Mechanism of Action of Prethioviridamide, an Anticancer Ribosomally

Jul 31, 2019 - Thioviridamide, prethioviridamide, and JBIR-140, which are ribosomally synthesized and post-translationally modified peptides (RiPPs) ...
0 downloads 0 Views 4MB Size
Download PDF
Articles Cite This: ACS Chem. Biol. XXXX, XXX, XXX−XXX

pubs.acs.org/acschemicalbiology

Mechanism of Action of Prethioviridamide, an Anticancer Ribosomally Synthesized and Post-Translationally Modified Peptide with a Polythioamide Structure Shohei Takase,†,‡ Rumi Kurokawa,† Yasumitsu Kondoh,§ Kaori Honda,§ Takehiro Suzuki,∥ Teppei Kawahara,⊥,+ Haruo Ikeda,# Naoshi Dohmae,∥ Hiroyuki Osada,§ Kazuo Shin-ya,∇,○ Tetsuo Kushiro,‡ Minoru Yoshida,*,†,○,◆,¶ and Ken Matsumoto*,†,◆ Downloaded via GUILFORD COLG on August 1, 2019 at 11:52:34 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

†

Chemical Genomics Research Group, RIKEN Center for Sustainable Resource Science, Saitama 351-0198, Japan School of Agriculture, Meiji University, 1-1-1 Higashimita, Tama-ku, Kawasaki, Kanagawa 214-8571, Japan § Chemical Biology Research Group, RIKEN Center for Sustainable Resource Science, Saitama 351-0198, Japan ∥ Biomolecular Characterization Unit, RIKEN Center for Sustainable Resource Science, Saitama 351-0198, Japan ⊥ Japan Biological Informatics Consortium (JBIC), 2-4-7 Aomi, Koto-ku, Tokyo 135-0064, Japan # Kitasato Institute for Life Sciences, Kitasato University, 1-15-1 Kitasato, Sagamihara, Kanagawa 252-0373, Japan ∇ National Institute of Advanced Industrial Science and Technology (AIST), 2-4-7 Aomi, Koto-ku, Tokyo 135-0064, Japan ○ Collaborative Research Institute for Innovative Microbiology, The University of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657, Japan ◆ Seed Compounds Exploratory Unit for Drug Discovery Platform, Drug Discovery Platforms Cooperation Division, RIKEN Center for Sustainable Resource Science, Saitama 351-0198, Japan ¶ Department of Biotechnology, Graduate School of Agricultural and Life Sciences, The University of Tokyo, Yayoi, Tokyo 113-8657, Japan ‡

S Supporting Information *

ABSTRACT: Thioviridamide, prethioviridamide, and JBIR140, which are ribosomally synthesized and post-translationally modified peptides (RiPPs) possessing five thioamide bonds, induce selective apoptosis in various cancer cells, especially those expressing the adenovirus oncogene E1A. However, the target protein of this unique family of bioactive compounds was previously unknown. To investigate the mechanism of action, we adopted a combined approach of genome-wide shRNA library screening, transcriptome profiling, and biochemical identification of prethioviridamidebinding proteins. An shRNA screen identified 63 genes involved in cell sensitivity to prethioviridamide, which included translation initiation factors, aminoacyl tRNA synthetases, and mitochondrial proteins. Transcriptome profiling and subsequent analysis revealed that prethioviridamide induces the integrated stress response (ISR) through the GCN2-ATF4 pathway, which is likely to cause cell death. Furthermore, we found that prethioviridamide binds and inhibits respiratory chain complex V (F1Fo-ATP synthase) in mitochondria, suggesting that inhibition of complex V leads to activation of the GCN2ATF4 pathway. These results imply that the members of a unique family of RiPPs with polythioamide structure target mitochondria to induce the ISR.

T

complex biological systems and develop novel strategies for drug discovery. To date, a large number of compounds have been isolated from nature that exhibit potent and specific bioactivities via unique modes of action. However, most of their mechanisms remain unknown, primarily due to the

he structural and functional diversity of secondary metabolites produced by plants or microorganisms has contributed greatly not only to improvement of human health by serving as medicines, but also to basic research as biochemical tools. Examples include FK506 for research on the immune response,1 rapamycin for nutrient signaling,2 trichostatin A for histone acetylation,3 and spliceostatin A for RNA splicing.4 All of these compounds have played pivotal roles in identifying key molecules that helped to unravel © XXXX American Chemical Society

Received: May 23, 2019 Accepted: July 18, 2019

A

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

Figure 1. Knockdown of aminoacyl tRNA synthetases and mitochondrial proteins affects sensitivity to prethioviridamide. (A) Structure of prethioviridamide. (B) HeLa S3 cells were treated with prethioviridamide for 48 h. WST-8 assays were performed to assess cell viability. (C) shRNA screening conditions. (D) An shRNA screen identified a total of 124 genes for which the corresponding shRNAs were enriched in prethioviridamide−treated cells. The hits were categorized as shown. (E) Eighty candidate genes, excluding those encoding ribosomal proteins and pseudogenes, were subjected to a secondary screen using siRNA pools targeting each gene. Cellular sensitivity to prethioviridamide upon individual knockdown of the 80 candidate genes could be categorized into four groups: apparently less responsive to prethioviridamide at all concentrations tested; less responsive to 10−100 nM prethioviridamide; less responsive to 100−1000 nM prethioviridamide; and no effect of knockdown on the sensitivity. (F) List of 63 hits from secondary screen. Genes whose products localize to mitochondria are shown in red (five stars of confidence in the COMPARTMENTS database) or pink (three or four stars). (G) Top 10 canonical pathways revealed by pathway analysis of the 63 hits.

difficulty of identifying the molecules responsible for their activities. Although several approaches to identifying target molecules for bioactive compounds have been developed, including phenotypic profiling, physical interaction, and genetic interaction, in general these efforts are still very challenging.5

Ribosomally synthesized and post-translationally modified peptides (RiPPs) are an emerging group of natural products that possess high chemical diversity, as with other families of compounds such as terpenoids, alkaloids, and polyketides.6,7 RiPPs are biosynthesized from ribosomally synthesized precursor peptides by multiple enzymes responsible for their post-translational modification and proteolysis. They have B

