Targeting the lysosome by an aminomethylated Riccardin D triggers DNA damage through cathepsin B‐mediated degradation of BRCA1

Abstract RD‐N, an aminomethylated derivative of riccardin D, is a lysosomotropic agent that can trigger lysosomal membrane permeabilization followed by cathepsin B (CTSB)‐dependent apoptosis in prostate cancer (PCa) cells, but the underlying mechanisms remain unknown. Here we show that RD‐N treatment drives CTSB translocation from the lysosomes to the nucleus where it promotes DNA damage by suppression of the breast cancer 1 protein (BRCA1). Inhibition of CTSB activity with its specific inhibitors, or by CTSB‐targeting siRNA or CTSB with enzyme‐negative domain attenuated activation of BRCA1 and DNA damage induced by RD‐N. Conversely, CTSB overexpression resulted in inhibition of BRCA1 and sensitized PCa cells to RD‐N‐induced cell death. Furthermore, RD‐N‐induced cell death was exacerbated in BRCA1‐deficient cancer cells. We also demonstrated that CTSB/BRCA1‐dependent DNA damage was critical for RD‐N, but not for etoposide, reinforcing the importance of CTSB/BRCA1 in RD‐N‐mediated cell death. In addition, RD‐N synergistically increased cell sensitivity to cisplatin, and this effect was more evidenced in BRCA1‐deficient cancer cells. This study reveals a novel molecular mechanism that RD‐N promotes CTSB‐dependent DNA damage by the suppression of BRCA1 in PCa cells, leading to the identification of a potential compound that target lysosomes for cancer treatment.


| INTRODUCTION
RD-N, an aminomethylated derivative of riccardin D, is reported to be a potential anti-cancer agent to induce cell death through lysosomal rupture. It was suggested that cathepsin B (CTSB) released from the lysosomes after RD-N treatment triggered apoptosis in prostate cancer (PCa) cells, 1 but the underlying molecular mechanisms responsible for the effects remained unknown.
Targeting lysosomes has great therapeutic potential in cancer, because it not only triggers apoptotic cell death pathways but also inhibits cytoprotective autophagy. 2,3 Lysosomal membrane permeabilization results in the release of lysosomal enzymes into the cytosol, which can initiate caspase-dependent or -independent cell death.
Cathepsin B is one of the key lysosomal cysteine proteases that play important roles in migration and invasion of human cancer cells. 4,5 On the other hand, cysteine cathepsins have been shown to mediate cancer cell apoptosis. 6,7 In addition to be localized in lysosomes, cysteine cathepsins and their splice variants are also detected and function in the nucleus, 8,9 plasma membrane 10,11 and extracellular milieu. 12 In addition to being as primary lysosomal protein recycling machine, cysteine cathepsins are involved in multiple physiological and pathological processes through regulation of protein stability, initiation of proteolytic cascade and fusion with plasma membrane. For example, CTSB, once released from the lysosomes into the cytosol, induces cleavage of pro-apoptotic factor Bid, which leads to cytochrome c release from mitochondria and ultimately caspase-dependent apoptosis. 6 However, CTSB-mediated cell death is also observed in a caspase-independent manner. 13,14 Despite considerable amounts of work published, the mechanisms underlying the activity of CTSB in the nucleus to induce cell death are largely unknown.
We have previously identified RD-N, an aminomethylated derivative of bisbibenzyls Riccardin D, as a potential antitumour agent that was able to cause lysosomal membrane permeabilization. 1 This study went a further step and showed that RD-N-induced cell death is significantly dependent on the translocation of CTSB to the nucleus after treatment. Pro-apoptotic effect of CTSB, particularly the active enzyme domain of CTSB, is associated with suppression of breast cancer 1 protein (BRCA1) activity in the nucleus. Impairment of BRCA1 by CTSB facilitated DNA damage and cell death in response to RD-N.

