Search

Figures S1-S2-S3 + Suppl. Fig. legends + Suppl. Methods

Pathway Enrichment_RNASeq

Proteomics_CRPC

SLFN5_binding sites

RNASeq_SLFN5 KO

ATF4_binding sites

List of antibodies

Androgen deprivation therapy (ADT) is the standard of care for treatment of nonresectable prostate cancer. Despite high treatment efficiency, most patients ultimately develop lethal castration-resistant prostate cancer (CRPC). In this study, we performed a comparative proteomic analysis of three in vivo, androgen receptor (AR)-responsive orthograft models of matched hormone-naïve prostate cancer and CRPC. Differential proteomic analysis revealed that distinct molecular mechanisms, including amino acid (AA) and fatty acid metabolism, are involved in the response to ADT in the different models. Despite this heterogeneity, Schlafen family member 5 (SLFN5) was identified as an AR-regulated protein in CRPC. SLFN5 expression was high in CRPC tumors and correlated with poor patient outcome. In vivo, SLFN5 depletion strongly impaired tumor growth in castrated conditions. Mechanistically, SLFN5 interacted with ATF4 and regulated the expression of LAT1, an essential AA transporter. Consequently, SLFN5 depletion in CRPC cells decreased intracellular levels of essential AA and impaired mTORC1 signaling in a LAT1-dependent manner. These results confirm that these orthograft models recapitulate the high degree of heterogeneity observed in patients with CRPC and further highlight SLFN5 as a clinically relevant target for CRPC.Significance:This study identifies SLFN5 as a novel regulator of the LAT1 amino acid transporter and an essential contributor to mTORC1 activity in castration-resistant prostate cancer.

List of primers

Significantly down-reg genes in SLFN5 KO cells and tumours

Androgen deprivation therapy (ADT) is the standard of care for treatment of nonresectable prostate cancer. Despite high treatment efficiency, most patients ultimately develop lethal castration-resistant prostate cancer (CRPC). In this study, we performed a comparative proteomic analysis of three in vivo, androgen receptor (AR)-responsive orthograft models of matched hormone-naïve prostate cancer and CRPC. Differential proteomic analysis revealed that distinct molecular mechanisms, including amino acid (AA) and fatty acid metabolism, are involved in the response to ADT in the different models. Despite this heterogeneity, Schlafen family member 5 (SLFN5) was identified as an AR-regulated protein in CRPC. SLFN5 expression was high in CRPC tumors and correlated with poor patient outcome. In vivo, SLFN5 depletion strongly impaired tumor growth in castrated conditions. Mechanistically, SLFN5 interacted with ATF4 and regulated the expression of LAT1, an essential AA transporter. Consequently, SLFN5 depletion in CRPC cells decreased intracellular levels of essential AA and impaired mTORC1 signaling in a LAT1-dependent manner. These results confirm that these orthograft models recapitulate the high degree of heterogeneity observed in patients with CRPC and further highlight SLFN5 as a clinically relevant target for CRPC. Significance: This study identifies SLFN5 as a novel regulator of the LAT1 amino acid transporter and an essential contributor to mTORC1 activity in castration-resistant prostate cancer.