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

Figure 2. The GCN2-ATF4 pathway is activated by prethioviridamide, as determined by transcriptome analysis. (A) Top five enriched canonical pathways revealed by the Ingenuity Pathway Analysis (IPA). The gene list uploaded in the IPA software was collected from genes upregulated or downregulated by treatment with prethioviridamide in two DNA microarray analyses. (B) List of top 10 probes upregulated in prethioviridamide− treated cells and their fold change values (treated/untreated) revealed by one of the DNA microarray analyses. (C) HeLa S3 cells were treated with 0.1 μM or 1 μM prethioviridamide or 0.5 μM thapsigargin (Tg) for 2, 4, or 8 h, and the levels of ATF3, DDIT3, and GDF15 mRNAs were analyzed by qPCR. Data represent means ± SD from three independent experiments. *p < 0.05, **p < 0.01, and ***p < 0.001. (D) Cells were treated with 1 μM prethioviridamide or 0.5 μM thapsigargin for 1, 2, or 4 h. Protein levels of ATF4 and phosphorylation of GCN2 and PERK were analyzed by Western blotting. (E) ATF4 mRNA in cells treated with 0.1 μM or 1 μM prethioviridamide for 24 h was quantified by PCR. Data represent means ± SD from three independent experiments. *p < 0.05. (F) Cells were treated with 1 μM prethioviridamide for 2 or 4 h, or with 25 μM cycloheximide (CHX) for 5 min, and then pulse-labeled with 10 μg/mL puromycin for 10 min. Proteins were analyzed by Western blotting with antipuromycin, ATF4, and α-tubulin antibodies. (G) HeLa S3 cells were transfected with siRNA pools against seven candidate genes. Protein levels of ATF4, GCN2, and eIF2α and phosphorylation of GCN2 and eIF2α were analyzed by Western blotting.

pounds that contain thioamides are very rare, several thioviridamide-like RiPPs containing three or four thioamides were recently isolated by genome mining of Amycolatopsis alba and Streptomyces sp. These compounds include prethioviridamide, neothioviridamide/thiostreptamide S87/thioholga-

attracted increasing attention due to their wide range of biological activities, including anticancer, antiviral, antifungal, antibacterial, and allelopathic activities. Thioviridamide is a structurally unique RiPP that contains five consecutive thioamide bonds.8 Although natural comC

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

in each knockdown culture relative to that in control cells could be categorized into three classes. Knockdown of 28 genes made cells less responsive to prethioviridamide at all ranges of prethioviridamide concentrations, whereas knockdown of 26 genes was effective in the concentration range 10− 100 nM, and knockdown of 9 genes had an effect only in the range 100−1000 nM. When the cellular localization of gene products was analyzed using the COMPARTMENTS database (https://compartments.jensenlab.org/), many proteins were predicted to localize to mitochondria (Figure 1F). In addition, a pathway analysis for the 63 genes revealed that tRNA aminoacylation and translation initiation were the most enriched biological functions (Figure 1G). These results suggest that treatment of prethioviridamide affects mitochondria and translational initiation. The GCN2-ATF4 Pathway Is Activated by Prethioviridamide. We next examined the changes in mRNA expression following prethioviridamide treatment. For this purpose, we treated HeLa S3 cells with 4 μM prethioviridamide for 6 h, and analyzed total RNA by DNA microarray, using untreated cultures as a baseline for comparison. Two independent experiments identified 46 upregulated genes and 8 downregulated genes, respectively (Figure S3). A pathway analysis of the upregulated genes revealed a significant enrichment of genes involved in the unfolded protein response (UPR) and amino acid biosynthesis (Figure 2A). Furthermore, Connectivity Map analysis of these upregulated genes identified the ER stress inducer thapsigargin as the compound yielding expression signatures most similar to those of prethioviridamide (https://www.broadinstitute.org/connectivity-mapcmap). Indeed, expression of genes involved in the UPR, such as GDF15, DDIT3/CHOP, and ATF3, was upregulated in the DNA microarray analysis (Figure 2B), and this was confirmed by quantitative PCR (qPCR) analysis (Figure 2C). GDF15, DDIT3, and ATF3 are transcriptional targets of ATF4, a transcription factor central to the ISR, and are involved in induction of apoptosis in cancer cells.16,17 Various cellular stresses activate one of the four protein kinases that can phosphorylate the alpha subunit of eukaryotic translation initiation factor eIF2 (eIF2α). Phosphorylated eIF2α acts as an inhibitor of eIF2B, a guanine nucleotide exchange factor for eIF2, which in turn decreases the level of the eIF2-GTP-Met-tRNAiMet ternary complex, causing a downregulation of global translation.18 Under these conditions, mRNAs with short upstream open reading frames such as ATF4 are translationally activated. ATF4 was significantly induced within 1 h upon treatment with prethioviridamide or thapsigargin at both the transcriptional and translational levels (Figure 2D and E). Next, we sought to determine which of the four eIF2α kinases is activated upon prethioviridamide treatment by examining their autophosphorylation. We found that prethioviridamide induced phosphorylation of GCN2, whereas thapsigargin induced PERK phosphorylation in order to activate the ER stress response (Figure 2D). GCN2 is activated under stress conditions, such as amino acid deficiency or oxidative stress, by binding to uncharged tRNAs through its histidyl-tRNA synthetase−like region, which depends on its interaction with GCN1.19,20 In accordance with this model,18 prethioviridamide treatment for 2 or 4 h inhibited global translation, as assessed by pulse-labeling of nascent polypeptides by puromycin (Figure 2F). Therefore, we conclude that

mide A, thioalbamide, thiostreptamide S4, and thioholgamide B, which are cytotoxic against various human cancer cell lines (Figure S1).9−11 JBIR-140, a derivative of thioviridamide that contains five thioamide bonds, is biosynthesized from a 12 amino acid precursor peptide, VMAAAASIALHC, in Streptomyces avermitilis SUKA17.12 In addition to their structural uniqueness, these thioviridamide-like compounds induce selective apoptosis in cells transformed by adenoviral oncogenes.8,12,13 The unique characteristic features and essential genes of the thioamide-containing RiPP biosynthesis pathway have already been characterized,9,10,13 but little is known about mechanism of action of the thioviridamide family of molecules. In this study, we used multiple approaches to identify the target pathway of prethioviridamide. A combination of transcriptome profiling, genome-wide shRNA library screening, and target protein fishing using prethioviridamide-conjugated beads revealed that prethioviridamide induces the integrative stress response (ISR) through activation of the GCN2-ATF4 pathway by directly binding to mitochondrial respiratory chain complex V (F1Fo-ATP synthase).

