Lipids, either endogenously synthesized or exogenous, have been linked to human cancer. Here we found that PML is frequently co-deleted with PTEN in metastatic human prostate cancer (CaP). We demonstrated that conditional inactivation of Pml in the mouse prostate morphs indolent Pten-null tumors into lethal metastatic disease. We identified MAPK reactivation, subsequent hyperactivation of an aberrant SREBP prometastatic lipogenic program, and a distinctive lipidomic profile as key characteristic features of metastatic Pml and Pten double-null CaP. Furthermore, targeting SREBP in vivo by fatostatin blocked both tumor growth and distant metastasis. Importantly, a high-fat diet (HFD) induced lipid accumulation in prostate tumors and was sufficient to drive metastasis in a nonmetastatic Pten-null mouse model of CaP, and an SREBP signature was highly enriched in metastatic human CaP. Thus, our findings uncover a prometastatic lipogenic program and lend direct genetic and experimental support to the notion that a Western HFD can promote metastasis.
Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
We thank all the members of the laboratory of P.P.P. for critical comments, and L. Southwood and E. Stack for editing the manuscript. We are grateful to G. Augusto dos Santos for insightful discussions. We thank the BIDMC Histology Core facility and M. Yuan in the BIDMC Mass Spectrometry Core for help with the lipidomics experiments. M.C. was supported in part by a DOD Prostate Cancer Research Program (PCRP) Postdoctoral Training Award. This work was supported by NIH grants R01CA142780, R01CA142874 and R35CA197529 to P.P.P. This work was also partially supported by NIH grants P01CA120964 and R35CA197459 to J.M.A.
Integrated supplementary information
(a) Bar graph showing the percentage of co-deletion of PTEN with 58 high-confidence TSGs32 in the Grasso et al. dataset of the mCRPC samples14, respectively (4 out of 62 TSGs from the Walker et al gene list, data not available). The genes co-deleted with PTEN only in metastatic disease among the top 25 TSGs are highlighted in Red. (b,c) Bar graph showing the percentage of deletion of PTEN or PML (b), or PTEN and PML (c) in mCRPC samples from the Robinson et al. dataset25. (d) Representative homozygous or hemizygous focal PML deletion from the Robinson et al. dataset (38% (17/45) of PML deletion was focal)25. Copy number plots with x-axis representing chromosomal 15q and y-axis referring to copy number level. Red open circle indicates genomic position of PML. (e) Representative IHC staining of PTEN or PML showing examples of low, medium and high staining. Scale bar, 50 μm. (f) Table showing the significant correlation of co-loss of PTEN and PML protein expression during the disease progression. The number of cases in each expression category was listed together with Gleason score. (g-j) Overall survival curves for CaP patients after radical prostatectomy based on the expression of PML protein (g), the expression of PTEN protein (h), Gleason score (i), or pathologic stage (j). In f, Pearson’s chi-squared test was used to determine significance.
(a) Schematic map of the WT Pml locus (top), targeting vector (upper middle) and predicted targeted allele (lower middle) and floxed allele (bottom). The Pml genomic sequence was cloned and inserted into the pEZ-LOX-FRT-DT vector. Black triangles mark the location of loxP sites that were utilized to excise the exon 2. Blue triangles mark the location of FRT sites that were utilized to excise the neomycin resistant cassette. The probes for Southern blot analysis are indicated (5′ and 3′ probes). BamHI digestion of genomic DNA from targeted ES cells was use to distinguish WT and targeted allele. BamHI (B), ScaI (S), NotI (N). (b) Southern blot analysis of recombined ES cell clones after digestion with BamHI and hybridization with the 5’ probe (upper panel) and the neomycin (lower panel) probe. ES cell clones with corrected homologous recombination are highlighted in red. (c) Southern blot analysis of tail DNA from F2 mice after digestion with BamHI and hybridization with the 5′ probe (top), 3′ probe and neomycine probe (bottom). The mice with deletion of neomycin resistant cassette are highlighted in red.
Supplementary Figure 3 Pml loss drives MAPK reactivation and metastatic progression in Pten-null CaP.
