Abstract
Driver genes with a mutually exclusive mutation pattern across tumor genomes are thought to have overlapping roles in tumorigenesis. In contrast, we show here that mutually-exclusive prostate cancer driver alterations involving the ERG transcription factor and the ubiquitin ligase adaptor SPOP are synthetic sick. At the molecular level, the incompatible cancer pathways are driven by opposing functions in SPOP. ERG up-regulates wild type SPOP to dampen androgen receptor (AR) signaling and sustain ERG activity through degradation of the bromodomain histone reader ZMYND11. Conversely, SPOP-mutant tumors stabilize ZMYND11 to repress ERG-function and enable oncogenic androgen receptor signaling. This dichotomy regulates the response to therapeutic interventions in the AR pathway. While mutant SPOP renders tumor cells susceptible to androgen deprivation therapies, ERG promotes sensitivity to high-dose androgen therapy and pharmacological inhibition of wild type SPOP. More generally, these results define a distinct class of antagonistic cancer drivers and a blueprint toward their therapeutic exploitation.
Normal cells transform into cancer cells by the acquisition of genetic aberrations in so called driver genes. In some instances, the functional redundancy of mutations in different genes result in a mutually-exclusive mutation pattern across tumor genomes because one alteration is sufficient to activate the specific oncogenic pathway. Based on this assumption, bioinformatic tools have been generated to search for functional redundancy of mutated genes in larger cancer genome data sets1,2.
In prostate cancer, recurrent gene fusions involving the ERG transcription factor and point mutations in the ubiquitin ligase adaptor SPOP are two truncal mutations that are mutually exclusively distributed across tumor genomes (Fig. 1a and Supplementary Fig. 1a)3–7. The underlying cause for this exquisite pattern remains controversial. While earlier reports suggested a functional redundancy between mutant SPOP and ERG based on the finding that mutant SPOP stabilizes the ERG oncoprotein8,9, more recent studies challenge this view by showing descriptive evidence for divergence in tumorigenesis3,10.
a Distribution of genetic alterations in SPOP and ERG across 333 primary prostate cancers in TCGA database1. b 3D growth of mouse prostate epithelial organoids derived from C57BL/6 mice over-expressing the indicated SPOP and ERG species (bar represents 20 μm) (n=3, technical replicates). Representative bright field pictures and haematoxylin and eosin (H&E) stained sections are shown. c In vivo growth of VCaP xenografts over-expressing the indicated SPOP species in immune-compromised mice (each group, n=10). d Immunoblot of VCaP cells over-expressing the indicated SPOP species and corresponding quantification of the indicated protein levels depicted as heatmap. Protein expression changes were normalized to β-ACTIN and Control cell line. e Tumor growth kinetics of xenografts established from LuCaP-147 PDX (SPOP-Y83C) stably overexpressing ΔERG or Control vector (each group, n=10). Corresponding immunoblot and quantification depicted as heatmap. Protein expression changes were normalized to Vinculin (VCL) and Control cell line. f Dose-response curve of VCaP cells overexpressing the indicated SPOP species and treated with the ETS-inhibitor YK-4-279. All error bars, mean + s.e.m. P values were determined by one-way ANOVA (b) or two-way ANOVA (c,e,f) with multiple comparisons and adjusted using Benjamini-Hochberg post-test *P < 0.05, **P < 0.01, ***P < 0.001. Molecular weights are indicated in kilodaltons (kDa).
RESULTS
Activation of the ERG oncogene and missense mutation in SPOP are synthetic sick
To shed light into the functional relationship of these recurrent driver genes, we assessed the impact of SPOP mutations and ERG activation on the cellular growth of mouse prostate epithelial organoids. To do so, we first established and validated the organoids by the presence of multilayered structure with expression of CK5 and CK8 in basal and luminal cells, respectively (Supplementary Fig. 1b). In agreement with recent reports, lentiviral-transduced point mutants of SPOP (SPOP-Y87C, SPOP-W131G) or a truncated version of ERG, which typically results from gene fusion with androgen-regulated genes in prostate cancer (ΔERG, amino acids 33-486), promoted cell growth (Fig. 1b, Supplementary Fig. 1c)11–14. While SPOP mutant organoids displayed a round shape, the over-expression of ΔERG gave rise to characteristic finger-like protrusions. Surprisingly, the joint expression of both drivers considerably diminished cell growth and reduced finger-like protrusions, implying a synthetic sick relationship between the two genetic alterations. Cytological follow up analysis revealed reduced proliferation evidenced by reduced Ki-67 and increased p16, P-HP1y positivity and vacuolization of the cytoplasm compatible with senescence induction (Supplementary Fig. 1d).
We wondered if the observed synthetic sick relationship also applied to established cancer cells from advanced, castration-resistant metastatic disease. Forced expression of mutant SPOP (SPOP-Y87C, SPOP-W131G) promoted 3D growth of ERG fusion-negative LAPC-4 human prostate cancer cells (Supplementary Fig. 2a). The oncogenic effect was paralleled by an increase in the expression of the oncogenic transcription factors MYC and HOXB13 and a decrease in the cell cycle inhibitor p21 as seen also in an organoid line derived from SpopF133V-mutant transgenic mice (Supplementary Fig. 2a,b)11. In contrast, we observed the opposite phenotypic and molecular changes in VCaP human prostate cancer cells harboring the recurrent TMPRSS2-ERG fusion (Fig. 1c, d & Supplementary Fig. 2c, d). In this setting, mutant SPOP (SPOP-Y87C, -F102C, -W131G, -F133S) dramatically decreased the proliferation of cancer cells in culture and the growth of xenograft tumor models in vivo. Similarly, to the mouse prostate organoids, induction of senescence was evidenced by an increase of senescence-associated β-galactosidase (SA-β-gal) positive cells and upregulation of p21 and GDF15 protein levels. In line with this, the transfer of conditioned medium from VCaP cells expressing mutant SPOP (SPOP-Y87C, SPOP-W131G) reduced also the proliferation of parental VCaP cells, indicating a contribution of Senescence-Associated Secretory Phenotype (SASP) to the synthetic sick relationship (Fig. 1d and Supplementary Fig. 2f,j).
Conversely, forced expression of ΔERG significantly reduced the growth of SPOP-Y83C mutant LuCaP-147 patient-derived xenograft (PDX) cancer cells in vivo and in culture (Fig. 1e, Supplementary Fig. 3a, b)15, adding orthogonal support for a synthetic sick relationship between mutant SPOP and ΔERG in advanced prostate cancer. In addition, over-expression of MYC promoted cancer cell growth in both VCaP and LuCaP-147 cells (Supplementary Fig. 3c, d). The latter finding may suggest that the over-expression system per se is not the underlying cause of the synthetic sick relationship mentioned above.
Next, we wondered if genetic or pharmacologic suppression of ERG signaling may revert the growth suppressing function of mutant SPOP in VCaP cells. Indeed, knockdown of ERG by short-hairpin RNA interference decreased the growth of VCaP control cells and of cells over-expressing wild-type SPOP, while it promoted the growth of cells over-expressing SPOP-W131G (Supplementary Fig. 3e). In addition, low doses of the ETS inhibitor YK-4-279 promoted specifically the growth of VCaP cells over-expressing mutant SPOP (Fig. 1f). We noted a similar effect when VCaP cells were co-treated with a small molecule inhibitor of SPOP (Supplementary Fig. 3f)16. In aggregate, the data support an antagonistic relationship between oncogenic activation of ERG and a loss of SPOP function in prostate cancer cells.
Mutant SPOP-induced androgen receptor signaling antagonizes ERG activity
To assess the underlying molecular biology of the antagonistic relationship between ERG-fused and SPOP-mutants tumors, we interrogated the transcriptomes from the TCGA cohort to nominate differences across these tumor subtypes. Indeed, the unbiased principal component analysis (PCA) revealed major differences in the transcriptional output (Fig. 2a). The differences were maintained in castration-resistant prostate cancers (CRPC) from the (SU2C) cohort using a single-sample gene-set enrichment analysis approach (Fig. 2b). Furthermore, derived (PDX) models also retained analogue transcriptional differences, as demonstrated by different behavior shown by SPOP-Mutant (LuCaP-78, -147) and ERG-Fused (LuCaP-35, -23.1, VCaP cells) models (Fig. 2b).
a PCA-analysis based on RNA-Seq derived mRNA expression levels (TCGA cohort). ERG-fused (violet) and SPOP-mutant (green). Individuals were annotated into subtypes as described in Material and Methods. b Boxplots representing the transcriptional activity of SPOP integrated-signature (see Materials and Methods) applied to CRPC samples (SU2C-2019 cohort, left) and PDX-models. Scores are determined genes-signatures derived from primary prostate tumors (TCGA-cohort). ERG-fused samples are depicted in violet, SPOP-mutant samples are depicted in green. Samples not harboring SPOP mutations or ERG rearrangements are represented in grey. P-values were determined using Wilcoxon-test. c Boxplots depicting the amount of normalized reads per binding site across all the differentially bound (DB) regions (FDR = 0.05) resulting from the comparison between ERG-fused (violet) and SPOP-mutant (green) samples.
