 mice, as described in our related work⁷, visible polyps at the gastroduodenal junction began emerging by 9 months after tamoxifen treatment, with pronounced pathology evident at 12 months. Accordingly, gastrointestinal phenotyping for Stk11^flox/+;Myh11-Cre/ERT2 and Stk11^flox/+;Pdgfrα-Cre/ERT2 mice was conducted at 12 months post-induction.

Vascular ultrasound

Vascular ultrasound was performed using a Vevo 3100 Imaging System (FUJIFILM VisualSonics, Toronto, ON, Canada). Mice were anesthetized with isoflurane (1–3% vol/vol, titrated to effect) and positioned on a temperature-controlled heating platform (37 °C) to minimize procedural stress and prevent hypothermia. A consistent imaging protocol was applied to all animals: the abdominal aorta was visualized in B-mode using a 40 MHz transducer (MX550D). Aortic wall motion was recorded in M-mode at predefined anatomical landmarks along the infrarenal aorta (e.g., distal to the renal arteries), with all images acquired at standardized depth and gain settings.

Systolic diameter (Ds) was defined as the maximum aortic lumen diameter during ventricular systole, and diastolic diameter (Dd) as the minimum lumen diameter at end-diastole. Both were measured from M-mode tracings using the leading-edge-to-leading-edge method at consistent anatomical sites. Measurements were taken over three consecutive cardiac cycles and averaged. Circumferential cyclic strain was calculated as (Ds–Dd)/Dd^10,11.

Aortic pulse wave velocity (PWV) was determined by dividing the distance between two measurement points by the time shift of the waveforms¹². For the abdominal aorta, blood flow and physiological signals were recorded at two locations: the suprarenal branch and approximately 1 mm proximal to the left renal artery branch. The time lag in flow between these sites was measured relative to the ECG R-wave. The pulse transit time (PTT or ΔT) was calculated as Δt₁ – Δt₂, where Δt₁ and Δt₂ represent the time lags at the distal (suprarenal) and proximal sites, respectively. The distance (S) between sites was measured from color Doppler images, and PWV was computed as S / ΔT. All echocardiographic procedures, including acquisition and analysis, were performed by an investigator blinded to sample identity.

Contrast-enhanced micro-computed tomography (micro-CT) scanning imaging

Prior to imaging procedures, experimental mice were anesthetized in a standardized acrylic induction chamber using 2–4% isoflurane inhalation. Upon achieving surgical-level anesthesia (confirmed by absence of pedal reflex), subjects received intravenous administration of 100–150 µL blood-pool contrast agent (ExiTron nano 12000, nanoPET Pharma GmbH, Berlin, Germany) via tail vein catheterization. Continuous anesthesia maintenance (1–2% isoflurane) was ensured through a nose cone delivery system during subsequent positioning in the imaging cradle.

Imaging protocols were initiated immediately post-contrast administration using a high-resolution SkyScan 1176 micro-CT system (Bruker MicroCT, Kontich, Belgium). Raw projection data were reconstructed using NRecon software (v1.7.1.0) with beam hardening correction and ring artifact reduction algorithms. The reconstructed volumetric datasets were converted to DICOM format through DicomCT conversion software for quantitative analysis in CVI42 post-processing workstation (Circle Cardiovascular Imaging, Calgary, Canada), where vascular architecture was evaluated using dedicated vessel analysis modules.

Aortic dimensions were quantified at three anatomical landmarks: (1) thoracic segment at the ventricular apex level, (2) supra-celiac abdominal segment (immediately distal to the celiac trunk origin), and (3) infrarenal segment (proximal to the aortic bifurcation). Medium arterial evaluation encompassed bilateral common iliac arteries (LCIA/RCIA) and femoral arteries (LFA/RFA).

Tissue collection, processing, and histology

Mice were euthanized by inhalation of 5% isoflurane and cervical dislocation where appropriate. Tissue samples for pathological analysis were fixed in 10% neutral buffered formalin and embedded in paraffin. Sections of 5 μm thickness were prepared from paraffin-embedded tissues and stained with hematoxylin and eosin (H&E) according to standard protocols^13,14. All images were recorded using an Olympus digital camera (Tokyo, Japan).