■

RESULTS AND DISCUSSION shRNA Screen with Prethioviridamide. Thioviridamide is an acetone adduct of prethioviridamide, the true product of the biosynthetic gene cluster (Figure 1A and Figure S1).13 Like thioviridamide, prethioviridamide inhibits proliferation of rat 3Y1 cells transformed with adenoviral oncogenes13 as well as HeLa S3 human cervical carcinoma cells (IC50 = 0.36 μM) (Figure 1B). To explore the mode of action of the thioviridamide family of compounds, we searched for genes that modified sensitivity to prethioviridamide by performing a pooled shRNA library screen. We targeted ∼15,000 human genes using the DECIPHER lentiviral library Human Modules 1, 2, and 3, each of which contains five or six shRNAs against ∼5,000 independent genes, in separate transductions. HeLa S3 cells transduced with the libraries were treated or not treated with 0.3 μM prethioviridamide for 7 days (Figure 1C). The shRNA-specific barcodes were amplified from the genomic DNA of prethioviridamide-treated or untreated cells, and the PCR products were subjected to deep sequencing. The normalized read counts of each barcode were compared between treated and untreated cells. We obtained 92 genes with two or more shRNAs with a fold change (ratio of read counts in prethioviridamide-treated vs untreated) ≤ 0.5 (Table S2), and 124 genes with four or more shRNAs with a fold change ≥2.5 (Figure 1D and Table S3), including genes encoding 42 ribosome proteins. Because many ribosomal genes are essential for cell growth,14,15 it would be difficult to assess the effect of knockdown of these genes on sensitivity to cytotoxic compounds. Therefore, we excluded the genes encoding ribosomal proteins, as well as two pseudogenes, for further analysis. We reasoned that if a gene is involved in the pathway(s) targeted by prethioviridamide, then depletion of the gene product by siRNA knockdown should inhibit cell growth in a similar way to prethioviridamide and make cells apparently less responsive to prethioviridamide. Therefore, we focused on the remaining 80 genes with fold change ≥ 2.5, and performed secondary screening by knockdown using siRNA pools against individual genes (Figures 1E and S2). Knockdown of 63 out of the 80 genes made cells apparently less responsive to prethioviridamide. The degree of prethioviridamide sensitivity D

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology prethioviridamide activates the ISR via the GCN2-ATF4 pathway. An shRNA screen followed by siRNA knockdown experiments identified 63 candidate genes whose knockdown reduced sensitivity to prethioviridamide (Figure 1E). To look for genes whose knockdown induces ATF4 activation even in the absence of prethioviridamide, we rescreened the 80 candidates according to the effect of their knockdown on expression of the ATF4 target mRNAs ATF3 and DDIT3. We identified seven genes whose knockdown induced an increase in both ATF3 and DDIT3 mRNA expression (Figure S4), and also induced ATF4 protein expression (Figure 2G). Among the seven genes, four (ABCB7, HSPA9, HSPE1, and PHB2) encode mitochondrial proteins, whereas the other three (EIF2S3, QARS, and SARS) encode a subunit of eIF2 and aminoacyl tRNA synthetases, in which defects are known to activate GCN2 by increasing the level of uncharged tRNAs. Prethioviridamide Affects Mitochondrial Morphology. The findings described above implied that mitochondria are involved in the mode of action of prethioviridamide. Thapsigairgin, which inhibits sarco/endoplasmic reticulum Ca2+-ATPase and thereby activates the ER stress response, is known to induce the mitochondrial fragmentation by a transient increase in the mitochondrial Ca2+ levels.21 Accordingly, we examined the morphology and activity of mitochondria in HeLa S3 cells treated with prethioviridamide. MitoTracker Red staining revealed that the tubular network of mitochondria was partly lost within 1 h and almost completely lost 2 h after the addition of 1 μM prethioviridamide (Figure 3). Co-staining with antibodies against Tom20, which localizes at the mitochondrial outer membrane, revealed that doughnutlike structures were formed by a 2 h treatment with prethioviridamide. Some of the doughnut-like structures were devoid of MitoTracker Red signal, which depends on mitochondrial membrane potential. For comparison, we also examined the morphology of mitochondria in cells treated with oligomycin A, an F1Fo-ATP synthase inhibitor, and carbonyl cyanide-p-trifluoromethoxyphenylhydrazone (FCCP), an uncoupler of the respiratory chain. Consistent with previous work,22 both treatments resulted in fragmented and dot- or doughnut-like mitochondria, as determined by anti-Tom20 immunostaining. FCCP induced the complete disappearance of the MitoTracker Red signal. In addition, the dotlike mitochondria structures were seen upon prethioviridamide treatment in E1A-3Y1 cells, in which the network of mitochondria was already shorter than 3Y1 cells even in the absence of the drug (Figure S5). These results further underscore the role of mitochondria in the mode of action of prethioviridamide. Prethioviridamide Targets F1Fo-ATP Synthase. We next tried to identify proteins that specifically bind prethioviridamide. Because we had no information about structure−activity relationship, it was difficult to determine the optimal way to chemically conjugate prethioviridamide to beads without impairing its activity. Therefore, we used the beads with a photoaffinity linker containing an aryl diazirine group to make prethioviridamide-affinity beads. This approach enables immobilization of a compound of interest in the linker in a random orientation following UV irradiation.23 HeLa cell lysates were precleared with control beads, preincubated with or without 50 μg/mL prethioviridamide, and then incubated with prethioviridamide−conjugated beads (Figure 4A). Many proteins coprecipitated specifically with prethioviridamide