In SPOP-mutant prostate cancer, several dysregulated SPOP substrates (e.g. NCOA3, TRIM24, BET proteins) have been shown to boost the AR pathway leading ultimately to high levels of AR target genes (Supplementary Fig. 4a)3,11,17–24. In contrast, ERG-fused cancer cells express typically lower levels of AR target genes as illustrated by the widely adopted AR score (Supplementary Fig. 4a)3. To get more insights into this different behavior, we further performed differential expression analysis between the two tumor subtypes. Gene-set enrichment analysis resulted in a clear up regulation of the canonical androgen response pathway in SPOP-mutant versus ERG-fused tumors, as defined by the respective Hallmark gene-set, curated by the Molecular Signature Database (MSigDB) (Supplementary Fig. 4b). In line with the divergence of transcriptome profile identified in the PCA plot of Fig. 2a, a clear division between SPOP-mutant and ERG-fused tumors was also reported in their respective cistrome counterpart25. By re-analyzing ChIP-Seq data, we could determine that most differentially bound regions between both tumor types, are indeed characterized by increased AR-binding in SPOP-mutant tumors (Fig. 2c).
Next, we analyzed the transcriptome changes of the VCaP cells overexpressing SPOP mutants (SPOP-MTs; SPOP-Y87C, -F102C, -F133S). The unbiased hallmark analysis showed a dramatic increase in the androgen response, recapitulating the changes identified in primary prostate cancer (Supplementary Fig. 4b, 5a and 5b). Based on these results, we posited that differential levels of androgen receptor (AR) signaling in SPOP-mutant versus ERG-fused cancers might be at the root of the incompatibility between the driver events. Thus, we analyzed in particular AR- and ERG-related transcription in VCaP cells, and generated a custom signatures using ChIP-seq data and matched RNA-seq samples (Supplementary Table 1)25. As expected, SPOP-MTs increased the transcription of genes bound by AR and induced by its ligand dihydrotestosterone (DHT), whereas genes bound by AR and repressed by DHT were further reduced (Fig. 3a, Supplementary Fig. 5c, and Supplementary Table 1). Remarkably, we observed the opposite effect on genes bound only by ERG. Mutant SPOP downregulated ERG-induced genes (e.g. MYC) and upregulated ERG-repressed genes, respectively (Fig. 1d). In line with these findings, gene ontology analysis of AR-ERG co-bound gene signature in VCaP cells indicated that the most striking transcriptional changes were linked to cellular differentiation and cell cycle arrest that are directly induced by DHT and repressed by ERG (e.g. HOXA genes, CDKN1A/p21, Fig. 1d, Fig. 3b, Supplementary Fig. 5d). To reduce the number of genes falling within our custom signatures, we used a particularly restrictive approach and considered as co-bound only genes where AR and ERG binding sites were overlapping for at least 1bp. As a result, some genes (i.e. CDKN1A/p21) which are bound both by AR and ERG in their promoter region, but which bindings do not overlap, are not included in this category despite being bona fide co-bound targets.
a Gene-set enrichment analysis of VCaP cells overexpressing SPOP mutant (SPOP-MTs; SPOP-Y87C, -F102C, -W131G) versus SPOP-wild type (-WT), based on RNA-seq data. Experiments were performed using three replicates for each condition. Enrichments are determined on custom gene-sets of direct androgen receptor (AR) and ERG target genes (Supplemental Table 1). Enrichments and FDR-adjusted p-values are computed with Camera (pre-ranked) b Venn Diagram and heatmap depicting the expression of genes included in the custom gene-set of AR/ERG co-bound genes that are repressed by ERG and induced by DHT in VCaP cells overexpressing SPOP-MTs (SPOP-Y87C, F102C, W131G), SPOP-WT and vector Control. Genes (rows) and samples (columns) were clustered using Euclidean distance. Gene expression values were normalized using variance stabilizing transformation (vst) and subsequently scaled and centered by row prior of clustering. Columns represent average expression of three replicates for each condition. c, Two-dimensional network representing overlaps between the 10 most significantly enriched Hallmark and custom gene-sets, identified when comparing SPOP-MTs (SPOP-Y87C, F102C, W131G) to SPOP-wild type (-WT) overexpressing VCaP cells. Thickness of edges is proportional to the significance of the overlap of the connected nodes measured by Fisher test. Only edges with FDR value <0.05 are shown. Size of nodes is proportional to gene-set enrichment significance and equals to −10 x log10 (FDR). d, Heatmap representing gene-set activity stratified according to tumor subtype, derived from TCGA cohort. For each tumor group, the average value of single sample GSEA scores was considered. Values were scaled and referenced to samples that did not harbor any ETS-fusion (ERG, ETV1, and ETV4) or point mutations in SPOP1.
The dramatic upregulation of this gene set was paralleled by a downregulation of cell cycle genes (e.g. E2F and MYC targets), implying a direct link between the induction AR/ERG co-bound genes, the repression of ERG targets, cell differentiation and the synthetic sick relationship of ERG and mutant SPOP (Fig. 1d, Supplementary Fig. 5a-e). The relationship of AR- and ERG-related custom signatures to the hallmark gene sets are highlighted in Fig. 3c in a two-dimensional network. Moreover, independently generated signatures of senescence-associated transcripts were enriched in VCaP overexpressing SPOP mutants, further corroborating our data of a senescence-induced cell cycle arrest (Supplementary Fig. 5f)26,27.
Conversely, we assessed the consequence of ERG over-expression in LNCaP cells under low DHT levels where mutant SPOP triggers AR signaling and tumor growth (Supplementary Fig. 6a,b)17,21. Over-expression of ΔERG in this setting robustly reverted the induction of signatures related to cell proliferation (e.g. E2F and MYC targets) and AR signaling. Taken together, the data implies a reciprocal incompatibility of mutant SPOP induced AR signaling and the function of the ERG oncogene.
Next, we verified if corresponding transcriptional changes were found in clinical tissue samples. Indeed, ERG-regulated genes culled from VCaP cells were up-regulated in ERG-fused and down-regulated in SPOP-mutant primary tumors (Fig. 3d, Supplementary Fig. 4c)3. Importantly, the most striking changes between the two groups were found again in the AR/ERG co-bound gene set in primary prostate cancers (Supplementary Fig. 4d)3,6. The results underscore both the relevance of our cell culture-based data and highlight the transcriptional differences among ERG- and SPOP-driven tumors.
ZMYND11 is a de novo SPOP substrate
Using tandem mass tag (TMT)-based quantitative mass-spectrometry, we set out to search for SPOP substrates that may influence the activity of AR and ERG and thereby may cause to the synthetic sick relationship between mutant SPOP and ERG in VCaP cells overexpressing mutant SPOP (SPOP-MTs; SPOP-Y87C, -F102C, -W131G, Fig. 4a). Because recurrent loss-of-function SPOP mutants impair substrate ubiquitylation and proteasomal degradation, we searched for proteins which expression levels increase without concomitant increase in mRNA levels (Fig. 4b, Supplementary Fig. 7a). Overall, we noted a strong correlation of protein with mRNA expression changes with consistent changes of our AR and ERG custom signatures at the protein level (Fig. 4b, Supplementary Fig. 7b, c). In addition, we found a marked upregulation of the known SPOP substrate and AR activator TRIM24 at the protein level (Fig. 1d & 4a, Supplementary Table 2)20,21 and subsequently assessed if TRIM24 and more generally AR is implicated in the synthetic sick relationship between mutant SPOP and ERG. Indeed, knockdown of TRIM24 by two short hairpin RNAs partially reverted the growth inhibition mediated by mutant-SPOP in VCaP cells and reduced AR signaling (Supplementary Fig. 7d-f), while over-expression of AR was sufficient to decrease cellular growth (Supplementary Fig. 7g, h).
a Schematic illustration showing the design of the proteomics experiments. Tandem Mass Tag (TMT)-based quantitative mass-spectrometry (n=2, biological replicates) was used in VCaP cells overexpressing Control vector (Control), SPOP-WT, or three different SPOP mutants (SPOP-Y87C, SPOP-F102C and SPOP-W131G). b Scatter plot comparing transcriptomic and proteomic derived fold changes resulting from the comparison between SPOP-MTs (average across SPOP-Y87C, -F102C, -W131G) and SPOP-WT VCaP cells. (n=3 biological replicates for transcriptome, n=2 biological replicates for proteome). Genes belonging the DHT-Induced/ERG-Repressed gene-signature (AR+ERG co-bound) are highlighted in red. TRIM24 and ZMYND11 are the most upregulated proteins without changes at mRNA levels and are highlighted in green. CDKN1A (p21) upregulated at both mRNA and protein levels, and MYC downregulated at both mRNA and protein levels, are highlighted in black. c Over-expression of HA-ZMYND11 and SPOP-WT in 293T cells and subsequent expression analysis of the indicated proteins by immunoblotting. d Whole cell extracts (WCE) of 293T cells over-expressing HA-ZMYND11-WT and different SPOP species and corresponding anti-HA-immunoprecipitation (HA-IP). Expression of the indicated proteins was analyzed by immunoblotting. e Domain structure of ZMYND11 with indicated SPOP-degron and ubiquitin sites. f Forced expression of SPOP-WT together with HA-ZMYND11-WT or two degron deficient mutants (DMT1 & DMT2) in 293T cells. g In vivo ubiquitylation assay of HA-ZMYND11 in 293T cells. Cell were transiently transfected with the indicated constructs and histidine-tagged (his-tag), ubiquitylated proteins were pulled down using nickel beads. Ubiquitylated HA-tagged ZMYND11 was detected by immunoblotting. h Over-expression of HA-ZMYND11 and SPOP-Y87C in 293T cells and subsequent expression analysis of the indicated proteins by immunoblotting after proteasomal inhibition with MG132. i Whole cell extracts (WCE) and corresponding anti-HA-immunoprecipitation (HA-IP) of 293T cells over-expressing HA-ZMYND11-WT and different SPOP-MTs species as indicated. Expression of the indicated proteins was analyzed by immunoblotting. j In vivo ubiquitylation assay of HA-ZMYND11 in 293T cells. Cell were transiently transfected with the indicated constructs and histidine-tagged (his-tag), ubiquitylated proteins were pulled down using nickel beads. Ubiquitylated HA-tagged ZMYND11 was detected by immunoblotting. k Immunoblots of indicated proteins in VCaP, LNCaP and LAPC4 human prostate cancer cells overexpressing the indicated SPOP species. Molecular weights are indicated in kilodaltons (kDa).