| Reagents
RD-N was the aminomethylated derivative of riccardin D and its structure was identified as reported previously. 1 The compound was dissolved in dimethyl sulfoxide (DMSO, Sigma-Aldrich, St. Louis, MO) at 10 mmol/L as stock solution. E64d and z-VAD-fmk were obtained from Enzo Life Sciences. CA074Me was acquired from Calbiochem. Z-RR-AMC was purchased from the EMD Chemicals. Propidium iodide (PI) was purchased from Sigma-Aldrich.

| Cell culture and treatments
Human PCa cell lines PC3, DU145 and LNCaP were obtained from the American Type Culture Collection (Manassas, VA). All the cell lines were cultured in RPMI-1640 (Hyclone, Logan, UT, USA) medium containing 10% foetal bovine serum (Gibco, Gaithersburg, MD, USA), 100 U/mL of penicillin and 100 μg/mL of streptomycin and maintained in a humidified incubator of 5% CO2 at 37°C. When growing cells reached approximately 50%-70% confluence, they were treated with RD-N or other chemicals as indicated. Vehicle treatment served as a control. Transient transfections were performed using Lipofectamine 2000 (Invitrogen). Prior to transfection, Lipofectamine (~2 μL/μg of DNA) was added to Optimum medium without foetal bovine serum. After a 5-minute incubation, the mixture was then added on the top of the DNA constructs. After a 20minute incubation at 37°C, the whole solution was added on the cells. GFP-tagged CTSB was a kind gift from Ted Hupp (University of Edinburgh).

| xCELLigence
Experiments were carried out using the RTCADP instrument (Roche, Germany) which was placed in a humidified incubator maintained at 37°C with 5% CO2. For time-dependent cell response profiling,

| Flow cytometry
We used Annexin V (0.1 mg/mL) for the assessment of PS exposure, and PI (0.5 mg/mL) for cell viability. Cell death was recorded in a FACScan cytometry (FACSCalibur, Becton Dickinson) in total population (10 000 cells/nuclei). For γH2AX detection, PC3 cells, untreated or RD-N-treated, were fixed in 4% paraformaldehyde, permeabilized with 0.1% Triton X-100, incubated with anti-γH2AX (Cell Signaling Technology) antibody and detected by flow cytometry (Becton Dickinson).
Following washing with TBST and incubating with peroxidase-conjugated appropriate secondary antibodies, immunoblot proteins were visualized by enhanced chemiluminescence detection system (Millipore) and exposed to X-ray films.

| Subcellular fractionation
Nuclear and cytoplasmic extracts were prepared by using the

| Neutral comet assays
To assess DNA double-strand breaks (DSBs), neutral comet assays were performed using CometSlide assay kits (Trevigen). Briefly, PCa cells were treated with RD-N (6 μmol/L) and were incubated at 37°C for 0-24 hours. Cells were embedded in agarose, lysed and subjected to neutral electrophoresis. Immediately before image analysis, cells were stained with SYBR Green and visualized under a fluorescence microscope (Olympus, Japan). Olive comet moment was calculated by multiplying the percentage of DNA in the tail by the displacement between the means of the head and tail distributions, as described. 15 We used the program CometScore software to calculate Olive Comet Moment. A total of 30 comets were analysed per sample in each experiment.

| CTSB activity
Cathepsin B activity was measured by using the fluorogenic substrate Z-RR-AMC from the EMD Chemicals following the manufacturer's instructions. Briefly, 10 6 cells were lysed in Lysis Buffer (100 mmol/L phosphate buffer, pH 6; 0.1% polyethylene glycol (PEG); 5 mmol/L DTT; 0.25% Triton X-100), substrates were added at 20 μmol/L final concentration in 100 μL Lysis Buffer in the presence or absence of inhibitors for CTSB (E64d, CA074Me). A total of 100 μg of protein extract was used per sample. Cleaved Z-RR-AMC substrate was detected by fluorescence reader (Exc: 380 nm; Emi: 460 nm).