Figure 3. Prethioviridamide affects mitochondrial morphology. HeLa S3 cells were treated with 1 μM prethioviridamide, 20 μM oligomycin A, or 10 μM FCCP for 1 or 2 h and stained with MitoTracker Red CMXRos. Cells were then fixed and immunostained with antiTom20 antibody. Bars: 10 μm.

beads. Among them, a protein band around 50 kDa disappeared when free prethioviridamide was added as a competitor. Mass spectrometry analysis revealed that this ∼50 kDa band contained ATP5B, a subunit of mitochondrial F1 ATP synthase. Western blotting analysis confirmed the binding of prethioviridamide to ATP5B in HeLa cell lysates, whereas neither histone H3 nor β-actin coprecipitated with prethioviridamide beads (Figure 4B). Prethioviridamide binding to ATP5B became undetectable in the presence of free prethioviridamide, suggesting that the interaction with ATP5B was specific. We next investigated whether prethioviridamide affects the F1Fo-ATPase activity in vitro. Like oligomycin A, prethioviridamide inhibited ATPase activity in a dose-dependent manner (Figure 5A and B). Prethioviridamide and oligomycin A induced similar morphological changes in mitochondria (Figure 3). Analysis of the bioenergetic profile revealed that E

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

Figure 4. Prethioviridamide binds the mitochondrial F1Fo-ATP synthase. HeLa S3 cell lysates were precleared with control beads, preincubated with DMSO or 50 μg/mL prethioviridamide, and then incubated with prethioviridamide beads. The reacted beads were washed, and bound proteins were eluted in SDS sample buffer, subjected to SDS-PAGE, and visualized by silver staining (A). Cell lysates (lane C) and eluted proteins were analyzed by Western blotting (B).

in 3Y1 cells. E1A binds to a tumor suppressor retinoblastoma protein (pRB), thereby perturbing the interaction between pRB and a transcription factor E2F, resulting in the activation of E2F-dependent promoters (reviewed in ref 24). Consistent with the fact that GCN2 and eIF2α are included in a database of known E2F target genes (GSEA HALLMARK_E2F_TARGETS, http://software.broadinstitute.org/gsea/msigdb/cards/ HALLMARK_E2F_TARGETS),25 the levels of GCN2 and eIF2α mRNAs in E1A-3Y1 cells were twice as high as those in 3Y1 cells, whereas the amounts of ATF4 and PERK mRNAs did not significantly differ (Figure 6B). These results suggest that GCN2-ATF4 signaling is reinforced in E1A-expressing cells. Conclusions. To identify target pathway(s) and protein(s) of prethioviridamide, we performed pooled barcoded shRNA library screening, transcriptome profiling, and biochemical identification of prethioviridamide-binding proteins (Figure 6C). Our data showed that prethioviridamide targets mitochondrial F1Fo-ATP synthase and induces the ISR through the GCN2-ATF4 pathway. The connection between mitochondrial defects and ISR activation has been observed in several organisms.26,27 It remains unclear, however, how the GCN2-mediated ISR is induced upon inhibition of mitochondrial function. It seems possible that a rapid decrease in mitochondrial ATP impairs production of aminoacyl AMP, the first step in the reaction catalyzed by aminoacyl-tRNA synthetase, which requires ATP. Decrease in the aminoacylation activity would lead to the accumulation of uncharged tRNAs, which activate GCN2. The mechanism by which activation of ATF4 and its downstream targets results in distinct cell fates (cell survival or cell death) is also unknown. Although the activation of ATF4 target genes and inhibition of global translation upon exposure to mitochondrial stressors may protect cells during the early phase of a stress response, prolonged exposure to compounds such as FCCP27 and prethioviridamide (this study) is likely to induce apoptosis because these drugs induce expression of CHOP, an important pro-apoptotic protein.17,28 Targeting mitochondrial functions in cancer is an attractive therapeutic approach.29,30 Although energy production in cancer cells is reprogrammed to depend on aerobic glycolysis, inhibition of mitochondrial function severely impacts cell

prethioviridamide caused time-dependent inhibition of O2 consumption, as does oligomycin A, indicating that it inhibits mitochondrial function (Figure 5C). Consistent with this, combined treatment with the glycolytic inhibitor 2-deoxyglucose sensitized HeLa S3 cells to prethioviridamide (Figure 5D). To further confirm the synthetic effect of prethioviridamide with glycolysis inhibition, we knocked down hexokinase 2 (HK2), the enzyme that phosphorylates glucose at the first step of glycolysis and localizes to the outer membrane of mitochondria. siRNA-mediated downregulation of HK2, but not HK1, made cells more sensitive to prethioviridamide than treatment with control siRNAs (Figure S6), consistent with our initial shRNA screen, in which HK2 was one of the genes whose shRNAs were most significantly depleted in prethioviridamide-treated cells relative to untreated cells (Table S2). Importantly, E1A-3Y1 cells were more sensitive to oligomycin A than parental 3Y1 cells (Figure S6). These results indicate that selective killing of E1A-3Y1 cells by prethioviridamide was due to the inhibition of mitochondrial F1Fo-ATP synthase activity. In addition, we investigated whether inhibition of F1Fo-ATP synthase activates the GCN2-ATF4 pathway (Figure 5E). Indeed, treatment with 0.1 μM oligomycin A or 1 μM prethioviridamide induced a dramatic increase in ATF4. Moreover, this effect was significantly diminished by knockdown of GCN2, but not by knockdown of other eIF2α kinases. E1A Expression Accelerates the GCN2-ATF4 Pathway. As reported for thioviridamide,8 prethioviridamide and oligomycin A also exhibited selective cytotoxicity against E1A-3Y1 cells with an ∼50-fold lower IC50 than 3Y1 cells (Figure S6). However, it is not known why cells expressing adenovirus E1A are more sensitive to these compounds than the corresponding parental cells. To determine whether the GCN2-ATF4 pathway is involved in this selective killing of E1A-expressing cells, we examined the expression of ATF4 in E1A-3Y1 and 3Y1 cells treated with prethioviridamide. Prethioviridamide induced a marked increase in ATF4 expression in E1A-3Y1 cells in a time-dependent manner, similar to the effect in HeLa S3 cells, but to a lesser extent in 3Y1 cells (Figure 6A, compare with Figure 2D). We noted that the basal protein levels of GCN2 and eIF2α and the induced protein level of ATF4 in E1A-3Y1 cells were higher than those F