The most striking upregulation was noted for the bromodomain histone reader ZMYND11 (Fig. 4b). In line with a SPOP substrate, wild type SPOP bound and decreased the expression of HA-ZYMND11 in a proteasome-dependent manner (Fig. 4c, d). We found two degron sequences that were required for efficient SPOP-mediated ubiquitylation and protein degradation (Fig. 4e-g, Supplementary Fig. 8a.). As expected, SPOP mutants failed to bind and adequately ubiquitylate HA-ZMNYD11-WT (Fig. 4h-j)8,9,18–20,22–24. Finally, we confirmed that expression of mutant SPOP prolonged the half-life of endogenous ZMYND11 in VCaP cells and upregulated ZMYND11 expression in other prostate cancer cells (Fig. 4k and Supplementary Fig. 8b).
ZMYND11 induces AR signaling pathway and represses ERG activity
Next, we assessed if ZMYND11 protein upregulation also contributed to the synthetic sick relationship. In support, forced expression of the degron-deficient variants of ZMYND11 (HA-ZMYND11-DMT1/DMT2) was sufficient to diminish the growth of VCaP cells (Fig. 5a, b, Supplementary Fig. 8c), while knockdown of ZMYND11 partially reverted the growth inhibition mediated by mutant SPOP (Fig. 5c).
a,b Immunoblot of indicated proteins (a) and corresponding 2D proliferation assay (b) of VCaP cancer cells overexpressing HA-ZMNYD11-WT and derived degron-deficient mutants (DMT1/2) (n=3). Correlation between cell viability and ZMYND11 protein expression changes (Prot. Exp. Changes), as quantified by immunoblot in the same cell lines. P values were calculated using Pearson rank correlation. c Fold-change cell viability of VCaP cancer cells over-expressing the indicated SPOP species with and without ZMYND11 knockdown using two different short hairpin RNAs, at day 16 (n=3). Protein expression of the indicated proteins was analyzed by immunoblotting. d Chord-diagram of transcriptionally regulated genes by either SPOP-MTs or HA-ZMYND11-DMT2 in VCaP cells (FDR<0.05). Strings, whose thickness is proportional to the number of shared elements, represent common genes between sets. e Gene-set enrichment analysis of VCaP cell overexpressing HA-ZMYND11-DMT2 compared to Control, based on RNA-seq data. Enrichments are performed on custom gene-sets of direct androgen receptor (AR) and ERG target genes. FDR-adjusted p-values are computed with Camera (pre-ranked). f Heatmap of ChIP-seq signals around TSS regions (+/− 4kb) at which ZMYND11 bindings were identified by peak calling procedure (Macs2) in VCaP cells overexpressing the indication constructs. g IGV-derived screenshots representing loglikelihood ratio of ZMYND11 bindings in mutant SPOP (SPOP-Y87C) vs wild-type SPOP over-expressing VCaP cells. Reported are MYC (up) and CDKN1A (bottom). All error bars, mean ± s.e.m. P values were determined by two-way ANOVA (b) or one-way ANOVA with multiple comparisons and adjusted using Benjamini-Hochberg post-test (c). **P < 0.01, ***P < 0.001. Molecular weights are indicated in kilodaltons (kDa).
We postulated that ZMYND11 up-regulation could contribute to the synthetic sick relationship by repressing the transcriptional activity of the ERG oncogene or enhancing AR signaling. To this end, expression changes induced by HA-ZMYND11-DMT2 largely overlapped with genes perturbed by mutant SPOP while the opposite was noted when ZMYND11 expression was reduced by RNA interference (Fig. 5d and Supplementary Fig. 8d). In comparison to mutant SPOP, AR and ERG target genes were similarly dysregulated by HA-ZMYND11-DMT2 (Fig. 5e). Because the PWWP domain of ZMYND11 has been involved in the regulation of transcription through its ability to bind H3K36me3 histone marks28, we tested the contribution of this domain to the overall transcriptional output. Indeed, the PCA of VCaP cells over-expressing either HA-ZMYND11-DMT2 or a PWWP domain deficient mutant (W294A) revealed a major contribution of this domain to the ZMYND11 induced transcriptional changes (Supplementary Fig. 8e).
Subsequently, we mapped the genomic occupancy of ZMYND11 in VCaP cells expressing the SPOP-Y87C mutant by chromatin immunoprecipitation sequencing (ChIP-seq) and found an enrichment of ZMYND11 binding sites at promoter regions controlling ERG-induced genes (e.g. MYC,) and AR/ERG co-bound genes (e.g. p21/CDKN1A) (Fig. 5f, g and Supplementary Fig. 9a-e). The data imply a critical enhancer function of ZMYND11 in boosting AR signaling and repressing ERG signaling downstream of mutant SPOP.
Wild type SPOP is required for ERG oncogenic function
We reasoned that ERG-driven tumors might require wild type SPOP to degrade ZMYND11 and thereby unlock the oncogenic function of ERG. In support, over-expression of wild type SPOP increased the 3D growth of mouse prostate epithelial organoids and VCaP cells only when ERG was over-expressed (Fig. 1b, Supplementary Fig. 2c, 10a, b). Remarkably, ERG-fused human tumor tissues displayed also the highest SPOP mRNAs levels (Fig. 6a). Thus, we wondered if ERG itself may directly upregulate SPOP transcription to support its own oncogenic activity. Indeed, mining ERG ChIP-seq data in VCaP cells revealed ERG bindings sites in the promoter region of SPOP (Supplementary Fig. 10c). Moreover, knockdown of ERG reduced SPOP protein levels in VCaP cells, while forced expression of a ΔERG led to the upregulation of SPOP mRNA and protein levels in PC3 cells (Fig. 6b, Supplementary Fig. 3e, and 10d).
a SPOP mRNA expression levels in 333 primary prostate cancer tissues stratified according to the indicated driver mutations1. Error bars, mean ± s.d. b SPOP mRNA and protein levels in response to forced expression of ΔERG in PC3 prostate cancer cells by qPCR and immunoblotting, respectively. Error bars, mean + s.e.m. (n=3). P values were determined by unpaired, two-tailed Student’s t-test. #P < 0.05; Control versus ΔERG for SPOP expression levels. ***P < 0.001; Control versus ΔERG for ERG expression levels. c Transwell Matrigel invasion assay of PC3 cells with forced expression of ΔERG and knockdown of SPOP using two different short hairpin RNAs. Protein expression of the indicated proteins was assessed in parallel by immunoblotting. Error bars, mean ± s.e.m. (n=3). d Transwell Matrigel invasion assay of PC3 cells with forced expression of ΔERG and HA-ZMYND11-DMT2 and corresponding immunoblot analysis. Error bars, mean ± s.e.m. (n=3). e Analysis of the ΔERG-and HA-ZMYND11-DMT2-induced transcriptional changes in the ERG target genes PLAU and PLAT. All error bars, mean ± s.e.m. P values were determined by one-way ANOVA with multiple comparisons and adjusted using Benjamini-Hochberg post-test (a,c,d,e). NS, not significant. **P < 0.01, ***P < 0.001. Molecular weights are indicated in kilodaltons (kDa).
We then asked if the elevated SPOP levels in the context of forced ΔERG expression have a functional impact on the oncogenic activity of ΔERG in the androgen-independent PC3 cells, in which ERG promotes tumor cell invasion29. Indeed, the reduction of SPOP levels by RNA interference reduced the ability of ΔERG to invade into matrigel (Fig. 6c). Similarly, knockdown of SPOP in VCaP cells reduced cell growth in 3D cell culture and impaired ERG-mediated gene transcription (Supplementary Fig. 10e, f). In accordance with the ability of mutant SPOP to repress the function of endogenous wild type SPOP in a dominant-negative manner, the over-expression of mutant SPOP (SPOP-Y87C, -F102C,-W131G, -F133S) phenocopied the effect of SPOP knockdown on ERG-mediated invasion in PC3 cells (Supplementary Fig. 10g, h). In agreement with the established repressive function of ZMYND11 on ERG, we found that over-expression of HA-ZMNYD11-DM2 was sufficient to repress ERG-induced invasion and established target genes in PC3 cells (Fig. 6d, e). Taken together, the data imply the existence of a positive feed-forward loop, in which ΔERG promotes the expression of SPOP to sustain its oncogenic activity.