Background Prostate cancer is highly prevalent worldwide. Androgen deprivation therapy (ADT) remains the treatment of choice for incurable prostate cancer, but majority of patients develop disease recurrence following ADT. There is therefore an urgent need for early detection of treatment resistance. Methods Isogenic androgen-responsive (CWR22Res) and castration-resistant (22Rv1) human prostate cancer cells were implanted into the anterior lobes of the prostate in CD-1 Nu mice to generate prostate orthografts. Castrated mice bearing CWR22Res and 22Rv1 orthografts mimic clinical prostate cancer following acute and chronic ADT, respectively. 18F-Fluciclovine (1-amino-3-fluorocyclobutane-1-carboxylic acid) with a radiochemical purity of > 99% was produced on a FASTlab synthesiser. Ki67 staining in endpoint orthografts was studied. Western blot, quantitative RT-PCR and next-generation sequencing transcriptomic analyses were performed to assess the expression levels of amino acid transporters (including LAT1 and ASCT2, which have been implicated for Fluciclovine uptake). Longitudinal metabolic imaging with 18F-Fluciclovine-based positron emission tomography (PET) was performed to study tumour response following acute and chronic ADT. Results Both immunohistochemistry analysis of endpoint prostate tumours and longitudinal 18F-Fluciclovine imaging revealed tumour heterogeneity, particularly following ADT, with in vivo 18F-Fluciclovine uptake correlating to viable cancer cells in both androgen-proficient and castrated environment. Highlighting tumour subpopulation following ADT, both SUVpeak and coefficient of variation (CoV) values of 18F-Fluciclovine uptake are consistent with tumour heterogeneity revealed by immunohistochemistry. We studied the expression of amino acid transporters (AATs) for 18F-Fluciclovine, namely LAT1 (SLC7A5 and SLC3A2) and ASCT2 (SLC1A5). SLC7A5 and SLC3A2 were expressed at relatively high levels in 22Rv1 castration-resistant orthografts following chronic ADT (modelling clinical castration-resistant disease), while SLC1A5 was preferentially expression in CWR22Res tumours following acute ADT. Additional AATs such as SLC43A2 (LAT4) were shown to be upregulated following chronic ADT by transcriptomic analysis; their role in Fluciclovine uptake warrants investigation. Conclusion We studied in vivo 18F-Fluciclovine uptake in human prostate cancer orthograft models following acute and chronic ADT. 18F-Fluciclovine uptakes highlight tumour heterogeneity that may explain castration resistance and can be exploited as a clinical biomarker.

Androgen-deprivation therapy (ADT) is the standard of care for the treatment of non-resectable prostate cancer (PCa). Despite high treatment efficiency, most patients ultimately develop lethal castration-resistant prostate cancer (CRPC). In this study, we perform a comparative proteomic analysis of three in vivo, androgen receptor (AR)–driven, orthograft models of CRPC. Differential proteomic analysis reveals that distinct molecular mechanisms, including amino acid (AA) and fatty acid (FA) metabolism, are involved in the response to ADT between the different models. Despite this heterogeneity, we identify SLFN5 as an AR-regulated biomarker in CRPC. SLFN5 expression is high in CRPC tumours and correlates with poor patient outcome. In vivo, SLFN5 depletion strongly impairs tumour growth in castrated condition. Mechanistically, SLFN5 interacts with ATF4 and regulates the expression of LAT1, an essential AA transporter. Consequently, SLFN5 depletion in CRPC cells decreases intracellular levels of essential AA and impairs mTORC1 signalling in a LAT1-dependent manner.

Androgen-deprivation therapy (ADT) is the standard of care for the treatment of non-resectable prostate cancer (PCa). Despite high treatment efficiency, most patients ultimately develop lethal castration-resistant prostate cancer (CRPC). In this study, we perform a comparative proteomic analysis of three in vivo, androgen receptor (AR)–driven, orthograft models of CRPC. Differential proteomic analysis reveals that distinct molecular mechanisms, including amino acid (AA) and fatty acid (FA) metabolism, are involved in the response to ADT between the different models. Despite this heterogeneity, we identify SLFN5 as an AR-regulated biomarker in CRPC. SLFN5 expression is high in CRPC tumours and correlates with poor patient outcome. In vivo, SLFN5 depletion strongly impairs tumour growth in castrated condition. Mechanistically, SLFN5 interacts with ATF4 and regulates the expression of LAT1, an essential AA transporter. Consequently, SLFN5 depletion in CRPC cells decreases intracellular levels of essential AA and impairs mTORC1 signalling in a LAT1-dependent manner.