Morphometric and histological analysis

All morphometric measurements were performed using ImageJ software (National Institutes of Health, Bethesda, MD, USA) on high-resolution images of intact organs or histological sections. For intestinal length measurement, the entire small intestine was carefully dissected from the pylorus to the ileocecal junction, placed gently on moist filter paper without stretching, and imaged alongside a scale ruler. Length was measured in triplicate along the intestinal midline. For colon length measurement, the entire colon was dissected from the cecum to the distal rectum, handled without stretching under moist conditions, and photographed next to a reference scale. Colon length was similarly determined in triplicate using ImageJ by tracing the central axis of the colon.

For cross-sectional diameter assessments (including the aorta, femoral artery, duodenum, and ileum), three non-consecutive H&E-stained sections per animal were analyzed. All measurements were performed by two independent investigators who were blinded to the genotype. Values are presented as mean ± SEM (Standard Error of the Mean).

Statistical analysis

Statistical analyses were performed using GraphPad Prism 8.0 (GraphPad Software, San Diego, CA, USA). Unpaired two-tailed Student’s t-tests were used to calculate significant differences between two groups. Multiple comparison correction analysis was performed using one-way ANOVA followed by Bonferroni’s post hoc test. P < 0.05 was considered statistically significant.

Results

Aortic dilation in SMC-specific homozygous, but not heterozygous, Stk11-deficient mice

Our previous investigations established that SMC-specific homozygous deletion of Stk11 (Stk11^flox/flox;Myh11-Cre/ERT2) drives progressive aortic pathobiology, culminating in arterial/aortic dilation, aneurysm formation, and fatal rupture^8,9. To systematically delineate the phenotypic consequences of heterozygous versus homozygous Stk11 deletion in mature SMCs, we performed comparative analyses across three genotypes following tamoxifen induction: Stk11^+/+;Myh11-Cre/ERT2 (wild-type, WT), Stk11^flox/+;Myh11-Cre/ERT2 (heterozygous, HT), and Stk11^flox/flox;Myh11-Cre/ERT2 (homozygous, HO).

Non-invasive vascular ultrasonography revealed striking genotype-dependent vascular remodeling. Homozygous mutants exhibited significant aortic dilation consistent with our prior findings^8,9 (Fig. 1A–C), whereas heterozygous mice maintained normal aortic morphology throughout extended observation periods (over 12 months post-induction) (Fig. 1A–C). Moreover, homozygous mice displayed marked vascular stiffening, characterized by a 39.05% reduction in vascular strain and 142% increase in pulse wave velocity compared to WT littermates (Fig. 1D,E).

Fig. 1.

Aortic dilation in mice with smooth muscle-specific homozygous deletion of Stk11 revealed by vascular ultrasound. (A) Representative ultrasonographic images of the abdominal aorta in Stk11^+/+;Myh11-Cre/ERT2 (wild-type, WT), Stk11^flox/+;Myh11-Cre/ERT2 (heterozygous, HT), and Stk11^flox/flox;Myh11-Cre/ERT2 (homozygous, HO) mice. Images were acquired at 4 months post-tamoxifen induction for WT and HO mice, and at 4, 8, and 12 months for HT mice. Red dashed arrows indicate the internal diameter of the aorta. (B,C) Quantification of diastolic (B) and systolic (C) aortic luminal diameters across genotypes. (D,E) Analysis of circumferential cyclic strain (D) and aortic stiffness evaluated by pulse wave velocity (PWV) (E) in the abdominal aorta among genotypes. Data are presented as mean ± SEM (WT, n = 8; HT-4 M, n = 4; HT-8 M, n = 12; HT-12 M, n = 4; HO-4 M, n = 7). P-values were calculated using one-way ANOVA followed by Bonferroni’s post hoc test (B–E). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001; NS, not significant.

To further assess vascular morphology, we performed computed tomography angiography (CTA) across all genotypes. Quantitative analyses confirmed pronounced aortic enlargement in homozygous knockouts compare with both heterozygous and WT controls. The dilation involved multiple regions, including the descending thoracic and abdominal aorta at defined anatomical landmarks (the celiac trunk origin and aortic bifurcation). Systemic arterial dilation also extended to the bilateral common iliac (LCIA/RCIA) and femoral arteries (LFA/RFA), with homozygous mice exhibiting markedly increased diameters across all assessed vessels (Fig. 2A,B). These results indicate that complete Stk11 ablation induces pan-vascular dilation phenotype rather than localized vascular changes.

Fig. 2.