ERG and mutant SPOP trigger different responses to therapeutic interventions
Based on the above-mentioned differences in tumorigenesis, we speculated that ERG or mutant SPOP could also trigger different therapeutic responses. In light of the dependency of ERG-driven tumors on wild type SPOP function, we hypothesized that ERG-fusion positive cells may be particularly sensitive to pharmacological inhibition of SPOP. We analyzed the response of the SPOP small molecule inhibitor compound 6b (SPOP-i) in ERG-fused, SPOP mutant and other prostate cancer cell lines and patient-derived xenograft models (PDX)16. The SPOP inhibitor increased the protein but not the mRNA levels of established SPOP substrates and ZMNYD11 while the related inactive analog compound 6c did not (Supplementary Fig. 11a-c). The latter did also not exert any activity in 3D culture models (Supplementary Fig. 11d). In agreement with our previous results, we found that ERG-fused cells (VCaP, LuCaP-23.1, -35) were more sensitive to SPOP-i than ERG-negative cells (22Rv1, LNCaP, PC3), while SPOP mutant cells (LuCaP-78, -147) were particularly insensitive in 3D culture models and in xenograft tumor models in vivo (Fig. 7a-f, Supplementary Fig. 11e). We further validated our results in the mouse prostate epithelial organoids and confirmed the increased sensitivity of ΔERG-expressing cells to SPOP inhibition in this isogenic system (Fig. 7g).
a SPOP inhibitor (SPOP-i, compound 6b) mediated 3D growth inhibition in methylcellulose in the indicated prostate cancer cell lines. b Tumor growth kinetics with (n =10) or without (vehicle; n = 10) SPOP-i treatment in xenografts established from LuCaP-147 (SPOP-Y83C) PDX cells. c Tumor growth kinetics with (n = 4) or without (vehicle; n = 4) SPOP-i treatment in xenografts established from LuCaP-78 (SPOP-W131G) PDX cells. d Tumor growth kinetics with (n = 11) or without (vehicle; n = 11) SPOP-i treatment in xenografts established from VCaP. e Tumor growth kinetics with (n = 8) or without (vehicle; n = 10) SPOP-i treatment in LuCaP-35 (ERG-positive) PDX. f Tumor growth kinetics with (n = 6) or without (vehicle; n = 8) SPOP-i treatment in LuCaP-23.1 (ERG-positive) PDX. All SPOP-i treatment initiated when tumors reached an average of 100mm3. g Dose-response curves to SPOP-i treatment of Mouse Prostate Organoids overexpressing ΔERG, SPOP-Y87C and Control vector. All error bars, mean + s.e.m. P values were determined by two-way ANOVA with multiple comparisons and adjusted using Benjamini-Hochberg post-test. NS, not significant. **P < 0.01, ***P < 0.001.
Given the notion that wild type SPOP dampens AR function in the context of ERG to sustain tumor growth, we asked if VCaP cells are particularly susceptible to increased DHT levels. Indeed, exposure to high-dose of testosterone in vivo or DHT in vitro induced similar molecular changes as for the over-expression of mutant SPOP and greatly suppressed the growth of ERG-fusion positive cells but not of SPOP mutant cells in vitro and in vivo (Fig. 1c, 8a-f and Supplementary Fig. 12a-e). Moreover, signatures of senescence-associated transcripts were also enriched in VCaP cells upon treatment with DHT, further corroborating our data of a senescence-induced cell cycle arrest (Supplementary Fig. 12f). Strikingly, the sensitivity to SPOP-i and to high testosterone in vivo correlated well with ERG protein expression levels in the respective ERG-fusion positive cell line and PDX model (Fig. 8g, h). The data suggests a therapeutic opportunity for SPOP inhibition or high-dose androgen therapy in prostate cancers that express high levels of ERG.
a Dose-response curves to DHT treatment of VCaP, LuCaP-35, LuCaP-78 and LuCaP-147 PDX cancer cells. Prior to DHT treatment, PDX were grown in standard media without DHT. VCaP were starved for 24h in CSS medium (RPMI + 10% charcoal-stripped serum). Cell viability was assessed after 2 weeks. b Tumor growth kinetics with (n = 10) or without (vehicle; n = 10) testosterone treatment in xenografts established from LuCaP-147 (SPOP-Y83C). c Tumor growth kinetics with (n = 4) or without (vehicle; n = 4) testosterone treatment in xenografts established from LuCaP-78 (SPOP-W131G) cells. d Tumor growth kinetics with (n = 6) or without (vehicle; n = 10) testosterone treatment in xenografts established from VCaP (ERG positive) cells. e Tumor growth kinetics with (n =12) or without (vehicle; n = 12) testosterone treatment in xenografts established from LuCaP-23.1 (ERG positive) cells. f Tumor growth kinetics with (n = 10) or without (vehicle; n = 10) testosterone treatment in xenografts established from LuCaP-35 (ERG positive) cells. g Sensitivity to Testosterone and SPOP-i treatment in xenograft and PDX models. LuCaP-23.1, LuCaP-35 and VCaP are ERG-positive cancer cells. LuCaP-147 and LuCaP-78 are SPOP mutant cancer cells (respectively SPOP-Y83C and SPOP-W131G). Growth inhibition is calculated using the last tumor measurements as shown in b-f and Fig 7.b-f. h Correlation of sensitivity to SPOP-i or testosterone treatment shown in Extended Data Fig 8.f-j and 9e-i, with ERG protein levels, as quantified by immunoblot, in PDX models and xenografts. P values were calculated using Pearson rank correlation. Corresponding immunoblot and quantification of AR and ERG protein levels depicted as a heatmap. Protein expression changes were normalized to GAPDH and LuCaP-78. All error bars, mean + s.e.m. P values were determined by two-way ANOVA with multiple comparisons and adjusted using Benjamini-Hochberg post-test (b-e) or by unpaired, two-tailed Student’s t-test (g), NS, not significant. **P < 0.01, ***P < 0.001. Molecular weights are indicated in kilodaltons (kDa).
Conversely, and because SPOP mutant cancers are driven predominantly by androgen signaling and consequently display high-level activation of AR-related transcripts in human tumor tissues, we speculated that these tumors may be particularly susceptible to androgen deprivation or anti-androgen therapies (ADT) (Supplementary Fig. 4c). Indeed, the prevalence of SPOP mutations in primary tumors -and tumors that had progressed after initial surgery or radiotherapy-is consistently higher as compared to tumors that had become resistant to subsequent ADT (also referred as castration-resistant prostate cancer, CRPC, Supplementary Fig. 13a). In line with the notion that this difference may be related to a better response of SPOP mutant tumors to ADT, SPOP mutant tumor display a trend towards better overall survival despite progressing faster after initial therapy (Fig. 9a,b). To functionally analyze the response of androgen deprivation or the anti-androgen enzalutamide, we chose to ectopically expressed different SPOP variants and ΔERG in the androgen-dependent human LAPC4 prostate cancer cells that are wild-type for both driver genes. In accordance with the clinical observation, the presence of mutant SPOP (SPOP-Y87C, SPOP-W131G) rendered LAPC4 cells more susceptible to either ADT or enzalutamide in comparison to cells expressing control vector (Fig. 9c, Supplementary Fig. 13b). In contrast, ΔERG rendered the same cells more resistant to enzalutamide. In line with the previous findings in VCaP and LuCaP-147 cells, ΔERG expression rendered LAPC4 cells susceptible to high levels of DHT, while mutant SPOP had the opposite effect (Supplementary Fig. 13b). Taken together, the different responses to established and experimental therapeutic modalities observed between mutant SPOP and ERG add further credence to their divergent roles of the AR pathway related to tumorigenesis.
a Progression-free survival of prostate cancer patients derived from the TCGA-cohort. Curves representing TMPRSS2-ERG rearranged and SPOP-mutant patients are indicated in violet and green, respectively. The area around the curves represents 80% confidence interval. The bar plot in the lower left corner indicates the percentage of SPOP-mutant tumors within all patients who were diagnosed with prostate cancer (DIAG) and within the individuals who developed a progression of the disease (PROG). b Overall survival of prostate cancer patients derived from the MSK-IMPACT cohort. Curves representing TMPRSS2-ERG rearranged and SPOP-mutant patients are indicated in violet and green, respectively. The area around the curves represents 80% confidence interval. The bar plot in the lower left corner indicates the percentage of SPOP-mutant tumors within all patients who were diagnosed with prostate cancer (DIAG), within individuals who developed a metastatic progression of the disease (PROG), and within individuals who developed castration-resistant prostate cancer (CRPC). P values for Kaplan-Meier curves were determined using log-rank test. c Enzalutamide sensitivity of LAPC4 cells overexpressing ΔERG or SPOP mutant species (Y87C, W131G). All error bars, mean + s.e.m. P values were determined by unpaired, two-tailed Student’s t-test (c), NS, not significant. *P < 0.05, **P < 0.01. Molecular weights are indicated in kilodaltons (kDa).
DISCUSSION
Although multiple studies over recent years have uncovered different genetically-defined subtypes of primary prostate cancer, their biological understanding and therapeutic implications remain largely unexplored territory. Here, we report two diametrically different paths toward tumorigenesis triggered by either highly recurrent missense mutation in SPOP or gene fusion involving the ERG oncogene. Importantly, wild type SPOP emerges as a critical component that enforces oncogenic ERG signaling in part through dampening AR activity, while mutant SPOP drives tumorigenesis through activation of AR signaling. Moreover, several studies have previously highlighted the importance of AR target genes in the context of SPOP mutants and ERG positive tumors 17,30,31. Based on the incompatibility between the two tumor subtypes, our work enabled the development of specific custom signatures related to AR and ERG transcript that are necessary to drive proliferation and tumorigenesis in the context of ERG-positive and SPOP-mutants tumors. In addition, we show that the bromodomain histone reader ZMYND11 is a SPOP substrate implicated downstream of SPOP in the opposing regulation of the ERG and AR pathway in the two tumor subtypes (Fig. 10). The AR and ERG pathways have been previously reported to have a partially antagonistic relationship31,32, further corroborating our findings.