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

Figure 5. Prethioviridamide inhibits mitochondrial F1Fo-ATPase activity. (A and B) The F1Fo-ATPase activity was measured in the presence of an increasing amount of prethioviridamide (A) and oligomycin A (B). Data represent means ± SD from three independent experiments. (C) Oxygen consumption rate (OCR) was analyzed following treatment with 1 μM prethioviridamide or 125 nM oligomycin A. (D) HeLa S3 cells were treated for 4 days with various concentrations of prethioviridamide in the presence of 0, 0.3, or 1 mM 2-DG. WST-8 assays were performed, and values are shown relative to a sample not treated with prethioviridamide (top graph). Viability of cells treated with 0, 0.3, or 1 mM 2-DG is shown (bottom graph). Data represent means ± SD from three independent experiments. (E) Cells were transfected with siRNA pools against one of the four eIF2α kinases or control siRNA (Ctrl siRNA), and then treated with 1 μM prethioviridamide for 4 h or 0.1 μM oligomycin A for 2 h. Proteins were analyzed by Western blotting, and mRNA knockdown was validated by qPCR. Data represent means ± SD from three independent experiments.

perturbations of the interaction between pRB and E2F.24,32,33 Expression of E1A in various cultured cells increases sensitivity to multiple cytotoxic reagents, such as cisplatin, paclitaxel, doxorubicin, gamma irradiation, and TNF-α treatment.34,35 E1A makes cells sensitive to nitric oxide by inducing mitochondrial dysfunction−mediated apoptosis.36 We observed that mitochondrial morphology was different in 3Y1 and E1A-3Y1 cells. Therefore, it is possible that expression of E1A could modify mitochondrial functions or structures, making cells more sensitive to mitochondrial stresses such as prethioviridamide treatment. We also found that the mRNA

viability, suggesting that metabolic reprogramming renders cancer cells susceptible to mitochondrial perturbations. We examined the stability of prethioviridamide in growth medium31 and found that it has a half-life of ∼40 min; accordingly, its effect on mitochondrial morphology was slightly weakened by preincubation for 2 h in growth medium (Figure S7). Improved pharmacological stability of polythioamide-containing peptides will be needed to represent a novel seed compound with therapeutic potential. Adenovirus E1A exerts its functions through altering protein−protein interactions, as exemplified by E1A-mediated G

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology

Figure 6. ATF4 induction determines prethioviridamide sensitivity. (A) 3Y1 and E1A-3Y1 cells treated with 1 μM prethioviridamide for 1, 2, or 4 h. Protein levels of ATF4, GCN2, and eIF2α were analyzed by Western blotting. (B) Relative levels of GCN2, eIF2α, ATF4, and PERK mRNAs were analyzed by qPCR. Data represent means ± SD from three independent experiments. *p < 0.05 and ***p < 0.001. n.s., not significant. (C) Model for the mode of action of prethioviridamide. changes in mitochondria morphology, cells were stained with 100 nM MitoTracker Red CMXRos (Thermo Fisher Scientific) at 37 °C for 30−45 min, washed in warm phosphate-buffered saline (PBS), and then fixed with warm 4% paraformaldehyde in PBS at 37 °C for 20 min. The cells were examined under a fluorescence microscope (IX81, Olympus). To examine the effect on protein synthesis activity, puromycin labeling was performed as described previously.41,42 Briefly, after treatment with the compounds, the cells were incubated for 10 min with growth medium containing 10 μg/mL puromycin. The cell lysates were analyzed by Western blotting with antipuromycin antibody. Quantitative PCR. Total cellular RNA was prepared using the RNeasy mini kit (Qiagen), and cDNA was synthesized using the the PrimeScript RT reagent kit (Takara Bio). Expression of target genes was analyzed using SYBR Premix Ex TaqII (Takara Bio) on a Thermal Cycler Dice Real Time system (Takara Bio). Primers are listed in Table S4. Data were analyzed by two-sided unpaired Student’s t test. Differences were considered statistically significant when p was less than 0.05. Lentiviral shRNA Screening. DECIPHER barcoded shRNA libraries (Human Modules 1, 2, and 3) were obtained from Cellecta. Packaging into lentiviral particles using 293LTV cells and transduction of HeLa S3 cells were performed as described previously.43,44 Aliquots of transduced cells were treated or untreated with 0.3 μM prethioviridamide. The cells were passaged as necessary to maintain exponential growth in the presence or absence of the compounds for 7 days and then collected. Genome DNA extraction, PCR amplification of barcodes with indexed primers, and barcode quantitation by next-generation sequencing were performed as described previously.44 siRNA Screening. HeLa S3 cells were transfected with 10 nM siRNA pools (Human siGENOME SMARTpool, GE Dharmacon) containing four oligos against each gene, or with siGENOME NonTargeting siRNA #2, using Lipofectamine RNAiMAX reagent

and protein levels of GCN2 and eIF2α, both of which are E2F targets, were significantly upregulated in E1A-3Y1 cells, suggesting that E1A elevates expression of these genes in 3Y1 cells. Therefore, it is also possible that E1A sensitizes cells to prethioviridamide by intrinsically elevating the expression of the components of the GCN2−eIF2α−ATF4 pathway. Although identification of targets or modes of action of bioactive compounds remains difficult, the combined use of phenotypic profiling, genetic interaction, and physical interaction has proven to be a powerful approach.5,37,38 In this study, we verified the effectiveness of these multifaceted approaches to search for the determinants of sensitivity to a bioactive compound. The combination of genome-wide shRNA library screening, transcriptome profiling, and binding protein screening using affinity beads provided an unbiased and systematic means of identifying targets of compounds. Further studies aimed at identifying target molecules or pathways by this method will unlock the potential of various natural products as therapeutic seeds and biochemical tools.