Systemic arterial dilation in Stk11 homozygous mutants revealed by computed tomography angiography (CTA). (A) Representative CTA cross-sectional images of aortic segments including descending thoracic aorta (TA), supraceliac abdominal aorta at celiac trunk origin (AA-L1), infrarenal abdominal aorta proximal to bifurcation (AA-L2), right common iliac artery (RCIA), and right femoral artery (RFA) in Stk11^+/+;Myh11-Cre/ERT2 (WT), Stk11^flox/+;Myh11-Cre/ERT2 (heterozygous, HT), and Stk11^flox/flox; Myh11-Cre/ERT2 (homozygous, HO) mice at 4 months post-tamoxifen induction. (B) Quantitative analysis of mean luminal diameters across aortic segments (TA, AA-L1, AA-L2) and peripheral arteries (LCIA, RCIA, LFA, and RFA), revealing genotype-dependent vascular remodeling (6-n = 5-7 per group). Data are presented as mean ± SEM. P-values were calculated using one-way ANOVA followed by Bonferroni’s post hoc test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Histopathological examination via H&E staining confirmed the findings from CTA. Sections of the thoracic and abdominal aortas from Stk11^flox/flox;Myh11-Cre/ERT2 mice displayed marked luminal dilation compared with heterozygous and WT littermates (Fig. 3A–E). Notably, cells with typical chondrocyte morphology were present exclusively in aortic sections of homozygous knockout mice (Fig. 3A), consistent with our previous report⁸. Likewise, cross-sections of femoral arteries from homozygous knockout animals exhibited pronounced luminal expansion, indicative of aneurysm formation (Fig. 3A,F,G). The consistency between radiographic and histological measurements confirms the development of systemic arteriomegaly specifically in homozygous Stk11-knockout mice. In contrast, heterozygous mice exhibited vascular dimensions comparable to WT controls, underscoring that complete loss of Stk11, rather than haploinsufficiency, is required to compromise arterial wall integrity.

Fig. 3.

Histological evidence of aortic and femoral artery dilation in Stk11 homozygous knockout mice. (A) H&E staining of thoracic aorta, abdominal aorta, and femoral artery in Stk11+/+;Myh11-Cre/ERT2 (WT), Stk11flox/+;Myh11-Cre/ERT2 (heterozygous, HT), and Stk11flox/flox; Myh11-Cre/ERT2 (homozygous, HO) mice at 4–5 months post-tamoxifen induction. Scale bars: 100 μm (main panels), 200 μm (insets). (B–G) Morphometric analysis of the internal and external elastic laminae (IEL and EEL, respectively) perimeters in thoracic aorta (B,C), abdominal aorta (D,E), and femoral artery (F,G) (WT, n = 5; HT, n = 5; HO, n = 7-9). Data are presented as mean ± SEM. P-values were calculated using one-way ANOVA followed by Bonferroni’s post hoc test (B,C). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Severe bowel dilatation exclusive to homozygous Stk11 conditional knockout mice

Beyond the established vascular abnormalities, Stk11-deficient mice (Stk11^flox/flox;Myh11-Cre/ERT2) exhibited extensive gastrointestinal remodeling. Comprehensive morphometric analyses revealed progressive increases in absolute gastrointestinal mass and elevated organ-to-body weight ratios compared to WT controls (Fig. 4A,B). Homozygous mutants also showed substantial elongation of intestinal segments, with small intestine length increased by 36.98% (44.82 ± 5.38 cm vs. WT 32.72 ± 3.29 cm, p < 0.0001) and colon length by 30.05% (8.18 ± 0.60 cm vs. WT 6.29 ± 0.68 cm, p < 0.0001) (Fig. 4C,D).

Fig. 4.

Severe bowel dilatation in Stk11 homozygous SMC-specific knockout mice. (A,B) Absolute (A) and normalized (B) GI tract mass in Stk11+/+;Myh11-Cre/ERT2 (WT) and Stk11flox/flox;Myh11-Cre/ERT2 (homozygous, HO) mice at 1 and 4-5-months post-tamoxifen induction (WT/1 M: n = 8; HO/1 M: n = 9; WT/4-5 M: n = 16; HO/4-5 M: n = 13). (C,D) Lengths of small intestine (C) and colon (D) in WT and Stk11 HO mice at 4–5 months post-tamoxifen induction (WT: n = 16; HO: n = 17). (E-G) Duodenal architectural analysis: Representative H&E-stained sections (E), with quantifications of luminal diameter (F) and muscularis propria thickness (G) in WT and HO cohorts at 4–5 months post-tamoxifen induction (WT: n = 8; HO: n = 7). (H–J) Ileal structural evaluation: Representative H&E staining (H), and quantifications of luminal diameter (I) and muscularis propria thickness (J) in corresponding groups (n = 6 per group). Scale bars: 500 μm (E,H). Data are presented as mean ± SEM. P-values were calculated by two-tailed Student’s t-test. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Histopathological profiling revealed progressive architectural disruption along the gut axis. Serial cross-sections from duodenum (Fig. 4E–G) to ileum (Fig. 4H–J) demonstrated marked luminal expansion concurrent with muscularis propria thinning. Remarkably, homozygous mutants displayed compromised intestinal wall integrity characterized by structural disorganization and villus fragmentation (Fig. 4E,H). These findings suggest that STK11 signaling is critical for maintaining both the muscular architecture and epithelial homeostasis within the gastrointestinal tract.