Because activation of the androgen receptor by androgens represents a key lineage specific oncogenic pathway in prostate cancer, androgen deprivation/antagonization therapies (ADT) remain the uniform treatment modality up to this very day. That said, the responses to ADT are highly variable and may last from a few weeks up to many years. Here, we provide functional evidence that pre-existing prostate cancer founder mutations influence the treatment response. Most notably, SPOP mutations promote susceptibility to androgen deprivations therapies. In agreement with our findings, earlier reports have shown underrepresentation of SPOP mutant tumors in cohorts of castration-resistant disease and a more favorable response to the abiraterone and enzalutamide33,34.
Conversely, we show that the presence of the ERG oncogene increases the susceptibility of tumor cells to high-dose androgen therapy, while cells expressing mutant SPOP remain largely unaffected. This is of clinical interest because testosterone treatment of patients with advanced castration-resistant disease has recently shown to trigger anti-tumor responses in around one third of the patients35. It is tempting to speculate that these insights may help to discern responders from non-responders.
In addition, we provide evidence that the antagonistic relationship between mutant SPOP and ERG may be used towards the development of new therapeutic avenues. More specifically, we show that ERG-driven cancer cells are particularly sensitive to the inhibition of wild-type SPOP using recently developed small molecule inhibitors16. Our preclinical data suggests that SPOP inhibition may be effective in clinical settings where ERG is robustly expressed (e.g. neo-adjuvant setting or early metastatic disease).
More generally, our results identify another paradigm for antagonistic driver genes in prostate cancer that has recently emerged also for other cancer types 36–38. In analogy to prostate cancer, truncal point mutations in DNMT3A and gene fusions in PML-RARA are mutually exclusive drivers in acute myeloid leukemia (AML). Similarly to SPOP, intact DNMT3A has been found to be critical for PML-RARA-driven leukemia (Supplementary Fig. 14a, b) 39,40. Importantly, we demonstrate here for prostate cancer that the concept of antagonistic driver genes can be exploited to identify therapeutic opportunities.
Author Contributions
T.B. and J.P.T. originally developed the concept, further elaborated on it and designed the experiments together with G.E.T. and M.B.; T.B. and G.E.T. performed experiments and analyzed the data together with N.D.U., S.A.C., A.M., T.S., L.P.B., F.S., M.Z., V.C., Anna R., H.J., D.B., M.C., D.A. and R.G‥ T.B. and G.E.T. performed xenograft tumor experiments in immunodeficient mice. Z.D. and C.G.Y. provided SPOP 6b and SPOP 6c compounds. M.S., S.D. and S.M. performed immunohistochemical experiments and J.P.T the subsequent analysis. H.M. and P.S. provided prostate cancer samples. M.B., A.V. and Andrea R. analyzed genomic and RNA-Seq data. J.P.T, H.M., S.A.C., M. A. R., M.K.J., A.A., F.B., W.Z. and G.M.C provided funding and resources. J.P.T., T.B., G.E.T. and M.B. interpreted the data and wrote the paper. T.B., G.E.T. and M.B. contributed equally to this work.
Author Information
MAR is listed as a co-inventor on US and International patents in the diagnostic and therapeutic fields of ETS gene fusion prostate cancers (Harvard and University of Michigan) and SPOP mutations (Weill Cornell Medicine). JPT has received funding for the venue of scientific conferences from Astellas, MSD, and Janssen/Cilag. The remaining authors declare no competing financial interests. Correspondence and requests for materials should be addressed to J.P.T. (JeanP_Theurillat{at}ior.iosi.ch)
Methods
Cell culture, Transfection and Infection
VCaP, LNCaP, PC3, 22Rv1, HEK293 cells were purchased from ATCC. LAPC-4 were a gift from Prof. Helmut Klocker. VCaP and HEK293 were grown in DMEM with Glutamax (Gibco); LNCaP, PC3, 22Rv1, LAPC-4 in RPMI medium (Gibco); all were supplemented with 10% full bovine serium (FBS; Invitrogen), or 10% charcoal-stripped serum (CSS; One Shot Fetal Bovine Serum, Charcoal Strippped, Gibco) for androgen deprivation therapy response, and 1% Penicilin/Streptomycin. LuCaP-147 were grown in StemPro medium (hESC SFM StemPro, Gibco) with regular supplements. All cells were incubated at 37°C and 5% CO2 and routinely tested for mycoplasma contamination.
For stable knockdown experiments, cells were infected with pLKO-1 vectors (Sigma) and the following clones were used; SPOP: TRCN0000140431 (shSPOP_1) and TRCN000013911 (shSPOP_2); TRIM24: TRCN000021262 (shTRIM24_1) and TRCN0000195528 (shTRIM24_2); ERG: TRCN0000429354 (shERG_1) and TRCN0000432394 (shERG_2); ZMYND11: TRCN0000275479 (shZMYND11_1) and TRCN0000275542 (shZMYND11_2). After infection, cells were selected in the presence of puromycin (2 μg/ml).
For SPOP, ΔERG, HA-ZMYND11-WT, HA-ZMYND11-DMT1, HA-ZMYND11-DMT2, MYC and AR over-expression a derivate of the pLX304 vector was used throughout in which the CMV promoter has been exchanged to a PGK promoter and the blasticidin cassette left unchanged (ΔERG constructs) or exchanged by a puromycin resistance cassette (SPOP constructs) (pLX_TRC_307, available at Addgene as Plasmid 41392, pCW107). All ORFs were cloned into pLX_TRC_307 using Nhe1 and Mlu1. Tumors from PDX LuCaP-78, -147, -35,-23.1 were collected, dissociated and cultured as previosly described41.
Chemicals
MG-132 (M7449) and Cycloheximide (CHX, C4859) were purchased from Sigma and used at 20 μM and 100 μg/ml in all experiments, respectively. SPOP inhibitor (SPOP-i, compound 6b) and its inactive analog (compound 6c), were provided by the laboratory of. Yang (State Key Laboratory of Drug Research, Shanghai Institute of Materia Medica). DHT (5α-Dihydrotestosterone) was purchased from Sigma (D-073), MDV3100 (Enzalutamide) was purchased from APExBIO (A3003). YK-4-279 (ETS inhibitor) was purchased from Selleckchem. All chemicals were used at the indicated concentration.
Dose-response curves and cell-growth assays
Cells were seeded (between 1×103 and 1×104 per well) in a 96-well plate. Cells were subsequently treated with serial dilutions of DHT (in 10% CSS medium), or enzalutamide; SPOP inhibitor, ETS inhibitor to determine dose-response curves or were left untreated for cell-growth assays. Proliferation at corresponding time points was assessed by MTT (Methylthiazolyldiphenyl-tetrazolium bromide) assay according to the manufacturer’s recommendations (Sigma). For each time point, absorbance (OD, 590 nm) was measured in a microplate reader.
SA-β-galactosidase staining of VCaP cells
Senescence-associated-β-galactosidase (SA--β-gal) staining was performed using the Senescence β-Galactosidase Staining Kit (Cell Signaling, #9860) following the manufacturer’s instructions. To avoid false-positive staining, we adapted the β-Galactosidase Staining Solution to pH 7. Cells incubated with glycerol 70% were observed under a bright field microscope.
Matrigel Invasion assay
Invasion assay was performed as previously described42. Briefly, equal number of PC3 cells were seeded into 10cm dishes and starved with a medium without fetal bovine serum for 24 hr.; subsequently 1×105 cells were resuspended in 100 μl of starved medium and seeded onto the basement of a Boyden chamber (CLS3422; Sigma) coated with Matrigel. RPMI with 10% fetal bovine serum was added to the lower chamber. After 48hr, invaded cells were fixed with 10% of formalin and stained with crystal violet. Absorbance was measured at 560 nm.
Clonogenic assay in methylcellulose
Cells were seeded (between 5×103 and 1×104) in methylcellulose (Methocult H4100, StemCell Technologies) in triplicate. Cells were left untreated for cell-growth assay. For SPOP inhibitor assay, cells were treated with vehicle (0.1% DMSO) or drug (SPOP-i) at corresponding concentration. For androgen therapy, cells were treated with vehicle (0.01% Methanol) or DHT at corresponding concentration. Cells were incubated at 37°C and 5% CO2 for 7-28 days and colonies were stained with MTT solution at 37°C overnight and absorbance (OD, 590 nm) was measured in a microplate reader.
Mouse Prostate Organoid Generation and Experiments
Prostate tissue was extracted from euthanized mice, digested and seeded in matrigel as previously described43. To overexpress SPOP species and ΔERG genes, mouse prostate cells were virally infected by spinoculation for 1hr at 600g at 32 °C and selected with puromycin. For the ‘‘organoid formation assay’’ 1.5×104 single cells were plated per well onto 40μl of Matrigel on day 1 and organoids were grown in “revised human prostate organoids medium” as previously described41. Briefly, the medium included the following reagents: adDMEM/F12, glutaMAX (2mM), Pen-strep 100u/ml, HEPES (10mM), B27 (1X), EGF (5ng/ml), AB3-01 (500nM), Noggin (100ng/ml), R-Spondin 1 (500ng/ml), DHT (1nM), FGF-2 (5ng/ml), FGF-10 (10ng/ml), Prostaglandin E2 (1000nM). The number of formed organoids that reached 100 μM of diameter was counted on days 14 post plating with cellSens software (Olympus). For the Dose-Response experiment 1×104 mouse prostate cells were plated in 40μl of Matrigel and treated with vehicle (0.1% DMSO) or drug (SPOP-i) at indicated concentration for 7 days. Live/dead staining was performed using Calcein AM (final concentration 1uM) and Ethidium Homodimer I (EthD-1, final concentration (1.33uM) dissolved in the medium for 1 hour. GFP or RFP positive organoids were analyzed under a fluorescence microscope. The percentage of dead organoids (RFP positive) was normalized to the Etoposide-treated (positive control) organoids. The genetically engineered Mouse Prostate Organoids, derived from PbCre;R26F133V, were generated as previously described11.