| Immunofluorescence
Cells growing in coverslips were fixed for 10 minutes in ice-cold methanol/acetone (1:1), followed by three washes in PBS. After blocking in 3% BSA in PBS with 0.1% Triton X-100 for 20 minutes, cells were incubated with CTSB, γH2AX or p-BRCA1 antibodies overnight at 4°C, washed three times and incubated 1 hour at 37°C with secondary antibodies. After washing three times in PBS, cells were counterstained with 4',6-diamidino-2-phenylindole (DAPI) and coverslips mounted on slides. Fluorescence images were captured using a confocal microscopy (Carl Zeiss, Germany).

| Protein modelling
We used the known crystal structure of BRCA1 and CTSB for protein docking. Crystal structure of BRCA1 ring domain (PDB ID: 1JM7) 16 and BRCT domains (PDBID: 1JNX) 17 were docked to the structure of CTSB (PDB ID: 3K9M) 18 by ZDOCK. 19 Two sets of 2000 structure complexes were generated and ranked according to the ZRANK scoring function. 20

| Microscopy
To visualize chromatin condensation, we used Hoechst33342 or DAPI to stain DNA in the nuclei. Briefly, PC3 cells cultured on cover glasses were incubated with 5 μg/mL Hoechst33342 or DAPI for 15 minutes. The cells were then washed with PBS and nuclear fluorescence was detected using fluorescence microscope (Olympus).
Alternatively, apoptotic cells were identified using an in situ cell death detection TUNEL kit (Roche). The staining was performed according to manufacturer's instruction and observed using fluorescence microscope (Olympus).

| Immunohistochemical analysis
CTSB, p-BRCA1 and γH2AX protein expression were evaluated by immunohistochemistry on tumour sections using CTSB, p-BRCA1 and γH2AX-specific antibody. The sections were deparaffinized first.
For antigen retrieval, the sections were boiled in 10 mmol/L citrate buffer (pH 6.0) for 30 minutes, and endogenous peroxidase activity was blocked using 3% H2O2 for 5 minutes. The sections were then blocked in antibody diluent for 1 hour at room temperature. CTSB, p-BRCA1 and γH2AX antibodies were diluted 1:200 in blocking solution, and sections were incubated overnight at 4°C. After washing, the slides were incubated with anti-goat biotinylated antibody for 30 minutes at room temperature. Sections were counterstained with haematoxylin, dehydrated and permanently mounted.

| Statistical analysis
The values represent the mean ± SD for triplicate experiments. Statistical differences were assessed using an unpaired Student's t test and one-way ANOVA P < 0.05 was considered statistically significant.

| RD-N induces DNA damage in PCa cells
We first validated the pro-apoptotic effect of RD-N on PC3 cells using the xCELLigence system. As shown in Figure 1A Figure 2A and B, CTSB depletion was markedly rescued RD-N-mediated apoptosis in PC3 cells. The fraction of apoptotic cells was 23.4% at 24 hours compared to 42.4% in scramble siRNA-treated cells (Figure 2A,B). Also, CTSB silencing did not present similar amounts of chromatin condensation or TUNEL positivity to the levels of scramble siRNA-treated cells in response to RD-N ( Figure S2A,B). In addition, we re-introduced the CTSB into PC3 cells to confirm the role of CTSB in RD-N-mediated apoptosis. The results in Figure 2C showed that ectopic expression of CTSB exacerbated cell death induced by RD-N. These findings demonstrated that CTSB is required for RD-N-mediated chromatinolysis and cell death. We therefore examined the response of CTSB in cells treated with RD-N. As shown in Figure 2D