■

METHODS

Cell Culture. Rat fibroblast 3Y1 and E1A-3Y1 cells were obtained from RIKEN BioResource Center Cell Bank.39,40 The human cervical cancer cell line HeLa S3 was tested for nine short tandem repeat loci and the amelogenin gene in February 2017 and April 2019. These cells were maintained in growth medium (Dulbecco’s modified Eagle’s medium [DMEM] supplemented with 10% heat-inactivated fetal bovine serum (FBS) and penicillin/streptomycin) at 37 °C under 5% CO2. The WST-8 [2-(2-methoxy-4-nitrophenyl)-3-(4nitrophenyl)-5-(2,4-disulfophenyl)-2H-tetrazolium] assay using Cell Count Reagent SF (Nacalai Tesque) was performed to measure cell viability in an iMark microplate reader (Bio-Rad). To examine H

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology (Thermo Fisher Scientific). After 48 h, transfected cells were split and seeded in 96-well plates and then treated with various concentrations of compounds for 48 h. Western Blotting. Preparation of cell lysates and Western blotting were performed as described previously.44 Primary antibodies used were antibodies against ATF4, PKR, PERK, GCN2, eIF2α, phospho-eIF2α (Cell Signaling Technology), α-tubulin, β-actin (Sigma), phospho-GCN2, histone H3, ATP5B (Abcam), and puromycin (Merck). Immunocytochemistry. Cells were stained with MitoTracker Red CMXRos and fixed as described above. The following steps were performed at RT, except for incubation with the first antibody (antiTom20, Santa Cruz Biotechnology), which was performed at 4 °C overnight. The fixed cells were washed with 0.3% Triton X-100 in PBS and blocked with 1% bovine serum albumin in PBS. After blocking, immunocytochemistry was performed as described previously.45,46 Cells were examined under a confocal microscope (Olympus FV1200) using a UPLSAPO100XS 100× silicon oil immersion lens (NA 1.35).

■

work was inspired by the Asian Chemical Biology Initiative, JSPS.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acschembio.9b00410.

■

REFERENCES

(1) Harding, M. W., Galat, A., Uehling, D. E., and Schreiber, S. L. (1989) A receptor for the immunosuppressant FK506 is a cis-trans peptidyl-prolyl isomerase. Nature 341, 758−760. (2) Brown, E. J., Albers, M. W., Shin, T. B., Ichikawa, K., Keith, C. T., Lane, W. S., and Schreiber, S. L. (1994) A mammalian protein targeted by G1-arresting rapamycin-receptor complex. Nature 369, 756−758. (3) Yoshida, M., Kijima, M., Akita, M., and Beppu, T. (1990) Potent and specific inhibition of mammalian histone deacetylase both in vivo and in vitro by trichostatin A. J. Biol. Chem. 265, 17174−17179. (4) Kaida, D., Motoyoshi, H., Tashiro, E., Nojima, T., Hagiwara, M., Ishigami, K., Watanabe, H., Kitahara, T., Yoshida, T., Nakajima, H., Tani, T., Horinouchi, S., and Yoshida, M. (2007) Spliceostatin A targets SF3b and inhibits both splicing and nuclear retention of premRNA. Nat. Chem. Biol. 3, 576−583. (5) Yoshida, M. (2019) Recent advances in target identification of bioactive natural products. Biosci., Biotechnol., Biochem. 83, 1−9. (6) Arnison, P. G., Bibb, M. J., Bierbaum, G., Bowers, A. A., Bugni, T. S., Bulaj, G., Camarero, J. A., Campopiano, D. J., Challis, G. L., Clardy, J., Cotter, P. D., Craik, D. J., Dawson, M., Dittmann, E., Donadio, S., Dorrestein, P. C., Entian, K. D., Fischbach, M. A., Garavelli, J. S., Goransson, U., Gruber, C. W., Haft, D. H., Hemscheidt, T. K., Hertweck, C., Hill, C., Horswill, A. R., Jaspars, M., Kelly, W. L., Klinman, J. P., Kuipers, O. P., Link, A. J., Liu, W., Marahiel, M. A., Mitchell, D. A., Moll, G. N., Moore, B. S., Muller, R., Nair, S. K., Nes, I. F., Norris, G. E., Olivera, B. M., Onaka, H., Patchett, M. L., Piel, J., Reaney, M. J., Rebuffat, S., Ross, R. P., Sahl, H. G., Schmidt, E. W., Selsted, M. E., Severinov, K., Shen, B., Sivonen, K., Smith, L., Stein, T., Sussmuth, R. D., Tagg, J. R., Tang, G. L., Truman, A. W., Vederas, J. C., Walsh, C. T., Walton, J. D., Wenzel, S. C., Willey, J. M., and van der Donk, W. A. (2013) Ribosomally synthesized and post-translationally modified peptide natural products: overview and recommendations for a universal nomenclature. Nat. Prod. Rep. 30, 108−160. (7) Ortega, M. A., and van der Donk, W. A. (2016) New Insights into the Biosynthetic Logic of Ribosomally Synthesized and Posttranslationally Modified Peptide Natural Products. Cell Chem. Biol. 23, 31−44. (8) Hayakawa, Y., Sasaki, K., Adachi, H., Furihata, K., Nagai, K., and Shin-ya, K. (2006) Thioviridamide, a novel apoptosis inducer in transformed cells from Streptomyces olivoviridis. J. Antibiot. 59, 1−5. (9) Frattaruolo, L., Lacret, R., Cappello, A. R., and Truman, A. W. (2017) A Genomics-Based Approach Identifies a Thioviridamide-Like Compound with Selective Anticancer Activity. ACS Chem. Biol. 12, 2815−2822. (10) Kjaerulff, L., Sikandar, A., Zaburannyi, N., Adam, S., Herrmann, J., Koehnke, J., and Muller, R. (2017) Thioholgamides: ThioamideContaining Cytotoxic RiPP Natural Products. ACS Chem. Biol. 12, 2837−2841. (11) Kawahara, T., Izumikawa, M., Kozone, I., Hashimoto, J., Kagaya, N., Koiwai, H., Komatsu, M., Fujie, M., Sato, N., Ikeda, H., and Shin-Ya, K. (2018) Neothioviridamide, a Polythioamide Compound Produced by Heterologous Expression of a Streptomyces sp. Cryptic RiPP Biosynthetic Gene Cluster. J. Nat. Prod. 81, 264− 269. (12) Izumikawa, M., Kozone, I., Hashimoto, J., Kagaya, N., Takagi, M., Koiwai, H., Komatsu, M., Fujie, M., Satoh, N., Ikeda, H., and Shin-ya, K. (2015) Novel thioviridamide derivative–JBIR-140: heterologous expression of the gene cluster for thioviridamide biosynthesis. J. Antibiot. 68, 533−536. (13) Izawa, M., Nagamine, S., Aoki, H., and Hayakawa, Y. (2018) Identification of essential biosynthetic genes and a true biosynthetic product for thioviridamide. J. Gen. Appl. Microbiol. 64, 50−53. (14) Blomen, V. A., Majek, P., Jae, L. T., Bigenzahn, J. W., Nieuwenhuis, J., Staring, J., Sacco, R., van Diemen, F. R., Olk, N.,