Immunohistochemistry
Cytoblocks were prepared from the pellets of organoids by adding plasma and thrombin in order to obtain a solid matrix. Once solidified, the organoids were fixed in 10% formalin (Thermo Scientific, 5701) and embedded in paraffin as a normal tissue. Sections of 4 μm were used for IHC analyses and hematoxylin and eosin (H&E) staining (Diapath, C0303) and (Diapath, C0363) respectively. Once dried the sections were treated with OTTIX plus solution (Diapath, X0076) and OTTIX shaper solution (Diapath, X0096) to dewax and rehydrate the sections. Antigen retrieval was performed using pH 6 solutions at 98°C for 20 min. Successively the endogenous peroxidases and non-specific binding sites were blocked using 3% H2O2 (VWR chemicals, 23615.248) and Protein-Block solution (DAKO Agilent technologies, X0909) respectively, for 10 min. Sections were then stained for anti-p16 (ab211542, Abcam, 1:1200), anti-Ki67 (Clone SP6; Lab Vision Corporation #RT-9106-R7, RTU) anti-Phospho-HP1y (Ser83) Antibody (CST #2600, 1:200), anti-CK8 (ab,59400, Abcam), anti-CK5 (ab52635, Abcam). IHC analyses were performed using the Imagescope software.
In vivo experiments
All animal experiments were carried out in male athymic nude mice (Balb/c nu/nu, 6-8 weeks old), NSG mice (NOD Scid Gamma, 6-8 weeks old), and NRG (NOD Rag gamma, 6-8 weeks old) accordingly to protocol approved by the Swiss Veterinary Authority (No. TI-14-2014, TI-38-2018, TI-39-2018 and TI-42-2018). Patient-derived xenografts (PDX) LuCaP-147, -78, -35, -23 were provided by Eva Corey (University of Washington) and maintained as previously described44. 2×106 VCaP cells, 5×106 LuCaP-147, LuCaP-23.1, LuCaP-35 and LuCaP-78 were resuspended in 100 μl of PBS and Matrigel 1/1 and subcutaneously injected into both of the dorsal flanks of the mice. Tumor growth was recorded using digital caliper and tumor volumes were calculated using the formula (L x W 2) /2, where L=length and W=width of tumor. For the testosterone propionate (25mg/kg) and SPOP inhibitor (SPOP-i, 50mg/kg) treatment, the mice were grouped randomly and the treatment started when the mean tumor volume reached 100m3. Tumor volume and weight were measured 2 times per week. Testosterone propionate was resuspended first in ethanol (150mg/kg) and then in Corn oil (Sigma) at a final concentration of 25mg/kg. SPOP inhibitor was resuspended in Dulbecco’s phosphate buffered saline (PBS) at a final concentration of 50 mg/kg. At the end of the experiment, mice were euthanized, tumors extracted and weighted. Testosterone level was measured using the Human Testosterone ELISA Kit from Abcam (ab174569). In order to recapitulate the levels of supraphysiological testosterone administrated in clinical trials35, mice reaching at least 3 times the testosterone levels measured before the treatment initiated were included in the depicted data.
Antibodies, Immunoblotting, and Immunoprecipiation
Antibodies used in immunoblotting and immunoprecipitation assays were: anti-SPOP (ab81163, Abcam), anti-TRIM24 (Sc-271266, Santa Cruz), anti-ß-ACTIN (4967, Cell Signaling), anti-AR (Sc-7305, Santa Cruz), anti-GADPH (Sc-47724, Santa Cruz), anti-ERG (Sc-271048, Santa Cruz), anti-α-Tubulin (3873S, Cell Signaling), anti-ZMYND11 (NBP2-20960, Novus Biologicals), anti-HA (H3663, Sigma), anti-BRD2 (A302-583A, Bethyl Labs), anti-NCOA3 (2126, Cell Signaling), anti-DEK (610948, BDBioscience), anti-p21 (2947S, Cell Signaling), anti-c-MYC (5605S, Cell Signaling), anti-HOXB13 (Sc-28333, Santa Cruz), anti-PTEN (9559, cell signaling), anti-p21 (ab188224, Abcam), anti-HOXB13 (NBP2-43655, Novus biologicals), anti-GDF-15 (27455-1-AP, proteintech). All antibodies were employed at dilutions suggested by the manufacturers.
For immunoblotting, cells were washed with PBS and subsequently lysed in RIPA buffer (Sigma) and sonicated. Protein concentration was determined using the BCA reagent (ThermoFisher), same amounts of protein were separated by SDS-PAGE (Biorad) and transferred onto PVDF membrane (ThermoScientific). The membrane was incubated for one hour in 5% nonfat dry milk/TBS-T blocking buffer followed by incubation with the primary antibody overnight at 4°C. The membrane was washed with TBS-T followed by incubation with horseradish peroxidase-conjugated secondary antibody (Promega).
To detect interactions of SPOP and ZMYND11, cells were lysed in 1 % NP40 buffer (50mM Tris-HCl pH 7.4, 150 mM NaCl, 1 % NP40) with 2x protease inhibitor cocktail (Complete, Roche), sonicated, and 3 mg of lysate were incubated overnight with 2 μg of anti-HA-tag or control mouse IgG antibody (sc-2025, Santa Cruz Biotechnology) at 4 °C. Subsequently, antibodies were collected by 25 μl protein A/G magnetic beads (88803, Fisher Scientific) for 2h, followed by 2 washing steps with 1 % NP40 buffer. Proteins were eluted by addition of 1x SDS-sample buffer under reducing conditions at 95 °C for 5 min.
In Vivo Ubiquitylation Assay
293T cells were transiently transfected with indicated plasmids: pCW107-HA-ZMYND11-WT or HA-ZYMND11-DMT1/DMT2 (2 μg), pCW107-SPOP-WT or SPOP-MT (2 μg), CMV-8x Ubi-His (2 μg). 42 hours later, cells were treated with MG-132 (20μM) or DMSO for additional 7 hours. Cells were then washed with PBS and collected by centrifugation. Small amount of cells was lysed in RIPA buffer and the rest in Buffer C (6M guanidine – HCL, 0.1 M Na2HPO4/NaH2PO4, 10mM Imidazole, pH=8). The whole cells extract was sonicated and incubated with 60 μl of Ni-NTA agarose (Sigma) overnight at 4°C. Next, Ni-NTA beads were washed once with Buffer C, twice with Buffer D (1 volume of Buffer C: 3 volumes of Buffer E) and once with Buffer E (25 mMTris-HCL, 20 mM Imidazole, pH=6.8). Elution of bound proteins was processed by boiling in 1x SDS loading buffer containing 300 mM Imidazole. Samples were loaded, separated by SDS-PAGE, and detected by immunoblotting.
Gene Expression Studies
RNA was extracted using the RNeasy kit (Qiagen) and processed by Kapa SybrFAST one-Step qRT-PCR kit according to manufacturer’s instructions. qRT-PCR was undertaken on an Applied Biosystems StepOnePlus System. The target mRNA expression was quantified using ΔΔCt method and normalized to Actin expression. The following primers were used: SPOP, forward 5’-GAAATGGTGTTTGCGAGTAAACC-3’, reverse 5`-TACCTACGCTTCCAGTCTCTG-3’; ERG, forward 5’-TGTATGCCAGCATTTGTTTCTT-3’, reverse 5’-TTGCTGGTCTTGCCATTCCT-3’; β-ACTIN, forward 5’-AAGGAGCCCCACGAAAAAT-3’, reverse 5’-ACCGAACTTGCATTGATTCCAG-3’; PLAU, forward 5’-TACGGCTCTGAAGTCACCACCAAAAT-3’, reverse 5’-CCCCAGCTCACAATTCCAGTCAA-3’; PLAT, forward 5’-CACTGGGCCTGGGCAAACATA-3’, reverse 5’-CACGTCAGCCTGCGGTTCTTC-3’; TMPRSS2, forward 5’-CAGGAGTGTACGGGAATGTGATGGT-3’, reverse 5’-GATTAGCCGTCTGCCCTCATTTGT-3’; KLK2, forward 5’-CTGCCCATTGCCTAA AGAAG-3’, reverse 5’-GTAGAGCGGGTGTGGGAAG-3’; PSA forward 5’-GAGCACCCCTATCAACCCCCTATT -3’, reverse 5’-AGCAACCCTGGACCTCACACCTAA-3’; ZMYND11, forward 5’-ATGGCACGTTTAACAAAAAGACG-3’, reverse 5’-CGGTCAATGTTGGCAATCTGC-3’; BRD2, forward 5’-CTACGTAAGAAACCCCGGAAG-3’, reverse 5’-GCTTTTTCTCCAAAGCCAGTT-3; TRIM24, forward 5`-CAGCCACAAATGCCTAAGCAG-3’, reverse 5’-GTGTTGGGAACTTGGATAACTGG-3; NCOA3, forward 5’-AGACGGGAGCAGGAAAGTAAA-3’, reverse 5’-GTAAAAGCGGTCCTAAGGAGTC-3’; DEK, forward 5’-AACTGCTTTACAACAGGCCAG-3’, reverse 5’-ATGGTTTGCCAGAAGGCTTTG-3’.