| CTSB is required for RD-N induced DNA damage
Since RD-N induces the permeabilization of lysosomes and releases CTSB to the nucleus, we are promoted to investigate whether CTSB translocation to nucleus has a causative effect on DNA damage. PC3 cells were pre-incubated with CTSB inhibitors (E64d and CA074Me) prior to RD-N treatment. As shown in Figure 3A, inhibition of CTSB significantly reduced the levels of phospho-Chk2, γH2AX in PC3 cells in the presence of RD-N but had minimal effect on the phospho-Chk1 and total H2AX, suggesting that ATM/ChK2 may be more important in RD-N-mediated effect which is consistent with the results in Figure 1F.
Similarly, inhibition of CTSB also attenuated the H2AX activation in DU145 and LNCaP cells exposed to RD-N ( Figure 3B). In the presence of specific protease inhibitors, we found that inhibition of the lysosomal protease CTSB (E64d or CA074Me), but not of caspases (z-VADfmk), strongly prevented formation of γH2AX foci in the nucleus (Figure 3C) and decreased H2AX phosphorylation ( Figure 3D). DNA strand-break formation also provided evidence that inhibition of CTSB greatly alleviated RD-N-induced DNA damage ( Figure 3F). The data demonstrated a key role for CTSB translocation to the nucleus in mediating the DNA damage induced by RD-N.

| Suppression of BRCA1 is essential for conferring CTSB-induced DNA damage in response to RD-N
It was noted that p-BRCA1 started to drop down at 12 hours and disappeared after 24 hours treatment, at which CTSB was evidenced  in the nucleus associated with the increases in PARP cleavage in response to RD-N ( Figures 1F and 2). We hypothesized that CTSB acts as a proteolysis enzyme to induce DNA damage through downregulation of BRCA1 in cells exposed to RD-N. The results in Figure 4A indicated that p-BRCA1 was observed after 12 hours treatment with RD-N, however, RD-N-induced BRCA1 phosphorylation was enhanced in the presence of CA074Me. Depletion of endogenous CTSB with specific targeting siRNA also significantly facilitated BRCA1 phosphorylation that was induced in PC3 cells after 12 hours treatment with RD-N ( Figure 4B). Immunofluorescence confocal microscopy further demonstrated that, in scramble siRNA cells, CTSB was confined to the lysosome, whereas p-BRCA1 was localized in the nucleus ( Figure 4C). Upon RD-N treatment, CTSB moved to the nucleus and associated with the decreased p-BRCA1 at 24 hours treatment ( Figure 4C). However, CTSB deficiency caused an accumulation of p-BRCA1 in the nucleus (Figure 4C), which attenuated H2AX phosphorylation exposed to RD-N ( Figure 4D). These findings supported that the migration of CTSB to the nucleus by RD-N was important for down-regulation of phosphor-BRCA1 and induction of DNA damage. Since CTSB could process many proteins to be degraded, 23,24 we hypothesized that suppression of p-BRCA1 ascribed to the degradation of BRCA1 by CTSB upon treatment with RD-N. As shown in Figure 4E, total BRCA1 protein expression was steadily induced until 4 hours by RD-N, and gradually faded thereafter. Accordingly, p-BRCA1 was time dependently accumulated, and impaired after 8 hours treatments ( Figure 4E), demonstrating the proteolysis ability of CTSB on BRCA1 degradation.
The ability of CTSB on BRCA1 was further examined in cells transfected with a CTSB expression plasmid. As shown in Figure 5A, p-BRCA1 was induced in the nucleus after 12 hours treatment which was consistent with the observation in Figure 1F, however, overexpression of CTSB significantly reduced the levels of BRCA1 and p-BRCA1 in cells exposed to RD-N ( Figure 5A). Immunofluorescence analysis confirmed that forced expression of CTSB markedly abol-  Figure 5A). Also, forced expression of CTSB without enzyme domain was able to alleviate the PARP cleavage and γH2AX expression when compared to that of wild type of CTSB ( Figure 5A). Immunofluorescence staining revealed that overexpression of CTSB markedly reduced phosphor-BRCA1 that was induced in cells by RD-N at 12 hours treatment ( Figure 5B), which was associated with an increase in DNA damage as indicated by elevated γH2AX foci ( Figure 5C). However, ΔCTSB lost its ability to reduce RD-N-activated BRCA1 which in turn attenuated DNA damage triggered by RD-N, compared to the CTSB plus RD-N under same conditions ( Figure 5B,C). These results strengthened the importance of proteolytic activity of CTSB in the suppression of BRCA1. To further determine whether nuclear localization of CTSB plays a role in the degradation of BRCA1, we constructed a mutant CTSB expression plasmid lacking nuclear localization signal (CTSB-ΔNLS). 26 Expression of CTSB-ΔNLS predominantly reduced its nuclear localization upon treatment with RD-N ( Figure 5E), importantly, the level of phosphor-BRCA1 was restored in cells when expression of CTSB-ΔNLS in the presence of RD-N ( Figure 5D). Also, forced expression of CTSB-ΔNLS without nuclear localization signal was able to reduce γH2AX expression when compared to that of wild type of CTSB ( Figure 5D). Finally, the available crystal structures of BRCA1 ring domain and BRCT domains (PDB ID: 1JNX) and CTSB (PDB ID: 3K9M) were used to predict the potential interaction sites. Docking images indicated that the BRCA1 ring domain was bound to the catalytic cysteine residue of CTSB, which strongly supported the observations that hydrolysis region of CTSB is required for reducing p-BRCA1 protein level ( Figure 5F). However, it is difficult to confirm the binding of the endogenous CTSB to p-BRCA1 by immuneprecipitation, probably due to the fast degradation of a substrate when an enzyme bound. Therefore, inhibition of BRCA1 by CTSB was important for RD-N-mediated DNA damage. The changes in CTSB, p-BRCA1 and γH2AX were further examined in animal tissue samples after treatment with RD-N. As shown in Figure 5H