Gastrointestinal polyposis resulting from heterozygous Stk11 deletion in mature SMCs

Distinct from the homozygous phenotype, Stk11^flox/+;Myh11-Cre/ERT2 mice developed a pronounced predisposition to polyp formation at the gastroduodenal junction (Fig. 5A,B). This observation demonstrates that heterozygous deletion of Stk11 in SMCs is sufficient to drive PJS-like polyposis. Notably, this phenotype contrasts with the complete absence of macroscopic gastrointestinal polyps in Stk11^flox/flox;Myh11-Cre/ERT2 mice, even at end-stage (Fig. 4E), consistent with our prior report⁷.

Fig. 5.

Heterozygous Stk11 deletion in SMCs induces gastrointestinal polyposis. (A) Macroscopic characterization of gastrointestinal polyposis in Stk11^+/+;Myh11-Cre/ERT2 (WT) and Stk11^flox/+;Myh11-Cre/ERT2 (heterozygous, HT) mice, demonstrating polyp formation (red arrows) localized to the gastroduodenal junction (demarcated by black dashed line) in HT mice at 12 months post-tamoxifen induction. (B) H&E staining showing normal duodenum morphology in WT controls versus stromal-rich polyps in age-matched HT mice. Scale bars: 1 mm.

Histopathological analysis revealed that gastrointestinal polyps in heterozygous mice exhibited characteristic mesenchymal hyperplasia, with prominent stromal cell accumulation within the polyp architecture (Fig. 5B). Collectively, these findings underscore a critical role for SMC-expressed STK11 in suppressing gastrointestinal polyposis.

Gastrointestinal polyposis resulting from homozygous and heterozygous Stk11 deletion in Pdgfra⁺ fibroblasts

To further characterize polyps induced by Lkb1 deficiency, we developed complementary mouse models targeting distinct stromal compartments and identified distinct spatiotemporal patterns of polyp development compared to the Myh11-driven knockout (Fig. 6A–D). In Lkb1^flox/flox;Pdgfra-Cre/ERT2 mice, tamoxifen-induced recombination efficiently ablated Stk11 in Pdgfrα⁺ stromal cells⁷ and produced an acute-onset phenotype: large gastroduodenal-junction polyps appeared within two months of induction (Fig. 6A,B). This rapid progression contrasts sharply with the delayed polyposis observed in Stk11^flox/+;Myh11-Cre/ERT2 mutants, as well as Lkb1^flox/+;Pdgfra-Cre/ERT2 mice. Partial Stk11 inactivation in Pdgfrα⁺ progenitors resulted in age-dependent penetrance, with gastrointestinal polyps developing significantly beyond 12 months post-induction (Fig. 6C,D). The differential latency between homozygous and heterozygous knockout models reflects the multi-hit nature of gastrointestinal tumorigenesis and reinforces the role of Lkb1 as a key gatekeeper gene in stromal homeostasis.

Fig. 6.

Homozygous and heterozygous deletion of Stk11 in Pdgfra⁺ fibroblasts leads to gastrointestinal polyposis. (A) Acute-onset gastrointestinal polyposis (red arrows) at the gastroduodenal junction (black dashed line) in Stk11^flox/flox;Pdgfra-Cre/ERT2 mice versus Stk11^+/+;Pdgfra-Cre/ERT2 (WT) controls at 2 months post-tamoxifen induction. (B) Duodenal H&E staining: normal mucosa in WT versus polypoid lesions in age-matched Stk11^flox/flox;Pdgfra-Cre/ERT2 mice. (C) Delayed polyp emergence (red arrows) in Stk11^flox/+;Pdgfra-Cre/ERT2 heterozygous mice compared to WT littermates at 12 months post-induction. (D) Comparative histopathology: preserved duodenal architecture in WT versus polyps in age-matched Stk11flox/+;Pdgfra-Cre/ERT2 mice. Scale bars: 1 mm (histological).