RNA-Seq of VCaP, LNCaP and LuCaP cells
RNA sequencing for all experiments involving LuCaP xenografts, VCaP and LNCaP cells was performed at the Institute of Oncology Research using Next Ultra II Directional RNA Library Prep Kit for Illumina and sequenced on the Illumina NextSeq500 with single-end, 75 base pair long reads. The overall quality of sequencing reads was evaluated using a variety of tools, namely FastQC (Andrews S., 2010), RSeQC45, AfterQC46 and Qualimap47. Sequence alignments to the reference human genome (GRCh38) was performed using STAR 48 (v.2.5.2a). Gene-expression was quantified at gene level by using the comprehensive annotations made available by Gencode49. Specifically, we used v27 release of the Gene Transfer File (GTF). Raw-counts were further processed in the R Statistical environment and downstream differential expression analysis was performed using the DESeq250 pipeline.
Genes being expressed at very low levels were automatically filtered out through the Independent Filtering feature embedded in DESeq2 (alpha = 0.05). Differential-expression results were ranked according to the computed Wald-statistics values. Subsequently, gene-set enrichment testing was performed using Camera51 pre-ranked (inter-gene correlation equal to 0.1, parametric test procedure). Statistical enrichments were assessed for gene-sets belonging to the Hallmark collection, which is curated by the Molecular Signature DataBase52,53 (MSigDB), and for custom ERG and DHT-specific gene-signatures. All enrichments were corrected for multiple testing using Benjamini and Hochberg FDR adjusted p-value.
Identification of ERG and AR related gene signatures
We retrieved RNA-seq data from GEO Dataset GSE8365214 to identify transcriptional perturbations in VCaP cells following treatment with DHT or following silencing of ERG. To this purpose we completely reprocessed samples SRR3713255-57, SRR3713267-72 using STAR and DESeq2 as previously described for VCaP cells. In addition, to identify direct targets, we integrated information relative to AR and ERG chromatin binding sites, which we derived from GEO Dataset GSE2895025. To maximize the number of peaks and to reduce false negatives, we merged results of experiments performed at different time points, namely 2h and 18h after DHT exposure. De-multiplexed reads were aligned to hg38 release of the human reference genome using bwa-mem54 (0.7.15). MACS55 (v.2.1.0) was used to perform peak calling procedure using a cutoff FDR q-value of 0.01 and a mappable genome size optimized for hg38 equal to 2.9 gigabases. Downstream analysis was performed in R statistical environment. We identified binding sites overlapping promoters by using bedtools56.
Promoters were defined as DNA regions ranging from 1500 bp upstream to 500 bp downstream of Transcription Start Sites (TSSs). To discriminate between ERG- and AR-specific transcriptional responses we stratified genes into three main classes: genes whose promoter regions are bound by AR but not by ERG, genes whose promoters are bound by ERG but not by AR, and finally, genes whose promoters are co-bound by both AR and ERG. AR bound only genes were further subdivided into two sets, those being significantly (FDR<0.05) induced following DHT treatment and those being significantly repressed. A similar approach was applied to ERG bound only genes, where genes were subdivided into ERG-induced and ERG-repressed gene-sets, if they were respectively down or up-regulated following ERG silencing. To be more stringent in the definition of AR-specific and ERG-specific signatures, we excluded genes from the ERG-induced set that were also significantly up-regulated following DHT treatment, vice-versa we excluded ERG-repressed genes that were significantly down-regulated following DHT-treatment. The same criteria were applied for DHT-specific gene-sets. Finally, defined an additional gene-set (DHT-induced/ERG-repressed) consisting of genes being co-bound by AR and ERG in their promoter region, which were significantly up-regulated following DHT treatment but also significantly upregulated following ERG-silencing. All gene-sets are detailed in Supplementary Table 1. Overlap between custom derived gene-signatures and the most represented Hallmark’s gene-sets was assessed using GeneOverlap R package (Shen L, Sinai M, 2013). Two-dimensional network visualization was generated with Cytoscape. 57.
Gene-set testing and RNA-Seq data processing of clinical samples
Publicly available RNA-Seq data for primary prostate cancer were obtained from The Caner Genome Atlas3 (TCGA) database and retrieved from Genomics Data Commons (GDC) in form of gene-centric raw counts, using TCGAbiolinks package58. We selected individuals characterized by either SPOP or ERG fusion, and a third group defined as “others”, which includes all remaining samples, excluding those patients exhibiting any other ETS-rearrangement. Differential expression and gene-set enrichment between samples harboring ERG fusions and SPOP-mutations were performed using DESeq2 and Camera (pre-ranked) as previously described for prostate cancer cells. Single-sample gene-set enrichment analysis (GSVA59 package) was applied to measure, for each individual patient, the overall activity of the custom gene-sets that were previously generated in VCaP cells. Following differential expression analysis between ERG-rearranged and SPOP mutant primary tumors, we defined two gene-sets consisting of SPOP-upregulated (n = 443, log2FC >1, FDR<0.05) and ERG-upregulated (n = 359, log2FC >1, FDR<0.05) genes. PolyA+ RNASeq data for metastatic prostate cancer were obtained from SU2C cohort60. Normalized RPKM values, retrieved through cBioportal, were log transformed and patient’s categorization (SPOP/ERG/OTHER) was performed in the same manner as for primary tumors. To evaluate whether transcriptional differences occurring between ERG-rearranged and SPOP-mutant individuals were also conserved in CRPC setting, we quantified the above mentioned SPOP-upregulated/ERG-upregulated signatures in the SU2C 2019 cohort, using single-sample gene-set enrichment analysis. The obtained ssGSEA scores were scaled in a range between -1 and 1 (SPOP-Upregulated) and between 1 and -1 (ERG-upregulated, inverted). Subsequently we averaged these rescaled values in order to obtain an aggregate score.
Circular representation of interactions between gene-sets
Chord diagrams were generated using circlize61 package in R statistical environment. Strings, whose thickness is proportional to the number of shared elements, represent common genes between sets.
ZMYND11 ChIP-seq in VCaP cells
ChIP-seq using anti-ZMYND11 antibody (NBP2-20960, Novus Biologicals) was performed in VCaP cells, overexpressing either wild-type SPOP or mutant SPOP harboring Y87C point mutation. Briefly, to isolate chromatin, cells (120.000.000 per IP) were cross-linked using 1% Formaldehyde cross-link protein-DNA complexes and crosslinking was terminated by the addition of 1/10 volume 1.25 M glycine for 5 min at room temperature followed by cell lysis and sonication, resulting in an average chromatin fragment size of 200 bp. Samples lysis was performed as previously described using MNase enzyme 1000 gel units=1 uL62. After adding the MNase sonication buffer, the samples were sonicated for 30 cycles, 30 sec ON and 30 sec OFF at high voltage. ChIP and input DNA (50 ng) were used for indexed library preparation using NEBNext Ultra II DNA Library Prep kit and subjected to 75 bp single-end sequencing on the Illumina NextSeq500. All procedures were performed at the Institute of Oncology Research. De-multiplexed reads were aligned to hg38 release of the human reference genome using bwa-mem54 (0.7.15). MACS55 (v.2.1.0) was used to perform peak calling procedure using a cutoff FDR q-value of 0.01 and a mappable genome size optimized for hg38 equal to 2.9 gigabases. Downstream analysis was performed in R statistical environment. ChIPseeker63 was used to annotate peaks and to represent the distribution of ZMYND11 binding sites relative to Transcription Start Sites (TSSs). The R package chipenrich64 was subsequently used to determine enrichment or depletion of ZMYND11 peaks in regions surrounding TSSs of genes that are included in Hallmarks or custom gene-set collections. Surrounding regions were defined as ranging from 5kb upstream to 5kb downstream of their TSSs (locusdef = 5kb), which is in line with the overall behavior of ZMYND11 binding sites around TSSs (Supplementary Fig. 6f-g).
Identification of AR-binding sites in primary prostate cancer specimen
Publicly available ChIP-Seq data were retrieved from GSE1207383. ChIP-seq data were reprocessed as described for ZMYND11 samples. Differential binding affinity of AR between ERG-rearranged and SPOP-mutant tumors was performed using DiffBind (Stark R and Brown G, 2011).
Frequency of SPOP mutations across patients’ cohorts
We defined the percentage of SPOP-mutant and TMPRSS2-ERG positive tumors across different patients’ cohorts originating from multiple sources. Patients with primary/loco-regional prostate tumors were derived from TCGA and MSK-IMPACT Clinical Sequencing cohorts34. Patients with tumor-progression (non-castrate) were derived from MSK-IMPACT and TCGA cohorts, by including from the latter only individuals that showed tumor-progression based on survival information. Castration resistant prostate cancer patients were retrieved from MSK-IMPACT, Beltran et. al65 and from the SU2C60.datasets. Neuroendocrine prostate cancer samples were retrieved from the SU2C cohort (samples annotated with neuroendocrine features) and from Beltran et al65. Total number of SPOP-mutant and TMPRSS2-ERG tumors were determined based on the clinical annotations of the individual studies and integrated with fusion information from TCGA Fusion Gene Database (www.tumorfusions.org). Survival analysis was performed in R statistical environment using the TCGA and MSK-IMPACT clinical sequencing cohort.
