Effects of alpelisib treatment on murine Pten-deficient lipomas

Abstract

Phosphatase and tensin homolog (PTEN) hamartoma tumour syndrome (PHTS) is a rare disorder caused by germline mutations in the tumour suppressor gene PTEN, a key negative regulator of phosphatidylinositol 3-kinase (PI3K)/AKT signalling. Children with PHTS often develop lipomas, for which only surgical resection is available as treatment. We investigated the effects of the selective PI3K-inhibitor alpelisib on Pten-deficient lipomas. After incubation with alpelisib or the non-selective PI3K inhibitor wortmannin, we analysed histology, gene expression, and Pi3k pathway in lipoma and control epididymal adipose tissue (epiWAT). Alpelisib increased adipocyte area in lipomas compared to epiWAT. Baseline gene expression showed higher levels of markers for proliferation (Pcna), fibrosis (Tgfb1), and adipogenesis (Pparg) in lipomas, while hormone-sensitive lipase expression was lower than in epiWAT. Following alpelisib incubation, target genes of Pi3k signalling and extracellular matrix factors were reduced. We confirmed Pi3k inhibition through detecting decreased Akt levels compared to control treatment. Human lipoma samples treated with alpelisib showed variable lipolysis responses, suggesting variability in therapeutic outcomes. We established an ex vivo model to study alpelisib effects on Pten-deficient lipomas. These results underscore the therapeutic potential of targeted PI3K inhibition in the treatment of PHTS-associated lipomas, particularly in cases that are inoperable.

Keywords: Adipose tissue; phosphatase and tensin homolog; phosphatidylinositol 3-kinase; tissue slice culture; wortmannin.

Figure 1.

Schematic overview of PI3K/AKT/mTOR pathway and experimental design. a) PI3K activation induces conversion of phosphatidylinositol-4,5-bisphosphate (PIP2) to phosphatidylinositol-3,4,5-trisphosphate (PIP3). PIP3 is bound by AKT leading to AKT activation. Activated AKT indirectly stimulates mTOR regulating proliferation, lipolysis, adipogenesis, cell survival and growth. Alpelisib and wortmannin are direct inhibitors of PI3K. PTEN counteracts PI3K by dephosphorylating PIP3 to PIP2. b) Representative image of a 12- week-old homozygous Osx-Cre; Rb1^fl/fl; Pten^fl/fl mouse with an axillary lipoma. Second column: 2 lipomas in PBS (upper image) and matching epiWAT control tissue in culture media (lower image). Third column: lipoma on sterile filter membranes before cutting with tissue chopper. Fourth column: 500 μm thick lipoma slices in culture media. Fifth column: 3 lipoma slices cultured on liquid-air-interface. c) Representative H&E staining of lipoma tissue cross sections from d0 and treatment conditions (control, alpelisib and wortmannin) after 72 h of incubation. Scale bar = 100 um. Tissue heterogeneity and morphology is well maintained across all conditions.

Figure 2.

Adipocyte area in lipoma and epiWAT control tissue at d0 and after incubation with PI3K inhibitors. Each panel shows the area distribution curves on the left and the percentage of adipocytes below and above 547.85 µm² as bar charts on the right. The cut-off value of 547.85 µm² was determined by calculating the curve intersection point of the probability density functions for lipoma and epiWAT adipocytes and is indicated as a hatched line. The area distribution curves reflect the distribution of adipocyte area values within the sample, i.e. The probability of adipocyte area occurrence for any given adipocyte area value. In the accompanying bar charts, the percentage of adipocytes below/above 547.28 µm² is depicted. a) Lipoma tissue with higher number of small adipocytes compared to epiWAT at d0: before incubation lipoma adipocyte size peaked at 280 µm² compared to 250–1300 µm² in epiWAT. 29.9% of lipoma and 19.3% of epiWAT adipocytes exhibited a mean area ≤547.85 µm². b) epiWAT adipocytes after 72 h incubation in control medium showed similar size distribution as epiWAT adipocytes on d0, whereas lipoma adipocytes after 72 h incubation in control medium showed subtle area decrease (epiWAT: 1.66% less adipocytes ≤547.85 µm², lipoma: 4.7% less adipocytes ≤547.85 µm² after 72 h incubation in control medium). c) After 72 h incubation alpelisib-treated lipoma tissue showed subtle area increase (highest number of adipocytes in control medium: 280 µm², in wortmannin: 245 µm², in alpelisib 400 μm²). Number of adipocytes ≤547.85 µm² decreased by 9.1% in alpelisib-incubated slices compared to control medium incubation. Increase of lipoma adipocytes ≤547.85 µm² in wortmannin-incubated tissue compared to control medium incubation. d) After 72 h incubation, epiWAT slices both treated with alpelisib and with wortmannin showed a adipocyte area reduction (control medium: 400–1600 µm², alpelisib 400–1000 µm², wortmannin: 320–800 µm²). After alpelisib incubation, epiWAT adipocytes ≤547.85 µm² increased by 5.8% compared to control medium (control medium 17.7%, alpelisib 23.5%). After wortmannin incubation, epiWAT adipocytes ≤547.85 µm² increased by 8.2% (control medium 17.7%, wortmannin 25.9%). e) Comparison of changes of adipocyte area of epiWAT and lipoma tissue slices after 72 h E) alpelisib incubation and f) wortmannin incubation.