Broad dilatation of Hollow organs following SMC-specific homozygous Stk11 ablation

In addition to gastrointestinal manifestations, Stk11^flox/flox;Myh11-Cre/ERT2 mice exhibited profound visceral distention involving multiple hollow organs, including the ureter, bladder and gallbladder (Fig. 7). All Stk11^flox/flox;Myh11-Cre/ERT2 mice (10/10) spontaneously developed hydronephrosis and hydroureter, likely attributable to impaired urinary voiding mechanics, with consequent urine retention observed throughout the renal pelvis, ureters, and bladder (Fig. 7A–C).

Fig. 7.

Hollow organ dilatation in Stk11 homozygous SMC-Specific knockout mice. (A,B) Comparative analysis of urinary tract abnormalities in Stk11^+/+;Myh11-Cre/ERT2 (WT) and Stk11^flox/flox;Myh11-Cre/ERT2 (homozygous, HO) mice at 4–5 months post-tamoxifen induction, demonstrating macroscopic hydronephrosis/hydroureter (A) and corresponding H&E staining analysis (B). (C) Incidence rates of hydronephrosis in WT versus Stk11 HO cohorts. (D) H&E-stained ureters showing pathological luminal expansion in Stk11 HO mice. (E) Morphometric quantification of ureteral diameters (WT: n = 6; Stk11 HO: n = 5). (F,G) Bladder remodeling: gross distension (F) and detrusor muscle layer attenuation (G) visualized through macroscopic imaging and H&E histopathology. (H) Macroscopic gallbladder morphology comparison. Scale bars: 2 mm (B); 200 μm (D,G). Data are presented as mean ± SEM. P-values were calculated using two-tailed Student’s t-test (E). *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

D

Histopathological evaluation revealed dramatic ureteral luminal enlargement accompanied by aberrant cellular morphological changes in the urothelium (Fig. 7D,E). Concurrent pathologies included significant bladder and gallbladder dilation (Fig. 7F,H), with bladder specimens demonstrating both severe thinning/disorganization of the smooth muscle layers and near-complete loss of transitional epithelium (Fig. 7G).

These multiorgan phenotypes collectively underscore the essential role of STK11 in maintaining visceral smooth muscle contractility, epithelial-stromal integrity, and mural architecture, with homozygous Stk11 deletion precipitating catastrophic functional failure across urinary and biliary systems. The stark contrast between these structural collapses and the absence of gastrointestinal polyps in Stk11^flox/flox;Myh11-Cre/ERT2 mice suggests tissue-specific thresholds for LKB1’s tumor-suppressive versus homeostatic functions.

Discussion

Our study delineates a striking phenotypic dichotomy between homozygous and heterozygous Stk11 deletion in mature SMCs, underscoring previously underappreciated pleiotropic roles of this tumor suppressor in visceral homeostasis. While Stk11 haploinsufficiency in Myh11⁺ SMCs recapitulates the classic PJS phenotype with gastrointestinal polyposis, complete Stk11 ablation precipitates catastrophic multiorgan dysfunction resembling human megacystis-microcolon intestinal hypoperistalsis syndrome (MMIHS)^15,16 or visceral myopathy^17,18. This paradigm-shifting observation expands the recognized clinical spectrum of STK11-related pathologies, positioning Stk11 as a master regulator of both neoplastic suppression and structural maintenance within mesenchymal lineages.

Because constitutive Stk11 knockout causes embryonic lethality³, conditional models have been indispensable for dissecting its tissue-specific functions^{5–9,19–21}. The complete penetrance of PJS-like polyposis in Stk11^flox/+;Myh11-Cre/ERT2 mice aligns with previous findings from mesenchymal knockout models using Twist1-Cre and Tagln-Cre^5,6, thereby reinforcing the stromal origin theory of PJS. In contrast, Stk11^flox/flox;Myh11-Cre/ERT2 mice fail to develop polyps—an outcome that stands in sharp contrast to Stk11^flox/flox;Pdgfra-Cre/ERT2 mice, which exhibit acute-onset gastroduodenal polyps within two months of tamoxifen induction. This rapid disease progression differs markedly from the delayed (≈ 12 months) polyp development observed in Stk11^flox/+Myh11-Cre/ERT2 and Stk11^flox/+;Pdgfra-Cre/ERT2 mice, in which only partial loss of Stk11 occurs. Three major insights arise from these comparative analyses: (1) complete Stk11 loss in Pdgfra⁺ stromal progenitors drives rapid proximal gut pathology; (2) haploinsufficiency necessitates extended latency for phenotypic manifestation; (3) the Myh11-Cre vs. Pdgfra-Cre models recapitulate distinct temporal phases of human PJS evolution. Nevertheless, the mechanisms underlying the distinct phenotypic outcomes resulting from homozygous Stk11 ablation in Myh11⁺ SMCs versus Pdgfrα⁺ stromal progenitors remain to be fully elucidated and represent a significant avenue for future research.