Detailed methods; 13C NMR and 1H NMR data for prethioviridamide; candidate genes from shRNA library screen; qPCR primers used in the study; thioviridamidelike peptide structures; up- and downregulated genes identified by DNA microarray; effect of prethioviridamide treatment on mitochondrial morphology; prthioviridamide stability in serum (PDF)

AUTHOR INFORMATION

Corresponding Authors

*E-mail: [email protected]. *E-mail: [email protected]. ORCID

Haruo Ikeda: 0000-0003-3977-7856 Hiroyuki Osada: 0000-0002-3606-4925 Kazuo Shin-ya: 0000-0002-4702-0661 Tetsuo Kushiro: 0000-0002-7598-4791 Ken Matsumoto: 0000-0002-7864-3394 Present Address +

T. Kawahara: Graduate School of Pharmaceutical Sciences, Kumamoto University, 5-1 Oe-honmachi, Chuo-ku, Kumamoto City 862-0973, Kumamoto, Japan. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS The authors wish to thank N. Suzuki, M. Tagami, and S. Aoki (GeNAS, RIKEN) for their help with deep sequencing, K. Fukumoto (Support Unit for Bio-Material Analysis, RRD, CBS, RIKEN) for DNA microarray analysis, and the RIKEN CBS-Olympus Collaboration Center for imaging equipment and software. This work was supported in part by JSPS under Grants-in-Aid for Scientific Research (S) (Grant Number 26221204), Scientific Research (C) (18K06678), and Scientific Research on Innovative Areas (16H06276) from the Ministry of Education, Culture, Sports, Science and Technology of Japan, and by AMED-CREST, AMED. This I

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX

Articles

ACS Chemical Biology Stukalov, A., Marceau, C., Janssen, H., Carette, J. E., Bennett, K. L., Colinge, J., Superti-Furga, G., and Brummelkamp, T. R. (2015) Gene essentiality and synthetic lethality in haploid human cells. Science 350, 1092−1096. (15) Hart, T., Chandrashekhar, M., Aregger, M., Steinhart, Z., Brown, K. R., MacLeod, G., Mis, M., Zimmermann, M., FradetTurcotte, A., Sun, S., Mero, P., Dirks, P., Sidhu, S., Roth, F. P., Rissland, O. S., Durocher, D., Angers, S., and Moffat, J. (2015) HighResolution CRISPR Screens Reveal Fitness Genes and GenotypeSpecific Cancer Liabilities. Cell 163, 1515−1526. (16) Wek, R. C., Jiang, H. Y., and Anthony, T. G. (2006) Coping with stress: eIF2 kinases and translational control. Biochem. Soc. Trans. 34, 7−11. (17) Pakos-Zebrucka, K., Koryga, I., Mnich, K., Ljujic, M., Samali, A., and Gorman, A. M. (2016) The integrated stress response. EMBO Rep. 17, 1374−1395. (18) Jackson, R. J., Hellen, C. U., and Pestova, T. V. (2010) The mechanism of eukaryotic translation initiation and principles of its regulation. Nat. Rev. Mol. Cell Biol. 11, 113−127. (19) Garriz, A., Qiu, H., Dey, M., Seo, E. J., Dever, T. E., and Hinnebusch, A. G. (2009) A network of hydrophobic residues impeding helix alphaC rotation maintains latency of kinase Gcn2, which phosphorylates the alpha subunit of translation initiation factor 2. Mol. Cell. Biol. 29, 1592−1607. (20) Lageix, S., Zhang, J., Rothenburg, S., and Hinnebusch, A. G. (2015) Interaction between the tRNA-binding and C-terminal domains of Yeast Gcn2 regulates kinase activity in vivo. PLoS Genet. 11, No. e1004991. (21) Hom, J. R., Gewandter, J. S., Michael, L., Sheu, S. S., and Yoon, Y. (2007) Thapsigargin induces biphasic fragmentation of mitochondria through calcium-mediated mitochondrial fission and apoptosis. J. Cell. Physiol. 212, 498−508. (22) Fu, D., and Lippincott-Schwartz, J. (2018) Monitoring the Effects of Pharmacological Reagents on Mitochondrial Morphology. Curr. Protoc Cell Biol. 79, No. e45. (23) Kanoh, N., Honda, K., Simizu, S., Muroi, M., and Osada, H. (2005) Photo-cross-linked small-molecule affinity matrix for facilitating forward and reverse chemical genetics. Angew. Chem., Int. Ed. 44, 3559−3562. (24) DeCaprio, J. A. (2009) How the Rb tumor suppressor structure and function was revealed by the study of Adenovirus and SV40. Virology 384, 274−284. (25) Cao, A. R., Rabinovich, R., Xu, M., Xu, X., Jin, V. X., and Farnham, P. J. (2011) Genome-wide analysis of transcription factor E2F1 mutant proteins reveals that N- and C-terminal protein interaction domains do not participate in targeting E2F1 to the human genome. J. Biol. Chem. 286, 11985−11996. (26) Baker, B. M., Nargund, A. M., Sun, T., and Haynes, C. M. (2012) Protective coupling of mitochondrial function and protein synthesis via the eIF2alpha kinase GCN-2. PLoS Genet. 8, No. e1002760. (27) Quiros, P. M., Prado, M. A., Zamboni, N., D’Amico, D., Williams, R. W., Finley, D., Gygi, S. P., and Auwerx, J. (2017) Multiomics analysis identifies ATF4 as a key regulator of the mitochondrial stress response in mammals. J. Cell Biol. 216, 2027−2045. (28) Jiang, H. Y., and Wek, R. C. (2005) Phosphorylation of the alpha-subunit of the eukaryotic initiation factor-2 (eIF2alpha) reduces protein synthesis and enhances apoptosis in response to proteasome inhibition. J. Biol. Chem. 280, 14189−14202. (29) Fulda, S., Galluzzi, L., and Kroemer, G. (2010) Targeting mitochondria for cancer therapy. Nat. Rev. Drug Discovery 9, 447− 464. (30) Biondani, G., and Peyron, J. F. (2018) Metformin, an Antidiabetic Drug to Target Leukemia. Front. Endocrinol. (Lausanne, Switz.) 9, 446. (31) Jenssen, H., and Aspmo, S. I. (2008) Serum stability of peptides. Methods Mol. Biol. 494, 177−186.