Quantitative liquid-chromatography-mass spectrometry (LC-MS/MS)
In solution digestion VCaP cell pellets were lysed at 4 ºC in 8 M urea, 50 mM Tris-HCl pH 8.0, 150 mM NaCl, 1 mM EDTA, 2 μg/μl aprotinin (Sigma-Aldrich), 10 μg/μl leupeptin (Roche), and 1 mM phenylmethylsulfonyl fluoride (PMSF) (Sigma). Protein concentration was determined using a bicinchoninic acid (BCA) protein assay (Pierce). Proteins were reduced with 5 mM (DTT) for 45 min at room temperature (RT), followed by alkylation with 10 mM iodoacetamide for 30 min at room temperature in the dark. The urea concentration was reduced to 2 M using 50 mM Tris-HCl, pH 8. Samples were digested for 2 h at 25 ºC with endoproteinase Lys-C (Wako Laboratories) at an enzyme-to-substrate ratio of 1:50. Samples were subsequently digested overnight at 25 ºC with sequencing grade trypsin (Promega) at an enzyme-to-substrate ratio of 1:50. Following overnight digestion, samples were acidified to a final concentration of 1% formic acid.
Peptide samples were desalted on a 100 mg tC18 Sep-Pak SPE cartridge (Waters). Cartridges were conditioned with 1 mL of 100% MeCN, 1 mL of 50% MeCN/0.1% FA, and 4x with 1 mL of 0.1% TFA. The sample was loaded, and washed 3x with 1 mL of 0.1% TFA, 1x with 1 mL of 1% FA, and eluted 2x with 600 μl of 50% MeCN/0.1% FA.
TMT labeling of peptides
Peptides were labeled with TMT 10-plex isobaric mass tagging reagents (Thermo Fisher Scientific). Each TMT reagent was resuspended in 41 μL of MeCN. Peptides were resuspended in 100 μL of 50 mM HEPES and combined with TMT reagent. Samples were incubated at RT for 1 h while shaking. The TMT reaction was quenched with 8 μL of 5% hydroxylamine at RT for 15 min with shaking. TMT labeled samples were combined, dried to completion, reconstituted in 100 μL of 0.1% FA, and desalted on StageTips or 100 mg SepPak columns as described above.
Basic Reverse Phase (bRP) Fractionation
The TMT labeled samples were fractionated using offline high pH reversed-phase chromatography (bRP) as previously described [Mertins et al Nat Prot]. Samples were fractionated using Zorbax 300 Extend C18 column (4.6 × 250 mm, 300 Å, 5 μm, Agilent) on an Agilent 1100 series high-pressure liquid chromatography (HPLC) system. Samples were reconstituted in 900 μL of 4.5 mM ammonium formate (pH 10) in 2% (vol/vol) acetonitrile (MeCN) (bRP solvent A). Samples were injected with Solvent A at a flow rate of 1 mL/min and separated using a 96 min gradient. The gradient consisted of an initial increase to 16% solvent B (90% MeCN, 5 mM ammonium formate, pH 10), followed by 60 min linear gradient from 16% solvent B to 40% B and successive ramps to 44% and 60% at a flow rate of 1 mL/min. Fractions were collected in a 96-deep well plate (GE Healthcare) and pooled in a non-contiguous manner into final 24 proteome fractions. Pooled fractions were dried to completeness using a SpeedVac concentrator.
Liquid chromatography and mass spectrometry
Desalted peptides were resuspended in 3% MeCN/0.1% FA and analyzed by online nanoflow liquid chromatography tandem mass spectrometry (LC-MS/MS) using Q-Exactive plus mass spectrometer (Thermo Fisher Scientific) coupled on-line to a Proxeon Easy-nLC 1200 as previously described [Mertins et al Nature Protocols]. Briefly, 1 ug of each sample was loaded onto a microcapillary column (360 μm outer diameter × 75 μ m inner diameter) containing an integrated electrospray emitter tip (10 μm), packed to approximately 22 cm with ReproSil-Pur C18-AQ 1.9 μm beads (Dr. Maisch GmbH) and heated to 50 ºC. Samples were analyzed with 110 min LC-MS method. The 110 min method contained a mobile phase with a flow rate of 200 nL/min, comprised of 3% acetonitrile/0.1% formic acid (Solvent A) and 90% acetonitrile/0.1% formic acid (Solvent B), with the following gradient profile: (min:%B) 0:2; 1:6; 85:30; 94:60; 95:90; 100:90; 101:50; 110:50 (the last two steps at 500 nL/min flow rate). The Q-Exactive plus MS was operated in the data-dependent mode acquiring HCD MS/MS scans (r =35,000) after each MS1 scan (r = 70,000) on the 12 most abundant precursor ions using an MS1 target of 3 × 10^6 and an MS2 target of 5 × 10^4. The maximum ion time utilized for MS/MS scans was 120 ms; the HCD-normalized collision energy was set to 30; the dynamic exclusion time was set to 20 s, isotope exclusion function was enabled, and peptide match function was set to preferred. Charge exclusion was enabled for charge states that were unassigned, 1 and >6.
MS Data Analysis
All data were analyzed using Spectrum Mill software package v 6.1 pre-release (Agilent Technologies). Similar MS/MS spectra acquired on the same precursor m/z within +/− 60 s were merged. MS/MS spectra were excluded from searching if they were not within the precursor MH+ range of 750-4000 Da or if they failed the quality filter by not having a sequence tag length >0. MS/MS spectra were searched against UniProt human database. All spectra were allowed +/− 20 ppm mass tolerance for precursor and product ions, 30% minimum matched peak intensity, and “trypsin allow P” enzyme specificity with up to 4 missed cleavages. The fixed modifications were carbamidomethylation at cysteine, and TMT at N-termini and internal lysine residues. Variable modifications included oxidized methionine and N-terminal protein acetylation. Individual spectra were automatically designated as confidently assigned using the Spectrum Mill autovalidation module. Specifically, a target-decoy based false-discovery rate (FDR) scoring threshold criteria via a two-step auto threshold strategy at the spectral and protein levels was used. First, peptide mode was set to allow automatic variable range precursor mass filtering with score thresholds optimized to yield a spectral level FDR of 1 %. A protein polishing autovalidation was applied to further filter the peptide spectrum matches using a target protein-level FDR threshold of 0. Following autovalidation, a protein-protein comparison table was generated, which contained experimental ratios. For all experiments, non-human contaminants and reversed hits were removed. Furthermore, data were filtered to only consider proteins with 2 or more unique peptides and was median normalized.
Statistical analysis
GraphPad Prism version 8.3 (GraphPad Software) was used for statistical analysis. Data are depicted as mean + s.e.m. unless otherwise specified. The number of independent experiments or mice used is indicated in each figure legends. Unpaired Student’s t-test was used for comparisons between two groups, one-way analysis of variance (ANOVA) with multiple comparisons for two groups or more, and two-way ANOVA with multiple comparisons for repeated measurements. Multiple comparisons tests were corrected by controlling the False Discovery Rate (FDR) using Benjamini and Hochberg’s method‥ Correlation analyses were performed using Pearson correlation coefficients.
Data Availability
The original mass spectra have been deposited in the public proteomics repository MassIVE and are accessible at ftp://MSV000082915@massive.ucsd.edu when providing the dataset password: prostate. If requested, also provide the username: MSV000082915. This data will be made public upon acceptance of the manuscript.
RNA-Seq data generated have been deposited in the ArrayExpress database at EMBL-EBI and were assigned the accessions E-MTAB-7165, E-MTAB-7170, E-MTAB-7173. These data are accessible when providing the datasets username and password as follow:
Username: Reviewer_E-MTAB-7165 Password: 4sJqpicv
Username: Reviewer_E-MTAB-7170 Password: AAA65MMy
Username: Reviewer_E-MTAB-7173 Password: sIpcob0o.
CHiP-Seq data generated have been deposited in the ArrayExpress database at EMBL-EBI and were assigned the accessions E-MTAB-7174. These data are accessible when providing the dataset username and password as follow:
Username: Reviewer_E-MTAB-7174 Password: onwwwwni
All these data will be made public upon acceptance of the manuscript.
Acknowledgements
We thank the University of Washington and Eva Corey and Donna M. Peehl for providing the Patient-Derived Xenografts (PDX) models. We thank Martina Storz, Susanne Dettwiler, Carmen Gavrisan, and Christiane Mittmann for histology assistance. We thank Enrica-Mira Catò and all members of the IOR/IRB animal core facility for technical assistance and the animal work. In addition, we thank all member of the laboratory for scientific discussions. The results shown here are in whole or part based upon data generated by the TCGA Research Network: https://www.cancer.gov/tcga. J.P.T is funded by a Swiss National Science Foundation Professorship (PP00P3_150645 & PP00P3_179072) grant, and grants by the Swiss Cancer League, the Lega Ticinese contro il cancro, and the Fidinam Foundation. This work was also supported in part by grants from the National Cancer Institute (NCI) Clinical Proteomic Tumor Analysis Consortium grants NIH/NCI U24-CA210986 and NIH/NCI U01 CA214125 (to S.A.C.).
References