The visceral myopathy phenotype in homozygous Stk11^flox/flox;Myh11-Cre/ERT2 mice unveils the fundamental role of Stk11 in SMC homeostasis. Homozygous deletion of Stk11 in mature SMCs leads to extensive dilation of multiple hollow organs, including the aorta, arteries, ureter, bladder, gallbladder, and gastrointestinal tract. These mice exhibit systemic arteriomegaly alongside features highly reminiscent of human MMIHS, a severe congenital visceral myopathy characterized by massive bladder enlargement and intestinal dysmotility^15,16. Mechanistically, the dilatation across organs is associated with thinning of the muscularis propria and loss of structural integrity, suggesting profound disruption of SMC function and tissue homeostasis.

The remarkable phenotypic concordance between the homozygous mutant phenotype and human visceral myopathies positions this model as a robust preclinical system for dissecting the pathophysiology of these rare disorders. By faithfully recapitulating hallmark disease features such as multi-organ dilation, smooth muscle hypocontractility, and epithelial disorganization, this system provides a unique platform to dissect the role of STK11/LKB1 signaling in visceral smooth muscle maintenance and degeneration. It also opens avenues for developing novel therapeutic strategies aimed at mitigating disease progression in conditions like MMIHS.

Together with prior studies^7–9, our findings highlight the central importance of Stk11 in mesenchymal biology. Although our data clearly demonstrate its essential role in SMC contractility and lineage maintenance, the precise molecular mechanisms governing these processes remain to be fully elucidated. The phenotypic divergence between heterozygous and homozygous models underscores a gene dosage-dependent functionality - Stk11 haploinsufficiency permits survival with neoplastic predisposition, while complete loss disrupts basic cellular contractility and structural integrity.

These distinct outcomes highlight the context-dependent roles of STK11 in smooth muscle physiology and point toward promising yet unexplored therapeutic implications. Further investigation remains essential to elucidate key unresolved questions, including: (1) How does Stk11 coordinate its tumor-suppressive and homeostatic roles in SMCs? (2) What molecular switches differentiate progenitor vs. mature SMC responses to Stk11 loss? (3) Do compensatory pathways mitigate polyposis in homozygous mutants?

Collectively, our work compels an updated conceptual framework where Stk11 functions as both gatekeeper—preventing neoplastic transformation—and guardian—preserving contractility and structural integrity—within mesenchymal tissues. Future investigations using single-cell transcriptomics and in vivo CRISPR screening will be instrumental in delineating the spatiotemporal regulation of these dual functions and in identifying actionable targets for both PJS and visceral myopathies.

Author contributions

H.C., G.Z., Y.J., and M.L. executed the experiments and analyzed the data. Y.H., L.F., L.C., D.X., and Y.H. provided technical assistance and advice. B.H. conceived and designed the experiments and revised the manuscript. Z.C. conceived and designed the experiments, analyzed the data, wrote and revised the manuscript.

Funding

This work was supported by the National Natural Science Foundation of China (grant nos. 82300489 and 82130012), the Shanghai Pujiang Program (grant no. 23PJD084), the Medicine-Engineering Interdisciplinary Grant at Shanghai Jiao Tong University (grant no. KYXJJQR202101), the Shanghai Arrhythmia Research Center Project (grant no. 2022ZZ01008), the Clinical Research Plan of SHDC (grant no. SHDC2020CR1039B), the SINO-German Mobility Programme (grant no. M-0526), the Innovative Research Team of Highlevel Local Universities in Shanghai (grant no. SHSMU-ZLCX20212302), and the Nurture projects for basic research of Shanghai Chest Hospital (grant no. 2022YNJCQ03). The funders had no role in the study design, data collection, data analyses, interpretation, or writing of the manuscript.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Competing interests

The authors declare no competing interests.