| BRCA1-deficient cells are sensitive to RD-N or combined with cisplatin
Since RD-N induced BRCA1 degradation through CTSB, we sought to determine whether down-regulation of BRCA1, which is critical in maintaining genomic integrity by promoting homologous recombination (HR), would affect RD-N mediated apoptosis of cancer cells. As shown in Figure 6A, knockdown of the endogenous BRCA1 resulted   effects. 30 Targeting lysosomes therefore have great therapeutic potential in cancer, because it not only triggers apoptotic and lysosomal cell death pathways but also reverses drug resistance. 29,31 We provide evidence here that targeting the lysosome by RD-N causes a cathepsin-dependent apoptosis via induction of DNA damage.
Enhanced translocation of CTSB to the nucleus by RD-N promoted the degradation of phosphor-BRCA1, failing to repair damaged-DNA. Importantly, RD-N significantly increased cisplatin cytotoxicity, because lysosomal transporters are shown to mediate cellular resistance to cisplatin. 32 Despite the substantial investigations performed on CTSB, little is known about the mechanisms by which this protein promotes caspase-independent apoptosis. 33 CTSB may play two opposing roles in malignancy: as an executioner of apoptosis in cytotoxic signalling cascades and a mediator of tumour invasion. 34,35 In this study, we provided a novel mechanism for the pro-apoptotic action of CTSB. Lysosomotropic RD-N-mediated apoptosis, at least in part, required the nuclear translocation of CTSB and BRCA1 rather than caspase activation. Together with our previous results, 1 we propose a novel sequence of molecular events for CTSB-dependent cell death; namely, RD-N-induced the translocation of CTSB from the lysosomes to the nucleus, where CTSB triggered DNA damage, and mediates degradation of BRCA1 to cause DNA repair defects. The degradation effect of CTSB on BRCA1 may be similar to the action of trypsin, but CTSB is a highly specific enzyme. 36 Some studies have shown that CTSB may exert its digest effect intracellularly or extracellularly, depending on the cell type and location of CTSB. 37,38 Our study demonstrated that migration of CTSB to the nucleus was critical in mediating the DNA damage induced by RD-N. Unlike to RD-N, DNA-damaging agent etoposide-mediated apoptosis seems not require the nuclear redistribution of CTSB.
Therefore, CTSB-mediated DNA damage was variable in response to agents.
It has been known that acidic proteases, but not the proteasome or calpain, degradeBRCA1 in DU145, SKBR3 and MCF7 cells, 39 however, no specific cathepsin was identified for the degradation. We showed here that CTSB could negatively regulate BRCA1 expression, and enzyme-negative domain of CTSB failed to suppress the phosphor-BRCA1. Although CTSB and phospho-BRCA1 did not co-immunoprecipitate in RD-N-treated cells, probably it is difficult to get an enzyme-substrate complex, the available crystal structures of BRCA1 ring domain was able to bind to the catalytic cysteine residue of CTSB. We may conclude that degradation of BRCA1 by CTSB, at least partially, leads to increased level of γH2AX in cells treated with RD-N. In response to DNA damage, cells could undergo apoptosis if the damaged DNA fails to be repaired. 40,41 Thus, the specific redistribution of CTSB in the nucleus appears to be essential to the CTSB/ BRCA1-axis action, leading to the CTSB-dependent cell death in response to RD-N.
One of the first responses to the production of DSBs in DNA is the rapid generation of phosphorylated H2AX at Ser139 near the DNA break point. Our data showed that H2AX is phosphorylated after RD-N treatment for 12 hours, at which the phosphorylation is crucial for CTSB-dependent DNA damage. Each type of DNA damage elicits a specific cellular repair response. 42,43 BRCA1 is a nuclear tumour suppressor that is critical for resolving DSBs and interstrand crosslinks by HR. BRCA1-deficient cancers are highly sensitive to chemotherapeutic agents. 44,45 In our study, BRCA1 was rapidly activated upon RD-N treatment, followed by significant down-regulation after 12 hours exposure, indicating the involvement of BRCA1 in RD-N-mediated DNA damage. As expected, BRCA1 depletion resulted in enhancement of cell death induced by RD-N. Importantly, targeting lysosomes by RD-N caused genomic instability, which in turn increased cell response to cisplatin in cells, and more evident in BRCA1-deficient cells.
In summary, targeting lysosomes by RD-N engages a novel cell death pathway that can be independent of classical apoptotic pathways. This offers a novel option for induction of cell death in tumour cells resistant to DNA damage agents. In particular, the use of RD-N as a single agent or in combination with DNA damage agents is a potential strategy for cancer treatment, especially for HR-deficient tumours. Lysosomes are therefore promising drug targets because lysosomal sequestration of chemotherapeutics is implicated in mediating drug resistance. This study might prove a promising agent that targets lysosomes with a novel mechanism for PCa treatment.