Based on a molecular classification of prostate cancer using gene expression pathway signatures, we derived a set of 48 genes in critical pathways that significantly predicts clinical outcome in all tested patient cohorts. We tested these genes in a functional genomics screen in a panel of three prostate cancer cell lines (LNCaP, PC3, DU145), using RNA interference. The screen revealed several genes whose knockdown caused strong growth inhibition in all cell lines. Additionally, we tested the gene set in the presence of docetaxel to see whether any gene exhibited additive or synergistic effects with the drug. We observed a strong synergistic effect between DLGAP5 knockdown and docetaxel in the androgen-sensitive line LNCaP, but not in the two other androgen-independent lines. We then tested whether this effect was connected to androgen pathways and found that knockdown of the androgen receptor by si-RNA attenuated the synergy significantly. Similarly, androgen desensitized LNCaP-AI cells had a higher IC50 to docetaxel and did not exhibit the synergistic interaction. Short-term exposure to enzalutamide did not significantly alter the behaviour of parental LNCaP cells. An immunofluorescence analysis in LNCaP cells suggests that under the double insult of DLGAP5 knockdown and docetaxel, cells predominantly arrest in metaphase. In contrast, the knockdown of the androgen receptor by siRNA appears to assist cells to progress through metaphase in to anaphase, even in the presence of docetaxel. Our data suggest that DLGAP5 has a unique function in stabilizing spindle formation and surviving microtubule assault from docetaxel, in an androgen-regulated cell cycle system.

PPFIA1 (liprin α1) is a member of the liprin (LAR protein-tyrosine phosphatase-interacting protein) family and is located at the 11q13 amplification region, which is one of the most common amplicons in multiple epithelial cancers. In squamous cell carcinoma of head and neck (HNSCC), this amplification occurs in 30-50% of the cases. In epithelial cancers, the size and structure of 11q13 amplicon is complex and consists of multiple amplicon cores that differ between different tumor types. In HNSCC, the amplicon core contains 10 genes that are amplified in cases with 11q13 amplification. This region contains well-established oncogenes, such as CCND1 and CTTN (cortactin), yet the role of several genes in this region is unknown. Our aim is to study the role of PPFIA1, which has recently been found to be involved in cell motility, modeling of the cell edge dynamics and axon guidance. PPFIA1 is amplified and overexpressed in several epithelial cancers, but the role of it is controversial. PPFIA1 is required for breast and cervical cancer cell invasion, but in contrast depletion of it results in increased invasion of HNSCC cells. We have overexpressed liprin α1 and cortactin in HNSCC cells with no 11q13 amplification and related phenotypes have been analysed in in vitro assays. Migration assays using matrigel coated inserts have shown that overexpression of cortactin increases invasive properties of the HNSCC cells with no 11q13 amplicon whereas liprin α1 has an opposite effect. We have also shown that liprin α1 localizes with cortactin in same vesicular structures, but they do not show physical interaction. Our aim is now to explore the molecular mechanisms behind these findings to assess the role of liprin α1 in vesicular transport and cell motility. We are also carrying out immunoprecipitation and mass spectrometry analyses to identify novel interaction partners for this protein. Our study will provide novel information on the potential specialized functions of liprin α1 in different epithelial cancers, which is likely to improve our understanding on the functional role of 11q13 amplification in malignant potential of cancer cells. Citation Format: Henna Pehkonen, Tatiana Lepikhova, Riku Louhimo, Nina Peitsaro, Misa Imai, Rafael Martinez, Juha Kleftsröm, Sampsa Hautaniemi, Outi Monni. Functional properties of PPFIA1, located at the 11q13 amplification region, in epithelial cancer cells. [abstract]. In: Proceedings of the 104th Annual Meeting of the American Association for Cancer Research; 2013 Apr 6-10; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2013;73(8 Suppl):Abstract nr 4933. doi:10.1158/1538-7445.AM2013-4933