(32) Ferrari, R., Pellegrini, M., Horwitz, G. A., Xie, W., Berk, A. J., and Kurdistani, S. K. (2008) Epigenetic reprogramming by adenovirus e1a. Science 321, 1086−1088. (33) King, C. R., Zhang, A., Tessier, T. M., Gameiro, S. F., and Mymryk, J. S. (2018) Hacking the Cell: Network Intrusion and Exploitation by Adenovirus E1A. mBio 9, e00390-18. (34) Chen, M. J., Holskin, B., Strickler, J., Gorniak, J., Clark, M. A., Johnson, P. J., Mitcho, M., and Shalloway, D. (1987) Induction by E1A oncogene expression of cellular susceptibility to lysis by TNF. Nature 330, 581−583. (35) Brader, K. R., Wolf, J. K., Hung, M. C., Yu, D., Crispens, M. A., van Golen, K. L., and Price, J. E. (1997) Adenovirus E1A expression enhances the sensitivity of an ovarian cancer cell line to multiple cytotoxic agents through an apoptotic mechanism. Clin. Cancer Res. 3, 2017−2024. (36) Radke, J. R., Figueroa, I., Routes, J. M., and Cook, J. L. (2018) PIDD-dependent activation of caspase-2-mediated mitochondrial injury in E1A-induced cellular sensitivity to macrophage nitric oxide-induced apoptosis. Cell Death Discov 4, 97. (37) Kuai, L., Ong, S. E., Madison, J. M., Wang, X., Duvall, J. R., Lewis, T. A., Luce, C. J., Conner, S. D., Pearlman, D. A., Wood, J. L., Schreiber, S. L., Carr, S. A., Scolnick, E. M., and Haggarty, S. J. (2011) AAK1 identified as an inhibitor of neuregulin-1/ErbB4-dependent neurotrophic factor signaling using integrative chemical genomics and proteomics. Chem. Biol. 18, 891−906. (38) Farha, M. A., and Brown, E. D. (2016) Strategies for target identification of antimicrobial natural products. Nat. Prod. Rep. 33, 668−680. (39) Kimura, G., Itagaki, A., and Summers, J. (1975) Rat cell line 3y1 and its virogenic polyoma- and sv40- transformed derivatives. Int. J. Cancer 15, 694−706. (40) Shimura, H., Ohtsu, M., Matsuzaki, A., Mitsudomi, T., Onodera, K., and Kimura, G. (1988) Selective cytotoxicity of phospholipids and diacylglycerols to rat 3Y1 fibroblasts transformed by adenovirus type 12 or its E1A gene. Cancer Res. 48, 578−583. (41) Schmidt, E. K., Clavarino, G., Ceppi, M., and Pierre, P. (2009) SUnSET, a nonradioactive method to monitor protein synthesis. Nat. Methods 6, 275−277. (42) Silva, R. C., Sattlegger, E., and Castilho, B. A. (2016) Perturbations in actin dynamics reconfigure protein complexes that modulate GCN2 activity and promote an eIF2 response. J. Cell Sci. 129, 4521−4533. (43) Kobayashi, H., Nishimura, H., Matsumoto, K., and Yoshida, M. (2015) Identification of the determinants of 2-deoxyglucose sensitivity in cancer cells by shRNA library screening. Biochem. Biophys. Res. Commun. 467, 121−127. (44) Takase, S., Kurokawa, R., Arai, D., Kanemoto Kanto, K., Okino, T., Nakao, Y., Kushiro, T., Yoshida, M., and Matsumoto, K. (2017) A quantitative shRNA screen identifies ATP1A1 as a gene that regulates cytotoxicity by aurilide B. Sci. Rep. 7, 2002. (45) Matsumoto, K., Kose, S., Kuwahara, I., Yoshimura, M., Imamoto, N., and Yoshida, M. (2018) Y-box protein-associated acidic protein (YBAP1/C1QBP) affects the localization and cytoplasmic functions of YB-1. Sci. Rep. 8, 6198. (46) Tanaka, K. J., Ogawa, K., Takagi, M., Imamoto, N., Matsumoto, K., and Tsujimoto, M. (2006) RAP55, a cytoplasmic mRNP component, represses translation in Xenopus oocytes. J. Biol. Chem. 281, 40096−40106.

J

DOI: 10.1021/acschembio.9b00410 ACS Chem. Biol. XXXX, XXX, XXX−XXX