ACKNOWLEDG EMENTS
We thank Dr. Ted Hupp (University of Edinburgh) for kindly providing us the plasmid of GFP-CTSB and the professor of Zhihui Feng (University of Shandong) for kindly providing us the plasmid of pcDNA-BRCA1. We also thank Dr. Wanjun Chen (NIH/NIDCR) for F I G U R E 6 Breast cancer 1 protein (BRCA1) down-regulation is essential for RD-N-induced cell death. (A) Scramble siRNA and BRCA1 siRNA cells were untreated (control) or treated with RD-N (6 μmol/L, 8 h), labelled with PI and analysed by flow cytometry. (B) Scramble siRNA and BRCA1 siRNA PC3 cells were treated with RD-N (6 μmol/L) and stained with DAPI to visualize nuclei. Scale bar, 15 µm. (C) PC3 cells were transduced with an siRNA for depletion of BRCA1 (siBRCA1) or a control siRNA (siScramble) and BRCA1 levels were assessed by Western blot. (D) siScramble and siBRCA1 cells were incubated with poly (ADP-ribose) polymerase inhibitor (Pi) or cisplatin and RD-N, and the extent of cell death was assessed by flow cytometry. Data are the means of three independent experiments ±SD, *P < 0.05, **P < 0.01. (E) Western blot to detect the level of BRCA1, γH2AX in prostate cancer cells treated with RD-N (6 μmol/L) and plasmid (pcDNA 3.1, pcDNA BRCA1). (