The androgen receptor-lncRNASAT1-AKT-p15 axis mediates androgen-induced cellular senescence in prostate cancer cells.

The bipolar androgen therapy (BAT) to treat prostate cancer (PCa) includes cycles of supraphysiological androgen levels (SAL) under androgen-deprivation therapy (ADT). We showed previously that SAL induces cellular senescence in androgen-sensitive PCa cells and in ex vivo-treated patient PCa tumor samples. Here, we analyzed the underlying molecular pathway and reveal that SAL induces cellular senescence in both, castration-sensitive (CSPC) LNCaP and castration-resistant PCa (CRPC) C4-2 cells through the cell cycle inhibitor p15INK4b and increased phosphorylation of AKT. Treatment with the AKT inhibitor (AKTi) potently inhibited SAL-induced expression of p15INK4b and cellular senescence in both cell lines. Proximity-ligation assays (PLA) combined with high-resolution laser-scanning microscopy indicate that SAL promotes interaction of endogenous androgen receptor (AR) with AKT in the cytoplasm as well as in the nucleus detectable after three days. Transcriptome sequencing (RNA-seq) comparing the SAL-induced transcriptomes of LNCaP with C4-2 cells as well as with AKTi-treated cell transcriptomes revealed landscapes for cell senescence. Interestingly, one of the identified genes is the lncRNASAT1. SAL treatment of native patient tumor samples ex vivo upregulates lncRNASAT1. In PCa tumor tissues, lncRNASAT1 is downregulated compared with nontumor tissues of the same patients. Knockdown indicates that the lncRNASAT1 is crucial for SAL-induced cancer-cell senescence as an upstream factor for pAKT and for p15INK4b. Further, knockdown of lncRNASAT1 enhances cell proliferation by SAL, suggesting that lncRNASAT1 serves as a tumor suppressor at SAL. Interestingly, immunoprecipitation of AR detected lncRNASAT1 as an AR-interacting partner that regulates AR target-gene expression. Similarly, RNA-ChIP experiments revealed the interaction of AR with lncRNASAT1 on chromatin. Thus, we identified a novel AR-lncRNASAT1-AKT-p15INK4b signaling axis to mediate SAL-induced cellular senescence.

Antithetic hTERT Regulation by Androgens in Prostate Cancer Cells: hTERT Inhibition Is Mediated by the ING1 and ING2 Tumor Suppressors.

The human telomerase is a key factor during tumorigenesis in prostate cancer (PCa). The androgen receptor (AR) is a key drug target controlling PCa growth and regulates hTERT expression, but is described to either inhibit or to activate. Here, we reveal that androgens repress and activate hTERT expression in a concentration-dependent manner. Physiological low androgen levels activate, while, notably, supraphysiological androgen levels (SAL), used in bipolar androgen therapy (BAT), repress hTERT expression. We confirmed the SAL-mediated gene repression of hTERT in PCa cell lines, native human PCa samples derived from patients treated ex vivo, as well as in cancer spheroids derived from androgen-dependent or castration resistant PCa (CRPC) cells. Interestingly, chromatin immuno-precipitation (ChIP) combined with functional assays revealed a positive (pARE) and a negative androgen response element (nARE). The nARE was narrowed down to 63 bp in the hTERT core promoter region. AR and tumor suppressors, inhibitor of growth 1 and 2 (ING1 and ING2, respectively), are androgen-dependently recruited. Mechanistically, knockdown indicates that ING1 and ING2 mediate AR-regulated transrepression. Thus, our data suggest an oppositional, biphasic function of AR to control the hTERT expression, while the inhibition of hTERT by androgens is mediated by the AR co-repressors ING1 and ING2.

Mouse models for hereditary spastic paraplegia uncover a role of PI4K2A in autophagic lysosome reformation.

Hereditary spastic paraplegia (HSP) denotes genetically heterogeneous disorders characterized by leg spasticity due to degeneration of corticospinal axons. SPG11 and SPG15 have a similar clinical course and together are the most prevalent autosomal recessive HSP entity. The respective proteins play a role for macroautophagy/autophagy and autophagic lysosome reformation (ALR). Here, we report that spg11 and zfyve26 KO mice developed motor impairments within the same course of time. This correlated with enhanced accumulation of autofluorescent material in neurons and progressive neuron loss. In agreement with defective ALR, tubulation events were diminished in starved KO mouse embryonic fibroblasts (MEFs) and lysosomes decreased in neurons of KO brain sections. Confirming that both proteins act in the same molecular pathway, the pathologies were not aggravated upon simultaneous disruption of both. We further show that PI4K2A (phosphatidylinositol 4-kinase type 2 alpha), which phosphorylates phosphatidylinositol to phosphatidylinositol-4-phosphate (PtdIns4P), accumulated in autofluorescent deposits isolated from KO but not WT brains. Elevated PI4K2A abundance was already found at autolysosomes of neurons of presymptomatic KO mice. Immunolabelings further suggested higher levels of PtdIns4P at LAMP1-positive structures in starved KO MEFs. An increased association with LAMP1-positive structures was also observed for clathrin and DNM2/dynamin 2, which are important effectors of ALR recruited by phospholipids. Because PI4K2A overexpression impaired ALR, while its knockdown increased tubulation, we conclude that PI4K2A modulates phosphoinositide levels at autolysosomes and thus the recruitment of downstream effectors of ALR. Therefore, PI4K2A may play an important role in the pathogenesis of SPG11 and SPG15.Abbreviations: ALR: autophagic lysosome reformation; AP-5: adaptor protein complex 5; BFP: blue fluorescent protein; dKO: double knockout; EBSS: Earle's balanced salt solution; FBA: foot base angle; GFP: green fluorescent protein; HSP: hereditary spastic paraplegia; KO: knockout; LAMP1: lysosomal-associated membrane protein 1; MAP1LC3B/LC3: microtubule-associated protein 1 light chain 3 beta; MEF: mouse embryonic fibroblast; SQSTM1/p62: sequestosome 1; PI4K2A: phosphatidylinositol 4-kinase type 2 alpha; PtdIns3P: phosphatidylinositol-3-phosphate; PtdIns4P: phosphatidylinositol-4-phosphate; RFP: red fluorescent protein; SPG: spastic paraplegia gene; TGN: trans-Golgi network; WT: wild type.

A mouse model for SPG48 reveals a block of autophagic flux upon disruption of adaptor protein complex five.

Hereditary spastic paraplegia is a spastic gait disorder that arises from degeneration of corticospinal axons. The subtype SPG48 is associated with mutations in the zeta subunit of the adaptor protein complex five (AP5). AP5 function and the pathophysiology of SPG48 are only poorly understood. Here, we report an AP5 zeta knockout mouse, which shows an age-dependent degeneration of corticospinal axons. Our analysis of knockout fibroblasts supports a trafficking defect from late endosomes to the transGolgi network and reveals a structural defect of the Golgi. We further show that both autophagic flux and the recycling of lysosomes from autolysosomes were impaired in knockout cells. In vivo, we observe an increase of autophagosomes and autolysosomes and, at later stages, the accumulation of intracellular waste in neurons. Taken together, we propose that loss of AP5 function blocks autophagy and thus leads to the aberrant accumulation of autophagic cargo, which finally results in axon degeneration.