(a) IHC staining for Pml in the VP tissues from WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 12 weeks of age. (b) H&E and IHC staining of the DLP tissues from Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 20 weeks of age. Note that tumors in Ptenpc−/−Pmlpc−/− mice acquired invasive feature. Invasiveness was confirmed by the absence of SM-α-actin staining along with high level of Ki67 staining in the cancer cells. Arrows indicate invasive cancer. (c) Higher magnification of Ptenpc−/−Pmlpc−/− tumors at 13 months of age showing predominate adenocarcinoma (arrows) in the presence of focal features of sarcomatoid carcinoma with high-grade pleomorphic spindle cells (arrowheads). (d) Immunoblot (IB) analysis of tissue lysates for Pml from a WT mouse. (e) H&E-stained low-grade PIN in the VP and DLP tissues from a Pmlpc−/− mouse at 12 months of age. Insets show crowding cells with large nuclei. (f) H&E and IHC staining of lumbar lymph node metastases from three Ptenpc−/−Pmlpc−/− mice. Arrows indicate metastases. (g) IHC staining for phosphor-ERK in the DLP tissues from three pairs of Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 12 weeks of age. (h) IB analysis of the DLP tissue lysates from WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 12 weeks of age. (i) IHC staining for Pml and phosphor-ERK in the DLP tissues from a Ptenpc−/−Pmlpc+/− mouse at 12 weeks of age. Arrows indicate areas with lower level of Pml, but higher level of p-ERK. Arrowheads indicate areas with higher level of Pml, but lower level of p-ERK. Scale bars in all panels, 50μm. Uncropped images in d and h are shown in Supplementary Fig. 7.
Supplementary Figure 4 Transcriptome and lipidomics profiling of WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− prostates.
(a) Representative H&E staining of DLP from WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 12 weeks of age. Scale bar, 50μm. (b) Heat map of the SREBP targets in the microarray analysis of prostate tissues from the three genotypes of mice. (c-f) GSEA enrichment plot for the targets of LXR, ChREBP, PPARγ, and USF. The up- to down-regulated genes from the ranked gene list were analysed with the GSEA algorithm for enrichment of all gene sets in MSigDB among WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− prostates. (g,h) The relative intensity of all the identifiable 35 lipid classes (g) or the 30 most abundant fatty acyl chains (h) in prostate tissues from the three genotypes of mice. (i) Heat map of the top 70 most regulated lipid ions in prostate tissues from the three genotypes of mice. (j) The validation of the expression changes of hypoxia-induced target genes by the qPCR among WT, Ptenpc−/− and Ptenpc−/−Pmlpc−/− prostates. Data shown in g, h and j are mean ± s.e.m.
Supplementary Figure 5 SREBP-dependent lipogenesis is critical for PML-loss-induced CaP growth and metastasis.
(a,b) Representative images and quantitation of migrated and invaded PC3 cells transfected with siRNA against PML or/and SREBP-1 (a), or LNCaP cells transfected siRNA against SREBP-2 (b), in the migration and invasion assays. CaP cells were transfected with control or indicated siRNA for 48 hrs. PC3 cells were then subjected to 24-hr migration and invasion assay, while LNCaP cells were subjected to 24-hr migration and 48-hr invasion assay. (c) H&E and IHC staining of metastases in the lumbar lymph node of two vehicle-treated Ptenpc−/−Pmlpc−/− mice. Arrows indicate metastases. In a and b, the results of one representative experiment are shown (n = 3). Data are from three independent cultures (4 fields per insert). Data shown are mean ± s.e.m. Student′s t-test (two-tailed) was used to determine significance. Scale bars in all panels, 50 μm.
Supplementary Figure 6 A HFD drives metastatic progression in mouse models of CaP and increases lipid abundance in prostate tumors.
(a,b) H&E and IHC staining of metastases in the lung of a representative Ptenpc−/−Pmlpc−/− mouse (a) or a Ptenpc−/− mouse (b). Arrows indicate metastases. (c) The survival analysis of Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice upon 3-month HFD feeding beginning at 12 months of age. (d,e) The relative intensity of all the identifiable 36 lipid classes (d) or the 30 most abundant fatty acyl chains (e) in prostate tissues from chow- or HFD- fed Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice. (f) The ORO staining of vehicle or dietary lipids treated PC3 cells. (g) Representative images and quantitation of migrated or invaded PC3 cells in the migration and invasion assay. PC3 cells were pretreated with BSA, 2% lipid mixture, BSA-conjugate palmitic acid or oleic acid for 7 days, then subjected to 24-hr migration and invasion assay. (h) The serum testosterone levels in chow- or HFD- fed Ptenpc−/− and Ptenpc−/−Pmlpc−/− mice at 14–15 months of age. In g, the results of one representative experiment are shown (n = 5). Data are from three independent cultures (4 fields per insert). Data shown in d, e, g and h are mean ± s.e.m. Student′s t-test (two-tailed) was used to determine significance. Scale bars in all panels, 50 μm.
Supplementary Figures 1–7 and Supplementary Tables 8, 9 and 13–16
Summary of TMA Results and Clinical Information
The Microarray Results
GO and GSEA analysis of Microarray Data.
The Nomenclature of Lipids
Lipidomics Data from Mouse Prostates at 12 Weeks of Age
Lipidomics Data from Chow- or HFD-fed Mouse Prostates