Figure 3.

Gene expression patterns in Pten deficient lipoma versus epiWAT. a) Compared to control epiWAT, Pten was significantly lower expressed in lipoma tissue (p = 0.001), whereas Pcna (p = 0.008), Pparg (p = 0.015) and Tgfb1 (p = 0.003) expression was elevated. Srebf1 (p = 0.008), Lipe (p < 0.0001), Lox (p = 0.0013) and Eln (p < 0.0001) expression was significantly decreased in lipoma. No difference was detected in Pnpla2 (p = 0.672) and Ptger3 (p = 0.0649) expression levels, with a trend towards lower expression of Ptger3. Data are presented as fold changes lipoma/control adipose tissue, which is depicted by a line at 1 (mean±sd, n = 7–8 independent experiments). b-d) Pparg, Pcna and Tgfb1 expression in alpelisib-treated versus control-treated lipoma and epiWAT slices. Expression of Pcna (p = 0.0066), Tgfb1 (p = 0.046) and Pparg (p = 0.059) was significantly reduced in alpelisib-incubated compared to control incubated lipoma slices. Alpelisib-treated epiWAT slices did not show significant gene expression differences compared to control treatment, with a trend towards lower Pcna expression in alpelisib vs control-treated epiWAT. Data were normalized to the mean of solvent controls and presented as mean±sd (n = 3–10 independent experiments). Abbreviations: Pcna = proliferating cell nuclear antigen, Tgfb1 = transforming growth factor beta, Pparg = peroxisome proliferator activated receptor gamma, Srebf1 = sterol regulatory element-binding factor 1, Lox = lysyl oxidase, Eln = elastin, Lipe = hormone-sensitive lipase, Pnpla2 = patatin-like phospholipase domain-containing protein 2.

Figure 4.

Expression differences of PI3K targets, genes involved in lipolysis and in extracellular matrix remodelling in alpelisib-treated lipoma. a) Pcna expression was reduced (n = 8), while b) Pgk1 expression was not changed and c) Glut1 expression was significantly elevated (n = 7) in lipoma tissue incubated with alpelisib compared to control-treated lipoma slices. Lipoma slices incubated with alpelisib exhibited d) reduced Tgfb1 (n = 6) and e) Eln expression (n = 3), while F) Lox was not altered compared to control-treated lipoma tissue slices. G) Pnpla2 and h) Gnai expression was not altered, in contrast i) Lipe (n = 3) and j) Ptger3 (n = 3) expression was diminished after incubation with alpelisib compared to control treatment. Opposite effects were seen regarding k) MAO A and l) MAO B expression levels after alpelisib compared to control medium incubation. Lipoma slices incubated with wortmannin showed no gene expression changes compared to controls. Data were normalized to the mean of solvent controls and presented as mean±sem (n = 2–8 independent experiments).

Figure 5.

Pten, Akt and Hsl protein was altered by alpelisib treatment in lipoma and epiWAT. a) Pten protein was significantly reduced in lipoma tissue compared to epiWAT, regardless of inhibitor treatment. b) Phospho(Ser473)Akt was higher in lipoma compared to epiWAT slices after 72 h in control medium. Alpelisib incubation for 72 h reduced phospho(Ser473)Akt both in lipoma and epiWAT, with a significant difference in lipoma slices compared to control treatment (n = 3, p = 0.0015). c) Total Akt was significantly reduced in both, lipoma and control tissue. d) Phospho(Ser660)Hsl levels were increased in alpelisib-treated lipoma and epiWAT tissue, with a significant increase in lipoma tissue compared to control-treated phospho(Ser660)Hsl and epiWAT. e) Total Hsl abundance tended to be reduced in lipoma compared to epiWAT. Wortmannin treatment did not induce significant differences in lipoma or epiWAT tissue compared to control treatment. Results from densitometric analyses of 2–4 independent experiments are shown as mean±sd with a representative Western blot image for each graph.

Figure 6.

Lipolysis in human lipoma adipocytes is not influenced by alpelisib. a) Glycerol release and b) HSL phosphorylation in human lipoma adipocytes was significantly increased after isoproterenol incubation and remained unaltered following alpelisib incubation. Three out of nine human lipoma samples showed an increased glycerol release and corresponding HSL phosphorylation after treatment with alpelisib. Data were normalized to solvent control treated values for each experiment shown as line at 1 and presented as mean±sd of 9 independent experiments.